Per-N-methylated analogues of an antitumor bicyclic hexapeptide RA-VII

Article abstract of DOI:10.1016/j.bmc.2011.02.003

Penta-N-methyl and hexa-N-methyl analogues of RA-VII, an antitumor bicyclic hexapeptide of plant origin, were prepared. In the former, the nitrogens of d-Ala-1 and Ala-4 and in the latter, those of d-Ala-1, Ala-2, and Ala-4 were methylated under the phase

Full text of DOI:10.1016/j.bmc.2011.02.003

Bioorganic & Medicinal Chemistry 19 (2011) 2458–2463
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Per-N-methylated analogues of an antitumor bicyclic hexapeptide RA-VII
Yukio Hitotsuyanagi, Ji-Ean Lee, Saori Kato, Ik-Hwi Kim, Hideyuki Kohashi, Haruhiko Fukaya,
⇑
Koichi Takeya
School of Pharmacy, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Penta-N-methyl and hexa-N-methyl analogues of RA-VII, an antitumor bicyclic hexapeptide of plant ori-
Received 15 December 2010
Revised 29 January 2011
Accepted 1 February 2011
Available online 5 March 2011
gin, were prepared. In the former, the nitrogens of D-Ala-1 and Ala-4 and in the latter, those of D-Ala-1,
Ala-2, and Ala-4 were methylated under the phase-transfer catalysis conditions. Their solution structures
were established by NOESY experiments and the crystal structures by X-ray crystallography. Those two
methylated analogues showed much weaker cytotoxicity against P-388 leukemia cells than the parent
RA-VII.
Keywords:
RA-VII
Ó 2011 Elsevier Ltd. All rights reserved.
Bouvardin
N-Methylation
X-ray crystallography
Rubia cordifolia
Cytotoxicity
1. Introduction
RA-VII (1),1,2 isolated from Rubia cordifolia L. and Rubia akane
Nakai (Rubiaceae), and bouvardin (NSC 259968, 2)3 from Bouvardia
ternifolia (Cav.) Schltdl. (Rubiaceae) (Fig. 1) are bicyclic hexapep-
tides which are structurally closely related to each other, both hav-
ing potent antitumor activity. Their antitumor action is considered
to be due to inhibition of protein synthesis through interaction
with eukaryotic ribosomes.4,5 Recently, peptide 1 was shown to
cause conformational change of F-actin and induce G2 arrest
through the inhibition of cytokinesis.6 These peptides exist as a
mixture of two or three stable conformers in solution,7,8 and the
most populated conformer of 1, having trans, trans, trans, trans,
cis, and trans (t-t-t-t-c-t) configuration in the peptide bonds be-
Tyr-3
Ala-2
O
O
OR2
Me
R1
N
Me
O
N
HN
D
-Ala-1
Me
O
Me
Me
Ala-4
NH
O
N
R3
N
Me
O
H
Tyr-6
Tyr-5
tween
Tyr-6, and Tyr-6/
tive conformer.9–11 In this conformer, the intramolecular hydrogen
bondings between -Ala-1 C@O and Ala-4 NH and between Ala-4
-Ala-1 NH contribute to the stabilization of this
D
-Ala-1/Ala-2, Ala-2/Tyr-3, Tyr-3/Ala-4, Ala-4/Tyr-5, Tyr-5/
O
OR4
D-Ala-1, respectively, has been identified as an ac-
D
RA-VII ( ): R1 = R3 = H, R2 = R4 = Me
1
C@O and
D
Bouvardin ( ): R = R4 = H, R2 = Me, R3 = OH
5: R1 = R2 = R4 = Me, R3 = H
1
2
conformation.8
In the present work, to study the roles and meanings of those
intramolecular hydrogen bondings or of the backbone peptide
conformation in the biological activities of 1, we prepared its N-
methylated analogues, 3 and 4, in which the amide protons of
RA-II (8): R1 = R2 = R3 = H, R4 = Me
Figure 1. Structures of RA-VII (1), bouvardin (2), analogue 5, and RA-II (8).
D-Ala-1 and Ala-4 were replaced by methyl groups. In thus
prepared 3 and 4, the intramolecular hydrogen bondings were
interfered and their original peptide backbone conformations were
affected as expected. Their new backbone conformation and their
cytotoxicity were studied.
⇑
Corresponding author. Tel.: +81 42 676 3007; fax: +81 42 677 1436.
E-mail address: takeyak@ps.toyaku.ac.jp (K. Takeya).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.02.003

Y. Hitotsuyanagi et al. / Bioorg. Med. Chem. 19 (2011) 2458–2463
2459
2. Results and discussion
2.1. Preparation of N-methylated analogues 3 and 4
When subjected to the previously known N-methylation proce-
dure by using iodomethane and potassium fluoride–alumina in
1,2-dimethoxyethane8 or by using iodomethane and 50% aqueous
sodium hydroxide with tetrabutylammonium bromide in dichloro-
methane,12 1 produced exclusively mono-methylated product 5, in
which only the nitrogen of Ala-2 was methylated. The N-methyla-
tion of 1 at Ala-2 is known to scarcely affect the backbone peptide
conformation and the cytotoxic activity.8,12 The amide proton of
Ala-2 is externally oriented and solvent-exposed to be more easily
methylated, whereas those of D-Ala-1 and Ala-4 are oriented in-
ward and are involved in the intramolecular hydrogen bonding
so that they are less reactive and need stronger reaction conditions
for that. When peptide 1 was stirred vigorously with 2 mol equiv of
tetrabutylammonium iodide and excess amounts of iodomethane
and powdered sodium hydroxide in dichloromethane at room tem-
perature for 24 h, hexa-N-methyl product 3 was obtained in 78%
yield (Scheme 1). Since 3 did not give good crystals for crystallog-
raphy, its bromide 3a was prepared by treating 3 with pyridinium
hydrobromide perbromide and sodium acetate in a mixture of
methanol and acetic acid. Its X-ray crystallography showed that
the stereochemistry of the chiral centers of 3 was the same as that
of the parent peptide 1 (Fig. 2). N-Methylation of peptides nor-
Figure 2. ORTEP representation of analogue 3a.
