Anomalies in the stereoselectivity of the petasis reaction using styrenyl boronic acids

Article abstract of DOI:10.1071/CH10341

The Petasis three-component coupling reaction of N-benzylphenylglycinol, glyoxylic acid, and styrenylboronic acids allows for the efficient synthesis of functionalized homoarylalanine derivatives. The reactions were shown to proceed in high yield but low selectivity, regardless of the nature of the substituent on the styrenylboronic acid component. Anomalies in the stereoselectivity of these reactions compared with previously reported results have been traced to the source of the organoboronic acid. Asymmetric dihydroxylation of the unsaturated amino acid derivatives enables a highly efficient route to dihydroxyhomoarylalanine derivatives. CSIRO 2011.

Full text of DOI:10.1071/CH10341

Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2011, 64, 62–67
www.publish.csiro.au/journals/ajc
Anomalies in the Stereoselectivity of the Petasis Reaction
Using Styrenyl Boronic Acids
A
A
A
Quentin I. Churches, James K. Johnson, Nathan L. Fifer,
and Craig A. Hutton
A B
,
A
School of Chemistry and Bio21 Molecular Science and Biotechnology Institute,
The University of Melbourne, Vic. 3010, Australia.
B
Corresponding author. Email: chutton@unimelb.edu.au
The Petasis three-component coupling reaction of N-benzylphenylglycinol, glyoxylic acid, and styrenylboronic acids
allows for the efficient synthesis of functionalized homoarylalanine derivatives. The reactions were shown to proceed in
high yield but low selectivity, regardless of the nature of the substituent on the styrenylboronic acid component. Anomalies
in the stereoselectivity of these reactions compared with previously reported results have been traced to the source of the
organoboronic acid. Asymmetric dihydroxylation of the unsaturated amino acid derivatives enables a highly efficient
route to dihydroxyhomoarylalanine derivatives.
Manuscript received: 14 September 2010.
Manuscript accepted: 1 December 2010.
Introduction
corresponding phenylacetylenes 6 (Scheme 3) or obtained from
commercial sources. Petasis reactions were conducted accord-
ing to the standard method,[12] with the solvent evaporated and
the crude reaction mixture analyzed by 1H NMR spectroscopy.
The substituted styrenylboronic acids all gave the corresponding
oxazinone products 5 in good yield but poor diastereoselectivity
(dr ꢀ3:1) (Table 1).
The Petasis reaction, or boronic acid Mannich reaction, involves
the three-component coupling of an amine, aldehyde, and
organoboronic acid.[1–5] One of the most important uses of the
Petasis reaction is the synthesis of a-amino acids using glyoxylic
acid as the aldehyde component (Scheme 1).[6,7] The use of
chiral amines allows the stereoselective production of a-amino
acid derivatives.[8–11]
With the extended results at hand, it was apparent that nearly
all reactions proceed with poor diastereoselectivity and among
these there was only a weak trend relating diastereoselectivity
to electronic character. The most striking result, though, was the
one obvious outlier – use of the unsubstituted styrenylboronic
acid 1a, which resulted in a dr of 20:1. In order to account for the
anomalously high diastereoselectivity in this reaction, we won-
dered if it was a result of the sourcing of the styrenylboronic
acids. To this point, the substituted boronic acids had all been
synthesized in our laboratory, whereas the unsubstituted com-
pound was obtained commercially. We therefore sourced some
of the substituted styrenylboronic acids commercially, which
were at this point becoming increasingly available, and also
prepared a sample of styrenylboronic acid 1a by hydroboration
of phenylacetylene. While the commercial substituted styrenyl-
boronic acid samples 1e, f, i unsurprisingly gave identical results
to those obtained using prepared samples, intriguingly, the use
of the prepared sample of styrenylboronic acid 1a proceeded
with very low diastereoselectivity (dr 1.3:1, Table 1). While
the selectivity of this reaction was now in accord with the
substituted boronic acids 1b–i, the anomaly now shifted to the
difference between the commercial sample of styrenylboronic
acid and the sample prepared in our laboratories. A new batch of
commercial styrenylboronic acid was sourced and subjected to
standard Petasis reaction conditions, and was also found to give
rise to low diastereoselectivity. It was apparent, therefore, that
We recently reported the use of N-benzylphenylglycinol 2
as a suitable chiral secondary amine in Petasis reactions with
styrenylboronic acids.[12] In these reactions, the initial Petasis
reaction product 4 cyclizes to give the corresponding oxazinone
5. The use of styrenylboronic acid 1a gave the oxazinone 5a in
excellent yield and 20:1 diastereomeric ratio (Scheme 2). We
then investigated the use of substituted styrenylboronic acids to
give functionalized homophenylalanine derivatives. However,
the Petasis reactions with substituted styrenylboronic acids
either failed to proceed (those containing strongly electron-
withdrawing substituents), or proceeded with low levels of
diastereoselectivity (those containing electron-donating substi-
tuents). These results were initially interpreted as the influence
of the electronic nature of the substituent upon both the rate
and diastereoselectivity of the reactions.[12] While the lack of
reactivity with the strongly electron-withdrawing nitro-group is
consistent with decreased nucleophilicity of the boronic acid,
the perceived trend of a decrease in diastereoselectivity as more
electron donating substituents are incorporated into the styre-
nylboronic acid substrate was surprising. In order to discern the
subtle factors affecting the stereoselectivity of these Petasis
reactions, we have investigated a wider range of substituted
styrenylboronic acids, and report our results herein.
