Designer discodermolide segments via ozonolysis of vinyl phosphonates

Article abstract of DOI:10.1021/ol802503h

(Chemical Equation Presented) To apply our collection of enantiopure 7-ring vinyl sulfones to probe the anticancer SAR of a series of computer-designed (+)-discodermolide analogs, the ozonolytic reactivity of transposed cyclic vinyl phosphonates was explo

Full text of DOI:10.1021/ol802503h

ORGANIC
LETTERS
2009
Vol. 11, No. 3
543-546
Designer Discodermolide Segments via
Ozonolysis of Vinyl Phosphonates1
Xavier Mollat du Jourdin, Mohammad Noshi, and P. L. Fuchs*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907
pfuchs@purdue.edu
Received October 29, 2008
ABSTRACT
To apply our collection of enantiopure 7-ring vinyl sulfones to probe the anticancer SAR of a series of computer-designed (+)-discodermolide
analogs, the ozonolytic reactivity of transposed cyclic vinyl phosphonates was explored. Successful preparation of the desired aldehyde-
esters and lactones from vinyl phosphonates via an oxidative cleavage-phosphite/methanol exchange sequence is described.
Our laboratory’s recent work on exploiting the intermediacy
of vinyl sulfones for the synthesis of natural products has
demonstrated the utility of five-, six- and seven-membered
vinyl sulfones to elaborate contiguous chiral centers.2 This
strategy was expanded to provide acyclic arrays especially
including complete collections of dipropionate units via
oxidative cleavage of vinyl sulfones with OsO4, RuO4, and
O3.2,3 Trapping the intermediate acyl sufone-aldehydes
provided termini-differentiated acyclic products ready for
sequential coupling.
However, although ozonolysis of electron-rich olefins is widely
used and fully documented, ozonolytic cleavage of electron-
deficient alkenes is quite rare because of their diminished
reactivity; electron-rich olefins are 10,000 times more reactive
to ozone than are electron-deficient olefins.4 Griesbaum and
Fuchs reported the first ozonolysis of vinyl nitriles,5 vinyl
(3) El-Awa, A.; Mollat du Jourdin, X.; Fuchs, P. L. J. Am. Chem. Soc.
2007, 129, 9086–9093.
(4) Pryor, W. A.; Giamalva, D.; Church, D. F. J. Am. Chem. Soc. 1983,
105, 6858–6861.
(5) Griesbaum, K.; Huh, T. S.; Gutsche, S. H. Tetrahedron Lett. 1990,
31, 3299–3300.
Among the favored techniques for oxidative cleavage, ozone
presents indubitable advantages including low cost, limited
production of toxic byproduct, and simple reaction protocols.
(6) (a) El-Awa, A; Fuchs, P. L. Org. Lett. 2006, 8, 2905–2908. (b) Kim,
Y; Fuchs, P. L. Org. Lett. 2007, 9, 2445–2448.
(7) (a) Gao, J.; Martichonok, V.; Whitesides, G. M. J. Org. Chem. 1996,
61, 9538–9540. (b) Chan, T-H.; Xin, Y-C. Chem. Commun. 1996, 905–
906.
(1) Synthesis via vinyl sulfones 96. Chiral Carbon Collection 18.
(2) El-Awa, A.; Noshi, M. N.; Mollat du Jourdin, X.; Fuchs, P. L. Chem.
ReV., submitted for publication.
(8) Noshi, M. N.; El-Awa, A.; Torres, E.; Fuchs, P. L. J. Am. Chem.
Soc. 2007, 129, 11242–11247.
10.1021/ol802503h CCC: $40.75
Published on Web 12/31/2008
2009 American Chemical Society

