Polyamino acids as catalysts in asymmetric synthesis

Article abstract of DOI:10.1016/S0968-0896(99)00144-3

The use of polyamino acids in asymmetric organic synthesis is reviewed. Particular emphasis is placed on the asymmetric epoxidation of α,β- unsaturated ketones with hydrogen peroxide in the presence of polyalanine or polyleucine, and further transformatio

Full text of DOI:10.1016/S0968-0896(99)00144-3

Bioorganic & Medicinal Chemistry 7 (1999) 2145±2156
Review Article
Polyamino Acids as Catalysts in Asymmetric Synthesis
Michael J. Porter, Stanley M. Roberts* and John Skidmore
Department of Chemistry, University of Liverpool, Liverpool, PO Box 147, L69 7ZD, UK
Received 1 July 1998
AbstractÐThe use of polyamino acids in asymmetric organic synthesis is reviewed. Particular emphasis is placed on the asymmetric
epoxidation of a,b-unsaturated ketones with hydrogen peroxide in the presence of polyalanine or polyleucine, and further trans-
formations of the epoxide products. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
leads to hydroxylation at a position remote from pre-
existing functionality.5±7 The oxidation of aromatic
compounds to provide phenols is also an important
transformation and has been carried out on a large
scale.8 The conversion of aromatic compounds into
cyclohexadienediols is a biotransformation that is
impossible to emulate in one step using other chemical
technology. Such dienediols have been exploited in a
most elegant fashion in organic synthesis.9,10
Enzymes are complex macromolecules consisting of up
to 20 di€erent amino acids. Typically, neighbouring
amino acid residues are di€erent [(1) R1 ˆ R2 ˆ R3. . .],
and an enzyme contains few regions of homogeneity.
This structural diversity leads to the wide variety of
enzymes that are found in microorganisms, plants and
animals. However the very complexity of protein struc-
ture has the consequence that the vast majority of
enzymes are not amenable to chemical synthesis;
recourse to cloning of the gene and over-expression of
the protein is normally the mechanism by which large
quantities of enzymes are obtained.
One area of biotransformation that is notably uncom-
mon is the conversion of alkenes into epoxides (oxi-
ranes). While there are isolated examples of
epoxidations (Scheme 1) giving optically active pro-
ducts, these usually involve terminal alkenes and cer-
tainly the biocatalytic process does not have wide
applicability.11,12
In contrast to the dearth of naturally occurring proteins
which can catalyse epoxidation, Julia reported in 1980
that a synthetic polypeptide consisting solely of alanine
residues, i.e. (1) R1=R2=R3=CH3, is able to catalyse
the epoxidation of chalcone (4) and similar enones
(Scheme 2).13 It was demonstrated that the oxidation
reaction occurs with a high degree of enantioselectivity;
the authors concluded, however, that the protocol was
only applicable to chalcones, that is compounds of the
type Ar1COCHˆCHAr2.
1
Enzymes are able to catalyse a wide range of reac-
tions.1±3 The hydrolysis of esters and amides, the ester-
i®cation of alcohols and the reduction of ketones are
popular biotransformations amongst synthetic organic
chemists, not least because enzymes frequently confer a
high degree of chemo-, regio- and/or stereoselectivity to
the process. Furthermore, the use of bio-hydroxylation
reactions is widespread in academia and industry. The
conversion of acyclic, heterocyclic and alicyclic mole-
cules (e.g. steroids) into alcohols is well established;4 the
employment of mono-oxygenases in whole cells often
This reaction was the ®rst example of asymmetric cata-
lysis by a synthetic polypeptide; subsequent to this
work, a number of further examples have been reported.
In this review we shall survey the preparation and use of
polyamino acids in asymmetric synthesis.14,15 The
review focuses largely on asymmetric epoxidation of
enones, highlighting recent developments that expand
the range of epoxides which may be generated.16±21
Key words: Amino acids and derivatives; peptides and polypeptides;
polymers; reviews.
* Corresponding author. Tel.: +44-151-794-3501; fax: +44-151-794-
3587; e-mail: sj11@liv.ac.uk
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00144-3