An extended range of 4-substituted styrenylboronic acids
including 1e–i was prepared by hydroboration of the
Ó CSIRO 2011
10.1071/CH10341
0004-9425/11/010062

Anomalies in the Stereoselectivity of the Petasis Reaction
63
R1
N
the original commercial sample of styrenylboronic acid,
which had been used for several years and by all researchers
in the group involved in studies of the Petasis reaction up to
2008, was the only sample that gave a highly stereoselective
reaction.
R1
N
B(OH)2
CO2H
O
CO2H
R2
R3
ϩ
ϩ
R2
H
H
R3
We surmised that an impurity was present in the original
batch of styrenylboronic acid 1a that was influencing the
diastereoselectivity of the reaction. Unfortunately, with none
of the original sample remaining, and despite an extensive
worldwide search for samples from the same batch number,
we were left to guess as to the nature of the impurity. A series
of reactions were conducted with a wide range of additives to
search for conditions that restored high levels of selectivity.
The additives included: water, molecular sieves, boric acid,
catechol, BHT, boron trifluoride, triethylamine, and use of an
excess of each of the reactants. In addition, the reaction was
performed with styrenyl boroxine 7 and styrenyl trihydroxy-
borate 8 as the organoboron source (Fig. 1). In all examples,
low levels of diastereoselectivity were observed. As such,
the mystery surrounding the origin of the high levels of dia-
stereoselectivity observed from the original batch of styrenyl-
boronic acid 1a remains as inexplicable as ever.
With low levels of diastereoselectivity now the norm, we
transferred our focus to methods for isolating the diasteromeric
b,g-dehydrohomophenylalanine products 5 in pure form.
Attempts to equilibrate the isomers upon treatment with base
according to the method of Jiang resulted only in isomerization
to the corresponding a,b-dehydrohomophenylalanine products
9.[12,13] However, it was noticed that the isomerization of the
diastereomeric substrates 5 proceeded at significantly different
rates, and as such the (R,R)-isomer could be isolated through a
kinetic resolution process. Specifically, treatment of a 1.3:1
mixture of b,g-dehydrohomophenylalanine diastereomers 5a
with triethylamine in chloroform at room temperature for 18 h
gave the a,b-dehydrohomophenylalanine 9a derivative in
58% yield, with recovery of the (R,R)-isomer 5a in 38% yield
(Scheme 4). That is, complete isomerization of the (R,S)-isomer
of 5a to the a,b-dehydrohomophenylalanine 9a occurred, but
only a small amount of the (R,R)-isomer of 5a underwent
isomerization to the a,b-dehydrohomophenylalanine 9, with
67% of (R,R)-5a recovered unchanged.
Scheme 1. Synthesis of a-amino acids by the Petasis reaction.
OH
Ph
BnN
OH
Ph
BnN
O
O
B(OH)2
B(OH)2
O
Ph
NHBn
O
2
OHC–CO2H
3
1a
(R,R)-5a
4
Scheme 2.
Ph
BnN
OH
O
B(OH)2
O
(i) Catechol borane,
Ph
NHBn
THF, 70ЊC
2
(ii) H2O
OHC–CO2H
3
R
R
6
1
R
5
Scheme 3.
Table 1. Comparison of Petasis reaction yields and stereoselectivities
from Scheme 3
Product
R
Yield [%]
dr
5a
5b
5c
5d
5e
5f
H
89
92
82
90
90
95
87
99
92
20:1A 1.3:1B
1.3:1A
1.2:1A
1:1A
CH3
OAc
OTPS
F
The different rates of isomerization are attributed to the
different torsion angles between the Ca–H bond (C2–H in
Fig. 2) and the C¼O bond. For the (R,R)-isomer of 5a, the
crystal structure shows this torsion angle to be 238 (Fig. 2),[12]
with consequential poor overlap between the Ca–H s-bond and
the C¼O p-bond, rendering this hydrogen only weakly acidic.
The (R,S)-isomer would presumably have the Ca–H bond in
the same position as the Ca–Cb bond in the (R,R)-isomer
3:1B
Cl
2.2:1B
1.4:1B
1.9:1B
1.1:1B
5g
5h
5i
Ph
OCF3
OMe
AFrom ref. [12].
BThis work.
Ph
O
Ph
BnN
Ph
BnN
O
O
OH
BnN
O
HO
OH
O
O
O
3
B
B
DBU or Et3N
Na
ϩ
R
R
R
5
9
(R,R)-5
7
8
Scheme 4.
Fig. 1. Alternative styrenylboron reagents.

64
Q. I. Churches et al.
(i.e. pseudo-axial rather than pseudo-equatorial, equivalent to
C2–C5 position in Fig. 2), with the Ca–H bond therefore ideally
positioned for abstraction (H–Ca–C¼O torsion angle ,908).