triflates, and vinyl sulfones, respectively, in the 1990s.2,3 More
recently, ozonolysis of seven-membered vinyl sulfones was
more extensively studied for the synthesis of polypropionate
fragments of apoptolidine and aplyronine A.6 Ozonolysis of
acyclic vinyl phosphonates was first reported in the late 1990s
by Whitesides and Chan7 and was then successfully applied
to cyclic systems by Noshi and Fuchs for the synthesis of
the C15-C20 fragment of aplyronine A.8
rameter at play as vinyl sulfone 5 is unexpectedly unreactive
to ozone even at -30 °C and at high concentration (0.5 M).
To circumvent this lack of reactivity, vinyl sulfones 3-5
were transposed to vinyl phosphonates 3b-5b using Noshi’s
method (Scheme 1).8 This strategy, developed with relatively
In the course of these studies, it was observed that
conformation and substitution pattern greatly affected the
reactivity of cyclic vinyl sulfones. Most importantly, it
appeared that the ꢀ′-substituent dominated the reactivity of
the olefin. As the size of the substituent increased, the
reactivity of the olefin decreased (Table 1). Vinyl sulfones
Scheme 1
.
Vinyl Sulfone to Vinyl Phosphonate Transposition of
Stereotetrads
Table 1. Ozonolysis of Vinyl Sulfones, ꢀ′ Substituent Effect
a DEP ) diethyl phosphite. b 3b-Si refers to the fully TBS-protected
intermediate 3b.
nonhindered substrates, satisfactory extends to sterically more
demanding stereotetrads and does not suffer from extensive
ꢀ-elimination.
TMS protection followed by sulfone to phosphonate
transposition and selective TMS deprotection gave 69-79%
yield of the desired vinyl phosphonates 3b and 4b over 3
steps. TBS protection and transposition with vinyl sulfone
3 and 5 provided good to moderate yields of the desired vinyl
phosphonates 3b-Si and 5b.
a See ref 9. b No oxidative cleavage is observed. c The reaction was run
at 0.5 M in CH2Cl2/MeOH (2:1) with NaHCO3 (3 equiv) at -78 to -30
°C for 1 h.
Early ozonolytic reactions with simple model cyclic vinyl
phosphonates 7 and 8 under acidic conditions and vinyl
phosphonate 98 using methanol and NaHCO3 for 15 min at
-78 °C afforded the desired acetal-methyl esters 7a and 8a
and aldehyde-methyl ester 9a (Table 2), respectively. At this
juncture, it was assumed that the reaction proceeded via the
intermediacy of a simple R-keto phosphonate, as such species
are known to be good acylating agents.10
1 and 2 were ozonized in 15 min at -78 °C to provide good
to excellent yields of the desired lactols as anomeric
mixtures.9 Increasing the size of the ꢀ′-substituent was
detrimental to the rate of the reaction as vinyl sulfone 3
required 45-60 min at -30 °C to afford moderate to good
yields of lactols. Substrates bearing an isopropyl group at
the ꢀ′-position escaped oxidative cleavage of vinyl sulfone
4, with oxidation of the alcohol moiety to ketone 4a being
the preferred pathway at -78 °C, to the exclusion of
subsequent ozonolysis even with increase in time and/or
temperature. However, ꢀ′-substitution is not the only pa-
Although oxidative cleavage of substrates 5b and 11b
under Noshi’s conditions (Table 3, entries 2 and 4) was
successfully monitored by TLC, only traces of the desired
aldehyde-methyl esters were observed after extended reaction
times (10 h), suggesting a slow phosphite-methanol ex-
change. Although disappointing, these results shed light on
(9) Torres, E.; Chen, Y.; Kim, I. C.; Fuchs, P. L. Angew. Chem., Int.
Ed. 2003, 42, 3124–3131.
544
Org. Lett., Vol. 11, No. 3, 2009

Table 2. Ozonolysis of Model Vinyl Phosphonatesa
Table 3. Ozonolysis of Cyclic Vinyl Phosphonates
a Typical procedure: (i) O3, NaHCO3 (5 equiv), CH2Cl2/MeOH (4:1),
-78 °C, 15 min; (ii) PPh3 (1.1 equiv) or Me2S (10 equiv), -78 °C to rt,
2-3 h; (iii) p-TSOH was added to obtain acetals 7a and 8a.
the difference of reactivity of acyl phosphonate-aldehyde E
versus its acyl sulfone counterpart (Scheme 2). Extensive
Scheme 2. Hypothesis for Ozonolysis of Vinyl Phosphonates
a (i) O3, CH2Cl2/MeOH, pyridine; (ii) cat. DMAP; (iii) Bh3, t-BuNH2.
b (i) O3, DCM/MeOH, NaHCO3; (ii) Me2S. c (i) O3, AcOEt, NaHCO3; (ii)
MeOH; (iii) Me2S. d (i) O3, AcOEt/MeOH; (ii) Me2S; (iii) cat. DMAP. e (i)
O3, AcOEt, NaHCO3; (ii) Me2S, CH2Cl2/EA (9:1). f Acyl phosphonate could
not be purified, yield was estimated by 1H NMR. g 3c and 4c were converted
to 3d and 4d, respectively, during purification and could not be isolated.
h Estimated yield by 1H NMR for mixture 3c and 3d (6:1). i Estimated yield
optimization performed on substrates 10, 5b, 11b, 3b, and
3b-Si demonstrated that a rapid phosphite-methanol ex-
change could be achieved in two ways. First, ozonolysis
could be performed in a mixture of MeOH/CH2Cl2/pyri-
dine11,12 or MeOH/AcOEt to provide, respectively, acyl
phosphonate-aldehyde E directly or R-hydroxymethyl hy-
droperoxides D, which could be quenched with excess Me2S
1
by H NMR for mixture 4c and 4d (6.2:1).
to provide E in a sequential fashion (Scheme 2). Acyl
phosphonate-aldehyde E was then treated under basic condi-
tions. After screening bases (pyridine, 2,6-lutidine, DABCO,
DMAP, NEt3, NaHCO3), it was revealed that catalytic DMAP
(20-50 mol %) was sufficiently basic and mild enough to
(10) Serine, M.; Kume, A.; Hata, T. Tetrahedron Lett. 1981, 22, 3617–
3620
.
(11) Schwartz, C.; Raible, J.; Mott, K.; Dussault, P. H. Org. Lett. 2006,
8, 3199–3201
.
(12) Mohammad Noshi, PhD thesis, Purdue University.
Org. Lett., Vol. 11, No. 3, 2009
545