2146
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Applications of Poly-ꢀ-amino Acids in Synthesis
Synthetic polyamino acids have found various applica-
tions including use as chiral stationary phases for chro-
matography37,38 and as matrices for controlled drug
delivery systems.39,40 In this review, however, we shall
concentrate on their use as catalysts in asymmetric
synthesis.
Scheme 1. Reagents and conditions: Micrococcus sp. M90C, 30ꢀC, 14±
24 h.
The ®rst report detailing the use of a polypeptide in
asymmetric synthesis was by Akabori et al. in 1956.41
Using palladium supported on the structural protein
silk ®broin, moderate enantiomeric excesses were
obtained in the hydrogenation of prochiral carbon±car-
bon and carbon±nitrogen double bonds (Scheme 4). For
example, hydrogenation of the oxazolinone 10 followed
by hydrolysis a€orded unnatural d-phenylalanine (11)
in 16% yield with 36% ee. Similarly, hydrogenation of
oxime derivative 12 a€orded l-glutamic acid (13) in
39% yield with ca. 7% ee.
Scheme 2. Reagents and conditions: aq. H2O2, poly-l-alanine, aq.
NaOH, toluene, 24 h, 85% yield, 93% ee.
In addition, we shall discuss brie¯y some of the reactions
which have been performed on the chiral a,b-epoxy-
ketones so produced.
Preparation of Poly-ꢀ-amino Acids
The use of polyamino acids as ligands for palladium
was also reported by Alper et al.42 The palladium-cata-
lysed carbonylation of but-2-en-1-ol (14) proceeds in the
presence of high molecular weight (ca. 22,000) poly-l-
leucine to give (R)-a-methyl-g-butyrolactone (15) in
49% yield and 61% ee (Scheme 5). This optical purity
was superior to those obtained with ``conventional''
ligands such as diethyl tartrate or BINAP (2,20-bis(di-
phenylphosphino)-1,10-binaphthalene).
The most commonly used method for polymerisation of
a-amino acids involves initial activation of the amino
acid 6 towards nucleophilic attack by formation of an
N-carboxyanhydride (NCA, 7) (Scheme 3). This is
achieved by use of phosgene22,23 or an equivalent such
as diphosgene24,25 or triphosgene;26 alternatively an N-
alkoxycarbonyl-protected amino acid 8 may be cyclised
using thionyl chloride.27,28 Polymerisation is then initi-
ated by addition of a nucleophile, typically water,29 an
amine30 or diamine31 to give polymer 9.32,33 This
method gives a range of chain lengths in the polymer
product, although the average length can be controlled
by varying the ratio of NCA to initiator.
Nonaka and co-workers have reported the use of
polyamino acids as electrode coatings for asymmetric
electrochemical oxidations and reductions. Thus,
reduction of 4-methylcoumarin (16) on a graphite
cathode coated with poly-l-valine a€orded (S)-3,4-
dihydro-4-methylcoumarin (17) in 43% ee, albeit with
a chemical yield of only 8% (Scheme 6).43 Likewise,
reduction of citraconic acid (18) yielded (S)-methyl-
succinic acid (19) with an ee of 25%.43±45 Curiously,
use of a poly-d-valine-coated electrode was reported to
give the same enantiomer of 19 with lower optical
purity.44
The polymers prepared using simple initiators have
proved dicult to separate from the reaction products.
This problem can to some extent be alleviated by use of
polyamino acids attached to a polystyrene support;
these are prepared by use of a hydroxy-34 or amino-
functionalised35 cross-linked polystyrene as the initiator.
An alternative method of polyamino acid synthesis, and
one which may prove most useful in the elucidation of
reaction mechanisms, is the controlled generation of
de®ned chain length polymers using standard solid-
phase peptide synthesis techniques.36
Reductions of a-keto acids, oximes and a geminal
dibromide were less successful than the conjugate
reductions, with the reported enantiomeric excesses
being in the range 0±17%.46
Scheme 3.

M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
2147
mixture of aqueous sodium hydroxide and toluene. In a
series of papers, Julia and Colonna systematically
examined and optimised the conditions for this trans-
formation:13,34,49±51
. Carbon tetrachloride and toluene were reported as
the optimal organic solvents for the reaction,
whilst cyclohexane and n-hexane were found to
give poor asymmetric induction.49 Julia and
Colonna also reported that the use of methanol as
a solvent results in racemic product; it was sug-
gested that the alcoholic solvent disrupts hydrogen
bonding which was postulated to be important in
setting up the chiral environment essential for an
asymmetric reaction. In a subsequent study (by
Lantos, Novak et al.) on the asymmetric epoxida-
tion of the naphthyl ketone 30 (Table 1), di€erent
solvent e€ects were found.52 In this case n-hexane
was reported to be the optimal organic solvent
with carbon tetrachloride and cyclohexane also
being e€ective, whilst toluene gave little or no
reactivity. The origin of these di€ering solvent
dependences is not clear.
Scheme 4.
Similarly, the asymmetric oxidation of alkyl aryl sul-
phides to sulphoxides using a platinum electrode coated
®rst with polypyrrole and then with a poly-l-amino acid
proceeds with good enantioselectivity (Scheme 7).47,48
The highest ee, 93%, was obtained in the oxidation of
tert-butyl phenyl sulphide (20) to the corresponding (S)-
sulphoxide 21 on a poly-l-valine-coated electrode.
The most extensively studied reaction that uses a poly-
amino acid as a catalyst is the asymmetric epoxidation
of conjugated enones discovered by Julia et al. in 1980.13
Initial results showed that poly-l-alanine could cata-
lyse the epoxidation of chalcone (4) (for example by
using hydrogen peroxide and sodium hydroxide in a
toluene±water mixture), a€ording (2R,3S)-epoxy-
chalcone (5) in good yield and 93% ee (Scheme 8). More
details of this reaction are documented in the following
section.
. It was found that the relative proportions of the
three phases (solid catalyst, organic solvent and
aqueous) has a signi®cant e€ect on the chemical
yield and optical purity. Optimum conditions for
the epoxidation of 1 g of chalcone were reported to
be 8.8 cm3 of water, 13.9 cm3 of toluene and 800
mg of catalyst.13
. The length of the polyamino acid chain has a sig-
ni®cant bearing on the outcome of the reaction.
Initial studies were carried out using a polymer
with a mean length, n, of 10 residues;53 subse-
quently it was reported that, on examination of a
range of polymers with n=5±30, increased poly-
mer length resulted in an increase in the optical
purity of the product.49
. A range of di€erent amino acid monomers was
investigated;51 it was found that poly-l-alanine,
poly-l-leucine and poly-l-iso-leucine are good
catalysts for the epoxidation of chalcone (4). Poly-
l-valine was also reported to catalyse the reaction
but with reduced rate and optical purity of pro-
duct (4% yield and 30% ee after 12 days reaction).
Poly-l-phenylalanine and poly-l-proline were found
to give essentially racemic product.34 Interpretation
Julia±Colonna Epoxidation
The initial report of asymmetric epoxidation by Julia
and co-workers utilised insoluble poly-l-alanine as cat-
alyst and aqueous hydrogen peroxide as oxidant in a
Scheme 5.
Scheme 7.
Scheme 8. Reagents and conditions: aq. H2O2, poly-l-alanine, aq.
NaOH, toluene, 24 h, 85% yield, 93% ee.
Scheme 6.