Preferential deprotonation of the minor (R,S)-isomer and sub-
sequent isomerization therefore proceeds, enabling isolation of
the pure major (R,R)-isomer of 5a in reasonable yield.
the asymmetric dihydroxylation of 5a. The IND-ligands have
only one alkaloid group and thus possess a more accessible
‘L’-shaped binding cleft. Treatment of 5a with catalytic osmium
tetroxide in the presence of DHQD-IND gave the dihydroxy-
lated product 10 as a single diastereomer in 69% yield
(Scheme 5). In the absence of crystallographic analysis, the
configuration of the diol was tentatively assigned by correlation
with the Sharpless mnemonic for the [DHQ(D)]2PHAL ligands.
An analogous reaction in the presence of the pseudoenantio-
meric DHQ-IND ligand proceeded much more slowly, with only
17% yield isolated after 72 h. The low rate of dihydroxylation
presumably reflects the mismatched case where the configura-
tion of the stereocentres on the oxazine ring adversely affects
binding to and/or reaction at the DHQ-IND ligand complex.
Despite the negative influence of this mismatched effect on
reaction rate, the dihydroxylation nevertheless proceeds with
high diastereoselectvity, yielding only the complementary dia-
stereomer 11 to that produced in the presence of the DHQD-IND
ligand. As such, the use of the indolinyl ligands in the dihy-
droxylation of dehydrophenylalanine derivative 5a allows for
stereoselective formation of each of the stereoisomers, 10 and
11, of the corresponding syn-diol.
With substantial quantities of the dehydrophenylalanine
derivative in hand, we turned our focus to the elaboration of
the olefin to generate highly functionalized amino acid deriva-
tives. b,g-Dihydroxyhomotyrosine is a constituent of echino-
candin B, a member of the echinocandin family of anti-fungal
cyclic peptides.[14] The echinocandin family includes caspo-
fungin, a semi-synthetic variant of pneumocandin B0 that is
in clinical use for the treatment of Aspergillus infection.[15]
Despite intense research into the synthesis of the echinocandin
peptides and their analogues, no total synthesis of the most
potent member, echinocandin B, has been forthcoming, and only
one route to the synthesis of a dihydroxyhomotyrosine deriva-
tive has been reported.[16] We envisaged that asymmetric
dihydroxylation of Petasis reaction products such as dehydro-
phenylalanine derivative 5a, would represent a highly efficient
and general route to dihydroxyhomoarylalanine derivatives,
applicable to the preparation of dihydroxyhomotyrosine.
Accordingly, dehydrophenylalanine derivative 5a was trea-
ted with the commercial AD-mix-a and AD-mix-b reagents.
However, under standard asymmetric dihydroxylation condi-
tions, no reaction was observed and the starting material was
recovered. In order to assess reactivity of the olefin, compound
5a was treated with osmium tetroxide in the absence of a chiral
ligand. Dihydroxylation of the olefin was observed, providing
the dihydroxyhomophenylalanine derivative in 60% yield as a
1:1 mixture of diastereomers 10 and 11. The lack of reactivity of
5a with the AD-mixes was attributed to the bulky oxazinyl group
hindering binding of the olefin into the tight ‘U’-shaped cleft
of the [DHQ(D)]2PHAL ligands.[17] The poor performance of
phthalazine-based ligands in the asymmetric dihydroxylation of
olefins with bulky groups in close proximity to the double bond
is well established.[17] We therefore investigated the use of the
indolinyl-derived ligands, DHQD-IND and DHQ-IND,[18,19] in
Conclusion
In conclusion, we have unearthed surprising and inexplicable
anomalies in the stereochemical course of the Petasis reaction.
Nevertheless, the tandem Petasis reaction–asymmetric dihy-
droxylation sequence represents a highly efficient route to
dihydroxyhomoarylalanine derivatives that could be exploited
in the preparation of non-proteinogenic amino acid components
of biologically active cyclic peptides.
Experimental
(E)-2-Arylethenylboronic Acids 1
Styrenyl boronic acids 1 were purchased from Sigma–Aldrich
and/or prepared by hydroboration of the corresponding
arylacetylene according to our published method.[12]
C(10)
Ph
O
C(11)
C(9)
N
C(12)
Bn
O
C(8)
OH
C(35)
C(7)
DHQD-IND
OsO4
HO
Ph
Bn
K3Fe(CN)6
C(6)
C(5)
O
N
10
O
O(2)
C(2)
C(15)
C(1)
Ph
Bn
C(14)
C(16)
C(17)
O
N(1)
O(1)
C(13)
C(21)
N
C(3)
DHQ-IND
OsO4
K3Fe(CN)6
O
C(19)
C(4)
C(20)
OH
C(22)
HO
C(18)
C(25)
5a
C(23)
C(24)
11
Fig. 2. ORTEP diagram of oxazinone 5a.
Scheme 5.

