Synthetic strategy for bicyclic tetrapeptides HDAC inhibitors using ring closing metathesis

Article abstract of DOI:10.1007/s12039-015-0922-y

Cyclic peptides show diverse biological activities and are considered as good therapeutic agents due to structural rigidity, receptor selectivity and biochemical stability. We have developed bicyclic tetrapeptide HDAC inhibitors based on different cyclic

Full text of DOI:10.1007/s12039-015-0922-y

c
J. Chem. Sci. Vol. 127, No. 9, September 2015, pp. 1563–1569. ꢀ Indian Academy of Sciences.
DOI 10.1007/s12039-015-0922-y
Synthetic strategy for bicyclic tetrapeptides HDAC inhibitors using ring
closing metathesis
MD NURUL ISLAMa,b,∗, MD SHAHIDUL ISLAMa,b, MD ASHRAFUL HOQUEa,c,
TAMAKI KATOa and NORIKAZU NISHINOa
aGraduate School of Life Science and Systems Engineering, Kyushu Institute of Technology,
Kitakyushu 808-0196, Japan
bDepartment of Chemistry, Faculty of Science, University of Rajshahi, Rajshahi 6205, Bangladesh
cDepartment of Biochemistry and Molecular Biology, Faculty of Science, University of Rajshahi,
Rajshahi 6205, Bangladesh
e-mail: mnurulchem@gmail.com
MS received 9 February 2015; revised 27 May 2015; accepted 27 May 2015
Abstract. Cyclic peptides show diverse biological activities and are considered as good therapeutic agents
due to structural rigidity, receptor selectivity and biochemical stability. We have developed bicyclic tetrapeptide
HDAC inhibitors based on different cyclic tetrapeptide scaffolds. For the synthesis of these bicyclic tetrapep-
tides, two cyclization steps, namely, peptide cyclization and fusion of aliphatic side chains by ring closing
metathesis (RCM) were involved. In the course of these syntheses, we have established two facts: a lower limit
of aliphatic loop length and better synthetic route for bicyclic tetrapeptide synthesis. It was found that nine
methylene loop length is the lower limit for aliphatic loop and the synthetic route selection depended on the
configuration of amino acids in the cyclic tetrapeptide scaffold. RCM followed by peptide cyclization was the
proper route for LDLD configuration and the reverse route was suitable for LLLD configuration.
Keywords. Bicyclic tetrapeptides; ring-closing metathesis; aliphatic loop length; HDAC inhibitors.
1. Introduction
of aliphatic loop length for RCM and the order of the
two cyclization steps involved in bicyclic tetrapeptide
synthesis.
Cyclic peptides show a wide spectrum of biological
activity, such as antibacterial activity, immunosuppres-
sive activity, anti-tumor activity, and so on.1–5 Structural
rigidity, receptor selectivity and biochemical stability
are general features of cyclic peptides. These features
allow cyclic peptides to be good therapeutic agents, and
efforts have been made to develop synthetic cyclic and
bicyclic peptides with biological activity.5,6
Ring closing metathesis (RCM) have recently appeared
as a versatile, chemo-selective means to create carbon-
carbon bonds in a range of molecules including natu-
ral products, polymers, and biomolecules.7–14 We also
designed and synthesized cyclic tetrapeptides Cyl-1,
CHAP31, trapoxin B and HC-toxin I (figure 1) based
bicyclic tetrapeptide HDAC inhibitors using RCM.15–18
For the synthesis of these bicyclic tetrapeptides, two
cyclization steps i.e., peptide cyclization and fusion of
aliphatic side chains by RCM were involved. Selection
of the synthetic route depended on the configuration of
the amino acids whose side chains take part in RCM.
We herein describe the establishment of a lower limit
2. Experimental
2.1 General
Unless otherwise noted, all solvents and reagents
were reagent grade and used without purification.
