Total synthesis of the prenylated cyclopeptide trunkamide A, a cytotoxic metabolite from Lissoclinum sp.

Article abstract of DOI:10.1016/S0040-4039(01)00194-0

A total synthesis of the doubly prenylated cyclic peptide trunkamide A of marine origin, and also its C45 epimer, is described.

Full text of DOI:10.1016/S0040-4039(01)00194-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 2573–2577
Total synthesis of the prenylated cyclopeptide trunkamide A, a
cytotoxic metabolite from Lissoclinum sp.
Benedict McKeever and Gerald Pattenden*
School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK
Received 8 January 2001; accepted 1 February 2001
Abstract—A total synthesis of the doubly prenylated cyclic peptide trunkamide A of marine origin, and also its C45 epimer, is
described. © 2001 Elsevier Science Ltd. All rights reserved.
The marine environment has delivered a prodigious
variety of structurally novel and biologically important
secondary metabolites in recent years, many of which
have lured chemists into their total synthesis and an
evaluation of their potential as chemotherapeutic
agents. Amongst some of the most intriguing marine
natural products are heterocyclic based cyclic peptides
produced by ascidians of the genus Lissoclinum.1 These
unusual compounds show structures which are based
on the inclusion of thiazoline and oxazoline rings in
their peptide backbone and, sometimes, modified by the
presence of reverse prenyl units in their side chains;
examples include mollamide 1,2 patellin 6 2,3 and trunk-
amide A 3.3 Most of the cyclic peptides isolated from
ascidians show moderate cytotoxicity, but trunkamide
A is reported to have particularly pronounced and
promising antitumour activity.4 Synthetic work within
the various families of natural cyclic peptides has been
intense,5 and recently we described the first synthesis of
a reverse prenylated member, i.e. mollamide 1.5d In
continuation of our studies of the synthesis,5 self assem-
bly,6 and ionophore properties7 of novel marine cyclic
peptides and their analogues, we now report a synthesis
of trunkamide A 3a, and of its C45 epimer 3b.8
Following our successful synthesis of mollamide 1, we
designed a synthesis of the structure 3b reported for
naturally derived trunkamide A 3 using a similar strat-
egy, whereby the thiazoline ring in the natural product
was to be produced in the final step from the thioamide
cyclic peptide 4. The cyclic peptide 4 was to be derived
from the heptapeptide 5 which, in turn, would be
elaborated from the thioamide 6 and the prenylated
amino acids 7 and 8 (Scheme 1).9
Thus the pentapeptide thioamide 6 was first prepared
starting from serine methyl ester and proceeding via the
thioamide 9 and then adding proline, alanine and
isoleucine sequentially (Scheme 2). After removal of the
Boc-protecting group in 6, a coupling reaction with 7
Ph
O
O
O
S
O
S
S
N
H
O
N
H
45
O
N
H
N
N
O
NH
N
O
N
O
N
O
N
HN
Ph
O
O
HN
HN
N
Ph
NH
O
O
HN
NH
O
H
N
HN
H
N
O
O
NH
O
O
N
O
O
O
O
O
1
2
3a; C45 R
3b; C45 S
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00194-0

2574
B. McKee6er, G. Pattenden / Tetrahedron Letters 42 (2001) 2573–2577
Ph
Ph
O
O
S
S
OH
O
OH
N
H
N
H
HN
HN
N
N
CO2Me
O
O
3b
O
HN
TcBocHN
HN
O
NH
NH
HN
H
N
H
N
O
O
O
O
O
O
O
O
O
4
5
Ph
S
OH
N
H
HN
N
CO2Me
TcBocHN
HO2C
O
O
NH
8
NHTcBoc
NHBoc
O
HO2C
O
7
6
Scheme 1.
