Synthesis and inhibitory action on HMG-CoA synthase of racemic and optically active oxetan-2-ones (β-Lactones)

BIOORGANIC &

MEDICINAL

CHEMISTRY

Bioorganic & Medicinal Chemistry 6 (1998) 1255±1272

Synthesis and Inhibitory Action on HMG-CoA Synthase of

Racemic and Optically Active Oxetan-2-ones (ꢀ-Lactones)

Daniel Romo,^a,* Paul H. M. Harrison,^b,yStephen I. Jenkins,^b,{

R. William Riddoch,^b,xKaapjoo Park,^a,{Hong Woon Yang,^aCunxiang Zhao^a

and Gerard D. Wright^c

^aDepartment of Chemistry, Texas A&M University, P.O. Box 300012, College Station, TX 77843-3012, USA

^bDepartment of Chemistry, McMaster University, 1280 Main Street West, Hamilton, Ontario L8S 4M1, Canada

^cDepartment of Biochemistry, McMaster University, 1280 Main Street West, Hamilton, Ontario L8S 4M1, Canada

Received 25 March 1998; accepted 8 June 1998

AbstractÐA homologous series of both C3-unsubstituted and C3-methyl substituted oxetan-2-ones (b-lactones) was

investigated as potential inhibitors of yeast 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) synthase. Several

reported methods for racemic b-lactone synthesis were studied for preparation of the target series. In addition, a novel

aluminum-based Lewis acid obtained by combination of Et₂AlCl with (1R,2R)-2-[(diphenyl)hydroxymethyl] cyclo-

hexan-1-ol was studied for the asymmetric [2+2] cycloaddition of aldehydes and trimethylsilylketene. This Lewis acid

exhibited good reactivity but variable enantioselectivity (22-85% ee). In in vitro assays using both native and recom-

binant HMG-CoA synthase from Saccharomyces cerevisiae, oxetan-2-ones mono-substituted at C4 with linear alkyl

chains gave IC₅₀s that decreased monotonically with chain length up to 10 carbons and then rose rapidly for longer

chains. The trans isomers of 3-methyl-4-alkyl-oxetan-2-ones showed a similar trend but had 1.3- to 5.6-fold lower

IC₅₀s. The results imply a substantial hydrophobic pocket in this enzyme that interacts with both C-3 and C-4 sub-

Introduction

lanosterol into cholesterol and other sterols.³Inhibition

of the many enzymes along this biosynthetic pathway thus

oers the potential to regulate sterol biosynthesis; indeed

the commercial statin drugs such as lovastatin and ¯uva-

statin, which inhibit HMG-CoA reductase,⁴have proved

to be highly eective at reducing sterol levels in humans.⁵

Inhibitors of this pathway can also act as potent anti-

microbial agents, especially for cells such as fungi which

possess active sterol biosynthetic machinery.⁶HMG-CoA

synthase is also involved in ketogenesis.^1a,6

The enzyme 3-hydroxy-3-methylglutaryl-coenzyme A

(HMG-CoA) synthase¹catalyzes the formal aldol con-

densation of 1 mol of acetyl-coenzyme A (AcCoA) and

1 mol of acetoacetyl-coenzyme A (AcAcCoA). The

reaction product, (3S)-HMG-CoA, is the universal pre-

cursor of terpenes and steroids,²being converted by the

action of HMG-CoA reductase to mevalonate and

thence via isopentenyl diphosphate and dimethylallyl

diphosphate through to geranyl (GPP), farnesyl (FPP),

or other diphosphates.^2,3GPP and FPP act as progeni-

tors for mono- and sesquiterpenes, respectively, while in

the sterol sequence, FPP is converted via squalene and

Key words: b-Lactones; HMG-CoA synthase; inhibition;

asymmetric synthesis; chiral Lewis acid.

*Corresponding author. Tel.: 409 845 9571; Fax: 409 862 7963;

E-mail: romo@chemvx.chem.tamu.edu

yCorresponding author regarding biological studies. Tel.: (905)

525-9140 ext. 27290; Fax: (905) 522-2509;

E-mail: pharriso@ mcmail.cis.mcmaster.ca

{Major contributors to synthetic aspects of this work.

xMajor contributor to biological aspects of this work.

PII: S0968-0896(98)00114-X

1256

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

HMG-CoA synthase has been isolated from a range of

species, including yeast,⁷and the livers of ox,⁸chicken⁹

and man.¹⁰The human^10band avian liver⁹enzymes have

been cloned and puri®ed to homogeneity;¹⁰however no

cloning work has been undertaken on the yeast enzyme.

Detailed kinetic studies have been performed on the

synthase from a range of these species to determine the

sequence of events during catalysis of this condensation

reaction; the enzyme exhibits BiBi ping-pong kinetics

with AcCoA binding ®rst to form an acetyl-enzyme

covalent intermediate. The latter is isolable,⁹and the

point of attachment has been shown to be a conserved

cysteine residue (Cys-129 in both man^10aand avian

liver⁹). Binding of AcAc-CoA, via an interaction with

His-264 (human enzyme residue),¹¹then occurs; binding

to free enzyme results in an abortive complex, account-

ing for the observed substrate inhibition, while binding

to the acetyl±enzyme intermediate leads to aldol con-

densation and hence to a CoA-HMG-enzyme covalent

intermediate.¹²Hydrolysis of the latter generates

HMG-CoA and free synthase ready for another cata-

lytic cycle.

F-244.²⁰This result established that the lactone ring is

the only structural component required for irreversible

inhibition, and suggests that the ring substituents play a

role solely in guiding the inhibitor into the enzyme

active site. In order to explore the role of substituents at

C3 and C4 of the oxetan-2-one ring, we thus prepared a

series of b-lactones bearing either hydrogen or methyl at

C-3 and alkyl chains of increasing length at C4, as well

as several b-lactones containing variations on these

substitution patterns. The biological activity of these

simple F-244 analogs in in vitro assays with HMG-CoA

synthase is described herein.

As well as the statins, there exists an array of other

natural products which act as inhibitors of the sterol

pathway. However, aside from inhibition by one of the

substrates (AcAcCoA) and other primary metabolic

CoA esters such as succinyl-CoA,¹³the sole known

metabolite which is speci®cally active against HMG-

CoA synthase is the antibiotic F-244 (1a),¹⁴also known

as L-659,699^14b,15and 1233A.¹⁶This b-lactone, which

has been isolated from Scopulariopsis, Fusarium, and

Cephalosporium sp.,^14±16along with a range of synthetic

analogues which possess similar structures,¹⁷has been

shown to be an irreversible inactivator of the enzyme in

in vitro assays,^{10a,14,15,17,18}as well as to be a potent

inhibitor in in vivo systems.^14,15,17,19The observation

that both the methyl ester 1b and O-methyl ether 1c of

F-244^{14a,15b,17d,18b}exhibited little if any reduction in

inhibition indicates that neither the carboxy nor

hydroxy groups of F-244 participate in strong hydro-

gen-bond donor interactions with the enzyme active site,

and therefore suggests that hydrophobic interactions are

important in determining the anity of F-244 and

analogs for the enzyme. Thus a variety of non-polar

substituents have been attached to the oxetan-2-one

(b-lactone) C4-carbon, including 2-arylethyl^17b,c,fand

decyl.^17gHowever, a systematic investigation of the

eect of the length of alkyl groups attached at this site

has not been conducted to the best of our knowledge.

A range of methods for preparing racemic b-lactones

has been reported and we investigated several of these

routes to prepare our target series of compounds. This

enabled a comparative study of these methods. The

most ecient routes are those which involve a single-pot

transformation such as the [2+2] cycloaddition of

ketenes and aldehydes,²¹aldol-lactonizations,²²and

most recently the tandem Mukaiyama aldol-lactoniza-

tion (TMAL) reaction.²³

We were also interested in determining if the absolute

stereochemistry of simple 4-alkyloxetan-2-ones would

have a bearing on the inhibition of HMG-CoA syntha-

se.^17aThe [2+2] cycloaddition of ketenes and aldehydes

is one of the most concise routes to b-lactones. The ®rst

example of this reaction was reported by Staudinger in

1911,^21ahowever more recently, Zaitseva showed that

trimethylsilyl ketenes can participate in this reaction

with Lewis acid catalysis.^21bRecent studies with tri-

methylsilylketene and aldehydes have shown that high

diastereoselectivity can be obtained with bulky, achiral

Lewis acids.²⁴In addition, the recent use of chiral Lewis

acids has allowed access to optically active b-lactones.²⁵

However, a general chiral Lewis acid that provides high

reactivity and enantioselectivity with a broad range of

aldehyde substrates has yet to be found. We have also

been studying this cycloaddition reaction as a concise

entry into optically active b-lactones. In this regard, we

have recently reported that the TiCl₂-TADDOL catalyst

exhibits low to good enantioselectivity (9±80% ee) in

this reaction.²⁶Herein, we disclose a novel aluminum

In an earlier report, we showed that b-butyrolactone (2),

a simple b-lactone analogue of F-244, was an irrever-

sible inhibitor of the synthase, with an inactivation rate

constant (k_inact) similar to F-244, but with much weaker

binding anity: K_Iwas ca. 10⁵-fold higher than that for

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1257

Lewis acid for this cycloaddition. While this Lewis acid

provides the highest enantioselectivities reported to date

for a-unbranched, aliphatic aldehydes, it exhibits enantio-

selectivity which appears to be substrate dependent.

A chiral promoter based on the chiral diol, (1R,2R)-2-

[(diphenyl)hydroxymethyl] cyclohexan-1-ol (13),²⁹was

studied for the [2+2] cycloaddition of trimethylsilyl-

ketene and aldehydes. Diol 13 was readily prepared in

two steps from b-ketoester 11 employing the modi®ed

Noyori reduction conditions reported by Taber in 67%

yield and 64% ee.³⁰Addition of excess phenyl magne-

sium bromide to the trans-b-hydroxyester 12 aorded

the diol 13 in 72% yield. After a single recrystallization,

the diol 13 could be enriched to 99% ee. Treatment of

this diol, which was sparingly soluble in toluene with a

slight de®ciency of Et₂AlCl, resulted in a homogeneous

solution of the presumed Lewis acid 14. The structure of

the Lewis acid has not been determined and the repre-

sentation provided is meant only to show the presumed

Lewis acid composition (Scheme 4).

Results

Preparation of oxetan-2-ones (ꢀ-lactones). Several of

the reported methods for the preparation of racemic

b-lactones were investigated. The desired b-lactones

were prepared by the following methods: (1) the one-pot

aldol-lactonization procedure reported initially by Dan-

heiser^22aand subsequently modi®ed by Schick^22b

(Scheme 1, Table 1); (2) an aldol followed by a separate

cyclization of the hydroxy acid derivative by the meth-

ods of Wemple²⁷and Masamune²⁸(Scheme 1, Table 1);

(3) the tandem Mukaiyama-aldol lactonization (Scheme

2, Table 2); or (4) the catalyzed [2+2] cycloaddition of

Use of 20 mol of this promoter led to a variety of opti-

cally active a-silyl-b-lactones 10 with enantioselectivities

ranging from 22±85%. In some cases, the b-lactones

were more readily isolated after desilylation which gave

the monosubstituted b-lactones 7 (Table 4).

aldehydes and ketene using both BF₃OEt₂²¹(Scheme 3,

.

Table 3) and a new, chiral aluminum-based Lewis acid

(Scheme 5, Table 4).

Scheme 1.

Table 1. Synthesis of racemic b-lactones via one-pot or two-step aldol-lactonizations

Entry

b-Lactone

R¹

R²

Ester

trans/cis Ratio (7)

Method^a

% Yield

1

2

3

4

5

6

7

7a

7b

7c

7d

C₇H₁₅

C₈H₁₇

C₉H₁₉

Me

H

3b

3d

3a

3b

3d

3b

>19:1^b

Ð

A

B

C

A

B

C

32

39

56^c

61^c

37

H

Ð

Me

>19:1^b

1.2:1

35

66^c

^aMethod A: Danheiser one-pot aldol-lactonization (ref 22a) but the aldehyde was added at 100 ^ꢀC. Method B: Schick one-pot aldol-

lactonization (ref 22b). Method C: Two-step procedure involving aldol reaction according to the method of Wemple (ref 27) and

subsequent lactonization by the method of Masamune (ref 28).