Flash chromatography was performed using silica gel
60 (230–400 mesh) eluting with solvents as indi-
cated. All compounds were routinely checked by
thin layer chromatography (TLC) and/or high perfor-
mance liquid chromatography (HPLC). TLC was per-
formed on aluminum-backed silica gel plates (Merck
DC-Alufolien Kieselgel 60 F254) with spots visual-
ized by UV light or charring. Analytical HPLC was
performed on a Hitachi instrument equipped with
a chromolith performance RP-18e column (4.6 ×
100 mm, Merck). The mobile phases used were A:
H2O with 10% CH3CN and 0.1% TFA, B: CH3CN
with 0.1% TFA using a solvent gradient of A-
B over 15 min with a flow rate of 2 mL/min,
with detection at 220 nm. HR FAB-mass spectra
∗For correspondence
1563

1564
Md Nurul Islam et al.
concentrated to remain an oily substance, which was
purified by silica gel chromatography using a mixture
of chloroform and methanol (99:1) to yield Boc-L-Ae8-
D-Pro-OtBu (1.60 g, 78%) as an oil. The protected
dipeptide (1.55 g, 3.8 mmol) was dissolved in 4 M
HCl/dioxane (16 mL) and the mixture was kept at room
temperature for 30 min. The reaction was monitored by
TLC. After completion of the reaction HCl/dioxane was
evaporated. The residue was dissolved in EtOAc and
washed with saturated Na2CO3, dried over anhydrous
Na2CO3 and EtOAc was evaporated to get H-L-Ae8-D-
Pro-OtBu as a heavy oil (0.930 g, 79%). To a cooled
solution of H-L-Ae8-D-Pro-OtBu (0.930 g, 3 mmol),
Boc-D-Ae8-OH (0.940 g, 3.3 mmol) and HOBt.H2O
(0.460 g, 3 mmol) in DMF (6 mL), DCC (0.742 g, 3.6
mmol) was added and stirred for 12 h at room tempera-
ture. The product Boc-D-Ae8-L-Ae8-D-Pro-OtBu was
obtained in the same manner as described earlier as a
heavy oil (1.40 g, 84%). HPLC, retention time (tR) 8.02
min. To a solution of linear tripeptide Boc-D-Ae8-L-
Ae8-D-Pro-OtBu (1.10 g, 2 mmol) in anhydrous and
degassed dichloromethane (DCM) (250 mL), a solution
of Grubbs first generation ruthenium catalyst (0.330 g,
0.4 mmol) in anhydrous and degassed DCM (50 mL)
was added. The reaction mixture was stirred at room
temperature for 48 h. After the completion of reac-
tion, DCM was evaporated and the residue was purified
by silica gel chromatography using a mixture of chlo-
roform and methanol (99:1) to yield linear tripeptide
with unsaturated fused ring as a foam which on cat-
alytic hydrogenation in presence of Pd-C (0.100 g) in
acetic acid (10 mL) yield both side protected tripeptide
with saturated fused ring (0.920 g, 87%). HPLC, tR 7.96
min. The protected tripeptide (0.530 g, 1 mmol) was
dissolved in 4 M HCl/dioxane (5 mL) and the mixture
was kept at room temperature for 30 min. The reaction
was monitored by TLC. After completion of the reac-
tion HCl/dioxane was evaporated. The residue was dis-
solved in EtOAc and washed with saturated Na2CO3,
dried over anhydrous Na2CO3 and EtOAc was evap-
orated to get tripeptide free amine (0.310 g, 73%).
To a cooled solution of the free amine (0.310 g, 0.73
mmol), Boc-L-Asu(OBzl)-OH (0.295 g, 0.78 mmol)
and HOBt.H2O (0.112 g, 0.73 mmol) in DMF (2 mL),
DCC (0.161 g, 0.78 mmol) was added and stirred for
12 h at room temperature. The product linear tetrapep-
tide with fused ring (0.450 g, 79%) was obtained in
the same manner as described earlier as a foam. HPLC,
tR 10.42 min. The protected tetrapeptide (0◦.445 g, 0.57
mmol) was dissolved in TFA (3 mL) at 0 C and kept
for 3 h. After evaporation of TFA, the residue was solid-
ified using ether and petroleum ether to yield TFA salt
of the linear tetrapeptide (0.425 g, 100%). To DMF
Figure 1. Some natural and synthetic cyclic tetrapeptide
HDAC inhibitors.
were measured on a JEOL JMS-SX 102A instrument.
