Dolastatin 11 conformations, analogues and pharmacophore

Article abstract of DOI:10.1016/j.bmc.2005.04.040

Twenty analogues of the natural antitumor agent dolastatin 11, including majusculamide C, were synthesized and tested for cytotoxicity against human cancer cells and stimulation of actin polymerization. Only analogues containing the 30-membered ring were

Full text of DOI:10.1016/j.bmc.2005.04.040

Bioorganic & Medicinal Chemistry 13 (2005) 4138–4152
Dolastatin 11 conformations, analogues and pharmacophore
Md. Ahad Ali,a Robert B. Bates,a,* Zackary D. Crane,a Christopher W. Dicus,a
Michelle R. Gramme,a Ernest Hamel,b Jacob Marcischak,a David S. Martinez,a
Kelly J. McClure,a Pichaya Nakkiew,a George R. Pettit,c Chad C. Stessman,a
Bilal A. Sufia and Gayle V. Yaricka
aDepartment of Chemistry, University of Arizona, Tucson, AZ 85721, USA
bNational Cancer Institute, Frederick, MD 21702, USA
cDepartment of Chemistry and Biochemistry and Cancer Research Institute, Arizona State University, Tempe, AZ 85287, USA
Received 8 March 2005; revised 15 April 2005; accepted 15 April 2005
Available online 5 May 2005
Abstract—Twenty analogues of the natural antitumor agent dolastatin 11, including majusculamide C, were synthesized and tested
for cytotoxicity against human cancer cells and stimulation of actin polymerization. Only analogues containing the 30-membered
ring were active. Molecular modeling and NMR evidence showed the low-energy conformations. The amide bonds are all trans
except for the one between the Tyr and Val units, which is cis. Since an analogue restricted to negative 2-3-4-5 angles stimulated
actin polymerization but was inactive in cells, the binding conformation (most likely the lowest-energy conformation in water)
has a negative 2–3–4–5 angle, whereas a conformation with a positive 2–3–4–5 angle (most likely the lowest energy conformation
in chloroform) goes through cell walls. The highly active R alcohol from borohydride reduction of dolastatin 11 is a candidate for
conversion to prodrugs.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
2. Conformations of dolastatin 11 (1)
The isolation from the sea hare Dolabella auricularia,
structure determination, and strong antitumor activity
of dolastatin 11 (1) were reported in 1989.1 Its synthesis
was reported in 19972 and its cytotoxicity was reported
to involve disruption of microfilaments by stimulation
of the polymerization of actin in 2001.3 An X-ray fiber
diffraction study of the F-actin/dolastatin 11 complex
showed dolastatin 11 (1) to stabilize microfilaments by
binding to two strands of F-actin.4 We now report the
synthesis of 20 analogues (2–21) of dolastatin 11 (1)
and the results of testing them for cytotoxicity against
human tumors and for stimulation of actin polymeriza-
tion. These results are interpreted in terms of the main
conformations which appear to be present from NMR
studies and molecular modeling.
NMR evidence from NOEs (Table 1) indicated that in
the predominant form of dolastatin 11 (1) in chloroform
solution, the ester bond and all of the amide bonds are
trans except for the amide bond between the Tyr and
Val units, which is cis from the very strong NOE
observed between the Tyr and Val a-Hꢀs. The NOE
peaks showing these configurations are in italics in
Table 1.
Molecular modeling calculations on dolastatin 11 (1)
with the configurations of the amide and ester bonds re-
stricted to those determined by NMR were performed
for chloroform and water solutions: chloroform to
understand the conformations in the NMR solvent
(deuterochloroform; not soluble enough in D2O for
NMR) which may also be favored for dolastatin 11
and analogues in the cell wall, and water because the cal-
culations in water should apply well to the conforma-
tions favored in blood and intracellular fluid. Fifteen
conformations (C1–C15) were calculated to be within
10 kJ/mol of the lowest-energy form in chloroform
Keywords: Dolastatin 11; Pharmacophore; Depsipeptide; Anticancer.
*
Corresponding author. Tel.: +1 520 621 6317; fax: +1 520 621
8407; e-mail: batesr@u.arizona.edu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.04.040

M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
4139
Structures of dolastatin 11 (1) and derivatives 2–21a
Compound
Map
Ala
R3
Ibu
Leu
Hmp
R1
R2
Et
R4
R5
R6
R7
R8
H
R9
R10
Me
Dolastatin 11 (1)
Map analogues
3-Nor (2)
3,4-Bis-nor (3)
R,R-Acc (3-epimer) (4)
S,S-Acc (4-epimer) (5)
Me
Me
Me
Me
@O
Me
Me
H
H
H
bCH2CH2CH2
CH2CH2CH2
b
Ala analogues
7-Epi (6)
7-Nor (7)
b
H
Ibu analogues
10-Nor (8)
10,10-Bis-nor (9)
11R-OH (10)
11R-OAc (11)
12-Epi (12)
H
H
H
OH
OAc
b
Leu analogue
majusculamide C (13)
Me
H
Hmp analogue
57-Nor (14)
H
Map, Ala analogue
3,4,7-Tris-nor (15)
H
H
H
H
H
Map, Ibu analogue
3,4-Bis-nor-11-OH (16)
OH
OH
OH
Ala, Ibu analogue
7-Nor-11-OH (17)
H
H
Map, Ala, Ibu analogue
3,4,7-Tris-nor-11-OH (18)
H
H
Open analogues
1,2-Seco (19)
13,14-Seco (20)
Unit-deletion analogue
des-Hmp (21)
Only the groups which differ from those in dolastatin 11 (1) are shown for the derivatives.
a Acc = 2-aminocyclopentanecarboxylic acid; Hmp = (2S,3S)-2-hydroxy-3-methylpentanoic acid; Ibu = (S)-4-amino-
2,2-dimethyl-3-oxopentanoic acid; Map = (2S,3R)-2-methyl-3-aminopentanoic acid.
b Epimer at this center.
(C1), and eighteen (W1–W18) to the lowest conforma-
tion in water (W1); eight were common to both media.
These 25 conformations included eight different confor-
mations of the 30-membered ring, listed in Table 2, with
the others involving different side-chain rotations, as
shown in Tables 3 and 4. The two lowest-energy confor-
mations calculated in chloroform (C1and C2) and the
lowest (W1) in water are depicted in Figure 1. From Fig-
ure 1 and Table 2, it can be seen that the 15–16–17–18
cis-amide bond between Tyr and Val and the 28–29–
30–1 bond in the Hmp unit have angles close to zero
in all of these conformations, and that there is not much
variation in the conformation of the shorter (13-atom)
Val-Gly1-Leu-Gly2 connecting chain; the largest change
in this shorter chain comes in C2, where the 25–26–27–
28 and 26–27–28–29 angles in Gly2 are 109ꢁ and ꢀ110ꢁ,
respectively. This shorter chain of ring atoms is nearly

4140
M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
Table 1. NOE peaks of dolastatin 11 (1) and ROESY peaks of 7-epidolastatin 11 (6)
Protona
Proton #
NOE peaks in 1
ROESY peaks in 6
Map-a-CH
Map-b-CH3
Map-b-CH
Map-c-CH2
Map-c-CH2
Map-d-CH3
Map-NH
3
3a
3a, 4, 4a0, 4b
3, 4, 5
3, 3a, 4a, 4a0, 4b
4, 4b, 5
3, 4, 4b
3, 4, 4a, 4a0, 5
3a, 4a, 4b, 7
5,b 7a, 8
7, 8
3a, 4
3, 4, 4a0, 5, 28
3, 3a, 4a, 4a0, 4b, 8
4, 4a0, 4b, 5
4
4a
4a0
3a, 4, 4a, 4b
4, 4a, 4a0, 5
4b
5
3a, 4a, 4b, 7, 28
5, 7a
7, 8
Ala-a-CH
Ala-b-CH3
Ala-NH
7
7a
8
7, 7a, 10b
12
4, 7a, 10a, 10b
8, 10b
8, 10a, 12
Ibu-b-CH3
Ibu-b-CH3
Ibu-c-CH
Ibu-d-CH3
Ibu-NH
10a
10b
12
8
10a, 12a, 13
12, 13
12, 12a, 15
10b, 12a, 13
12
12a
13
12, 15, 19a
Tyr-a-CH
Tyr-b-CH2
Tyr-b-CH2
Tyr-d-CH
Tyr-e-CH
OCH3
15
13, 15a, 15aꢀ, 15c, 18
15, 15a0, 15c
15, 15a, 15c
13, 15a 15a0, 15c, 18
15, 15a0, 15c, 18
15, 15a, 15c
15a
15a0
15c
15d
15f
16a
18
15, 15a, 15a0, 15d, 16a, 18, 18c
15, 15a, 15a0, 15d, 16a, 18, 18b
15c, 15f, 18c
15d
15c
15c, 15f, 18b
15d
Tyr-NCH3
Val-a-CH
Val-b-CH
Val-c-CH3
Val-c-CH3
Val-NCH3
Gly1-a-CH2
Gly1-a-CH2
Gly1-NH
15c
15, 15a, 15c, 18b, 18c
18b, 18c, 19a
15, 15c, 18a, 18b, 18c
18, 18b, 18c, 19a
18, 18a, 19a
18a
18b
18c
19a
21a
21b
22
15c, 15d, 18, 18a, 18c, 19a
18, 18a, 18b, 19a
13, 18a, 18b, 18c, 21a, 21b
19a, 21b, 22
15c, 15d, 18, 18a
18a, 18b, 21a, 21b
19a, 21b, 22
19a, 21a, 22
19a, 21a, 22
21a, 21b, 24
21a, 21b, 24
Leu-a-CH
Leu-b-CH2
Leu-b-CH2
Leu-c-CH
Leu-d-CH3
Leu-d-CH3
Leu-NCH3
Gly2-a-CH2
Gly2-a-CH2
Gly2-NH
24
22, 24a, 24a0, 24b, 24c
24, 24a0, 24b, 24c, 24d, 25a
24, 24a, 24b, 24c, 24d, 25a
24, 24a, 24a0
24, 24a, 24a0
24a, 24a0
22, 24a, 24a0, 24c
24, 24a0, 24c
24a
24a0
24b
24c
24d
25a
27a
27b
28
24, 24a, 24b, 24d, 25a
24a0, 24c, 24d, 25a
24, 24a, 24b
24a0, 24b
24a, 24a0, 27a
24a0, 24b, 27a, 27b
25a, 27b, 28
25a, 27a, 28
25a, 27b, 28
27a, 28
27a, 27b, 30
3a, 5, 27a, 27b, 30
28, 30a, 30b, 30c, 30d
30, 30b, 30b0, 30c, 30d
30, 30a, 30b0, 30c, 30d
30a, 30b, 30c, 30d
30, 30a, 30b, 30b0
30, 30a, 30b, 30b0
Hmp-a-CH
Hmp-b-CH
Hmp-c-CH2
Hmp-c-CH2
Hmp-d-CH3
Hmp-c-CH3
30
28, 30a, 30b, 30b0, 30c, 30d
30, 30b, 30b0, 30c, 30d
30, 30a, 30bꢀ, 30c, 30d
30, 30a, 30b, 30c, 30d
30, 30a, 30b, 30b0
30, 30a, 30b, 30b0
30a
30b
30b0
30c
30d
a Within CH2s and C(CH3)2s, upfield absorbing group is listed first.
b Italicized numbers in the table indicate the main peaks which show amide configurations.
extended to maximum length, with most angles near
180ꢁ. In contrast, the longer (17-atom) Hmp-Map-Ala-
Ibu-Tyr chain shows much variation among these eight
ring conformations. The conformations divide into two
groups (+ and ꢀ) based on the sign of the 2–3–4–5 angle
(around the only sp3-sp3 bond in the large ring), which is
either about +60ꢁ or ꢀ60ꢁ; the significance of this is dis-
cussed below in the section on analogues.
provide similar five-membered ring hydrogen bonds
in all eight conformations. C2, with its four 10- to
15-membered cross-ring hydrogen bonds of the b-sheet
type, is very different from all of the others. C7 differs
from C4 only in the Ibu-NH to Ibu-1-CO bond, C15
from C1 only in the Gly-1-NH to Ibu-1-CO bond,
W1 from C3 only in the Map-NH to Gly-2-CO bond,
and W5 from W1 largely in the Gly-1-NH to Ibu-1-
CO bond.
Table 5 shows nitrogen to carbonyl oxygen distances
for likely intramolecular hydrogen bonds in the eight
30-membered ring conformations; not shown in Table
5 are the 2.24–2.36A distances between Gly-2-N and
O1 (the non-carbonyl ester oxygen) which probably
All of the NMR evidence, including the NOEꢀs shown in
Table 1 for dolastatin 11 (1) and strong additional evi-
dence from the spectra of restricted rotation derivatives
given below, is consistent with the view that its predom-
˚

