Eco-friendly combination of the immobilized PGA enzyme and the s-phacm protecting group for the synthesis of cys-containing peptides

Article abstract of DOI:10.1002/chem.201201370

Enzyme-labile protecting groups have emerged as a green alternative to conventional protecting groups. These groups introduce a further orthogonal dimension and eco-friendliness into protection schemes for the synthesis of complex polyfunctional organic molecules. S-Phacm, a Cys-protecting group, can be easily removed by the action of a covalently immobilized PGA enzyme under very mild conditions. Herein, the versatility and reliability of an eco-friendly combination of the immobilized PGA enzyme and the S-Phacm protecting group has been evaluated for the synthesis of diverse Cys-containing peptides.

Full text of DOI:10.1002/chem.201201370

DOI: 10.1002/chem.201201370
Eco-Friendly Combination of the Immobilized PGA Enzyme and the
S-Phacm Protecting Group for the Synthesis of Cys-Containing Peptides
Miriam Gꢀngora-Benꢁtez,[a, b] Alessandra Basso,[c] Thomas Bruckdorfer,[d]
Miriam Royo,[e] Judit Tulla-Puche,*[a, b] and Fernando Albericio*[a, b, f, g]
Abstract: Enzyme-labile protecting
groups have emerged as a green alter-
native to conventional protecting
groups. These groups introduce a fur-
ther orthogonal dimension and eco-
friendliness into protection schemes for
the synthesis of complex polyfunctional
organic molecules. S-Phacm, a Cys-pro-
tecting group, can be easily removed
by the action of a covalently immobi-
lized PGA enzyme under very mild
conditions. Herein, the versatility and
reliability of an eco-friendly combina-
tion of the immobilized PGA enzyme
and the S-Phacm protecting group has
been evaluated for the synthesis of di-
verse Cys-containing peptides.
Keywords: cysteine
· enzymes ·
green chemistry · peptides · protect-
ing groups
Introduction
“plus” for a protecting group is that it should be removable
under mild conditions. Nowadays, this term “mild condi-
tions” has a double meaning. Thus, deblocking conditions
not only should leave other parts of the molecule unaltered
but should also minimize—or avoid—harm to the environ-
ment. A paradigm of this ideal protecting group are
enzyme-labile protecting groups.[4] These groups offer feasi-
ble alternatives to classical chemical methods because they
can be removed selectively under mild conditions, thereby
installing an additional orthogonal dimension and eco-
friendliness in protection schemes.
The amino-acid sequence of a peptide or protein deter-
mines its 3D structure and, therefore, its biological activity.
In particular, Cys residues are involved in the disulfide-bond
scaffold, which is a key structural feature that has been im-
plicated in the folding and structural stability of many natu-
ral peptides and proteins. Thus, it is not surprising that Cys
is a potential target for incorporating modifications into the
peptide structure. For instance, the artificial introduction of
extra disulfide bridges into peptides or proteins, thereby al-
lowing the generation of conformational constraints, may
improve biological activity, bioavailability, and/or stability.
The inclusion of an extra Cys residue is commonly accom-
plished for further chemoselective derivatization, such as la-
beling or bioconjugation, by using a variety of electrophilic
molecules (haloacetyl groups, maleimides, disulfides, sulf-
The synthesis of complex polyfunctional organic molecules,
such as peptides, oligonucleotides, oligosaccharides, and the
conjugates thereof, as well as other challenging small com-
pounds, demands the implementation of an efficient protec-
tion scheme. The literature describes a myriad of protecting
groups, which were usually first developed for peptide
chemistry[1] and then rapidly adapted by organic chemists[2]
for the pursuit of attractive target molecules. The prepara-
tion of polyfunctional molecules, such as peptides, often re-
quires the concourse of a set of orthogonal protecting
groups, which are defined as those that are removed in any
order by means of a different chemical mechanism.[3]
A
[a] M. Gꢀngora-Benꢁtez, Dr. J. Tulla-Puche, Prof. F. Albericio
Institute for Research in Biomedicine, 08028-Barcelona (Spain)
Fax : (+34)93 4037126
E-mail: judit.tulla@irbbarcelona.org
fernando.albericio@irbbarcelona.org
[b] M. Gꢀngora-Benꢁtez, Dr. J. Tulla-Puche, Prof. F. Albericio
CIBER-BBN, 08028-Barcelona (Spain)
[c] Dr. A. Basso
Purolite International Ltd., Llantrisant, South Wales, CF72 8LF (UK)
[d] Dr. T. Bruckdorfer
IRIS Biotech GmbH, D-95615 Marktredwitz (Germany)
AHCTUNGTREGaNNNU midates, a,b-unsaturated esters, among others) for many
[e] Dr. M. Royo
research purposes.[5] Furthermore, the Cys residue is re-
quired to perform the native chemical ligation strategy,[6]
which is a potent approach for the synthesis of small-to-
medium-sized proteins.
Barcelona Science Park, 08028-Barcelona (Spain)
[f] Prof. F. Albericio
Department of Organic Chemistry, University of Barcelona
08028-Barcelona (Spain)
[g] Prof. F. Albericio
Therefore, suitable protection of the nucleophilic thiol
group of Cys calls for the preparation of complex Cys-con-
taining peptides, which may require the presence of various
Cys-protecting groups for regioselective disulfide-bond for-
mation.
School of Chemistry, University of KwaZulu Natal
Durban 4000 (South Africa)
Supporting information for this article, including experimental details
of peptide elongation by using solid-phase peptide synthesis, is availa-
ble on the WWW under http://dx.doi.org/10.1002/chem.201201370.
16166
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176

FULL PAPER
Over the past few decades, the rapid development of
enzyme-immobilization technology has promoted the emer-
gence of enzymatic catalysis in organic synthesis as an at-
tractive and sustainable alternative to the conventional
chemical approaches.[7] The activity and stability of enzymes
mainly depend on the particular operating and storage con-
ditions and they are strongly influenced by factors such as
temperature, pH, ionic strength, and solvent properties. In
general, immobilized enzymes have enhanced stability and
activity over a broad range of temperatures and pH values,
as well as higher tolerance toward organic solvents, com-
pared to soluble enzymes. Furthermore, immobilized en-
zymes can be easily separated from the reaction media by a
simple and rapid filtration, thereby allowing for their reuse
and application in continuous processes and in commercial
bulk-production processes. Thus, these features lead to sig-
nificant savings in terms of operating time, enzyme con-
sumption, and production costs by efficient recycling and
control of the process.[8]
Scheme 1. Schematic representation of the removal of S-Phacm by the
The covalent immobilization of penicillin G acylase
(PGA) from E. coli (E.C. 3.5.1.1) on an amino-acrylic resin,
which recognizes the phenylacetyl moiety, exhibits promis-
ing applications in peptide synthesis and opens up new per-
spectives for further applications.[9] Herein, we explore the
power of immobilized-PGA-enzyme catalysis and evaluate
several features of this process, including its high efficiency,
recyclability, its capacity to operate under mild conditions,
and its environmental friendliness, for the synthesis of Cys-
containing peptides.
immobilized PGA enzyme.
followed by extensive thiol treatment) or to directly render
a disulfide bond through oxidation (mediated by I2 or TlIII
salts). The requirement of toxic heavy metals and their diffi-
cult removal from the crude peptides have made some of
these procedures unviable, whilst the iodine-mediated de-
protection of S-Acm and oxidation has been the cornerstone
of numerous successful syntheses in peptide chemistry.[11]
Nonetheless, several side-reactions have been reported to
occur under S-Acm-deblocking conditions, for example, S!