NMR spectra were recorded on a JEOL JNM A500
MHz spectrometer in DMSO-d6 solutions with refer-
ence to TMS. All 1H shifts are given in parts per million
(s = singlet; d = doublet; t = triplet; m = multiplet).
Assignments of proton resonances were confirmed,
when possible, by correlated spectroscopy.
Ring closing metathesis was performed by the aid of
benzylidene-bis(tricyclohexylphosphine)dichlororuthen-
ium (Grubbs’first generation ruthenium catalyst).Coupling
reactions were performed by standard solution-
phase chemistry using dicyclohexyl-carbodiimide
(DCC) and 1-hydroxybenzotriazol (HOBt). Peptide
cyclization was mediated by N−[(dimethylamino)-
1H-1,2,3-triazolo[4,5-b]pyridin-1-yl-methylene]-N-
methylmethanaminium hexafluorophosphateN-oxide
(HATU). The artificial amino acids Boc-L-2-amino-
6-heptenoic acid (Boc-L-Ae7-OH), Boc-L-2-amino-
7-octenoic acid (Boc-L-Ae8-OH), Boc-L-2-amino-8-
nonenoic acid (Boc-L-Ae9-OH), Boc-D-2-amino-7-
octenoic acid (Boc-D-Ae8-OH) and Boc-D-2-amino-8-
nonenoic acid (Boc-D-Ae9-OH) used for the synthesis
of bicyclic tetrapeptides were synthesized by the
reported procedure.12
2.1a Synthesis of bicyclic tetrapeptide hydroxamic
acid with ten CH2 loop length (4): To a cooled solu-
tion of H-D-Pro-OtBu (0.855 g, 5 mmol), Boc-L-Ae8-
OH (1.29 g, 5 mmol) and HOBt.H2O (0.766 g, 5 mmol)
dimethylformamide (DMF) (10 mL), DCC (1.24 g, 6
mmol) were added. The mixture was stirred for 12 h
at room temperature. After completion of the reaction,
DMF was evaporated and the residue was dissolved
in ethyl acetate (EtOAc) and successively washed with
10% citric acid, 4% sodium bicarbonate and brine. The
EtOAc solution was dried over anhydrous MgSO4 and

Synthetic strategy for bicyclic tetrapeptides
1565
solvent (570 mL), the TFA salt (0.425 g, 0.57 mmol), 2.1c Synthesis of bicyclic tetrapeptide hydroxamic
HATU (0.326 g, 0.86 mmol), and DIEA (0.24 mL, 1.42 acid with eleven CH2 loop length (5): Compound 5
mmol) were added in separate five portions in every (0.195 g, yield: 75%, HPLC, tR 7.39 min) was synthe-
30 min with stirring for the cyclization reaction. After sized following the same procedure as described in case
completion of the reaction, DMF was evaporated under of compound 4 using Boc-L-Ae9-OH instead of Boc-
1
vacuum; the residue was dissolved in ethyl acetate and L-Ae8-OH. H NMR (500 MHz, DMSO-d6 30 C): δH
washed with citric acid (10%) solution, sodium bicar- 1.10–1.99 (m, 32H), 2.50 (m, 4H), 3.57 (dd, J = 18.2,
bonate (4%) solution, and brine, successively. It was 8.4 Hz, 1H), 3.86 (ddd, J = 9.5, 9.5, 3.8 Hz, 1H), 4.19
then dried over anhydrous MgSO4 and filtered. After (dd, J = 16.7, 7.7 Hz, 1H), 4.41 (m, 1H), 4.53 (d, J =
evaporation of ethyl acetate, the residue was purified by 7.5 Hz, 1H), 4.70 (d, J = 7.6 Hz, 1H), 7.13 (d, J = 9.1
silica gel chromatography using a mixture of chloro- Hz, 1H), 7.23 (d, J = 9.8 Hz, 1H), 8.42 (d, J = 9.2 Hz,
form and methanol (99:1) to yield the bicyclic tetrapep- 1H), 8.61 (s, 1H), 10.31 (s, 1H); HR FAB-MS [M+H]+
tide (0.143 g , 41%). HPLC, tR 7.85 min. The bicyclic 550.3603 for C28H48N5O6 (calcd 550.3605).