1

^btrans/cis Ratio determined by 200 MHz H NMR on the puri®ed b-lactones.

^cYields are for the two-step aldol-lactonization sequence.

1258

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

Scheme 2.

Table 2. Synthesis of racemic b-lactones via the tandem Mukaiyama aldol-lactonization (TMAL) reaction

Entry

b-Lactone

R¹

R²

Ketene acetal

trans/cis Ratio (7)^a

% Yield

1

7e^b

7f^b

7g^b

7c

Ph(CH₂)₂

t-Bu

H

8b

8a

8b

8c

8d

8c

Ð

66

42

65

59

34

36

39

60

35

38

41

40

32

36

47

16

2

H

3

C₇H₁₅

C₉H₁₉

C₄H₉

H

Ð

4

H

5

7h

7i

Me

>19/1

39/1

6

C₅H₁₁

C₆H₁₃

C₇H₁₅

C₈H₁₇

C₉H₁₉

C₁₀H₂₁

7

7j

8

7a^b

7k

7d

9

>19/1

<1/19

>19/1

10

11

12

13

14

15

16

17

7l

7m

7n

C

₁₁H₂₃

₁₃H₂₇

₁₇H₃₅

7o

7p^b

7q^b

7r^b

p-NO₂Ph

TBSO(CH₂)₄

c-C₆H₁₁

1

^aRatios estimated or determined by H NMR (200 MHz) or by GC, respectively.

^bThese b-lactones have been described previously (refs. 23b and c).

Table 3. Synthesis of racemic b-lactones via [2+2] cycloaddi-

tions of silylketenes and aldehydes

Entry

b-Lactone

Silylketene

R¹

% Yield (7)^a

1

2

3

4

5

6

7

8

7s

10t

7u

7g

7b

7c

9a

9b

9a

C₄H₉

C₅H₁₁

C₆H₁₃

C₇H₁₅

C₈H₁₇

C₉H₁₉

74

94^b

74

80

78

84

82

79

7v

7w

C₁₀H₂₁

C₁₁H₂₃

^aYield is for the two steps of [2+2] cycloaddition and desilylation.

^bYield is for a-silyl-b-lactone 10t as a mixture of trans/cis dia-

stereomers (1/4.7).

Scheme 3.

Isolation of HMG-CoA synthase. Native HMG-CoA

synthase was isolated from yeast as previously descri-

bed,²⁰and puri®ed through the hydroxylapatite step to

GenBank (accession no. Z50178), bearing HindIII and

NdeI restriction endonuclease sites at the 3⁰and 5⁰ter-

mini, respectively, were prepared. These primers were

then used to amplify the putative HMG-CoA synthase

gene from a genomic S. cerevisiae DNA preparation

by PCR. The resulting product (1.5 kb) was digested

with HindIII and NdeI, purifed and cloned into the

1

a speci®c activity of 0.13 mmol min ¹mg (units/mg).

For the recombinant synthase, oligonucleotide primers

complementary to the Saccharomyces cerevisiae puta-

tive HMG-CoA synthase sequence obtained from

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1259

exchange (Q-Sepharose) and size exclusion (S-200)

chromatography to a ®nal speci®c activity of 0.85 units/

mg protein (Table 5).

IC₅₀values for ꢀ-lactones. Initially, we screened the b-

lactones 7a±d, 7g±o and 7s±w for inhibitory activity

against HMG-CoA synthase using an in vitro assay

against either native or recombinant enzyme. While IC₅₀

is a poor measure of activity for irreversible inhibitors,

this value does provide a comparison between inactiva-

tors determined with the same sample under identical

conditions, as was the case in this study. Since IC₅₀is

much more readily obtained, compared to the more

fundamental constants of inhibition, k_inactand K_I, it was

practical to obtain this value for all the inhibitors. All

assays were performed by measuring the percentage of

enzyme inhibited, relative to an inhibitor-free control,

during ®ve minutes of preincubation of inhibitor with

enzyme, followed by addition of adequate substrates to

prevent further inactivation of the enzyme, as demon-

strated by linear substrate consumption during the assay

period (Fig. 1). The IC₅₀values were then obtained by

plotting % inhibition versus concentration of inhibitor

for a series of assays performed at inhibitor concentra-

tions above, below and around the estimated IC₅₀, fol-

lowed by extrapolation of the IC₅₀value (Fig. 2). For F-

244, as well as several of the synthetic b-lactones studied,

the IC₅₀values were determined repetitively in several

independent trials, and statistical analysis gave estimates

of errors associated with these determinations. The

results are shown in Table 6.

Scheme 4.

corresponding restriction sites on the plasmid pET-22b.

After transforming E. coli BL21 (DE3) cells with this

plasmid, plasmid DNA from three clones was prepared.

The entire 1500 bases of interest were sequenced using

primers for the T7 promoter and terminator sequences

present in the plasmid, and an additional primer which

was complementary to a central region of the gene. The

results show that in all three clones there are three silent

mutations relative to the original database sequence, as

well as one non-silent mutation (nt 223 from T to C)

corresponding to codon 75 being changed from serine to

proline. Induction of expression of HMG-CoA synthase

[MW 47.5 kD] in one of these clones was achieved by

IPTG as assessed by SDS±PAGE of total cell protein.

After cell lysis, the crude cell-free extract possessed a

speci®c activity of synthase of 0.04 units/mg, compared

to uninduced and insert-free controls which were inactive.

Several other synthetic b-lactones were also investigated

for inhibitory ecacy against the synthase. The results

for IC₅₀in these cases are shown in Table 7.

Discussion

Synthetic studies

Synthesis of racemic ꢀ-lactones. Danheiser and Nowick

reported the ®rst one-pot condensation of enolates with

aldehydes to give variously substituted b-lactones.^22a

They employed thiophenyl ester enolates (4a or 4b) and

the presumed, initially formed aldolates 5 sponta-

neously lactonized on warming (Scheme 1). However,

The recombinant enzyme was partially puri®ed by a

sequence of ammonium sulfate precipitation, anion

Scheme 5.

1260

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

Table 4. Synthesis of optically active b-lactones via [2+2] cycloadditions of trimethylsilyl-ketene and aldehydes

Entry

b-Lactone

R¹

cis/trans (10)

% Yield (7)^a

% ee

Con®g. (C4)

1

2

3

4

5

6

7

8

9

7x^b

7s

c-C₆H₁₁

n-Bu

t-Bu

>99:1^c

Ð

>99:1^c

>19:1^g

83^d

86^d

no rxn

82^d

45^d

60

84^e

85^e

Ð

28^e

75^h

36^h

46^j

22^j

56^j

S^f

R^f

7y^b

Ð

7z

Ph

PhCH₂

ND

Rⁱ

S^f

7aa

7e^b

7bb

7cc

7dd

PhCH₂CH₂

TSBO(CH₂)₅

CH₂=CH(CH₂)₇

(CH₃CH₂)₂CH

55

71

ND

46

^aYield of 7 for two steps (not including recovered aldehyde).

^bThese b-lactones have been described previously (refs 23b±c and 26).

^cDetermined on the crude reaction mixtures by GC.

^dYield of 10.

^e% ee's are for 10 and were determined by chiral GC(TBS-cyclodextrin, ref 31).

^fAbsolute con®guration were assigned after alcoholysis of 7 to the corresponding known, b-hydroxy esters and comparison of optical

rotation data.

^gDetermined on the crude reaction mixtures by ¹H NMR (200 or 300 MHz).

^h% ee's are for 7 and were determined by chiral HPLC (Chiralcel OD).

ⁱAbsolute con®guration was assigned by comparison of optical rotation data to the known b-lactone (ref 32).

^j% ee's of 7 determined by chiral GC (TBS-cyclodextrin).

Table 5. Puri®cation of recombinant yeast HMG-CoA synthase

Step

Vol. (mL)

Total protein (mg)

Speci®c activity (units/mg)^a

Yield (%)

Fold puri®cation

Cell-free Extract

(NH₄)₂SO₄(25±40%)

Q-Sepharose

20

5

30

50

3

0.04

0.10

0.5

100

104

63

1

2.5

12

21

10

S-200

1

0.85

35

^aOne unit=1 mmol/min.

condensation of the enolate derived from phenyl thio-

acetate with hexanal, octanal or decanal failed to give

any a-unsubstituted-b-lactones in our hands. In an

analogous one-step lactonization, Schick and co-work-

ers^22bused the enolates of phenyl esters (4c or 4d).

However, this method also failed to provide the expected

b-lactone 7t by condensation of the enolate of phenyl

acetate with hexanal. In contrast, addition of either

octanal or decanal ( 100 ^ꢀC) to either the phenyl or

thiophenyl propionate (3b or 3d) derived enolates (4)

gave lactones 7a and 7d in yields of 32±39% (Table 1,

entries 1±2 and 5±6) with a trans/cis ratio of >19/1

(200 MHz ¹H NMR). The lower reaction temperature

was crucial to obtain the desired b-lactones. Thus, both

methods provide good selectivity but low yields in the

preparation of 3,4-disubstituted oxetan-2-ones, and

appear to be unsuitable for the preparation of C4-

monosubstituted-b-lactones.

Figure 1. Inhibition of HMG-CoA Synthase by b-Lactone 7c.

Enzyme was incubated with the concentrations of 7c shown for

5 min, then acetoacetyl-CoA (AcAcCoA) was added (time 0).

Absorbance of AcAcCoA at 303 nm was monitored for 150 s

(providing a control for AcAcCoA thiolase), then acetyl-CoA

was added, and the reaction was monitored until time 300 s.

An alternative approach to C-4-monosubstituted oxe-

tan-2-ones involves isolation of the aldol adduct 6, fol-

lowed by a separate lactonization (Scheme 1). Wemple

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1261

2, Table 2).²³Condensation of the TBS(tert-butyldi-

methylsilyl)ketene acetal 8c with a variety of aldehydes

using ZnCl₂as a promoter, aorded the corresponding

3-methyl-4-alkyloxetan-2-ones in yields of 32±47% with

high diastereoselectivities (trans/cis, >19/1). As pre-

viously described,^23b,cthis method could also be used to

access C4-monsubstituted-b-lactones unavailable by

other one-pot aldol-lactonization methods described

above. We have also previously reported that higher

yields of b-lactones using propionate and especially

acetate derived thiopyridylketene acetals could be

achieved by employing the triethylsilyl (TES) ketene

acetals 8b and 8d. This was further demonstrated in this

study (i.e. entries 1, 3±4 and 8, Table 2).^23c

The Lewis acid catalyzed [2+2] cycloaddition of tri-

methylsilylketene and aldehydes ®rst reported by Zait-

seva²⁷was also investigated. Trimethylsilylketene³³was

prepared and its [2+2] cycloaddition with a range of

saturated aldehydes aorded monosubstituted b-lac-

tones 7b±c, 7g, 7s±7w in good yields (74±84%, Table 3)

after a subsequent desilylation step of the a-silyl-b-lac-

tones 10. This two step process provides a concise route

to racemic, C4-monosubstituted b-lactones. Limitations

to this method are the somewhat tedious preparation of

ethoxyacetylene³⁴required for the synthesis of tri-

methylsilylketene and the fact that it is a two step

process. In summary, for trans-3-methyl-4-alkyloxetan-

2-ones, the methods of Danheiser and Schick are com-

parable in both yield and diastereoselectivity, removing

the requirement for diastereomer separation inherent

in the method due to Wemple. The TMAL reaction

provides a concise, and highly diastereoselective route

to both C4-monosubstituted and C3, C4-disubstituted-

b-lactones. In most cases, these methods are com-

plementary to one another since for example, the

single-pot aldol-lactonizations proceed eciently with

ketone substrates but less so with aldehyde substrates.

In contrast, the TMAL reaction provides good yields

with aldehyde substrates including the use of acetic acid

derived ketene acetals leading to C4-monosubstituted b-

lactones. In some cases, comparable diastereoselectivity

but higher yields can be obtained via the TMAL reac-

tion in comparison to the methods of Danheiser and

Schick (cf. entry 1 and 2, Table 1 and entry 8, Table 2).

