Solid-phase synthesis of a library of C-terminal agmatine dipeptides using a backbone amide linker (BAL) strategy

Article abstract of DOI:10.1016/j.tetlet.2012.08.002

In this Letter, we report a novel solid-phase strategy using a backbone amide linker (BAL) attached to a polystyrene support for the synthesis of C-terminal agmatine dipeptides. Our method eliminates the need to purify intermediates by column chromatography and enables us to build rapidly an 18-member library of C-terminal agmatine dipeptides which are subsequently screened for inhibitory activity against a viral enzyme. 2012 Elsevier Ltd. All rights reserved.

Full text of DOI:10.1016/j.tetlet.2012.08.002

Tetrahedron Letters 53 (2012) 5511–5514
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Solid-phase synthesis of a library of C-terminal agmatine dipeptides using
a backbone amide linker (BAL) strategy
⇑
Melgious Jin Yan Ang, Huichang Annie Lim, Cheng San Brian Chia
Experimental Therapeutics Centre, Agency for Science, Technology and Research (A⁄STAR), 31 Biopolis Way, Nanos #03-01, Singapore 138669, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
In this Letter, we report a novel solid-phase strategy using a backbone amide linker (BAL) attached to a
polystyrene support for the synthesis of C-terminal agmatine dipeptides. Our method eliminates the need
to purify intermediates by column chromatography and enables us to build rapidly an 18-member library
of C-terminal agmatine dipeptides which are subsequently screened for inhibitory activity against a viral
enzyme.
Received 14 June 2012
Revised 19 July 2012
Accepted 1 August 2012
Available online 9 August 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Agmatine
Backbone amide linker (BAL)
Protease inhibitor
Peptides containing a C-terminal agmatine (4-aminobutylgua-
nidine) have been shown to inhibit competitively trypsin-like ser-
ine proteases such as thrombin, plasmin, and furin in micro- to
nanomolar potencies (Fig. 1a).1,2 Recently, we reported that such
peptidomimetics also inhibited the West Nile virus (WNV) NS2B/
NS3 serine protease at low micromolar potencies, the most active
chloride resin (Fig. 2a) was unsuccessful as acid treatment during
resin cleavage led to a complex mixture of products.7 Another re-
port used an arenesulfonyl chloride linker (Figure 2b), which re-
quired hydrogen fluoride for resin cleavage and hence needed a
suitably equipped laboratory.8 A more acid-labile, Fmoc-chemistry
compatible chroman (3,4-dihydro-2,5,7,8-tetramethyl-2H-1-ben-
zopyranyl acetic acid) sulfonyl chloride linker was later developed
(Fig. 2c) but to the best of our knowledge, this linker is not com-
mercially available and required a six-step synthetic procedure.9
These reports suggested that anchoring a guanidine moiety to a
solid support was not trivial and thus, unsuitable for the synthesis
of a C-terminal agmatine peptide library. To circumvent these
shortcomings, we adopted a backbone-amide linker (BAL) strategy,
first described by Jensen et al., for the synthesis of C-terminal mod-
ified peptides.10 In their strategy, an amino acid with an O-allyl-
protected carboxyl group was first anchored to an aldehyde-func-
tionalized resin via its N-terminal amine followed by peptide chain
elongation using standard Fmoc SPPS (see Boas et al.,11 for an
excellent review on the various uses of the BAL resin). In our adap-
tation, we anchored 1,4-diaminobutane to a 2-(3,5-dimethoxy-4-
formylphenoxy)ethyl (DFPE) polystyrene resin before reducing
the resulting imine to a secondary amine. Next, the agmatine moi-
ety was generated on resin by reacting the remaining free primary
amine with a guanylating reagent. The peptide chain was next
built on the secondary amine using standard Fmoc SPPS. Finally,
the target C-terminal agmatine dipeptide was released from the re-
sin by trifluoroacetic acid (TFA) treatment (Scheme 1).
compound exhibiting an IC50 value of 4.7 1.2 l
M (Fig. 1b).3 Based
on the conventional terminology put forward by Schechter and
Berger,4 the P1, P2, and P4 regions of such inhibitors have been
studied extensively and optimized.5,6 However, there are limited
studies on the P3 region and a panel of inhibitors with P3 amino
acid mutations could provide better insights into inhibitor-enzyme
specificity and binding in the development of new anti-virals.
