Hemisynthesis of bistramide D by stereoselective reduction of bistramide A. Partial determination of relative and absolute configurations

Article abstract of DOI:10.1016/S0040-4039(00)01323-X

The hemisynthesis of bistramide D by stereoselective reduction of bistramide A is reported. The absolute configuration of C4 in bistramide D was also determined and the relative configurations of C6, C9 and C11 in the pyran part, as well as those of C22, C23, C27 and C31 in the spiro moiety of bistramides A and D were also determined by NMR. (C) 2000 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01323-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 7747–7750
Hemisynthesis of bistramide D by stereoselective reduction
of bistramide A. Partial determination of relative and
absolute configurations
a,
a
b
Guy Solladi e´ , * Claude Bauder and Jean-Fran c¸ ois Biard
a
Laboratoire de St e´ r e´ ochimie associ e´ au CNRS, Universit e´ Louis Pasteur, ECPM, 25 rue Becquerel,
7087 Strasbourg, Cedex 2, France
Institut Substances et Organisme de la mer (ISOMER), SMAB, Facult e´ de Pharmacie, Universit e´ de Nantes,
4035 Nantes, Cedex 01, France
6
b
4
Received 22 June 2000; revised 24 July 2000; accepted 1 August 2000
Abstract
The hemisynthesis of bistramide D by stereoselective reduction of bistramide A is reported. The
absolute configuration of C4 in bistramide D was also determined and the relative configurations of C6,
C9 and C11 in the pyran part, as well as those of C22, C23, C27 and C31 in the spiro moiety of
bistramides A and D were also determined by NMR. © 2000 Published by Elsevier Science Ltd.
Bistramides, a family of bioactive macrolides, were isolated from Lissoclinum bistratum
1a,b
Sluiter, an ascidian from New Caledonia. Five members (A, B, C, D and K) were identified.
Bistramide A was also isolated from the same ascidian collected in Australia and described as
1
c
2
3
bistratene A. This product has antiparasitic and immunomodulatory properties; it induces
4
5
atypical differentiation in HL-60 and in NSCLC-N6 cell lines. However, it is too toxic in vivo
for therapeutic purposes. Bistramides D and K, which are far less abundant, have the same
activity in the NSCLC-N6 cell line but are less toxic and were evaluated for their in vivo
6
antitumoral activity. We therefore intended to produce large amounts of bistramide D by
reduction of bistramide A. The first task was to determine the configurations of the chiral
centres. Faced with the noncrystallinity of such compounds, we assigned their absolute and
relative configurations by combining NMR spectroscopy and a synthetic correlation of natural
bistramide A with bistramide D (bst D).
To transform naturally available bistramide A (1) into bistramide D, we investigated the
reduction of the C4 carbonyl (Scheme 1) with different reducing agents, as shown in Table 1.
The first experiment with
L
-selectride (entry 1) gave a mixture of diastereomers D and D .
1
2
*
Corresponding author. Fax: 33 3 88 13 6949; e-mail: solladie@chimie.u-strasbg.fr
0
040-4039/00/$ - see front matter © 2000 Published by Elsevier Science Ltd.
PII: S0040-4039(00)01323-X

7748
Scheme 1.
Table 1
Reduction conditions of bistramide A to natural bistramide D
a
b
c
Entry
Reducing agent (equiv.)
-Selectride (5.3)
Solvent/time (h)
T (°C)
D (%)
D (%)
D3
Yield
1
2
1
2
3
4
5
6
L
THF/0.5
THF/16
Ether/11
THF/4
THF/4
THF/0.3
−78
−78
−78
−75
−75
−75
24
23
22
73
65
76
77
78
27
35
95
Yes
Yes
Yes
Yes
Yes
No
Quant.
93
LiAlH (6.5)
4
Zn(BH ) (4.6)
99
95
4
2
NaBH (5.9)
4
LiBH (5.5)
98
80
4
d
LiBH(Et) (3)
\5
3
a
1
Diastereomers % determined by H NMR of the crude product; TLC R values (acetone/CH Cl 6:4): D 0.7, D
2
f
2
2
1
0
.6.
b
D corresponds to the natural bst D.
2
c
Yield given for the crude product (D +D +D ).
1
2
3
d
The isolated yield of a mixture of D and D , no side product D detected by NMR.
1
2
3
These two epimers were separated and characterized by NMR. We have shown by TLC
1
13
comparison with natural bistramide D, by H and C NMR, as well as by HPLC, that the D2
isomer was identical to natural bistramide D.
We observed that by changing the reducing agents and the experimental conditions, the
diastereoselectivity could be drastically changed. LiAlH afforded a similar diastereoselectivity
4
(
(
50%, entry 2), the D isomer being the main product. Zinc borohydride gave a similar result
2
entry 3), while sodium and lithium borohydride led to the opposite stereoselectivity, D being
1
the main product (entries 4 and 5). In all these cases we observed the formation of an impurity,
D , which was not identified. We could, however, avoid the formation of the impurity D by
3
3
using 3 equivalents of superhydride so obtaining D in more than 90% de (entry 6). Alkoxy-
2
aluminium hydrides afforded only degradation products and no reduction was observed with
DIBAL-H.
7
In order to assign the absolute configuration of C4 in D , we used Mosher’s method:
2
esterification of the C4 hydroxyl group with the two enantiomers of 2-methoxy-2-phenyl-2-tri-
fluoromethyl acetyl chloride (MPTA-Cl), followed by an NMR study of the two resulting
diastereomers. It was first necessary to protect the two hydroxyl groups on C39 and C15 in bst
A (1) (Scheme 2) by acetylation (Ac O, pyridine, 96% yield) and then to reduce 4 with
2
superhydride to obtain 5 (80% yield). The configuration of the C4 hydroxyl in 5 was shown to
be the same as in bistramide D by reduction of the two acetoxy groups with LiAlH at low
4
temperature (80% yield), giving a sample identical to the natural product.

