Aromatic capping surprisingly stabilizes furan moieties in peptides against acidic degradation

Article abstract of DOI:10.1039/c2ob25548k

We herein describe the synthesis of furan containing peptides for further post-synthetic derivatisation in solution through our recently developed furan-oxidation-labeling technology. Previously, it was reported by others that during acidic cleavage of fu

Full text of DOI:10.1039/c2ob25548k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 3999
www.rsc.org/obc
COMMUNICATION
Aromatic capping surprisingly stabilizes furan moieties in peptides against
acidic degradation†
Kurt Hoogewijs, Annelies Deceuninck and Annemieke Madder*
Received 13th March 2012, Accepted 3rd April 2012
DOI: 10.1039/c2ob25548k
deprotection. For C-terminal labeling the carboxylic group can
be left to react with a fluorophore amine group such as amino-
coumarin using a variety of coupling reagents.5–8 However, low
yields have been observed due to the reduced nucleophilicity of
aromatic amines, and epimerization seems to be a severe
problem. As for the internal labeling of peptides, mostly the
natural available nucleophilic functionalities are used. Amine
groups of lysine residues can be modified by reaction with iso-
thiocyanates,9,10 active esters,11,12 sulfonyl chlorides13 or alde-
hydes,14,15 while thiols can be modified by reaction with
maleimides.16,17 However, these nucleophilic functionalities are
widely occurring in peptides and proteins, rendering it difficult
to obtain monolabeled or site-specifically labeled species.
Therefore, it would be of use to introduce a unique electrophi-
lic functionality into the peptide. As has been shown and
reported by others earlier, site specific incorporation of an alde-
hyde functionality still remains a hurdle as the available building
blocks require lengthy syntheses, mostly combined with
additional protecting group chemistries.18,19 Introduction of a
furan moiety as a masked reactive enal functionality appears to
be an attractive alternative to existing methods, since commer-
cially available building blocks such as 3-(2-furyl)-propenoic
acid for N-terminal introduction or N-α-Fmoc-β-(2-furyl)-
L-alanine for internal incorporation are ready to use in solid
phase peptide synthesis.
In continuation of our previous work where a solid-phase
based peptide labelling method was developed,2 exploiting a
furan moiety internally incorporated in the peptide, here we set
out to explore the translation of our furan-oxidation technology
towards solution conditions. However, in the first experiments
major difficulties were experienced upon cleavage of the furan-
peptides from the solid support. In the context of our research on
peptide–DNA binding studies,20 we envisaged the synthesis of a
furan containing analogue of the PhoB peptide, a minimized
epitope of the DNA-binding domain of the transcription factor
PhoB from Escherichia coli, described by Eckel and co-
workers.21 As illustrated in Scheme 1, in a first attempt the
DNA-binding PhoB peptide was synthesized introducing a furan
moiety through N-terminal capping with 3-(2-furyl)propanoic
acid for further DNA binding studies with labeled versions of
this peptide. Upon using a standard cleavage cocktail containing
trifluoro acetic acid (TFA), dithiothreitol (DTT) and H2O, clea-
vage of 1 from the chlorotrityl resin yielded a very complex
We herein describe the synthesis of furan containing peptides
for further post-synthetic derivatisation in solution through
our recently developed furan-oxidation-labeling technology.
Previously, it was reported by others that during acidic clea-
vage of furan-modified peptides, furan moieties can suffer
from degradation. We demonstrate here that this degradation
is position dependent and can be fully suppressed through
introduction of proximate aromatic residues. Versatile intro-
duction of 2-furylalanine at internal, C-terminal as well as
the sensitive N-terminal positions has now been proven
possible.
Recently our group developed a highly efficient nucleic acid
crosslinking methodology based on the selective oxidation of a
furan moiety incorporated into oligonucleotides.1 It was sub-
sequently shown that translation of the methodology to the area
of peptide chemistry was possible. Indeed, we further reported
on the development of a method for the labeling of peptides on
solid support by incorporation of a furan residue into the solid
phase bound peptide. Following oxidative ring opening of the
furan ring into a highly reactive enal and subsequent reductive
amination, labeling could be carried out with the peptide still
bound to the solid support.2 As cleavage of peptides from resins
often requires acidic or basic conditions that are not always com-
patible with the introduced labels, it is of interest to further
develop the methodology for post-synthetic furan based labeling
in solution. As for the labeling methodology of peptides in sol-
ution, the introduction of a fluorescent label at the N-terminus
has been widely described. The amine reactive probe FITC can
be selectively introduced at the N-terminus under buffered con-
ditions at pH 8.5.3 Alternatively, peptides can be prepared by
using the Boc/Bzl strategy, using a Fmoc-protecting group for
lysine residues.4 Following acidic cleavage, the N-terminus can
be selectively labeled with a fluorophore after which the final
N-terminally labeled peptide can then be obtained upon lysine
Laboratory for Organic and Biomimetic Chemistry, Department of
Organic Chemistry, Krijgslaan 281, S4, B-9000 Gent, Belgium.