In addition the one step TMAL reaction gives

comparable yields to the two-step aldol-lactonization

(Wemple/Masamune) for the synthesis of C4-mono-

substituted b-lactones (cf. entry 4, Table 1 vs entry 4,

Table 2). However, the [2+2] cycloaddition of aldehydes

and ketenes provides the highest overall yields for C4-

monosubstituted b-lactones but is a two step procedure.

Figure 2. Representative inhibition curves for HMG-CoA syn-

thase. Data from Figure 1 and similar plots for b-lactones 7b,

7g, 7u, 7v, and 7w are re-plotted to show percent inhibition as a

function of b-lactone concentration. IC₅₀values were deter-

mined by extrapolation.

condensed the enolate of phenyl thioacetate with

aldehydes at low temperature, then quenched the result-

ing aldol alkoxide 5 to give b-hydroxythiophenylesters

6.²⁷Masamune cyclized b-hydroxythiophenylesters using

Hg(O₃SCH₃)₂to give various substituted b-lactones.²⁸

This sequential route furnished racemic 4-octyl-(7b) and

4-nonyl-oxetan-2-one (7c) from thiophenyl acetate and

nonanal, or decanal in two steps in 56 and 61% overall

yield, respectively (entries 3 and 4, Table 1). However,

the corresponding condensation of phenyl thiopropio-

nate with decanal followed by lactonization gave the

lactone 7d in a ca. 1/1.2 ratio of cis/trans diastereomers

and 66% overall yield. Thus, while this method provides

a route to both C3, C4-disubstituted as well as C4-

monsubstituted-b-lactones, undesirable features include

the use of mercury salts, the low diastereoselectivity

with the propionate enolate, and the fact that it is a two

step process. We next investigated the use of the 2-thio-

pyridylketene acetals 8 for the single pot preparation of

b-lactones, according to a method ®rst reported by

Hirai and further developed in our laboratories (Scheme

Synthesis of optically active ꢀ-lactones. In the design

stages of the aluminum complex 14, we envisioned a tet-

rahedral complex (A or B) formed upon complexation of

1262

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

Table 6. IC₅₀values for inhibition of HMG-CoA synthase with b-lactones of varying chain length^a

n

R=H

Enz^b

R=Me

Enz^b

Entry

b-Lactone

IC₅₀(mM)^c

Entry

b-Lactone

IC₅₀(mM)^c

Ð

0

1

2

15

2

N

N,R

R

500 (5)

2,000 (1)40^d

200 (4); 150 (4)^e

Ð

13

14

15

16

17

18

19

20

21

22

23

24

Ð

7h

7i

Ð

R

N

R

N

R

Ð

3

7s

7t

88 (5)

26 (1)

10 (1)

4

Ð

N

5

7u

7g

7b

7c

7v

7w

Ð

13 (1)

9.0 (8)

7j

6

N

7a

7k

7d

7l

1.6 (1)

1.3 (1)

1.0 (1)0.28^d

0.28 (1)

2.7 (1)

280 (8)

7

N

6.3 (5)

8

N,R

R

2.0 (2)0.17^f

1.4 (2)0.08^g

2.0 (1)

9

10

12

16

10

11

12

7m

7n

7o

Ð

^aDetermined for 5 min inactivation.

^bEnz=enzyme source, N=native, R=recombinant.

^cValues in parentheses are errors in the last decimal place from a single inhibition plot; errors from statistical analysis of several repeat

experiments are shown as one standard deviation.

^dAverage and standard deviation () from ®ve determinations.

^eFirst value is for racemic, second for b-lactone (4R)-7s (85% ee).

^fAverage and standard deviation () from six determinations.

^gAs for note f, from four determinations.

Table 7. IC₅₀values for inactivation of HMG-CoA synthase by b-lactones with varying substitution^a

Entry

b-Lactone

R₁

R₂

Con®g.

IC₅₀(mM)^b

1

2

3

4

5

6

7

8

9

10t

7x^d

C₅H₁₁

c-C₆H₁₁

SiMe₂t-Bu

RS-trans/cis^c

RS

4-S

>2,000

350 (5)

184 (4)

H

7x^d

7e^d

7bb^d

7p^d

16^d

17^d

(F-244) 1

Ph(CH₂)₂

p-NO₂Ph

CH₂CH(OH)CH₃

(CH₂)₅

H

RS

RS-trans

1800 (50)

1500 (50)

>>2,000

283 (15)

Me

H

RS-cis

3R,4R,2⁰S

achiral

4-R-trans

2300 (50)^e

0.01000.0006^f

Ð

^aDetermined for 5 min inactivation.

^bValues in parentheses are errors in the last decimal place from a single inhibition plot; errors from statistical analysis of several repeat

experiments are shown as one standard deviation.

^cThis was a 4.7/1 ratio of cis/trans b-lactones.

^dThese b-lactones have been described previously, see refs 22a, 23c, and d.

^eReversible inhibitor.

^fAverage and standard deviation () from six determinations.

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1263

Figure 3. Proposed conformations for complexation of aldehydes with monomeric Lewis acid 14.

Inhibition studies

the (monomeric) Lewis acid with an aldehyde (Fig. 3).

In this situation, assuming the chlorine adopts a pseu-

doequatorial position upon complexation, the aldehyde

would be in a pseudoaxial position. Two conformations

are likely, A and B, and it was not clear at the outset

which would predominate. These conformations avoid

1,3-diaxial type interactions between the formyl hydro-

gen or R group of the aldehyde and the axial phenyl

group. However, it appeared that conformation A

would provide low facial selectivity while conformation

B may lead to high facial selectivity since the pseudoax-

ial phenyl eectively blocks the Re face of the aldehyde.

The possibility of formyl C-H±O hydrogen bonding to

oxygen, recently proposed by Corey to explain a num-

ber of enantioselective transformations,³⁵found in con-

former B was also considered as a possible control

element to favor conformation B. However, analysis of

the complex by ²⁷Al NMR indicates that the Lewis acid

14 in toluene-d₈may exist as two species, one of which is

a dimer, making a rationalization of the results more

dicult.³⁶

Although the native HMG-CoA synthase from yeast is

available, as we have described previously,²⁰diculties

in the puri®cation of this enzyme to homogeneity, cou-

pled with the availability of the sequence of the yeast

genome, led us to investigate

a genetic cloning

approach. Cloning of the putative S. cerevisiae HMG-

CoA synthase gene, which has ca. 40% sequence simi-

larity to the known human sequence, into an E. coli host

under control of the T7 promoter, followed by induc-

tion gives HMG-CoA synthase of de®ned sequence,

which is fully catalytically competent and resembles the

native enzyme in many respects. The putative role of

this gene in yeast is thus con®rmed. The mutation of

Ser-75 to Pro-75 which was detected in this study has

little eect on the competency of the protein. The IC₅₀

values for several of the b-lactone inhibitors (2, 7s, 7c,

and 7v) were determined with both recombinant and

native enzyme preparations, and in each case the values

obtained were found to be essentially identical. The

recombinant and native HMG-CoA synthases behave

similarly during puri®cation; however the higher initial

speci®c activity in the recombinant case allows for a

purer preparation overall (ca. 25% pure based on titra-

tion with F-244). The IC₅₀for both b-lactone 2 and F-

244 (1) was determined at various stages of puri®cation

The Lewis acid 14 derived from diol 11 and Et₂AlCl led

to good reactivity but variability in the enantioselec-

tivity for the [2+2] cycloaddition of trimethylsilyl

ketene and various aldehydes (Table 4). Interestingly,

the same facial selectivity is observed with most alde-

hydes for which the absolute con®gurations have been

determined (entries 1, 2 and 5 but not entry 6, Table 4)

and is consistent with attack on the Si face of the alde-

hyde on conformer B. It should be noted that the dif-

ference in absolute con®guration for the b-lactones

derived from cyclohexanecarboxaldehyde and n-butanal

(cf. entry 1 and 2, Table 4) is due to a change in

substituent priority rather than a change in facial

selectivity of the aldehyde. While these are the highest

enantioselectivities reported to date for a-unbranched,

aliphatic aldehydes (i.e. entry 2, Table 4) in this [2+2]

cycloaddition, the reactivity and generality must obviously

be improved before this becomes a practical procedure for

the preparation of optically active b-lactones.

Figure 4. Proposed HMG-CoA synthase binding/active site for

b-lactone inhibitors.

1264

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

of both enzyme preparations, and was found to be

invariant with enzyme purity, indicating that there are

no interferences with inhibition through processes such

as protein±protein interactions, and allowing further

studies of the inhibition process with the partially pur-

i®ed material. An overall 21-fold puri®cation of the

recombinant synthase was obtained in three steps; com-

pletion of the puri®cation is currently in progress.

These results are consistent with data from a range of

previous studies that have shown, for example, that

F-244 methyl ester, F-244 O-methyl ether and several

oxetan-2-ones possessing phenalkyl groups at C4 exhibit

little loss of potency compared to the parent antibiotic.

However, the present set of data rigorously establish the

approximate size requirements for the hydrophobic

cavity at the active site of HMG-CoA synthase. Thus,

Nature has selected the optimal chain length in the nat-

ural inhibitor F-244 (1).

Although the amino acid sequence for yeast synthase is

substantially dierent than that for human enzyme, the

results of this study con®rm that the yeast synthase is a

valid model for the human protein. Thus, the IC₅₀value

for F-244 was found to be essentially identical to values

reported previously.

The results with several lactones which are reported in

Table 7 demonstrate that the enzyme is not highly tol-

erant of major changes in structure of the inhibitor at

either C3 or C4. Thus, replacement of a C3-methyl by

tert-butyldimethylsilyl gives an extremely poor inhibitor

(cf. entry 16, Table 6 with entry 1, Table 7). Likewise, a

p-nitrophenyl substituent at C4 resulted in only weak

activity. The results with b-lactones 7e and 7bb show

that a phenethyl substituent at C4 gives poor activity

against yeast HMG-CoA synthase. This result is in

contrast to those of Hashizume with rat liver enzyme.^17b

This observation suggests a dierence between the two

proteins for these inhibitors, whereas F-244 shows simi-

lar activity with both enzymes. That such a dierence

could be exploited in species selective inhibition of the

synthase is under active investigation. In contrast, 2⁰-

hydroxypropyl or cyclohexyl substituents at C4 give

inhibitory activities which approximate those of the

straight-chain compounds which possess the same chain

length (cf. entry 2 and 7, Table 7 with entry 3, Table 6).

In the case of the cyclohexyl substituent, enantiomeri-

cally enriched material exhibited an IC₅₀of approxi-

mately one-half that of the racemic compound (entries

2 and 3, Table 7), again showing the binding to be

enantioselective.

Two sets of b-lactones bearing either hydrogen or

methyl at C-3 and alkyl chains from one to 18 carbons

at C-4 were evaluated for inhibitory ecacy. The results

show a roughly monotonic decrease in IC₅₀with C-4

chain length until an optimum length of 10 carbons is

reached. Thereafter, the IC₅₀rises sharply again. The

results are most simply interpreted in terms of a sub-

stantial hydrophobic binding pocket on the enzyme;

increasing alkyl chain length gives rise to tighter bind-

ing, and hence to reduced IC₅₀(Fig. 4). Alkyl chains of

greater than 10 carbons `over¯ow' out of the pocket,

perhaps into a hydrophilic region or into the aqueous

environment, making this arrangement disfavoured.

Comparison of the C3-methyl series with the C4-mono-

substituted b-lactones as inhibitors shows a consistent

improved anity for the enzyme when the 3-methyl

group is present, by a factor of between 1.3 and 5.6.

This result implies that a substituent at C-3 can also

provide a substantial hydrophobic interaction with the

enzyme. Although this study was performed largely with

racemic lactones, due to ease of synthesis, lactone 7a in

enantiomerically enriched form was investigated, to

establish that the inhibitors were acting enantioselec-

tively. Comparison of the results with racemic and

chiral materials shows that the enantiomer correspond-

ing in con®guration to F-244 is most eective, suggest-

ing that the compound is binding in the synthase active

site in a similar manner to F-244. It is interesting to note

from Fig. 2 that b-lactones in the odd n series abolished

synthase activity completely, but that even n b-lactones

appeared to permit some residual enzyme activity even

at high concentrations. The source of this eect is cur-

rently unclear, and investigations of this phenomenon

are ongoing. The acute loss of activity for very long

chains, as well as the discrimination for one enantiomer

of 7a are both inconsistent with inhibitory activity

depending on aggregation of the hydrophobic b-lac-

tones. Thus, the critical micellar concentration of long-

chain b-lactones is expected to decrease with increasing

chain length to 18 carbons.