tetrapeptide (0.125 g, 0.2 mmol) was dissolved in acetic
2.1d Synthesis of bicyclic tetrapeptide hydroxamic
acid with twelve CH2 loop length (6): Compound 6
(0.195 g, yield: 71%, HPLC, tR 7.52 min) was synthe-
sized following the same procedure as described in case
of compound 4 using Boc-D-Ae9-OH instead of Boc-
acid (5 mL) and Pd-C (0.050 g) was added. The mixture
was stirred under H2 for 10 h. After filtration of Pd-C,
acetic acid was evaporated to yield bicyclic tetrapeptide
carboxylic acid (0.104 g, 100%). The bicyclic tetrapep-
tide carboxylic acid (0.090 g, 0.17 mmol) was dissolved
in DMF (0.5 mL) at 0C, and O-benzylhydroxylamine
hydrochloride (0.041 g, 0.26 mmol), HOBt.H2O (0.026
g, 0.17 mmol), triethylamine (0.037 mL, 0.26 mmol)
and DCC (0.054 g, 0.26 mmol) were added. The mix-
ture was stirred for 24 h. The product (0.062 g, 60%)
obtained was dissolved in acetic acid (1 mL), and Pd-
BaSO4 (0.020 g) was added. The mixture was stirred
under H2 for 24 h. After filtration of Pd-BaSO4, acetic
acid was evaporated and crystallized with ether to yield
compound 4 (0.036 g, yield: 70%, HPLC, tR 6.03 min).
1H NMR (500 MHz, DMSO-d6, 40 C): δH 1.14–1.28
(m, 16H), 1.31–1.38 (m, 4H), 1.42–1.54 (m, 6H), 1.64–
1.79 (m, 4H), 1.85–2.00 (m, 4H), 3.38–3.45 (m, 2H),
3.57 (dd, J = 18, 8.2 Hz, 1H), 3.92 (ddd, J = 9.5,
9.5, 3.5 Hz, 1H), 4.21 (m, 1H), 4.44 (m, 1H), 4.58 (d,
J = 7.9 Hz, 1H), 4.70 (m, 1H), 7.13 (d, J = 9.0
Hz, 1H), 7.26 (d, J = 10.1 Hz, 1H), 8.29 (d,J =
9.1 Hz, 1H), 8.64 (s, 1H), 10.33 (s, 1H); HR FAB-
MS [M+Na]+ 558.3263 for C27H45N5O6Na (calcd
558.3268).
1
D-Ae8-OH. H NMR (500 MHz, DMSO-d6 30 C): δH
1.17–2.10 (m, 34H), 2.50 (m, 4H), 3.41 (m, 1H), 3.58
(dd, J = 17.7, 8.7 Hz, 1H), 3.74 (m, 1H), 3.81 (ddd,
J = 9.7, 9.5, 3.5 Hz, 1H), 4.06 (m, 1H), 4.19 (dd, J =
16.5, 7.5 Hz, 1H), 4.34 (t, J = 10.7 Hz, 1H), 4.55 (d,
J = 8.2 Hz, 1H), 4.70 (d, J = 7.3 Hz, 1H), 7.16 (d, J =
9.5 Hz, 1H), 7.23 (d,J = 10.0 Hz, 1H), 8.44 (d, J =
9.2 Hz, 1H), 8.62 (s, 1H), 10.34 (s, 1H); HR FAB-MS
[M+H]+ 564.3691 for C29H50N5O6 (calcd 564.3761).
