Serendipitous discovery of an efficient method for the synthesis of dimeric-RGD analogues using DMAP-photoirradiation

Article abstract of DOI:10.1039/c5nj01810b

We describe a novel disulfide reaction via UV/N,N-dimethylaminopyridine (DMAP) methodology for efficient construction of alkyl and aryl symmetrical disulfides. Compared with other chemical strategies, our methodology is distinguished in that the dimerization reaction can proceed efficiently without metal catalysts, expensive reagents or forcing conditions. This methodology was successfully applied to the preparation of complex dimeric biological peptide-based molecules, and the dimeric RGD peptides produced by this methodology had better binding affinity than the commercially available E-[RGDfK]₂. Our methodology will greatly extend the method to the construction of a complex disulfide bond under mild conditions.

Full text of DOI:10.1039/c5nj01810b

NJC
View Article Online
View Journal | View Issue
LETTER
Serendipitous discovery of an efficient method for
the synthesis of dimeric-RGD analogues using
DMAP-photoirradiation†
Cite this: New J. Chem., 2015,
3
9, 7750
Received (in Montpellier, France)
a
b
a
a
c
Ruiping Zhang,‡* Yao Sun,‡ Ying Qiao, Jianding Li and Jun Xie
11th July 2015,
Accepted 14th August 2015
DOI: 10.1039/c5nj01810b
www.rsc.org/njc
We describe a novel disulfide reaction via UV/N,N-dimethyl-
Dimerization of cysteine residues has become a vital method
aminopyridine (DMAP) methodology for efficient construction of for maintaining biologically active conformations of physio-
alkyl and aryl symmetrical disulfides. Compared with other chemical logically important peptides such as somatostatin and vaso-
1
2
strategies, our methodology is distinguished in that the dimerization pressin. The development of new and efficient protocols for
reaction can proceed efficiently without metal catalysts, expensive the preparation of complex disulfide biomolecules under mild
reagents or forcing conditions. This methodology was successfully reaction conditions is an important challenge for medicinal
applied to the preparation of complex dimeric biological peptide- chemistry.
based molecules, and the dimeric RGD peptides produced by this
methodology had better binding affinity than the commercially many biological processes. The synthesis of multivalent pep-
available E-[RGDfK] . Our methodology will greatly extend the tides has further enhanced the interaction of individual ligands
method to the construction of a complex disulfide bond under with their receptors. For example, the RGD (arginine–glycine–
mild conditions. aspartic acid) tripeptide motif plays an essential role in the
molecular recognition of integrin a , which is overexpressed in
Multivalent interactions are known to play a critical role in
1
3
2
v 3
b
1
4
The conversion of thiols to the corresponding disulfides has various types of tumors. Monomeric RGD-based probes have
1
gained significant applications both in chemical industry and been successfully prepared and exhibit selectivity for integrin
2
15
biological science. Disulfides are valuable intermediates for avb3 in vitro and in vivo. However, monomeric RGD-based
3
16
the production of sulfenyls and sulfinyls in organic chemistry.
probes exhibited low cellular uptake in vitro. To overcome this
Furthermore, disulfide bonds are the principal entities respon- issue, multimeric RGD ligands have been developed, and they
sible for stabilizing the secondary or tertiary structure of demonstrate higher receptor binding affinity in vitro and better
4
5
17
proteins. Procedures involving the use of halogen derivatives,
tumor retention in vivo due to their multivalent composition.
Therefore, there still remains a need for mild reaction conditions
6
7
8
9
transition metal salts, peroxides, molecular oxygen, nitric oxide,
,6-dicarboxypyridinium chlorochromate and cerium salts have capable of preparing multimeric RGD peptides with high flexi-
1
0
11
2
been introduced for oxidative coupling of thiols to disulfides. bility, efficiency, and chemoselectivity.