2148
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Table 1. Substrates epoxidised under triphasic reaction conditionsa
Substrate
Product
Method
a
Yield %
85
ee %
93
Ref
[13]
4
5
a
83
82
[49]
22
23
24 R1=R2=H
26 R1=OMe, R2=H
28 R1=R2=OMe
25 R1=R2=H
27 R1=OMe, R2=H
29 R1=R2=OMe
b
b
b
65
64
74
38
66
84
[55]
[55]
[55]
c
d
a
a
e
d
e
e
e
f
82
82
96
30
75
70
92
85
85
90
57
51
95
>96
80
[52]
[59]
[49]
[49]
[59]
[31]
[31]
[31]
[62]
[63]
[62]
[62]
30
32
34
36
38
40
42
44
46
48
50
52
31
33
35
37
39
41
43
45
47
49
51
53
70
>96
94
>98
90
77
>99
90
g
e
>94
(continued on next page)

M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
2149
Table 1 (continued)
Substrate
Product
Method
e
Yield %
60
ee %
90
Ref
[60]
54
56
58
60
62
55
57
59
61
63
65
e
e
g
g
f
Ð
74
>98
>99
>95
>95
>96
[59]
[31]
[62]
[62]
[31]
100
66
78
64
a
Reagents and conditions: a. aq. H2O2, aq. NaOH, toluene, poly-l-alanine; b. aq. H2O2, aq. NaOH, CCl4, poly-l-alanine; c. aq. NaOH, n-hexane,
EDTA, poly-l-leucine; d. aq. H2O2, aq. NaOH, DCM, poly-d-leucine; e. aq. H2O2, aq. NaOH, DCM, poly-l-leucine; f. aq. H2O2, aq. NaOH, n-
hexane, poly-l-leucine; g. aq. H2O2, aq. NaOH, toluene, CLAMPS-poly-l-leucine.
of these results in terms of the expected secondary
structure of the various polyamino acids was
attempted. Poly-l-alanine and poly-l-leucine are
reported to form a-helices, whereas poly-l-valine
is known to adopt a b-sheet structure. This sug-
gests that increased a-helical character might be
associated with catalytic function and degree of
asymmetric induction. Further support for this
hypothesis was provided in the form of results
obtained with poly-dl-alanine; this is known to
adopt the b-sheet conformation and, as expected,
generates 5 in poor yield. Not all the results were
consistent with the a-helix hypothesis: poly-l-iso-
leucine is reported to exist as a b-sheet; however it
proved to be an excellent catalyst. Moreover, poly-
b-benzyl-l-aspartate and poly-g-benzyl-l-gluta-
mate adopt a-helices with opposite handedness,
but each gives the same major enantiomer in the
epoxidation reaction. Unfortunately both were
relatively poor catalysts for the reaction (enantio-
meric excesses 3 and 11.6%, respectively). In a
later study Lantos and Novak52 examined a range
of polyamino acids and reported that poly-l-leu-
cine consistently gave the highest yields and enan-
tioselectivities.
poly-l-alanine. It was suggested that the reduced
e€ectiveness of the recycled catalyst stems from
degradation of the catalyst via hydrolysis. The
increased stability of poly-l-leucine under the
reaction conditions was ascribed to its greater
steric bulk. In contrast to these results Lantos and
Novak52 found that they could re-use poly-l-leu-
cine (the polymer length was not reported) six
times and observed no erosion in either yield or ee.
. The physical form of the polyamino acid catalysts
changes under the triphasic conditions; the cata-
lyst swells to form a gel. A number of authors
recommend an initial swelling stage to the reac-
tion, prior to addition of the substrate, in which
the catalyst, oxidant, base and solvents are stirred
for a number of hours.52
. In early work little attention was paid to the nat-
ure of the initiator employed in the polymerisation
reaction used to generate the catalyst (vide supra),
with some workers employing water52 and others
amine nucleophiles.13 However, in 1990 Itsuno
and co-workers35 examined the use of a cross-linked
aminomethyl polystyrene (CLAMPS) polymer as an
initiator. Polyamino acids generated with this initia-
tor are easier to handle due to the larger particle size
which facilitates rapid and easy ®ltration of the cat-
alyst after the reaction has ®nished.
. As the polyamino acid catalysts are insoluble in
the reaction medium it is possible to ®lter, wash
and then re-use them. Julia and Colonna reported
reduced catalytic ability and asymmetric induction
with recycled poly-l-alanine.51 The reduction in
catalytic ability was less if a longer polymer was
employed or if poly-l-leucine was used instead of
. Generally the oxidant employed is an excess of
aqueous hydrogen peroxide, which has the advan-
tage of being cheap and readily available.
Attempts have been made to extend the metho-
dology to allow the use of hydrogen peroxide