So far, conventional strategies used to synthesize C-terminal
agmatine peptides involved solution-phase chemistry where an
appropriately protected agmatine moiety is coupled to Fmoc-pro-
tected amino acids in a stepwise process.1–3 These approaches
were time-consuming and required significant quantities of organ-
ic solvents for silica gel column purification of the intermediates
after each coupling step. As C-terminal agmatine peptides are valu-
able compounds that can be used to probe peptide sequence spec-
ificities of various trypsin-like proteases, we decided to develop a
new strategy that would allow us to synthesize conveniently an
agmatine peptide library with minimal solvent wastage.
Toward this end, we adopted an Fmoc-chemistry solid-phase
peptide synthesis (SPPS) approach. A number of solid-phase strat-
egies have been reported, all of which involved anchoring a guani-
dine moiety to a resin. The earliest report utilizing 2-chlorotrityl
Herein, we report a novel solid-phase BAL strategy used for the
synthesis of an 18-member library of C-terminal agmatine dipep-
tides for inhibitory activity screening against the WNV NS2B/NS3
⇑
Corresponding author. Tel.: +65 64070348; fax: +65 64788768.
E-mail address: cschia@etc.a-star.edu.sg (C.S.B. Chia).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.08.002

5512
M. J. Y. Ang et al. / Tetrahedron Letters 53 (2012) 5511–5514
(a)
(b)
P4
P3
P2
P1
NH2
O
NH
NH2
NH
NH2
H
N
H
N
H
N
Peptide
N
H
N
H
N
H
O
O
O
NH2
Figure 1. (a) Schematic of a peptide linked to a C-terminal agmatine (in blue); (b) the most potent WNV NS2B/NS3 serine protease inhibitor.3 The molecule is demarcated
into four regions (P1–P4) based on the terminology put forward by Schechter and Berger.4
H
N
O
O
S
O
Cl
Cl
Cl
O
Cl
S
O
O
O
(c)
(a)
(b)
= polystyrene resin
Figure 2. Various linkers used for anchoring guanidino moieties: (a) 2-chlorotrityl chloride; (b) arenesulfonyl chloride; (c) chroman sulfonyl chloride.
H3CO
HN
O
H3CO
H3CO
O
H3CO
HN
O
O
BocHN
a, b
c
d, e
O
H
2
NBoc
NHBoc
NBoc
NHBoc
OCH3
OCH3
OCH3
OCH3
N
O
N
H
NH2
H2N
N
H
= polystyrene resin
f, e, g
H3CO
O
BocHN
H2N
O
2
2
O
H
NBoc
NHBoc
NH
NH2
OCH3
H
H
N
h
N
N
O
N
N
H
N
H
N
H
N
H
O
R1
O
R1
O
Scheme 1. Solid-phase synthesis of C-terminal agmatine dipeptides. Reagents and conditions: (a) resin swelling in DMF, 25 °C, 1 h; (b) 1,4-diaminobutane, NaCNBH3, 1%
AcOH in DMF (v/v), microwave, 60 °C, 10 min, repeat; (c) N,N0-bis(tert-butoxycarbonyl)thiourea, DIPEA, DMF, microwave, 60 °C, 10 min; (d) Fmoc-Lys(Boc)-OH, DIC, HOBt,
90% CH2Cl2 in DMF (v/v), microwave, 60 °C, 10 min, repeat; (e) 20% piperidine in DMF (v/v), microwave, 60 °C, 5 min; (f) Fmoc-amino acid-OH, DIC, HOBt, 90% CH2Cl2 in DMF
(v/v), microwave, 60 °C, 10 min; (g) 4-biphenylacetic acid, DIC, HOBt, 90% CH2Cl2 in DMF (v/v), microwave, 60 °C, 10 min; (h) 95% TFA in CH2Cl2 (v/v), 25 °C, 30 min.
serine protease [inhibitory data will be published elsewhere in an
appropriate journal].
twice, resulted in maximal anchoring of the diamine to the resin
(Table 1).