Despite these promising examples, there are still several drawbacks
Thiol–yne click chemistry has become an important tool for
including the need for expensive reagents used in excess, long the construction of both multivalent molecules of biological
18
reaction times, and forcing reaction conditions. Additionally, these interest and assorted materials (Scheme 1a). More recently, Sun
methods are mainly focused on producing a simple disulfide bond and co-workers have successfully applied thiol–yne click chemistry
1
9
based on small molecules, and it is uncertain whether these to the construction of multivalent peptide-based imaging probes.
protocols are amenable to the synthesis of complex biomolecules. It is worth noting that UV-induced thiol–yne reaction inevitably
2
0
produces some amount of disulfide as the main byproduct.
a
b
c
Therefore, we envisioned that the UV-irradiation of thiol–yne
reaction provides a possible method to make disulfide bonds
under mild conditions. More recently, good yields have been
reported for disulfide reaction catalyzed by bases such as
Department of Radiology, First Hospital, Shanxi Medical University,
Taiyuan 030000, China. E-mail: zrp_7142@126.com
Key Laboratory of Combinatorial Biosynthesis and Drug Discovery,
Wuhan University School of Pharmaceutical Science, Wuhan 430071, China
Department of Molecular Biology, Shanxi Medical University, Taiyuan 030000,
China
2
1
3
tetramethylguanidine and Et N. Inspired by both facts, herein
we developed a photoirradiation methodology for the efficient
preparation of dimeric RGD analogue products via substituting
†
‡
Electronic supplementary information (ESI) available. See DOI: 10.1039/c5nj01810b
R. P. Zhang and Y. Sun contributed equally to this work.
7750 | New J. Chem., 2015, 39, 7750--7753
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
Letter
NJC
low-molecular organic base catalyst with a powerful effect on
2
2
many organic reactions. Hence, we speculate a possible base-
catalyzed mechanism via DMAP in our methodology.
The scope of the dimerization reaction was investigated using
a range of substrates and the optimized UV/DMAP conditions
(Table 2). A similar yield was obtained with the RGD analogue
RADfC (entry 1, 75%), and the dimeric RADfC was confirmed by
MALDI-TOF-MS (Fig. S3, m/z = 1184.3, ESI†) and ESI-MS (Fig. S4,
m/z = 1183.6, ESI†). Then a more complex dimeric-peptide (AE105,
Scheme 1 (a) UV-induced thiol–yne click chemistry; (b) UV/DMAP
method for dimerization of RGD.
1
1-mer peptide antagonist) was also successfully prepared via our
the typical thiol–yne click reaction catalyst 2,2-dimethoxy-2-
phenylacetophenone (DMPA) for N,N-dimethylaminopyridine
methodology (entry 2, 54%), and the dimeric peptide was con-
firmed by MALDI-TOF-MS (Fig. S5, m/z = 2792.3, ESI†). In light of
these results, we then turned our attention to the suitability of our
method for preparing small disulfides. To our delight, every thiol
substrate tested including aryl and alkyl thiols produced the
corresponding disulfides in good to excellent yields (entries 3–8,
(DMAP) (Scheme 1b).
Initially, we chose RGDfC (1) as a model reaction to establish
the best conditions for dimerization of the RGD analogue (2)
under various conditions (Table 1). The desired product was
obtained in 65% yield after reaction in water for 3 h under UV
irradiation (4 W, 365 nm) in the presence of DMAP (entry 1).
The results of MALDI-TOF-MS (ESI,† Fig. S1, m/z = 1156.223)
and ESI-MS (Fig. S2, m/z = 1155.5, ESI†) confirmed the dimer-
RGDfC as the desired product. A better yield was obtained using
DMF as solvent (yield 75%, entry 2), while the optimal result was
achieved using DMSO (yield 78%, entry 3). The role of polar
aprotic solvents having high dielectric constants, such as DMF
and DMSO are known to increase the oxidation rate of thiol with
7
4
8–88%), with Fmoc-cysteine giving only moderate yield (entry 9,
2%). It is possible that the low yield of dimeric Fmoc-cysteine is
attributable to DMAP-mediated Fmoc cleavage during the reac-
tion. The structure of every disulfide product was confirmed by
1
13
ESI-MS, HNMR and CNMR (ESI†). Based on the above results,
our methodology mainly has the following advantages. Firstly, our
research efforts have established that the bulky macrocycles of
RGD analogues or small molecule thiols could be effectively
converted into the corresponding disulfides in the presence of
simple and cheap DMAP/UV. Secondly, this disulfide reaction can
occur under mild reaction conditions and can tolerate a wide
range of functional groups such as –OH and –COOH.