2150
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
adducts which are more easily and safely handled
than aqueous hydrogen peroxide. Roberts et al.
have shown that sodium percarbonate and sodium
perborate can both be used. There is some varia-
tion in ee and reaction time with di€erent oxi-
dants; indeed on some occasions sodium perborate
and sodium percarbonate have proved the more
e€ective.31 A number of attempts have been made
to use tert-butyl hydroperoxide and, although it has
been shown to e€ect the Julia±Colonna reaction,
greatly reduced enantioselectivities are observed.31,49
Scheme 9. Reagents and conditions: aq. H2O2, aq. NaOH, DCM, poly-
l-leucine, 168 h.
In 1997 a modi®ed Julia±Colonna epoxidation protocol
was reported which served to overcome these limita-
tions.64 Under the new ``biphasic conditions'' the reac-
tion is performed in a non-aqueous solvent, using a
water-free source of hydrogen peroxide and a non-
nucleophilic amine base. A range of solvents was sur-
veyed and tetrahydrofuran, 1,2-dimethoxyethane, tert-
butyl methyl ether, dimethylsulphoxide, N,N-dimethyl-
formamide and ethyl acetate were all found to be sui-
table; less e€ective solvents included dichloromethane,
acetonitrile and toluene. The most e€ective base of
those tested was DBU (1,8-diazabicyclo[5.4.0]undec-7-
ene). In order to minimise the amount of water in the
reaction, it was necessary to add the oxidant in the form
of a complex; examples of such complexes include those
formed with urea65 and DABCO (1,4-diazabicy-
clo[2.2.2]octane).66 The ready availability of the urea±
hydrogen peroxide complex (popularised by Heaney
and commonly known as UHP) made this the oxidant
of choice.
Triphasic conditions Ð substrate range
Julia and Colonna employed a relatively narrow range
of substrates; essentially all reported examples were
analogues of chalcone (4). Subsequent workers have
expanded the substrate range somewhat; Ferreira and
Bezuidenhoudt54±58 and Lantos and Novak52 have
reported a number of examples with di€erent aryl sub-
stituents. In a series of papers31,59±62 Roberts et al.
reported the results of a systematic attempt to extend
the substrate range and to ®nd the limits of reactivity.
Recently, in a paper concerned with the development of
copper catalysed epoxy-stannane cross-coupling, Falck
has demonstrated the ®rst example of a vinyl stannane
48 undergoing the Julia±Colonna epoxidation.33 Table 1
illustrates selected results from all these workers.
In addition to chalcone, a range of aryl-substituted
substrates, bearing both electron-donating and electron-
withdrawing groups has been reported. It is possible to
replace either aryl substituent with a heteroaryl ring,
such as pyridine, thiophene or furan. Introduction of
aliphatic chains can result in reduced enantioselectivity,
although tert-butyl and cyclopropyl ketones are parti-
cularly good substrates. A number of sulphur(II) con-
taining compounds have been epoxidised without
oxidation of the sulphur moiety. Roberts et al. have
oxidised a range of substrates containing more than one
double bond and/or carbonyl group clearly mapping
out the selectivities obtainable with such systems.
Under the biphasic conditions reaction times are greatly
reduced; for example the epoxidation of chalcone (4) is
complete in 30 min and the diene 64 is converted to the
mono-epoxide 65 (85% conversion, >95% ee) in 3 h.
Moreover, previously unreactive substrates could be oxi-
dised, for example the methyl ketone 68 was converted
into 69 in 70% yield and 80% ee in 4 h. Table 2 illustrates
the range of substrates that has been epoxidised under the
new conditions.64,67,68
In addition to a wider range of alkyl substituents, the
new conditions may be used to oxidise dienones. Epox-
idation of trisubstituted double bonds is slower than
disubstituted cases, thus it is possible to oxidise 77 selec-
tively. One case of a relatively reactive trisubstituted
enone is the a-tetralone derivative 79.
Biphasic conditions
Despite the range of substrates which can be oxidised
using the Julia±Colonna procedure a number of pro-
blems limit the applicability of the methodology. The
reaction times are long; reactive enones, such as chal-
cone (4), can be epoxidised in 24 h but a signi®cant
number of substrates prove to be essentially unreactive
or are prohibitively slow to react. For example, oxida-
tion of the diene 64 a€orded the epoxide 65 in 78%
yield and >96% ee under the triphasic conditions, but
the reaction took 72 h. In particular, enolisable sub-
strates, such as the iso-propyl 66 and methyl 68 ketones
were found to be epoxidised very slowly or not at all
(Scheme 9).62 Furthermore, additional equivalents of
aqueous hydrogen peroxide must be added at regular
intervals. The triphasic conditions also limit the poten-
tial range of substrates to which the methodology may
be applied, i.e. to those compounds which are stable in
the presence of nucleophilic, aqueous base.
Even under these accelerated conditions various classes
of substrate remain unreactive; thus far, attempts to
epoxidise cinnamate esters and vinyl chlorides have
proved unsuccessful,67 whilst an a,b-unsaturated nitro
compound has been readily epoxidised but with no
enantioselectivity. Unlike the triphasic case the poly-
amino acid catalyst appears not to swell appreciably
under the biphasic conditions, instead the polymer
adopts the appearance of a paste. Interestingly, it has
been found that the rate and enantioselectivity of the
reaction are maximised if the poly-l-leucine catalyst is
®rst activated by stirring in a mixture of aqueous
sodium hydroxide and an organic solvent, then ®ltered,
washed and dried before use. The exact e€ect of this
activation procedure is currently unclear.72

M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Table 2. Substrates epoxidised under biphasic conditionsa
2151
Substrate
Product
Method
a
Yield %
85
ee %
>95
Ref
[64]
4
5
a
a
a
a
a
b
a
a
c
100
70
81
91
76
80
57
70
60
97
80
[69]
[68]
[70]
[68]
[68]
[67]
[67]
[67]
[71]
66
68
70
72
42
64
75
77
79
67
69
71
73
43
74
76
78
80
>98
89
94
90
86
92
86
a
Reagents and conditions: a. UHP, DBU, CLAMPS-poly-l-leucine, THF; b. UHP, DBU, CLAMPS-poly-d-leucine, THF; c. UHP, DBU,
CLAMPS-poly-l-leucine, EtOAc.
Noting that under the biphasic conditions the poly-l-
leucine does not swell appreciably it has been postulated
that the modi®ed Julia±Colonna epoxidation might be
applicable to a ®xed-bed reactor approach to epoxide
synthesis. In an initial small-scale study the epoxidation
of chalcone (4) was found to be most e€ective on a col-
umn of CLAMPS-poly-l-leucine and DABCO-hydro-
gen peroxide using tert-butyl methyl ether containing
DBU (0.5% v/v) as the eluent. A residence time of
20 min ensured the generation of epoxychalcone (5) in
>97% conversion and >98% ee.36
This section is not intended to be an exhaustive review of
the chemistry of a,b-epoxy-carbonyl compounds. It con-
centrates on transformations reported on the products of
Julia±Colonna epoxidation reactions rather than the sub-
strate class in general. The chemistry can be classi®ed into
two main areas, namely reaction at the carbonyl group
and opening or rearrangement of the epoxide. Often the
latter transformations are performed after manipulation
of the ketone moiety, i.e. a typical approach is to elaborate
the carbonyl group prior to opening the epoxide ring. This
re¯ects the relative paucity of regioselective epoxide ring
openings that may be accomplished at the ketone oxida-
tion level. A third class of reaction, currently limited to a
single example, involves carbon±carbon bond formation
via an epoxy-stannane.63
Elaboration of the epoxides
Epoxides generated using the Julia±Colonna reaction
have been elaborated into a number of optically active
structures of biological interest, including Diltiazem,68
the Taxol1 side chain,68 (+)-clausenamide36 and a
number of ¯avonoid derivatives.54±58
Reaction at the carbonyl group. The three main transfor-
mations of the carbonyl moiety are Baeyer±Villiger oxi-
dation, addition of alkyl metal reagents and reduction.