M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
4141
Table 2. Thirty-membered ring torsion angles for dolastatin 11 (1) conformations C1–C4, C7, C15, W1, and W5
Unit
Map
Angle
C1
C2
C3
C4
C7
C15
W1
14
W5
1–2–3–4
2–3–4–5
ꢀ122
76
82
ꢀ42
ꢀ30
ꢀ23
ꢀ21
ꢀ115
69
76
21
ꢀ85
120
ꢀ86
88
ꢀ81
99
ꢀ64
147
ꢀ65
149
ꢀ79
125
3–4–5–6
4–5–6–7
173
95
174
99
171
168
168
127
170
124
171
106
173
127
180
133
Ala
Ibu
5–6–7–8
6–7–8–9
7–8–9–10
ꢀ157
172
ꢀ155
173
72
ꢀ167
ꢀ166
141
ꢀ167
180
112
ꢀ168
178
165
ꢀ164
171
154
ꢀ164
ꢀ177
126
ꢀ158
ꢀ174
129
8–9–10–11
9–10–11–12
10–11–12–13
11–12–13–14
12–13–14–15
13–14–15–16
14–15–16–17
15–16–17–18
16–17–18–19
17–18–19–20
18–19–20–21
19–20–21–22
20–21–22–23
21–22–23–24
22–23–24–25
23–24–25–26
24–25–26–27
25–26–27–28
26–27–28–29
27–28–29–30
28–29–30–1
29–30–1–2
30–1–2–3
ꢀ159
ꢀ93
56
39
131
ꢀ50
109
ꢀ47
96
ꢀ77
64
ꢀ68
71
ꢀ46
109
ꢀ47
108
ꢀ146
178
126
ꢀ158
171
110
ꢀ153
175
76
ꢀ163
ꢀ179
62
ꢀ153
ꢀ175
107
ꢀ158
ꢀ178
100
ꢀ158
180
73
ꢀ157
180
89
Tyr
ꢀ126
ꢀ2
ꢀ121
0
103
ꢀ125
ꢀ11
85
ꢀ139
ꢀ5
ꢀ128
ꢀ7
ꢀ135
0
81
ꢀ130
ꢀ6
ꢀ136
ꢀ8
Val
80
94
84
86
78
ꢀ129
178
ꢀ106
ꢀ175
ꢀ158
ꢀ175
ꢀ177
109
ꢀ133
ꢀ179
ꢀ150
ꢀ148
175
ꢀ127
ꢀ176
ꢀ167
ꢀ178
176
ꢀ137
ꢀ175
ꢀ176
ꢀ176
180
ꢀ140
ꢀ176
ꢀ176
ꢀ176
ꢀ179
120
ꢀ131
ꢀ179
ꢀ151
ꢀ147
177
ꢀ137
173
Gly-1
Leu
ꢀ163
ꢀ170
172
ꢀ150
177
ꢀ179
158
117
120
113
ꢀ137
177
109
ꢀ133
180
120
ꢀ143
176
ꢀ150
ꢀ177
ꢀ174
ꢀ177
ꢀ179
ꢀ1
ꢀ126
178
109
ꢀ147
ꢀ178
ꢀ175
ꢀ164
171
ꢀ121
ꢀ178
ꢀ172
ꢀ175
178
ꢀ126
ꢀ179
ꢀ166
179
Gly-2
Hmp
ꢀ170
ꢀ170
172
ꢀ170
ꢀ171
171
ꢀ160
179
179
ꢀ110
177
179
ꢀ10
ꢀ69
ꢀ169
ꢀ7
ꢀ7
ꢀ8
ꢀ1
ꢀ32
ꢀ72
171
ꢀ35
ꢀ73
173
ꢀ82
ꢀ173
ꢀ68
ꢀ176
ꢀ71
176
ꢀ71
176
ꢀ81
ꢀ172
Table 3. Energies (kJ/mol>minimum in that solvent), sign of angle 2–3–4–5, 30-membered ring type, and side chain conformations of the 25
conformations calculated to be within 10 kJ/mol of the minima for dolastatin 11 (1) in chloroform (C) and water (W); see Table 4 for key to side
chain rotation numbers
Conformation
2–3–4–5 Sign
Ring type
Side chain rotations
Map
Tyr
Val
Leu
Hmp
v1
v1
v1
v1
v2
v1
v2
C1 (0) = W2 (1.6)
C2 (2.2) = W18 (9.9)
C3 (5.4) = W3 (2.7)
C4 (7.7) = W4 (5.1)
C5 (7.9) = W11 (8.3)
C6 (8.0)
+
C1
C2
1
1
1
2
1
1
2
1
1
1
1
1
1
2
1
1
1
1
1
3
1
2
2
2
1
1
2
2
2
1
1
1
1
1
1
2
1
1
2
1
2
2
2
1
2
1
2
1
2
2
1
1
1
1
1
1
1
1
1
2
1
1
1
1
2
1
2
1
1
1
1
1
1
1
2
1
1
1
2
1
1
2
1
1
1
2
1
1
1
2
1
2
1
2
2
1
1
2
1
2
1
1
1
2
1
3
2
2
1
1
2
1
1
1
2
1
2
1
2
2
3
1
2
1
2
1
1
1
1
1
1
1
1
2
1
1
3
3
1
1
2
2
1
2
1
1
2
2
2
2
1
1
2
1
3
1
1
1
1
1
1
1
1
1
1
1
1
3
1
1
1
1
3
3
3
ꢀ
ꢀ
ꢀ
+
C3
C4
C1
C1
+
C7 (8.3)
C8 (8.3)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
+
C7
C7
C9 (8.4) = W8 (7.1)
C10 (8.5) = W14 (9.1)
C11 (9.1)
C7
C7
C7
C3
C12 (9.3)
C13 (9.4)
C14 (9.7)
C1
C7
ꢀ
+
C15 (9.8) = W7 (6.8)
W1 (0)
W5 (5.2)
C15
W1
W5
W1
C15
W5
C7
ꢀ
ꢀ
ꢀ
+
W6 (6.0)
W9 (7.1)
W10 (7.9)
W12 (8.9)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
+
W13 (9.1)
W15 (9.6)
W1
W1
W1
C15
W16 (9.6)
W17 (9.9)