2
1b
oxygen.
Therefore, the better yields obtained from organic
solvents were consistent with the literature. It was also found that
the DMAP loading presents a critical determinant of the reaction
efficiency in this reaction; decreasing the amount of DMAP from
To evaluate whether dimeric RGD analogues prepared via
our methodology maintained binding affinity and specificity for
1
equiv. to 0.1 equiv. can extend the reaction time from 5 to 12 h
and reducing yields from 78% to 35% (entries 3–5). Finally, we
investigated whether DMAP or UV could catalyze this dimerization
reaction exclusively. It is noteworthy that DMAP alone could Table 2 Synthesis of disulfides under UV/DMAP
catalyze the reaction to furnish moderate yields (46%) of the
dimer, albeit with extended reaction time (entry 6), while only a
little desired product was observed under UV irradiation in
a
Entry Thiols (1)
Product (2)
Time (h) Yield (%)
the absence of a catalyst after a long time of 12 h (entry 7, 8%).
It is well known that DMAP is a good example of a modern
1
2
3
RADfC (1a)
AE105(1b)
Dimer RADfC (2a)
Dimer AE105 (2b)
3
5
0.5
75
54
80
Table 1 Optimization for dimerization of RGDfC
4
0.5
82
5
6
0.5
0.5
87
88
a
b
Entry
Catalyst
Solvents
Time (h)
Yield (%)
7
0.5
82
1
2
3
4
5
6
7
DMAP/UV (1 equiv.)
DMAP/UV (1 equiv.)
DMAP/UV (1 equiv.)
DMAP/UV (0.5 equiv.)
DMAP/UV (0.1 equiv.)
DMAP (1 equiv.)
UV
H
DMF
DMSO
DMSO
DMSO
DMSO
DMSO
2
O
3
3
3
65
75
78
74
35
48
8
8
9
0.5
0.5
78
42
5
12
12
12
a
Reaction conditions: [RGDfC] = 0.01 M in solvent, UV source: UVGL-55
b
a
Handheld UV Lamp, 4 W, 365 nm. Isolated yield.
Isolated yields.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem., 2015, 39, 7750--7753 | 7751

View Article Online
NJC
Letter
Table 3 IC50 values for mono and dimer RGD analogues
Acknowledgements
Entry
Peptides
IC50/nM
We would like to thank Dr Samuel D. Banister (University of
1
2
3
4
5
Monomeric-RGDfC
Dimeric-RGDfC
68.0 Æ 7.8 nM Stanford) for his insightful suggestions and his contributions
25.0 Æ 5.1 nM
to the preparation of the manuscript. This study was supported
E-[RGDfK]
2
46.0 Æ 6.7 nM
by grants from the National Natural Science Foundation of
Monomeric-RADfC
Dimeric-RADfC
4800 nM
4800 nM
China (81371628, 81571747), Postdoctoral Science Foundation
of China (2014T70233, 2013M541206) and International Science
and Technology Cooperation Projects of Shanxi Province
125
integrin a b , competitive cell binding assay using I-echistatin
as the integrin a
U87MG human glioblastoma cells. The mono- and dimeric
v
3
(
2015081035).
v 3
b specific radioligand was performed on
2
3
RGD and RAD analogues, and the obtained IC50 values are Notes and references
summarized in Table 3. As expected, the dimeric RGDfC peptide
1
(a) Y. Kanda and T. J. Fukuyama, J. Am. Chem. Soc., 1993,
15, 8451–8452; (b) B. Schmidt, S. Lindman, W. Tong,
showed higher binding affinity (IC50 = 25.0 Æ 5.1 nM) compared
to the monomeric-RGDfC (IC50 = 68.0 Æ 7.8 nM). The dimeric
RGD peptide constructed by our method exhibited better binding
affinity than the commercially available dimeric RGD (E-[RGDfK]2,
cyclic RGD, IC50 = 46.0 Æ 6.7 nM). It is possible that the presence
1
G. Lindeberg, A. Gogoll, Z. Lai, M. Thornwall, B. Synnergren,
A. Nilson, C. J. Welch, M. Sohtell, C. Westerlund, F. Nyberg,
A. Karlen and A. J. Hallberg, J. Med. Chem., 1997, 40, 903–919.
(a) A. Dhakshinamoorthy, M. Alvaro and H. Garcia, Chem.
