Cobalt versus Osmium: Control of Both trans and cis Selectivity in Construction of the EFG Rings of Pectenotoxin 4

Article abstract of DOI:10.1002/anie.201708278

Catalytic oxidative cyclisation reactions have been employed for the synthesis of the E and F rings of the complex natural product target pectenotoxin 4. The choice of metal catalyst (cobalt- or osmium-based) allowed for the formation of THF rings with either trans or cis stereoselectivity. Fragment union using a modified Julia reaction then enabled the synthesis of an advanced synthetic intermediate containing the EF and G rings of the target.

Full text of DOI:10.1002/anie.201708278

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Cobalt versus osmium: control of both trans and cis selectivity in
construction of the EFG rings of pectenotoxin 4
Authors: Ahria Roushanbakhti, Yifan Liu, Paul Winship, Michael
Tucker, Wasim Akhtar, Daryl Walter, Gail Wrigley, and
Timothy James Donohoe
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201708278
Angew. Chem. 10.1002/ange.201708278
Link to VoR: http://dx.doi.org/10.1002/anie.201708278
http://dx.doi.org/10.1002/ange.201708278

10.1002/anie.201708278
Angewandte Chemie International Edition
disease could be identified. It was discovered that the digestive
glands of the scallops contained PTXs-1-5 along with the polyether
fatty acids dinophysistoxin (DTX)-1 and -3. Since this report, a
series of other pectenotoxin natural products (approximately 20)
have been isolated. These more recently isolated natural products
have structures that are similar to the original PTX compounds, with
variation mainly occurring at the spirocyclic acetal centre together
with differing oxygenation patterns on the macrocyclic polyether
backbone.[2]
Synthesis of the pectenotoxin 4 EFG rings
DOI: 10.1002/anie.200123456
Cobalt versus osmium: control of both trans and cis
selectivity in construction of the EFG rings of
pectenotoxin 4
Given their intriguing structure and biological activity it is not
surprising that the pectenotoxins have attracted the interest of
synthetic organic chemists. Despite the fact that an impressive set of
synthetic methods have been focused on preparing these natural
products[3] there have only been two successful syntheses of any
pectenotoxins to date, by Evans (PTX-4) [4] and Fujiwara (PTX-2).[5]
We were attracted to the pectenotoxins because of the prevalence
of multiply substituted THF rings bearing a range of stereochemical
arrangements (compare rings C, E and F). These targets provide a
challenging test for methodology we have developed which uses
catalytic osmium to construct THF rings with a high degree of
stereochemical control.[6] Indeed, our previous work in this area has
described a short route to the C1-C16 fragment of PTX-4 using an
osmium catalysed double oxidative cyclisation in concert with a
hydride initiated spiroketalisation sequence (Scheme 1a).[7]
Ahria Roushanbakhti, Yifan Liu, Paul C. M. Winship,
Michael J. Tucker, Wasim M. Akhtar,¥ Daryl S. Walter,≠ Gail
Wrigleyɸ and Timothy J. Donohoe*
Abstract: Catalytic oxidative cyclisation reactions have been
employed for the synthesis of the E and F rings of the
complex natural product target pectenotoxin 4. The choice of
metal catalyst (cobalt or osmium based) allowed for the
formation of THF rings with either trans or cis
stereoselectivity. Fragment union using a modified Julia
reaction then enabled the synthesis of an advanced synthetic
intermediate containing the EF and G rings of the target.
With the ABC fragment completed, we now wished to synthesise
the EFG portion of this natural product (Fragment B, Scheme 1b).
We envisaged that B could be broken into two smaller fragments
(the E ring and the FG unit) which could be coupled by a key Julia
reaction (Scheme 1b), followed by formation of the hemi-acetal
containing G ring as the final step.