E-mail: annemieke.madder@ugent.be; Fax: (+) 32-9-264 49 98
†Electronic supplementary information (ESI) available: Detailed pro-
cedures for peptide synthesis and cleavage with relevant LC-MS analysis
for peptides 1 and 2. HPLC and MALDI-TOF analysis for peptides
14–18 and their oxidation products. ESI-MS analyses for peptides 23
and 24. See DOI: 10.1039/c2ob25548k
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3999–4002 | 3999

View Article Online
Scheme 1 HPLC analysis after cleavage of N-terminal furan contain-
ing PhoB peptide with TFA, DTT and water.
Scheme 3 Products obtained upon cleavage of an internally FurAla
modified peptide from the resin.
Table 1 Ratio of products 6–13 formed after cleavage of the internally
FurAla modified peptide 5 from the resin
Scheme 2 HPLC-analysis of an N-terminal furan containing peptide
pentamer after cleavage with TFA, TIS and water.
Cleavage cocktail
Productsa
reaction mixture as evidenced by RPHPLC (Scheme 1). Further-
more LC-MS analysis showed that none of the mass signals
could be correlated to the expected product. As in 2004 Schulz
and coworkers already reported the partial degradation of furan
as a consequence of using TFA and scavengers,22 we figured that
the main problems occurred during the acidolytic cleavage of the
peptide. To analyse the problems in detail and possibly remedi-
ate, the simpler pentamer sequence 2, for which side products
are more easily analysed by LC-MS, was prepared as a test
peptide on Wang resin. We decided not to use DTT as a scaven-
ger and replaced it by triisopropylsilane (Scheme 2). However,
when testing a cleavage cocktail of TFA, TIS and water for the
cleavage of 2, though the situation improved, under these con-
ditions hydrogenation of the aromatic furan ring proved to
produce a side product which is difficult to separate from the
desired compound. A mixture containing reduced tetrahydrofura-
nyl peptide 4 as a major compound was obtained.
We reasoned that the situation might improve when the furan
moiety is more sterically hindered, i.e. located closer to the back-
bone of the peptide as is the case for a β-(2-furyl)-L-alanine
(FurAla) residue. It was therefore analyzed whether incorporat-
ing a FurAla residue towards the interior of a peptide would give
improvement. A new peptide was therefore conceived on Wang
resin containing an internally incorporated FurAla residue.
Additionally, to allow evaluation of the influence of nucleophilic
side chain functionalities a Lys residue was introduced. The
obtained solid phase bound peptide 5 was treated with a range of
cleavage cocktails (Scheme 3 and Table 1). Careful LC-MS
analysis and characterisation of the cleaved products showed that
under the influence of TIS partial or complete hydrogenation of
the furan ring can take place (cf. compounds 8, 9 and 10), while
thiol scavengers can lead to thioacetal formation (compounds 11
to 13). From this comparative study it can be observed that com-
binations in which either phenol or EDT are used as scavengers
lead to further unidentified products and give rather low yields
of the desired furan containing peptide. On the other hand, com-
binations containing either TIS or DTT lead to more favorable
ratios of the desired product.
TFA : H2O
TFA : TIS
TFA : TIS : H2O
TFA : phenol
TFA : phenol : H2O
TFA : phenol : H2O : TIS 88 : 5 : 5 : 2
TFA : EDT
TFA : EDT : H2O
TFA : DTT
TFA : DTT : H2O
95 : 5
95 : 5
6 (50%) + 7
6 (65%) + 8 + 9 + 10
95 : 2.5 : 2.5 6 (55%) + 7 + 8 + 9 + 10
95 : 5 6 (60%) + unident. prod.
95 : 2.5 : 2.5 6 (30%) + 7 + unident. prod.
6 (45%) + 8 + 9 + 10
6 (25%) + unident. prod.
95 : 5
95 : 2.5 : 2.5 6 (25%) + unident. prod.
95 : 5 6 (40%) + 11 + 12 + 13
95 : 2.5 : 2.5 6 (80%) + 11 or 12
a As determined from the crude chromatogram.
