A synthetic approach to the phorboxazoles – Synthesis of the C20–C32 central core

Article abstract of DOI:10.1016/j.tetlet.2017.07.009

An enantiospecific synthesis of the C20–C32 central core of the phorboxazole scaffold, including the non-macrocyclic oxazole is detailed in 17 steps (longest linear sequence) from methacrolein in 7.8% overall yield. All of the stereocenters are communicated from a single Evans aldol reaction, and the final compound is suitably functionalized for further elaboration to the natural products.

Full text of DOI:10.1016/j.tetlet.2017.07.009

Accepted Manuscript
A Synthetic Approach to the Phorboxazoles – Synthesis of the C20-C32 Central
Core
James W. Leahy, Stephen J. Boyer
PII:
S0040-4039(17)30844-4
DOI:
http://dx.doi.org/10.1016/j.tetlet.2017.07.009
TETL 49091
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
31 May 2017
26 June 2017
1 July 2017
Please cite this article as: Leahy, J.W., Boyer, S.J., A Synthetic Approach to the Phorboxazoles – Synthesis of the
C20-C32 Central Core, Tetrahedron Letters (2017), doi: http://dx.doi.org/10.1016/j.tetlet.2017.07.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A Synthetic Approach to the Phorboxazoles – Synthesis of the C20-C32 Central
Core
James W. Leahya* and Stephen J. Boyerb
aDepartment of Chemistry, University of South Florida, CHE 205, 4202 E. Fowler Avenue,
Tampa, FL 33620; Florida Center of Excellence for Drug Discovery and Innovation, University
of South Florida, 3720 Spectrum Boulevard, Suite 303, Tampa, FL 33612 and Department of
Molecular Medicine, Morsani College of Medicine, University of South Florida, MDC 7, 12901
Bruce B. Downs Boulevard, Tampa, FL 33612. bDepartment of Chemistry, University of
California, Berkeley, CA 94720
Abstract
An enantiospecific synthesis of the C20-C32 central core of the phorboxazole scaffold, including
the non-macrocyclic oxazole is detailed in 17 steps (longest linear sequence) from methacrolein
in 7.8% overall yield. All of the stereocenters are communicated from a single Evans aldol
reaction, and the final compound is suitably functionalized for further elaboration to the natural
products.
Keywords
Phorboxazole; enantiospecific synthesis; oxazole; pentasubstituted pyran

The phorboxazoles (1 and 2) and hemiphorboxazole (3) are a family of marine macrolides
that possess potent cytostatic and antifungal activity.1, 2 This activity has been ascribed to their
ability to regulate cyclin-dependent kinase 4,3 one of the presumptive targets of Pfizer’s breast
cancer drug palbociclib.4 Not surprisingly, the novel architecture and considerable biological
activity of these macrolides has attracted significant synthetic interest.5, 6 Ultimately, that
resulted in several completed total syntheses as well as provided synthetic access to a number of
novel synthetic analogs and chemical probes. We report here our synthetic approach to the
central C20-C32 core of the natural product, including the exocyclic oxazole and the
pentasubstituted pyran.
Our retrosynthetic approach to the phorboxazoles was to introduce the exocyclic pyran as a
dithiane nucleophile (4)7 to the macrolide appropriately adorned with a bromomethyloxazole.
This macrolide would emanate from a Julia olefination at C19/C20 and a C1 esterification, thus