Accessing both the E and F THF rings would require two
different oxidative cyclisation approaches. Of particular interest was
the possibility of using either cobalt (2,5-trans, F ring)[8] or osmium
(2,5-cis, E-ring)[6] catalysis to install the required stereocentres in a
divergent manner. The two different stereochemical outcomes have
their origins in the way that the alcohol substrate binds to the metal
prior to cyclisation. Mono-dentate binding, as exhibited by cobalt,
leads to trans stereochemistry (see C→D), while bidentate binding
(of a 1,2-diol) to osmium enforces cis stereochemistry in the THF
product (see E→F). We sought to utilise both of these catalytic
cyclisations to our advantage in the synthesis of PTX-4.
The pectenotoxin (PTX) family of natural products are a series of
macrocyclic polyethers which boast intriguing structural complexity
and potent biological (eg anti-tumour) activity.[1] The first members
of the family, PTXs-1-5, were isolated by Yasumoto and co-workers
off the coast of Japan, in 1985.[1] Local folklore had described the
occurrence of gastroenteritis following the ingestion of shellfish
harvested in late spring to summer. Therefore, Yasumoto and co-
workers
examined
native scallops
in the hope
that the toxins
causing
this
[]
Prof. T. J. Donohoe, Mr A. Roushanbahkti, Dr Y. Lui, Dr P. C.
M. Winship, Dr M. J. Tucker, Mr W. M. Akhtar
Department of Chemistry,
University of Oxford,
Chemistry Research Laboratory,
Mansfield Road,
Oxford,
OX1 3TA (UK)
Fax: (+44) 1865 275708
E-mail: (timothy.donohoe@chem.ox.ac.uk)
ɸ
[ ] AstraZeneca,
IMED Oncology
Darwin Building, Unit 310,
Cambridge Science Park,
Cambridge
CB4 0WG (UK)
[≠]
GlaxoSmithKline Medicines Research Centre,
Gunnels Wood Road,
Stevenage,
SG1 2NY (UK)
[] We thank the EPSRC, AstraZeneca, GlaxoSmithKline and the
China Scholarship Council for supporting this project and Dr K.
Luscombe for carrying out preliminary experiments. Professor
D. A. Evans is also thanked for his assistance.
Scheme 1. Cis and trans THF cyclisation modes using Os and Co
catalysis
[¥]
Author to whom correspondence regarding the X-ray structure
should be addressed.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.201708278
Angewandte Chemie International Edition
Scheme 2 shows the key sulfone sub-target G required for the
Julia coupling; the stereochemistry at C37 could be installed by a
Sharpless AD reaction on trans-alkene H, itself derived by cross-
metathesis of the cobalt cyclisation product I.
group at C-36 must be oxidised selectively later in the synthesis.
Pleasingly, reaction of 13 with di-tert-butylsilyl ditriflate
(DTBSOTf) afforded the cyclic bis-protected compound 14 in 65%
yield. Presumably compound 14 is formed after initial reaction of
the primary alcohol with the silylditriflate, followed by selective
ring closure to form a seven, rather than eight, membered ring.
Finally, the C-36 hydroxyl was protected with a labile TMS group in
91% yield to form product 15 as a crystalline solid whose structure
was confirmed by Single Crystal X-ray diffraction studies.[13]
Pleasingly, this structure proves beyond doubt the trans selectivity
of the Co-catalysed cyclisation, the sense of diastereoselectivity
during dihydroxylation and the regioselectivity of triol 13 protection
using the silylditriflate reagent.
Our synthesis of the pectenotoxin F ring fragment I began
with commercially available meso diol 1, which was desymmetrised
using Pancreatin catalysis to give a mono-acetate 2 of known
absolute configuration (Scheme 2).[9] Protection of the remaining
hydroxyl group as a PMB ether was accomplished under acidic
conditions to furnish compound 3. This material was determined to
have ≥95:5 enantiomeric ratio by a sequence that involved acetate
deprotection and Mosher’s ester analysis of the unprotected alcohol.
Subsequent ozonolysis of 3 was followed immediately by a double
methylenation reaction on the unstable aldehyde 4 under Wittig
conditions to produce the cyclisation substrate 5 (after in situ acetate
deprotection). Finally, the key cobalt catalysed trans THF forming
reaction transformed 5 into 6 in acceptable yield and as a single
(trans) diastereoisomer (via an intermediate such as J), as proven by
nOe experiments.