2.1e Synthesis of HC-toxin-I and trapoxin based
bicyclic tetrapeptide hydroxamic acid (7 and 8):
Detailed synthetic procedures for these compounds
have been published.17
2.1f Synthesis of CHAP31 based cyclic tetrapeptide
hydroxamic acid (9): Compound 9 (0.250 g, yield:
91%, HPLC, tR 7.51 min) was synthesized by the same
solution phase synthetic method described for com-
pound 4 using Boc-L-Ae7-OH and Boc-D-Ae7-OH in
place of Boc-L-Ae8-OH and Boc-D-Ae8-OH respec-
tively, and by skipping RCM step. 1H NMR (500 MHz,
DMSO-d6) δH 0.81–0.88 (m, 6H),1.18-1.30 (m, 18H),
1.44 (m, 4H), 1.62–1.74 (m, 4H), 1.87–1.97 (m, 4H),
3.52 (dd, = 18.5, 8.5 Hz, 1H), 3.77 (ddd, J = 10, 9.5, 4
Hz, 1H), 4.19–4.27 (m, 2H), 4.63 (dd, J = 17.3, 8 Hz,
1H), 4.71 (d, J = 7.3 Hz, 1H), 7.06 (d, J = 10.1 Hz,
1H), 7.21 (d, J = 9.5 Hz, 1H), 8.43 (d, J = 9.2 Hz,
1H), 8.63 (s, 1H), 10.31 (s, 1H); HR FAB-MS [M+H]+
538.3543 for C27H48N5O6 (calcd 538.3605).
2.1b Synthesis of bicyclic tetrapeptide hydroxamic
acid with nine CH2 loop length (3): Compound 3
(0.010 g, yield: 73%, HPLC, tR 4.74 min) was synthe-
sized following the same procedure as described in case
of compound 4 using Boc-L-Ae7-OH instead of Boc-
L-Ae8-OH. 1H NMR (500 MHz, DMSO-d6): δH 1.15–
1.40 ( m, 18H), 1.47-1.78 (m, 10H), 1.83–1.98 (m, 4H),
3.60 (dd, J = 17.5, 8.7 Hz, 1H), 3.85 (ddd, J = 9.7,
9.5, 3.5 Hz, 1H), 4.06 (m, 1H), 4.13 (m, 1H), 4.20 (dd,
J = 17.3, 8 Hz, 1H), 4.61 (d, J = 7.9 Hz, 1H), 7.18
(d, J = 9.5, 1H), 7.91 (s, 1H), 8.09 (s, 1H), 8.66 (s,
1H), 10.36 (s, 1H); HRFAB-MS [M+Na]+ 544.3141
for C26H43N5O6Na (calcd 544.3111).
2.1g Synthesis of HC-toxin-I and trapoxin based
cyclic tetrapeptide hydroxamic acid (10 and 11):
Detailed synthetic procedures of these compounds have
been published.17

1566
Md Nurul Islam et al.
After successful synthesis of CHAP31 based bicyclic
2.2 Circular dichroism (CD) measurements
tetrapeptide with ten methylene loop length (4) through
RCM followed by peptide cyclization (scheme 1), we
reduced the loop length to nine methylene units and
synthesized bicyclic tetrapeptide (3). However, synthe-
sis of bicyclic tetrapeptide with eight methylene loop
length (2) failed in the RCM step. Instead of obtaining
expected linear tripeptide with fused ring, two dimers
with two different HPLC retention times were obtained
corresponding to parallel and anti-parallel arrangement
of the monomer during RCM (figure 3).
CD spectra were recorded on a JASCO J-820 spectropo-
larimeter using a quartz cell of 1 millimeter path length
at room temperature. Peptide solutions (0.1 mM) were
prepared in methanol and CD spectra were recorded in
terms of molar ellipticity, [θ]M (deg cm2 dmol−1).