4142
M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
Table 4. Key to side chain rotation numbers in Table 3
Unit
Map
1
2
3
Far Me anti
to CH
Far Me anti
to N
Far Me anti
to H
—
Tyr v1
Val v1
Leu v1
v2
Ar anti to CO
Hꢀs anti
i-Pr anti to N
Ar anti to N
H anti to N
i-Pr anti to CO
—
—
Me anti to CH Me0 anti to CH H anti to CH
Hmp v1 Et anti to H
Et anti to CO
Far Me anti
to Me
Et anti to OCO
Far Me anti
to H
v2
Far Me anti
to CH
inant conformation in chloroform solution is C1. It was
not possible to determine its preferred conformation(s)
in water by NMR due to its low solubility.
3. Synthesis of dolastatin 11 (1) analogues 2–21
Analogues 2–21 were made by syntheses following
that of dolastatin 11 (1)2 with three improvements: (1)
In the last step in the preparation of Map, the p-toluene-
sulfonyl group was removed not with HBr but with
sodium in liquid ammonia. (2) In the workup of the
preparation of N-Me-L-Val-O,N-diMe-L-Tyr-OBnÆHCl,
evaporation of most of the benzyl alcohol and addi-
tion of ether caused the product to crystallize and
avoided multiple extractions and lyophilization. (3)
Some of the final cyclizations were carried out
at higher dilution (Procedure B, Experimental Section),
doubling the yield from the earlier procedure (Proce-
dure A).2
The NMR data on these analogues are summarized in
Table 6. Analogue 20, the last intermediate in the dolast-
atin 11 (1) synthesis, is not included; its NMR parame-
ters were reported earlier.2 The spectra of analogues 19
and 20, which lack the 30-membered ring but have three
N-methyl amide groupings, are complicated by the
occurrence of significant amounts of up to eight rota-
mers; when the 30-membered ring is closed, as noted
earlier, the 16,17-cis, 19,20-trans, 25,26-trans rotamer
predominates greatly over the others, making the spec-
tra much easier to interpret. 2D NMR studies were car-
ried out on analogues 6 and 7, and the assignments for
the other analogues were made where possible by com-
parisons with these and the spectra of dolastatin 11
Figure 1. The two lowest-energy conformations calculated for dolast-
atin 11 (1) in chloroform (C1, top; C2, middle) and the lowest in water
(W1, bottom). Dashed lines represent intramolecular hydrogen bonds.
(1). ꢁObscꢀ means the absorption was obscured and
asterisks mean the assignments may be reversed with
Table 5. Intramolecular hydrogen bonding (with N to carbonyl O distances in Angstroms) in the eight lowest-energy 30-membered ring
˚
conformations of dolastatin 11 (1); weak bonds (2.7–3.1 A) are in parentheses
Conformation
Map-N to
Gly-2
Ala-N to
Gly-2
Ibu-N to
Ibu-3
Gly-1-N to
Ibu-1
Gly-2-N to
Gly-2 Ala
Ala
Ibu-1
Gly-1
Gly-1
Ibu-3
H-bond ring size ! 12
5
15
—
7
5
—
10
5
16
2.08
—
14
—
1.83
—
—
—
—
—
—
5
11
—
C1 = W2
C2 = W18
C3 = W3
C4 = W4
C7
1.81
—
(2.84)
(2.83)
2.12
2.41
2.44
(2.84)
2.43
2.37
1.97
—
(3.03)
1.89
—
2.16
2.30
2.43
2.17
2.10
2.14
2.50
2.15
2.08
(2.90)
2.21
2.12
2.11
2.12
2.10
2.10
1.86
2.29
2.63
(2.89)
—
1.76
2.10
—
—
—
—
—
—
—
(2.90)
(2.88)
2.02
2.03
(3.08)
(3.01)
1.99
(2.79)
(2.79)
(2.90)
1.95
—
1.98
1.95
1.81
2.69
—
—
—
—
—
—
—
—
C15 = W7
W1
W5
—
—
—
2.63
2.66
—

Table 6. 1H NMR shifts (d, CDCl3), multiplicities, and coupling constants (J, in parentheses) for compounds 1–19 and 21
Dolastatin 11 (1)
3-Nor (2)
3,4-Bis-nor (3)
R,R-Acc (4)
S,S-Acc (5)
7-Epi (6)
7-Nor (7)
Map
a-CH
2.79qd(7,2.5)
CH2 2.33dd(14.5,11)
2.73dd(14.5,2.5)
CH2
2.62ddd(16,10,2)
3.33q(8)
2.49q(8)
3.03 m
2.83 m
2.74ddd(16,8,3)
—
b-CH3
b-CH
c1-CH2
c2-CH2
d-CH3
NH
1.10d(7)
4.47m
1.46m
—
CH2 1.77,2.17m
4.90m
2.18m
CH2 1.80,2.10m
4.82m
2.20m
1.23d(7)
3.77m
1.25m
1.15d(7)
4.24m
1.38m
4.91m
1.46m
1.56m
0.95t(7)
7.09d(9)
CH2 3.60,3.85m
—
1.56m
0.93t(7.5)
7.09d(10.5)
—
—
2.18m
CH2 1.50m
7.31m
2.20m
CH2 1.50m
6.75d(9)
1.76dqd(14,7,3)
0.98t(7)
8.17d(7)
1.63m
0.96t(7.5)
7.42d(9)
7.47br t(7)
Ala
a-CH
b-CH3
NH
4.44p(7)
1.07d(7)
7.78d(8.5)
4.44p(6.5)
1.07d(7)
7.79d(8.5)
4.65p(7)
1.25d(7)
7.18d(7)
4.63p(7)
1.22d(7)
6.99d(7)
4.45p(6.5)
1.07d(6.5)
7.83d(9)
4.56p(7)
1.25d(7)
6.78d(8)
CH2 3.59dd(16,5) 4.05dd(16,6)
—
7.53t(5.5)
Ibu
b1-CH3
b2-CH3
c-CH
d-CH3
NH
1.44s
1.49s
1.45s
1.48s
1.38s
1.57s
1.30s
1.49s
1.43s
1.43s
1.39s
1.57s
1.45s
1.53s
4.91p(7)
1.13d(7)
7.15d(9)
4.93p(7)
1.13d(7)
7.13d(9)
4.98p(7)
1.30d(7)
6.49d(7.5)
4.88p(7)
1.41d(7)
6.45d(8)
4.93p(7)
1.09d(7)
7.30d(7)
4.88p(6.5)
1.29d(6.5)
6.50d(7)
4.89p(7)
1.20d(7)
6.72d(7)
Tyr
a-CH
b1-CH2
b2-CH2
d-Ar
e-Ar
OCH3
NCH3
5.11dd(8.5,6.5)
2.82dd(14,8.5)
3.25dd(14,6.5)
7.14d(8.5)
6.81d(8.5)
3.75s
5.12dd(8,6.5)
2.84dd(14.5,8)
3.25dd(14.5,6.5)
7.15d(8.5)
6.82d(8.5)
3.76s
5.13dd(10.5,5)
2.79dd(15,10.5)
3.37dd(15,5)
7.14br d(8.5)
6.82br d(8.5)
3.75s
5.05d(9,4)
2.75dd(14,9)
3.38dd(14,4)
7.14d(9)
6.82d(9)
3.75s
5.08t(7.5)
2.84dd(15,8)
3.24dd(15,7)
7.14d(8.5)
6.82d(8.5)
3.76s
5.07dd(11,4)
2.76dd(14.5,11)
3.43dd(14.5,4)
7.16d(8.5)
6.83d(8.5)
3.76s
5.08dd(9.5,5)
2.81dd(14.5,9.5)
3.34dd(14.5,5)
7.15d(8.5)
6.83d(8.5)
3.77s
2.96s
2.96s
2.95s*
2.92s*
3.00s*
2.89s
2.93s*
Val
a-CH
b-CH
c1-CH3
c2-CH3
NCH3
4.78d(10.5)
2.23m
4.79d(10.5)
2.24m
4.59d(10.5)
2.16dh(10.5,6.5)
0.17d(6.5)
0.65d(6.5)
2.89s*
4.60d(11)
2.16d(11)
0.22d(7)
0.66d(6.5)
2.90s*
4.88d(10.5)
2.23m
4.53d(10.5)
2.15dh(10.5,6.5)
0.17d(6.5)
0.63d(6.5)
2.87s
4.64d(10.5)
2.20dh(10.5,6.5)
0.28d(6.5)
0.68d(7)
0.37d(6.5)
0.74d(6.5)
2.95s
0.38d(6.5)
0.75d(6.5)
2.94s
0.45d(6.5)
0.78d(6.5)
2.98s*
2.91s*
Gly-1
a1-CH2
a2-CH2
NH
3.60dd(18,2)
4.42dd(18,7.5)
7.41dd(7.5, 2)
3.60dd(17.5,1.5)
4.42dd(17.5,7.5)
7.42dd(7.5,1.5)
3.75dd(17,2.5)
4.48dd(17,7)
6.84dd(7,2.5)
3.96dd(17.5,3)
4.24dd(17.5,4)
7.12m*
3.65dd(16,2)
4.40dd(16,8)
7.59dd(8,2)
3.86dd(17.5,3.5)
4.12dd(17.5,7)
6.94dd(7,3.5)
3.79dd(18,3)
4.42dd(18,6.5)
7.09dd(6.5,3)
Leu
a-CH
b1-CH2
b2-CH2
c-CH
5.37dd(11,5)
1.62m
5.37dd(11,5)
1.62m
5.20dd(11.5,5)
1.80ddd(13.5,8,5)
1.92ddd(13.5,11.5,4)
Obsc
5.26dd(9.5,6.5)
1.78m
1.96m
5.42dd(11,5)
1.62m
1.87m
5.35dd(11.5,4.5)
1.73ddd(14.5,10,4.5)
1.96ddd(14.5,11.5,4)
1.54m
5.32dd(11.5,5)
1.65ddd(15,10,5)
1.93ddd(15,11.5,4)
1.56m
1.87ddd(13.5,11,4)
1.56m
0.92d(6.5)
1.89ddd(13.5,11,4)
1.56m
0.94d(6)
1.55m
0.96d(6.5)
1.55m
0.91d(6.5)
d1-CH3
0.97d(6.5)
0.97d(6.5)
0.96d(6)
(continued on next page)