2152
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Flisak, Lantos et al. have reported the Baeyer±Villiger
oxidation of a number of epoxychalcones prepared by
Julia±Colonna oxidation of the corresponding enones.73
Oxidation was performed using m-CPBA (3-chloro-
peroxybenzoic acid) in dichloromethane and a€orded the
corresponding aryl esters in 57±80% yield and 87±>99%
ee after recrystallisation (e.g. 81 is produced from 5 in
74% yield and >99% ee); as expected the oxidation
does not a€ect the optical purity of the epoxide. The
methodology was subsequently extended by Roberts and
co-workers to include the readily available tert-butyl
ketones; for example 82 is generated from 43 in 94%
yield.68 Thus, it is possible to convert epoxy ketones into
either acid or base labile esters maximising the eciency of
the methodology for the synthesis of complex molecules.
borohydride in ether a€orded the corresponding ery-
thro-con®gured secondary alcohols 86±88 with good
stereocontrol and quantitative yields (Scheme 11).
Transformation of the epoxide. Relatively little work
has been reported on the selective intermolecular open-
ing of the epoxide moiety of homochiral a,b-epoxy
ketones. In a series of papers Ferreira has investigated
the conversion of protected, polyoxygenated epoxy-
chalcones, generated via Julia±Colonna epoxidation,
into ¯avonoids. Hydrogenolysis of epoxychalcones gen-
erates the a-hydroxydihydro-chalcone derivatives with
little or no loss of optical purity.56 For example, 29,
available in 84% ee from the corresponding chalcone,
can be converted into 89 (ee 76%) by catalytic
hydrogenolysis using palladium on barium sulphate
(Scheme 12).
Initial attempts to deprotect the MOM group of 29 and
to cyclise directly onto the epoxide moiety using the
phenol nucleophile led to low optical and chemical
yields of the desired product 92.55 These problems were
overcome by the development of a two step protocol,58
the ®rst step consisting of opening the epoxide with
concomitant deprotection of the MOM group using
benzylthiol in the presence of tin tetrachloride. For
example, treatment of 29 under these conditions a€or-
ded a mixture of the two diastereomeric sulphides 90
and 91 in a ratio of ca. 2.3:1 and an overall yield of
93%. Subsequent cyclisation of this mixture using a
thiophilic Lewis acid, silver tetra¯uoroborate, generated
a mixture of the separable diastereomeric dihydro-
¯avonoids 92 and 93 in an overall yield of 71% and a
ratio 92:93 of 79:21 (Scheme 13).
81
82
Roberts has also examined the addition of alkyl metal
reagents to the ketone. Alkyl lithium, organocerium and
Grignard reagents have been studied with the latter
giving the best diastereoselectivities.74,75 For example,
epoxychalcone 5 was treated with n-butyl and methyl
Grignard reagents to a€ord the corresponding tertiary
alcohols 83 and 84 in reasonable yields and excellent
diastereoselectivities (Scheme 10).
Stereoselective hydride reduction of the ketone has been
studied by Colonna et al.76 It was discovered that
treatment of epoxychalcones 5, 23 and 85 with zinc
Roberts has also prepared cyclic derivatives by reaction
of the epoxides with internal nucleophiles; amino chal-
cone 70 is readily converted to the epoxide 71 under the
biphasic conditions (Table 2). Subsequent cyclisation
proceeds smoothly with inversion to a€ord the chiral
tetrahydroquinolone 94 in 52% yield over two steps and
with >98% ee (Scheme 14).70
Scheme 10. Reagents and conditions: i. n-BuMgI, diethyl ether, 78ꢀC,
60%; ii. MeMgI, diethyl ether, 78ꢀC, 89%.
Scheme 11. Reagents and conditions: Zn(BH4)2, Et2O, 0ꢀC, 100%.
Scheme 12. Reagents and conditions: H2, Pd/BaSO4, MeOH, 88%.

M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
2153
Scheme 13. Reagents and conditions: i. BnSH, SnCl4, DCM, 20 to 0ꢀC, 93%; ii. AgBF4, DCM, 71%.
Scheme 14. Reagents and conditions: n-butanol, 85ꢀC, 64%.
Scheme 15. Reagents and conditions: o-aminothiophenol, PhMe, 110ꢀC,
90%.
The only examples of selective, intermolecular opening
of the epoxide moiety have been performed at the car-
boxylate oxidation state. In an ecient synthesis of the
calcium channel blocker Diltiazem, Roberts' group uti-
lised Inoue's epoxide opening77,78 which proceeds with
retention of con®guration to convert epoxide 95 into 96
in 90% yield (Scheme 15).68
Scheme 16. Reagents and conditions: i. LiOH, H2, MeOH, 95%; ii.
MeO2C(CH2)2SH, NaOMe, THF, MeOH; iii. NaOH, H2O; iv. HCl,
H2O, 76%.
In their synthesis of SK&F 104353 (98) Lantos and
Novak converted the ester 97 into a carboxylate salt
then used a thiolate to selectively attack the b-carbon of
the epoxide, with inversion of stereochemistry (Scheme
16).52
electrophiles mediated by copper (I) sulphide.63 For
example, epoxide 49, obtained in >99% ee from vinyl
stannane 48, was coupled with phenyl chloro-
thionoformate to a€ord thionoester 101 in good yield
(Scheme 18).
Roberts has also examined chemistry of the epoxide
after addition to the carbonyl group; the tertiary alco-
hols generated by the previously described Grignard
additions can be isomerised to the corresponding sec-
ondary alcohols via a silyl chloride-mediated Payne
rearrangement; for example treatment of 84 with tert-
butyldimethylsilyl chloride (TBDMS-Cl) and imidazole
in DMF a€orded the alcohol 99 in 69% yield. In addi-
tion, the epoxide was opened by reaction with tin tetra-
chloride at 78ꢀC to a€ord the chloro-diol 100 with
retention of con®guration (Scheme 17).74
Mechanism of polyleucine-catalysed epoxidation reactions
A discussion of the mechanistic investigations into
polyamino acid-catalysed epoxidation of electron-poor
alkenes is beyond the scope of this review.79,80 However,
in the light of the catalytic activity of homogeneous
poly-l-leucine36 it is now possible to make selected
``mutations'' of the peptide to investigate the mode of
action of the polymer in detail. We are currently active
in this area.
Carbon±carbon bond formation. Falck has recently
reported a cross-coupling of epoxy-stannanes and reactive