7749
Scheme 2.
Scheme 3. Mosher’s model
Mosher proposed the preferred conformation for MTPA esters in solution shown in Scheme
In the proton NMR, according to this conformation, the aromatic ring would be expected to
3
.
shield the protons of the G group in one diastereomer and the protons of the G group in the
1
2
other. By comparing the chemical shift displacements of the hydrogens close to the carbinol in
the two diastereomers, it is possible to assign the absolute configuration of the asymmetric
center. Many applications of this method have been reported in the literature to assign the
7
absolute configurations of carbinols.
From this model and from the data in Table 2, we can assign the R absolute configuration to
the carbon C4 of diacetylated bst D 5 and therefore to C4 of bst D 3.
Furthermore, the relative configurations of several chiral centers were assigned by a 2D 500
1
MHz H NMR analysis of diacetylated bst A 4. The NOESY spectrum showed, for the cyclic
pyran part, a strong long-range correlation between protons H6 and H12, indicating that the H6
proton, as well as the 12-CH methine were both axially oriented (Scheme 4).
2
Scheme 4.
In addition, a cis relationship between H9 and H11 was suggested on the basis of a clear
H– H NOESY correlation between these two nuclei. Thus the relative structure of the pyran
moiety was established as either (6R,9S,11S) or (6S,9R,11R).
1
1

7750
Table 2
1
H NMR (500 MHz) chemical shift differences (Dl) for MTPA derivatives 6 and 7
Protons
(S)-MTPA ester 6
(R)-MTPA ester 7
Dl
a
a
Me1
H2
H3
H5
H5b
1.59
5.96
5.41
2.16
1.56
3.60
1.58
5.95
5.53
2.11
1.57
3.39
+0.01
+0.01
−0.12
+0.05
−0.01
+0.21
a
H6
a
No significant change.
1
1
Further NOESY analysis showed a very clear H– H correlation between H22 and H31
showing the axial orientation of these two protons in the spiran part of the molecule. Moreover,
1
1
H22 exhibited a long-range H– H correlation with the adjacent methyl group on carbon C23,
which must be equatorial. The lack of any NOESY correlation between H22 and H23 indicated
that they have the trans configuration (Scheme 4). Therefore the two possible relative configura-
tions of the spiran part must be (22S,23R,27R,31R) or (22R,23S,27S,31S).
In conclusion we have reported a stereoselective transformation of bst A into bst D and
determined the absolute configuration of C4. We have also assigned by 2D NMR, the relative
configuration of three chiral centers in the pyran part and four in the spiro moiety of bst 4.
Acknowledgements
This work was supported by a grant from ANVAR (grant A 94 09 081 M 00), which is
gratefully acknowledged.
References
1. (a) Gouiffes, D.; Juge, M.; Grimaud, N.; Welin, L.; Sauviat, M. P.; Barbin, Y.; Laurent, D.; Roussakis, C.;
Henichart, J. P.; Verbist, J. F. Toxicon 1988, 26, 1129; (b) Biard, J. F.; Roussakis, C.; Kornprobst, J. M.; Verbist,
J. F.; Cotelle, P.; Foster, M. P.; Ireland, C. M.; Debitus, C. J. Nat. Prod. 1994, 57, 1336; (c) Degnan, B. M.;
.
Hawkins, C. J.; Lavin, M. F.; McCaffrey, E. J.; Parry, D. L.; Watters, D. J. J. Med. Chem. 1989, 32, 1354.
. Gautret, P.; Le Pape, P.; Biard, J. F.; Menard, D.; Verbist, J. F.; Marjolet, M. Acta Parasitol. 1998, 43, 50.
. Pusset, J.; Maillere, B.; Debitus, C. J. Nat. Toxins 1996, 5, 1.
2
3
4
. Johnson, W. E.; Watters, D.; Suniara, R. K.; Brown, G.; Bruce, C. M. Biochem. Biophys. Res. Commun. 1999, 260,
80.
5. Siavoshian, S.; Jacquot, C.; Biard, J. F.; Briand, G.; Roussakis, C. Anticancer Res. 1999, 19, 5361.
6. Biard, J. F.; Cortadellas, D.; Debitus, C.; Laurent, D.; Roussakis, C.; Verbist, J. F. PCT/FR 94/00256, 8 March
1994.
7
. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512. For some recent applications of Mosher’s method see:
Kusumi, T.; Ohtani, I.; Inouye, Y.; Kakisawa, H. Tetrahedron Lett. 1988, 29, 4731. Natsume, M.; Yasui, K.;
Kondo, S.; Marumo, S. Tetrahedron Lett. 1991, 32, 3087. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J.
Am. Chem. Soc. 1991, 113, 4092. Arce, E.; Carre n˜ o, M. C.; Cid, M. B.; Garcia-Ruano, J. L. J. Org. Chem. 1994,
59, 3421. Tse, B.; Blazey, C. M.; Tu, B.; Balkovec, J. J. Org. Chem. 1997, 62, 3236.