To our best knowledge, the BAL resin has not been used to syn-
thesize C-terminal agmatine peptides before so a number of exper-
imental conditions had to be optimized. The first involved resin
swelling (step a, Scheme 1) and we recommend that the DFPE resin
be soaked in DMF for at least an hour at room temperature (RT) to
ensure proper resin swelling, failing which would result in low
yields. The second step involved anchoring 1,4-diaminobutane to
the aldehyde-functionalized DFPE resin by reductive amination
(step b, Scheme 1). We discovered that mild acid (1% AcOH in
DMF) with 10 min microwave irradiation at 60 °C, conducted
We next proceeded to optimize the guanylation step (step c,
Scheme 1) using different conditions and solvents (Table 2). We
discovered that a 10-fold excess of the guanylating reagent, N,N0-
bis(tert-butoxycarbonyl)thiourea in DMF with microwave irradia-
tion for 10 min at 60 °C gave the best yield as quantified by the for-
mation of the intermediate: Fmoc-Arg(Pbf)-agmatine using ESI-MS
(Table 2). Interestingly, the secondary amine proximal to the resin
did not react with the excess guanylating reagent and no double-
guanylated intermediates were detected by MS. Similar observa-
tions on the stability of secondary amines toward alkyl-S-isourea

M. J. Y. Ang et al. / Tetrahedron Letters 53 (2012) 5511–5514
5513
Table 1
Table 4
Reductive amination conditions using NaCNBH3 (step b, Scheme 1). Crude yields (%)
were estimated based on mass spectrometry detection of the intermediate: Fmoc-
Arg(Pbf)-agmatine. RT represents room temperature
Structures of the synthesized C-terminal agmatine dipeptides. AA denotes the
conventional 3-alphabet notations of amino acids. All amino acids used in this study
were l-isomers unless otherwise stated. Overall yields (%) were calculated using the
resin loading factor (0.87 mmol/g) for the lyophilized product after HPLC purification.
Nle = norlecuine; Orn = ornithine
Entry Temp (°C) Solvent
Microwave Time (min) Crude yield (%)
1
2
3
4
5
6
7
8
RT
RT
60
60
60
60
60
60
DMF
MeOH
DMF
DMF
1% AcOH in DMF Yes
1% AcOH in DMF Yes
1% AcOH in DMF Yes
1% AcOH in DMF Yes
No
No
Yes
Yes
60
60
30
10
30
20
10
10 (ꢀ2)
ꢁ30
0
H2N
2
ꢁ50
ꢁ30
>90
ꢁ80
ꢁ70
>90
NH
NH2
H
AA
N
N
H
N
H
O
O
Table 2
Primary amine guanylation conditions using N,N0-di-(t-butoxycarbonyl)-S-meth-
ylisothiourea after experimental protocol entry 8, Table 1. Crude yields (%) were
estimated based on mass spectrometry detection of the intermediate: Fmoc-Arg(Pbf)-
agmatine. RT represents room temperature. Experimental details can be found in the
Experimental procedures section
Product
AA
Yield (%)
Product
AA
Yield (%)
1
2
3
4
5
Gly
Ala
Val
Leu
Nle
35
8
3
6
4
10
11
12
13
14
Gln
Ser
Orn
Lys
8
8
9
10
13
Entry Temp (°C) Solvent Guanylating Microwave Time
Crude
D-Lys
reagent
(min) yield (%)
6
7
8
9
Pro
Asp
Glu
Asn
10
7
9
15
16
17
18
Arg
His
Trp
Phe
9
9
7
18
(mol equiv)
1
2
3
4
RT
60
60
60
DMF
DMF
DMF
CH2Cl2
10
10
5
No
60
10
10
10
ꢁ50
>80
ꢁ50
ꢁ70
Yes
Yes
Yes
9
10
milligram quantities, but sufficient for our enzyme inhibition
screening regime. Our first target, compound 1, was obtained in
an overall yield of 35%. Closer data analysis suggested that the syn-
thesis of amino acids with branched alkyl side-chains (compounds
3–5) was not facile (3–6% overall yields), while amino acids with
polar and/or charged side-chains (compounds 7–15) provided bet-
ter overall yields (7–13%). We attributed this to solubility issues as
compounds 3–5 were relatively more hydrophobic. Amino acids
with aromatic side-chains (compounds 16 and 17) resulted in 7–
9% yields. Interestingly, compound 18 with a hydrophobic phenyl-
alanine gave an unexpectedly high yield of 18%.
In conclusion, we have successfully developed a novel solid-
phase BAL synthetic strategy which enabled us to synthesize rap-
idly 18 C-terminal agmatine peptides using commercially available
reagents in 3–35% overall yields. A major advantage of this ap-
proach over solution-phase chemistry was that it eliminated the
need to purify intermediates by silica gel chromatography, thereby
significantly reducing costs and organic solvent consumption. We
believe that this method can be employed to prepare libraries of
C-terminal agmatine peptide inhibitors of trypsin-like serine pro-
teases with various amino acids at the P2–P4 regions. Such li-
braries can be screened to probe enzyme specificities and aid in
the design of new drugs.