resulting in the two retrons 5 and 6. We wish to report here our approach to the synthesis of 5,
which resulted in the synthesis of a functional equivalent (31, vide infra).
While we were confident that our prior synthetic experience on the synthesis oxazole-
containing natural products was going to provide a significant advantage as we initiated our
approach to 5,8, 9 we realized that the methodology developed by Panek for the synthesis of the
ulapualides was superior and more practical.10 Hantzsch cyclization with ethyl bromopyruvate
using the readily available and inexpensive 7 gave the crystalline heterocycle 8, which was
oxidatively cleaved and reduced to provide 2-hydroxymethylated adduct 9. Following protection
of the alcohol, the ester was smoothly converted to the corresponding aldehyde (10), which was
ready to be coupled to the pyran component.
O
1) Br
CO2Et
Ph
NH2
O
Ph
1) OsO4, NMO
CO2Et, NaHCO3
N
O
2) (CF3CO)2O
(79%)
2) NaIO4
3) NaBH4
7
8
(60%)
CO2Et
CHO
N
1) NaH, PMBBr, Bu4NI
2) DIBAL
HO
PMBO
N
O
O
3) (ClCO)2, DMSO, NEt3
(75%)
9
10
The polypropionate composition of the pyran lent itself to consideration of an Evans aldol
approach.11 Toward this end, aldol addition to methacrolein provided the desired syn product
with excellent stereocontrol.12 Following silylation of the alcohol and reductive removal of the
Evans auxiliary, primary alcohol 12 was tritylated to provide the corresponding olefin.
Hydroboration with 9-BBN proceeded with complete stereocontrol, yielding the desired
stereotriad 13.12 The resulting alcohol was oxidized to the aldehyde, and a terminal olefin was
installed under Wittig conditions. This was in anticipation of cyclizing the ring using the elegant

carbonylative methodology developed by Semmelhack.13 Removal of the trityl group and
oxidation to the aldehyde allowed for the formation of β-ketophosphonate 16, which smoothly
underwent Horner-Wadsworth-Emmons olefination with 10 to afford the trisubstituted alkene
with good stereocontrol (>8:1 E:Z).14 Secondary alcohol 17 was unmasked with fluorosilicic
acid,15 and the stage was set to use an intramolecular Tishchenko reaction, which would
simultaneously set the desired stereocenter at C26 and install an ester at C24.16 We had done this
with an eye toward using an appropriate aldehyde at C1 to form the ester in the natural product,
and had even hoped to explore this as a novel macrocyclization strategy. To our surprise,
treatment of 17 with p-nitrobenzaldehyde and SmI2 instead resulted in the generation of racemic

18, which was formed via a retro-aldol of 17 followed by an aldol addition to p-
nitrobenzaldehyde and then an intramolecular Tishchenko reduction.17 Further confounding these
efforts was the observation that Evans-Saksena reduction, which typically proceeds with
excellent 1,3-anti control,18 gave us predominantly the unexpected 1,3-syn product 19
(confirmed via examination of 20 using the Rychnovsky acetonide model).19 Finally, attempts to
cyclize 19 under Pd-carbonylative conditions as a model of our approach to 5 (either as the diol
or various analogs with the non-allylic alcohol protected) failed to yield even a trace of the
desired product, forcing us to reconsider our cyclization strategy. Toward this end, we
recognized a need to prepare a more versatile advanced intermediate.
Aldehyde 14 was extended under Horner-Wadsworth-Emmons conditions, which allowed
for the facile generation of 21. This conversion to the primary allylic silyl ether proved critical,
as all attempts to install a β-ketophosphonate in the subsequent steps were incompatible with the
initial ester that was formed from 14. Selective trityl deprotection and Swern oxidation set the

stage for the introduction of the requisite β-ketophosphonate, and installation of the oxazole
proceeded as expected (with 12:1 E:Z selectivity). The primary alcohol could then be liberated
selectively with fluorosilicic acid over 2 hours. The latter reaction also apparently allowed for
the isomerization of the undesired Z-isomer, presumably through protonation of the carbonyl.
We recognized that Michael cyclization of a substrate such as 27 under equilibrating
conditions should greatly favor the desired stereochemistry, so Pinnick oxidation and
esterification of 24 yielded 25.20 Once again, all attempts to reduce the ketone to give 27 with
stereocontrol failed completely. A plausible explanation for this arises from the prospects that
intramolecular delivery of the hydride to give the anti-product 27 would, in this case, require an
unfavorable syn-pentane interaction,21 so the typically disfavored syn product (which forces a
pseudodiaxial interaction between the carbonyl and the γ-carbon) proceeds instead. The main