Anomalies in the Stereoselectivity of the Petasis Reaction
65
(E)-2-(4-Trifluoromethoxyphenyl)ethenylboronic Acid 1h
4-Benzyl-5R-phenyl-3RS-(E-2-(4-fluorophenyl)ethenyl)-
oxazin-2-one 5e
dH (d6-acetone, 500 MHz) 7.64 (m, 2H), 7.38 (br d, J 18.4, 1H),
7.32 (br d, J 8.7, 2H), 6.25 (dd, J 18.4, 2.5, 1H). dC (CD3OD,
125 MHz) 147.9, 139.5, 136.7, 130.5, 128.0, 127.6 (q, JCF.
227.8) (C–B not observed).
1H NMR (CDCl3, 500 MHz) (major (R,R)-diastereomer) 7.59
(br d, J 7.8, 2H), 7.49–7.22 (m, 10H), 7.08–7.01 (m, 2H), 6.63 (d,
J 15.8, 1H), 6.35 (dd, J 7.8, 15.8, 1H), 4.64 (dd, J 4.5, 11.4, 1H),
4.55 (dd, J 8.1, 11.3, 1H), 4.41–4.33 (m, 3H), 3.74 (d, J 13.7,
1H), 3.51 (d, J 13.7, 1H). dH (CDCl3, 500 MHz) (minor (R,S)-
diastereomer) 7.59 (br d, J 7.8, 2H), 7.49–7.22 (m, 10H), 6.70
(dd, J 15.9, 1.8, 1H), 6.11 (dd, J 5.6, 16.0, 1H), 4.39 (dd, J 1.8,
5.3, 1H), 4.41–4.33 (m, 3H), 4.22 (dd, J 3.5, 11.6, 1H), 4.09
(dd, J 3.5, 11.6, 1H), 3.89 (d, J 13.8, 1H), 3.68 (d, J 13.9, 1H).
dC (CDCl3, 125 MHz) 169.8, 169.0, 162.7 (d, JCF. 248.2), 162.6
(d, JCF. 241.2), 137.6, 137.0, 136.9, 136.4, 135.5, 132.1, 131.4,
129.4, 129.1, 129.0, 128.7, 128.6, 128.6, 128.5, 128.5, 128.3,
128.3, 128.2, 128.2, 128.1, 128.0, 127.6, 127.4, 127.2, 121.3,
121.3, 115.6 (d, JCF. 21.7), 115.5 (d, JCF. 21.7), 72.6, 71.0, 63.5,
63.5, 62.8, 62.6, 57.7, 56.9, 52.8. m/z (HR-ESI) calc. for
C25H23FNO2: 388.1707 [M þ H]þ. Found: 388.1708.
N-Benzyl-5-phenyloxazin-2-ones 5
Oxazinones were prepared according to our published
method.[12]
4-Benzyl-5R-phenyl-3S-(2E-phenylethenyl)-
oxazin-2-one 5a
dH (CDCl3, 500 MHz) (major (R,R)-diastereomer) 7.59 (d, J 7.3,
2H), 7.54–7.15 (m, 14H), 6.68 (d, J 15.8, 1H), 6.45 (dd, J 7.8,
15.8, 1H), 4.64 (dd, J 4.3, 11.1, 1H), 4.55 (dd, J 11.1, 8.4, 1H),
4.45–4.28 (m, 3H), 3.74 (d, J 13.7, 1H), 3.52 (d, J 13.7, 1H).
dH (CDCl3, 500 MHz) (minor (R,S)-diastereomer) 7.61 (d, J 7.4,
2H), 7.51–7.16 (m, 12H), 6.77 (d, J 16.0, 1H), 6.22 (dd, J 5.1,
16.0, 1H), 4.45–4.28 (m, 3H), 4.21 (dd, J 3.3, 11.6, 1H), 4.10
(br d, J 11.6, 1H), 3.90 (d, J 13.8, 1H), 3.70 (d, J 13.9, 1H).
dC (CDCl3, 125 MHz) 170.3, 169.2, 138.2, 137.1, 137.1, 136.7,
136.5, 133.4, 133.1, 132.3, 129.4, 129.3, 129.2, 129.0, 128.6,
128.5, 128.5, 128.3, 128.1, 128.0, 127.6, 127.4, 126.6, 121.5,
121.3, 72.6, 70.8, 62.8, 62.5, 57.6, 56.9, 52.7. m/z (HR-EI) calc.
for C25H23NO2: 369.1729 [M]þ. Found: 369.1729.