N and S!O transfer of the Acm group onto side-chains, the
modification of sensitive residues, such as Trp, Tyr, and Met,
and the formation of intramolecular tryptophan-2-thioether
as a side-product.[12]
Results and Discussion
Phenylacetamidomethyl (Phacm) is a Cys-protecting group
that is compatible with the two major synthetic strategies in
peptide chemistry (Boc and Fmoc). Moreover, it can be re-
moved under similar conditions to the acetamidomethyl
(Acm) group and, in addition, by the action of the PGA
enzyme. Because PGA is selective towards the phenylacetyl
moiety, S-Phacm can be smoothly deblocked, whilst the re-
sulting thioaminal is hydrolyzed into a free Cys residue
(Scheme 1).
The removal of S-Phacm by immobilized PGA and subse-
quent disulfide-bridge formation have been thoroughly stud-
ied and compared with the commonly used S-Acm-protec-
tion strategy for the synthesis of the urotensin-related pep-
tide (URP).[13] URP is an eight-residue peptide that contains
ꢀ
a disulfide bridge (Cys2 Cys7) and Trp and Tyr residues in
its sequence, which are susceptible to side-reactions. Two
parallel syntheses of URP, by following the Fmoc/tBu strat-
egy with S-Acm- and S-Phacm-protection strategies, were
carried out. Linear peptides were manually synthesized on a
2-chlorotrityl chloride (2-CTC) resin by using DIPCDI and
Oxyma in DMF,[14] with a 5 min pre-activation, for 1 h at
258C to incorporate the Cys residues. These conditions as-
sured the absence of racemization.[15] The other amino acids
were coupled by using COMU and DIEA in DMF for 1 h at
258C.[16]
After peptide elongation on a solid phase, a portion of
the peptidyl resin (1) from the S-Acm strategy was treated
three times with I2 (5 equiv) in DMF at 258C for 15 min and
the crude peptide was then cleaved from the resin. In con-
trast, the peptidyl resin from the S-Phacm strategy was
treated with a mixture of TFA/TIS/water (95:2.5:2.5) for 1 h
Following on from our pioneering work,[10] herein, we ex-
amined the combination of immobilized PGA and the S-
Phacm protecting group for the synthesis of Cys-containing
peptides of diverse lengths and structures. The potential
scope and limitations of this approach for oxidative cycliza-
tion reactions were addressed for peptides with one or two
disulfide bridges with various ring sizes, which involved Tyr
and Trp residues that were sensitive to oxidation. The reuse
of immobilized PGA, as well as the influence of organic sol-
vents, pH, and ionic strength on the biocatalytic reaction,
was also evaluated.
S-Phacm as a green alternative to the conventional S-Acm
group: The S-Acm protecting group can be removed to
either generate the free thiol (mediated by HgII or AgI salts,
Chem. Eur. J. 2012, 18, 16166 – 16176
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16167

J. Tulla-Puche, F. Albericio et al.
Figure 1. Chromatographic profiles of the oxidized peptide URP (3) from
the: a) S-Acm, and b) S-Phacm strategies (phenylacetic acid: N).
Influence of the co-solvent and pH on the biocatalytic reac-
tion: The use of organic media in enzymatic organic synthe-
ses offers several advantages over enzymatic reactions in
aqueous solution. For instance, the solubility of sparingly
water-soluble substrates can be greatly enhanced compared
to that in water. Thus, the use of co-solvents in aqueous
media[17] and the effect of pH were broadly examined for
the removal of S-Phacm and the further oxidation of URP
(Table 1). All of the experiments were carried out by adding
immobilized PGA to a solution of the crude peptide (2) in
the reaction medium and leaving this mixture to stand at
378C. The biocatalytic reactions were monitored by reverse-
phase HPLC (RP-HPLC).
Scheme 2. S-Phacm- and S-Acm-protection strategies for the synthesis of
URP.
The enzymatic activity and stability of immobilized PGA
remained almost intact in the presence of 5% of an extend-
ed variety of organic co-solvents (DMSO, MeCN, DMF, var-
ious alcohols, and Et2O). In some cases, an increase of the
ratio of the co-solvent to up to 20% led to a drop in enzy-
matic activity (MeCN, DMF, EtOH, 2-propanol, and Et2O),
whereas some other co-solvents were well-tolerated by the
biocatalyst (DMSO, MeOH, ethylene glycol, and glycerol).
Notably, fully oxidized URP (3) was obtained when DMSO
was present in the reaction media (Table 1, entries 2–3, 23–
24, and 28), whilst partially oxidized URP was afforded
when either other co-solvents or no co-solvents were used.
Therefore, not only is the use of DMSO as a co-solvent
better tolerated by immobilized PGA than other organic
solvents, but it also promotes the oxidation of thiols to disul-
fide.
at 258C and the resulting S-Phacm-protected peptide (2)
was deprotected by immobilized PGA (3 EU) and subse-
quently oxidized in a mixture of water/DMSO (95:5) at
378C for 24 h (Scheme 2).
The quality of the crude oxidized URP product (3) by
using the S-Phacm-protection strategy was clearly superior
to that prepared by following the S-Acm approach (40%
purity by using the S-Acm strategy versus 90% purity by
using the S-Phacm strategy, Figure 1). As expected, a di-
verse range of side-products were identified by RP-HPLC-
MS (ES) analysis of the crude peptide (3) that was obtained
by using the S-Acm strategy. Presumably, some of the extra
peaks corresponded to the incorporation of two S-Acm
groups onto the Trp and Tyr rings (+142 Da), over-oxida-
tion sub-products (+16 and +32 Da), the incorporation of
two iodine atoms (+252 Da) combined with over-oxidation
(+32 Da), and the formation of intramolecular tryptophan-
2-thioether side-products (see the Supporting Information).
A further advantage of the immobilized enzyme is that it
tolerates a relatively wide range of pH values. Although, the
optimal conditions were at pH 7.9 in 0.1 mm phosphate
buffer/DMSO (95:5; Table 1, entry 23), the use of pH 5.3
also allowed the clean removal of S-Phacm (Table 1,
entry 28). All of these results demonstrate a high S-Phacm-
16168
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176

Synthesis of Cys-Containing Peptides
FULL PAPER
Table 1. Influence of the co-solvent and pH on the biocatalytic removal of S-Phacm and the further oxidation of URP (3).[a]
Entry Reaction medium pH t [h] S-Phacm removal [%] Entry Reaction medium pH t [h] S-Phacm removal [%]
1
2
3
4
5
6
7
8
water
7.0 24
7.0 24
7.0 48
7.0 48
7.0 72
7.0 48
7.0 72
7.0 72
7.0 48
7.0 72
7.0 48
7.0 72
100
100
100
100
10
100
100
10
100
90
100
45
15
16
17
18
19
20
21
22
23
24
25
26
27
28
water/2-propanol (80:20)
water/ethylene-glycol (95:5)
water/ethylene-glycol (80:20)
water/glycerol (95:5)
water/glycerol (80:20)
water/Et2O (95:5)
7.0 72
7.0 48
7.0 72
7.0 24
7.0 40
7.0 48
7.0 72
7.9 16
10
100
100
100
100
100
50
100
100
100
100
5
water/DMSO (95:5)
water/DMSO (80:20)
water/MeCN (95:5)
water/MeCN (80:20)
water/DMF (95:5)
water/DMF (90:10)
water/DMF (80:20)
water/MeOH (95:5)
water/MeOH (80:20)
water/EtOH (95:5)
water/EtOH (80:20)
water/Et2O (80:20)
0.1 mm phosphate buffer
9
0.1 mm phosphate buffer/DMSO (95:5) 7.9
0.1 mm phosphate buffer/DMSO (80:20) 7.9 24
0.1 mm phosphate buffer/MeCN (95:5) 7.9 16
0.1 mm phosphate buffer/MeCN (80:20) 7.9 72
0.1 mm phosphate buffer 5.3 24
0.1 mm phosphate buffer/DMSO (95:5) 5.3 24
8
10
11
12
13
14
water/2-propanol (95:5) 7.0 72
water/2-propanol (90:10) 7.0 72
100
100
100
100
[a] All experiments were performed at a concentration of 8ꢃ10ꢀ5 m of linear S-Phacm-protected URP 2 with 24 mg (3 EU) of SPRIN imibond PGA (hy-
drolytic activity: 130 Ugwetꢀ1) at 378C. The percentage removal of S-Phacm was determined by RP-HPLC analysis.