Conclusion

For the synthesis of racemic b-lactones from alde-

hydes, the aldol-lactonization methods of Danheiser

and Schick are comparable in terms of yields and

selectivity for the preparation of C3, C4-disubstituted

b-lactones. However, these methods do not allow

access to C4-monosubstituted-b-lactones. The TMAL

reaction provides a single step route to C4-monsub-

stituted-b-lactones using readily available thiopyridyl

ketene acetals, however the two step procedure invol-

ving [2+2] cycloaddition of aldehydes and silylketenes

provides higher overall yields. In addition, the TMAL

reaction provides good yields and high diastereoselec-

tivity for the synthesis of C3,C4-disubstituted-b-lac-

tones. A new, aluminum chiral Lewis acid has been

applied to the [2+2] cycloaddition of aldehydes and

trimethylsilylketene and provides good reactivity but

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1265

variable enantioselectivity dependent on the substrate

structure. Both 4-alkyl- and 3-methyl-4-alkyl-oxetan-2-

ones exhibit a logical trend in inhibition of HMG-CoA

synthase with potencies which are dependent on the

length of the C4 substituent; a chain of 10±11 carbons

gives maximal inhibition. The results can be explained by

hydrophobic interactions with the synthase active site.

Varian VXR-300 (75 MHz) or XL 200E (50 MHz) and

chemical shifts are reported in ppm using CDCl₃(d

77.0) as internal reference. ²⁷Al NMR spectra were

recorded on a Varian XL 200 (52 MHz) in toluene-d₈

and chemical shifts are reported in ppm relative to

Al(acac)₂(d 0.0) as external reference.

Methods A,^22aB,^22band C^22,28(Table 1) were per-

formed according to the published procedures except

that in method A, the aldehyde was added at 100 ^ꢀC.

The characterization of b-lactones prepared by these

methods is described below.

Experimental

General synthetic procedures

All reactions were performed in ¯ame-dried glassware

under a positive pressure of nitrogen and magnetically

stirred unless otherwise indicated. Toluene was distilled

from sodium immediately prior to use. Methylene

chloride was distilled from CaH₂immediately prior to

use. Tetrahydrofuran and diethyl ether were distilled

from sodium-benzophenone ketyl radical immediately

prior to use. Acetonitrile was heated at re¯ux overnight

with P₂O₅, distilled, and kept under argon over acti-

vated 3 A molecular sieves. Commercially available

aldehydes were purchased from Aldrich Chemical Co.

and other aldehydes were prepared from the corre-

sponding alcohols by oxidation. All aldehydes were

distilled (Kugelrohr distillation) or puri®ed by ¯ash

chromatography (silica gel) immediately prior to use.

Commercial-grade reagents (Aldrich) and solvents

(Caledon) were used without further puri®cation except

as indicated below. Flash column chromatography was

accomplished using Merck 60 (230±400 mesh) silica gel

or Baxter S/P Silica Gel 60 A (230±400 mesh ASTM).

Thin layer chromatography was performed using Merck

60 F-254 plates, and b-lactones were visualized through

treatment with a solution of phosphomolybdic acid in

10% sulfuric acid (5 g/100 mL).

General procedure for the [2+2] cycloaddition of

aldehydes and trimethylsilyl ketene and desilylation as

described for 4-nonyl-oxetan-2-one (7c). A solution of

decanal (0.403 g, 2.62 mmol) and trimethylsilylketene

9a³⁴(0.334 g, 2.91 mmol, 1.1 equiv) dissolved in

dichloromethane (5 mL) was cooled to 0 ^ꢀC. While

stirring, boron tri¯uoride etherate (0.4 mL of a 0.1 M

solution in dichloromethane) was added dropwise

until IR indicated loss of trimethylsilylketene starting

material (ca. 50 min). The reaction was quenched with

two drops of water and the solution was ®ltered through

a short plug of anhydrous sodium sulfate. After solvent

removal on the rotary evaporator, the crude oil was

dissolved in 7 mL of acetonitrile. To this solution,

.

®nely crushed KF 2H₂O (0.496 g, 5.23 mmol, 2.0

equiv) was added and the mixture was vigorously

stirred for 20 min. The resulting mixture was ®ltered

through ¯orisil with ether washing and the solvent was

removed on a rotary evaporator. The crude oil was

puri®ed via ¯ash chromatography (EtOAc/hexanes,

10/90) to give 432 mg of lactone 7c (84%) as a color-

1

less oil. IR (thin ®lm) 2928, 1831 cm

;

¹H NMR

(200 MHz, CDCl₃) d 4.49 (m, 1H), 3.51 (dd, J=16.2,

5.7 Hz, 1H), 3.05 (dd, J=16.2, 4.3 Hz, 1H), 1.76 (m,

2H), 1.28 (m, 14H), 0.88 (t, J=6.9 Hz, 3H); ¹³C NMR

(50 MHz, CDCl₃) d 168.3, 71.5, 42.8, 34.6, 29.4 (2C),

29.2, 24.8, 22.6, 14.1.

Mass spectra were obtained on a VG analytical 70S

high-resolution, double focusing, sectored (EB) mass

spectrometer at the Center for Chemical Characteriza-

tion and Analysis (Texas A&M) or on a VG Analytical

ZAB-E mass spectrometer (McMaster). Enantiomeric

excess was determined by GC (Hewlett±Packard 5880A

gas chromatograph) analysis using a TBS-b-cyclodex-

trin column³¹or HPLC (RAININ SD-200 with

DYNAMAX UV-C detector) analysis using a Chiralcel

OD column. IR spectra were recorded on a Nicolet

Impact 410DSP (Texas A&M) or on a BIO-RAD

FTS-40 FT-IR spectrometer (McMaster). ¹H NMR

spectra were recorded on a Varian VXR-300 (300 MHz)

or 200E spectrometer (200 MHz) (Texas A&M) or Bru-

ker AC-200, AC-300, or DRX-500 spectrometers

(McMaster) and chemical shifts are reported in ppm

using tetramethylsilane (d 0.0) or CHCl₃(d 7.26) as

internal reference. ¹³C NMR spectra were recorded on a

4-Butyloxetan-2-one (7s). The [2+2] cycloaddition of

pentanal (0.197 g, 1.74 mmol) with trimethylsilylketene

9a (0.215 g, 1.82 mmol) and then desilylation with

potassium ¯uoride dihydrate (0.322 g, 3.41 mmol) was

performed according to the general procedure. Work-up

followed by puri®cation aorded 187 mg (74%) of pro-

1

duct 7s as a colorless oil. IR (thin ®lm) 2931, 1828 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.50 (m, 1H), 3.50 (dd,

J=16.2, 5.7 Hz, 1H), 3.04 (dd, J=16.2, 4.3 Hz, 1H),

1.86 (m, 2H), 1.30 (m, 8H), 0.88 (t, J=6.9 Hz, 3H); ¹³C

NMR (50 MHz, CDCl₃) d 168.3, 71.2, 42.8, 34.6, 31.5,

28.7, 24.7, 22.4, 13.9.

4-Hexyloxetan-2-one (7u). The [2+2] cycloaddition of

heptanal (0.197 g, 1.74 mmol) with trimethylsilylketene

1266

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

9a (0.215 g, 1.82 mmol) and then desilylation with

potassium ¯uoride dihydrate (0.322 g, 3.41 mmol) was

performed according to the general procedure. Work-up

product 7w as a colorless oil. IR (thin ®lm) 2926,

1831 cm ¹; ¹H NMR (200 MHz, CDCl₃) d 4.50 (m, 1H),

3.50 (dd, J=16.2, 5.7 Hz, 1H), 3.05 (dd, J=16.2,

4.3 Hz, 1H), 1.79 (m, 2H), 1.25 (m, 18H), 0.88 (t,

J=6.9 Hz, 3H); ¹³C NMR (50 MHz, CDCl₃) d 168.4,

71.3, 42.9, 34.7, 31.9, 29.6, 29.4 (3C), 29.3, 29.2, 24.9,

22.7, 14.1.

followed by puri®cation aorded 187 mg (74%) of pro-

1

duct 7u as a colorless oil. IR (thin ®lm) 2931, 1828 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.50 (m, 1H), 3.50 (dd,

J=16.2, 5.7 Hz, 1H), 3.04 (dd, J=16.2, 4.3 Hz, 1H),

1.86 (m, 2H), 1.30 (m, 8H), 0.88 (t, J=6.9 Hz, 3H); ¹³C

NMR (50 MHz, CDCl₃) d 168.3, 71.2, 42.8, 34.6, 31.5,

28.7, 24.7, 22.4, 13.9.

cis/trans-4-Butyl-3-tert-butyldimethylsilyloxetan-2-one

(10t). The [2+2] cycloaddition of hexanal (0.242 g,

2.4 mmol) with tert-butyldimethylsilyl ketene (0.390 g,

2.5 mmol) was performed according to the general pro-

cedure with the exception that the desilylation was not

performed. Puri®cation (EtOAc/hexanes, 10/90) gave

4-Heptyloxetan-2-one (7g). The [2+2] cycloaddition of

octanal (0.217 g, 1.63 mmol) with trimethylsilylketene 9a

(0.200 g, 1.73 mmol) and then desilylation with potas-

sium ¯uoride dihydrate (0.306 g, 3.24 mmol) was per-

formed according to the general procedure. Work-up

0.561 g (94%) of b-lactone 10t (cis/trans, 4.7/1,

1

200 MHz ¹H NMR). IR (thin ®lm) 2932, 1806 cm

;

followed by puri®cation aorded 214 mg (80%) of pro-

1

cis-10t: ¹H NMR (200 MHz, CDCl₃) d 4.61 (m, 1H),

3.58 (d, J=6.3 Hz, 1H), 1.76 (m, 2H), 1.31 (m, 6H),

0.91 (m, 12H), 0.16 (s, 3H), 0.12 (s, 3H); ¹³C NMR

(50 MHz, CDCl₃) d 172.1, 74.9, 43.8, 34.8, 32.0, 26.8,

duct 7g as a colorless oil. IR (thin ®lm) 2930, 1830 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.50 (m, 1H), 3.50 (dd,

J=16.2, 5.7 Hz, 1H), 3.04 (dd, J=16.2, 4.3 Hz, 1H),

1.80 (m, 2H), 1.27 (m, 10H), 0.88 (t, J=6.9 Hz, 3H); ¹³

C

23.2, 17.3, 14.2,

7.1,

7.3; trans-10t: ¹H NMR

NMR (50 MHz, CDCl₃) d 168.3, 71.4, 42.8, 34.6, 31.5,

28.9, 24.8, 22.4, 14.0.

(200 MHz, CDCl₃) d 4.36 (m, 1H), 3.58 (d, J=4.1 Hz,

1H), 1.76 (m, 2H), 1.31 (m, 6H), 0.91 (m, 12H), 0.16

(s, 3H), 0.12 (s, 3H); ¹³C NMR (50 MHz, CDCl₃) d

172.1, 73.7, 46.2, 36.3, 32.2, 26.8, 23.2, 17.3, 14.2,

7.1, 7.3.

4-Octyloxetan-2-one (7b). The [2+2] cycloaddition of

nonanal (0.319 g, 2.26 mmol) with trimethylsilylketene

9a (0.287 g, 2.52 mmol) and then desilylation with

potassium ¯uoride dihydrate (0.413 g, 4.46 mmol) was

performed according to the general procedure. Work-up

The tandem Mukaiyama aldol-lactonization (TMAL)

reaction was used to prepare the following b-lactones

and was performed as described previously.^23c

followed by puri®cation aorded 317 mg (78%) of pro-

1

duct 7b as a colorless oil. IR (thin ®lm) 2928, 1831 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.48 (m, 1H), 3.49 (dd,

J=16.2, 5.7 Hz, 1H), 3.04 (dd, J=16.2, 4.3 Hz, 1H),

4-Heptyl-3-methyloxetan-2-one (7a). The reaction of

octanal (0.216 g, 1.63 mmol) with ketene acetal 8c

(0.504 g, 1.78 mmol) was performed using 0.533 g

(2.35 mmol) zinc chloride in 10 mL dichloromethane

according to the general procedure. Work-up followed

1.79 (m, 2H), 1.26 (m, 10H), 0.86 (t, J=6.9 Hz, 3H); ¹³

C

NMR (50 MHz, CDCl₃) d 168.3, 71.2, 42.8, 34.6, 31.7,

31.5, 29.3, 29.1, 24.8, 22.6, 14.0.

by puri®cation aorded 7a (108 mg, 39%) as a clear oil.