Table 6 (continued)
Dolastatin 11 (1)
3-Nor (2)
3,4-Bis-nor (3)
R,R-Acc (4)
S,S-Acc (5)
7-Epi (6)
7-Nor (7)
d2-CH3
NCH3
0.98d(6.5)
3.14s
1.00d(6)
3.16s
1.03d(6.5)
3.09s
1.01d(6.5)
3.00s
0.99d(6.5)
3.22s
1.02d(6.5)
3.05s
1.00d(6.5)
3.08s
Gly-2
a1-CH2
a2-CH2
NH
3.58dd(16,4.5)
4.46dd(16,7)
7.35dd(7,4.5)
3.58dd(16,5)
4.46dd(16,7)
Obsc
3.62dd(17.5,4)
4.22dd(17.5,7.5)
7.33dd(7.5,4)
3.78dd(16.5,4)
4.26dd(16.5,4)
7.23m*
3.50dd(17,4.5)
4.59dd(17,7)
7.19m
4.01dd(18,3.5)
4.24dd(18,5)
7.14m
3.76dd(17,5.5)
4.38dd(17,6)
7.15t(6)
Hmp
a-CH
b-CH
c1-CH2
c2-CH2
c-CH3
d-CH3
5.19d(3.5)
2.07m
1.23m
5.24d(3)
2.07m
1.22m
5.18 d(4)
2.07m
1.25m
5.30d(3.5)
2.08m
1.17m
5.15d(3)
2.06m
1.27m
5.36d(3)
2.08m
1.19m
5.20d(3)
2.07m
1.27m
1.47m
0.89d(7)
0.88t(7.5)
1.47m
0.89d(8)
0.88t(7)
1.48m
0.95d(7)
0.90t(7.5)
1.48m
0.91d(7)
0.88t(7.5)
1.47m
0.90d(7)
0.90t(7.5)
1.42m
0.89d(6.5)
0.87t(7.5)
1.47m
0.89d(6.5)
0.89t(7.5)
12-Epi-7-nor (7a)
10-Nor (8)
10,10-Bis-nor (9)
11R-OH (10)
11R-OAc (11)
12-Epi (12)
Majusculamide C (13)
Map
a-CH
b-CH3
b-CH
c1-CH2
c2-CH2
d-CH3
NH
2.80m
Obsc
2.88qd(7,2.5)
1.16d(7)
4.22m
2.73qd(7,2.5)
1.14d(7.5)*
4.66m
3.12qd(7.5,2.5)
1.26d(7)
3.71m
2.75m
1.12d(7)
Obsc
2.80qd(8,2.5)
1.11d(7)
4.52m
1.15d(7)
4.37m
Obsc
1.16d(7)
4.53m
Obsc
Obsc
Obsc
Obsc
Obsc
Obsc
Obsc
Obsc
1.46m
1.56m
Obsc
0.94t(7)
6.95d(10)
Obsc
0.88-0.95
7.04d(8)*
1.78dqd(14,7.5,3)
0.94t(7.5)
7.72d(6.5)
0.96t(7)
6.98br d(4)*
0.94t(7.5)
6.72d(9)
0.95t(7)
6.91d(10)
0.94t(7.5)
7.08d(9.5)
Ala
a-CH
b-CH3
NH
CH2 3.56dd(16,6) 4.12dd(16,6)
—
7.53t(6)
4.50p(7)*
1.20d(7.5)*
7.62d(8)
4.54m
1.27d(7)*
Obsc
4.45p(7)
1.08d(7)*
7.48d(7.5)
4.73p(6.5)
1.01d(6.5)
7.00d(6.5)
4.61p(7)
1.08d(7)
7.52d(8)
4.44p(6.5)
1.06d(6)
7.80d(8.5)
Ibu
b-CH
b1-CH3
b2-CH3
c-CH
d-CH3
NH
—
1.46s
1.55s
—
—
3.54br s
1.17s
1.38s
4.84d(2.5)
1.11s
1.29s
—
1.52s
1.52s
—
1.49s
1.51s
1.19d(7)*
a-CH obsc
4.50p(7)*
1.33d(7)*
Obsc
a-CH2 3.29d(15)
a-CH2 3.32d(15)
4.54m
1.34d(7.5)*
6.91d(7)*
—
4.90p(7)
1.25d(6.5)
7.28d(8)
—
4.18br p(6.5)
0.81d(6.5)
7.32d(10.5)
—
4.01br p(6.5)
1.50d(6.5)
6.66d(8.5)
2.14s
4.98p(6.5)
1.22d(6.5)
7.31d(8)
—
4.93p(7)
1.14d(7)
7.34d(5.5)
—
Ac
—
Tyr
a-CH
b1-CH2
b2-CH2
d-Ar
e-Ar
OCH3
NCH3
4.98m
2.80m
5.08t(7.5)
2.71dd(14,7)
3.50dd(14,8)
7.21d(8)
6.92d(8)
3.86s
5.20t(7.5)
2.75dd(14,8)
3.48dd(14,7)
7.19d(8.5)
6.87d(8.5)
3.80s
4.92t(7)
4.95dd(11,3)
2.71dd(14.5,11)
3.44dd(14.5,3)
7.15d(8.5)
6.81d(8.5)
3.74s
5.04t(8)
5.16t(7.5)
2.82dd(14,7.5)
3.26dd(14,7)
7.15d(8.5)
6.82d(8.5)
3.76s
2.89dd(14,8)
3.20dd(14,6.5)
7.15d(8.5)
6.83d(8.5)
3.77s
2.76-2.79m
3.27dd(13.5,8)
7.14d(9)
6.83d(9)
3.77s
3.24dd(14,8)
7.15d(8.5)
6.83d(8.5)
3.77s
3.00s*
2.88s*
2.99s*
3.02s*
2.81s*
2.99s
2.97s

Val
a-CH
b-CH
c1-CH3
c2-CH3
NCH3
4.95d(10.5)
2.28dh(10.5,6.5)
0.50d(6.5)
0.79d(6.5)
2.98s*
4.93d(10.5)
2.31m
4.83d(11)
2.27m
4.83d(10.5)
2.23m
4.48d(10.5)
2.14m
4.89d(10.5)
2.28m
4.81d(11.5)
2.24m
0.65d(6.5)
0.81d(6.5)
2.95s*
0.47d(6.5)
0.76d(6.5)
2.90s*
0.40,0.29d(6.5)
0.74,0.62d(6.5)
2.98s*
0.14d(6.5)
0.62(6.5)
2.85s*
0.57d(6.5)
0.81d(6.5)
2.93s
0.42d(6.5)
0.76d(7)
2.96s
Gly1
a1-CH2
a2 -CH2
NH
3.65dd(17,2)
4.42dd(17,7)*
7.12m
3.67dd(18,2)*
4.43dd(18,7)
Obsc
3.76-3.82m
4.28dd(17.5,6.5)*
7.15br t(4)*
3.48dd(17.5,2)
4.52dd(17.5,8)
7.01br d(8)
3.91dd(18,2)
4.29dd(18,6)
7.00br d(6)
3.65dd(18,2)
4.44dd(18,7.5)
Obsc
3.93dd(17,2)
4.63dd(17,8)
7.60br d(8)
Leu
a-CH
5.20dd(10,5)
1.62m
5.36dd(9,6.5)
Obsc
1.72m
5.13dd(11,5)
Obsc
5.08dd(10,5)
Obsc
5.39dd(11,5)
1.70ddd(14.5,10,4)
1.91ddd(14.5,11,3.5)
1.78m
5.20dd(10.5,5.5)
Obsc
4.91d(10.5)
CH 2.08m
–
b1-CH2
b2-CH2
c-CH
1.82ddd(13,10,5)
Obsc
1.80ddd(14,11,4)
Obsc
Obsc
1.81ddd(14.5,10,4)
Obsc
1.81ddd(13,10.5,4)
Obsc
Obsc
CH2 1.08,1.43m
0.92d(6.5)
c-CH3 1.05d(6.5)
3.23s
d1-CH3
d2-CH3
NCH3
0.87–0.90
0.95d(6.5)
3.06s
0.88–0.95
0.88–0.95
3.04s
0.90d(6.5)
0.98d(6)
3.07s
0.94d(6.5)
0.98d(6.5)
2.94s
0.89–0.94
0.89–0.94
3.10s
Obsc
3.04s
Gly2
a1-CH2
a2-CH2
NH
3.80dd(17,2)
4.34dd(17,7)*
7.22m
3.69dd(17.5,2)*
4.50–4.54m
Obsc
3.76-3.82m
4.38dd(17,6.5)*
7.11br t(6)*
3.90dd(16,5.5)
4.37dd(16,6)
7.39t(6)
4.12dd(17,3.5)
4.34dd(17,7.5)
7.15dd(7.5,3.5)
3.71dd(16.5,4.5)
4.48dd(16.5,7)
7.34dd(7,4.5)
3.56dd(16,7)
4.45dd(16,6.5)
7.38t(6.5)
Hmp
a-CH
b-CH
5.28d(3)
2.08m
Obsc
5.22d(4)
1.99m
Obsc
5.22d(3)
2.08m
Obsc
5.35d(3)
2.09m
1.22m
5.32d(3.5)
2.04m
1.22m
5.27d(3)
2.07m
Obsc
5.21d(3.5)
2.09m
1.25m
c1-CH2
c2-CH2
c-CH3
d-CH3
1.49m
0.87-0.90
0.87-0.90
Obsc
0.88-0.95
0.88-0.95
Obsc
0.90d(7.5)
0.90t(7.5)
Obsc
0.88d(7)
0.88t(7.5)
1.44m
0.90d(7)
0.88t(7.5)
Obsc
0.89d(7.5)
0.88t(7.5)
1.44m
0.90d(6.5)
0.89t(7)
57-Nor (14)
3,4,7-Tris-nor (15)
3,4-Bis-nor-11-OH (16)
7-Nor-11-OH (17)
3,4,7-Tris-nor-11-OH (18)
1,2-Seco (19)
des-Hmp (21)
Map
a-CH
2.81m
CH2
2.62ddd(16.5,8,2.5)
CH2 2.65,2.66m
Obsc
CH2 2.66m
Obsc
2.72qd(7,4)
2.70ddd(16.5,9,3)
—
b-CH3
b-CH
c1-CH2
c2-CH2
d-CH3
NH
1.11d(7)
4.50m
Obsc
—
CH2 3.65,3.74m
1.17d(7)
4.30m
Obsc
—
CH2 Obsc
Obsc
Obsc
1.41d(7)*
Obsc
Obsc
CH2 3.52,3.95m
—
—
—
—
—
Obsc
Obsc
Obsc
0.95t(7.5)
7.08d(10)
—
—
Obsc
0.89–0.93
7.40d(9)
Obsc
0.88t(7.5)
6.72d(7)*
—
7.56t(6)
—
Obsc
0.85–1.05
6.28d(8)*
7.38d(9)
Ala
a-CH
4.45p(7)
CH2 3.84dd(17,4)
4.18dd(17,4)
—
4.70p(7)
CH2 3.64dd(15.5,5.5)
4.03dd(15.5,5.5)
—
CH2 3.61dd(16.5,3)
4.20dd(16.5,6.5)
Obsc
4.41p(7)
b-CH3
NH
1.08d(7)
7.74d(8.5)
1.21d(6.5)
6.78d(8)
—
Obsc
Obsc
7.32,7.40m*
1.19d(7)*
7.50d(8.5)
6.99br t(4)
7.41br s
(continued on next page)