4-Benzyl-5R-phenyl-3RS-(E-2-(4-chlorophenyl)ethenyl)-
oxazin-2-one 5f
1H NMR (CDCl3, 500 MHz) (major (R,R)-diastereomer) 7.55
(br d, J 7.8, 2H), 7.46–7.18 (m, 12H), 7.08–7.01 (m, 2H), 6.59 (d,
J 15.8, 1H), 6.38 (dd, J 7.8, 15.8, 1H), 4.61 (dd, J 4.5, 11.4, 1H),
4.52 (dd, J 8.1, 11.3, 1H), 4.38–4.30 (m, 3H), 3.70 (d, J 13.7,
1H), 3.51 (d, J 13.7, 1H). dH (CDCl3, 500 MHz) (minor (R,S)-
diastereomer) 7.59 (br d, J 7.8, 2H), 7.46–7.18 (m, 12H), 6.66
(dd, J 15.9, 1.8, 1H), 6.13 (dd, J 5.6, 15.9, 1H), 4.38–4.30 (m,
3H), 4.18 (dd, J 3.5, 11.6, 1H), 4.05 (dd, J 3.5, 11.6, 1H), 3.86 (d,
J 13.9, 1H), 3.64 (d, J 13.9, 1H). dC (CDCl3, 125 MHz) 169.7,
168.9, 137.5, 137.0, 136.8, 136.3, 135.4, 134.6, 134.4, 134.0,
133.6, 131.3, 129.4, 129.0, 129.0, 128.9, 128.8, 128.7, 128.6,
128.6, 128.5, 128.4, 128.4, 128.3, 128.1, 128.1, 128.0, 127.9,
127.9, 127.8, 127.6, 127.5, 122.3, 72.6, 71.1, 63.5, 62.8, 62.5,
57.7, 56.9, 52.9. m/z (HR-ESI) calc. for C25H2335ClNO2:
404.1412 [M þ H]þ. Found: 404.1412.
4-Benzyl-5R-phenyl-3RS-(E-2-(4-methylphenyl)ethenyl)-
oxazin-2-one 5b
dH (major (R,R)-diastereomer) 7.61 (d, J 7.4, 2H), 7.51–7.16 (m,
12H), 6.65 (d, J 15.8, 1H), 6.43 (dd, J 7.8, 15.8, 1H), 4.65 (dd,
J 4.5, 11.4, 1H), 4.56 (dd, J 11.4, 8.4, 1H), 4.43–4.35 (m, 3H),
3.75 (d, J 13.7, 1H), 3.54 (d, J 13.7, 1H), 2.39 (s, 3H). dH
(CDCl3, 500 MHz) (minor (R,S)-diastereomer) 7.61 (d, J 7.4,
2H), 7.51–7.16 (m, 12H), 6.77 (d, J 16.0, 1H), 6.20 (dd, J 5.2,
16.0, 1H), 4.43–4.35 (m, 3H), 4.21 (br d, J 11.6, 1H), 4.10 (br d, J
11.6, 1H), 3.90 (d, J 13.8, 1H), 3.72 (d, J 13.8, 1H), 2.38 (s, 3H).
dC (CDCl3, 125 MHz) 170.3, 169.2, 138.2, 137.9, 137.7, 137.1,
137.1, 136.7, 136.5, 133.4, 133.1, 132.3, 129.4, 129.3, 129.2,
129.0, 128.6, 128.5, 128.5, 128.3, 128.1, 128.0, 127.6, 127.4,
126.6, 126.5, 126.4, 72.6, 70.9, 62.9, 62.6, 57.7, 57.0, 52.8, 22.7,
21.2. IR: 3028, 2922, 1741, 1455, 907, 697 cmꢁ1. m/z (HR-ESI)
calc. for C26H26NO2: 384.1958 [M þ H]þ. Found: 384.1958.
4-Benzyl-5R-phenyl-3RS-(E-2-(40-biphenyl)ethenyl)-
oxazin-2-one 5g
1H NMR (CDCl3, 500 MHz) (major (R,R)-diastereomer) 7.65–
7.58 (m, 5H), 7.52–7.25 (m, 14H), 6.72 (d, J 15.8, 1H), 6.49 (dd,
J 7.8, 15.8, 1H), 4.66 (dd, J 4.5, 11.4, 1H), 4.56 (dd, J 11.4, 8.4,
1H), 4.46–4.36 (m, 3H), 3.76 (d, J 13.7, 1H), 3.54 (d, J 13.7, 1H).
dH (CDCl3, 500 MHz) (minor (R,S)-diastereomer) 7.61 (d, J 7.4,
2H), 7.51–7.16 (m, 12H), 6.84 (d, J 16.0, 1H), 6.25 (dd, J 5.2,
16.0, 1H), 4.46–4.36 (m, 3H), 4.23 (dd, J 3.5, 11.6, 1H), 4.10
(dd, J 3.5, 10.8, 1H), 3.92 (d, J 13.8, 1H), 3.71 (d, J 13.8, 1H).
dC (CDCl3, 125 MHz) 169.9, 169.1, 141.1, 140.8, 140.5, 140.5,
137.7, 137.1, 137.0, 136.4, 136.3, 135.2, 134.9, 132.0, 129.4,
129.0, 128.8, 128.8, 128.7, 128.6, 128.6, 128.5, 128.4, 128.1,
128.0, 127.6, 127.5, 127.4, 127.4, 127.3, 127.2, 127.1, 127.0,
127.0, 126.9, 126.9, 121.5, 72.6, 71.0, 63.4, 63.4, 62.9, 62.6,
57.7, 57.0, 52.8. m/z (HR-ESI) calc. for C31H28NO2: 446.2115
[MþH]þ. Found: 446.2115.