1

4-Decyloxetan-2-one (7v). The [2+2] cycloaddition of

undecanal (0.640 g, 3.69 mmol) with trimethylsilylketene

9a (0.475 g, 4.14 mmol) and then desilylation with

potassium ¯uoride dihydrate (0.708 g, 7.43 mmol) was

performed according to the general procedure. Work-up

IR (thin ®lm) 2930, 1825 cm

;

¹H NMR (200 MHz,

CDCl₃) d 4.14 (td, J=6.7, 4.0 Hz, 1H), 3.20 (qd, J=7.5,

4.0 Hz, 1H), 1.77 (m, 2H), 1.35 (d, J=7.5 Hz, 3H), 1.26

(m, 10H), 0.87 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.0, 79.5, 50.6, 34.0, 31.5, 31.3, 29.0, 24.6,

22.4, 13.8, 12.4.

followed by puri®cation aorded 651 mg (82%) of pro-

1

duct 7v as a colorless oil. IR (thin ®lm) 2926, 1830 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.50 (m, 1H), 3.50 (dd,

J=16.2, 5.7 Hz, 1H), 3.05 (dd, J=16.2, 4.3 Hz, 1H),

4-Nonyl-3-methyloxetan-2-one (7d). The reaction of

decanal (0.104 g, 0.647 mmol) with ketene acetal 8c

(0.203 g, 0.712 mmol) was performed using 0.203 g

(0.887 mmol) zinc chloride in 5 mL dichloromethane

according to the general procedure. Work-up followed

1.79 (m, 2H), 1.26 (m, 16H), 0.88 (t, J=6.9 Hz, 3H); ¹³

C

NMR (50 MHz, CDCl₃) 168.3, 71.3, 42.9, 34.7, 31.9,

29.4 (3C), 29.3, 29.1, 24.9, 22.6, 14.1.

by puri®cation aorded 7d (46 mg, 35%) as a clear oil.

1

4-Undecyloxetan-2-one (7w). The [2+2] cycloaddition of

dodecanal (0.433 g, 2.32 mmol) with trimethylsilylketene

9a (0.285 g, 2.49 mmol) and then desilylation with

potassium ¯uoride dihydrate (0.428 g, 4.63 mmol) was

performed according to the general procedure. Work-

up followed by puri®cation aorded 413 mg (79%) of

IR (thin ®lm) 2928, 1826 cm

;

¹H NMR (200 MHz,

CDCl₃) d 4.16 (td, J=6.7, 4.0 Hz, 1H), 3.20 (qd, J=7.5,

4.0 Hz, 1H), 1.79 (m, 2H), 1.37 (d, J=7.5 Hz, 3H), 1.26

(m, 14H), 0.87 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.1, 79.6, 50.8, 34.2, 31.9, 31.6, 29.5 (2C),

29.3, 25.0, 22.7, 14.5, 12.6.

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1267

4-Butyl-3-methyloxetan-2-one (7h). The reaction of pen-

tanal (0.125 g, 1.33 mmol) with ketene acetal 8c (0.416 g,

1.46 mmol) was performed using 0.429 g (1.85 mmol)

zinc chloride in 7 mL dichloromethane according to the

general procedure. Work-up followed by puri®cation

CDCl₃) d 4.15 (td, J=6.7, 4.0 Hz, 1H), 3.19 (qd, J=7.5,

4.0 Hz, 1H), 1.79 (m, 2H), 1.34 (d, J=7.5 Hz, 3H), 1.24

(m, 16H), 0.86 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.0, 79.5, 50.6, 34.1, 31.8, 29.5 (2C), 29.4,

29.2, 29.1, 24.9, 22.6, 14.0, 12.4.

aorded 7h (58 mg, 34%) as a clear oil. IR (thin ®lm)

1

2936, 1824 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.14

4-Undecyl-3-methyloxetan-2-one (7m). The reaction of

dodecanal (0.291 g, 1.64 mmol) with ketene acetal 8c

(0.508 g, 1.80 mmol, 1.1 equiv) was performed using

0.649 g (2.93 mmol, 1.6 equiv) zinc chloride in 7 mL

dichloromethane according to the general procedure.

Work-up followed by puri®cation gave 157 mg of

(td, J=6.7, 4.0 Hz, 1H), 3.18 (qd, J=7.5, 4.0 Hz, 1H),

1.78 (m, 2H), 1.32 (d, J=7.5 Hz, 3H), 1.24 (m, 4H), 0.89

(t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz, CDCl₃) d 172.0,

79.5, 50.6, 33.7, 26.9, 22.2, 13.7, 12.4.

4-Pentyl-3-methyloxetan-2-one (7i). The reaction of hex-

anal (0.163 g, 1.62 mmol) with ketene acetal 8c (0.509 g,

1.78 mmol) was performed using 0.479 g (2.12 mmol)

zinc chloride in 10 mL dichloromethane according to the

general procedure. Work-up followed by puri®cation

b-lactone 7m (41%) as a clear oil. IR (thin ®lm) 2929,

1826 cm

1

;

¹H NMR (200 MHz, CDCl₃) d 4.16 (dt,

J=6.7, 4.0 Hz, 1H), 3.20 (dq, J=7.6, 4.0 Hz, 1H), 1.77

(m, 2H), 1.37 (d, J=7.6 Hz, 3H), 1.25 (m, 18H), 0.87 (t,

J=6.7 Hz, 3H); ¹³C NMR (50 MHz, CDCl₃) d 172.1,

79.6, 50.7, 34.1, 31.9, 29.6 (3C), 29.4, 29.3, 29.2, 24.9,

22.7, 14.1, 12.5.

aorded 7i (83 mg, 36%) as a clear oil. IR (thin ®lm)

1

2933, 1825 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.15

(dt, J=6.7, 4.0 Hz, 1H), 3.20 (dq, J=7.5, 4.0 Hz, 1H),

1.79 (m, 2H), 1.36 (d, J=7.5 Hz, 3H), 1.24 (m, 6H), 0.89

(t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz, CDCl₃) d 172.0,

79.5, 50.6, 34.0, 31.3, 24.6, 22.4, 13.8, 12.4.

4-Tridecyl-3-methyloxetan-2-one (7n). The reaction of

tetradecanal (0.270 g, 1.29 mmol) with ketene acetal 8c

(0.406 g, 1.43 mmol) was performed using 0.396 g

(1.74 mmol) zinc chloride in 8 mL dichloromethane

according to the general procedure. Work-up followed

4-Hexyl-3-methyloxetan-2-one (7j). The reaction of hep-

tanal (0.183 g, 1.55 mmol) with ketene acetal 8c (0.488 g,

1.71 mmol) was performed using 0.489 g (2.15 mmol)

zinc chloride in 10 mL dichloromethane according to the

general procedure. Work-up followed by puri®cation

by puri®cation aorded 7n (131 mg, 40%) as a clear oil.

1

IR (thin ®lm) 2923, 1819 cm

;

¹H NMR (200 MHz,

CDCl₃) d 4.14 (td, J=6.7, 4.0 Hz, 1H), 3.19 (qd, J=7.5,

4.0 Hz, 1H), 1.75 (m, 2H), 1.41 (d, J=7.5 Hz, 3H), 1.23

(m, 22H), 0.86 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.0, 79.5, 50.6, 34.1, 31.9, 29.6, 29.4, 29.2,

24.9, 22.6, 14.0, 12.5.

aorded 7j (94 mg, 39%) as a clear oil. IR (thin ®lm)

1

2932, 1827 cm

;

¹H NMR (200 MHz, CDCl₃) d 4.15

(td, J=6.7, 4.0 Hz, 1H), 3.20 (qd, J=7.5, 4.0 Hz, 1H),

1.78 (m, 2H), 1.36 (d, J=7.5 Hz, 3H), 1.27 (m, 8H), 0.87

(t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz, CDCl₃) d 172.0,

76.4, 50.6, 34.1, 31.5, 28.8, 24.9, 22.4, 13.9, 12.4.

4-Heptadecyl-3-methyloxetan-2-one (7o). The reaction of

octadecanal (0.179 g, 0.679 mmol) with ketene acetal 8c

(0.217 g, 0.747 mmol) was performed using 0.231 g

(1.02 mmol) zinc chloride in 5 mL dichloromethane

according to the general procedure. Work-up followed

4-Octyl-3-methyloxetan-2-one (7k). The reaction of

nonanal (0.169 g, 1.19 mmol) with ketene acetal 8c

(0.34 g, 1.21 mmol) was performed using 0.364 g

(1.62 mmol) zinc chloride in 5 mL dichloromethane

according to the general procedure. Work-up followed

by puri®cation aorded 7o (67 mg, 32%) as a clear oil.

1

IR (thin ®lm) 2924, 1818 cm

;

¹H NMR (200 MHz,

CDCl₃) d 4.16 (td, J=6.7, 4.0 Hz, 1H), 3.21 (qd, J=7.5,

4.0 Hz, 1H), 1.75 (m, 2H), 1.38 (d, J=7.5 Hz, 3H), 1.25

(m, 30H), 0.90 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.0, 79.6, 50.7, 34.1, 31.9, 31.6, 29.7, 29.4,

29.2, 24.9, 22.7, 14.1, 12.5.

by puri®cation aorded 7k (110 mg, 35%) as a clear oil.

1

IR (thin ®lm) 2929, 1826 cm

;

¹H NMR (200 MHz,

CDCl₃) d 4.14 (td, J=6.7, 4.0 Hz, 1H), 3.19 (qd, J=7.5,

4.0 Hz, 1H), 1.76 (m, 2H), 1.35 (d, J=7.5 Hz, 3H), 1.24

(m, 12H), 0.85 (t, J=6.9 Hz, 3H); ¹³C NMR (50 MHz,

CDCl₃) d 172.0, 79.4, 50.6, 34.0, 31.7, 29.3, 29.1, 29.0,

24.9, 22.5, 14.0, 12.4.

Preparation of (1R,2R)-2-ethoxycarbonyl cyclohexan-1-

ol (12). This b-hydroxy ester was prepared according to

known procedures³⁰with a few modi®cations. To a

dried metal Parr reactor were added ethyl 2-cyclohex-

anonecarboxylate (10.98 g, 64.5 mmol), CH₂Cl₂(50 mL),

Ru-(R)-BINAP catalyst (4 mL; from 40 mg of (R)-

BINAP) which was prepared by the method of Taber,^30a

and Dowex-50 resin (700 mg; washed with water,

methanol, diethyl ether, and methanol then dried).

Hydrogenation was carried out at 200 psi H₂and 110 ^ꢀC

4-Decyl-3-methyloxetan-2-one (7l). The reaction of

undecanal (0.188 g, 1.04 mmol) with ketene acetal 8c

(0.328 g, 1.14 mmol) was performed using 0.336 g

(1.50 mmol) zinc chloride in 5 mL dichloromethane

according to the general procedure. Work-up followed

by puri®cation aorded 7l (83 mg, 38%) as a clear oil.

;

¹H NMR (200 MHz,

1

IR (thin ®lm) 2928, 1827 cm

1268

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

for 24 h. After cooling, the reaction mixture was ®ltered

through celite and concentrated in vacuo. Flash chro-

matography (EtOAc/Hexanes, 10/90) gave a mixture of

cis-b-hydroxy ester (3.2 g, 29%) and trans-b-hydroxy

ester 12 (7.4 g, 67%) as pale-yellow oils. Spectral data

for the trans-b-hydroxy ester 12 matched that previously

5/95) gave 167 mg (83%) of a-silyl-b-lactone 10x as a

colorless oil. Spectral data for this compound matched

that previously reported:³⁷84% ee; R_f0.35 (EtOAc/

hexanes, 10/90); ¹H NMR (300 MHz, CDCl₃) d 4.21

(dd, J=5.9, 10.5 Hz, 1H), 3.29 (d, J=5.9 Hz, 1H), 0.83±

2.09 (m, 11H), 0.25 (s, 9H); ¹³C NMR (75 MHz, CDCl₃)

d 171.1, 78.0, 45.9, 40.6, 28.9, 28.6, 25.9, 25.1, 25.0,

reported:^30c64% ee; R_f0.19 (EtOAc/hexanes, 20/80);

22

[a]_d

43.1^ꢀ(c 0.58, ether); ¹H NMR (200 MHz,

0.85; IR (thin ®lm) 2935, 2855, 1799, 1453, 1252,

1

CDCl₃) d 4.18 (q, J=7.1 Hz, 2H), 3.77 (ddd, J=4.4,

10.2, 10.2 Hz, 1H), 2.77 (bs, 1H), 2.24 (ddd, J=3.7,

10.1, 12.0 Hz, 1H), 2.04 (m, 2H), 1.75 (m, 2H), 1.30 (m,

4H), 1.25 (t, J=7.1 Hz, 3H).