2154
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Scheme 17. Reagents and conditions: i. TBDMS-Cl, imidazole, DMF, 69%; ii. SnCl4, DCM, 78ꢀC, 86%.
17. Bougauchi, M.; Watanabe, S.; Arai, T.; Sasai, H.; Shiba-
saki, M. J. Am. Chem. Soc. 1997, 119, 2329±2330.
18. Elston, C. L.; Jackson, R. F. W.; MacDonald, S. J. F.;
Murray, P. J. Angew. Chem. Int. Ed. Engl. 1997, 36, 410±412.
19. Enders, D.; Zhu, J.; Kramps, L. Liebigs Ann./Receuil 1997,
1101±1113.
20. Wang, Z.-X.; Shi, Y. J. Org. Chem. 1997, 62, 8622±8623.
21. Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1998, 39,
1599±1602.
Scheme 18. Reagents and conditions: PhOC(ˆS)Cl, Cu2S, THF, 60ꢀC,
89%.
Conclusion
22. Fuchs, F. Ber. 1922, 55, 189.
23. Baird, W.; Parry, E. G.; Robinson, S. Brit. Patent 646033,
1950.
24. Oya, M.; Katakai, R.; Nakai, H.; Iwakura, Y. Chem. Lett.
1973, 1143±1144.
25. Katakai, R.; Iizuka, Y. J. Org. Chem. 1985, 50, 715±716.
26. Daly, W. H.; Poche, D. Tetrahedron Lett. 1988, 29, 5859±
5862.
The Julia±Colonna oxidation has emerged as a useful
methodology for the preparation of optically active
epoxides from a,b-unsaturated ketones. Given the com-
mercial availability of the active catalyst81 it is clear that
further examples of the use of this simple protocol will
appear. However the application of polyamino acid
catalysis to other stereoselective oxidations and, indeed,
to completely di€erent processes must await a clear
understanding of the ordering of the reactants on the
chiral surface of the catalyst.
27. Leuchs, H. Ber. 1906, 39, 857±861.
28. Leuchs, H.; Geiger, W. Ber. 1908, 41, 1721±1726.
29. Woodward, R. B.; Schramm, C. H. J. Am. Chem. Soc.
1947, 69, 1551±1552.
30. Coleman, D. J. Chem. Soc. 1950, 3222±3229.
31. Lasterra-Sanchez, M. E.; Felfer, U.; Mayon, P.; Roberts,
S. M.; Thornton, S. R.; Todd, C. J. J. Chem. Soc., Perkin
Trans. 1 1996, 343±348.
References
1. Wong, C. H.; Whitesides, G. M. Enzymes in Organic
Synthesis; Pergamon: Oxford, 1994.
2. Drauz, K.; Waldmann, H., Eds. Enzyme Catalysts in
Organic Synthesis; VCH: Weinheim, 1995.
3. Faber, K. Biotransformations in Organic Chemistry;
Springer-Verlag: Berlin, 1996.
4. For some recent examples, see Roberts, S. M. J. Chem. Soc.,
Perkin Trans 1 1998, 157±169.
5. de Raadt, A.; Griengl, H.; Petsch, M.; Plachota, P.; Schoo,
N.; Weber, H.; Braunegg, G.; Kopper, I.; Kreiner, M.; Zeiser,
A.; Kieslich, K. Tetrahedron: Asymmetry 1996, 7, 467±472.
6. de Raadt, A.; Griengl, H.; Petsch, M.; Plachota, P.; Schoo,
N.; Weber, H.; Braunegg, G.; Kopper, I.; Kreiner, M.; Zeiser,
A. Tetrahedron: Asymmetry 1996, 7, 473±490.
32. For reviews on polymerisation of NCAs, see Sekiguchi, H.
Pure Appl. Chem. 1981, 53, 1689±1714 and ref. 33.
33. Kricheldorf, H. R. In Comprehensive Polymer Science;
Allen, G.; Bevington, J. C., Eds.; Pergamon: Oxford, 1989;
Vol. 3, pp 531±551.
34. Ban®, S.; Colonna, S.; Molinari, H.; Julia, S.; Guixer, J.
Tetrahedron 1984, 40, 5207±5211.
35. Itsuno, S.; Sakakura, M.; Ito, K. J. Org. Chem. 1990, 55,
6047±6049.
36. Cappi, M. W.; Chen, W.-P.; Flood, R. W.; Liao, Y.-W.;
Roberts, S. M.; Skidmore, J.; Smith, J. A.; Williamson, N. M.
Chem. Commun. 1998, 1159±1160.
37. Hirayama, C.; Ihara, H.; Tanaka, K. J. Chromat. 1988,
450, 271±276.
7. de Raadt, A.; Griengl, H.; Petsch, M.; Plachota, P.; Schoo,
N.; Weber, H.; Braunegg, G.; Kopper, I.; Kreiner, M.; Zeiser,
A. Tetrahedron: Asymmetry 1996, 7, 491±496.
8. See, for example, Dingler, C.; Ladner, W.; Krei, G. A.;
Cooper, B.; Hauer, B. Pesticide Science 1996, 46, 33±35.
9. Hudlicky, T.; Thorpe, A. J. Chem. Commun. 1996, 1993±
2000.
38. Sinibaldi, M.; Catellani, L.; Federici, F.; Messina, A. J.
Liq. Chromat. 1993, 16, 2977±2992.
39. Shyu, S. S.; Kuroyanagi, Y.; Seno, M. Colloid Polym. Sci.
1988, 266, 587±593.
40. Kuroyanagi, Y.; Kim, E.; Kenmochi, M.; Ui, K.;
Kageyama, H.; Nakamura, M.; Takeda, A.; Shioya, N. J.
Appl. Biomater. 1992, 3, 153±161.
10. Banwell, M.; De Savi, C.; Watson, K. Chem. Commun.
1998, 1189±1190.
41. Akabori, S.; Sakurai, S.; Izumi, Y.; Fujii, Y. Nature 1956,
178, 323±324.
11. See, for example, Mahmoudian, M.; Michael, A. Appl.
Microbiol. Biotechnol. 1992, 37, 23±27 and ref. 12.
12. Mahmoudian, M.; Michael, A. Appl. Microbiol. Bio-
technol. 1992, 37, 28±31.
42. Alper, H., Hamel, N. J. Chem. Soc., Chem. Commun. 1990,
135±136.
43. Abe, S.; Nonaka, T.; Fuchigami, T. J. Am. Chem. Soc.
1983, 105, 3630±3632.
13. Julia, S.; Masana, J.; Vega, J. C. Angew. Chem. Int. Ed.
Engl. 1980, 19, 929±931.
44. Nonaka, T.; Abe, S.; Fuchigami, T. Bull. Chem. Soc. Jpn.
1983, 56, 2778±2783.
14. For recent complementary reviews on this subject, see Ebra-
him, S.; Wills, M. Tetrahedron: Asymmetry 1997, 8, 3163±3173
and ref. 15.
45. Abe, S.; Nonaka, T. Chem. Lett. 1983, 1541±1542.
46. Abe, S.; Fuchigami, T.; Nonaka, T. Chem. Lett. 1983,
1033±1036.
15. Pu, L. Tetrahedron: Asymmetry 1998, 9, 1457±1477.
16. For other recently reported methods for asymmetric
epoxidation of enones, see Baccin, C.; Gusso, A.; Pinna, F.;
Strukul, G. Organometallics 1995, 14, 1161±1167 and refs. 17±21.
47. Komori, T.; Nonaka, T. J. Am. Chem. Soc. 1983, 105,
5690±5691.
48. Komori, T.; Nonaka, T. J. Am. Chem. Soc. 1984, 106,
2656±2659.