deblocking efficiency under a wide range of reaction condi-
tions, along with a broad pH-compatible range for the bio-
ACHTUNGTRENNUNGcatalytic removal of S-Phacm by immobilized PGA.
Reuse of the immobilized PGA enzyme: The recycling po-
tential of immobilized PGA was evaluated by reusing the
immobilized biocatalyst for the synthesis of peptide URP
(3) in up to five cycles (for the reaction conditions, see
Table 1, entry 2). At the end of each reaction cycle, phenyl-
acetic acid was removed because it is a strong inhibitor of
PGA. Therefore, the immobilized enzyme was filtered from
the reaction media and then washed three times with a
phosphate buffer (20 mm, pH 8.0, three-times the volume
with respect to the immobilized PGA), followed by a final
washing step with the same reaction medium that was used
for the biocatalytic reaction. After five recycling cycles,
100% enzyme activity was fully recovered. Therefore, each
biocatalytic cycle exhibited the same kinetics and rendered
a similar quality of the crude peptide (Figure 2). These re-
sults showed the possibility of applying immobilized PGA in
a continuous process with the potential for use in commer-
cial production.
Figure 2. Chromatographic profiles of the oxidized peptide URP (3)
after: a) cycle 1, and b) cycle 5 (phenylacetic acid: N).
Regioselective formation of disulfide bonds: The regioselec-
tive construction of disulfide bonds is a challenging and de-
sirable goal for the synthesis of complex Cys-containing pep-
tides. Several suitable combinations of orthogonal and/or
compatible protecting groups have been described for this
purpose.[18] One of the main drawbacks of the synthesis of
multiple-Cys-containing peptides is the disulfide-scrambling
phenomenon, which can be prevented or minimized by
using very mild conditions. The incorporation of S-Phacm,
which can be removed selectively under mild conditions, in-
stalls an additional orthogonal dimension into classical pro-
tection schemes and it has been shown to be appropriate for
the synthesis of complex Cys-containing peptides.
ACHTUGTNNERNUG(Phacm)-Pro-d-Val-CysACHUTNGTRENN(UGN Acm)-NH2 was synthesized manual-
ly and the protected peptide was then incubated for 30 h
with the biocatalyst to obtain the S-Acm-protected open
dimer. The second disulfide bond was formed upon treat-
ment of the S-Acm-protected peptide with I2 (10 equiv) in
AcOH/water (4:1) for 2 h at 258C to afford the target pep-
tide in excellent yield.[10]
Herein, we explored the combination of S-Trt and S-
Phacm protecting groups for the regioselective synthesis of
diverse peptide sequences.
The use of S-Phacm for the synthesis of an oxytocin ana-
logue: First, the orthogonality of S-Phacm to the oxidative
disulfide formation and the removal of S-Phacm in the pres-
ence of a disulfide bond was studied in the synthesis of an
oxytocin analogue (6). Oxytocin is a nine-residue peptide
The compatibility of S-Phacm and S-Acm has been dem-
onstrated in the preparation of the parallel cyclic dimer
(Ac-Cys-Pro-d-Val-Cys-NH2)2. The tetrapeptide Ac-Cys-
Chem. Eur. J. 2012, 18, 16166 – 16176
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16169

J. Tulla-Puche, F. Albericio et al.
ꢀ
that contains a disulfide bridge (Cys1 Cys6) and a Tyr resi-
could be subsequently derivatized (conjugation, labeling,
etc.) for biochemical purposes (Scheme 3).
due in its sequence. An oxytocin analogue with an extra Cys
residue that was linked to the N terminus through a b-ala-
nine residue was prepared by combining S-Trt for the native
Cys residues and S-Phacm for the extra Cys residue. After
deblocking, the thiol group of the additional Cys residue
Linear peptide 4 was manually synthesized on a Rink-
Amide AM resin by following the same procedure as that
described for the synthesis of linear URP. After the assem-
bly of the peptide chain on the solid phase, the peptide was
cleaved from the resin by treatment with TFA, with the con-
comitant removal of the S-Trt protecting groups, and the for-
mation of the disulfide bridge was attempted. Linear oxyto-
cin analogue 4 was not totally soluble in water or in any
aqueous buffer; consequently, a co-solvent was required to
carry out the oxidation. The use of DMF or DMSO (5-
20%) was appropriate for both the solubilization and oxida-
tion of analogue 4, whereas other co-solvents, such as
MeCN, 2-propanol, and MeOH, did not promote its solubili-
ty and, hence, its further cyclization. Once the formation of
Figure 3. Chromatographic profiles of: a) oxidized S-Phacm-protected
oxytocin analogue 5, and b) after the removal of S-Phacm (Table 2,
entry 9; phenylacetic acid: N).
Scheme 3. Synthesis of an oxytocin analogue that combines S-Phacm and
2 S-Trt groups.
Table 2. Disulfide-bond formation and removal of S-Phacm in oxytocin analogue (6).[a]
Entry
Disulfide-bond formation
Reaction medium
Removal of S-Phacm
pH
t [h]
Reaction medium
pH
t [h]
Observations
1
2
3
4
5
6
7
8
water/DMSO (90:10)
water/DMF (90:10)
water/DMSO (80:20)
water/DMSO (95:5)
0.1 mm phosphate buffer/DMSO (90:10)
0.1 mm phosphate buffer/DMF (90:10)
0.1 mm phosphate buffer/DMSO (90:10)
0.1 mm phosphate buffer/DMF (90:10)
water/DMSO (90:10)
water/DMSO (90:10)
water/DMSO (90:10)
water/DMSO (99:1)
water/DMSO (97:3)
7.0
7.0
7.0
7.0
7.0
7.0
5.2
5.2
7.0
7.0
7.0
7.0
7.0
16
48
16
16
24
72
72
120
16
16
16
24
24
water/DMSO (90:10)
water/DMF (90:10)
water/DMSO (80:20)
water/DMSO (95:5)
0.1 mm phosphate buffer/DMSO (90:10)
0.1 mm phosphate buffer/DMF (90:10)
0.1 mm phosphate buffer/DMSO (90:10)
0.1 mm phosphate buffer/DMF (90:10)
water
0.1 mm phosphate buffer
0.1 mm phosphate buffer
water/DMSO (99:1)
water/DMSO (97:3)
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
5.4
7.8
7.0
7.0
5
4
5
scrambling
some scrambling
scrambling
scrambling
scrambling
some scrambling
scrambling
some scrambling
–
5
24
24
24
24
4
4
4
4
4
9
10
11
12
13
–
–
some scrambling
scrambling
[a] All experiments were performed at a concentration of 8ꢃ10ꢀ5 m of linear S-Phacm-protected oxytocin analogue 4 with 24 mg (3 EU) of SPRIN imi-
bond PGA (hydrolytic activity: 130 Ugwetꢀ1) at 378C. The progress of the reactions, along with disulfide scrambling, were monitored by RP-HPLC analy-
sis.