hexa-N-methyl analogue, 3, under the phase-transfer catalysis
(PTC) conditions, however, did not affect the original diastereo-
meric integrity, which is noteworthy.
a
mally promotes base-catalyzed epimerization at the adjacent C
position.13 In fact, a simple treatment of 1 with base in a homoge-
Penta-N-methyl analogue 4 was prepared by first protecting the
most reactive Ala-2 nitrogen before N-methylation. Possible
neous medium produced its diastereomers at Tyr-5 or at Tyr-3 and
Tyr-5.14 The present N-methylation of peptide
1
to its
Tyr-3
Ala-2
Me
O
O
O
O
OMe
OMe
OMe
Me
N
Me
Me
N
Me
N
Me
N
Me
O
O
O
HN
Me
Me
N
HN
O Me
N
O Me N
Ala-4
D
-Ala-1
Me
O
Me
Me
Me
Me
Me
O
Me
Me
.
MeI, Bu4NI
C5H5NHBr Br2
NH
O
N
Me O
N
Me O
NaOH
CH2Cl2
78%
NaOAc
MeOH/AcOH
89%
N
O
Hb
Ha
N
N
α
α
Hb
Ha
N
N
N
β
ζ
β
O
O
O
Me
Me
δb
εb
Me
δa
γ
γ
Tyr-6
Tyr-5
Br
δa
εa
εa
δb
ζ
εb
O
O
O
OMe
OMe
OMe
1
RA-VII ( )
3
3a
O-(2,4-dinitrophenyl)hydroxylamine
Bu4NBr, 50% aq. NaOH, CH2Cl2
92%
O
O
O
O
OMe
OMe
OMe
Me
H2N
Me
H2N
Me
HN
N
Me
N
Me
N
Me
O
O
O
N
N
O Me
N
HN
O Me
N
Me
O
Me
Me
Me
O
Me
Me
Me
O
Me
Me
MeI, Bu4NI
NaNO2
N
Me O
NH
O
N
Me O
aq. HCl
THF
46%
NaOH
CH2Cl2
62%
N
N
N
N
N
N
O
Me
O
O
Me
Me
O
O
O
OMe
OMe
OMe
6
7
4
Scheme 1. Preparation of N-methylated analogues of RA-VII.

2460
Y. Hitotsuyanagi et al. / Bioorg. Med. Chem. 19 (2011) 2458–2463
protecting groups for the peptide nitrogens are very limited, but
we found that an N-amino group served as a good protecting group
The NMR spectra including the NOESY spectrum of 3a, prepared
for the X-ray crystallography, were quite similar to those of 3, indi-
cating that the solution structures of 3 and 3a were almost identi-
cal. Thus, the major conformer of 3a of population 93% was shown
to have the same amide configuration sequence as that of 3.
The 1H and 13C NMR spectra of penta-N-methyl analogue 4 in
CDCl3 (Tables 1 and 2) showed that it was present in two conform-
ers in a ratio of 99:1. In the major conformer of population 99%, the
in this case. When
1
was treated with O-(2,4-dinitro-
phenyl)hydroxylamine and 50% aqueous sodium hydroxide under
PTC conditions, an amino group was effectively introduced on
the nitrogen atom of Ala-2 to give an N-amino product 6 in 92%
yield. The protecting group remained intact during the subsequent
N-methylation of 6 with iodomethane and powdered sodium
hydroxide under PTC conditions to give 7 in 62% yield. The protect-
ing group at Ala-2 was removed easily when 7 was treated with so-
dium nitrite in an acidic medium to give penta-N-methyl analogue
4 in 46% yield. The X-ray crystallography of 4 confirmed that no
changes in the chirality of stereogenic centers in the molecule
had occurred through transformations (Fig. 3).
a
a
a
NOE correlations between D-Ala-1 H /Ala-2 NH, Tyr-3 H /Ala-4 H ,
a
a
a
a
D
Ala-4 H /Tyr-5 NMe, Tyr-5 H /Tyr-6 H , and Tyr-6 H / -Ala-1 NMe
indicated that its amide configurations between
Tyr-3/Ala-4, Ala-4/Tyr-5, Tyr-5/Tyr-6, and Tyr-6/
D
-Ala-1/Ala-2,
-Ala-1 were
D
trans, cis, trans, cis, and trans, respectively. The chemical shift val-
ues of the -protons of Ala-2 and Tyr-3 were so close that the con-
a
figuration of the amide bond between Ala-2/Tyr-3 could not be
determined by the NOESY data. However, it was considered to be
2.2. Structures of 3, 3a and 4 in solution and of 3a and 4 in
crystals
a
cis, because the 13C NMR chemical shift of Tyr-3 C was dC 59.3,
a value characteristic of a cis amide configuration at this position.8
Accordingly, the sequence of the amide configuration of the major
conformer of 4 was considered to be t-c-c-t-c-t as in the major con-
former of 3. The amide configuration between Tyr-3/Ala-4 was cis
both in 3 and 4, which is unique, because in the natural RA-series
peptides, including 1 from which 3 and 4 were derived, this amide
bond normally takes trans configuration. Both in the major con-
The solution structures of 3, 3a, and 4, in which N-methyl
groups were introduced and accordingly the intramolecular hydro-
gen bonding and the backbone conformation had been affected,
were studied by the NMR experiments. The 1H and 13C NMR spec-
tra of hexa-N-methyl analogue 3 in CDCl3 (Tables 1 and 2) demon-
strated that it was present in three conformers in a ratio of 93:6:1.