M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
2155
49. Julia, S.; Guixer, J.; Masana, J.; Rocas, J.; Colonna, S.;
Annuziata, R.; Molinari, H. J. Chem. Soc., Perkin Trans. 1
1982, 1317±1324.
50. Julia, S.; Masana, J.; Rocas, J.; Colonna, S.; Annunziata,
R.; Molinari, H. Anal. Quim. 1983, 79, 102±104.
51. Colonna, S.; Molinari, H.; Ban®, S.; Julia, S.; Masana, J.;
Alvarez, A. Tetrahedron 1983, 39, 1635±1641.
52. Flisak, J. R.; Gombatz, K. J.; Holmes, M. M.; Jarmas, A.
A.; Lantos, I.; Mendelson, W. L.; Novack, V. J.; Remich, J. J.;
Snyder, L. J. Org. Chem. 1993, 58, 6247±6254.
53. As the catalysts are prepared by an uncontrolled poly-
merisation reaction they are not homogeneous, instead a statis-
tical distribution of chain lengths is present, dependent on the
ratio of initiator to NCA employed in the polymerisation proce-
dure. It is common in the literature for a polymer to be referred
to as being of length n, where n is in fact the mean length.
54. Bezuidenhoudt, B. C. B.; Swanepoel, A.; Augustyn, J. A.
N.; Ferreira, D. Tetrahedron Lett. 1987, 28, 4857±4860.
55. Augustyn, J. A. N.; Bezuidenhoudt, B. C. B.; Ferreira, D.
Tetrahedron 1990, 46, 2651±2660.
Hursthouse; M. B.; Nugent; T. C.; Pulido; R.; Roberts; S. M.;
Wu, L. E. Chem. Commun. 1997, 739±740.
65. Cooper; M. S.; Heaney; H.; Newbold; A. J.; Sanderson;
W. R. Synlett 1990, 533±535.
66. Oswald, A. A.; Guertin, D. L. J. Org. Chem. 1963, 28,
651±657.
67. Allen, J. V.; Cappi, M. W.; Kary, P. D.; Roberts, S. M.;
Williamson, N. M.; Wu, L. E. J. Chem. Soc., Perkin Trans. 1
1997, 3297±3298.
68. Adger, B. M.; Barkley, J. V.; Bergeron, S.; Cappi, M. W.;
Flowerdew, B. E.; Jackson, M. P.; McCague, R.; Nugent, T. C.;
Roberts, S. M. J. Chem. Soc., Perkin Trans. 1 1997, 3501±3507.
69. Mukherjee, S., Roberts, S. M., unpublished results.
70. Chen, W.-P.; Egar, A. L.; Hursthouse, M. B.; Abdul
Malik, K. M.; Mathews, J. E.; Roberts, S. M. Tetrahedron
Lett. 1998, 39, 8495±8498.
71. Bentley, P. A., Roberts, S. M., unpublished results.
72. Allen, J. V.; Bergeron, S.; Griths, M. J.; Mukherjee, S.;
Roberts, S. M.; Williamson, N. M.; Wu, L. E. J. Chem. Soc.,
Perkin Trans. 1 1998, 3171±3179.
56. Augustyn, J. A. N.; Bezuidenhoudt, B. C. B.; Swanepoel,
A.; Ferreira, D. Tetrahedron 1990, 46, 4429±4442.
57. van Rensburg, H.; van Heerden, P. S.; Bezuidenhoudt, B.
C. B.; Ferreira, D. Chem. Commun. 1996, 2747±2748.
58. van Rensburg, H.; van Heerden, P. S.; Bezuidenhoudt, B.
C. B.; Ferreira, D. Tetrahedron 1997, 53, 14141±14152.
59. Lasterra-Sanchez, M. E.; Roberts, S. M. J. Chem. Soc.,
Perkin Trans. 1 1995, 1467±1468.
60. Kroutil, W.; Mayon, P.; Lasterra-Sanchez, M. E.;
Maddrell, S. J.; Roberts, S. M.; Thornton, S. R.; Todd, C. J.;
Tuter, M. Chem. Commun. 1996, 845±846.
61. Kroutil; W.; Mayon; P.; Lasterra-Sanchez; M. E.;
Maddrell; S. J.; Roberts; S. M.; Thornton; S. R.; Todd; C. J.;
Tuter, M. Chem. Commun. 1996, 2495.
62. Kroutil, W.; Lasterra-Sanchez, M. E.; Maddrell, S. J.;
Mayon, P.; Morgan, P.; Roberts, S. M.; Thornton, S. R.;
Todd, C. J.; Tuter, M. J. Chem. Soc., Perkin Trans. 1 1996,
2837±2844.
63. Falck; J. R.; Bhatt; R. K.; Reddy; K. M.; Ye; J. H. Synlett
1997, 481-482.
64. Bentley, P. A.; Bergeron; S.; Cappi; M. W.; Hibbs; D. E.;
73. Baures, P. W.; Eggleston, D. S.; Flisak, J. R.; Gombatz,
K.; Lantos, I.; Mendelson, W.; Remich, J. J. Tetrahedron Lett.
1990, 31, 6501±6504.
74. Gillmore, A. T.; Roberts, S. M.; Hursthouse, M. B.; Abdul
Malik, K. M. Tetrahedron Lett. 1998, 39, 3315±3318.
75. Carde; L.; Davies; H.; Pulido; R.; Roberts; S. M., unpub-
lished results.
76. Ban®, S.; Colonna, S.; Molinari, H.; Julia, S. Synth. Com-
mun. 1983, 13, 901±904.
77. Hashiyama, T.; Inoue, H.; Konda, M.; Takeda, M. J.
Chem. Soc., Perkin Trans. 1 1984, 1725±1732.
78. Schwartz, A.; Madan, P. B.; Mohacsi, E.; O'Brien, J. P.;
Todaro, L. J.; Co€en, D. L. J. Org. Chem. 1992, 57, 851±856.
79. For studies on the physical properties of the polyamino
acid catalyst, see Valencia-Parera, G.; Solans-Marsa, C.; Reig-
Isart, F.; Garcia-Anton, J. M. J. Colloid Interf. Sci. 1986, 114,
140±148 and ref. 80.
80. Bentley, P. A.; Kroutil, W.; Littlechild, J. A.; Roberts, S.
M. Chirality 1997, 9, 198±202.
81. Available from UFC-Pharma, Science Park, Manchester,
M15 6SY, UK.