Scheme 2. Formation of the trans THF F-ring of PTX-4 by cobalt
catalysed cyclisation.
In order to complete construction of the FG ring scaffold, the
corresponding cross metathesis coupling partner 10 was prepared
from (R)-methylsuccinic acid 7 using the four step sequence shown
in Scheme 3a.[10] The pivotal step revolved around regioselective
protection of the least hindered alcohol group of diol 8 with
TBDPSCl, as reported by Lautens. Note that the mixture of
regioisomers formed during this step could not be separated until the
formation of alkene 10.
Scheme 3. Cross metathesis and asymmetric dihydroxylation en-
route to the pectenotoxin FG ring system: MBT=
mercaptobenzothiazole.
Next, we embarked on a synthesis of the pectenotoxin E-ring
utilising the osmium catalysed oxidative cyclisation to set the
desired stereochemistry between the C-22 hydroxymethyl and C-25
methyl groups. Our route began with a regioselective 1,2-protection
of the triol derived from reduction of (R)-malic acid (16→17,
Scheme 4). Conversion of the remaining primary alcohol into the
iodide was followed by sequential SN2 displacement by the anion of
TMS-acetylene, and silyl deprotection, to furnish compound 18. A
nickel catalysed hydroalumination (followed by an iodination
quench) as developed by Hoveyda then allowed for the selective
formation of regioisomer 19,[14] which was coupled to a zinc reagent
derived from serine under Negishi conditions to form 1,1-
disubstituted alkene 20.[15] The osmium catalysed THF forming
reaction was then attempted under conditions that were acidic
enough to deprotect the acetal and so form the chelating 1,2-diol
Experimentation showed that it was desirable to transform
compound 6 into the (Julia) sulfone coupling partner 11 prior to
cross metathesis; this was achieved in
a
two-step
displacement/oxidation process, Scheme 3b. The cross metathesis
coupling between 10 (4 equivalents) and 11 then proceeded with
Hoveyda Grubbs (II) catalyst in 83% yield,[11] followed by TBDPS
deprotection with HCl/MeOH, to give the E-alkene 12 (92:8 dr, J=
15 Hz). Diastereoselective dihydroxylation of 12 was accomplished
with good selectivity (95:5 dr) using the (DHQD)2PHAL ligand for
a Sharpless asymmetric dihydroxylation reaction.[12] Interestingly,
selectivity of 95:5 dr for the alternative syn diol diastereoisomer
could be achieved by using a (DHQ)2PHAL ligand for the
dihydroxylation (structure not shown). Regioselective protection of
triol 13 was essential for further manipulation because the hydroxyl
2
This article is protected by copyright. All rights reserved.

10.1002/anie.201708278
Angewandte Chemie International Edition
motif in situ, this forming the basis for the cyclisation cis-
stereochemistry. Pyridine N-oxide (PNO) is a particularly effective
reoxidant for this transformation because it allows osmium to be
shuttled between the (IV) and (VI) oxidation states without forming
unwanted Os(VIII).[16] Transition structure K provides a plausible
explanation for the observed stereochemical outcome of the
reaction. Note that the initially formed diol L lactonised in situ so
that the product was 21, which was isolated as TBS ether 22 in 69%
yield over the two steps. The formation of lactone 21 was a planned
extension of the oxidative cyclisation that enabled us to form the
THF ring and allow selective protection of the hydroxyl groups
within L, all in one pot.
We were now able to examine the key C-C bond forming reaction
to join the two fragments; based on precedent from Evans, we chose
a modified Julia coupling to accomplish this step.
