Synthesis of the middle fragment of oxazolomycin

Article abstract of DOI:10.1016/S0040-4039(02)02408-5

A stereoselective and direct synthesis of an (E,E)-diene suitable for application to the synthesis of oxazolomycin and its analogues is reported.

Full text of DOI:10.1016/S0040-4039(02)02408-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 9629–9632
Synthesis of the middle fragment of oxazolomycin
Zhaoyang Wang and Mark G. Moloney*
The Department of Chemistry, Dyson Perrins Laboratory, The University of Oxford, South Parks Road, Oxford OX1 3QY, UK
Received 8 August 2002; revised 21 October 2002; accepted 24 October 2002
Abstract—A stereoselective and direct synthesis of an (E,E)-diene suitable for application to the synthesis of oxazolomycin and
its analogues is reported. © 2002 Elsevier Science Ltd. All rights reserved.
Oxazolomycin A 1a is the parent member of a class of
the membrane fusions essential for the ingestion of a
virion by endocytosis, its release into the host cell and
the expulsion of new infectious particles have been
1
antibiotics, other members being oxazolomycin B 1b
2
and C 1c, the 16-methylisomer 1d, and the curro-
3
8,9
mycins 1e,f. The oxazolomycins exhibit wide ranging
studied in detail. The intrinsically acidic pH of the
and potent antibiotic activity, including inhibitory
activity against Gram-positive bacteria, antiviral activ-
ity against vaccinia, herpes simplex type I and influenza
A, as well as in vivo antitumour activity, but with low
toxicity (LD50 of 10.6 mg/kg for intraperitoneal injec-
tion in mice). A recent report has described the potent
antibacterial and cytotoxicity of 16-methyloxa-
zolomycin 1d (MIC against Bacillus subtilis 5 mg/ml and
endosome is of primary importance for the dispersal of
the viral RNA replication machinery into the cytosol.
Under these acidic conditions, a significant conforma-
tional change (triggered by the protonation of six Asp
side chain carboxylates in a 19 residue spacing segment,
reducing charge–charge repulsions between them)
translocates a membrane-bound fusion peptide by some
100 A in such a way that it can bridge both the viral
,
4
10
IC against P388 leukaemia cells, 0.23 mg/ml). Some
and cellular membranes, thereby facilitating fusion.
Significantly, it has been shown that lipid soluble bases
(such as NH and Me N) can inhibit viral infection, by
5
0
of the details of the biosynthesis of this structurally
5
novel class of compounds have been established.
3
3
raising the pH of the normally acidic endosomes, and
preventing these protonations. This also forms the basis
of the action of the well-known anti-influenza drugs,
amantidine and rimantidine, and it is our hypothesis
that oxazolomycin acts similarly. However, the low
toxicity of oxazolomycin (vide supra) is surprising, and
this suggests that its protonophoric activity might be
selective for the endosomal membrane over the mito-
The broad-spectrum activity of the oxazolomycins has
been attributed to their protonophoric (i.e. proton carry-
ing) properties, and the pharmacological interest in
these compounds comes not only from their antibiotic
activity but also their unusual mode of action. The
significance of this mode of action is apparent from a
consideration of the life-cycle of influenza A, for which
6
,7
*
0
Corresponding author. E-mail: mark.moloney@chem.ox.ac.uk
040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02408-5

9
630
Z. Wang, M. G. Moloney / Tetrahedron Letters 43 (2002) 9629–9632
Figure 1.
double bond geometry and of the length of the spacer
group between the left and right hand fragments on
biological activity. We report here a direct and versatile
route to the middle fragment of the oxazolomycins
which makes use of the elegant work of Andrus for the
selective dihydroxylation of terminal alkenes over more
Scheme 1. Reagents and conditions: (i) FmocCl, py, DCM
15–17
substituted ones.
(
83%); (ii) n-Bu SnH, AlBN (73%); (iii) I , DCM (79%).