2156
M. J. Porter et al. / Bioorg. Med. Chem. 7 (1999) 2145±2156
Biography
Mike Porter was born in Harrogate, Yorkshire, UK on 2nd August 1970.
From 1988 he studied at Merton College, the University of Oxford,
where he received his B. A. degree in 1992. He remained in Oxford for his
D. Phil. studies, investigating the biosynthesis of penicillins under the
supervision of Professor Sir Jack Baldwin. In 1996 he was awarded a
NATO Postdoctoral Fellowship to work with Professor Philip Magnus
at the University of Texas at Austin, researching methods for the synth-
esis of Amaryllidaceae alkaloids. In February 1998 he returned to the
United Kingdom to take up a post as a Senior Research Assistant at the
University of Liverpool.
Stan Roberts was born in Hale, Cheshire, UK in 1945. After leaving
school he joined ICI as a technician and studied for two years part-time
to gain a Higher National Certi®cate in Chemistry. Two years of full time
study at the University of Salford, UK led to the award of a ®rst-class
Honours degree. After studying polyhaloaromatic chemistry at Salford
under the supervision of Professor Hans Suschitzky the award of Ph.D
was made in 1969. Post doctoral periods with Professors Dreiding (Zur-
ich) and Woodward (Harvard) followed. In 1972, Dr. Roberts was
appointed lecturer in biochemistry and organic chemistry at Salford
University, where he developed his interest in prostaglandin synthesis.
Industry then beckoned again and, shortly after being appointed Reader,
he left Salford in 1980 to join Glaxo (Greenford) as Head of Chemical
Research. During the next six years Stan led teams involved in anti-
microbial and anti-cancer chemotherapy. In 1986 he joined Exeter Uni-
versity as Professor of Organic Chemistry and set up a team in the area of
biotransformations. Finally, in 1995 he joined Liverpool University as
Heath Harrison Professor of Organic Chemistry, and has become fasci-
nated, over this latter period, in biomimetic catalysis.
John Skidmore was born in Maccles®eld, Cheshire, UK on January 13th
1972. He studied for his B. A. at St. Peters' College, the University of
Oxford, graduating in 1994. Remaining at Oxford, he carried out
research for his D. Phil. under the supervision of Dr. J. M. Peach, inves-
tigating the synthesis and properties of spin-labelled calcium ionophores.
In 1997 he moved to the University of Liverpool where he currently holds
the position of Senior Research Assistant, working with Professor S. M.
Roberts on the chemistry of polyamino acids.