Having performed this comparative analysis, we decided to
further investigate the influence of the exact positioning of the
furan moiety within the peptide by preparing a series of the orig-
inally envisaged PhoB 20-mer peptides (vide supra) varying the
position of the furylalanine (Scheme 4). A Rink amide linker
was chosen because of the more reliable loading of the first
residue compared to 2-chlorotrityl linker which was used for
peptide 1. When cleaving 20-mer peptide 14 which contains a
C-terminal β-(2-furyl)-L-alanine residue, the furan moiety was
found to be completely stable during the acidolytic cleavage
with TFA, TIS and water and pure furan-modified peptide was
obtained.
Likewise, for all other depicted sequences containing an
internal β-(2-furyl)-L-alanine moiety, pure products were
obtained corresponding to the desired peptides (Scheme 4, pep-
tides 15–18). No hydrogenation of the furan ring was observed.
However, when N-terminal furan introduction was carried out
with Fmoc-Fur-Ala again complete degradation was observed
upon cleavage from the solid support similar to the observations
previously made with an N-terminal 3-(2-furyl) propanoic acid
residue (cf. Scheme 2).
Surprisingly, while studying in parallel a few other DNA-
binding HMG-I(Y)-peptides, which were capped with 9-acri-
dine-carboxylic acid in an attempt to enhance the stability of the
peptide–DNA complex, we discovered that the stability of the
4000 | Org. Biomol. Chem., 2012, 10, 3999–4002
This journal is © The Royal Society of Chemistry 2012

View Article Online
Scheme 6 HPLC-analysis of oxidation of furan-containing peptide 23
in solution.
capped with acridine-9-carboxylic acid, while the second peptide
21 was capped with 2-naphthoic acid. Both peptides were
cleaved from the solid phase with a mixture of TFA, TIS and
water and could be obtained pure. The furan moiety remained
intact during the acidic cleavage. A tentative explanation for this
unexpected stability of the N-terminal furan ring capped with an
aromatic moiety can be found in a π–π stacking of the aromatic
moiety with the furan ring, possibly protecting this moiety
against side reactions during acidic cleavage.
Having established suitable synthesis and cleavage conditions
for furan-peptides, with variable positioning of the furan moiety
throughout the peptide, we further explored various oxidation
conditions for generation of the desired reactive unsaturated alde-
hyde functionality. Fast scanning of oxidation conditions was
carried out on the short FurAla containing peptide 23. One of
the above used N-terminal capping moieties was included to
ascertain and show that the acridine moiety does not negatively
influence the outcome of the oxidation reaction.
Scheme 4 HPLC-analysis after cleavage of C-terminally and internally
furan-modified peptides with TFA, TIS and water (95 : 2.5 : 2.5). The
position of the FurAla residue is indicated in bold.
During the optimisation studies, it was shown that treatment
of furan-peptides with 1 equivalent of NaOCl did not lead to
reaction and upon increasing the amount of oxidant, only partial
oxidation could be observed next to peptide degradation pro-
ducts and remaining starting material. Next, a combination of
Rose Bengal and visible light irradiation, known to generate
singlet oxygen, was applied. Though most of the starting
material could be converted into the desired enal, next to remain-
ing starting material also products of over oxidation could be
detected.
However, the use of a single equivalent of NBS led to clean
conversion into the desired furan-oxidized peptide 24 in 1 hour.
As evidenced from the RPHPLC analysis in Scheme 6, an extra
product was visible. Through MS analysis (see ESI†) it was
shown that the originally obtained reactive aldehyde was
accompanied by its hydrate as can be expected in aqueous sol-
ution. It was further established that also for the furan-containing
PhoB 20-mer peptides (vide supra, Scheme 4) oxidation could
be achieved using NBS (see ESI†). MALDI-TOF analysis of the
oxidation reactions revealed a concomitant loss of H2O in case
of peptides obtained from resins 14 and 17, resulting from an
intramolecular imine formation with the proximate lysine resi-
dues, illustrating the availability of the unsaturated aldehyde for
reaction with nucleophiles.
Scheme 5 HPLC-analysis after cleavage of N-terminally furan-
modified peptides capped with an aromatic moiety.
N-terminal furan ring can be enhanced by capping the N-termi-
nus with an aromatic moiety (Scheme 5). Two test peptides were
prepared with an N-terminal furylalanine residue which was
further capped with an aromatic unit. The first peptide 19 was
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3999–4002 | 4001

View Article Online
4 L. Fulop, B. Penke and M. Zarandi, J. Pept. Sci., 2001, 7, 397–401.
Conclusions
5 P. D. Edwards, R. C. Mauger, K. M. Cottrell, F. X. Morris, K. K. Pine,
M. A. Sylvester, C. W. Scott and S. T. Furlong, Bioorg. Med. Chem. Lett.,
2000, 10, 2291–2294.