1127 cm

.

cis-4-Butyl-3-(trimethylsilyl)oxetan-2-one (10s). This a-

silyl-b-lactone was obtained from valeraldehyde (95 mL,

0.892 mmol) and trimethylsilylketene 9a (153 mg,

1.34 mmol) using Lewis acid 14 prepared from Et₂AlCl

(0.1 mL of 1.8 M in toluene, 0.178 mmol) and chiral diol

13 (60.4 mg, 0.214 mmol) according to the general pro-

cedure for asymmetric [2+2] cycloaddition. Flash chro-

matography (EtOAc/hexanes, 5/95) gave 153 mg (86%)

of a-silyl-b-lactone 10s as a colorless oil. Spectral data

for this compound matched that previously reported:²⁶

85% ee; R_f0.36 (EtOAc/hexanes, 10/90); ¹H NMR

(300 MHz, CDCl₃) d 4.57 (m, 1H), 3.33 (d, J=6.0 Hz,

1H), 1.32±1.88 (m, 6H), 0.93 (t, J=7.2 Hz, 3H), 0.22 (s,

Preparation of chiral diol ligand ((1R,2R)-2-[(diphenyl)-

hydroxymethyl] cyclohexan-1-ol) 13. To a solution of

trans-b-hydroxy ester 12 (1.0 g, 5.8 mmol) in anhydrous

diethyl ether (60 mL) was added phenylmagnesium bro-

mide (20.3 mL of 1.0 M in THF, 20.3 mmol) slowly and

gradually at 0 ^ꢀC. After stirring for 1 h at 0 ^ꢀC, the

reaction mixture was heated to re¯ux for 2 h. The mix-

ture was cooled to ambient temperature and quenched

with saturated aqueous NH₄Cl solution. The resulting

precipitates were redissolved by the addition of aqueous

2 N HCl and the mixture was extracted with ether

(3Â50 mL). The combined organic layer was washed

with brine, dried over Na₂SO₄, and concentrated in

vacuo. Flash chromatography (EtOAc/hexanes, 10/90)

gave 1.17 g (72%) of diol as a pale-yellow solid. Opti-

cally pure diol (>99.9% ee) was obtained as white

crystals after a single recrystallization from CHCl₃.

Spectral data for this compound matched that pre-

viously reported:²⁹>99.9% ee; R_f0.37 (EtOAc/hex-

9H); IR (thin ®lm) 2958, 2933, 2875, 1800, 1492, 1260,

1

1123 cm

.

cis-4-Phenyl-3-(trimethylsilyl)oxetan-2-one (10z). This a-

silyl-b-lactone was obtained from benzaldehyde (91 mL,

0.892 mmol) and trimethylsilylketene 9a (153 mg,

1.34 mmol) using Lewis acid 14 prepared from Et₂AlCl

(0.1 mL of 1.8 M in toluene, 0.178 mmol) and chiral diol

13 (60.4 mg, 0.214 mmol) according to the general pro-

cedure for asymmetric [2+2] cycloaddition. Flash chro-

matography (EtOAc/hexanes, 5/95) gave 161 mg (82%)

of a-silyl-b-lactone 10z as a colorless oil. Spectral data

for this compound matched that previously reported:³⁷

28% ee; R_f0.44 (EtOAc:hexanes, 10/90); ¹H NMR

(300 MHz, CDCl₃) d 7.30±7.46 (m, 5H), 5.71 (d,

J=6.3 Hz, 1H), 3.72 (d, J=6.3 Hz, 1H), 0.13 (s, 9H);

¹³C NMR (75 MHz, CDCl₃) d 170.5, 136.8, 128.5, 128.3,

anes, 20/80); mp 176.5±177.5 ^ꢀC (lit. 177±178 ^ꢀC); [a]_d

22

+8.9^ꢀ(c 1.46, CHCl₃); H NMR (200 MHz, CDCl₃) d

1

7.25±7.55 (m, 10H), 3.80 (bs, 1H), 3.32 (ddd, J=4.6, 9.9,

9.9 Hz, 1H), 2.50 (ddd, J=2.7, 9.8, 12.2 Hz, 1H), 0.85±

1.95 (m, 8H).

General procedure for the asymmetric [2+2] cyclo-

addition reaction as described for cis-4-cyclohexyl-3-

(trimethylsilyl)oxetan-2-one (10x). To a solution of

chiral diol 13, (60.4 mg, 0.214 mmol) in toluene (6 mL)

was added Et₂AlCl (0.10 mL of 1.8 M in toluene,

0.178 mmol) at 0 ^ꢀC slowly. After stirring for 5 h at 0 ^ꢀC,

the reaction mixture was cooled to 78 ^ꢀC then cyclo-

hexanecarboxaldehyde (108 mL, 0.892 mmol) was

added. After stirring for 15 min at 78 ^ꢀC, a solution of

trimethylsilylketene 9a³⁴(153 mg, 1.34 mmol) in toluene

(2 mL) was added and the mixture was further stirred

for 30 min at 78 ^ꢀC. The resulting reaction mixture was

kept in a freezer ( 26 ^ꢀC) for 40 h. After addition of pH

7 buer (1 mL) at 26 ^ꢀC, the reaction mixture was

warmed to room temperature and ®ltered through celite.

The solution was diluted with ethyl acetate (10 mL),

washed with brine, dried over Na₂SO₄, ®ltered, and

concentrated. Flash chromatography (EtOAc/hexanes,

125.4, 72.6, 49.7, 1.86; IR (thin ®lm) 3067, 3036, 2955,

1

2901, 1811, 1453, 1255, 1110 cm

.

cis-4-Benzyl-3-(trimethylsilyl)oxetan-2-one (10aa). This

a-silyl-b-lactone was obtained from phenylacetaldehyde

(104 mL, 0.892 mmol) and trimethylsilylketene 9a

(153 mg, 1.34 mmol) using Lewis acid 14 prepared from

Et₂AlCl (0.1 mL of 1.8 M in toluene, 0.178 mmol) and

chiral diol 13 (60.4 mg, 0.214 mmol) according to the

general procedure for asymmetric [2+2] cycloaddition.

Flash chromatography (EtOAc/hexanes, 5/95) gave

94 mg (45%) of a-silyl-b-lactone 10aa as a colorless oil:

75% ee; R_f0.33 (EtOAc/hexanes, 10/90); [a]_d+96.1^ꢀ

22

(c 1.02, CHCl₃); ¹H NMR (300 MHz, CDCl₃) d 7.24±

7.36 (m, 5H), 4.83 (ddd, J=3.3, 6.3, 10.2 Hz, 1H), 3.44

(d, J=6.6 Hz, 1H), 3.15 (dd, J=10.2, 14.7 Hz, 1H), 3.05

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1269

1

(dd, J=3.3, 14.7 Hz, 1H), 0.29 (s, 9H); ¹³C NMR

(75 MHz, CDCl₃) d 170.5, 136.7, 128.7, 128.6, 127.0,

74.0, 46.5, 39.7, 1.1; IR (KBr) 1785 cm ¹; HRMS

calcd for C₁₃H₁₈O₂Si [M+Na]: 257.0974. Found:

257.0984.

anes, 1/5); H NMR (300 MHz, CDCl₃) d 7.20±7.36 (m,

5H), 4.45±4.55 (m, 1H), 3.48 (dd, J=5.7, 16.5 Hz, 1H),

3.03 (dd, J=4.5, 16.5 Hz, 1H), 2.65±2.88 (m, 2H), 2.01±

2.25 (m, 2H).

4-(5-tert-Butyldimethylsiloxypentyl)oxetan-2-one (7bb).

This b-lactone was obtained from (6-tert-butyldi-

methylsiloxy)hexanal (205 mg, 0.892 mmol) and tri-

methylsilylketene 9a (153 mg, 1.34 mmol) using Lewis

acid 14 prepared from Et₂AlCl (0.1 mL of 1.8 M in

toluene, 0.178 mmol) and chiral diol 13 (60.4 mg,

0.214 mmol) according to the general procedure for

asymmetric [2+2] cycloaddition followed by desilylation

according to the general procedure. Flash chromato-

graphy (EtOAc/hexanes, 10/90) gave 113 mg (55% yield

for two sequences) of b-lactone 7bb as a colorless oil:

46% ee; R_f0.47 (EtOAc/hexanes, 1/5); ¹H NMR

(S)-4-Cyclohexyloxetan-2-one (7x). The a-silyl-b-lactone

10x (150 mg, 0.66 mmol) was desilylated according to

the general procedure in CH₃CN (4 mL) using

.

KF 2H₂O (125 mg, 1.33 mmol) at ambient temperature.

Flash chromatography (5!10% ethyl acetate in hex-

anes) aorded 92 mg (91%) of b-lactone 7x as a color-

less oil, which exhibited spectral data which matched

that previously reported:^23cR_f0.22 (EtOAc/hexanes, 10/

90); 84% ee; [a]_d+15.9^ꢀ(c 1.67, CHCl₃); H NMR

(300 MHz, CDCl₃) d 4.20 (ddd, J=4.2, 5.7, 8.4 Hz, 1H),

3.43 (dd, J=5.7, 16.2 Hz, 1H), 3.12 (dd, J=4.2, 16.2 Hz,

1H), 1.55±2.00 (m, 5H), 1.0±1.37 (m, 6H).

22

1

(300 MHz, CDCl₃)

d 4.48±4.55 (m, 1H), 3.61 (t,

J=6.6 Hz, 2H), 3.52 (dd, J=6.0, 16.5 Hz, 1H), 3.07 (dd,

J=4.2, 16.5 Hz, 1H), 1.70±1.92 (m, 2H), 1.35±1.60 (m,

6H), 0.89 (s, 9H), 0.05 (s, 6H); ¹³C NMR (75 MHz,

CDCl₃) d 168.3, 71.2, 62.8, 42.8, 34.6, 32.4, 25.9, 25.4,

24.6, 18.3, 5.4; IR (thin ®lm) 1829 cm ¹; HRMS calcd

for C₁₄H₂₈O₃[M+H]: 273.1886. Found: 273.1890.

(R)-4-Butyloxetan-2-one (7s). This b-lactone was

obtained from a-silyl-b-lactone 10s (100 mg, 0.50 mmol)

.

and KF 2H₂O (94 mg, 1.0 mmol) according to the gen-

eral procedure for desilylation. Flash chromatography

(EtOAc/hexanes, 10/90) gave 58 mg (91%) of b-lactone

7s as a colorless oil: 85% ee; R_f0.23 (EtOAc/hexanes,

22

10/90); [a]_d

+24.7 (c 1.75, CHCl₃); ¹H NMR

4-(8-Nonenyl)oxetan-2-one (7cc). This b-lactone was

obtained from 9-decenal (138 mg, 0.892 mmol) and tri-

methylsilylketene 9a (153 mg, 1.34 mmol) using Lewis

acid 14 prepared from Et₂AlCl (0.1 mL of 1.8 M in

toluene, 0.178 mmol) and chiral diol 13 (60.4 mg,

0.214 mmol) according to the general procedure for

asymmetric [2+2] cycloaddition followed by desilyla-

tion according to the general procedure. Flash chroma-

tography (EtOAc/hexanes, 10/90) gave 124 mg (71%

yield for two sequences) of b-lactone 7cc as a colorless

(300 MHz, CDCl₃) d 4.51 (m, 1H), 3.51 (dd, J=6.0,

16.3 Hz, 1H), 3.06 (dd, J=4.2, 16.3 Hz, 1H), 1.70±1.90

(m, 2H), 1.36±1.50 (m, 4H), 0.93 (t, J=7.2 Hz, 3H); IR

1

(thin ®lm) 2957, 2930, 2860, 1828, 1125 cm

.