In the solution structure of the major conformer of population 93%,
formers of 3 and 4, characteristic NOE correlations were observed
a
b
the NOE correlations observed between
between
tween -Ala-1/Ala-2 was trans, and the correlation between Ala-2
D
-Ala-1 H /Ala-2 NMe and
between
D
-Ala-1 H3
/D-Ala-1 NMe, Tyr-3 NMe/Tyr-5 NMe, Ala-4
D
-Ala-1 H3b/Ala-2 NMe showed that the amide bond be-
NMe/Tyr-5 NMe, and Ala-4 NMe/Tyr-5 Hba, implying that their
peptide backbone structures in CDCl3 were almost identical.
The solid state structures of 3a and 4 were studied by crystal-
lography and the results were compared with that of RA-II (8),
which is considered to have conformational property identical to
that of 1, and which displayed weaker antitumor and cytotoxic
activities than 1 due to its absence of O-methyl group in Tyr-3
(Fig. 1).15–17 The results showed that the solid state structure of
3a was basically identical to that in solution of the major conform-
ers of 3 and 3a as determined by its NOESY experiments in CDCl3,
with the amide configuration of t-c-c-t-c-t (Table 3). The distances
between the protons (or the nearest methyl protons) in solid state
as estimated by X-ray crystallography and the relevant ones in-
volved in the NOE correlations in 3 in CDCl3 solution, which char-
D
a
a
H /Tyr-3 H , that the amide bond between Ala-2/Tyr-3 was cis
a
(Fig. 4). Analogously, from the correlations between Tyr-3 H /
a
a
a
a
a
Ala-4 H , Ala-4 H /Tyr-5 NMe, Tyr-5 H /Tyr-6 H , and Tyr-6 H /
-Ala-1 NMe, the amide bonds between Tyr-3/Ala-4, Ala-4/Tyr-5,
Tyr-5/Tyr-6, and Tyr-6/ -Ala-1 were determined to be cis, trans,
D
D
cis, and trans, respectively. Thus, the sequence of the amide config-
uration of the major conformer of 3 was shown to be t-c-c-t-c-t.
acterize the backbone structure of its major conformer, were as
b
follows: 2.24 Å (
D-Ala-1 H3
/D
-Ala-1 NMe), 3.04 Å (
D
-Ala-1 NMe/
a
a
Ala-4 H3b), 2.88 Å (Ala-2 H /Ala-4 H ), 2.78 Å (Tyr-3 NMe/Tyr-5
NMe), 2.93 Å (Ala-4 NMe/Tyr-5 NMe), and 2.62 Å (Ala-4 NMe/
Tyr-5 Hba). The distances by crystallography were quite reasonable
for producing NOE cross-peaks.
The solid state structure of 4 was shown to have the same t-c-c-
t-c-t configuration as 3a, with very similar back bone torsions (Ta-
ble 3). However, the side-chain torsion angles of Tyr-3, that is, the
a
c
N–C –Cb–C
dihedral
(v
1), were ꢀ163.5(3)° for 3a and
ꢀ65.59(19)° for 4. The presence of the difference in the dihedral
angles as demonstrated by superimposing the crystal structures
of 3a and 4 (Fig. 5) was suggested by a series of the NMR data of
the major conformers of 3 and 4. The methyl signal of Ala-2 in 4
resonated at dH 0.58, that in 3 at dH 1.17, and those in other ana-
logues 3a, 6, and 7 at dH 1.15–1.43. This rather unusual upfield shift
of the Ala-2 methyl signal may be explained by the presence of the
Tyr-3 phenyl ring right above the Ala-2 methyl group, as demon-
strated in the solid state structure of 4 (Fig. 3). Differences between
the chemical shifts and the vicinal coupling constants of the b-pro-
tons of Tyr-3 in 3 and 4 also suggested the conformational prefer-
ences of their Tyr-3 side chain. The pro-R b-proton of Tyr-3 in 3
resonated at dH 2.70 with Jvicinal = 4.9 Hz, whereas that in 4 at dH
3.05 with Jvicinal = 8.7 Hz, and the pro-S b-proton of Tyr-3 in 3
resonated at dH 3.50 with Jvicinal = 9.3 Hz, whereas that in 4 at dH
3.11 with Jvicinal = 5.4 Hz. There might be subtle differences
Figure 3. ORTEP representation of analogue 4.