3 2
We planned to use a similar approach to that of
1
2
chondrial membrane and therefore does not disrupt
general mitochondrial ATP synthesis, for which the
generation of a proton gradient is also required. Fur-
thermore, the action of the oxazolomycins is made even
more unusual and potentially valuable by its disruption
of membrane fusion early in the viral life-cycle. We
hypothesise that the antibacterial and cytotoxic nature
of these compounds results from a similar disruption of
essential membrane fusion activity.
Kende, by applying a Stille coupling for the diene
fragment. To this end, vinylstannane 3a was prepared
from propargylamine 2a by initial protection as the
Fmoc derivative 2b and subsequent hydrostannylation
12
using the literature procedure and this was converted
to vinyl iodide 3b using standard methodology (Scheme
1).
For the other component, Hiyama methodology involv-
ing Cr(II)-mediated addition of an allyl bromide to an
aldehyde, was used for the construction of the
18
In order to probe in more detail the molecular basis of
action of the oxazolomycins, we require access to both
the natural products and to analogues. The only extant
total synthesis of any of these compounds, namely
required anti stereochemistry (Scheme 2). Reaction of
19
stannane 4b, prepared from 4a by Swern or PCC
oxidation, with crotyl bromide in the presence of
chromium(III) chloride/lithium aluminium hydride
gave allylic alcohol 5a, whose anti-stereochemistry was
established using a combination of NOESY and molec-
ular modelling analysis (Fig. 2), consistent with the
literature precedent. Chem 3D analysis indicated a sin-
gle preferred minimum energy conformation in which
the C-1/C-6 backbone is such that the two double
bonds are approximately coplanar and parallel; strong
1
1
neooxazolomycin, is very lengthy, and relies on an
efficient Stille coupling approach for the diene frag-
1
2
ment. Using a simple retrosynthetic strategy based
upon amide and 1,2-diol disconnections, the three frag-
ments indicated in Fig. 1 were identified as pivotal
intermediate targets, and synthetic strategies for the left
13,14
and right hand fragments have been developed.
Key
questions worthy of examination include the effect of
Scheme 2. Reagents and conditions: (i) PCC/DCM or (ClCO) /DMSO (70%); (ii) crotyl bromide, CrCl /LaH (73%); (iii)
2
3
TBDMSCl, imidazole, DCM (90%); (iv) TBDPSCl, imidazole, DCM (90%); (v) AD-mix-a, t-BuOH, H O (72%); (vi) NaIO ,
2
4
THF, H O (89%); (vii) NaBH , EtOH (85%); (viii) 3b, Pd(CH CN) Cl /THF (65%).
2
4
3
2
2

Z. Wang, M. G. Moloney / Tetrahedron Letters 43 (2002) 9629–9632
9631
and its analogues, suitable for the elucidation of its
biological action, and details of this work will be pub-
lished in due course.
Acknowledgements
We acknowledge the use of the EPSRC Chemical Data-
22
base Service at Daresbury and the EPSRC National
Mass Spectrometry Service Centre at Swansea. Zhaoy-
ang Wang is grateful to the China Scholarship Council
for the award of a Fellowship.
Figure 2. NOESY enhancements for 5a (butyl residues on Sn
simplified for clarity).
enhancement sequences from H-1“H-4“H-6 and H-
3
“H-5“H-7 and H-5“CH are most consistent with
3
an anti- C3(Me)ꢀC4(OH) arrangement. Protection as
the TBDMS or TBDPS ether gave 5b,c in excellent
yield (90%). At this point, the sequence diverged, and
each protected derivative was used to investigate differ-
ent routes. TBDMS ether 5b was treated with AD-mix-
References
1
. Mori, T.; Takahashi, K.; Kashiwabara, M.; Uemura, D.;
Katayama, C.; Iwadare, S.; Shizuri, Y.; Mitomo, R.;
Nakano, F.; Matsuzaki, A. Tetrahedron Lett. 1985, 26,
1
5
a according to the Andrus protocol to afford diol 6 in
1073–1076.