16170
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176

Synthesis of Cys-Containing Peptides
FULL PAPER
the disulfide had been achieved and without isolation of the
S-Phacm-protected cyclic peptide (5), immobilized PGA was
added to remove the S-Phacm group. However, when DMF
or DMSO was used for S-Phacm deblocking, disulfide
scrambling occurred. As expected, this phenomenon was
more significant in the presence of DMSO than in DMF
(Table 2, entries 1, 3–5, and 7 vs. 2, 6, and 8). At this point,
a new strategy, which consisted of carrying out the oxidative
disulfide formation in the presence of DMSO (10%) in
aqueous media, followed by removal of the DMSO by lyo-
philization, was assayed. The enzymatic cleavage of S-
Phacm was then accomplished in water or in 0.1 mm phos-
phate buffer, thereby minimizing disulfide scrambling
(Table 2, entries 9–11; Figure 3).
TFA/TIS/water (95:2.5:2.5) for 1.5 h at 258C to afford the
linear S-Phacm-protected peptide (7, Scheme 4). The first
disulfide-bond formation of this peptide (Scheme 4a) was
achieved in either 1 mm or 20 mm phosphate buffer with or
Table 3. The use of S-Phacm for the regioselective construction of pep-
tide RGD-4C (9).[a]
Reaction medium
pH
tA [h]
tB [h]
1
2
3
4
5
6
7
8
1 mm phosphate buffer
8.0
8.0
8.0
6.9
6.9
6.9
8.0
8.0
8.0
6.9
6.9
6.9
240
48
24
360
72
24
48
24
24
240
48
16
16
16
16
16
16
16
16
16
16
16
16
1 mm phosphate buffer/DMSO (95:5)
1 mm phosphate buffer/DMSO (80:20)
1 mm phosphate buffer
1 mm phosphate buffer/DMSO (95:5)
1 mm phosphate buffer/DMSO (80:20)
20 mm phosphate buffer
20 mm phosphate buffer/DMSO (95:5)
20 mm phosphate buffer/DMSO (80:20)
20 mm phosphate buffer
20 mm phosphate buffer/DMSO (95:5)
20 mm phosphate buffer/DMSO (80:20)
9
10
11
12
The use of S-Phacm for the regioselective synthesis of pep-
tide RGD-4C: The double-cyclic peptide RGD-4C is an 11-
24
ꢀ
residue peptide that contains two disulfide bridges (Cys2
[19]
[a] All experiments were performed at a concentration of 8ꢃ10ꢀ5 m of
cyclic S-Phacm-protected RGD-4C peptide (8) with 24 mg (3 EU) of
SPRIN imibond PGA (hydrolytic activity: 130 Ugwetꢀ1) at 378C. The
progress of the reactions was monitored by RP-HPLC analysis (tA =reac-
tion time for the formation of the first disulfide bond; tB =reaction time
for the deblocking of S-Phacm and successive oxidation).
ꢀ
Cys10 and Cys4 Cys8). The linear peptide was manually
synthesized on 2-CTC resin by using DIPCDI and Oxyma in
DMF, with a 5 min pre-activation, for 1 h at 258C to incor-
porate all of the Fmoc-amino acids. After peptide elonga-
tion on the solid phase, the peptidyl resin was treated with
Figure 4. Chromatographic profiles of: a) cyclic S-Phacm-protected
RGD-4C peptide (8), b) final bicyclic peptide 9 from the biocatalytic re-
moval of S-Phacm and oxidation (Table 3, entry 1), and c) final bicylic
peptide (9) after the removal of S-Phacm and oxidation by treatment
with I2 (phenylacetic acid: N).
Scheme 4. Synthetic approach for the preparation of peptide RGD-4C
(9) through a combination of S-Phacm and S-Trt protecting groups.
Chem. Eur. J. 2012, 18, 16166 – 16176
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16171

J. Tulla-Puche, F. Albericio et al.
without DMSO as a co-solvent (Table 3, tA); complete oxi-
dation was faster when 20% DMSO was present in the reac-
tion medium (Table 3, entries 3, 6, 9, and 12). The first disul-
fide-bridge construction was favored at pH 8 in 20 mm phos-
phate buffer (Table 3, entry 7), whereas the neat oxidation
took much longer when performed at pH 6.9 in 1 mm phos-
phate buffer (Table 3, entry 4). The disulfide formation was
monitored by RP-HPLC analysis until the reaction was
completed (Figure 4a). Then, and without isolation of
after the formation of the first disulfide bond, the oxidized
cyclic S-Phacm-protected RGD-4C peptide (8) was dis-
solved in water/MeCN (1:1) and I2 (5 equiv) was added to
the reaction medium. The progress of the I2-mediated oxida-
tion reaction was monitored by RP-HPLC and the final oxi-
dized peptide (9) was obtained in low quality (Figure 4c).
The use of S-Phacm for the synthesis of peptide T22: T22
([Tyr5,12,Lys7]-polyphemusin II) is an 18-residue peptide
ꢀ
ꢀ
mono
cyclic peptide 8, immobilized PGA was added to
that contains two disulfide bridges (Cys4 Cys17 and Cys8
Cys13), four Tyr residues, and a Trp residue in its se-
quence.[20] The S-Phacm protecting group was applied for
the preparation of the peptide T22 by means of two synthet-
ic approaches that were carried out in parallel: 1) a regiose-
lective strategy that combined S-Trt and S-Phacm groups
(Scheme 5a, b) and 2) a random strategy that used S-Phacm
as a unique protecting group for Cys residues (Scheme 5c).
Linear peptides 10 and 11 were manually synthesized on
an AM ChemMatrix resin by following the same procedure
as that for the synthesis of linear peptide RGD-4C. After
the assembly of the peptides on the solid phase, the peptidyl
remove S-Phacm, with the concomitant formation of the
second disulfide bond (Scheme 4b). S-Phacm deblocking
and successive oxidation were achieved regardless of the re-
action media (Table 3). In all cases, the biocatalytic reaction
was complete after 16 h (Table 3, tB) and the final bicyclic
peptide RGD-4C (9) was obtained in high purity (Fig-
ure 4b).
In addition, the removal of S-Phacm and successive oxida-
tion to the disulfide was attempted by treatment with I2 to
compare chemical removal and oxidation with biocatalytic
deprotection and successive disulfide-bond formation. Thus,
Scheme 5. Synthetic approaches for the synthesis of peptide T22: regioselective strategy that combines the S-Phacm and S-Trt protecting groups (peptide
10, steps a and b) and a random strategy that uses S-Phacm for the Cys residues (peptide 11, step c).