(R)-4-Benzyloxetan-2-one (7aa). This b-lactone was

obtained from a-silyl-b-lactone 10aa (85 mg, 0.36 mmol)

.

and KF 2H₂O (68 mg, 0.73 mmol) according to the gen-

eral procedure for desilylation. Flash chromatography

(10/90, EtOAc/hexanes) gave 53 mg (91%) of b-lactone

7aa as a colorless oil. Spectral data for this compound

matched that previously reported:³²R_f0.44 (1/3,

EtOAc/hexanes); [a]_d²²+6.5 (c 4.77, CHCl₃); ¹H NMR

(300 MHz, CDCl₃) d 7.21±7.35 (m, 5H), 4.74 (dddd,

J=4.2, 5.7, 6.3, 6.3 Hz, 1H), 3.48 (dd, J=5.7, 16.5 Hz,

1H), 3.22 (dd, J=6.3, 14.4 Hz, 1H), 3.15 (dd, J=4.2,

16.5 Hz, 1H), 3.06 (dd, J=6.3, 14.4 Hz, 1H).

1

oil: 22% ee; R_f0.32 (EtOAc/hexanes, 10/90); H NMR

(300 MHz, CDCl₃) d 5.81 (ddt, J=6.6, 10.5, 17.1 Hz,

1H), 4.91±5.03 (m, 2H), 4.47±4.55 (m, 1H), 3.52 (dd,

J=5.7, 16.2 Hz, 1H), 3.06 (dd, J=4.5, 16.2 Hz, 1H),

2.00±2.08 (m, 2H), 1.68±1.92 (m, 2H), 1.30±1.50 (m,

10H); ¹³C NMR (75 MHz, CDCl₃) d 168.3, 138.8, 114.1,

71.2, 42.7, 34.5, 33.6, 29.1, 28.9, 28.7, 28.6, 24.7; IR

(thin ®lm) 1829 cm ¹; HRMS calcd for C₁₂H₂₀O₂:

219.1361. Found: 219.1359.

4-(2-Phenylethyl)oxetan-2-one (7e). This b-lactone was

obtained from hydrocinnamalaldehyde (117 mL,

0.892 mmol) and trimethylsilylketene 9a (153 mg,

1.34 mmol) using Lewis acid 14 prepared from Et₂AlCl

(0.1 mL of 1.8 M in toluene, 0.178 mmol) and chiral diol

13 (60.4 mg, 0.214 mmol) according to the general pro-

cedure for asymmetric [2+2] cycloaddition followed by

desilylation according to the general procedure. Flash

chromatography (EtOAc/hexanes, 10/90) gave 94 mg

(60% yield for two sequences) of b-lactone 7e as a color-

less oil. Spectral data for this compound matched that

previously reported:^23c36% ee; R_f0.27 (EtOAc/hex-

4-(1-Ethylpropyl)oxetan-2-one (7dd). This b-lactone was

obtained from 2-ethylbutyraldehyde (110 mL, 0.892 mmol)

and trimethylsilylketene 9a (153 mg, 1.34 mmol) using

Lewis acid 14 prepared from Et₂AlCl (0.1 mL of 1.8 M

in toluene, 0.178 mmol) and chiral diol 13 (60.4 mg,

0.214 mmol) according to the general procedure for

asymmetric [2+2] cycloaddition followed by desilyla-

.

tion using KF 2H₂O according to the general procedure.

Flash chromatography (EtOAc/hexanes, 10/90) gave

58 mg (46% yield for two sequences) of b-lactone 7dd as

a colorless oil: 56% ee; R_f0.46 (EtOAc/hexanes, 20/80);

1270

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

¹H NMR (200 MHz, CDCl₃) d 4.36 (ddd, J=4.6, 5.6,

8.4 Hz, 1H), 3.47 (dd, J=6.0, 16.2 Hz, 1H), 3.11 (dd,

J=4.4, 16.2 Hz, 1H), 1.20±1.70 (m, 5H), 0.91 (t,

J=7.2 Hz, 6H); ¹³C NMR (50 MHz, CDCl₃) d 168.4,

73.7, 44.5, 41.5, 21.4, 20.8, 10.9, 10.2; IR (thin ®lm)

2971, 2936, 2871, 1835 cm ¹; HRMS calcd for C₈H₁₄O₂

[M⁺]: 142.0994. Found: 142.0982.

2.5 h. The cells were collected by centrifugation (8000 g,

5 min), washed with 0.85% NaCl, and resuspended in

lysis buer (50 mM HEPES, pH 8.0, 50 mM NaCl,

10 mM EDTA, 15 mL). The cells were twice passed

through the French Press (20,000 psi) and cellular debris

was removed by centrifugation (8000 g, 5 min).

Ammonium sulphate precipitation. To cell-free extract

(20 mL), cooled in an ice bath, was gradually added

®nely crushed ammonium sulphate with stirring to 25%

saturation. Stirring was continued for 20 min at 40 ^ꢀC.

The precipitate was removed by centrifugation

(10,000 g, 10 min). Additional ammonium sulphate was

then added to 45% saturation, and the precipitate was

collected by centrifugation as above. The pellet was

resuspended in 25 mM HEPES (pH 8.2), and dialyzed

for 3 h two times against 1L of the same buer.

Ethyl-(S)-3-hydroxy-3-cyclohexylpropanoate. This b-

hydroxy ester was prepared by the reaction of b-lactone

7x (26 mg, 0.17 mmol) with NaOEt (14 mg, 0.20 mmol,

1.2 equiv) in EtOH-THF at ambient temperature.³⁸

General work-up followed by puri®cation by ¯ash

chromatography (EtOAc/hexanes, 20/80) gave 29 mg

(86%) of b-hydroxy ethyl ester. Spectral data for this

compound matched that previously reported:³⁹63% ee;

22

R_f0.30 (EtOAc/hexanes, 1/6); [a]_d

15.2 (c 0.66,

22

CHCl₃); lit. [a]_d+27.8 (c 0.66, CHCl₃) for R-enan-

tiomer;^{38 1}H NMR (300 MHz, CDCl₃) d 4.18 (q,

J=7.2 Hz, 2H), 3.74±3.81 (m, 1H), 2.88 (d, J=4.2 Hz,

1H), 2.52 (dd, J=2.7, 16.5 Hz, 1H), 2.41 (dd, J=9.3,

16.5 Hz, 1H), 1.60±1.90 (m, 5H), 1.28 (t, J=7.2 Hz, 3H),

0.98±1.42 (m, 6H).

Q-Sepharose chromatography. The dialyzed solution

was applied onto a 50 mL Q-Sepharose FPLC column,

pre-equilibrated with loading buer (25 mM HEPES,

pH 8.2). After washing with one column volume of

loading buer (10 min), the column was eluted with a

continuous salt gradient (0±0.5 M NaCl, 5 mL/min, over

60 min). Fractions (5 mL) were collected and monitored

for HMG-CoA synthase activity. Maximum enzyme

activity was found in fractions 32±35, which were

pooled and concentrated to 0.5 mL in a Gelman con-

centrating centrifuge tube (MW cut-o=30 kD,

10000 g, 60 min).

Methyl-(R)-3-hydroxyheptanoate. This b-hydroxy ester

was prepared by the reaction of b-lactone 7a (45 mg,

0.35 mmol, 85% ee) with NaOMe (22 mg, 0.42 mmol,

1.2 equiv) in MeOH-THF at ambient temperature.³⁸

General work-up followed by puri®cation by ¯ash

chromatography (EtOAc/hexanes, 20/80) gave 47 mg

(85%) of b-hydroxy methyl ester. Spectral data for this

compound matched that previously reported:^30b85%

S-200 chromatography. The resulting solution was loa-

ded onto a S-200 column (600 mL) which had been pre-

equilibrated with loading buer (25 mM HEPES, pH

8.2, 50 mM NaCl). Elution (same buer, 0.3 mL/min)

gave fractions (6 mL). Active fractions (30±34) were

pooled and concentrated as above.

22

ee; R_f0.25 (EtOAc/Hexanes, 20/80); [a]_d

1.67 (c

22

2.58, EtOH); lit. [a]_d+1.85 (c 2.58, EtOH) for S-

enantiomer; ¹H NMR (200 MHz, CDCl₃) d 4.01 (m,

1H), 3.71 (s, 3H), 2.87 (bs, 1H), 2.30±2.60 (m, 2H),

1.20±1.60 (m, 6H), 0.91 (t, J=7.3 Hz, 3H); IR (thin ®lm)

1

3455, 2927, 1730 cm

.

Acknowledgements

General biochemical procedures. Biochemical reagents

were purchased from Sigma and chemical reagents from

Aldrich. HMG-CoA synthase assays, protein determi-

nations and other biochemical procedures were carried

out as described previously.²⁰Cloning of the HMG-

CoA synthase gene of S. cerevisiae into E. coli to give

strain Y-HMGS was performed according to standard,

established procedures.⁴⁰

Financial support for this work from the Natural Sci-

ences and Engineering Research Council of Canada

through grants to both P.H. and G.D.W., the Medical

Research Council of Canada (GDW) and the National

Science Foundation (Grant CHE-9624532 to D.R.) is

gratefully acknowledged. We thank Dr. Lloyd Sumner

and Dr. Barbara Wolf of the Texas A&M Center for

Characterization for mass spectral analyses obtained on

instruments acquired by generous funding from the

NSF (CHE-8705697) and the TAMU Board of Regents

Research Program.

Cell growth and lysis. The E. coli strain Y-HMGS was

grown overnight on LB-amp plates at 37 ^ꢀC. A single

colony was then used to inoculate LB-amp liquid med-

ium (50 mL) which was grown overnight at 37 ^ꢀC with

shaking (250 rpm). LB-amp liquid culture (1 L) was then

inoculated with the overnight culture (7 mL) and grown

to OD₆₀₀=0.6 (37 ^ꢀC, 250 rpm). Solid IPTG (95 mg) was

then added and incubation was continued for a further

References and Notes

1. For a review of this enzyme, see: Quant, P. A. Essays Bio-

chem. 1994, 28, 13. Qureshi, N.; Porter, J. W. In Biosynthesis of

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

1271

Isoprenoid Compounds; Porter, J. W.; Spurgeon, S. L., Eds;

John Wiley and Sons: New York, 1981; Vol. 1, pp 57±62.

17. (a) Tomoda, H.; Kumagai, H; Ogawa, Y.; Sunazuka, T.;

Hashizume, H.; Nagashima, H.; Omura, S. J. Org. Chem. 1997,

62, 2161. (b) Hashizume, H.; Ito, H.; Yamada, K.; Nagashima,

H.; Kanao, M.; Tomoda, H.; Sunazuka, T.; Kumagai, H;

Omura, S. Chem. Pharm. Bull. 1994, 42, 512. (c) Hashizume,

H.; Ito, H.; Kanaya, N.; Nagashima, H.; Usui, H.; Oshima, R.;

Kanao, M.; Tomoda, H.; Sunazuka, T.; Nagamitsu, T.;

Kumagai, H.; Omura, S. Heterocycles 1994, 38, 1551. (d)

Sunazuka, T.; Tsuzuki, K.; Kumagai, H.; Tomoda, H.;

Tanaka, H.; Nagashima, H.; Hashizume, H.; Omura, S. J.

Antibiot. 1992, 45, 1139. (e) Thompson, K. L.; Chang, M. N.;

Chiang, Y.-CP.; Yang, S. S.; Chabala, J. C.; Arison, B. H.;

Greenspan, M. D.; Hanf, D. P.; Yudkovitz, J. Tetrahedron

Lett. 1991, 32, 3337. (f) Mayer, R. J.; Louis-Flamberg, P.;

Elliott, J. D.; Fisher, M.; Leber, J. Biochem. Biophys. Res.

Commun. 1990, 169, 610. (g) Chabala, J. C.; Chiang, Y.-CP.;

Chang, M. N.; Graham, D. W.; Heck, J. V.; Yang, S. S. U.S.

Patent 4,806,564, 1989.

