Approaches to the total synthesis of phomactins

Article abstract of DOI:10.1016/S0040-4039(03)00341-1

The macrocyclic triene 34, an advanced intermediate for synthesis of phomactins, has been synthesized using the 2,3-Wittig rearrangement of the propargylic ether 21 as a key step.

Full text of DOI:10.1016/S0040-4039(03)00341-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 2713–2716
Approaches to the total synthesis of phomactins
Andrew S. Balnaves, Graham McGowan, Peter D. P. Shapland and Eric J. Thomas*
The Department of Chemistry, The University of Manchester, Manchester M13 9PL, UK
Received 14 January 2003; accepted 4 February 2003
Abstract—The macrocyclic triene 34, an advanced intermediate for synthesis of phomactins, has been synthesized using the
2,3-Wittig rearrangement of the propargylic ether 21 as a key step. © 2003 Elsevier Science Ltd. All rights reserved.
The phomactins, e.g. A 1, D 2, E 3, and Sch. 49028 4,
are diterpenes with novel structures and biological activ-
ity as platelet activating factor antagonists.1 Several
groups have reported synthetic studies2 directed towards
the phomactins, including total syntheses of phomactin
D 23 and A 1.4
By analogy with the synthesis of phomactin D,3 ketone
5 should be accessible by cyclisation of sulfone 6 which
could be synthesized by a 2,3-Wittig rearrangement of a
suitable (cyclohexenyl)methyl ether.6 Preliminary studies
have shown that whereas rearrangement of (cyclohex-
enyl)methyl dimethylallyl ethers takes place with the
undesired regioselectivity, rearrangement of analogous
propargyl ethers gives the required regioisomers, e.g. 8
An approach to the phomactins can be envisaged in
gives 9 albeit as a mixture of diastereoisomers.7,8 On this
which the trienyl ketone 5 is an advanced intermediate
basis, the propagylic ether 7 was identified as a suitable
able to provide access to several members of the
substrate for the Wittig rearrangement with the conver-
phomactin family. For example, stereoselective reduction
sion of the alkyne into the trisubstituted double-bond to
of the ketone followed by hydroxyl directed epoxidation
be carried out either before or after macrocyclisation. We
of the 3,4- and exocyclic double-bonds, reoxidation to the
here report a synthesis of the bicyclic trienyl alcohol 34,
bis-epoxyketone and isomerisation of the exocyclic b,g-
a congener of the ketone 5 based on this strategy.
epoxide, should lead to Sch. 49028 4 and hence to
phomactin A 1 after deprotection.5 Similar schemes
could be devised to access phomactins D and E from
ketone 5.
* Corresponding author. Tel.: 0044 (0)161 275 4614; fax: 0044 (0)161
275 4939; e-mail: e.j.thomas@man.ac.uk
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00341-1

2714
A. S. Balna6es et al. / Tetrahedron Letters 44 (2003) 2713–2716
The synthesis of the Wittig rearrangement precursor 21
is outlined in Scheme 1. Dissolving metal reduction9–
methylation of o-toluic acid 10 followed by esterifica-
tion and allylic oxidation10 gave the cyclohexadienone
11. Conjugate addition of the lower order lithium
dimethylcuprate to 11 was regioselective for the less
substituted double-bond, and stereoselective for addi-
tion cis to the methyl substituent, giving the cyclo-
hexenone 12 as the major product,11 stereoselectivity ca.
85:15, although the stereoisomers were not usually sep-
arated at this stage. Free-radical bromination then gave
a mixture of the corresponding bromomethyl com-
pounds from which the required major isomer 13 was
separated by chromatography. Luche reduction of the
enone was highly stereoselective giving alcohol 14 and
protection of the alcohol gave the SEM-ether 15.12
required sulfone 21 by reduction and substitution of the
mesylate of the alcohol so obtained with sodium thio-
phenoxide to give a sulfide. Oxidation of the sulfide to
the sulfone 21 was complicated by the presence of the
trisubstituted double bonds, but was best achieved
using ammonium molybdate and hydrogen peroxide in
methanol.15 Under these conditions an acceptable, 65%,
yield of the sulfone 21 was obtained together with some
of the intermediate sulfoxide which could be separated
and oxidised to give more sulfone.