16172
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176

Synthesis of Cys-Containing Peptides
FULL PAPER
resins were treated with TFA/TIS/water (95:2.5:2.5) for 3 h
at 258C to assure the complete deprotection of the five Pbf
(2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl)
tecting groups for the Arg residues.
pro-
Whilst the first disulfide bond from the protected S-
Phacm-protected T22 peptide (10, Scheme 5a) was easily
and rapidly obtained in water or in low-salt-concentration
buffers (0.1 mm and 1 mm), neither total nor partial S-
Phacm deblocking was observed under these reaction condi-
tions when immobilized PGA was added directly (Table 4,
entries 1–4). It is worth noting that increasing the salt con-
centration enabled the removal of the S-Phacm group,
thereby demonstrating the main dependence of the biocata-
lytic reaction on the ionic strength of the medium for this
peptide sequence (Table 4). An increase in ionic strength
has been reported to promote enzyme–substrate interac-
tions[21] and, in the case of PGA-catalyzed hydrolysis, an in-
crease in ionic strength favors the displacement of PhAcOH
from the active site, thus achieving higher yields. In addi-
tion, increasing the ionic strength guarantees buffered pH
control, thereby preventing the shift to a lower pH, which
would cause a decrease in activity and yield.
Table 4. The use of S-Phacm for the regioselective construction of pep-
tide T22 (13).[a]
Reaction medium
pH tA [h] tB [h] Removal of S-Phacm
1
2
3
4
5
6
7
8
9
water
water/DMSO (95:5)
0.1 mm phosphate buffer 7.4 24
1 mm phosphate buffer
20 mm phosphate buffer
20 mm phosphate buffer
20 mm phosphate buffer
20 mm phosphate buffer
50 mm phosphate buffer
7.0 48
7.0 24
72
72
72
72
72
72
72
72
48
48
72
48
48
no
no
no
no
7.7 24
9.0 72
8.0 24
7.0 24
6.0 48
8.1 24
partial
Figure 5. Chromatographic profiles of: a) final bicyclic T22 peptide (13)
from the regioselective strategy (Table 4, entry 10), b) final bicyclic T22
peptide (13) from the random strategy (Table 4, entry 10), and c) final bi-
cyclic T22 peptide (13) after the removal of S-Phacm and concomitant
oxidation of the cyclic S-Phacm-protected T22 peptide (12) by treatment
with I2 (phenylacetic acid: N).
almost complete
almost complete
partial
complete
complete
partial
10 100 mm phosphate buffer 8.1 24
11 20 mm NH4HCO3
12 50 mm NH4HCO3
13 100 mm NH4HCO3
8.2 24
8.1 24
8.1 24
complete
complete
Moreover, as for peptide RGD-4C, the removal of S-
Phacm and the concomitant oxidation of the cyclic T22 in-
termediate (12) by treatment with I2 were assayed for the
regioselective strategy. Therefore, after oxidization, the
cyclic S-Phacm-protected T22 peptide (12) was dissolved in
water/MeCN (1:1) and I2 (5 equiv) was added to the reac-
tion medium. The progress of the reaction was monitored
by RP-HPLC, which showed a complex chromatographic
profile after 30 min (Figure 5c).
For the random obtaining of peptide T22 (13, Scheme 5c),
the S-Phacm-protected linear T22 peptide (11) was treated
with immobilized PGA (6 EU) and incubated at 378C in the
same reaction media (Table 4). Similar behavior was ob-
served in response to salt-buffer concentration and pH.
Hence, at higher salt concentrations, S-Phacm deblocking
and successive random oxidation was accomplished and the
final crude product (13) was obtained in high quality (Fig-
ure 5b).
[a] All experiments were performed at a peptide concentration of 8ꢃ
10ꢀ5 m cyclic S-Phacm-protected T22 peptide (12) with 24 mg (3 EU) of
SPRIN imibond PGA (hydrolytic activity: 130 Ugwetꢀ1) at 378C. The
progress of the reactions was monitoring by RP-HPLC analysis (tA =re-
action time for the formation of the first disulfide bond; tB =reaction
time for the deblocking of S-Phacm and successive oxidation).
Therefore, by increasing the salt concentration up to
20 mm, the removal of the S-Phacm group by the immobi-
lized PGA was observed. The pH of the reaction media was
determinant for the progress of the biocatalytic reaction
with this peptide sequence (Table 4, entries 5–8). Thus, al-
though the S-Phacm groups were not completely deblocked
at pH 7 or 8 after 72 h, the mere partial removal of the pro-
tecting group was observed after the same reaction time at
pH 6 or 9.
At salt concentrations of 50 mm and 100 mm, the removal
of the S-Phacm group by the biocatalyst was faster and
clean (Table 4, entries 9, 10 and 12, 13) and the oxidized
crude peptide T22 (13) was superior in quality to those that
were rendered at lower salt concentrations (Figure 5a).
The use of S-Phacm group for the synthesis of peptides
through native chemical ligation (NCL): Based on the ca-
pacity of immobilized PGA to catalyze the removal of S-
Chem. Eur. J. 2012, 18, 16166 – 16176
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16173

J. Tulla-Puche, F. Albericio et al.
Phacm under a broad range of conditions, we were encour-
aged to exploit the use of the S-Phacm protecting group in
the presence of sensitive functional groups, such as thioest-
ers. Therefore, the enzymatic deblocking of S-Phacm in the
presence of a thioester moiety was applied to the synthesis
of a small cyclic peptide by using intramolecular NCL. Thus,
linear C-terminal thioester peptide 15 was prepared by
using S-Phacm as a protecting group for the b-thiol function
of the N-terminal Cys residue in a side-chain-anchoring ap-
proach[22] where an Fmoc-Asp-OAllyl residue was incorpo-
rated onto an Fmoc-Rink-amide AM polystyrene resin and
further chain elongation was carried out by using a standard
Fmoc strategy with Oxyma and DIPCDI in DMF
(Scheme 6a–d). After peptide elongation on the solid phase,
the N terminus of the linear peptide was protected with a
Boc group before the removal of the allyl protecting group
by Pd0 in the presence of PhSiH3. Subsequently, attachment
of previously prepared H-Gly-SACTHNUTRGNE(UNG CH2)2CO2Et (14) was ac-
complished by using HATU and DIEA as a coupling
system. Then, the required linear peptide thioester (15) was
cleaved from the resin by treatment with TFA/TIS/water
(95:2.5:2.5) with the concomitant removal of the N-terminal
Boc group (Scheme 6e–h).
Figure 6. Chromatographic profiles of: a) linear C-terminal thioester pep-
tide 15, b) final cyclic peptide 16 from the enzymatic removal of S-
Phacm and concomitant cyclization through an NCL reaction in water at
pH 7 (phenylacetic acid: N).
The formation of linear thioester hexapeptide 15 was de-
termined by analytical RP-HPLC analysis (Figure 6a).
Then, the removal of S-Phacm and successive head-to-tail
cyclization by NCL was attempted under different deblock-
ing conditions to obtain cyclic peptide 16 (Scheme 6i). In all
cases, the crude peptide was incubated with the immobilized
PGA at 378C and enzymatic deprotection and spontaneous
cyclization were monitored by RP-HPLC until the reaction
was completed. As expected, a clean chromatographic pro-
file was obtained in water or in potassium phosphate buffer
(1 mm and 20 mm) at pH 5, 6, and 7, whilst a complex chro-
matographic profile was observed when the biocatalytic de-
protection was performed at pH 8. RP-HPLC analysis of the
final crude product at pH 7 in water showed a major peak
that corresponded to the desired cyclic peptide (16, Fig-
ure 6b).