drive the reaction to completion and avoid detrimental
ꢀ-eliminations/epimerizations (Table 3, entries 1 and 6).12
Whereas the basicity of the reagent proved crucial for
activity, its nucleophilicity is clearly also an important factor.
A second successful protocol required addition of MeOH
after ozonolysis in AcOEt in the presence of NaHCO313 but
prior to quenching with Me2S (Table 3, entries 3, 5, and 8).
These results and the presence of free diethyl phosphite in
the reaction mixture14 prior to quenching suggest that the
phosphite was expelled from ozonide A to provide ozonide
C after generation of oxonium intermediate B suggested by
Griesbaum.15 This equilibrium exchange is driven to C due
to the large excess of MeOH (Scheme 2).
Initial attempts to drive lactonization to completion were
unsuccessful due to the high base sensitivity of 3c and 4c.
Problems included ꢀ-elimination, stereochemical erosion,
and/or phosphite addition to the aldehyde. Fortunately,
catalytic DBU (20 mol %) in CH2Cl2 could smoothly convert
4c to 4d without phosphite addition on the aldehyde (Scheme
4). Aldehyde 3c was less tolerant to DBU and required
Scheme 4.
Lactonizationa
Oxidative cleavage of 3b and 4b provided aldehydes 3c
and 4c16 (Scheme 3) along with the desired lactones 3d and
a Method A: NaHMDS (1 equiv), p-NO2C6H4CHO (4 equiv), THF, -10
°C. Method B: DBU (20 mol %), CH2Cl2, Room Temperature.
Scheme 3
.
Ozonolysis of Vinyl Phosphonates Bearing a Free
-OHa
dropwise addition of NaHMDS in the presence of p-
nitrobenzaldehyde to provide 3d while trapping the expelled
diethyl phosphite.19,20
Ozonolysis of vinyl sulfones and vinyl phosphonates to
acyl sulfones and acyl phosphonates reveals substantial
reactivity differences as acylating agents.
Acknowledgment. We thank Arlene Rothwell, Dr. Karl
Wood and Dr. Phillip Fanwick from Purdue University for
the MS and X-ray data and Purdue University for support
of this research.
a Reactions were run in AcOEt/CH2Cl2 and quenched with Me2S.
Supporting Information Available: Procedures, spec-
troscopic data, spectra, and CIF file. This material is available
free of charge via the Internet at http://pubs.acs.org.
4d as minor products in 6:1 ratio in AcOEt or CH2Cl2 (Table
3, entries 7 and 9). This result is in accord with the higher
reactivity of acyl phosphonates at the C center than the P
center10,17,18 and with Whitesides’ and Chan’s results.7
OL802503H
(16) The structure of R-hydroxyphosphonates 3c and 4c was confirmed
by 31P and 13C NMR spectroscopy.
(17) (a) Berlin, K. D.; Roy, N. K.; Claunch, R. T. J. Am. Chem. Soc.
1968, 90, 4494–4495. (b) Kim, D. Y.; Wiemer, D. F. Tetrahedron Lett.
2003, 44, 2803–2805. (c) Meier, C.; Laux, W. H. G Tetrahedron: Asymmetry
1995, 6, 1089–1092. (d) Demir, A. S.; Reis, O.; Kayalar, M.; Eymur, S.;
Reis, B. Synlett 2006, 3329–3333.
(13) Evans, D. A.; Johnson, J. S.; Olhava, E. J. J. Am. Chem. Soc. 2000,
122, 1635–1649.
(14) Prior to quenching, the MeOH/HP(O)(OEt)2 exchange takes place
at 25 °C. HP(O)(OEt)2 is usually visible by TLC with p-anisladehyde stain
as a polar white spot.
(18) Afarinkia, K.; Twist, A. J.; Yu, H. W. J. Organomet. Chem. 2005,
690, 2688–2691.
(15) (a) Griesbaum, K.; Volpp, W.; Huh, T. S. Tetrahedron Lett. 1989,
30, 1511–1512. (b) Griesbaum, K.; Schlindwein, K. J. Org. Chem. 1995,
60, 8062–8066.
(19) p-Nitrobenzaldehyde is used to trap diethylphosphite.
(20) Sardarian, A. R.; Kaboudin, B. Synth. Commun. 1997, 27, 543–
551.
546
Org. Lett., Vol. 11, No. 3, 2009