2. For reviews of terpene biosynthesis, including HMG-CoA

synthase, see: Dewick, P. M. Nat. Prod. Rep. 1997, 14, 111, and

previous reports in this series.

3. Herbert, R. B. The Biosynthesis of Secondary Metabolites;

Chapman and Hall: London, 1981; Ch. 4. Biosynthesis of Iso-

prenoid Compounds; Porter, J. W.; Spurgeon, S. L., Eds; John

Wiley and Sons: New York, 1981.

4. For selected recent reviews, see: (a) Hiyama, T. Pure Appl.

Chem. 1996, 68, 609. (b) Scheen, A. J. Rev. Med. Liege 1996,

51, 97. (c) Hampton, R.; Dimster-Denk, D.; Rine, J. Trends

Biochem. Sci. 1996, 21, 140.

5. For selected recent reviews, see: (a) Kong, S. X.; Crawford,

S. Y.; Gandhi, S. K.; Seeger, J. D.; Schumock, G. T.; Lam, N.

P.; Stubbings, J.; Schoen, M. D. Clin. Ther. 1997, 19, 778. (b)

Bjoerkhem, I.; Lund, E.; Rudling, M. Subcell. Biochem. 1997,

28, 2. 3. (c) Lennernas, H.; Fager, G. Clin. Pharmacokinet.

1997, 32, 403. (d) Mellies, M.; McGovern, M. Expert Opin.

Invest. Drugs 1997, 6, 31. (e) Hughes, A. D. J. Hum. Hypertens.

1996, 10, 387.

18. (a) Tomoda, H.; Kumagai, H.; Tanaka, H.; Omura, S. J.

Antibiot. 1993, 46, 872. (b) Greenspan, M. D.; Bull, H. G.;

Yudkovitz, J. B.; Hanf, D. P.; Alberts, A. W. Biochem. J. 1993,

289, 889.

6. (a) For anti-fungal activity of F-244, see: Garrity, G. M.;

Onishi, J. C.; Del Val, S. M. Eur. Pat. Appl. 234, 752, 1987. (b)

For use of F-244 in treatment of diabetic ketoacidosis, see:

Alberts, A. W.; Greenspan, M. D., U.S. Pat. 4,873,260, 1989.

19. Nagashima, H.; Kumagai, H.; Tomoda, H.; Omura, S. Life

Sci. 1993, 52, 1595.

20. Bell, K.; Saepudin, E.; Harrison, P. Can. J. Chem. 1996, 74,

24.

7. (a) Stewart, P. R.; Rudney, H. J. Biol. Chem. 1966, 241,

1212. (b) Middleton, B.; Tubbs, P. K. Biochem. J. 1972, 126, 2.

7. (c) Middleton, B. Biochem. J. 1972, 126, 35.

21. (a) Staudinger, H.; Bereza, S. Ann. 1911, 380, 243. (b)

Zaitseva, G. S.; Vinokurova, N. G.; Baukov, Y. I. Zh. Obshch.

Khim. 1975, 45, 1398. (c) Mead, K. T.; Samuel, B. Tetrahedron

Lett. 1988, 29, 6573.

8. Page, M. A.; Tubbs, P. K. Biochem. J. 1978, 173, 925.

9. (a) Clinkenbeard, K. D.; Sugiyama, T.; Reed, W. D.; Lane,

M. D. J. Biol. Chem. 1975, 250, 3124. (b) Misra, I.; Nara-

simhan, C.; Miziorko, H. M. J. Biol. Chem. 1993, 268, 12129.

22. (a) Danheiser, R. L.; Nowick, J. S.; Lee, J. H.; Miller, R.

F.; Huboux, A. H. Org. Synth. 1996, 73, 61 and references

cited. (b) Wedler, C.; Schick, H. Org. Synth. 1998, 75, 116 and

references cited.

10. (a) Russ, AP; Ruzicka, V.; Maerz, W.; Appelhans, H.;

Groû, W. Biochim. Biophys. Acta 1992, 1132, 329. (b) Rokosz,

L. L.; Boulton, D. A.; Butkiewicz, E. A.; Sanyal, G.; Cueto, M.

A.; Lachance P. A.; Hermes, J. D. Arch. Biochem. Biophys.

1994, 312, 1.

23. (a) Hirai, K.; Homma, H.; Mikoshiba, I. Heterocycles

1994, 38, 281. (b) Yang, H. W.; Romo, D. J. Org. Chem. 1997,

62, 5. (c) Yang, H. W.; Zhao, C.; Romo, D. Tetrahedron 1997,

63, 16471. (d) Yang, H. W.; Romo, D. J. Org. Chem. 1998, 63,

1344.

11. Misra, I.; Miziorko, H. M. Biochem. 1996, 35, 9610.

12. (a) Miziorko, H. M.; Lane, M. D. J. Biol. Chem. 1977, 252,

1414. (b) Miziorko, H. M.; Shortle, D.; Lane, M. D. Biochem.

Biophys. Res. Commun. 1976, 69, 92.

24. Maruoka, K.; Concepcion, A. B.; Yamamoto, H. Synlett

1992, 31.

25. (a) Tamai, Y.; Someya, M.; Fukumoto, J.; Miyano, S. J.

Chem. Soc., Chem. Commun. 1994, 1549. (b) Tamai, Y.; Yosh-

iwara, H.; Someya, M.; Fukumoto, J.; Miyano, S. J. Chem.

Soc., Chem. Commun. 1994, 2281. (c) Dymock, B. W.;

Kocienski, P. J.; Pons, J.-M. J. Chem. Soc., Chem. Commun.

1996, 1053.

13. (a) Lowe, D. M.; Tubbs, P. K. Biochem. J. 1985, 232, 37.

(b) Charlier Jr., H. A.; Narasimhan, C.; Miziorko, H. M. Bio-

chemistry 1997, 36, 1551.

14. (a) Omura, S.; Tomoda, H.; Kumagai, H; Greenspan, M.

D.; Yudkovitz, J. B.; Chen, J. S.; Alberts, A. W.; Martin, I.;

Mochales, S.; Monaghan, R. L.; Chabala, J. C.; Schwartz, R.

E.; Patchett, A. A. J. Antibiot. 1987, 40, 1356. (b) Greenspan,

M. D.; Yudkovitz, J. B.; Lo, C-YL.; Chen, J. S.; Alberts, A.

W.; Hunt, V. M.; Chang, M. N.; Yang, S. S.; Thompson, K. L.;

Chiang, Y-CP.; Chabala, J. C.; Monaghan, R. L.; Schwartz, R.

L. Proc. Natl. Acad. Sci. USA 1987, 84, 7488.

26. Yang, H. W.; Romo, D. Tetrahedron Lett. 1998, 2877.

27. Wemple, J. Tetrahedron Lett. 1975, 3255.

28. Masamune, S.; Hayase, Y.; Chan, W. K.; Sobczak, R. L. J.

Am. Chem. Soc. 1976, 98, 7874.

29. Seebach, D.; Beck, A. K.; Imwinkelried, R.; Roggo, S.;

Wonnacott, A. Helv. Chim. Acta. 1987, 70, 954.

15. Tomoda, H.; Kumagai, H.; Takahashi, Y.; Tanaka, Y.;

Iwai, Y.; Omura, S. J. Antibiot. 1988, 41, 247.

30. (a) Taber, D. F.; Silverberg, L. J. Tetrahedron Lett. 1991,

32, 4227. (b) Taber, D. F.; Deker, P. B.; Silverberg, L. J. J. Org.

Chem. 1992, 57, 5990. (c) Kitamura, M.; Ohkuma, T.; Toku-

naga, M.; Noyori, R. Tetrahedron: Asymmetry 1990, 1, 1.

16. (a) Aldridge, D. C.; Giles, D.; Turner, W. B. J. Chem. Soc.

D 1970, 639. (b) Aldridge, D. C.; Giles, D.; Turner, W. B. J.

Chem. Soc. C 1971, 3888. (c) Tomoda, H.; Kumagai, H.;

Tanaka, H.; Omura, S. Biochim. Biophys. Acta 1987, 922, 351.

31. Shitangkoon, A.; Vigh, G. J. Chromatogr. A 1996, 3. 1.

1272

D. Romo et al./Bioorg. Med. Chem. 6 (1998) 1255±1272

32. Capozzi, G.; Roelens, S.; Talami, S. J. Org. Chem. 1993,

58, 7932.

Engl. 1986, 25, 861. (b) van Vliet, M. R. P.; Buysingh, P.; van

Koten, G.; Vrieze, K.; Kojie-Prodie, B.; Spek, A. L. Organo-

metallics, 1985, 4, 1701. (c) Kriz, O.; Casensky, B.; Lycka, A.;

Fusek, J.; Hermanek, S. J. Mag. Res. 1984, 60, 375. Generally,

the higher the coordination number (CN) of the Al atom, the

greater the up®eld shift in ²⁷Al NMR: CN=3, d 250±300 ppm,

33. (a) Ruden, R. A. J. Org. Chem. 1974, 39, 3607. (b) Brady,

W. T.; Saidi, K. J. Org. Chem. 1979, 44, 733.

34. (a) Jones, E. R. H.; Eglinton, G.; Whiting, M. C.; Shaw, B.

L. Org. Syn. Coll. Vol. IV, p 404. (b) Private communication

from Prof. Howard T. Black (Eastern Illinois University).

Although ethoxyacetylene is commercially available, (Aldrich,

Lancaster) it is rather expensive (>$10/g) and, in our experi-

ence, the purity is variable and thus distillation is required prior

to use. We have prepared ethoxyacetylene by the above meth-

ods in 0±75% yields (from chloroacetaldehyde diethyl acetal)

and, also in our hands (as communicated by Black), very care-

ful quenching of the intermediate sodium acetylide was critical

for reproducible and reasonable yields (50±75%). Several other

reported methods for silylketene synthesis have been studied in

our laboratory but the silylation of ethoxyacetylene followed

by thermolytic retro-ene reaction (ref 33) provided the highest

purity and yields of silylketenes.

monomer, (e.g. (i-Bu)₃Al, 276 ppm, monomer); CN=4,

150±180 ppm, dimer, (e.g. Et₂AlCl, 167 ppm, dimer); CN=5,

100±130 ppm, dimer, (e.g. Et₂AlO(CH₂)₂OMe, 121 ppm,

d

dimer); CN=6, d 0±10 ppm, monomer, (e.g. Al(acac)₃, 0 ppm,

monomer). The ²⁷Al NMR spectrum of a toluene-d₈solution

(Al(acac)₃, 0 ppm, external reference) of catalyst 14 showed a

broad single peak at d 39 and a very small shoulder at d 90

indicating that the catalyst 14 may exist as two species under

these conditions, one of which is a dimer.

37. Concepcion, A. B.; Maruoka, K.; Yamamoto, H. J. Tetra-

hedron 1995, 51, 4011.

38. Arnold, L. D.; Kalatar, T. H.; Vederas, J. C. J. Am. Chem.

Soc. 1985, 107, 7105.

35. Corey, E. J.; Rohde, J. J. Tetrahedron Lett. 1997, 38, 37. A

related aluminum complex reported by Koga for use in Diels±

Alder reactions is discussed in this communication.

36. In organoaluminum compounds, the ²⁷Al NMR chemical

shifts are strongly dependent on the ligand character, on the

structural types, and on the coordination number (CN) of the

Al atom, see: (a) Benn, R.; Ru®nska, A. Angew. Chem. Int. Ed.

39. Carreira, E. M.; Singer, R. A.; Lee, W. J. Am. Chem. Soc.

1994, 116, 8837.

40. (a) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular

Cloning, A Laboratory Manual; 2nd Edn; Cold Spring Harbor:

Cold Spring Harbor. (b) Ausubel, F. M.; Brent. R.; Kingston, R.

E.; Moore, D. D.; Seidman, J. G.; Smith, J. A.; Struhl, K. Current

Protocols in Molecular Biology, John Wiley and Sons, New York.

Article Doi

Synthesis and inhibitory action on HMG-CoA synthase of racemic and optically active oxetan-2-ones (β-Lactones)

DOI: 10.1016/S0968-0896(98)00114-X

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S0968-0896(98)00114-X

Full text of DOI:10.1016/S0968-0896(98)00114-X

Products guided by the article

R&D Labs maybe for 39185-88-1

Relevant to this article

Hot Product