Table 6 (continued)
57-Nor (14)
3,4,7-Tris-nor (15)
3,4-Bis-nor-11-OH (16)
7-Nor-11-OH (17)
3,4,7-Tris-nor-11-OH (18)
1,2-Seco (19)
des-Hmp (21)
Ibu
b-CH
b1-CH3
b2-CH3
c-CH
d-CH3
NH
—
1.45s
1.51s
—
1.41s
1.59s
3.53dd(5,2.5)
1.17s
1.31s
3.56d(2)
1.20s
1.34s
3.55m
—
Obsc
—
1.34s
1.57s
1.21,1.16s
1.30,1.21s
4.05m
Obsc
Obsc
4.92p(7)
1.14d(7)
7.14m
4.94p(7)
1.26d(7)
6.44d(7)
4.01br p(7)
0.88,1.08d(6.5)
Obsc
4.05br p(7)
0.89–0.93
6.65d(8.5)
4.95p(7)
1.16d(7)*
6.83d(6)*
0.89–0.98,1.08d(7)
Obsc
Obsc
6.47,6.54m*
Tyr
a-CH
b1-CH2
b2-CH2
d-Ar
e-Ar
OCH3
NCH3
5.13t(7.5)
2.83dd(14,8)
3.26dd(14,7)
7.15d(8.5)
6.82d(8.5)
3.76s
5.10dd(10.5,4)
2.79dd(14.5,10.5)
3.38dd(14.5,4)
7.13d(8.5)
6.82d(8.5)
3.74s
4.89dd(9,6)
2.87dd(14,9)
3.25dd(14,6)
7.15d(8.5)
6.83d(8.5)
3.76s
4.86dd(8,6.5)
Obsc
4.86m
Obsc
Obsc
Obsc
Obsc
5.04dd(9,6)
2.75dd(14.5,9)
3.44dd(14.5,6)
7.15d(8.5)
6.84d(8.5)
3.76s
3.24dd(14,6)
7.15d(8.5)
6.83d(8.5)
3.77s
3.27,3.18dd(14,6)
7.15,7.13d(8.5)
6.83d(8.5)
3.76,3.77s
2.98,2.95s*
7.10-7.13d(8.5)
6.83d(8.5)
3.74s
2.95,2.98s*
2.96s*
2.93
2.99,2.93s*
3.00,2.85s*
2.95s*
Val
a-CH
b-CH
c1-CH3
c2-CH3
NCH3
4.80d(10.5)
2.25dh(10.5,6.5)
0.39d(6.5)
0.76d(6.5)
2.98s*
4.58d(10.5)
2.15dh(10.5,6.5)
0.16d(6.5)
0.63d(6.5)
2.88s
4.85d(10)
4.80d(10.5)
4.85,4.72d(10.5)
2.22m
5.10d(10.5)
Obsc
4.67d(10.5)
2.22m
2.22dh(10,6.5)
0.39,0.27d(6.5)
0.73,0.67d(6.5)
2.94,2.89s*
2.20dh(10.5,6.5)
0.38,0.16d(6.5)
0.73,0.60d(6.5)
2.96,2.84s*
0.42,0.40d(6.5)
0.75,0.71d(6.5)
2.97,2.91s*
0.27,0.48d(6.5)
0.68,0.70,0.75d(6.5)
2.88,2.89,2.90s*
0.37d(6.5)
0.70d(6.5)
2.94s*
Gly1
a1-CH2
a2-CH2
NH
3.61dd(18,2)
4.42dd(18,7.5)
7.41br d(7.5)
3.81dd(17.5,2)
4.42dd(17.5,6.5)
6.75dd(6.5,2)
3.75m*
3.69dd(17.5,2)
4.36dd(17.5,6.5)*
7.01m*
3.69dd(17.5,2)
4.35,4.45dd(17.5,8)
Obsc
Obsc
Obsc
6.90,6.95d(7)*
3.94dd(17.5, 2)
4.37dd(17.5, 7)
7.06br d(7)
4.44dd(17.5,6.5)
6.96m*
Leu
a-CH
5.38dd(11,5)
1.65m
5.24dd(11,5)
1.81ddd(14.5,9.5,5)
1.93ddd(14.5,11,4.5)
Obsc
5.38dd(10.5,6)
1.69ddd(15,9,6)
1.86ddd(15,10.5,5)
Obsc
5.28dd(10,4.5)
1.60ddd(14.5,10.5,5.5)
1.85ddd(14.5,10,4)
Obsc
5.39,5.36dd(10,5)
Obsc
Obsc
Obsc
Obsc
5.38dd(10,6)
Obsc
Obsc
b1-CH2
b2-CH2
c-CH
1.88ddd(14,11,4.5)
Obsc
0.94d(7)*
1.00d(6.5)
3.16s
Obsc
Obsc
Obsc
Obsc
d1-CH3
d2-CH3
NCH3
0.97d(6.5)*
0.93d(6.5)*
0.89–0.93
0.98d(6.5)
3.02,2.91s*
0.89–0.98
0.89–0.98
3.02,2.98s
0.85–1.05
0.85–1.05
3.02,3.06s
0.92d(6.5)
0.98d(6.5)
3.01s
1.03d(6.5)
3.06s
0.99d(6.5)
3.01,2.97s*
Gly2
a1-CH2
a2-CH2
NH
3.59dd(16,4.5)
4.47dd(16,7)
7.34dd(7,4.5)
3.64dd(17.5,4)
4.19dd(17.5,7)
7.69dd(7,4)
3.87dd(17.5,2)*
4.29dd(18,6)
7.13m*
3.83dd(17,4)
4.33dd(17,6)*
7.17m*
Obsc
4.28,4.35dd(18,6)
Obsc
Obsc
Obsc
6.69d(7)*
3.55dd(17,3)
4.70dd(17,9)
6.59br d(9)
Hmp
a-CH
b-CH
5.15d(3.5)
2.34hd(7,3.5)
—
5.11d(4)
2.07m
1.25m
5.30d(2.5)
2.08m
Obsc
5.26d(4)
2.05m
Obsc
5.13d(4.5)
2.08m
Obsc
Obsc
Obsc
—
—
—
—
—
—
c1-CH2
c2-CH2
c-CH3
d-CH3
Obsc
Obsc
CH3 0.93d(7)
0.93d(7)*
—
1.49m
0.96d(7)*
0.91t(7.5)
Obsc
0.95d(6.5)*
0.90t(7.5)
Obsc
0.89–0.93
0.89t(7.5)
Obsc
0.89-0.98
0.89–0.98
0.85–1.05
0.85–1.05

M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
4147
similar ones in the same column. All analogues were
characterized by mass as well as NMR spectra.
was prepared by omitting the step in which Hmp is
added. All other analogues had the 30-membered ring
and showed at least some activity.
About half as much of the 7-epimer 6 as dolastatin 11
(1) was formed as a by-product in the dolastatin 11 (1)
synthesis. It was separated by HPLC; retention times
for 7-epidolastatin 11 (6) and dolastatin 11 (1) were
43.2 and 45.2 min, respectively. To establish its identity,
this by-product was also synthesized starting with Boc-
D-Ala for the Ala unit.
The synthesis of majusculamide C (13), isolated from
the blue-green alga Lyngbya majuscula,7 verified that it
differs from dolastatin 11 (1) only in the substitution
of Ile for Leu. It was as active as dolastatin 11 (1) in
most tests, but did not bind quite as well to actin.3
The Hmp analogue 57-nordolastatin 11 (14, analogous
to 57-normajusculamide C,8 which accompanies majus-
culamide C (13) in L. majuscula) was almost as active as
dolastatin 11 (1).
The alcohols 10 and 16–18 were made by sodium boro-
hydride reduction of the corresponding ketones (1, 3, 7,
and 15). The method of Schmidt et al.5 using ether/eth-
anol at ꢀ20 ꢁC was generally used; the method of Win-
terfeldt et al.6 gave similar results. In each case, a new
multiplet appeared in the NMR at d3.53–3.56 for the
Ibu-b-CH proton, and large changes in chemical shifts
occurred for the other Ibu protons. In the reduction of
dolastatin 11 (1), 3,4-bis-nordolastatin 11 (3), and 7-
nordolastatin (7), one stereoisomer predominated to
the extent of at least 80%, presumably the 11R stereoiso-
mer since the ketonic carbonyl carbon is much more
accessible from the side which gives the 11R alcohol in
the low-energy conformations. In contrast, 3,4,7-tris-
nordolastatin (15) gave about equal amounts of both
stereoisomers, and the mixture was tested in this case.
In Table 6, the NMR parameters of alcohols are given
first for the stereoisomer believed to be 11R, and then
the most characteristic peaks are given for what is pre-
sumed to be the 11S stereoisomer.
The other analogues involved changes in the Map, Ala,
and Ibu units. Dolastatin 11 (1) stereoisomers 7-epido-
lastatin 11 (6), and 12-epidolastatin 11 (12) were pre-
pared in an effort to learn the identity of the
stereoisomer formed in about a 1:2 ratio with dolastatin
11 (1) in the synthesis. It proved to be 7-epidolastatin 11
(6), indicating that considerable epimerization occurs
when the Ala-Map amide bond is made. We have not
been able to reduce the amount of epimerization in this
step of the synthesis of dolastatin 11 (1) itself by chang-
ing the coupling agent, but have avoided this type of by-
product in some derivatives by removing the Ala chiral
center as described below. It is not known why this epi-
merization occurs; it may be related to Map being a
b-amino acid and/or Hmp being a hydroxy acid. The
formation of 7-epidolastatin 11 (6) reduces the yield of
dolastatin 11 (1) and necessitates their separation by
HPLC (as noted earlier, 7-epidolastatin 11 (6) comes
off 2 min before dolastatin 11 (1) under our best condi-
tions). Dolastatin 11 (1) epimers 6 and 12 unexpectedly
had greatly reduced activity compared to dolastatin 11
(1); possible reasons will be discussed in the next section.
4. Biological testing of dolastastin 11 (1) and
analogues 2–21
A summary of the results is given in Table 7. All ana-
logues were tested against six human cancer cell lines;
the cell lines used most often are listed. Where it was
measured, the activity against P388 murine leukemia is
given as well. All analogues were also tested for their
ability to stimulate the polymerization of actin.3 These
results will be discussed in the next section.
3,4-Bis-nordolastatin 11 (3), in which b-alanine replaces
Map, was prepared with a view to shortening the synthe-
sis by the eight steps used to prepare Map. Because it
would be much more economical to synthesize than
dolastatin 11 (1), analogue 3 was chosen for further test-
ing. It binds strongly to actin, and on the average has
about 20% of the activity against 60 human tumors as
dolastatin 11 (1; Table 8). It was not accompanied by
an appreciable amount of a 7-epi compound, possibly
because the Ala-Map amide bond is formed much faster
without steric hindrance from a 4-ethyl group.
The results of testing dolastatin 11 (1) and analogue 3
against about 60 human cancer cell lines are given in
Table 8. Dolastatin 11 (1) shows GI50s between 1.3 nM
(NCI-H460 lung cancer) and 2.8 mM, whereas 3,4-bis-
nordolastatin 11 (3) has GI50s between 12 nM (COLO
205 colon cancer) and 15 mM. Surprisingly, the ana-
logue 3 has the best LC50, 0.24 mM, again with COLO
205 colon cancer.
3-Nordolastatin 11 (2) was prepared by leaving out a
methylation step in the preparation of Map. It is inter-
mediate between dolastatin 11 (1) and 3,4-bis-nordolast-
atin 11 (3) in activity showing that both the 4-ethyl
group and the 3-methyl group contribute to the activity.
5. Design and SAR of dolastatin 11 (1) analogues
Analogues 19–21, which lack the 30-membered ring,
were inactive in all tests, suggesting that this large ring
may be necessary for activity. Compound 19 is the
hydroxy acid obtained by hydrolysis of dolastatin 11
(1) with mild alkali. Acyclic nonadepsipeptide 20 is the
last intermediate in the synthesis of dolastatin 11 (1).
des-Hmp-dolastatin 11 (21), with a 27-membered ring,
7-Nordolastatin 11 (7) was synthesized to avoid the
problem of getting a 7-epimer. It had strong cytotoxic
activity, but was unfortunately accompanied by about
30% of 12-epi-7-nordolastatin (7a), whose identity was
established by an independent synthesis starting from
Boc-D-Ala for the Ibu unit. The 12-epimer appears in
this synthesis and not in the synthesis of dolastatin 11