Y. Hitotsuyanagi et al. / Bioorg. Med. Chem. 19 (2011) 2458–2463
2461
Table 1
1H NMR data for the major conformers of 3, 3a, 4, 6, and 7 in CDCl3 at 300 Ka
Position
3
dH
3a
dH
4
dH
6
dH
7
dH
b
c
d
e
f
a
5.10 (q, 7.3)
5.13 (q, 7.4)
4.64 (q, 7.2)
5.25 (qd, 7.2, 6.8)
5.82 (q, 7.4)
D-Ala-1
b
1.30 (d, 7.3, 3H)
2.92 (s, 3H)
5.73 (q, 7.1)
1.33 (d, 7.4, 3H)
2.93 (s, 3H)
5.74 (q, 7.1)
1.26 (d, 7.2, 3H)
2.81 (s, 3H)
1.27 (d, 6.8, 3H)
6.41 (d, 7.2)
5.24 (q, 7.1)
1.30 (d, 7.4, 3H)
2.89 (s, 3H)
5.63 (q, 7.0)
NMe (NH)
a
Ala-2
Tyr-3
4.90g
b
1.17 (d, 7.1, 3H)
3.34 (s, 3H)
1.15 (d, 7.1, 3H)
3.36 (s, 3H)
0.58 (d, 6.4, 3H)
6.80 (d, 8.8)
1.43 (d, 7.1, 3H)
4.34 (s, 2H)
1.25 (d, 7.0, 3H)
4.52 (s, 2H)
5.00 (dd, 9.4, 4.7)
NMe (NH, NNH2)
a
ba
bb
d
e
5.17 (dd, 9.3, 4.9)
3.50 (dd, 13.4, 9.3)
2.70 (dd, 13.4, 4.9)
7.12 (d-like, 8.6, 2H)
6.83 (d-like, 8.6, 2H)
2.88 (s, 3H)
5.16 (dd, 9.1, 5.0)
3.49 (dd, 13.4, 9.1)
2.72 (dd, 13.4, 5.0)
7.13 (d-like, 8.7, 2H)
6.84 (d-like, 8.7, 2H)
2.89 (s, 3H)
4.89g
3.56 (dd, 10.4, 5.3)
3.35 (dd, 14.1, 5.3)
3.33 (dd, 14.1, 10.4)
7.04 (d-like, 8.6, 2H)
6.84 (d-like, 8.6, 2H)
2.86 (s, 3H)
3.11 (dd, 14.3, 5.4)
3.05 (dd, 14.3, 8.7)
7.08 (d-like, 8.7, 2H)
6.84 (d-like, 8.7, 2H)
2.91 (s, 3H)
3.54 (dd, 13.2, 9.4)
2.69g
7.11 (d-like, 8.6, 2H)
6.83 (d-like, 8.6, 2H)
2.88 (s, 3H)
NMe
OMe
3.78 (s, 3H)
3.78 (s, 3H)
3.76 (s, 3H)
3.80 (s, 3H)
3.78 (s, 3H)
Ala-4
Tyr-5
a
b
5.20 (q, 6.4)
1.15 (d, 6.4, 3H)
2.75 (s, 3H)
5.21 (q, 6.4)
1.16 (d, 6.4, 3H)
2.76 (s, 3H)
4.19 (q, 6.5)
1.36 (d, 6.5, 3H)
2.80 (s, 3H)
4.73 (dq, 7.6, 6.7)
1.07 (d, 6.7, 3H)
6.69 (d, 7.6)
5.44 (dd, 11.3, 2.9)
2.62 (dd, 11.3, 2.9)
3.69 (t, 11.3)
7.27 (dd, 8.4, 2.2)
7.42 (dd, 8.3, 2.2)
6.88 (dd, 8.4, 2.4)
7.21 (dd, 8.3, 2.4)
3.10 (s, 3H)
5.21 (q, 6.5)
1.16 (d, 6.5, 3H)
2.75 (s, 3H)
NMe (NH)
a
ba
bb
da
db
4.82 (dd, 10.3, 2.3)
2.81 (dd, 10.7, 2.3)
3.45 (t, 10.5)
7.26 (dd, 8.5, 2.2)
7.59 (dd, 8.5, 2.2)
6.87 (dd, 8.5, 2.4)
7.19 (dd, 8.5, 2.4)
2.50 (s, 3H)
4.81 (dd, 10.3, 2.2)
2.85 (dd, 10.7, 2.2)
3.43 (t, 10.6)
7.27 (dd, 8.4, 2.1)
7.62 (dd, 8.4, 2.1)
6.88 (dd, 8.4, 2.5)
7.16 (dd, 8.4, 2.5)
2.52 (s, 3H)
4.99 (dd, 10.5, 2.2)
4.78 (dd, 10.3, 2.4)
2.69g
2.91g
3.51 (t, 10.6)
7.28 (dd, 8.4, 2.2)
7.53 (dd, 8.4, 2.2)
6.88 (dd, 8.4, 2.5)
7.20 (dd, 8.4, 2.5)
2.57 (s, 3H)
3.39 (t, 10.3)
7.25 (dd, 8.4, 2.2)
7.61 (dd, 8.5, 2.2)
6.88 (dd, 8.4, 2.5)
7.18 (dd, 8.5, 2.5)
2.51 (s, 3H)
ea
eb
NMe
Tyr-6
a
ba
bb
da
db
4.66 (dd, 11.9, 3.8)
2.97 (dd, 18.2, 11.9)
2.86 (dd, 18.2, 3.8)
6.59 (dd, 8.4, 2.1)
4.48 (d, 2.1)
4.62 (dd, 12.2, 3.1)
2.98 (dd, 18.7, 12.2)
2.66 (dd, 18.7, 3.1)
4.67 (dd, 12.1, 3.5)
2.96 (dd, 18.2, 12.1)
2.79g
6.59 (dd, 8.3, 1.9)
4.45 (d, 1.9)
4.56 (dd, 12.1, 3.8)
3.13 (dd, 18.0, 12.1)
2.95 (dd, 18.0, 3.8)
6.58 (dd, 8.3, 2.0)
4.35 (d, 2.0)
4.65 (dd, 11.9, 3.6)
2.99 (dd, 18.4, 11.9)
2.89g
6.60 (dd, 8.3, 2.1)
4.48 (d, 2.1)
4.51 (s)
ea
6.81 (d, 8.4)
7.05 (s)
6.81 (d, 8.3)
6.80 (d, 8.3)
6.81 (d, 8.3)
NMe
OMe
2.68 (s, 3H)
3.93 (s, 3H)
2.66 (s, 3H)
3.94 (s, 3H)
2.68 (s, 3H)
3.93 (s, 3H)
2.72 (s, 3H)
3.94 (s, 3H)
2.68 (s, 3H)
3.93 (s, 3H)
a
b
c
d
e
f
Recorded at 500 MHz, chemical shifts referenced to residual CHCl3 (7.26 ppm); J-values given in Hz in parentheses.