A complication with this coupling was the fact that the E-ring
was prone to reversible ring opening via elimination of the anion
derived from sulfone 15; this led to formation of the coupled alkene
product containing an unwanted cis-THF ring. This problem had
originally been identified by Evans and, by using conditions which
involved a lithium anion and low temperatures, Evans’ protocol
allowed the formation of 28 as a 92:8 mixture of trans / cis THF
diastereoisomers, coupled with selective formation of the E-alkene
(J= 16 Hz).[4]
Opening of the lactone 22 was accomplished by reaction with N-
methyl-O-methyl hydroxylamine and trimethyl aluminium to form
23, with the Weinreb amide functionality installed for easy access to
the corresponding aldehyde later in the synthesis. Reduction of the
C-25 hydroxymethyl group was accomplished in two steps: firstly
formation of the iodide 24 with iodine and triphenyl phosphine
followed by hydrogenolysis using H2 gas and a palladium catalyst
(25). Thus, the C-25 hydroxymethyl group derived from the
oxidative cyclisation was transformed into the C-25 methyl group
with the correct relative stereochemistry. Finally, the Weinreb amide
was reduced to an aldehyde and olefinated with a methyl substituted
stabilised ylid to form (E)-alkene 26. The ester was reduced and
then transformed into aldehyde 27 in preparation for Julia coupling
with the previously prepared sulfone 15. Note that nOe
measurements on the ester 26 confirmed both the alkene and THF
ring stereochemistry as shown.
The synthesis of the C21-40 fragment 30 was then completed by
selective TMS deprotection at C36, followed by oxidation to the
ketone 29 using DMP. Finally, TASF promoted deprotection of all
silyl protecting groups liberated a triol that spontaneously formed
the desired cyclic hemi-acetal EFG system 30 in situ.[17]
Scheme 5. Completion of the synthesis of the pectenotoxin-4
EFG ring system
In conclusion, we have demonstrated the synthetic utility of both
the osmium and cobalt catalysed oxidative cyclisation reactions,
controlling the relative stereochemistry of the E- and F-rings present
in PTX-4. Both catalytic cyclisations performed well, and afforded
very high levels of diastereoselectivity for the desired products. The
two fragments were then combined using a Julia coupling, which
allowed for the synthesis of the C21-40 EFG fragment of the target
PTX-4.
Keywords oxidation · catalysis · pectenotoxin · osmium · cobalt
[1] (a) T. Yasumoto, M. Murata, Y. Oshima, M. Sano, G. K. Matsumoto,
J. Clardy, Tetrahedron 1985, 41, 1019–1025; (b) K. Sasaki, J. L. C.
Wright, T. Yasumoto, J. Org. Chem. 1998, 63, 2475–2480.
[2] (a) H.-D. Chae; T.-S. Choi, B.-M. Kim, J. H. Hung, Y.-J. Bang, D. Y.
Shin, Oncogene 2005, 24, 4813–4819; (b) D. Y. Shin, G. Y. Kim, N.
D. Kim, J. H. Jung, S. K. Kim, H. S. Hang, Y. H. Choi, Oncol. Rep.
2008, 19, 517–526.
[3] Selected references on approaches to the pectenotoxins: (a) M. Heapy,
T. W. Wagner, M. M. Brimble, Synlett 2007, 2359–2362; (b) R.
Halim, M. A. Brimble, J. Merten, Org. Lett. 2005, 7, 2659–2662; (c)
R. Halim, M. A. Brimble, J. Merten, Org. Biomol. Chem. 2006, 4,
Scheme 4. Osmium catalysed oxidative cyclisation
3
This article is protected by copyright. All rights reserved.

10.1002/anie.201708278
Angewandte Chemie International Edition
1387–1399; (d) S. Carley, M. A. Brimble, Org. Lett. 2009, 11, 563–
566; (e) D. Bondar, J. Liu, T. Müller, L. A. Paquette, Org. Lett. 2005,
7, 1813–1816; (f) P. D. O’Connor, C. K. Knight, D. Friedrich, X.
Peng, L. A. Paquette, J. Org. Chem. 2007, 72, 1747–1754; (g) S. D.
Lotesta, Y. Hou, L. J. Williams, Org. Lett. 2007, 9, 869–872; (h) S.
Joyasawal, S. D. Lotesta, N. G. Akhmedov, L. J. Williams, Org. Lett.
2010, 12, 988–991; (i) D. Vellucci, S. D. Rychnovsky, Org. Lett.
2007, 9, 711–714; (j) J. A. Aho, E. Salomäki, K. Rissanen, P. M.