4148
M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
Table 7. Biological testing results on 1–21: actin binding, murine leukemia, and six human tumors
Compound
Activity summaries EC50 lM ED50 (lg/mL)
bind actin
GI50, human cancer cell lines (lg/mL)
NCI-H460 OVCAR-3 BXPC-3 SF-295 KM20L2 DU-145
Actin
Cells
P388
leukemia
lung-nsc
ovarian
pancreas-a CNS
colon
prostate
Dolastatin 11 (1)
V strong V strong
9.5
0.0027
0.017
<0.001
0.0013a
0.0075
0.023a
0.034
0.0047
0.13a
0.31a
0.22a
Map analogues
3-Nor (2)
3,4-Bis-nor (3)
Strong
Strong
Strong
Medium
22
32
0.014
0.097
0.076a
8.7
0.092
0.66
0.063
0.34
>1
2.8a
12.5
3.8
0.62a
R,R-Acc (3-epimer) (4) Strong
S,S-Acc (4-epimer) (5) None
Weak
Weak
23
>50
26.0
>10
8.7
7.4
7.0
Ala analogues
7-Epi (6)
7-Nor (7)
None
Strong
Weak
Strong
>50
21
0.24
0.018
0.43
<0.001
3.4
0.70
2.9
0.36
3.5
0.38
1.3
0.063
3.4
0.64
Ibu analogues
10-Nor (8)
None
Strong
Medium >50
0.039
0.042
0.011
0.050
0.28
0.058
0.021
0.0003
0.032
0.44
0.90
0.59
0.18
0.63
4.1
0.46
0.37
0.075
0.41
3.4
0.51
0.42
0.16
0.45
3.6
0.38
0.23
0.039
0.27
2.9
0.63
0.48
0.25
0.37
2.4
10,10-Bis-nor (9)
11R-OH (10)
11R-OAc (11)
12-Epi (12)
Medium
Medium Strong
23
42
Medium Medium
None
41
>50
Weak
Leu analogue
Majusculamide C (13) Strong
V strong 19
0.0033
<0.001
0.0048
3.5
<0.001
0.033 <0.001
Hmp analogue
57-Nor (14)
Strong
Strong
Strong
Strong
Weak
Weak
22
20
21
0.030
0.014
10.4
5.7
0.19
>10
Map, Ala analogue
3,4,7-Tris-nor (15)
24.9
Map, Ibu analogue
3,4-Bis-nor-11-
OH (16)
3.8
9.2
>10
Ala, Ibu analogue
7-Nor-11-OH (17)
Strong
Strong
Medium
Weak
20
20
0.16
27.8
0.59
0.43
4.7
2.9
Map, Ala, Ibu analogue
3,4,7-Tris-nor-
11-OH (18)
>10
>10
Open analogues
1,2-Seco (19)
13,14-Seco (20)
None
None
None
None
>50
>50
>10
>10
>1
>10
>1
>1
>10
>1
>1
>10
>1
>10
Unit-deletion analogue
des-Hmp (21)
None
None
>50
>10
>10
>10
>10
>10
>10
a Results from the National Cancer Institute.
(1) and most of its analogues because most of the inter-
mediates for the Ibu-Ala unit in the latter syntheses are
crystalline, whereas those for the Ibu-Gly unit in the for-
mer are not crystalline; thus, products of partial epimer-
ization a to the ketone carbonyl group are removed by
recrystallization in the latter cases but not in the former
case.
rapidly than dolastatin 11 (1; hydrolysis of an amide
or ester bond would presumably cause loss of activity
as noted above). Analogue 4 still has all the carbon
atoms of dolastatin 11 (1) and should not be much more
readily metabolized than dolastatin 11 (1), so we postu-
late that it is inactive in cells because it does not get
through cell walls efficiently, as discussed further below.
3,4,7-Tris-nordolastatin 11 (15) combines the above sim-
plifications in the Map and Ala units. While it bound
well to actin, it had only weak activity against cancer
cells, either because it does not get through the cell wall
efficiently or because it is metabolized much more read-
ily than dolastatin 11 (1). Three other analogues (4, 16,
and 18) have strong actin-binding activity but only weak
activity in cells. 15, 16, and 18 lack two or three methyl
groups, and might well be metabolized much more
10-Nordolastatin 11 (8) and 10,10-bis-nordolastatin 11
(9) were made by omitting one or two methylation
steps, respectively. They had only medium activity,
suggesting that it is worthwhile to put in both methyl
groups.
Derivatives 4 and 5, in which (R,R)- and (S,S)-2-amino-
cyclopentanecarboxylic acids (Acc), respectively, replace
Map, were synthesized as analogues with restricted

M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
4149
Table 8. In vitro testing results for dolastatin 11 (1) and 3,4-bis-nordolastatin 11 (3), molar units
Panel
Cell line
Dolastatin 11 (1)
3,4-Bis-nordolastatin 11 (3)
TGI LC50
1.2Eꢀ05 >5.0Eꢀ05
GI50
TGI
LC50
GI50
Leukemia
CCRF-CEM
HL-60(TB)
K-562
4.0Eꢀ08
1.4Eꢀ08
8.9Eꢀ07
1.5Eꢀ06
2.7Eꢀ07
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
6.3Eꢀ06
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
3.2Eꢀ08
4.7Eꢀ08
1.5Eꢀ05
4.3Eꢀ08
4.6Eꢀ07
6.4Eꢀ07
3.4Eꢀ06
1.3Eꢀ06
5.1Eꢀ07
5.1Eꢀ06
2.8Eꢀ07
7.3Eꢀ06
3.1Eꢀ08
2.6Eꢀ08
1.2Eꢀ08
8.2Eꢀ08
2.9Eꢀ08
1.8Eꢀ06
1.9Eꢀ06
>5.0Eꢀ05
>5.0Eꢀ05
7.2Eꢀ06
1.3Eꢀ05
>5.0Eꢀ05
3.2Eꢀ05
5.4Eꢀ06
3.0Eꢀ06
2.9Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
2.9Eꢀ05
MOLT-4
RPMI-8226
SR
Lung
A549/ATCC
EKVX
HOP-62
1.9Eꢀ08
2.1Eꢀ07
>1.0Eꢀ05
HOP-92
2.1Eꢀ07
3.1Eꢀ07
3.7Eꢀ07
4.1Eꢀ07
1.3Eꢀ09
8.2Eꢀ07
7.1Eꢀ06
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
2.4Eꢀ07
NCI-H226
NIC-H23
NCI-H322M
NCI-H460
NCI-H522
COLO 205
HCC-2998
HCT-116
HCT-15
>1.0Eꢀ05
4.7Eꢀ06
>1.0Eꢀ05
Colon
4.9Eꢀ09
2.3Eꢀ08
1.7Eꢀ07
6.7Eꢀ07
2.2Eꢀ08
3.9Eꢀ07
2.7Eꢀ06
1.0Eꢀ07
1.3Eꢀ07
1.3Eꢀ08
2.8Eꢀ07
1.8Eꢀ07
2.7Eꢀ08
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
6.7Eꢀ07
5.5Eꢀ08
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
1.4Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
5.9Eꢀ06
2.8Eꢀ05
>5.0Eꢀ05
1.4Eꢀ05
3.1Eꢀ06
9.7Eꢀ07
1.3Eꢀ05
1.8Eꢀ05
2.8Eꢀ05
3.1Eꢀ06
2.6Eꢀ05
2.3Eꢀ07
1.4Eꢀ05
>5.0Eꢀ05
3.4Eꢀ06
1.6Eꢀ05
7.9Eꢀ07
>5.0Eꢀ05
1.9Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
1.5Eꢀ06
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
3.8Eꢀ05
HT29
KM12
SW-620
SF-268
4.9Eꢀ08
2.1Eꢀ07
2.7Eꢀ07
2.3Eꢀ07
2.8Eꢀ07
3.6Eꢀ07
4.8Eꢀ08
2.0Eꢀ07
3.5Eꢀ07
1.2Eꢀ06
1.8Eꢀ07
7.0Eꢀ08
6.9Eꢀ08
1.6Eꢀ07
2.9Eꢀ07
5.4Eꢀ07
6.4Eꢀ07
8.1Eꢀ08
2.6Eꢀ06
3.8Eꢀ06
CNS
SF-295
SF-539
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
4.6Eꢀ05
4.1Eꢀ08
SNB-19
SNB-75
U251
7.7Eꢀ07
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
2.1Eꢀ06
>1.0Eꢀ05
8.0Eꢀ06
Melanoma
LOX IMVI
MALMEꢀ3M
M14
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
7.3Eꢀ06
1.2Eꢀ07
>1.0Eꢀ05
>1.0Eꢀ05
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
IGROVI
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
SK-OV-3
786-0
8.3Eꢀ08
7.3Eꢀ08
2.9Eꢀ08
7.9Eꢀ08
1.4Eꢀ07
4.3Eꢀ08
2.3Eꢀ08
8.7Eꢀ08
1.2Eꢀ06
2.2Eꢀ07
2.5Eꢀ07
5.6Eꢀ07
>1.0Eꢀ05
2.5Eꢀ07
6.9Eꢀ06
>1.0Eꢀ05
9.7Eꢀ07
1.1Eꢀ07
8.1Eꢀ07
>1.0Eꢀ05
>1.0Eꢀ05
3.0Eꢀ06
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
3.4Eꢀ06
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
Ovarian
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
Renal
1.3Eꢀ06
2.2Eꢀ06
2.6Eꢀ06
1.6Eꢀ06
6.7Eꢀ07
2.4Eꢀ07
1.1Eꢀ06
7.8Eꢀ06
2.0Eꢀ07
2.1Eꢀ06
2.5Eꢀ07
2.1Eꢀ05
1.7Eꢀ07
6.0Eꢀ07
5.0Eꢀ07
3.7Eꢀ07
2.6Eꢀ06
9.4Eꢀ06
8.0Eꢀ06
2.9Eꢀ05
6.6Eꢀ06
1.9Eꢀ06
>5.0Eꢀ05
5.3Eꢀ06
1.7Eꢀ05
2.0Eꢀ05
3.1Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
7.4Eꢀ06
1.6Eꢀ05
3.6Eꢀ05
>5.0Eꢀ05
1.1Eꢀ05
>5.0Eꢀ05
2.6Eꢀ05
A498
ACHN
CAKI-1
4.4Eꢀ07
3.5Eꢀ08
4.3Eꢀ08
2.5Eꢀ08
3.0Eꢀ07
1.1Eꢀ07
1.7Eꢀ07
1.4Eꢀ07
2.2Eꢀ07
8.8Eꢀ08
2.8Eꢀ06
2.9Eꢀ08
4.0Eꢀ07
9.1Eꢀ08
1.1Eꢀ07
2.1Eꢀ07
4.6Eꢀ07
>1.0Eꢀ05
3.5Eꢀ07
1.7Eꢀ07
6.1Eꢀ08
>1.0Eꢀ05
3.9Eꢀ07
3.8Eꢀ07
4.7Eꢀ06
>1.0Eꢀ05
6.8Eꢀ07
>1.0Eꢀ05
1.1Eꢀ06
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
4.5Eꢀ07
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
6.1Eꢀ07
>5.0Eꢀ05
2.0Eꢀ05
RXF 393
SN12C
TK-10
7.8Eꢀ07
5.8Eꢀ06
>1.0Eꢀ05
2.6Eꢀ06
>5.0Eꢀ05
3.5Eꢀ05
UO-31
PC-3
8.7Eꢀ07
3.5Eꢀ05
Prostate
Breast
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
>1.0Eꢀ05
1.0Eꢀ06
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
>5.0Eꢀ05
4.6Eꢀ05
DU-145
MCF7
NCI/ADR-RES
MDA-MB-231/ATCC
HS 578T
MDA-MB-435
MDA-N
BT-549
T-47D
>1.0Eꢀ05
rotation about the 2–3–4–5 bond in the Map unit. R,R-
Acc is restricted to values of ꢀ60ꢁ to ꢀ180ꢁ for this
bond, whereas S,S-Acc has values of +60ꢁ to +180ꢁ.
The S,S-Acc derivative 5 should be a good model for