Mixture of one major and two minor conformers in a ratio of 93:6:1.
Mixture of two conformers in a ratio of 93:7.
Mixture of two conformers in a ratio of 99:1.
Mixture of two conformers in a ratio of 97:3.
Mixture of two conformers in a ratio of 92:8.
Multiplicity patterns were unclear due to signal overlapping.
g
between the peptide backbone structures of 3 and 4, which might
be responsible for such difference in their side chain torsion.
The unique structural features of 3 and 4 were highlighted by
superimposing the crystal structures of 3a and 4 over that of RA-
II (8), having essentially the same solid state structure as that of
their parent antitumor peptide 1 (Fig. 5). The part of the peptide
backbone at residues 1–4 in 3a and 4 significantly deviated form
that of 8, as expected by the difference in the configurations be-
tween Ala-2/Tyr-3 and between Tyr-3/Ala-4, which were cis and
cis in 3a and 4, and trans and trans in 8.
of the hydrogen bondings or of the conformation and its relation
to the activity, we prepared a series of analogues of 1 in which
the amide nitrogens of D-Ala-1 and Ala-4 were methylated. Intro-
duction of methyl groups and interfering of the hydrogen bonding
led to generation of unique backbone structures not observed in
the natural RA-series peptides. In those analogues, the N-methyl
group at Ala-2 affected the side-chain torsion angles of Tyr-3.
The quite obvious differences in the 18-membered peptide back-
bone between 3a and 4, and 8 as demonstrated in Figure 5 may
give further information about the peptide conformation–activity
relationships and also about the effect of different groups on the
three-dimensional molecular structures.
2.3. Cytotoxicity
Analogues 3, 3a, 4, 6, and 7 prepared in the present study, and,
as reference, 1, were evaluated for their cytotoxicity against P-388
leukemia cells, and the results are summarized in Table 4. Of them,
only the N-amino analogue 6 retained the potent cytotoxicity of
the original peptide 1, whereas the others 3, 3a, 4, and 7, having
4. Experimental
4.1. General
Melting points were determined on a Yanaco MP-3 apparatus
and recorded uncorrected. Optical rotations were measured on a
JASCO P-1030 digital polarimeter. IR spectra were recorded on a
JASCO FT/IR 620 spectrophotometer, NMR spectra on a Bruker
DRX 500 spectrometer (500 MHz for 1H NMR, 125 MHz for 13C
NMR), and mass spectra on a Micromass LCT spectrometer. Sin-
gle-crystal X-ray analyses were carried out on a Bruker AXS APEX
II ULTRA CCD area detector diffractometer with a rotating anode
N-methyl groups at D-Ala-1 and Ala-4, showed significantly re-
duced activity.