Ph
Ph
S
O
v, vi
OH
CO2Me
i, ii, iii, iv
S
OH
OH
CO2Me
N
H
BocHN
Boc-L-Pro
90%
HN
HN
H2N
NBoc
O
CO2Me
9
Ph
S
Ph
OH
O
N
H
S
HN
N
OH
CO2Me
v, vi
N
H
v, vii
v, vi
O
6
HN
N
CO2Me
7
Boc-L-Ala
60%
Boc-L-Ile
80%
O
NH
H
N
NHTcBoc
O
NHBoc
O
O
10
Ph
S
O
OH
N
H
HN
N
CO2
H3N
O
viii, x
viii, ix
xi x
5
4
8
NH
O
H
N
HN
O
O
O
O
11
L-Phe, DCC, HOBt, DIPEA, DCM, 74%; (ii) TBSCl, imidazole, DMF, 90%; (iii) Lawesson’s reagent,
Scheme 2. Reagents: (i) Boc-
benzene, 80°C, 92%; (iv) TBAF, THF, 93%; (v) AcCl, MeOH; (vi) DCC, HOBt, DIPEA, DCM; (vii) EDC, HOBt, DIPEA, DCM,
72%; (viii) Cd/Pb, 1 M NH4OAc/THF (1:1); (ix) EDC, HOBt, DCM, 78% (two steps); (x) 1 M NaOH, MeOH; (xi) DPPA,
DIPEA, DMF.
next produced 10, which was then deprotected and the
resulting free amine coupled with 8 leading to the
advanced doubly prenylated precursor 5.
After removal of the TcBoc and methyl ester protecting
groups in 5, we were disappointed to find that attempts
to induce macrolactamisation of the zwitterion 11 lead-

B. McKee6er, G. Pattenden / Tetrahedron Letters 42 (2001) 2573–2577
2575
ing to 4 failed and, instead, only intractable material
was recovered. We were persuaded that the thioamide
bond in 11 and its proximity to the site of macrolac-
tamisation was the likely cause of the problem. We
therefore decided to synthesise the thiazoline ring in the
target natural product prior to macrolactamisation,
hence removing any possible interference. This was a
somewhat risky decision since it is well documented
that chiral thiazolines are prone to undergo isomerisa-
tion under a range of acid/base conditions.10 In the
event we would expect to lose the stereochemical
integrity at C45 (and possibly C40) during any subse-
quent synthetic manipulations leading to the trunk-
amide target.
tide structure. However, as anticipated, analysis of the
material obtained clearly showed the presence of a
mixture of diastereoisomeric products. One of these
separated diastereoisomers, however, showed NMR
spectroscopic data which were identical with those
reported for natural trunkamide A, whereas another
showed NMR data corresponding to those reported for
the C45 epimer 3b of trunkamide A.3,8
Encouraged by this outcome, we speculated that if we
replaced the problematic thioamide bond in 11 with an
amide bond then our original macrocyclisation strategy,
cf. 5“4 (Scheme 1), may prove more successful and
allow us access to diastereomerically pure material.
Hence, we decided to synthesise the heptapeptide 14
and examine its macrocyclisation (Scheme 4). Thus,
starting from serine methyl ester, the heptapeptide 14
was prepared by the sequential addition of phenylala-
nine, proline, alanine and isoleucine, to give 13, fol-
lowed by the addition of the two reverse prenylated
amino acids 7 and 8.
Hence, treatment of the heptapeptide thioamide 5 with
Burgess’ reagent resulted in cyclodehydration leading to
the corresponding substituted thiazoline 12 (Scheme 3).
Removal of the protecting groups from 12 followed by
macrolactamisation of the resulting v-amino acid, and
chromatography, successfully produced the cyclopep-
Ph
Ph
O
S
N
O
S
N
H
45
H
N
N
N
N
CO2Me
40
O
O
O
i
ii-iv
O
5
HN
TcBocHN
NH
O
NH
H
N
HN
H
N
O
HN
O
O
O
O
O
O
O
12
Scheme 3. Reagents: (i) Burgess’ reagent, THF, 65°C, 92%; (ii) Cd/Pb, 1 M NH4OAc/THF (1:1), 93%; (iii) 1 M NaOH, MeOH;
(iv) DPPA, DIPEA, DMF.
Ph
Ph
O
O
O
O
OH
N
H
OH
N
H
HN
N
HN
CO2Me
N
OH
O
steps
steps
CO2Me
O
TcBocHN
H2N
CO2Me
O
NH
NH
H
N
HN
O
O
NHBoc
O
O
O
13
14
Ph
S
Ph
O
O
O
OH
O
N
H
N
H
N
HN
N
N
O
O
O
iv, v
iv
72%
i-iii
O
O
HN
4
HN
HN
59%
(2 steps)
NH
NH
HN
H
N
H
N
O
O
O
O
O
O
O
O
3b
15
Scheme 4. Reagents: (i) Cd/Pb, 1 M NH4OAc/THF (1:1), 98%; (ii) TBAH, THF, 0°C; (iii) DPPA, DIPEA, DMF, 35% (two
steps); (iv) DAST, DCM; (v) H2S, Et3N, MeOH.