4150
M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
the C1 and C15 conformations (with values of +76ꢁ and
+69ꢁ, respectively), especially the former, which is calcu-
lated to be the lowest energy conformation of 5 as well
as 1. The R,R-Acc derivative 4 should serve as a model
for the other low-energy conformations: C2–C4, C7,
W1, and W5, which have values from ꢀ64ꢁ to ꢀ86ꢁ.
When all efforts to resolve a mixture of R,R- and S,S-
Acc9 failed, the racemic mixture was carried through
the synthesis with the hope that the final cyclic depsipep-
tides 4 and 5 could be separated and identified by NMR;
fortunately, this was the case. Surprisingly, neither 4 nor
5 was very active in cells. However, the R,R derivative 4
bound very strongly to actin, whereas the S,S stereoiso-
mer 5 was inactive. The most likely interpretation of
these results is that dolastatin 11 (1) goes into cells in
an S,S-type conformation but binds in an R,R-type con-
formation. If this is correct, an analogue restricted to
one of these conformations will not be a suitable drug,
and only analogues flexible enough to take both confor-
mations have drug potential.
Figure 2. Superposition of the low-energy conformations calculated
for 1 and 5 in chloroform.
mation to dolastatin 11 (1). Molecular mechanics calcu-
lations on many of these derivatives supported this view.
In contrast, two derivatives which clearly appear to be
like 4, with predominantly negative angle 2–3–4–5 val-
ues, are 6 and 11, in which the two main strands of
the depsipeptide are pushed apart by a methyl group
and an acetate group, respectively, destabilizing confor-
mation C1. 3,4-Bis-nordolastatin 11 (3) and 3,4,7-tris-
nordolastatin 11 (15) also favor forms with negative
2–3–4–5 angles from their Val and Ibu shifts, but some
of the shifts in their other amino acids suggest that 3 and
15 do not share the same make-up of such forms as do 4,
6, and 11. A ROESY spectrum of 7-epidolastatin 11
(6, Table 1) showed that it has the same amide configu-
rations as dolastatin 11 (1).
In which conformation does dolastatin 11 (1) go
through the cell membrane? Probably C1 = W2, which
is by far the lowest energy conformation with a positive
value for angle 2–3–4–5 and was calculated to be the
lowest-energy conformation for most close derivatives
as well as for dolastatin 11 (1) in chloroform.
Which is the binding conformation? There are several
strong candidates with negative values for angle 2–3–
4–5: W1 (with the lowest calculated energy in water),
C2 = W18, C3 = W3, and C4 = W4. This will be dis-
cussed further in the next section.
The predominant conformation of 4, 6, and 11 in chlo-
roform is very likely C2 (the lowest-energy calculated
conformation for dolastatin 11 (1) having a negative
2–3–4–5 angle) for several reasons. First, of the low-en-
ergy conformations, only this one has significant
changes from C1 in the angles of the 30-membered ring
in the Map, Ibu, Val, and Gly2 units (Table 2) and in the
Tyr side chain (Table 3) to account for the large differ-
ences between the NMR spectra of 1 and the 4, 6, 11
group. Second, the NH proton shift changes observed
in going from dolastatin 11 (1) to 7-epidolastatin 11
(6) favor a C2-like conformation for 6: the Map-NH
absorption moves downfield by 1.1 ppm (from strong
intramolecular H-bonding in 1 to none in 6), whereas
the other four NH absorptions move upfield due to
stronger intramolecular H-bonds in 6 (Table 5). Third,
conformation C2 was the lowest-energy conformation
calculated for 4, 6, and 11 in chloroform (Table 9).
Alcohols 10 and 16–18 were prepared to learn their activ-
ities and to provide a handle for the preparation of pro-
drugs. All four bind strongly to actin. The three lacking
alkyl groups have greatly reduced activity in cells, but
10, from reduction of dolastatin 11 (1), retains strong activ-
ity in cells (it is better than dolastatin 11 (1) against NCI-
H460 lung cancer, with GI50 0.3 nM). The acetate 11 made
from alcohol 10 is a possible prodrug for alcohol 10.
6. Conformations of the analogues
The NMR data in Table 6 and molecular mechanics cal-
culations on some of the analogues provide information
about their preferred conformations. The most revealing
analogues are 4 and 5, which as noted above are limited
to ꢀ and + values of the 2–3–4–5 angle, respectively.
The only low-energy conformation calculated for 5 is
like C1 (Fig. 2 shows a superposition of the low-energy
conformations of 1 and 5), and since the NMR shifts for
dolastatin 11 (1) are similar to those of 5 (Table 6), it is
likely that they share this as their predominant confor-
mation. Many of the shifts for dolastatin 11 (1) are dis-
placed slightly toward those of 4, consistent with the
presence of small amounts of conformations which, like
4, have negative values for the 2–3–4–5 angle.
Most of the chemical shifts of 7-nordolastatin (7) are
about halfway between those of dolastatin 11 (1) and
7-epidolastatin 11 (6).
What is the most likely binding conformation of dolast-
atin to F-actin? Since R,R-Acc derivative 4 bound
strongly to actin, the binding conformation must have
a negative 2-3-4-5 angle, which rules out C1. The low-
energy candidates with angle 2–3–4–5 negative are
C2–C4 and W1. Though C2 is apparently the conforma-
tion that is seen in chloroform solution by NMR for
Most of the derivatives (2, 5, 8–10, 12–14, and 16–18)
had shifts, which suggest that they are similar in confor-