Commun., 2010, 46, 6476–6478; (b) A. R. Hajipour, S. E.
Mallakpour and H. Adibi, J. Org. Chem., 2002, 67,
8666–8668; (c) M. Kirihara, Y. Asai, S. Ogawa, T. Noguchi,
A. Hatano and Y. Hirai, Synthesis, 2007, 3286–3289.
2
2
of a mini-PEG linker in E-[RGDfK] decreases the binding avidity
of RGD in this dimer. Finally, mono- and dimeric RAD showed
non-specific binding to the integrin a , consistent with the
v 3
b
24
in vitro study by Garanger et al.
In conclusion, we have reported a novel and efficient method
for the construction of disulfides from thiols using UV irradia-
tion in the presence of DMAP. This method has been success-
fully applied to the construction of a library of dimeric RGD
analogues. Moreover, the dimeric RGD analogues exhibited
higher binding affinity than commercially available dimers.
Finally, this methodology is amenable to the synthesis of other
dimeric peptide-based small molecules or biomolecules. Their
great versatility and flexibility are very important for future
applications.
3
4
5
(a) B. Douglass and R. V. Norton, J. Org. Chem., 1968, 33,
2
104–2106; (b) B. Douglass, J. Org. Chem., 1974, 39, 563–564.
R. Fatemeh, K. Tayeb and R. S. Mohammad, Catal. Commun.,
013, 40, 13–17.
(a) J. Drabowicz and M. Mikocayczyk, Synthesis, 1980, 32–34;
b) G. W. Kabalka, M. S. Reddy and M. L. Yao, Tetrahedron
2
(
Lett., 2009, 50, 7340–7342.
(a) T. J. Wallace, J. Org. Chem., 1966, 31, 1217–1221;
6
7
(b) M. Oba, K. Tanaka, K. Nishiyama and W. Ando, J. Org.
Chem., 2011, 76, 4173–4177.
V. Kesavan, D. Bonnet-Delpon and J. P. Begue, Synthesis,
2
000, 223–225.
Experimental
Synthesis of AE105
8
9
K. T. Liu and Y. C. Tong, Synthesis, 1978, 669–670.
W. A. Pryor, D. F. Church, C. K. Govindan and G. Crank,
J. Org. Chem., 1982, 47, 156–159.
0 M. Tajbakhsh, R. Hosseinzadeh and A. Shakoori, Tetra-
hedron Lett., 2004, 45, 1889–1893.
1 (a) H. Firouzbadi, N. Iranpoor and H. A. Parham, Synth.
Commun., 1984, 14, 717–724; (b) C. C. Silveira and S. R.
Mendes, Tetrahedron Lett., 2007, 48, 7469–7471.
2 (a) N. L. Daly, K. J. Rosengren and D. J. Craik, Adv. Drug
Delivery Rev., 2009, 61, 918–930; (b) M. Almeida, R. Li and
S. H. Gellman, J. Am. Chem. Soc., 2012, 134, 75–78.
The peptide AE105 (Cys-Gly-Asp-Cha-Phe-(D)Ser-(D)Arg-Tyr-Leu-
Trp-Ser-NH ) was synthesized on Tentagel S RAM resin using
2
traditional Fmoc solid-phase peptide chemistry. After deprotec-
tion and cleavage from the resin using 93% TFA, 5% TIPS, and
1
1
2 2
2% H O for 2 h, the peptide was precipitated in cold Et O and
washed with Et O three times. The dried peptide was purified
2
by prep-HPLC and checked by MALDI-MS: m/z 1397.0.
1
Cell binding assay
5
U87MG cells (1 Â 10 ) were suspended in 500 mL of DMEM 13 M. Mammen, S. K. Choi and G. M. Whitesides, Angew.
seeded in 12-well tissue culture plates and incubated at 37 1C
overnight. The plate was incubated with
Chem., Int. Ed., 1998, 37, 2755–2794.