Scheme 6. Synthesis of linear thioester-peptide intermediate 15 that fea-
tures side-chain-anchoring to a Rink-amide AM polystyrene resin and ul-
terior enzymatic removal of S-Phacm with concomitant cyclization by
using NCL to obtain the final cyclic peptide (16): a) piperidine/DMF
(1:4); b) Fmoc-Asp-OAllyl (3 equiv), DIPCDI (3 equiv), Oxyma
(3 equiv) in DMF for 1 h at 258C; c) standard peptide-chain elongation
by using Fmoc chemistry; d) piperidine/DMF (1:4); e) (Boc)2O
(10 equiv), DIEA (10 equiv) in DMF for 1 h at 258C; f) [Pd
(0.1 equiv), PhSiH3 (10 equiv) in CH2Cl2 (3ꢃ15 min); g) H-Gly-S-
(CH2)2CO2Et (14, 10 equiv), HATU (10 equiv), DIEA (20 equiv) in
DMF for 30 min at 258C; h) TFA/TIS/water (95:2.5:2.5) for 1 h at 258C;
i) deblocking of S-Phacm was performed at a concentration of 8ꢃ10ꢀ5
ACHTUNGTRNE(NUNG PPh3)4]
ACHTUNGTRENNUNG
m
of S-Phacm-protected thioester-peptide 15 and 24 mg (3 EU) of SPRIN
imibond PGA (hydrolytic activity: 130 Ugwetꢀ1) for 5 h at 378C. Fmoc=
9-fluorenylmethoxycarbonyl, Boc=tert-butoxycarbonyl, DIPCDI=N,N’-
diisopropylcarbodiimide, Oxyma=ethyl 2-cyano-2-(hydroxyimino)ace-
tate, DIEA=N,N-diisopropylethylamine, HATU=2-(7-aza-1H-benzo-
triazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate, TFA=tri-
fluoroacetic acid, TIS=triisopropylsilane.
Conclusion
S-Phacm has been proven to be a useful alternative to
chemically removable protecting groups such as S-Acm. Its
16174
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176

Synthesis of Cys-Containing Peptides
FULL PAPER
piperidine/DMF (1:4, v/v; 1ꢃ1 min, 2ꢃ5 min). Washing between the de-
protection, coupling, and final deprotection steps were carried out with
DMF (5ꢃ1 min) and CH2Cl2 (5ꢃ1 min). Peptide-synthesis transforma-
tions and washes were performed at 258C.
mild biocatalytic removal by an immobilized PGA enzyme
avoids the concourse of harsh reagents, such as toxic heavy
metals and I2, which favor back-alkylation and other modifi-
cations of the peptide chain and whose residual products are
difficult to remove and are potentially harmful to the envi-
ronment.
The biocatalytic removal of S-Phacm by immobilized
PGA, which can be easily separated from the reaction
media by a simple and rapid filtration, exhibits the same ki-
netics and renders the crude peptide with a similar quality
after repeated reuse of the immobilized enzyme. The immo-
bilized biocatalyst demonstrates complete recovery of enzy-
matic activity after each cycle, which opens up the possibili-
ty of its implementation in a continuous process.
The removal of S-Phacm is sufficiently mild to prevent di-
sulfide scrambling. Therefore, it enables the deblocking of
additional Cys residues in peptide analogues for their subse-
quent derivatization and can be applied in the native chemi-
cal ligation strategy. The deblocking of S-Phacm in the pres-
ence of sensitive functional groups, such as thioesters, was
applied successfully to the synthesis of a small cyclic peptide
by using NCL.
The combination of S-Trt and S-Phacm has been success-
fully applied to the regioselective synthesis of peptides
RGD-4C and T22, whilst the use of S-Phacm as a unique
protecting group has shown notable results for the random
synthesis of the bicyclic peptide T22.
One of the main characteristics of the removal of S-
Phacm by immobilized PGA is that it depends on the pep-
tide sequence. Thus, some sequences tolerate a broad range
of conditions (pH, ionic strength, and the presence of organ-
ic co-solvents), whereas others require the optimal fine-
tuning of the conditions. Thus, we propose the artisanal
preparation of multi-Cys-containing peptides, as opposed to
conventional methods that apply a general approach to all
peptides and, consequently, render unsatisfactory results. In
our opinion, the capacity of the immobilized PGA to work
under a broad range of conditions is a major advantage and
makes the S-Phacm protecting group a suitable tool for use
in peptide chemistry.
An XBridge BEH130 C18 RP-HPLC analytical column (4.6 mmꢃ
100 mm, 3.5 mm) was obtained from Waters (Ireland). Analytical RP-
HPLC was performed on a Waters instrument that comprised a separa-
tion module (Waters 2695), an automatic injector, a photodiode-array de-
tector (Waters 2998), and a system controller (Empower login). UV de-
tection was performed at 220 and 254 nm and linear gradients of MeCN
(+0.036% TFA) into water (+0.045% TFA) were performed at a flow
rate of 1.0 mLminꢀ1 over 8 min. RP-HPLC-MS (ES) was performed on a
Waters Micromass ZQ spectrometer with a SunFireTM C18 RP-HPLC an-
alytical column (2.1 mmꢃ100 mm, 5 mm). Linear gradients of MeCN
(+0.07% formic acid) into water (0.1% formic acid) were performed at
a flow rate of 0.3 mLminꢀ1 over 8 min.
Immobilized PGA: The immobilized PGA was stored in a mixture of
20 mm phosphate buffer/glycerol (20:80) at 48C. Before use, the immobi-
lized PGA (1 g) was washed with 20 mm potassium phosphate buffer
(pH 8, 5ꢃ4 mL), and the glycerol-free immobilized PGA was stored at
48C for two months.
Removal of S-Acm and the oxidation of URP on a solid phase (3): A
portion of peptidyl resin 1 (10 mg) was treated with iodine (5 equiv,
6 mg) in DMF (500 mL) at 258C for 15 min and the treatment was repeat-
ed two more times. Then, the resin was washed with DMF (5ꢃ1 mLꢃ
1 min), piperidine/DMF (1:4) (5ꢃ1 mLꢃ1 min), and CH2Cl2 to remove
any excess iodine from the resin. Next, the resin was treated with a mix-
ture of TFA/TIS/water (95:2.5:2.5, 2 mL) for 1 h at 258C to render cyclic
URP 3. The oxidized peptide was obtained in 40% purity, as determined
by analytical RP-HPLC (linear gradient from 20% to 60% MeCN over
8 min; tR =4.6 min). RP-HPLC-MS (ES) showed the formation of the
target peptide (linear gradient from 0% to 50% MeCN over 8 min; tR =
7.8 min): m/z calcd for C49H64N10O10S2: 1017.2; found: 1018.3 [M+H]+,
509.7 [(M+2H)/2]2+ (M is the MW of the oxidized peptide URP 3).