4-Benzyl-5R-phenyl-3RS-(E-2-(4-acetoxyphenyl)ethenyl)-
oxazin-2-one 5c
1.2:1 mixture of diastereomers (82%). dH (major (R,R)-
diastereomer) 7.58 (d, J 7.8, 2H), 7.44–7.22 (m, 12H), 6.64 (d, J
15.8, 1H), 6.39 (dd, J 7.8, 15.8, 1H), 4.62 (dd, J 4.5, 11.4, 1H),
4.53 (dd, J 8.2, 11.4, 1H), 4.35–4.32 (m, 3H), 3.72 (d, J 13.7, 1H),
3.49 (d, J 13.7, 1H), 2.32 (s, 3H). dH (CDCl3, 500 MHz) (minor
diastereomer) 7.58 (d, J 7.8, 2H), 7.44–7.22 (m, 12H), 6.73 (dd, J
16.0, 1.8, 1H), 6.15 (dd, J 5.6, 16.0, 1H), 4.39 (dd, J 1.8, 5.3, 1H),
4.35–4.32 (m, 3H), 4.20 (dd, J 3.5, 11.4, 1H), 4.08 (dd, J 3.5, 11.4,
1H), 3.89 (d, J 13.8, 1H), 3.67 (d, J 13.8, 1H), 2.31 (s, 3H). dC
(CDCl3, 125MHz) 169.6, 169.3, 150.9, 150.6, 137.9, 137.31,
137.28, 136.7, 136.0, 134.2, 134.0, 131.8, 129.73, 129.69, 129.40,
129.37, 129.3, 129.01, 128.99, 128.97, 128.92, 128.89, 128.84,
128.79, 128.78, 128.74, 128.65, 128.4, 128.2, 128.1, 128.0,
127.92, 127.85, 127.7, 127.66, 122.2, 122.1, 122.04, 121.97, 77.6,
77.3, 77.1, 72.9, 71.2, 63.7, 63.2, 62.8, 58.0, 57.3, 53.1, 21.4. IR:
3031, 1742, 1506, 1190, 907, 697 cmꢁ1. m/z (HR-ESI) calc. for
C27H26NO4: 428.1856 [Mþ H]þ. Found: 428.1856.
4-Benzyl-5R-phenyl-3RS-(E-2-(4-trifluoromethoxyphenyl)
ethenyl)-oxazin-2-one 5h
1H NMR (CDCl3, 500 MHz) (major (R,R)-diastereomer) 7.60–
7.59 (m, 2H), 7.50–7.21 (m, 12H), 6.66 (d, J 15.8, 1H), 6.42 (dd,
J 7.8, 15.8, 1H), 4.65 (dd, J 4.5, 11.4, 1H), 4.57 (dd, J 8.1, 11.3,
1H), 4.38–4.33 (m, 3H), 3.75 (d, J 13.7, 1H), 3.51 (d, J 13.7, 1H).

66
Q. I. Churches et al.
dH (CDCl3, 500 MHz) (minor (R,S)-diastereomer) 7.60–7.59
(m, 2H), 7.50–7.21 (m, 12H), 6.73 (dd, J 15.9, 1.8, 1H), 6.17 (dd,
J 5.5, 15.9, 1H), 4.41 (dd, J 1.8, 5.5, 1H), 4.38–4.33 (m, 3H),
4.23 (dd, J 3.5, 11.6, 1H), 4.09 (dd, J 3.5, 11.6, 1H), 3.90 (d,
J 13.9, 1H), 3.68 (d, J 13.9, 1H). dC (CDCl3, 125 MHz) 169.7,
168.9, 149.0, 149.0, 137.6, 137.0, 136.8, 136.4, 135.1, 134.9,
134.7, 131.1, 129.5, 129.1, 128.7, 128.7, 128.6, 128.6, 128.5,
128.4, 128.1, 128.0, 127.9, 127.7, 127.5, 122.9, 121.1, 121.1,
120.5 (q, JCF. 257.3), 72.6, 71.1, 63.6, 63.5, 62.8, 62.6, 62.6,
57.7, 56.9, 52.9. m/z (HR-ESI) calc. for C26H22F3NO3Na:
476.1444 [M þ Na]þ. Found: 476.1444.
(2.5% wt in t-butanol) (0.3 mL, 0.02 mmol), K3Fe(CN)6 (1.19 g,
3.6 mmol), K2CO3 (500 mg, 3.6 mmol), and methane sulfona-
mide (1.14 g, 1.2 mmol) were stirred in 1:1 t-butanol/water
(13 mL) at room temperature for 48 h. This solution was cooled
to 08C and treated with sodium metabisulfite (1.52 g, 8.0 mmol)
then warmed to room temperature. The product was extracted
with ethyl acetate (3 ꢂ 30 mL) and the combined organic
extracts were washed with water (2 ꢂ 20 mL), brine (20 mL),
and dried with MgSO4. The solvent was then evaporated under
reduced pressure. Purification by flash chromatography (25%
ethyl acetate/petroleum spirits) gave the product 10 as a white
solid (229 mg, 69%); mp 143–1478C, [a]2D2 ꢁ55 (c 0.47, CHCl3).
dH (CDCl3, 400 MHz) 7.45–7.07 (m, 15 H, ArH), 4.94 (m, 2H),
4.87–4.80 (m, 1H), 4.71 (s, 1H), 4.20 (d, J 5.2, 1H), 3.98 (d,
J 5.2, 1H), 3.85 (d, J 12.8, 1H), 3.56 (d, J 12.8, 1H). dC (CDCl3,
400 MHz) 169.9, 129.5, 129.2, 129.1, 128.8, 128.5, 128.4,
128.1, 127.6, 127.5, 126.0, 125.9, 74.3, 73.5, 62.7, 56.1. m/z
(ESIþ) 404.18 [M þ H]þ.