The 2,3-Wittig rearrangement of the sulfone 21 was
carried out using 2.5 equivalents of n-butyllithium at
−78°C and gave an 82:18 mixture of the rearrangement
products 22 and 23. The configurations of these at C-1
were established by NMR but the configurations of the
hydroxyl bearing carbons were not confirmed until later
in the synthesis.16
The configuration of the hydroxyl group in the alcohol
14 is the opposite of that required at C-14 for incorpo-
ration into phomactin A 1 and Sch. 49028 4. However,
the configuration at this centre was not changed at this
stage since it was thought that macrocyclisation of
intermediates with the unnatural configuration at C-14
(phomactin numbering) would be less susceptible to
steric hindrance.
At this point it was necessary to decide whether to form
the macrocyclic ring first and then convert the alkyne
into the required trisubstituted alkene en route to the
target ketone 5, or to introduce the alkene before
macrocyclisation. In order to gain experience of the
macrocyclisation procedure, it was decided to study the
former option first.17
The acetylenic component 19 was prepared from the
protected propargylic alcohol 16 by reaction of the
corresponding lithium acetylide with oxetane13 to give
the alcohol 17. A Wittig reaction of the corresponding
aldehyde then gave the (E)-a,b-unsaturated ester 18
and a reduction, protection and selective deprotection
sequence led to the required alcohol 19. Alkylation14 of
this alcohol using the cyclohexenylmethyl bromide 15
gave the ether 20 which was taken through to the
The major Wittig rearrangement product 22 was pro-
tected as its p-methoxybenzyl ether 24 which was con-
verted into the allylic bromide 25 by selective removal
of the tert-butyldiphenylsilyl ether, mesylation and dis-
placement of the allylic mesylate using lithium bromide
(Scheme 2). Cyclisation was achieved by syringe pump
addition (over 1 h) of sodium hexamethyldisilazide to a
solution of the sulfone-bromide in THF at 0°C and,
Scheme 1. Reagents and conditions: (i) Li, liq. NH3, THF, −78°C then MeI (91%); (ii) AcCl, MeOH, rt (88%); (iii) PDC, Celite,
TBHP, C6H6, 0°C (90%); (iv), LiCuMe2, toluene, 0°C (88%; 80:20); (v) NBS, AIBN, CCl4, heat under reflux (72%); (vi) NaBH4,
CeCl3·7H2O, MeOH, −78°C (91%); (vii) SEM-Cl, i-Pr2NEt, DCM, 0°C–rt (80%); (viii) n-BuLi, oxetane, BF3·Et2O, −78°C (85%);
(ix) DMSO, (COCl)2, Et3N, DCM, −78°C–rt; (x) Ph3PCMe·CO2Et, −78°C–rt (91% from 17); (xi) LiBHEt3, THF, −78°C (84%);
(xii) TBDPS-Cl, imid., DCM, 0°C (67%); (xiii) 1:1 MeOH:CCl4, 50°C, sonicate (86%); (xiv) NaH, 15-crown-5, TBAI, THF, rt
(65%); (xv) LiBHEt3, THF, 0°C (88%); (xvi) (a) MsCl, Et3N, DCM, 0°C, (b) NaSPh, DMF, heat under reflux (85%); (xvii)
ammonium molybdate, H2O2, EtOH, −20°C–rt (65%).

A. S. Balna6es et al. / Tetrahedron Letters 44 (2003) 2713–2716
2715
Desulfonylation and oxidative removal of the p-
methoxybenzyl group then gave the alcohol 27. This
sequence confirmed that intramolecular sulfone alkyla-
tion could be used to assemble the bicyclic ring system
present in the phomactin A quite effectively.3 However,
preliminary attempts, using Red-Al or LiAlH4 with an
iodine quench,19 to convert the alkyne present in 27
into a trisubstituted alkene returned unchanged starting
material,20 and so it was necessary to effect this conver-
sion before macrocyclisation.
Oxidation of the mixture of Wittig rearrangement prod-
ucts 22 and 23 using the Dess Martin periodinane gave
the ketone 28 (Scheme 3). Preliminary attempts to add
lithium dimethylcuprate directly to this enone gave
approximately 1:1 mixtures of the geometrical isomers
of the corresponding enone and so stereoselective addi-
tion of benzenethiol was investigated. Using
a
Scheme 2. Reagents and conditions: (i) NaH, PMB-Cl, TBAI,
DMF, THF, 0°C–rt (58%); (ii) TBAF, THF, rt (87%); (iii)
MsCl, Et3N, DCM, rt, 30 min then LiBr, acetone, 0°C, 30
min (92%); (iv) NaHMDS, THF, 0°C, 1.5 h (68%); (v) 5%
Na/Hg, Na2HPO4, MeOH, THF, 0°C (93%); (vi) DDQ, pH 7,
DCM, H2O (72%).
methanolic solution of benzenethiol and triethyl-
amine,21 a separable 72:28 mixture of the (Z)- and
(E)-vinyl sulfides was obtained from which the required
(Z)-isomer 29 could be isolated in a 69% yield. Treat-
ment of this sulfide with lithium dimethyl cuprate then
gave the required (E)-alkene 30, with retention of
configuration, in excellent yield.22
after stirring for an extra 30 min, gave a 68% yield of
macrocyclic product which was isolated essentially as
the single diastereoisomer 26.