Removal of S-Phacm and the oxidation of URP in solution (3): Linear S-
Phacm-protected URP 2 (5 mg) was dissolved in a mixture of water/
DMSO (95:5, 45 mL, 8ꢃ10ꢀ5 m), immobilized PGA (120 mg, 15 EU) was
added, and the reaction was left to stand for 24 h at 378C and 50ꢃ
10 rpm. Next, the immobilized biocatalyst was removed by filtration from
the media and the aqueous mixture was lyophilized. Then, the crude pep-
tide was precipitated with cold Et2O (10 mL) and centrifuged 3 times to
render the cyclic peptide URP (3) in 95% purity, as determined by ana-
lytical RP-HPLC (linear gradient from 20% to 60% MeCN over 8 min;
tR =4.6 min). RP-HPLC-MS (ES) showed the formation of the target
peptide (linear gradient from 0% to 50% MeCN over 8 min; tR =
7.8 min): m/z calcd for C49H64N10O10S2: 1017.2; found: 1018.2 [M+H]+,
509.8 [(M+2H)/2]2+ (M is the MW of the oxidized peptide URP 3).
Oxidation of the oxytocin analogue and the removal of S-Phacm (6):
(Table 2, entry 9) Crude peptide 4 (1 mg) was dissolved in a mixture of
water/DMSO (90:10, 9 mL, 8ꢃ10ꢀ5 m) at 258C and the disulfide-bond for-
mation was monitoring by RP-HPLC analysis (linear gradient from 20%
to 60% MeCN over 8 min; tR =4.3). After the oxidation was completed,
RP-HPLC-MS (ES) analysis showed the formation of the target peptide
(linear gradient from 10% to 60% MeCN over 8 min; tR =7.7 min): m/z
calcd for C60H87N15O16S3: 1370.6; found: 1370.9 [M+H]+, 686.2 [(M+2H)/
2]2+ (M is the MW of the oxidized S-Phacm-protected oxytocin analogue
5). Next, the mixture was lyophilized to completely remove the DMSO
Experimental Section
Fmoc-amino-acid derivatives, Fmoc-Rink-OH linker, 2-CTC resin, and
Fmoc-Rink-amide polystyrene resin were obtained from IRIS Biotech
(Marktredwitz, Germany). DIEA and DIPCDI were obtained from Al-
drich (Milwaukee, WI), TFA was obtained from Scharlau (Barcelona,
Spain), Oxyma was obtained from Luxembourg Industries Ltd. (Tel Aviv,
Israel), and COMU, KH2PO4, and K2HPO4·3H2O were obtained from
Sigma–Aldrich (St Louis). PGA from E. coli (E.C.3.5.1.1) that had been
from the reaction media and the crude peptide (5) was redissolved in
ꢀ1
water (9 mL, 8ꢃ10ꢀ5 m). Immobilized PGA (24 mg, 3 EU, 130 Ugwet
)
covalently immobilized on an amino-acrylic resin (0.15–0.30 mm (96%),
was added to the reaction mixture, which was left to stand at 378C for
4 h to afford the thiol-free oxytocin analogue (6). The completion of the
reaction was determined by RP-HPLC analysis (linear gradient from
20% to 60% MeCN over 8 min; tR =3.2 min). RP-HPLC-MS (ES) analy-
sis showed the formation of the target peptide (linear gradient from 10%
to 60% MeCN over 8 min; tR =6.9 min): m/z calcd for C51H78N14O15S3:
1223.4; found: 1223.6 [M+H]+, 612.6 [(M+2H)/2]2+ (M is the MW of oxy-
tocin analogue 6).
ꢀ1
130 Ugwet
) was obtained from SPRIN technologies. DMF, CH2Cl2,
Et2O, DMSO, piperidine, and MeCN (HPLC grade) were purchased
from SDS (Peypin, France). All commercial reagents and solvents were
used as received.
Solid-phase syntheses were carried out manually in polypropylene syring-
es that were fitted with a porous polyethylene disc. Solvents and soluble
reagents were removed by suction. The Fmoc group was removed with
Chem. Eur. J. 2012, 18, 16166 – 16176
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
16175

J. Tulla-Puche, F. Albericio et al.
Regioselective disulfide-bond formation of peptide RGD-4C (9): Crude
peptide 7 (1 mg) was dissolved in the reaction medium (9 mL, 8ꢃ10ꢀ5 m;
Table 3) at 258C and disulfide-bond formation was monitoring by RP-
HPLC analysis (linear gradient from 15% to 40% MeCN over 8 min;
tR =6.9). After the oxidation was completed, RP-HPLC-MS (ES) analysis
showed the formation of the target peptide (linear gradient from 0% to
70% MeCN over 8 min; tR =6.7 min): m/z calcd for C60H80N16O18S4:
1441.6 [M+H]+; found: 1442.9, 722.2 [(M+2H)/2]2+ (M is the MW of the
oxidized S-Phacm-protected RGD-4C peptide 8). Next, the immobilized
PGA enzyme (24 mg, 3 EU, 130 Ugwetꢀ1) was added to the reaction mix-
ture, which was left to stand at 378C for 16 h to afford the bicyclic RGD-
4C peptide (9). Completion of the reaction was determined by RP-HPLC
analysis (linear gradient from 5% to 50% MeCN over 8 min; tR =
4.1 min). RP-HPLC-MS (ES) analysis showed the formation of the target
peptide (linear gradient from 5% to 100% MeCN over 8 min; tR =
[1] A. Isidro-Llobet, M. ꢄlvarez, F. Albericio, Chem. Rev. 2009, 109,
2455–2504.
[2] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis,
3rd ed., Wiley, New York, 1999.
[3] G. Barany, F. Albericio, J. Am. Chem. Soc. 1985, 107, 4936–4942.
[4] a) T. Pathak, H. Waldmann, Curr. Opin. Chem. Biol. 1998, 2, 112–
120; b) T. Pathak, H. Waldmann, in Stereoselective Biocatalysis,
(Ed.: R. N. Patel, M. Dekker), CRC, New York, 2000, pp. 775–797;
c) D. Kadereit, H. Waldmann, Chem. Rev. 2001, 101, 3367–3396.
[5] G. T. Hermanson, Bioconjugate Techniques, 2nd ed., Academic
Press, 2008.
[6] a) M. Schnolzer, S. B. H. Kent, Science 1992, 256, 221–25; b) P. E.
Dawson, T. W. Muir, I. Clark-Lewis, S. B. H. Kent, Science 1994,
266, 776–79.
4.6 min): m/z calcd for C42H60N14O16S4: 1145.3; found: 1146.3ACTHNUTRGNEUNG
[M+H]+,
[7] J. Brask, in The Power of Functional Resins in Organic Synthesis
(Eds.: J. Tulla-Puche, F. Albercicio), Wiley-VCH, Weinheim, 2008,
pp. 365–380.
573.9 [(M+2H)/2]2+ (M is the MW of the oxidized bicyclic RGD-4C pep-
tide 9).
[8] a) L. Cao, Carrier-bound Immobilized Enzymes: Principles, Applica-
tion and Design, Wiley-VCH, Weinheim, 2005; b) R. Wohlgemuth,
Curr. Opin. Biotechnol. 2010, 21, 713–724.
[9] a) A. Basso, P. Braiuca, S. Cantone, C. Ebert, P. Linda, P. Spizzo, P.
Caimi, U. Hanefeld, G. Degrassi, L. Gardossi, Adv. Synth. Catal.
2007, 349, 877–886; b) A. Basso, L. De Martin, C. Ebert, L. Gardos-
si, P. Linda, J. Mol. Catal. B Enzym. 2001, 16, 73–80.