Table 3
Amino acid coupling conditions after experimental protocol entry 2, Table 2. Crude
yields (%) were estimated based on mass spectrometry detection of the intermediate:
Fmoc-Arg(Pbf)-agmatine. RT represents room temperature. Experimental details can
be found in the Experimental procedures section
Entry Temp (°C) Solvent
Coupling
reagent
(5 mol eq)
Microwave Time Crude
(min) yield (%)
1
2
3
4
5
RT
RT
60
60
60
CH2Cl2
CH2Cl2
CH2Cl2
DMF
HBTU
DIC
DIC
no
no
yes
yes
yes
60
60
10
10
10
ꢁ10
ꢁ40
ꢁ70
ꢁ20
ꢁ70
DIC
CH2Cl2:DMF DIC
(9:1 v/v)
CH2Cl2:DMF DIC + HOBt yes
(9:1 v/v) (1:1)
6
60
10
>90
guanylating reagents have also been reported12 and we suspect
this was due to steric effects.
The next stage involved various reactions to determine the opti-
mal amino acid coupling conditions (Table 3). The solvent system
CH2Cl2:DMF (9:1 v/v) was found to be optimal for dissolving the
reagents and the addition of hydroxybenzotriazole (HOBt) to the
symmetrical anhydride N,N0-diisopropylcarbodiimide (DIC) cou-
pling reagent (1:1) gave the best yield for the intermediate:
Fmoc-Arg(Pbf)-agmatine. Coupling of the Fmoc-protected amino
acid to the secondary amine was repeated to ensure maximum
coupling.
Using the optimized reaction conditions (Scheme 1), we were
able to synthesize rapidly 18 agmatine dipeptides in milligram
quantities with overall yields of 3–35% using standard SPPS tech-
niques (Table 4).
Experimental procedures. The aldehyde-functionalized DFPE re-
sin (0.05 mmol, 58 mg) was stirred and allowed to swell in DMF
(5 mL) at 25 °C for 1 h. Next, a pre-mixed solution of 1,4-diamin-
obutane (0.5 mmol, 51 lL) and NaCNBH3 (0.5 mmol, 32 mg) in 1%
AcOH in DMF (v/v, 5 mL) was added to the resin. The mixture
was subjected to microwave irradiation while stirring at 60 °C for
10 min. This step was repeated. Unreacted reagents were removed
by filtration and the resin washed with DMF (15 ꢀ 3 mL) and
CH2Cl2 (15 ꢀ 3 mL) followed by DMF (15 mL). Next, a pre-mixed
solution of N,N0-bis(tert-butoxycarbonyl)thiourea (0.5 mmol,
146 mg), N,N-diisopropylethylamine (DIPEA, 0.5 mmol, 88 lL)
It must be emphasized that the overall percentage yields were
calculated using the manufacturer-stated resin loading factor of
0.87 mmol/g for the final lyophilized product after HPLC purifica-
tion over the 9-step process. Starting with 0.05 mmol (58 mg) of
resin, we were able to obtain our HPLC-purified targets in low
and DMF (5 mL) was added to the resin and the reaction mixture
subjected to microwave irradiation while stirring at 60 °C for
10 min. Unreacted reagents were removed by filtration and the re-
sin washed with DMF (15 ꢀ 3 mL) and CH2Cl2 (15 ꢀ 3 mL). A pre-
mixed solution of Fmoc-Lys(Boc)-OH (0.5 mmol, 235 mg), DIC

5514
M. J. Y. Ang et al. / Tetrahedron Letters 53 (2012) 5511–5514
(0.25 mmol, 27
lL), and HOBt (0.25 mmol, 33.8 mg), dissolved in
4.21–4.31 (2H, m, a
-H), 7.30–7.62 (9H, m, aromatics); 13C NMR
90% CH2Cl2 in DMF (v/v, 5 mL) was added to the resin. The mixture
was subjected to microwave irradiation while stirring at 60 °C for
10 min. This step was repeated. Unreacted reagents were removed
by filtration and the resin washed with DMF (15 ꢀ 3 mL) and
CH2Cl2 (15 ꢀ 3 mL) followed by DMF (15 ꢀ 3 mL). 20% piperidine
in DMF (v/v, 5 mL) was added to the resin. The mixture was sub-
jected to microwave irradiation while stirring at 60 °C for 5 min.
Unreacted reagents were removed by filtration and the resin
washed with DMF (15 ꢀ 3 mL) and CH2Cl2 (15 ꢀ 3 mL). A pre-
mixed solution of an appropriate Fmoc-protected amino acid
(100 MHz, CD3OD) d 22.4, 22.5, 25.7, 26.0, 26.5, 26.7, 28.1, 30.0,
38.4, 39.1, 40.7, 41.5, 53.1, 53.9, 126.4, 126.8, 127.1, 128.6, 129.3,
134.5, 139.8, 140.5, 157.3, 172.7, 173.0, 173.2. ESI-TOF-MS: m/z
calcd C31H48N8O3 (M+H+) 581.3928, found 583.3832.