3. Conclusion
In antitumor bicyclic hexapeptides of RA series, the hydrogen
bondings between -Ala-1 and Ala-4 take an active part in
D
retaining the conformation considered to be the active form of
the molecule. In the present study, to further relate the meanings
source (Mo K
a radiation, k = 0.71073 Å). Preparative HPLC was car-
ried out on a Shimadzu LC-6AD pump unit equipped with an

2462
Y. Hitotsuyanagi et al. / Bioorg. Med. Chem. 19 (2011) 2458–2463
Table 2
Table 3
13C NMR data for the major conformers of 3, 3a, 4, 6, and 7 in CDCl3 at 300 Ka
X-ray calculated backbone dihedrals (degree) in analogues 3a and 4
Residue 3a
-Ala-1
Position
3
dC
3a
dC
4
dC
6
dC
7
dC
4
u
93.0(3)
97.59(16)
D
a
49.5
49.6
52.2
46.3
49.8
D-Ala-1
w
x
u
w
x
u
w
x
u
w
x
u
w
x
u
w
x
ꢀ131.8(3)
ꢀ173.3(3)
ꢀ104.4(3)
121.3(3)
2.6(5)
ꢀ89.51(17)
b
14.6
174.3
31.0
14.7
174.3
31.0
45.2
16.7
172.9
30.8
58.4
14.3
169.8
30.1
43.2
19.7
19.6
176.0
14.7
176.5
31.3
46.3
15.2
ꢀ175.09(14)
ꢀ151.92(15)
125.20(16)
6.6(2)
C@O
NMe
a
Ala-2
Tyr-3
Ala-4
Tyr-5
Tyr-6
Ala-2
Tyr-3
45.2
16.6
48.4
13.4
172.6
b
ꢀ149.1(3)
ꢀ129.04(16)
C@O
NMe
a
b
c
d
e
f
C@O
NMe
OMe
a
172.9
30.7
172.6
173.8
89.9(4)
83.44(18)
ꢀ6.1(2)
11.3(6)
58.4
37.5
59.3c
35.8
68.1
32.5
58.7
37.4
ꢀ146.6(4)
ꢀ137.16(16)
65.10(18)
ꢀ171.41(13)
ꢀ143.27(14)
92.51(17)
37.5
67.9(4)
128.4
130.7b
114.3b
158.8
167.5
29.7
128.5
130.7b
114.4b
158.8
167.6
29.6
55.3
54.6
17.0
169.6
29.1
128.4
130.5b
114.7b
159.0
167.4
29.2
55.4
54.1
17.5
130.6
130.2b
114.1b
158.5
167.8
39.6
55.3
46.2
18.3
171.6
128.4
130.6b
114.4b
158.8
167.0
29.6
55.3
54.7
16.8
169.4
29.1
ꢀ177.4(3)
ꢀ152.5(3)
98.8(3)
ꢀ20.4(4)
ꢀ80.2(4)
152.0(3)
172.5(3)
ꢀ13.5(2)
ꢀ86.03(18)
159.52(14)
168.71(13)
55.3
54.6
17.0
Ala-4
Tyr-5
b
C@O
NMe
a
169.5
29.1
168.2
29.1
60.7
36.6
60.7
36.8
59.2c
36.5
54.5
37.0
60.9
b
36.6
c
da
db
137.1
132.6
131.4
123.5
125.9
157.2
169.2
30.1
137.4
132.6
131.7
123.2
125.5
156.4
169.4
30.2
136.6
132.8
131.3
123.6
126.0
157.3
168.6
29.7
135.4
132.8
131.0
124.2
125.9
158.1
169.1
30.5
137.0
132.6
131.3
123.6
125.8
157.2
169.4
30.3
e
e
f
a
b
C@O
NMe
a
Tyr-6
54.9
54.9
54.6
57.3
55.2
b
32.2
33.5
32.0
35.5
32.2
c
da
db
128.0
120.9
113.2
112.2
153.1
146.6
172.5
29.5
127.0
114.6
113.8
116.1
152.5
147.3
172.2
29.4
127.8
121.0
113.0
112.3
153.0
146.7
172.3
29.4
128.3
121.0
113.4
112.2
153.1
146.5
170.3
29.4
128.2
120.9
113.2
112.2
153.1
146.5
172.2
29.5
e
e
f
a
b
C@O
NMe
OMe
Figure 5. Superposition of the crystal structures of 3a (red), 4 (blue), and 8 (green).
56.2
56.4
56.2
56.1
56.2
a
b
c
Recorded at 125 MHz, chemical shifts referenced to CDCl3 (77.03 ppm).
Two carbons.
Assignments may be reversed.
Table 4
Cytotoxicity of RA-VII (1) and analogues 3,
3a, 4, 6, and 7 against P-388 leukemia cells
Compound
IC50 (lM)
RA-VII (1)
3
0.0022
3.1
3a
4
>110
10
6
7
0.0057
49
SPD-10A UV detector (254 nm) and an Inertsil ODS-3 column
(5
m, 20 ꢁ 250 mm, GL Sciences Inc.), and analytical HPLC with
a Sunniest C18 column (5
l
l
m, 4.6 ꢁ 250 mm, ChromaNik Technol-
ogies Inc.). In vitro cytotoxicity assays were performed as de-
scribed previously using the P-388 murine lymphocytic leukemia
cell line.10
4.2. Per-N-methylation of 1
Iodomethane (0.5 mL, 8.0 mmol), tetrabutylammonium iodide
(96.0 mg, 0.260 mmol), and powdered NaOH (200 mg, 5.00 mmol)
were added to a solution of 1 (100.2 mg, 0.130 mmol) in dichloro-
methane (1.5 mL), and the mixture was stirred vigorously at room
temperature for 24 h. Water (5 mL) was added to the mixture, and
the whole was extracted with CH2Cl2 (3 ꢁ 10 mL). The combined
Figure 4. Selected NOE correlations for 3. The side chain of Tyr-3 was omitted for
clarity.