2576
B. McKee6er, G. Pattenden / Tetrahedron Letters 42 (2001) 2573–2577
Ph
N
Ph
N
O
O
S
S
N
H
N
H
N
N
O
O
Pyridine
O
O
O
O
HN
HN
MeOH
50-55oC
NH
NH
HN
HN
H
N
H
N
O
O
O
O
O
O
O
O
3b
3a
Scheme 5.
After our failure to induce macrocyclisation from the
thioamide heptapeptide 5 we were delighted to find
that, following removal of the TcBoc and methyl ester
protecting groups in 14, macrocyclisation proceeded
smoothly and provided the cyclopeptide 15 in 35% yield
(Scheme 4). The macrocycle 15 was then elaborated to
the trunkamide structure 3b using a strategy described
by Wipf et al.8,11 Hence cyclodehydration of 15 to the
corresponding oxazoline, and subsequent thiolysis with
H2S led to the thioamide cyclic peptide 4 which, on
reaction with DAST gave 3b in 72% yield. The cyclic
peptide 3b was identical in all respects with the trunk-
amide structure reported by Uto and Wipf8 but differed
significantly with spectroscopic data reported by Bow-
den et al. for natural trunkamide A.3,12a Following the
report of Uto and Wipf revealing the correct structure
of trunkamide A, i.e. 3a rather than 3b,5a we then
decided to attempt to synthesise this epimeric structure.
Instead of carrying out a new total synthesis of natural
2. Bowden, B. F.; Carroll, A. R.; Coll, J. C.; Hockless, D.
C. R.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1994,
47, 61.
3. Bourne, D. J.; Bowden, B. F.; Carroll, A. R.; Coll, J. C.;
Ireland, C. M.; MacLeod, J. K.; Zabriskie, T. M. Aust. J.
Chem. 1996, 49, 659.
4. Bowden, B. F.; Garcia, G. D. W O9,739,025 A1, Oct. 23,
1997; Chem. Abstr. 1997, 127, 290919m.
5. For recent total syntheses, see: (a) Uto, Y.; Wipf, P. J.
Org. Chem. 2000, 65, 1037; (b) Boden, C. D. J.; Patten-
den, G. J. Chem. Soc., Perkin Trans. 1 2000, 875; (c)
Boden, C. D. J.; Norley, M. C.; Pattenden, G. J. Chem.
Soc., Perkin Trans. 1 2000, 883; (d) McKeever, B.; Pat-
tenden, G. Tetrahedron Lett. 1999, 40, 9317; (e) Kigoshi,
H.; Yamada, S. Tetrahedron 1999, 55, 12301; (f) Aguilar,
E.; Downing, S. V.; Meyers, A. I. J. Org. Chem. 1999, 64,
826; (g) Freeman, D. J.; Pattenden, G. Tetrahedron Lett.
1998, 39, 3251; (h) Bagley, M. C.; Moody, C. J. J. Chem.
Soc., Perkin Trans. 1 1998, 601.
6. (a) Blake, A. J.; Hannam, J. S.; Jolliffe, K. A.; Pattenden,
G. Synlett 2000, 1515; (b) Bertram, A.; Pattenden, G.
Synlett 2000, 1519; (c) Bertram, A.; Gonza´lez-Lo´pez de
Turiso, F.; Hannam, J. S.; Jolliffe, K. A.; Pattenden, G.
Synlett 1999, 1723.
7. (a) Drake, A. F.; Freeman, D. J.; Pattenden, G.; Sili-
gardi, G. J. Chem. Soc., Perkin Trans. 2 1998, 129; (b)
Michael, J. P.; Pattenden, G. Angew. Chem., Int. Ed.
Engl. 1993, 32, 1.