Acknowledgments
We thank A⁄STAR Biomedical Research Council for financial
support and Dr. Manfred Raida for mass spectrometric analysis.
(0.5 mmol), DIC (0.25 mmol, 27 lL), and HOBt (0.25 mmol,
Supplementary data
33.8 mg), dissolved in 90% CH2Cl2 in DMF (v/v, 5 mL) was added
to the resin. The mixture was subjected to microwave irradiation
while stirring at 60 °C for 10 min. Unreacted reagents were re-
moved by filtration and the resin washed with DMF (15 ꢀ 3 mL)
and CH2Cl2 (15 ꢀ 3 mL). 20% Piperidine in DMF (v/v, 5 mL) was
added to the resin. The mixture was subjected to microwave irra-
diation while stirring at 60 °C for 5 min. Unreacted reagents were
removed by filtration and the resin washed with DMF
(15 ꢀ 3 mL) and CH2Cl2 (15 ꢀ 3 mL). A pre-mixed solution of 4-
General experimental procedures and high-resolution mass
spectroscopic data for all compounds and supplementary data
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.tetlet.2012.08.002.
References and notes
1. Wiley, M. R.; Chirgadze, N. Y.; Clawson, D. K.; Craft, T. J.; Gifford-Moore, D. S.;
Jones, N. D.; Olkowski, J. L.; Schacht, A. L.; Weir, L. C.; Smith, G. F. Bioorg. Med.
Chem. Lett. 1995, 23, 2835.
biphenylacetic acid (0.5 mmol, 106 mg), DIC (0.25 mmol, 27 lL),
and HOBt (0.25 mmol, 33.8 mg), dissolved in 90% CH2Cl2 in DMF
(v/v, 5 mL) was added to the resin. The reaction mixture was sub-
jected to microwave irradiation while stirring at 60 °C for 10 min.
Unreacted reagents were removed by filtration and the resin
washed with DMF (15 ꢀ 3 mL) and CH2Cl2 (15 mL). The target
agmatine peptidomimetic was cleaved from the resin using 95%
TFA in CH2Cl2 (v/v, 1.5 mL) at 25 °C over 30 min. Excess TFA was re-
moved by a gentle stream of N2(g) to give a brown oil which was
dissolved in MeOH (1 mL) and purified by HPLC (H2O and MeCN
solvent). The target compound was lyophilized overnight in vacuo
to give an off-white powder (3–35%).
2. Becker, G. L.; Sielaff, F.; Than, M. E.; Lindberg, I.; Routhier, S.; Day, D.; Lu, Y.;
Garten, W.; Steinmetzger, T. J. Med. Chem. 2010, 53, 1067.
3. Lim, H. A.; Joy, J.; Hill, J.; Chia, C. S. B. Eur. J. Med. Chem. 2011, 46, 3130.
4. Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun. 1967, 27, 157.
5. Knox, J. E.; Ma, N. L.; Yin, Z.; Patel, S. J.; Wang, W.-L.; Chan, W.-L.; Rao, K. R. R.;
Wang, G.; Ngew, X.; Patel, V.; Beer, D.; Lim, S. P.; Vasudevan, S. G.; Keller, T. H. J.
Med. Chem. 2006, 49, 6585.
6. Stoermer, M. J.; Chappell, K. J.; Liebscher, S.; Jensen, C. M.; Gan, C. H.; Gupta, P.
K.; Xu, W.-J.; Young, P. R.; Fairlie, D. P. J. Med. Chem. 2008, 51, 5714.
7. Bernhardt, A.; Drewello, M.; Schutkowski, M. J. Pept. Res. 1997, 50, 153.
8. Zhong, H. M.; Greco, M. N.; Maryanoff, J. D. J. Org. Chem. 1997, 62, 9326.
9. García, O.; Nicolás, E.; Albericio, F. Tetrahedron Lett. 2003, 44, 5319.
10. Jensen, K. J.; Alsina, J.; Songster, M. F.; Vágner, J.; Albericio, F.; Barany, G. J. Am.
Chem. Soc. 1998, 120, 5441.
Spectral data of compound 13: 1H NMR (400 MHz, CD3OD) d
1.29–1.96 (16H, m), 2.82–3.23 (8H, m), 3.64 (2H, s, Ar–CH2–CO–),
11. Boas, U.; Brask, J.; Jensen, K. J. Chem. Rev. 2009, 109, 2092.
12. Castagnolo, D.; Schenone, S.; Botta, M. Chem. Rev. 2011, 111, 5247.