Considerable difficulty was experienced during attempts
to reduce the ketone 30 to the alcohol 31; the recovered
starting material was obtained using Luche’s conditions
and over-reduction with LiAlH4. In the end, the reduc-
tion was achieved using NaBH4 in ethanol, but only in
a 57% yield. Nevertheless this alcohol was protected as
its benzyloxymethyl (BOM) ether 32, and selective
The indicated stereochemistry was assigned to this
macrocyclic product on the basis of NMR experi-
ments.18 Of interest is the fact that high dilution condi-
tions were not necessary for this cyclisation with 90 mg
(0.119 mmol) of 25 being cyclised in 1.5 cm3 THF.
Scheme 3. Reagents and conditions: (i) DMP, DCM, rt (89%); (ii) PhSH, Et3N, THF, MeOH, −20°C (69% together with 26% of
the 3,4-(E)-isomer); (iii) LiCuMe2, Et2O, −78°C (97%); (iv) NaBH4, EtOH, rt (57%); (v) BOM-Cl, i-Pr2NEt, TBAI, THF, rt
(85%); (vi) TBAF, THF, rt (79%); (vii) MsCl, Et3N, DCM then LiBr, acetone (69%); (viii) NaHMDS, THF, 0°C; (ix) Na, liq.
NH3, THF, EtOH, −60°C (45% from 33).

2716
A. S. Balna6es et al. / Tetrahedron Letters 44 (2003) 2713–2716
deprotection of the primary alcohol and conversion
into the allylic bromide 33 was carried out as before.
Macrocyclisation was achieved using syringe pump
addition of sodium hexamethyldisilazide and simulta-
neous removal of the phenyl sulfone and BOM groups
gave the required bicyclic trienyl alcohol 34 in a 45%
yield over both the cyclisation and reduction steps.
6. (a) Marshall, J. A. In Comp. Organic Synthesis; Trost, B.
M.; Fleming, I., Eds.; Pergamon Press: Oxford, 1991;
Vol. 3, Chapter 3.11; (b) Mikami, K.; Nakai, T. Synthesis
1991, 594; (c) Nakai, T.; Mikami, K. Org. React. 1994,
46, 105.
7. Marsden, A.; Thomas, E. J. Arkivoc 2002, part ix.
8. Nakai, T.; Mikami, K.; Taya, S.; Fujita, Y. J. Am. Chem.
Soc. 1981, 103, 6492.
The alcohol 34 corresponds to the key intermediate
ketone 5 albeit with the opposite configuration at C-14.
Present work is concerned with the development of the
final stages of a synthesis of phomactins from this
alcohol.
9. (a) Acheson, R. M.; Flowerday, R. F. J. Chem. Soc.,
Perkin Trans. 1 1974, 2339; (b) Linker, T.; Frohlich, L. J.
Am. Chem. Soc. 1995, 117, 2694.
10. Schultz, A. G.; Taveras, A. G.; Harrington, R. E. Tetra-
hedron Lett. 1988, 29, 3907.
11. The stereochemistry of the major enone 12 was confirmed
by X-ray diffraction. Details will be published in a full
paper.
Acknowledgements
12. The stereochemistry of the reduction product 14 was
confirmed by NOE studies, e.g. a significant enhancement
of CHMe on irradiation of CHOH.
We thank the EPSRC for support (to A.S.B., G.M. and
P.D.P.S.), GlaxoSmithKline and Merck Sharp and
Dohme for CASE awards (to P.D.P.S. and G.M.,
respectively), Dr. D. Morgan (GSK) and Dr. L. Street
(MSD) for helpful discussions, and the EPSRC mass
spectrometry service at Swansea for mass spectra.
13. Eis, M. J.; Wrobel, J. E.; Ganem, B. J. Am. Chem. Soc.
1984, 106, 3693.