3. Results and Discussion
3.1 Design and synthesis
The dimers were confirmed by LC-MS ([M+H]+
987.4, calcd. 987.64). We changed the route and syn-
thesized cyclic tetrapeptide first and then RCM was
applied to obtain bicyclic tetrapeptide (scheme 2). Once
again, we obtained dimer instead of monomer. As the
bicyclic tetrapeptides having nine or more methylene
units were successively synthesized, nine methylene
units is the minimum loop length for bicyclic tetrapep-
tides using Grubbs first generation ruthenium catalyst.
We have also synthesized two more CHAP31 based
bicyclic tetrapeptides with eleven (5) and twelve (6)
methylene loop length to optimize the cyclization yield
and to check the better route for bicyclic tetrapeptide
synthesis.
For CHAP31 based bicyclic tetrapeptide series, RCM
and peptide cyclization yields were affected by the
aliphatic loop length (table 1). In general, both RCM
and peptide cyclization yields were increased with
increasing aliphatic loop length. This may be due to
the increase in flexibility of loop that facilitates the
RCM and peptide cyclization. As for compound 5 and
6, RCM and peptide cyclization yields were almost
the same, eleven methylene loop length was sufficient
to remove stress imposed by the loop. Our previous
report16 on these compounds showed that the loop
length has a significant impact on in vivo biological
We first designed Cyl-1-based (LLLD configuration)
bicyclic tetrapeptide with ten methylene loop length
(1) and tried to synthesize through RCM followed by
peptide cyclization but failed in the peptide cycliza-
tion step; whereas, peptide cyclization was successful
in case of cyclic tetrapeptide with the same configura-
tion. Komatsu et al. also synthesized cyclic tetrapeptide
CHAP30 with the same configuration.19 We assumed
that cyclization of Cyl-1 based tetrapeptide with
unfavourable configuration (LLLD) is impossible due
to the steric hindrance imposed by the aliphatic loop.
By changing the configuration, we successfully synthe-
sized CHAP31-based (LDLD configuration) bicyclic
tetrapeptide with the same loop length of ten methy-
lene units (4). Subsequently, a series of CHAP31-based
bicyclic tetrapeptides with different loop lengths (2, 3, 5
and 6) were designed to determine the lower limit of the
aliphatic loop length and favourable route for bicyclic
tetrapeptide synthesis. Finally, we designed HC-toxin
and trapoxin B based bicyclic tetrapeptides with ten
methylene loop length (7 and 8, figure 2) to confirm the
favourable synthetic route. We also designed and syn-
thesized compounds 9, 10 and 11 as open chain ref-
erence of compounds for 4, 7 and 8, respectively, for
comparing peptide backbone conformational changes.
Figure 2. Designed bicyclic and cyclic tetrapeptides.

Synthetic strategy for bicyclic tetrapeptides
1567
Scheme 1. Synthesis of CHAP31 based bicyclic tetrapeptide with ten methy-
lene loop length (4). Reagents and conditions: (a) Grubbs first generation cat-
alyst, DCM, 48 h; (b) AcOH, Pd-C, H2, 12 h; (c) 4 M HCl/dioxane, 30 min;
(d) Boc protected amino suberic acid benzyl ester (Boc-L-Asu(OBzl)-OH),
DCC, HOBt, DMF, 12 h; (e) TFA, 3 h; (f) HATU, DIEA, DMF, 4 h; (g)
HCl.H2NOBzl, DCC, HOBt, TEA, DMF, 24 h; (h) Pd-BaSO4, AcOH, H2.
activity. A sharp increase in p21 promoter-inducing In this case, the side chains are directed to the
activity was observed with the increase in loop length opposite sides of the cyclic tetrapeptide scaffolds which
from nine to eleven methylene groups. The change in is unfavourable for RCM. Therefore, RCM followed
biological activity may be due to the change in both the by peptide cyclization can be considered as the proper
hydrophobicity and stress on cyclic tetrapeptide back- synthetic route for bicyclic tetrapeptides with LDLD
bone. To check the better route for bicyclic tetrapeptide configuration. The successful synthesis with apprecia-
synthesis, compound 6 was synthesized through a dif- ble yields in peptide cyclization and RCM steps for
ferent route i.e., through peptide cyclization followed compound 7 (LDLD configuration, reverse to CHAP31)
by RCM. In this case, the RCM yield was remarkably supports our proposed route.