4-Benzyl-5R-phenyl-3RS-(E-2-(4-methoxyphenyl)ethenyl)-
oxazin-2-one 5i
1H NMR (CDCl3, 500 MHz) ((R,R)-diastereomer) 7.59 (d, J 7.4,
2H), 7.48–7.23 (m, 12H), 6.92–6.87 (m, 2H), 6.60 (d, J 15.8, 1H),
6.29 (dd, J 7.8, 15.8, 1H), 4.63 (dd, J 4.5, 11.4, 1H), 4.53 (dd,
J 11.4, 8.4, 1H), 4.43–4.32 (m, 3H), 3.85 (s, 3H), 3.73 (d,
J 13.7, 1H), 3.51 (d, J 13.7, 1H). dH (CDCl3, 500 MHz) ((R,S)-
diastereomer) 7.59 (d, J 7.4, 2H), 7.48–7.23 (m, 12H), 6.92–6.87
(m, 2H), 6.69 (d, J 16.0, 1H), 6.07 (dd, J 5.2, 16.0, 1H), 4.43–4.32
(m, 3H), 4.20 (dd, J 3.5, 11.5, 1H), 4.08 (dd, J 3.5, 11.5, 1H), 3.89
(br d, J 14.9, 1H), 3.84 (s, 3H), 3.69 (d, J 13.8, 1H). dC (CDCl3,
125 MHz) 170.1, 169.3, 159.7, 159.5, 137.7, 137.1, 136.5, 136.3,
131.8, 129.4, 129.0, 128.9, 128.7, 128.5, 128.5, 128.5, 128.3,
128.1, 128.0, 127.9, 127.8, 127.6, 127.3, 125.2, 118.8, 114.0,
113.9, 72.6, 70.9, 63.4, 62.9, 62.6, 57.7, 57.0, 55.3, 52.7. m/z (HR-
ESI) calc. for C26H26NO3: 400.1907 [M þ H]þ. Found: 400.1907.
Dihydroxylation of 5a Using DHQ-IND
4-Benzyl-5-phenyl-3-(2-phenylethenyl)-oxazin-2-one 5a
(80 mg, 0.22 mmol), DHQ-IND (26 mg, 0.06 mmol), OsO4
(2.5% wt in t-butanol) (0.15 mL, 0.01 mmol), K3Fe(CN)6
(214 mg, 0.65 mmol), K2CO3 (94 mg, 0.68 mmol), and methane
sulfonamide (25 mg, 0.26 mmol) were stirred in 1:1 t-butanol/
water (10 mL) at room temperature for 72 h. The mixture
was cooled to 08C and treated with sodium metabisulfite
(274 mg, 0.22 mol) and the resulting slurry was warmed to room
temperature. The product was extracted with ethyl acetate
(3 ꢂ 25 mL) and the combined organic extracts were washed
with water (25 mL), brine (2 ꢂ 25 mL), and dried (MgSO4). The
solvent was then evaporated under reduced pressure. Purifica-
tion by flash chromatography (25% ethyl acetate/petroleum
spirits) gave the product 11 as a white powder (15 mg, 17%).
dH (CDCl3, 300 MHz) 7.53–7.07 (m, 15H, ArH), 5.30 (s, 1H,
OH), 5.01–4.99 (m, 1H), 4.88–4.72 (m, 1H), 4.56–4.47 (m, 2H),
3.77 (d, J 2.5, 1H), 3.60 (d, J 13.9, 1H, NCHHPh), 3.51 (d,
J 13.9, 1H, NCHHPh). dC (CDCl3, 400 MHz) 169.8, 129.5,
129.2, 129.1, 128.8, 128.5, 128.4, 128.0, 127.6, 127.6, 126.0,
125.9, 74.3, 73.5, 62.7, 56.1. m/z (ESIþ) 404.16 [M þ H]þ.
Kinetic Resolution Procedure to Yield (R,R)-b,g-
dehydrooxazinones 5
Triethylamine (14.0 mL, 0.1 mmol) was added to a solution of a
mixture of diastereomers of the 3-(2-arylethenyl)-oxazinone 5
(0.1 mmol) in CDCl3 (5 mL), and the mixture was stirred at rt
with monitoring by NMR. Once the reaction was complete,
saturated ammonium chloride solution (5 mL) was added and
the organic layer separated, dried (MgSO4), filtered, and the
solvent was removed under reduced pressure. Flash chromato-
graphy eluting with 10% ethyl acetate in petroleum spirits gave
a,b-dehydro compound 9 and the recovered (R,R)-b,g-dehydro
compound 5.