1
25
I-echistatin in the 14 (a) N. Assa-Munt, X. Jia, P. Laakkonen and E. Ruoslahti,
presence of increasing concentrations of different RGD and RAD
peptide analogues (0–1000 nM). After the cells were incubated
for 2 h, the supernatant was removed and washed with binding
buffer. Radioactivity was determined using a gamma counter.
The best-fit 50% inhibitory concentration (IC50) values for the
Biochemistry, 2001, 40, 2373–2378; (b) E. Ruoslahti and
M. D. Pierschbacher, Science, 1987, 238, 491–497;
(c) L. Sernissi, M. Petrovic, D. Scarpi, A. Guarna, A. Trabocchi,
F. Bianchini and E. G. Occhiato, Chem. – Eur. J., 2014, 20,
11187–11203.
U87MG cells were calculated by fitting the data with non-linear 15 (a) M. A. Dechantsreiter, E. Planker, B. Matha, E. Lohof,
regression using Graph-Pad Prism (GraphPad Software, Int.). G. Holzemann, A. Jonczyk, S. L. Goodman and H. Kessler,
7752 | New J. Chem., 2015, 39, 7750--7753
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015

View Article Online
Letter
J. Med. Chem., 1999, 42, 3033–3040; (b) R. Haubner,
NJC
(d) H. Q. Peng, C. L. Sun, J. F. Xu, L. Y. Niu, Y. Z. Chen,
L. Z. Wu, C. H. Tung and Q. Z. Yang, Chem. – Eur. J., 2014,
20, 11699–11702.
B. Kuhnast, C. Mang, W. A. Weber, H. Kessler, H. J. Wester
and M. Schwaiger, Bioconjugate Chem., 2004, 15, 61–69.
16 (a) X. Chen, S. Liu, Y. Hou, M. Tohme, R. Park, J. R. Bading 19 Y. Sun, X. W. Ma, K. Cheng, B. Wu, J. Duan, H. Chen, L. Bu,
and P. S. Conti, Mol. Imaging Biol., 2004, 6, 350–359;
b) X. Chen, M. Tohme, R. Park, Y. Hou, J. R. Bading and
R. Zhang, X. Hu, Z. Deng, L. Xing, X. Hong and Z. Cheng,
Angew. Chem., Int. Ed., 2015, 54, 5981–5984.
(
P. S. Conti, Mol. Imaging Biol., 2011, 6, 169–177.
7 (a) W. Liu, G. Y. Hao, M. Long, T. Anthony, J. T. Hsieh and
20 M. Minozzi, A. Monesi, D. Nanni, P. Spagnolo and
N. Marchetti, J. Org. Chem., 2011, 76, 450–459.
1
X. K. Sun, Angew. Chem., Int. Ed., 2009, 48, 7346–7349; 21 (a) T. J. Wallance, N. Jocobson and A. Schriesheim, Nature,
(
b) S. L. Liu, Z. B. Li, L. P. Yap, C. W. Huang, R. Park and
1964, 201, 609–610; (b) J. L. G. Ruano, A. Parra and
J. Aleman, Green Chem., 2008, 10, 706–711.
P. S. Conti, Chem. – Eur. J., 2011, 17, 10222–10225;
(
c) G. Thumshirn, U. Hersel, S. L. Goodman and H. Kessler, 22 U. Ragnarsson and A. L. Grehn, Acc. Chem. Res., 1998, 31,
Chem. – Eur. J., 2003, 9, 2717–2725. 494–501.
8 (a) J. W. Chan, C. E. Hoyle and A. B. Lowe, J. Am. Chem. Soc., 23 W. Cai, X. Zhang, Y. Wu and X. Chen, J. Nucl. Med., 2006, 47,
009, 131, 5751–5753; (b) R. Hoogenboom, Angew. Chem., 1172–1180.
Int. Ed., 2010, 49, 3415–3417; (c) D. Konkolewicz, A. W. Gray 24 E. Garanger, D. Boturyn, J. L. Coll, M. C. Favrot and
and S. Perrier, J. Am. Chem. Soc., 2009, 131, 18075–18077; P. Dumy, Org. Biomol. Chem., 2006, 4, 1958–1965.
1
2
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2015
New J. Chem., 2015, 39, 7750--7753 | 7753