Y. Hitotsuyanagi et al. / Bioorg. Med. Chem. 19 (2011) 2458–2463
2463
CH2Cl2 extracts were washed sequentially with aqueous
hydrochloric acid (2 M, 10 mL) and water (5 mL), dried over sodium
sulfate, and filtered, and the solvent removed in vacuo. The residue
was subjected to HPLC (MeOH/H2O 70:30) to give 3 (82.5 mg, 78%)
0.1 M aqueous hydrochloric acid (10:1, 1 mL) over 30 min. The mix-
ture was further stirred at 0 °C for 30 min and then at room temper-
ature for 3 h. Water (5 mL) was added to the mixture, and the whole
was extracted with CHCl3 (3 ꢁ 10 mL). The combined CHCl3 extracts
were washed with brine (5 mL), dried over sodium sulfate, and fil-
tered, and the solvent removed in vacuo. The residue was subjected
to HPLC (MeCN/H2O 45:55) to give 4 (8.1 mg, 46%) as colorless
as an amorphous solid, ½a D26
ꢀ89 (c 0.14, CHCl3). IR (film) mmax
ꢂ
2934, 1650, 1514 cmꢀ1 1H and 13C NMR data, in Tables 1 and 2,
;
respectively; HRESIMS m/z 813.4141 ([M+H]+, calcd for C44H57
N6O9, 813.4187); Analytical HPLC (MeCN/H2O 55:45, flow rate
0.53 mL/min) 19.1 and 29.8 min.
prisms, mp 208–211 °C (MeOH), ½a D25
ꢀ120 (c 0.07, CHCl3). IR (film)
ꢂ
mmax 3307, 3006, 2934, 1651, 1515 cmꢀ1 1H and 13C NMR data, in
;
Tables 1 and 2, respectively; HRESIMS m/z 799.4014 ([M+H]+, calcd
4.3. Bromination of 3
for C43H55N6O9, 799.4031).
Sodium acetate (2.7 mg, 0.033 mmol) and pyridinium hydro-
bromide perbromide (ca. 85%, 12.4 mg, 0.033 mmol) were added
to an ice-cooled solution of 3 (18.0 mg, 0.0221 mmol) in a mixture
of MeOH/AcOH (1:1, 1 mL). The mixture was stirred at 0 °C for 1 h
and then at room temperature for 2 days. The mixture was diluted
with CHCl3 (20 mL), washed sequentially with aqueous NaHSO3
(5%, 5 mL) and brine (10 mL), and dried over Na2SO4. The mixture
was filtered and the solvent removed in vacuo. The residue was
subjected to silica gel column chromatography (CHCl3/MeOH
20:1) to afford 3a (17.5 mg, 89%) as colorless prisms, mp 222–
4.7. Crystallography of 3a and 4
Compound 3a: C88.52H112.52Br2Cl1.55N12O18; M = 1847.29; 0.24 ꢁ
0.20 ꢁ 0.15 mm, monoclinic; space group C2; a = 30.6207(9) Å;
b = 12.3675(4) Å; c = 13.3862(4) Å; b = 109.200(2)°; V = 4787.4
(3) Å3; Z = 2; DX = 1.282 Mg mꢀ3
;
l
(Mo K
reflections collected; 12,587 unique (Rint = 0.0336), R1 = 0.0673,
wR2 = 0.1844 [I > 2 (I)], GOF = 1.021; R1 = 0.0880, wR2 = 0.1995
(all data), absolute structure parameter 0.029(8).
Compound 4:
a
) = 0.959 mmꢀ1; 37,271
r
C
43H54N6O9, 2(CH4O), M = 863.01, 0.48 ꢁ
225 °C (MeOH/H2O), ½a D26
ꢂ
ꢀ75 (c 0.12, CHCl3). IR (film) mmax 3006,
0.27 ꢁ 0.06 mm, orthorhombic, space group P212121, a = 15.685
2935, 1652, 1501, cmꢀ1
;
1H and 13C NMR data, in Tables 1 and 2,
(2) Å, b = 15.782(2) Å, c = 18.147(3) Å, V = 4492.2(12) Å3, Z = 4,
respectively; HRESIMS m/z 891.3354 ([M+H]+, calcd for
DX = 1.276 Mg mꢀ3
,
l
(Mo K
collected, 10,404 unique (Rint = 0.0334), R1 = 0.0423, wR2 = 0.1198
[I > 2 (I)], GOF = 1.033; R1 = 0.0436, wR2 = 0.1212 (all data).
a
) = 0.092 mmꢀ1, 52,252 reflections
C44H56N6O9Br, 891.3292).
r
4.4. N-Amination of 1
The structures were solved by direct methods using SHELXS-97,18
and refined by full-matrix least-squares on F2 using SHELXL-97.19
Crystallographic data for compounds 3a and 4 reported in this pa-
per have been deposited with the Cambridge Crystallographic Data
Centre under the reference numbers CCDC 677161 and 795575,
respectively. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/data_request/cif, or by e-mailing data_re-
quest@ccdc.cam.ac.uk, or by contacting The Cambridge Crystallo-
graphic Data Centre, 12, Union Road, Cambridge CB2 1EZ, UK;
fax: +44 1223 336033.
A solution of O-(2,4-dinitrophenyl)hydroxylamine (65.1 mg,
0.327 mmol) in dichloromethane (3.5 mL) was added over 1.5 h
to a vigorously stirred mixture of 1 (100.8 mg, 0.131 mmol), tetra-
butylammonium bromide (21.1 mg, 0.0655 mmol), dichlorometh-
ane (3.5 mL), and aqueous NaOH (50%, 1.5 mL). The mixture was
further stirred for 1 h. Aqueous KOH (1 M, 20 mL) was added to
the mixture, and the whole was extracted with dichloromethane
(50 mL). The extract was washed with brine (5 mL), dried over
Na2SO4, and filtered, and the solvent removed in vacuo. The residue
was subjected to silica gel column chromatography (CH2Cl2/MeOH
10:1) and then to HPLC (MeOH/H2O 65:35) to give 6 (94.4 mg, 92%)
References and notes
as colorless needles, mp >300 °C (MeOH/CHCl3), ½a D25
ꢂ
ꢀ194 (c 0.16,
1. Itokawa, H.; Takeya, K.; Mihara, K.; Mori, N.; Hamanaka, T.; Sonobe, T.; Iitaka, Y.
Chem. Pharm. Bull. 1983, 31, 1424.