Pihko, Org. Lett. 2008, 10, 4183–4186; (k) H. Helmboldt, J. A. Aho,
P. M. Pihko, Org. Lett. 2008, 10, 4179–4182; (l) J. A. Aho, A. Piisola,
K. S. Krishnan, P. M. Pihko, Eur. J. Org. Chem. 2011, 12, 1682–
1694; for a review see (m) R. Halim, M. A. Brimble, Org. Biomol.
Chem. 2006, 4, 4048–4058; (n) A. Kouridaki, M. Sofiadis, T.
Montagnon, Eur. J. Org. Chem. 2015, 7240–7243.
Information (CIF); Crystallographic data have been deposited with the
Cambridge Crystallographic Data Centre (CCDC 1551234) and can be
obtained via www. ccdc.cam.ac.uk/data_request/cif.
[14] F. Gao, A. H. Hoveyda, J. Am. Chem. Soc. 2010, 132, 10961-10963.
[15] See, E. Negishi, Q. Hu, Z. Huang, M. Qian, G. Wang, Aldrichimica
Acta, 2005, 38, 71-87.
[16] T. J. Donohoe, K. M. P. Wheelhouse, P. J. Lindsay-Scott, P. A.
Glossop, I. A. Nash, J. S. Parker, Angew. Chem. 2008, 120, 2914-
2917; Angew. Chem. Int. Ed. 2008, 47, 2872-2875.
[17] We have assumed that the stereochemistry at C36 (hemiacetal) is
formed as the thermodynamically more stable stereoisomer with an
axial C-36 hydroxyl group; this assumption is based on the behaviour
of related systems as reported in the two previous synthesis of the
pectenotoxins.[4,5]
[4] (a) D. A. Evans, H. A. Rajapakse, D. Stenkamp, Angew. Chem. 2002,
114, 4751-4755; Angew. Chem. Int. Ed. 2002, 41, 4569–4573; (b) D.
A. Evans, H. A. Rajapakse, A. Chiu, D. Stenkamp, Angew. Chem.
2002, 114, 4755-4758; Angew. Chem. Int. Ed. 2002, 41, 4573–4576.
[5] K. Fujiwara, Y. Suzuki, N. Koseki, Y. I. Aki, Y. Kikuchi, S. I.
Murata, F. Yamamoto, M. Kawamura, T. Norikura, H. Matsue, A.
Murai, R. Katoono, H. Kawai, T. Suzuki, Angew. Chem. 2014, 126,
799-803; Angew. Chem. Int. Ed. 2014, 53, 780–784.
[6] Representative examples: (a) M. de Champdore, M. Lasalvia, V.
Piccialli, Tetrahedron Lett. 1998, 39, 9781; (b) T. J. Donohoe, S.
Butterworth, Angew. Chem. 2003, 115, 978; Angew. Chem. Int. Ed.
Engl. 2003, 42, 948; (c) D. M. Walba, G. S. Stoudt, Tetrahedron Lett.
1982, 23, 727-730; (d) T. J. Donohoe, S. Butterworth, Angew. Chem.
2005, 117, 4844-4846; Angew. Chem. Int. Ed. 2005, 44, 4766-4768;
(f) T. J. Donohoe, P. C. M. Winship, D. S. Walter, J. Org. Chem.
2009, 74, 6394-6397; (g) H. Cheng, C. B. W. Stark, Angew. Chem.
2010, 122, 1632-1635; Angew. Chem. Int. Ed. 2010, 49, 1587–1590;
(h) H. Sugimoto, T. Kanetake, K. Maeda, S. Itoh, Org. Lett. 2016, 18,
1246–1249. For reviews: (i) T. J. Donohoe, B. S. Pilgrim, J. Org.
Chem. 2013, 78, 2149-2167; (j) N. S. Sheikh, Org. Biomol. Chem.
2014, 12, 9492–9504; (k) J. Adrian, L. J. Gross, C. B. W. Stark,
Beilstein J. Org. Chem. 2016, 12, 2104–2123.