6 K. T. Nguyen, X. B. Hu and D. H. Pei, Bioorg. Chem., 2004, 32, 178–
191.
7 E. Onitsuka, T. Okumura, H. Murakami, N. Nishino and F. Morimatsu,
Biosci., Biotechnol., Biochem., 2006, 70, 2836–2843.
8 V. F. Pozdnev, Int. J. Pept. Protein Res., 1994, 44, 36–48.
9 Z. Georgoussi, A. Evangelopoulos, M. G. Ludwig and Heilmeyer,
Biochem. Pharmacol., 1986, 35, 4571–4573.
10 A. R. Morales, K. J. Schafer-Hales, A. I. Marcus and K. D. Belfield, Bio-
conjugate Chem., 2008, 19, 2559–2567.
11 W. Y. Leung, P. A. Trobridge, R. P. Haugland and F. Mao, Bioorg. Med.
Chem. Lett., 1999, 9, 2229–2232.
12 X. Liu, H. Wang, S. C. Liang and H. S. Zhang, Chromatographia, 2001,
53, 326–330.
13 G. Weber, Biochem. J., 1952, 51, 155–167.
In summary, the introduction of a furan moiety in peptides via
solid phase peptide synthesis and subsequent release of the furan
containing peptides in solution was optimized. Next to C-term-
inal and internal modification of the peptide with a furan moiety
proceeding smoothly using N-α-Fmoc-β-(2-furyl)-L-alanine, N-
terminally furan-modified peptides can be obtained in pure form
through capping of the peptide with an aromatic moiety in order
to protect the furan ring against degradation during acidolytic
cleavage. It was further shown that the furan moiety in the result-
ing peptides can be selectively converted into a reactive enal
upon oxidation with NBS. In that sense current methodology
allows for easy and site selective introduction of a caged electro-
philic functionality into peptides allowing oxidation triggered
activation at any desired moment for further labeling or conju-
gation purposes.
14 S. C. Beale, Y. Z. Hsieh, D. Wiesler and M. Novotny, J. Chromatogr., A,
1990, 499, 579–587.
15 J. P. Liu, Y. Z. Hsieh, D. Wiesler and M. Novotny, Anal. Chem., 1991,
63, 408–412.
16 K. Tyagarajan, E. Pretzer and J. E. Wiktorowicz, Electrophoresis, 2003,
24, 2348–2358.
17 J. E. T. Corrie, J. Chem. Soc., Perkin Trans. 1, 1994, 2975–2982.
18 P. Johannesson, G. Lindeberg, W. M. Tong, A. Gogoll, A. Karlen and
A. Hallberg, J. Med. Chem., 1999, 42, 601–608.
19 J. C. Spetzler and T. Hoeg-Jensen, Tetrahedron Lett., 2002, 43, 2303–
2306.
20 D. Verzele, L. L. G. Carrette and A. Madder, Drug Discovery Today:
Technol., 2010, 7, e115–e123.
21 R. Eckel, S. D. Wilking, A. Becker, N. Sewald, R. Ros and
D. Anselmetti, Angew. Chem., Int. Ed., 2005, 44, 3921–3924.
22 A. Schulz, A. Busmann, E. Kluver, M. Schnebel and K. Adermann,
Protein Pept. Lett., 2004, 11, 601–606.
Notes and references
1 (a) F. L. van Delft, A. M. Jawalekar, M. O. de Beeck and A. Madder,
Chem. Commun., 2011, 47, 2796–2798; (b) A. Madder, K. Stevens,
D. D. Claeys, S. Catak, S. Figaroli, M. Hocek, J. M. Tromp, S. Schurch
and V. Van Speybroeck, Chem.–Eur. J., 2011, 17, 6940–6953;
(c) A. Madder and M. Op de Beeck, J. Am. Chem. Soc., 2011, 133, 796–
807; (d) K. Stevens and A. Madder, Nucleic Acids Res., 2009, 37, 1555–
1565.
2 A. Deceuninck and A. Madder, Chem. Commun., 2009, 340–342.
3 M. J. Little, D. M. Paquette, M. D. Harvey and P. R. Banks, Anal. Chim.
Acta, 1997, 339, 279–288.
4002 | Org. Biomol. Chem., 2012, 10, 3999–4002
This journal is © The Royal Society of Chemistry 2012