M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
4151
Table 9. Energies (kJ/mol) calculated for conformations of dolastatin 11 (1) and three derivatives which favor a conformation with a negative 2–3–4–
5 angle in chloroform solution
Conformation!
C1 = W2
C2 = W18
C3 = W3
C4 = W4
W1
2–3–4–5 Sign!
+
ꢀ
2.2
0
ꢀ
ꢀ
ꢀ
D11 (1)
0
5.4
8.8
1.1
4.5
7.7
2.7
14.6
34.4
25.9
8.8
12.0
4.6
R,R-Acc (4)
7-Epi (6)
11R-OAc (11)
Impossible
7.2
27.3
0
0
Table 10. Energies (kJ/mol) calculated for conformations of dolastatin 11 (1) and selected derivatives in water
Conformation!
C1 = W2
C2 = W18
C3 = W3
C4 = W4
W1
2–3–4–5 Sign!
+
ꢀ
ꢀ
ꢀ
ꢀ
D11 (1)
3-Nor (2)
1.6
0
9.9
2.7
4.8
11.9
4.9
0
5.1
23.0
6.7
0
0
29.5
20.7
16.2
8.3
1.1
8.9
3.6
1.2
16.9
19.0
10.3
3,4-Bis-nor (3)
R,R-Acc (4)
7-Epi (6)
0
Impossible
2.5
0
5.2
6.2
19.1
9.6
11R-OH (8)
11R-OAc (11)
12-Epi (12)
38.5
12.3
0
26.9
0
14.0
19.4
14.6
derivatives like 4, 6, and 11, it is calculated to be much
higher in energy than the others in the intracellular fluid,
water (Table 10), and thus is unlikely to be the binding
conformation. We favor W1 since it is the lowest energy
conformation in water for the most active compound (1)
and is of low energy in the other highly active
compounds, but C3 and C4 are also of fairly low
energies in most strongly active derivatives and are
reasonable possibilities as well.
except for the one between the Tyr and Val units, which
is cis. Cyclic 3S,4S analogue 5 with restricted rotation
about the 3,4 bond, which modeled the lowest-energy
conformation of dolastatin 11 in chloroform (C1), was
inactive in cells and with actin. Its 3R,4R stereoisomer
4, which modeled most of the other low-energy confor-
mations, strongly stimulated actin polymerization but
was inactive in cells. Thus, the lowest-energy conforma-
tion in chloroform (C1) is not the binding conformation,
but may well be the conformation in which these
molecules go through cancer cell walls. The binding
conformation is most likely W1. The binding side of
the molecule is probably the side from which the mole-
cule is usually viewed. Thus, an active derivative should
have a low-energy C1 conformation to go through the
cell wall, a low-energy W1 conformation to bind, and
the right shape to bind strongly.
From most evidence, the binding side of the molecule
appears to be approximately the side from which the
molecule is usually viewed (1 and Fig. 1), since omitting
the ꢁbacksideꢀ 3- and 7-methyl and 4-ethyl groups and
making a five-membered ring in the Map unit (4) do
not decrease the binding much (Table 7). However,
10,10-bis-nordolastatin 11 (9), which lacks two methyl
groups on the front side in W1, also binds strongly.
One other piece of evidence: 7-epidolastatin 11 (6) has
low-energy C1 and W1 conformations, yet is not very
active; this may indicate that the 7-methyl group of
dolastatin 11 (1) is involved with the binding site.
8. Experimental section
NMR spectra were obtained on Bruker spectrometers at
250, 500, or 600 MHz in CDCl3 unless otherwise stated.
Mass spectra were obtained on JOEL HX100A (FAB)
and Finnigan LCQ spectrometers (ESI). HPLC was per-
formed on a Rainin instrument eluting with water–ace-
tonitile with UV detection. Unless otherwise stated, all
reagents and solvents were from Sigma–Aldrich except
1-cyclopentenecarboxylic acid, which was from Alfa–
Aesar Co. Most reactions were carried out under argon.
Solvent extracts were dried over anhydrous magnesium
sulfate before solvent evaporation. HPLC purifications
were done on a 21.4 mm ID · 25 cm reverse-phase C-
18 column with elution with water-acetonitrile mixtures
from 0% to 100% acetonitrile over 40 min at 8 mL/min,
except that a 30 min protocol at 3 mL/min was used for
difficult separations; the UV detector was set at 280 nm
unless otherwise specified.
7. Summary
Twenty analogues of the natural antitumor agent dolast-
atin 11 (1), including the natural product majusculamide
C (13), were synthesized and tested for cytotoxicity
against human cancer cells and stimulation of actin poly-
merization. Analogues lacking the 30-membered ring
(19–21) were inactive. The simplified analogue 3 in which
b-alanine replaces Map retained good activity, as did the
11R alcohol 10 from borohydride reduction of dolastatin
11 (1). The epimers of dolastatin 11 in the Ala and Ibu
units (6 and 12) had greatly reduced activity. Molecular
modeling and NMR evidence were used to study the
low-energy conformations. The amide bonds are all trans

4152
M. A. Ali et al. / Bioorg. Med. Chem. 13 (2005) 4138–4152
8.1. Analogue synthesis
256MB RAM. Data workup used SGI Indigo2 GUI-Ex-
treme Graphics, 150 MHz MIPS R4400/R4010 FPU
and 64MB RAM. Graphics used 50 MHz MIPS R400/
R4010 FPU 48MB RAM. An average of 100 struc-
tures/day were generated and minimized. Amide bonds
were locked in the trans configuration except for the
Tyr-Val bond, which was locked cis.
Analogues 2–22 were synthesized using the synthetic
route developed for dolastatin 11 (1),2 except for the
three significant improvements in that synthesis which
are included below. The other new reactions are in-
cluded in the supplementary material.
8.1.1. (2S,3R)-2-Methyl-3-aminopentanoic acid (Map).
Ammonia (850 mL) was distilled into a solution of N-
tosyl-Map (17 g, 60 mmol) in THF (50 mL) at ꢀ70 ꢁC.
Sodium (8.2 g, 356 mmol) was added in two portions
over 30 min. After 30 min more, the blue soln was
warmed to ꢀ30 ꢁC for 30 min. The blue color was elim-
inated by adding solid NH4Cl, sodium was destroyed by
adding water (cautiously!), and the ammonia was
allowed to evaporate. The residue was rinsed into a crys-
tallizing dish with NaOH soln (3 M, 25 mL) and water
(100 mL), and water and ammonia were evaporated in
a hood. The pH was adjusted to 1 with concd HCl
and the solution was washed with CH2Cl2 (4 · 50 mL).
The pH was adjusted to 6.5 with NaOH pellets and
the solution was evaporated in a hood. The residual
crystals were triturated with EtOH (5 · 60 mL). The
soln was filtered, the EtOH was evaporated in a hood,
and the residue was dried in a vacuum dessicator to give
Map (7.8 g, 100%). As Map is not easily purified by
recrystallization, it was converted to Boc-Map, which
was easily recrystallized.2
Acknowledgements
For financial support, we thank the NIH (grant R01
CA78750), the Camille and Henry Dreyfus Foundation
(Senior Scientist Mentor Initiative), the Research Cor-
poration, the Bristol-Myers Squibb Company, and the
Aldrich Chemical Company. We are also very grateful
to the NCI Developmental Therapeutics Program for
testing two compounds against 60 human tumors and
to Prof. Richard E. Moore for a generous gift of majus-
culamide C.
Supplementary data
Yield, NMR, and MS data for the intermediates and
analogues. Supplementary data associated with this arti-
cle can be found, in the online version, at doi:10.1016/
j.bmc.2005.04.040.
References and notes
8.1.2. N-Me-L-Val-O,N-diMe-L-Tyr-OBnꢁHCl. Thionyl
chloride (50 mL) was added dropwise to a solution of
Boc-N-Me-L-Val-O,N-diMe-L-Tyr-OH (10 g, 23.7
mmol) in benzyl alcohol (200 mL) at 0 ꢁC. After heating
at 55 ꢁC for 1 h and evaporation in a hood to 1/4 of the
volume, ether (150–200 mL) was added until cloudiness
occurred. After 16 h at ꢀ20 ꢁC, the crystals were filtered
and dried in a vacuum dessicator over P2O5, giving N-
Me-L-Val-O,N-diMe-L-Tyr-OBnÆHCl (8.3 g, 78%).
1. Pettit, G. R.; Kamano, Y.; Kizu, H.; Dufresne, C.; Herald,
C. L.; Bontems, R. J.; Schmidt, J. M.; Boettner, F. E.;
Nieman, R. A. Heterocycles 1989, 28, 553.
2. Bates, R. B.; Brusoe, K. G.; Burns, J. J.; Caldera, S.;
Cui, W.; Gangwar, S.; Gramme, M. R.; McClure, K.
J.; Rouen, G. P.; Schadow, H.; Stessman, C. C.;
Taylor, S. R.; Vu, V. H.; Yarick, G. V.; Zhang, J.;
Pettit, G. R.; Bontems, R. J. Am. Chem. Soc. 1997,
119, 2111.
3. Bai, R.; Verdier-Pinard, P.; Gangwar, S.; Stessman, C. C.;
McClure, K. J.; Sausville, E. A.; Pettit, G. R.; Bates, R. B.;
Hamel, E. Mol. Pharmacol. 2001, 59, 462.
4. Oda, T.; Crane, Z. D.; Dicus, C. W.; Sufi, B. A.; Bates, R.
B. J. Mol. Biol. 2003, 328, 319.
5. Schmidt, U.; Griesser, H.; Haas, G.; Kroner, M.; Riedl,
B.; Schumacher, A.; Sutoris, F.; Haupt, A.; Emling, F. J.
Pept. Res. 1999, 54, 146.
6. Winterfeldt, E.; Nelke, J. M.; Korth, T. Chem. Ber. 1971,
104, 802.
8.1.3. Dolastatin 11 (1), procedure B. To Ibu-L-Ala-
Map-Hmp-Gly-NMe-L-Leu-Gly-NMe-L-Val-O,N-di-
Me-L-Tyr-OHÆTFA (16.7 mg, 0.015 mmol) in DMF
(1.5 mL) was added a solution of HBTU (34 mg,
0.090 mmol) and Et3N (17 lL, 0.120 mmol) in DMF
(1.5 mL) dropwise with stirring over 30 min. After 6 h,
the solvent was evaporated in an open dish and the res-
idue was triturated with CH3CN for HPLC, which gave
dolastatin 11 (1, 6.5 mg, 44%), after separation from 7-
epi-dolastatin (6, 4.0 mg, 27%).
7. Carter, D. C.; Moore, R. E.; Mynderse, J. S.; Niemczura,
W. P.; Todd, J. S. J. Org. Chem. 1984, 49, 236.
8. Mynderse, J. S.; Hunt, A. H.; Moore, R. E. J. Nat. Prod.
1988, 51, 1299.
8.2. Molecular modeling
9. Yamazaki, T.; Zhu, Y.-F.; Probstl, A.; Chadha, R. K.;
Goodman, M. J. Org. Chem. 1991, 56, 6644.
10. MacroModel, Interactive Molecular Modeling System.
Version 6.0, Department of Chemistry, Columbia Univer-
sity, New York, NY, 1977.
Molecular modeling studies involving Monte Carlo cal-
culations and energy minimizations were performed
using MacroModel10 (v. 6.0 and 7.0) on an IBM RS/
6000 590 with 66 MHz POWER2 architecture and