[7] T. J. Donohoe, R. M. Lipiński, Angew. Chem. 2013, 125, 2551-2554;
Angew. Chem. Int. Ed. 2013, 52, 2491–2494.
[8] (a) T. Mukaiyama, S. Isayama, S. Inoki, K. Kato, T. Yamada, T.
Takai, Chem. Lett. 1989, 18, 449-452; (b) S. Inoki, K. Kayo, T. Takai,
S. Isayama, T. Yamada, T. Mukaiyama, Chem. Lett. 1989, 18, 515-
518; (c) S. Inoki, T. Mukaiyama, Chem. Lett. 1990, 19, 67-70; (d) K.
Kato, T. Yamada, T. Takai, S. Inoki, S. Isayama, Bull. Chem. Soc.
Jpn. 1990, 63, 179-186; (e) B. Perez, D. Schuch, J. Hartung, Org.
Biomol. Chem. 2008, 6, 3532-3541; (f) C. Palmer, N. Morra, A.
Stevens, B. Bajtos, B. Machin, B. Pagenkopf, Org. Lett. 2009, 11,
5614-5617; (g) G. A. Phillips, C. Palmer, A. C. Stevens, M. L.
Piotrowski, D. S. R. Dekruyf, Tetrahedron Lett. 2015, 56, 6052–6055.
[9] F. Theif, H. Schick, G. Winterb, G. Reck, Tetrahedron 1991, 47,
7569-7582.
[10] M. Lautens, T. A. Stammers, Synthesis 2002, 1993.
[11] See: A. H. Hoveyda, D. G. Gillingham, J. J. Van Veldhuizen, O.
Kataoka, S. B. Garber, J. S. Kingsbury, J. P. A. Harrity, Org. Biomol.
Chem. 2004, 2, 8-23.
[12] See: H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, Chem. Rev.
1994, 94, 2483-2547.
[13] Low temperature single crystal X-ray diffraction data were collected
using a (Rigaku) Oxford Diffraction SuperNova A diffractometer and
the structure was solved by charge-flipping using 'Superflip’ [L.
Palatinus, G. Chapuis, J. Appl. Cryst. 2007, 40, 786]. The structure
was refined by full-matrix least squares on F2 using CRYSTALS suite
[P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout, D. J.
Watkin, J. Appl. Cryst. 2003, 36, 1487; Parois, P, R. I. Cooper, A. L.
Thompson, Chemistry Central Journal 2015, 9:30; R. I. Cooper, A. L.
Thompson, D. J. Watkin, J. Appl. Cryst. 2010, 43, 1100-1107]. The
Flack x parameter [H. D. Flack, Acta Cryst. 1983, A39, 876-881; H.
D. Flack, G. Bernardinelli, J. Appl. Cryst. 2000, 33, 1143-1148]
refined to 0.00(2). Bayesian analysis of the Bijvoet pairs [R. W. W.
Hooft, L. H. Straver, A. L. Spek, J. Appl. Cryst. 2008, 41, 96-103]
gave the Hooft y parameter as 0.022(14), and the probability that the
structure was the correct hand of >99.99% given that the crystal is
enantiopure. Full refinement details are given in the Supporting
4
This article is protected by copyright. All rights reserved.

10.1002/anie.201708278
Angewandte Chemie International Edition
Synthesis of the pectenotoxin 4 EFG rings
Two types of oxidative cyclisation reaction, utilising either
osmium or cobalt catalysis, provide complete control of the
relative stereochemistry of THF rings embedded in the complex
pectenotoxin-4 ring system. In this manner, rapid access to either
trans (Co) or cis (Os) 2,5-disubstituted THF rings was facilitated.
A significant portion of the natural product was then put together
using cross metathesis and Julia coupling as the key C-C bond
forming protocols.
Cobalt versus osmium: control of both trans and cis
selectivity in construction of the EFG rings of
pectenotoxin 4
Ahria Roushanbakhti, Yifan Liu, Paul C. M. Winship,
Michael J. Tucker, Wasim M. Akhtar, Daryl S. Walter,
Gail Wrigley and Timothy J. Donohoe
5
This article is protected by copyright. All rights reserved.