14. Aspinall, H. C.; Greeves, N.; Lee, W.-M.; McIver, E. G.;
Smith, P. M. Tetrahedron Lett. 1997, 38, 4679.
15. (a) Hardy, P. M.; Rydon, H. N.; Thompson, R. C.
Tetrahedron Lett. 1968, 2525; (b) Ferezou, J. P.; Julia, M.
Tetrahedron 1990, 46, 475.
References
16. The configuration at C-1 (phomactin numbering) in both
22 and 23 also follows from their successful conversion
into the bicyclic intermediates 27 and 34.
17. An alternative approach was also investigated which
involved cyclisation of ether 21 to give the macrocyclic
ether i. However, attempts to effect the required 2,3-Wit-
tig rearrangement of i were unsuccessful, perhaps because
of ring strain in the product. Details will be published in
a full paper.
1. (a) Sugano, M.; Sato, A.; Iijima, Y.; Oshima, T.; Furuya,
K.; Kuwano, H.; Hata, T.; Hanzawa, H. J. Am. Chem.
Soc. 1991, 113, 5463; (b) Chu, M.; Patel, M. G.; Gullo,
V. P.; Truumees, I.; Puar, M. S. J. Org. Chem. 1992, 57,
5817; (c) Chu, M.; Truumees, I.; Gunnarsson, I.; Bishop,
W. R.; Kreutner, W.; Horan, A. C.; Patel, M. G.; Gullo,
V. P.; Puar, M. S. J. Antibiot. 1993, 46, 554; (d) Sugano,
M.; Sato, A.; Iijima, Y.; Furuya, K.; Haruyama, H.;
Yoda, K.; Hata, T. J. Org. Chem. 1994, 59, 564; (e)
Sugano, M.; Sato, A.; Iijima, Y.; Furuya, K.; Kuwano,
H.; Hata, T. J. Antibiot. 1995, 48, 1188.
2. (a) Foote, K. M.; Hayes, C. J.; Pattenden, G. Tetra-
hedron Lett. 1996, 37, 275; (b) Chen, D.; Wang, J.; Totah,
N. I. J. Org. Chem. 1999, 64, 1776; (c) Seth, P. P.; Totah,
N. I. J. Org. Chem. 1999, 64, 8750; (d) Seth, P. P.; Chen,
D.; Wang, J.; Gao, X.; Totah, N. I. Tetrahedron 2000, 56,
10185; (e) Seth, P. P.; Totah, N. I. Org. Lett. 2000, 2,
2507; (f) Kallan, N. C.; Halcomb, R. L. Org. Lett. 2000,
2, 2687; (g) Chemler, S. R.; Danishefsky, S. J. Org. Lett.
2000, 2, 2695; (h) Chemler, S. R.; Iserloh, U.; Danishef-
sky, S. J. Org. Lett. 2001, 3, 2949; (i) Foote, K. M.; John,
M.; Pattenden, G. Synlett 2001, 365; (j) Mi, B.;
Maleczka, R. E., Jr. Org. Lett. 2001, 3, 1491; (k)
Houghton, T. J.; Choi, S.; Rawal, V. H. Org. Lett. 2001,
3, 3615; (l) Mohr, P. J.; Halcomb, R. L. Org. Lett. 2002,
4, 241.
3. Miyaoka, H.; Saka, Y.; Miura, S.; Yamada, Y. Tetra-
hedron Lett. 1996, 37, 7107.
4. Goldring, W. P. D.; Pattenden, G. Chem. Commun. 2002,
1736; Halcomb, R. L.; Mohr, P. J. J. Am. Chem. Soc.
2003, 125, in press.
5. The spontaneous rearrangement of Sch. 49028 into
phomactin A was observed by Pattenden et al. during
their synthesis, see Ref. 4.
18. Coupling constants and NOE observations established
the configurations indicated at C-2 and C-10 (phomactin
numbering) in 26.
19. Corey, E. J.; Katzenellenbogen, J. A.; Posner, G. H. J.
Am. Chem. Soc. 1967, 89, 4245.
20. In Rawal’s approach to the phomactins, conjugate addi-
tion of lithium dimethylcuprate to a bicyclic ketone
analogous to that which would be formed on oxidation
of 27, gave a 2.5:1 mixture of the (Z)- and (E)-enones in
favour of the unwanted (Z)-isomer; see Ref. 2k.
21. Gardiner, J. M.; Giles, P. E. Tetrahedron Lett. 1995, 36,
7519.
22. Dieter, R. K.; Silks, L. A., III J. Org. Chem. 1986, 51,
4687.