8. In contemporaneous studies Uto and Wipf described a
total synthesis of the stereostructure 3b assigned to trunk-
amide A by Bowden et al.3 However, the spectroscopic
data recorded for the synthetic and natural materials did
not correlate; Uto, Y.; Wipf, P. Tetrahedron Lett. 1999,
40, 5165. In a subsequent report these authors showed
that 3a was the correct stereostructure for natural trunk-
amide A (Ref. 5a).
trunkamide A from
D
- rather than
L
-phenylalanine,
however, we planned to effect conversion of epi-trunk-
amide A 3b into trunkamide A 3a by selective epimeri-
sation of the incorrect C45 stereocentre. Indeed, we
found that when 3b was treated with methanolic pyri-
dine at 50–55°C it was slowly converted into trunk-
amide A 3a with no detectable formation of any other
diastereoisomer (Scheme 5). The data for our synthetic
trunkamide12b were identical in every respect with those
reported for the natural product3 and those reported by
Uto and Wipf.5a
Acknowledgements
We thank Dr. P. Bradley and Dr. M. Norley for their
interest in this work and Knoll Pharmaceuticals for
their support via a research scholarship (to B.M.). We
also thank Professor B. F. Bowden and Professor P.
Wipf for their helpful correspondence.
9. The amino acids 7 and 8 were synthesised by Lewis
acid-mediated ring openings of chiral aziridines prepared
from homochiral serine methyl ester and threonine
methyl ester, respectively, using a procedure described by
Okawa et al.; Nakajima, K.; Neya, M.; Okawa, K.;
Yamada, S. Bull. Chem. Soc. Jpn. 1982, 55, 3049 (see also
Refs. 5a and d).
References
10. Fritch, P. C.; Wipf, P. Tetrahedron Lett. 1994, 35, 5397
and references cited therein.
11. Fritch, P. C.; Miller, C. P.; Venkatraman, S.; Wipf, P.
Tetrahedron Lett. 1995, 36, 6395.
1. For reviews see: (a) Faulkner, D. J. Nat. Prod. Rep. 2000,
17, 7; (b) Wipf, P. Chem. Rev. 1995, 95, 2115; (c) David-
son, B. S. Chem. Rev. 1993, 93, 1771.

B. McKee6er, G. Pattenden / Tetrahedron Letters 42 (2001) 2573–2577
2577
12. Satisfactory spectroscopic and mass spectrometry data
were obtained for all new compounds.
m/z calcd for C43H63N7NaO8S ([M+Na]+): 860.4357;
found: 860.4321; (b) Data for synthetic 3a: [h]D −21 (c
0.26, CHCl3); lH (360 MHz, CDCl3) 7.97 (1H, d, J=6.4
Hz), 7.56 (1H, d, J=7.8 Hz), 7.29–7.21 (4H, m), 7.17
(1H, d, J=7.7 Hz), 7.15–7.12 (2H, m), 6.32 (1H, d,
J=9.6 Hz), 5.93 (1H, dd, J=17.6 and 10.8 Hz), 5.75 (1H,
dd, J=17.6 and 10.9 Hz), 5.28 (1H, dd, J=17.5 and 0.6
Hz), 5.25 (1H, dd, J=10.8 and 0.7 Hz), 5.17–5.11 (1H,
masked ddd), 5.16 (1H, dd, J=10.9 and 0.9 Hz), 5.12
(1H, dd, J=17.5 and 0.9 Hz), 4.93 (1H, app t, J=9.6
Hz), 4.63–4.56 (3H, m), 4.51 (1H, dq, J=6.5 and 5.7 Hz),
4.39 (1H, app t, J=6.0 Hz), 4.04 (1H, dq, J=6.5 and 5.1
Hz), 3.91 (1H, dd, J=9.1 and 2.1 Hz), 3.73 (1H, dd,
J=11.3 and 9.7 Hz), 3.64 (1H, dd, J=11.2 and 9.7 Hz),
3.55–3.49 (2H, m), 3.46 (1H, dd, J=9.1 and 3.2 Hz), 3.25
(1H, dd, J=13.9 and 5.7 Hz), 2.95 (1H, dd, J=14.0 and
5.9 Hz), 2.40–2.33 (1H, m), 2.27–2.22 (1H, m), 1.96–1.87
(3H, m), 1.49 (3H, s), 1.39 (3H, s), 1.35–1.18 (2H, m),
1.27 (3H, s), 1.25 (3H, s), 1.22 (3H, d, J=6.6 Hz), 1.07
(3H, d, J=6.5 Hz), 0.96 (3H, d, J=6.9 Hz), 0.94 (3H,
app t, J=7.4 Hz); lC (100 MHz, CDCl3) 173.3 (s), 171.1
(s), 170.9 (s), 170.6 (s), 170.2 (s), 170.1 (s), 168.7 (s), 142.7
(d), 142.1 (d), 135.7 (s), 129.7 (d), 128.4 (d), 127.2 (d),
115.8 (t), 115.0 (t), 78.2 (d), 77.9 (s), 76.0 (s), 67.3 (d),
62.3 (t), 60.0 (d), 57.9 (d), 56.5 (d), 55.4 (d), 52.8 (d), 47.8
(d), 47.1 (t), 39.9 (t), 36.5 (d), 36.4 (t), 28.6 (t), 27.4 (q),
25.8 (q), 25.8 (q), 25.7 (q), 25.6 (t), 23.8 (t), 18.6 (q), 18.1
(q), 16.1 (q), 12.0 (q); HRMS (ES) m/z calcd for
C43H63N7NaO8S ([M+Na]+): 860.4357; found: 860.4320.