low (40%). This may be due to the spatial arrange-
As mentioned above, LLLD configuration is
ment of the two aliphatic side chains of the adjacent unfavourable for peptide cyclization; RCM before
amino acids (LD configuration) taking part in RCM. peptide cyclization imposes an extra hindrance for the
Figure 3. Dimers obtained from RCM and their HPLC profiles.

1568
Md Nurul Islam et al.
Scheme 2. Determination of lower limit for aliphatic loop length of bicyclic
tetrapeptides. Reagents and conditions: (a) Grubbs first generation catalyst,
DCM, 48 h; (b) 4 M HCl/dioxane, 30 min; (c) saturated Na2CO3; (d) Boc-
L-Asu(OBzl)-OH, DCC, HOBt, DMF; (e) TFA, 3 h; (f) HATU, DIEA, DMF,
4 h.
Table 1. Effect of loop length and synthetic route on RCM and peptide cyclization.
Compd. No.
Origin
Configuration
Loop size
Route
RCM yield (%)
PC yield (%)
1
2
3
4
5
6
6
7
8
Cyl-1
LLLD
LDLD
LDLD
LDLD
LDLD
LDLD
LDLD
LDLD
LLLD
-(CH2)10-
-(CH2)8-
-(CH2)9-
-(CH2)10-
-(CH2)11-
-(CH2)12-
-(CH2)12-
-(CH2)10-
-(CH2)10-
RCM→PC∗
RCM→PC
RCM→PC
RCM→PC
RCM→PC
RCM→PC
PC→ RCM
RCM→PC
PC→ RCM
90
Failed
50
87
85
85
40
63
40
Failed
–
15
40
65
67
65
51
41
CHAP31
CHAP31
CHAP31
CHAP31
CHAP31
CHAP31
HC-toxin I
Trapoxin B
∗PC: Peptide cyclization
Figure 4. CD spectra of (a) CHAP31 analogs; (b) HC-toxin analogs and (c) Trapoxin B analogs.
latter step. As a result, peptide cyclization becomes peptide cylcization step for Cyl-1 based compound 1
almost impossible whatever be the configuration of the is an important evidence in favour of this speculation.
adjacent amino acid bearing the aliphatic side chains. In such cases, bicyclic tetrapeptides are expected
The success in RCM followed by the failure in the to be obtained if RCM is carried out after peptide

Synthetic strategy for bicyclic tetrapeptides
1569
cyclization. Moreover, in this route, there is an extra Supplementary Information
advantage if the aliphatic side chain bearing amino
The electronic supplementary information (NMR spec-
tra, figures S1-S5) is included which is available at
www.ias.ac.in/chemsci.
acids have the same configuration (LL or DD). The
side chains are positioned in the same side of the cyclic
tetrapeptide scaffolds which make the RCM step eas-
ier. Therefore, RCM after peptide cyclization can be
considered as the proper synthetic route for bicyclic
tetrapeptides with LLLD configuration. The successful
synthesis with appreciable yields in peptide cyclization
and RCM steps for compound 8 (LLLD configuration)
strengthen our proposed route.
Acknowledgement
This study was supported by Kitakyushu Foundation
for the Advancement of Industry, Science and Technol-
ogy (FAIS) and Japan Student Services Organization
(JASSO).
3.2 Conformational analysis by CD
References
CD is a sensitive, qualitative tool for monitoring pep-
tide conformational changes.20 CD measurements of the
synthesized compounds were carried out in methanol as
solvent for analyzing peptide backbone conformation.