product formed was 26, which was confirmed using the aforementioned Rychnovsky model.19
Critically, all of our efforts either resulted in concomitant reduction of the ester or formation of
the 26 as the major product.
Our inability to conduct a directed reduction prompted us to consider using the inherent
Felkin-Ahn selectivity of the ketone.22 A number of reducing agents were explored, and
ultimately the bulky lithium tri-tert-butoxyaluminum hydride gave us our best results with > 4:1
selectivity. Liberation of the primary alcohol finally revealed cyclization precursor 29, which we
selectively acetylated with an eye toward closing via a Pd-mediated π-allyl process. Once we
recognized the futility of this strategy, we were delighted to find that the desired pyran 31 could
be formed by simple cyclization with mesyl chloride in the presence of lutidine.
In conclusion, this route provided an enantiospecific synthetic route for the C20-C32 central
core of the phorboxazoles where every stereocenter is derived from a single asymmetric Evans
aldol addition. The longest linear sequence for this portion of the molecule was 17 steps, and
proceeded in 7.8% overall yield. The product obtained in this manner is suitably functionalized
on each end to allow for the further elaboration of this material into phorboxazole, and additional
efforts in this direction will be reported in due course.

Acknowledgements
We gratefully acknowledge the National Science Foundation (NSF Career Award CHE-
9502149) and the Research Corporation (Cottrell Scholar Award) for financial assistance.
References
1
Searle, P. A.; Molinski, T. F.; Brzezinski, L. J.; Leahy, J. W. J. Am. Chem. Soc. 1996,
118, 9422.
2
Smith, A. B., III; Liu, Z.; Hogan, A.-M. L.; Dalisay, D. S.; Molinski, T. F. Org. Lett.
2009, 11, 3766.
3
4
Forsyth, C. J.; Lu, Y.; Chen, J.; La Clair, J. J. J. Am. Chem. Soc. 2006, 128, 3858.
Fry, D. W.; Harvey, P. J.; Keller, P. R.; Elliott, W. L.; Meade, M.; Trachet, E.; Albassam,
M.; Zheng, X.; Leopold, W. R.; Pryer, N. K.; Toogood, P. L. Mol. Cancer Ther. 2004, 3, 1427.
5
Haustedt, L. O.; Hartung, I. V.; Hoffmann, H. M. R. Angew Chem Int Ed Engl 2003, 42,
2711.
6
Shultz, Z.; Leahy, J. W. J. Antibiot. 2016, 69, 220.
7
Leahy, J. W.; Brzezinski, L. J. Tetrahedron Lett. 2016, 57, 4670.
Lafontaine, J. A.; Provencal, D. P.; Gardelli, C.; Leahy, J. W. J. Org. Chem. 2003, 68,
8
4215.
9
Zylstra, E. J.; She, M. W. L.; Salamant, W. A.; Leahy, J. W. Synlett 2007, 623.
Panek, J. S.; Beresis, R. T. J. Org. Chem. 1996, 61, 6496.
10
11
12
13
Evans, D. A. Aldrichimica Acta 1982, 15, 23.
Evans, D. A.; Fitch, D. M. J Org Chem 1997, 62, 454.
Semmelhack, M. F.; Kim, C.; Zhang, N.; Bodurow, C.; Sanner, M.; Dobler, W.; Meier,
M. Pure Appl. Chem. 1990, 62, 2035.
14 Maryanoff, B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863.

15
16
17
Pilcher, A. S.; DeShong, P. J. Org. Chem. 1993, 58, 5130.
Evans, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1990, 112, 6447.
The structure of 18 was deduced via a combination of mass spectrometry, NMR and
polarimetry. Retro-aldol reactions under Tishchenko conditions has been previously observed.
See: Simpura, I. N., Vesa Tetrahedron Lett. 2001, 42, 3905.
18
19
20
21
22
Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988, 110, 3560.
Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron Lett. 1990, 31, 945.
Bal, B. S.; Childers, W. E., Jr.; Pinnick, H. W. Tetrahedron 1981, 37, 2091.
Roush, W. R. J. Org. Chem. 1991, 56, 4151.
Mengel, A.; Reiser, O. Chem. Rev. (Washington, D. C.) 1999, 99, 1191.

A synthetic approach to the phorboxazoles is disclosed
An enantiospecific synthesis of the C20-C32 exocyclic portion is described
All stereocenters derived from a single chiral auxiliary via an Evans aldol reaction

Br
MeO
O
O
PMBO
R1
R2
N
O
OH
MeO
HO
N
O
O
O
OTBS
O
N
O
O
O