7
2% yield as a 1:1 mixture of diastereomers, which was
immediately converted to aldehyde 7a (NaIO , THF,
2. Kanzaki, H.; Wada, K.; Nitoda, T.; Kawazu, K. Biosci.
Biotechnol. Biochem. 1998, 62, 438–442.
3. Kanzaki, H.; Ichioka, T.; Kobayashi, A.; Kawazu, K.
Biosci. Biotechnol. Biochem. 1996, 60, 1535–1537.
4. Ryu, G.; Kim, S.-K. J. Antibiot. 1999, 52, 193–197.
4
H O) and thence to alcohol 7b (NaBH , EtOH) in 85%
2
4
yield over the two steps. Coupling of this compound
with iodide 3b under standard Stille conditions gave
2
0
diene 8 in 65% yield.
5
6
7
8
9
. Grafe, U.; Kluge, H.; Thiericke, R. Liebigs. Ann. Chem.
1992, 429–432.
. Grigorjev, P. A.; Schlegel, R.; Grafe, U. Pharmazie 1992,
In order to investigate an alternative coupling strategy
to generate intermediates which might prove to be
crystalline and therefore permit independent stereo-
chemical assignment, the silyl ether 5c was converted to
47, 707–709.
. Tonew, E.; Tonew, M.; Grafe, U.; Zopel, P. Acta Virol.
1992, 36, 166–172.
. Bullough, P. A.; Hughson, F. M.; Skehel, J. J.; Wiley, D.
C. Nature 1994, 371, 37–43.
vinyl iodide 9 (I , DCM, 93% yield) (Scheme 3) and
2
thence to diol 10 in 65% yield as a 1:1 mixture of
15
diastereomers by application of the Andrus protocol.
This product was converted to aldehyde 11a (92% yield)
and then to alcohol 11b (91% yield) via the same
strategy outlined above. These products however,
proved to be colourless oils. Stille coupling of this
compound with 3a gave the diene product 12a in 65%
yield. Alternatively, alcohol 11b was converted to bro-
mide 11c which was in turn coupled to vinyl stannane
. Carr, C. M.; Kim, P. S. Cell 1994, 73, 823–832.
1
1
0. Doms, R. W. Methods Enzymol. 1993, 221, 61–72.
1. Kende, A. S.; Kawamura, K.; deVita, R. J. J. Am. Chem.
Soc. 1990, 112, 4070–4072.
12. Kende, A. S.; deVita, R. J. Tetrahedron Lett. 1990, 31,
3
07–310.
13. Bulger, P. G.; Moloney, M. G.; Trippier, P. C. Synlett
002, 1871.
2
1
3a to give diene 12b in 65% yield.
2
This route represents a direct and versatile approach to
a key intermediate for the synthesis of oxazolomycin
14. Andrews, M. D.; Brewster, A. G.; Moloney, M. G.
Synlett 1996, 612–614.
Scheme 3. Reagents and conditions: (i) AD-mix-a, t-BuOH, H O (65%); (ii) NaIO , THF, H O (92%); (iii) NaBH , EtOH (91%);
2
4
2
4
(
iv) CBr , PPh , THF (85%); (v) 3a, Pd(CH CN) Cl /THF (65%).
4
3
3
2
2

9
632
Z. Wang, M. G. Moloney / Tetrahedron Letters 43 (2002) 9629–9632
1
1
1
1
5. Andrus, M. B.; Lepore, S. D.; Turner, T. M. J. Am.
Chem. Soc. 1997, 119, 12159–12169.
6. Andrus, M. B.; Lepore, S. D.; Sclafani, J. A. Tetrahedron
Lett. 1997, 38, 4043–4046.
7. Andrus, M. B.; Turner, T. M.; Asgari, D.; Li, W. J. Org.
Chem. 1999, 64, 2978–2979.
126.91, 127.57, 129.33, 129.76, 130.97, 135.38, 141.21,
143.80, 156.09 (CꢁO). HRMS: 530.2703 (M+Na ), calcd
+
for C H NO SiNa: 530.2703.