[10] M. Royo, J. Alsina, E. Giralt, U. Slomcyznska, F. Albericio, J. Chem.
Soc. Perkin Trans. 1 1995, 1095–1102.
[11] a) K. Adermann, K. Barlos, in Oxidative Folding of Peptides and
Proteins, (Eds.: J. Buchner, L. Moroder), RSC, 2009, pp. 297–317;
b) B. Kamber, A. Hartmann, K. Eisler, B. Riniker, H. Rink, P.
Sieber, W. Rittel, Helv. Chim. Acta 1980, 63, 899–915; c) C. S.
Samuel, F. Lin, M. A. Hossain, C. Zhao, T. Ferraro, R. A. D. Bath-
gate, G. W. Tregear, J. D. Wade, Biochemistry 2007, 46, 5374–5381.
[12] a) H. Lamthanh, C. Roumestand, C. Deprun, A. Menez, Int. J. Pept.
Prot. Res. 1993, 41, 85–95; b) H. Lamthanh, H. Virelizier, D. Frays-
sinhes, Pept. Res. 1995, 8, 316; c) M. Engebretsen, E. Agner, J. San-
dosham, P. M. Fischer, J. Pept. Res. 1997, 49, 341–346; d) N. M.
Alexander, J. Biol. Chem. 1974, 249, 1946–1952; e) P. Sieber, B.
Kamber, B. Riniker, W. Rittel, Helv. Chim. Acta 1980, 63, 2358–
2363.
[13] D. Chatenet, C. Dubessy, J. Leprince, C. Boularan, L. Carlie, I. Sꢅga-
las-Milazzo, L. Guilhaudis, H. Oulyadi, D. Davoust, E. Scalbert, B.
Pfeiffer, P. Renard, M. C. Tonon, I. Lihrmann, P. Pacaud, H. Vaudry,
Peptides 2004, 25, 1819–1830.
[14] R. Subirꢀs-Funosas, R. Prohens, R. Barbas, A. El-Faham, F. Alberi-
cio, Chem. Eur. J. 2009, 15, 9394–9403.
[15] a) Y. Han, F. Albericio, G. Barany, J. Org. Chem. 1997, 62, 4307–
4312; b) Y. M. Angell, J. Alsina, F. Albericio, G. Barany, J. Peptide
Res. 2002, 60, 292–299.
[16] A. El-Faham, R. Subirꢀs-Funosas, R. Prohens, F. Albericio, Chem.
Eur. J. 2009, 15, 9404–9416.
Regioselective disulfide-bond formation of peptide T22 (13): Crude pep-
tide 10 (1 mg) was dissolved in the reaction medium (10 mL, 8ꢃ10ꢀ5 m,
Table 4) at 258C and disulfide-bond formation was monitored by RP-
HPLC (linear gradient from 10% to 60% MeCN over 8 min; tR =
4.2 min). After the oxidation was completed, RP-HPLC-MS (ES) analy-
sis showed the formation of the target peptide (linear gradient from 5%
to 100% MeCN over 8 min; tR =4.7 min): m/z calcd for C127H184N40O24S4:
2783.3; found: 929.5 [(M+3H)/3]3+, 697.2 [(M+4H)/4]4+ (M is the MW of
the oxidized S-Phacm-protected T22 peptide 12). Next, the immobilized
PGA enzyme (24 mg, 3 EU, 130 Ugwetꢀ1) was added to the reaction mix-
ture, which was left to stand at 378C to afford the fully oxidized T22 pep-
tide (13). Completion of the reaction was determined by RP-HPLC anal-
ysis (linear gradient from 10% to 40% MeCN over 8 min; tR =3.9 min).
RP-HPLC-MS (ES) analysis showed the formation of the target peptide
(linear gradient from 0% to 30% MeCN over 8 min; tR =5.7 min): m/z
calcd for C109H164N38O22S4: 2487.0; found: 1245.4 [(M+2H)/2]2+, 830.6
[(M+3H)/3]3+, 623.2 [(M+4H)/4]4+ (M is the MW of the oxidized bicyclic
T22 peptide 13).
Random disulfide-bond formation of peptide T22 (13): Crude peptide 11
(1 mg) was dissolved in the reaction medium (12 mL, 8ꢃ10ꢀ5 m, Table 4)
and the immobilized PGA enzyme (48 mg, 6 EU, 130 Ugwetꢀ1) was added
to the reaction mixture, which was left to stand at 378C to afford the
fully oxidized T22 peptide (13). Completion of the reaction was deter-
mined by RP-HPLC analysis (linear gradient from 10% to 40% MeCN
over 8 min; tR =3.9 min). RP-HPLC-MS (ES) analysis showed the forma-
tion of the target peptide (linear gradient from 0% to 30% MeCN over
8 min; tR =5.9 min): m/z calcd for C109H164N38O22S4: 2487.0; found: 1245.2
[(M+2H)/2]2+, 830.6 [(M+3H)/3]3+, 623.2 [(M+4H)/4]4+ (M is the MW of
the oxidized bicyclic T22 peptide 13).
Removal of S-Phacm and cyclization through NCL (16): Crude peptide
15 (1 mg) was dissolved in the reaction medium (14 mL, 8ꢃ10ꢀ5 m), the
immobilized PGA enzyme (24 mg, 3 EU, 130 Ugwetꢀ1) was added, and
the reaction mixture was left to stand at 378C for 5 h to afford the head-
to-tail cyclic peptide (16). Completion of the reaction was determined by
analytical RP-HPLC (linear gradient from 25% to 50% MeCN over
8 min; tR =3.3 min). RP-HPLC-MS (ES) showed the formation of the
target peptide (linear gradient from 20% to 70% MeCN over 8 min; tR =
6.4 min): m/z calcd for C26H37N7O7S: 591.68; found: 592.9 [M+H]+ (M is
the MW of the cyclic peptide 16).
[17] a) M. G. Kim, S. B. Lee, J. Mol. Catal. B Enzym. 1996, 181–190;
b) M. G. Kim, S. B. Lee, J. Mol. Catal. B Enzym. 1996, 201–211.
[18] a) D. Andreu, F. Albericio, N. A. Sole, M. C. Munson, M. Ferrer, G.
Barany, in Peptide Synthesis Protocols (Eds.: M. W. Pennington,
B. M. Dunn), Humana, New York, 1994, pp. 91–169.
[19] N. Assa-Munt, X. Jia, P. Laakkonen, E. Ruoslahti, Biochemistry
2001, 40, 2373–2378.
[20] H. Nakashima, M. Masuda, T. Murakami, Y. Koyanagi, A. Matsu-
moto, N. Fujii, N. Yamamoto, Antimicrob. Agents Chemother. 1992,
36, 1249–1255.
[21] K. Kuttiyawong, S. Nakapong, R. Pichyangkura, Carbohydr. Res.
2008, 343, 2754–2762.
Acknowledgements
[22] J. Tulla-Puche, I. V. Getun, J. Alsina, F. Albericio, G. Barany, Eur. J.
Org. Chem. 2004, 4541–4544.
The work has been financed by the CICYT (CTQ2009-07758), the Gen-
eralitat de Catalunya (2009SGR1024), the Institute for Research in Bio-
medicine Barcelona (IRB Barcelona), and the Barcelona Science Park.
Received: April 21, 2012
Published online: October 18, 2012
16176
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 16166 – 16176