For CHAP31 and HC-toxin analogs (LDLD configura-
tion), the difference in CD spectra is quite remarkable
between open and fused chain compounds. In case of
CHAP31 analogs, the negative CD band near 240 nm
disappeared and the positive band near 228 nm was
shifted to longer wavelength (figure 4a), while the pos-
itive band near 250 nm shifted to longer wavelength
for HC-toxin analogs upon ring closing (figure 4b). In
case of trapoxin B analogs (LLLD configuration), sim-
ilar CD spectra for both open and closed chain com-
pounds were observed (figure 4c). It implies that ring
closing imposes a stress on the peptide backbone of
CHAP31 and HC-toxin analogs (LDLD configuration)
but not on that of trapoxin B analogs (LLLD configura-
tion) with the same loop length. Therefore, RCM faced
hindrance in case of cyclic tetrapeptides with LDLD
configuration.
1. Yokoyama F, Suzuki N, Haruki M, Nishi N, Oishi S,
Fujii N, Utani A, Kleinman H K and Nomizu M 2004
Biochemistry 43 13590
2. Hruby V J 2001 Acc. Chem. Res. 34 389
3. Bolla M L, Azevedo E V, Smith J M, Taylor R E, Ranjit
D K, Segall A M and McAlpine S R 2003 Org. Lett. 5
109
4. Humphrey J M and Chamberlin A R 1997 Chem. Rev.
97 2243
5. Joo S H 2012 Biomol. Ther. 20 19
6. Baeriswyl V and Heinis C 2013 Chem. Med. Chem. 8
377
7. Grubbs R H 2003 In Handbook of Metathesis
(Weinheim, Germany: Wiley-VCH)
8. Grubbs R H 2006 Angew. Chem. Int. Ed. 45 3760
9. Schrodi Y and Pederson R L 2007 Aldrichimica Acta 40
45
10. Hoveyda A H and Zhugralin A R 2007 Nature 450
243
11. Prescher J A and Bertozzi C R 2005 Nat. Chem. Biol. 1
13
12. Miller S J, Blackwell H E and Grubbs R H 1996 J. Am.
Chem. Soc. 118 9606
13. Ghalit N, Reichwein J F, Hilbers H W, Breukink E,
Rijkers D T S and Liskamp R M J 2007 Chem. Bio.
Chem. 8 1540
14. Berezowska I, Chung N N, Lemieux C, Wilkes B C and
Schiller P W 2006 Acta Biochim. Pol. 53(1) 73
15. Nishino N, Shivashimpi G M, Soni B, Bhuiyan M P I,
Kato T, Maeda S, Nishino T G and Yoshida M 2008
Bioorg. Med. Chem. 16 437
16. Islam N M, Kato T, Nishino N, Kim H J, Ito A
and Yoshida M 2010 Bioorg. Med. Chem. Lett. 20
997
17. Islam M N, Islam M S, Hoque M A, Kato T, Nishino
N, Ito A and Yoshida M 2014 Bioorg. Med. Chem. 22
3862
18. Hoque M A, Islam M S, Islam M N, Kato T, Nishino
N, Ito A and Yoshida M 2014 Amino Acids 46
2435
4. Conclusion
In conclusion, we have established the lower limit of
aliphatic loop length and a better synthetic route for
bicyclic tetrapeptide synthesis. We have found that nine
methylene loop length is the lower limit for aliphatic
loop and the synthetic route selection depends on the
configuration of amino acids in the cyclic tetrapeptide
scaffold. Ring closing metathesis followed by peptide
cyclization is the favourable route for bicyclic tetrapep-
tide with LDLD configuration and the reverse route for
the LLLD configuration. Yields in peptide cyclization
and ring closing metathesis steps and CD spectral anal-
ysis have been used as evidence in favour of our con-
clusion. This information is of immense importance for
those working in the field of peptide-based drug design
and synthesis.
19. Komatsu Y, Tomizaki K, Tsukamoto M, Kato T,
Nishino N, Sato S, Yamori T, Tsuruo T, Furumai R,
Yoshida M, Horinouchi S and Hayashi H 2001 Cancer
Res. 61 4459
20. Bierzyñski A 2001 Acta Biochim. Pol. 48 1091