30
41
4
2
1. Data for N-Fmoc-8-amino-1-bromo-3-(tert-butyldimethyl-
silanyloxy)-2-methyl-octa-4,6-diene 12b: R : 0.50(Petrol/
f
1
EA=3:1). H NMR (500 MHz, CDCl ) l : 0.96 (d, 3H,
3
H
8. Hiyama, T.; Kimura, K.; Nozaki, H. Tetrahedron Lett.
CH , J=6.8 Hz), 1.07 (s, 9H, C(CH ) ), 2.00 (m, 1H,
3
3 3
1
981, 22, 1037–1040.
H-2), 3.33 and 3.39 (m, 2H, H-1 and H-1%), 3.80 (br, 2H,
19. Jung, M.; Light, L. Tetrahedron Lett. 1982, 23, 3851.
0. Data for N-Fmoc-8-amino-3-(tert-butyldimethylsilanyl-
H-8), 4.23 (m, 2H, H-3 and CHCH ), 4.45 (d, J=6.8 Hz,
2
2
2
H, CHCH ), 4.75 (br, 1H, NH), 5.46 (m, 2H, H-4 and
2
oxy)-1-hydroxy-2-methyl-octa-4,6-diene 8: Rf: 0.40
H-7), 5.68 (q, 1H, H-5), 5.92 (m, 1H, H-6), 7.37 (m,
1
(
Petrol/EA=2:1). H NMR (500 MHz, CDCl ) l : 0.08
3
H
13
1
0H), 7.64 (m, 6H), 7.79 (d, 2H, J=7.5 Hz); C NMR
(
s, 3H, CH ), 0.10 (s, 3H, CH ), 0.92 (m, 12H, CH and
3
3
3
(
100.6) MHz) l : 14.21 (CH ), 19.29, 26.94, 37.13 (C-1),
C
3
C(CH ) ), 1.74 (m, 1H, H-2), 2.67 (br, 1H, OH), 3.58 (m,
3
3
41.76 (C-2), 42.49 (C-8), 47.13 (CHCH2), 66.55
1
1
H, H-1), 3.72 (m, 1H, H-1%), 3.88 (br, 2H, H-8), 4.11 (t,
(
CHCH ), 75.93 (C-3), 119.88, 124.88, 126.92, 127.24,
2
H, H-3), 4.23 (t, 1H, CHCH ), 4.44 (d, J=6.8 Hz, 2H,
2
1
1
27.42, 127.58, 129.35, 129.56, 130.88, 131.53, 132.29,
33.62, 133.94, 135.83, 135.87, 141.21, 143.90, 156.04
CHCH ), 4.82 (br, 1H, NH), 5.67 (m, 2H, H-4 and H-7),
2
6
7
.15 (m, 2H, H-5 and H-6), 7.33 (t, J=7.4 Hz, 2H, ArH),
.42 (t, J=7.4 Hz, 2H, ArH), 7.61 (d, J=7.4 Hz, 2H,
(
CꢁO). MS(ES): 711.26 (M+NH ), 684.20 (M+1), 610.19,
4
13
550.50, 536.17, 440.11. HRMS: 711.2604 (M+NH ), calcd
ArH), 7.78 (d, J=7.5 Hz, 2H, ArH); C NMR (125.73
MHz) l : −4.11 (CH ), 13.89 (CH ), 17.92 (CH ), 25.70
4
for (C H NO Si+NH ): 711.2618.
40 45
4
4
C
3
3
3
(
6
C(CH ) ), 41.03 (C-2), 42.57 (C-8), 47.12 (CHCH ),
22. Fletcher, D. A.; McMeeking, R. F.; Parkin, D. J. Chem.
Inf. Comput. Sci. 1996, 36, 746–749.
3
3
2
5.87 (C-1), 66.58 (CHCH ), 78.44 (C-3), 119.87, 124.88,
2