Accessory Publication
4-Benzyl-3-(1,2-dihydroxy-2-phenylethyl)-5-phenyl-
oxazin-2-ones 10 and 11
1H and 13C NMR spectra for compounds 5a–c, 5e–i, 10, 11 are
available on the Journal’s website.
Dihydroxylation of 5a with OsO4
4-Benzyl-5-phenyl-3-(2-phenylethenyl)-oxazin-2-one 5a
(109 mg, 0.30 mmol) and N-methylmorpholine-N-oxide (60%
w/w in water) (0.225 mL, 1.15 mmol) were dissolved in THF
(1.2 mL) and water (0.12 mL). Osmium tetroxide [2.5% w/w in
t-butanol] (0.123 mL, 0.01 mmol) was added and the mixture
was stirred for 78 h at room temperature. Flash chromatography
eluting with 10–25% ethyl acetate in petroleum spirits gave the
diols 10 and 11 as a 1:1 mixture (74 mg, 60%); mp 162–1678C.
dH (CDCl3, 300 MHz) 7.53–7.07 (m, 15H, ArH), 5.30 (s, 1H,
OH), 5.01–4.99 (m, 3H), 4.88–4.71 (m, 3H), 4.56–4.47 (m, 2H),
4.20 (d, J 5.2, 1H), 3.98 (d, J 5.2, 1H), 3.85 (d, J 12.8, 1H), 3.77
(d, J 2.5, 1H), 3.60 (d, J 13.9, 1H), 3.51 (d, J 13.9, 1H), 3.56 (d,
J 12.8, 1H).
Acknowledgements
This work was supported by the Australian Research Council.
References
[1] N. Selander, K. J. Szabo´, Asymmetric Synthesis and Application
of a-Amino Acids 2009, Vol. 1009, Ch. 12, pp. 190–202 (Eds V. A.
Soloshonok, K. Izawa) (ACS Symposium Series: Washington, DC).
[2] Boronic Acids – Preparation, Applications in Organic Synthesis and
Medicine 2005 (Ed. D. G. Hall) (Wiley-VCH: Weinheim).
[3] N. R. Candeias, L. F. Veiros, C. A. M. Afonso, P. M. P. Gois, Eur. J.
Org. Chem. 2009, 1859. doi:10.1002/EJOC.200900056
[4] P. F. Kaiser, Q. I. Churches, C. A. Hutton, Aust. J. Chem. 2007, 60, 799.
doi:10.1071/CH07103
[5] N. A. Petasis, Aust. J. Chem. 2007, 60, 795. doi:10.1071/
CH07360
[6] N. A. Petasis, A. Goodman, I. A. Zavialov, Tetrahedron 1997, 53,
Dihydroxylation of 5a Using DHQD-IND
4-Benzyl-5-phenyl-3-(2-phenylethenyl)-oxazin-2-one 5a
(302 mg, 0.82 mmol), DHQD-IND (321 mg, 0.68 mmol), OsO4
16463. doi:10.1016/S0040-4020(97)01028-4

Anomalies in the Stereoselectivity of the Petasis Reaction
67
[7] N. A. Petasis, I. A. Zavialov, J. Am. Chem. Soc. 1997, 119, 445.
doi:10.1021/JA963178N
[8] B. Jiang, C.-G. Yang, X.-H. Gu, Tetrahedron Lett. 2001, 42, 2545.
doi:10.1016/S0040-4039(01)00229-5
[13] B. Jiang, M. Xu, Org. Lett. 2002, 4, 4077. doi:10.1021/OL026826I
[14] D. W. Denning, J. Antimicrob. Chemother. 2002, 49, 889. doi:10.1093/
JAC/DKF045
[15] C. Wagner, W. Graninger, E. Presterl, C. Joukhadar, Pharmacology
[9] K. K. Nanda, B. W. Trotter, Tetrahedron Lett. 2005, 46, 2025.
doi:10.1016/J.TETLET.2005.01.151
[10] T. J. Southwood, M. C. Curry, C. A. Hutton, Tetrahedron 2006, 62,
2006, 78, 161. doi:10.1159/000096348
[16] C. Palomo, M. Oiarbide, A. Landa, J. Org. Chem. 2000, 65, 41.
doi:10.1021/JO990964C
[17] H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, Chem. Rev. 1994,
236. doi:10.1016/J.TET.2005.09.114
[11] N. A. Petasis, Z. D. Patel, Tetrahedron Lett. 2000, 41, 9607.
doi:10.1016/S0040-4039(00)01717-2
[12] Q. I. Churches, H. E. Stewart, S. B. Cohen, A. Shro¨der, P. Turner,
C. A. Hutton, Pure Appl. Chem. 2008, 80, 687. doi:10.1351/
PAC200880040687
94, 2483. doi:10.1021/CR00032A009
[18] L. Wang, K. B. Sharpless, J. Am. Chem. Soc. 1992, 114, 7568.
doi:10.1021/JA00045A042
[19] R. M. Davey, M. A. Brimble, M. D. McLeod, Tetrahedron Lett. 2000,
41, 5141. doi:10.1016/S0040-4039(00)00798-X