(a) Data for synthetic 3b: [h]D −10 (c 0.50, CHCl3); lH
(400 MHz, CDCl3) 8.45 (1H, d, J=7.1 Hz), 8.17 (1H, d,
J=7.1 Hz), 7.56 (1H, d, J=7.0 Hz), 7.30 (1H, d, J=5.5
Hz), 7.18–7.14 (3H, m), 7.12–7.07 (2H, m), 6.26 (1H, d,
J=10.0 Hz), 5.96 (1H, dd, J=17.6 and 10.8 Hz), 5.76
(1H, dd, J=17.6 and 10.9 Hz), 5.32 (1H, dd, J=17.5 and
0.4 Hz), 5.30 (1H, dd, J=10.8 and 0.7 Hz), 5.17 (1H, dd,
J=10.8 and 0.9 Hz), 5.14 (1H, dd, J=17.5 and 0.9 Hz),
5.07–4.97 (2H, m), 4.85 (1H, app d, J=7.5 Hz), 4.68–4.58
(1H, masked dq), 4.67 (1H, dd, J=10.0 and 3.2 Hz), 4.61
(1H, dd, J=6.8 and 5.5 Hz), 4.47 (1H, app dt, J=6.5 and
2.9 Hz), 4.00 (1H, dq, J=6.5 and 5.2 Hz), 3.91 (1H, dd,
J=9.3 and 2.3 Hz), 3.67 (1H, dd, J=11.4 and 9.5 Hz),
3.62 (1H, app t, J=11.3 Hz), 3.51 (1H, dd, J=9.3 and
3.4 Hz), 3.47–3.35 (2H, m), 3.22 (1H, dd, J=14.0 and 5.4
Hz), 2.95 (1H, dd, J=14.1 and 5.8 Hz), 2.59–2.55 (1H,
m), 2.46–2.40 (1H, m), 2.01–1.84 (2H, m), 1.68–1.60 (1H,
m), 1.51 (3H, s), 1.42 (3H, s), 1.36–1.26 (2H, m), 1.29
(3H, s), 1.26, (6H, s), 1.19 (3H, d, J=6.7 Hz), 0.97 (3H,
d, J=6.2 Hz), 0.95 (3H, app t, J=6.4 Hz); lC (100 MHz,
CDCl3) 172.6 (s), 171.8 (s), 170.7 (s), 170.4 (s), 170.3 (s),
169.8 (s), 169.2 (s), 142.7 (d), 142.2 (d), 136.4 (s), 129.6
(d), 128.2 (d), 126.8 (d), 115.8 (t), 115.1 (t), 77.9 (s), 77.8
(d), 76.0 (s), 67.3 (d), 62.0 (t), 59.7 (d), 57.6 (d), 56.4 (d),
56.1 (d), 53.6 (d), 47.8 (d), 47.3 (t), 40.5 (t), 36.5 (d), 35.8
(t), 29.8 (t), 27.4 (q), 25.9 (q), 25.7 (q), 25.5 (q), 25.2 (t),
23.9 (t), 18.9 (q), 18.4 (q), 16.2 (q), 12.2 (q); HRMS (ES)
.
.