2. Itokawa, H.; Takeya, K.; Hitotsuyanagi, Y.; Morita, H. In The Alkaloids; Cordell, G.
A., Ed.; Academic Press: NY, 1997; Vol. 49, p 301.
3. Jolad, S. D.; Hoffman, J. J.; Torrance, S. J.; Wiedhopf, R. M.; Cole, J. R.; Arora, S. K.;
Bates, R. B.; Gargiulo, R. L.; Kriek, G. R. J. Am. Chem. Soc. 1977, 99, 8040.
4. Zalacaín, M.; Zaera, E.; Vázquez, D.; Jiménez, A. FEBS Lett. 1982, 148, 95.
5. Sirdeshpande, B. V.; Toogood, P. L. Bioorg. Chem. 1995, 23, 460.
6. Fujiwara, H.; Saito, S.; Hitotsuyanagi, Y.; Takeya, K.; Ohizumi, Y. Cancer Lett.
2004, 209, 223.
CHCl3). IR (film) mmax 3388, 2934, 1680, 1632, 1513 cmꢀ1
;
1H and
13C NMR data, in Tables 1 and 2, respectively; HRESIMS m/z
786.3094 ([M+H]+, calcd for C41H52N7O9, 786.3827).
4.5. Per-N-methylation of 6
Iodomethane (0.15 mL, 2.4 mmol), tetrabutylammonium iodide
(28.1 mg, 0.0761 mmol), and powdered NaOH (60 mg, 1.5 mmol)
were added to a solution of 6 (29.9 mg, 0.0380 mmol) in dichloro-
methane (0.45 mL), and the mixture was stirred vigorously at room
temperature for 24 h. Water (5 mL) was added to the mixture, and
the whole was extracted with CH2Cl2 (3 ꢁ 10 mL). The combined
CH2Cl2 extracts were washed with brine (5 mL), dried over sodium
sulfate, and filtered, and the solvent removed in vacuo. The residue
was subjected to HPLC (MeOH/H2O 70:30) to give 7 (19.3 mg, 62%)
7. Bates, R. B.; Cole, J. R.; Hoffmann, J. J.; Kriek, G. R.; Linz, G. S.; Torrance, S. J. J. Am.
Chem. Soc. 1983, 105, 1343.
8. Morita, H.; Kondo, K.; Hitotsuyanagi, Y.; Takeya, K.; Itokawa, H.; Tomioka, N.;
Itai, A.; Iitaka, Y. Tetrahedron 1991, 47, 2757.
9. Itokawa, H.; Kondo, K.; Hitotsuyanagi, Y.; Isomura, M.; Takeya, K. Chem. Pharm.
Bull. 1993, 41, 1402.
10. Itokawa, H.; Saitou, K.; Morita, H.; Takeya, K.; Yamada, K. Chem. Pharm. Bull.
1992, 40, 2984.
11. Boger, D. L.; Zhou, J. J. Am. Chem. Soc. 1995, 117, 7364.
12. Hitotsuyanagi, Y.; Suzuki, J.; Takeya, K.; Itokawa, H. Bioorg. Med. Chem. Lett.
1994, 4, 1633.
13. Gund, P.; Veber, D. F. J. Am. Chem. Soc. 1979, 101, 1885.
14. Itokawa, H.; Morita, H.; Kondo, K.; Hitotsuyanagi, Y.; Takeya, K.; Iitaka, Y. J.
Chem. Soc., Perkin Trans. 2 1992, 1635.
15. Hitotsuyanagi, Y.; Sasaki, S.-i.; Matsumoto, Y.; Yamaguchi, K.; Itokawa, H.;
Takeya, K. J. Am. Chem. Soc. 2003, 125, 7284.
16. Itokawa, H.; Takeya, K.; Mori, N.; Sonobe, T.; Mihashi, S.; Hamanaka, T. Chem.
Pharm. Bull. 1986, 34, 3762.
as an amorphous solid, ½a D25
ꢂ
ꢀ68 (c 0.19, CHCl3). IR (film) mmax
3356, 2936, 1649, 1514 cmꢀ1
;
1H and 13C NMR data, in Tables 1
and 2, respectively; HRESIMS m/z 814.4125 ([M+H]+, calcd for
C43H56N7O9, 814.4140); Analytical HPLC (MeCN/H2O 55:45, flow
rate 0.53 mL/min) 18.3 and 25.1 min.
17. Itokawa, H.; Kondo, K.; Hitotsuyanagi, Y.; Nakamura, A.; Morita, H.; Takeya, K.
Chem. Pharm. Bull. 1993, 41, 1266.
18. Sheldrick, G. M. SHELXS-97: Program for the Solution of Crystal Structures;
University of Göttingen: Göttingen, Germany, 1997.
4.6. Des-N-amination of 7
Sodium nitrite (3.0 mg, 0.043 mmol) was added to a cooled (0 °C)
and stirred solution of 7 (17.9 mg, 0.0220 mmol) in tetrahydrofuran/
19. Sheldrick, G. M. SHELXL-97: Program for the Refinement of Crystal Structures;
University of Göttingen: Göttingen, Germany, 1997.