Enantioselective syntheses of 1-carbacephalosporins from chemoenzymically derived β-hydroxy-α-amino acids: Applications to the total synthesis of carbacephem antibiotic loracarbef

Article abstract of DOI:10.1016/S0040-4020(00)00416-6

Serine hydroxymethyltransferase (SHMT) derived from recombinant E. coli was found to be able to catalyze the condensation between glycine and 4-pentenaldehyde, affording enantiopure L-erythro-2-amino-3-hydroxy-6-heptenoic acid (AHHA) in high yield and throughput. Conversion of this chiral intermediate of biosynthetic origin to the oral carbacephalosporin antibiotic loracarbef (Lorabid) via β-lactam forming reactions and subsequent Dieckmann cyclization was achieved. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00416-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 5667±5677
Enantioselective Syntheses of 1-Carbacephalosporins from
Chemoenzymically Derived b-Hydroxy-a-Amino Acids:
Applications to the Total Synthesis of Carbacephem
Antibiotic Loracarbef
Billy G. Jackson,a Steve W. Pedersen,a Jack W. Fisher,a Jerry W. Misner,a John P. Gardner,a
Michael A. Staszak,a Christopher Doecke,a John Rizzo,a James Aikins,a Eugene Farkas,a
Kristina L. Trinkle,a Jeffrey Vicenzi,a Matt Reinhard,a Eugene P. Kroeff,a,b
Chris A. Higginbotham,a,b R. J. Gazaka,b and Tony Y. Zhanga,
*
aChemical Process Research and Development, Lilly Research Laboratories, Eli Lilly & Company, Lilly Corporate Center,
Indianapolis, IN 46285-4813, USA
bBioprocess Puri®cation Development, Lilly Research Laboratories, Eli Lilly & Company, Lilly Corporate Center, Indianapolis,
IN 46285-4813, USA
The authors wish to dedicate this article to the memory of Dr Lowell D. Hat®eld, who lost his life in the ®ght against lung cancer on 13 June 1999
Received 10 November 1999; accepted 21 February 2000
AbstractÐSerine hydroxymethyltransferase (SHMT) derived from recombinant E. coli was found to be able to catalyze the condensation
between glycine and 4-pentenaldehyde, affording enantiopure l-erythro-2-amino-3-hydroxy-6-heptenoic acid (AHHA) in high yield and
throughput. Conversion of this chiral intermediate of biosynthetic origin to the oral carbacephalosporin antibiotic loracarbef (Lorabidw) via
b-lactam forming reactions and subsequent Dieckmann cyclization was achieved. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
Cephalosporins (1) have been the most widely used anti-
biotics in the world to eradicate bacterial infections.1±8
Replacement of the sulfur atom by a methylene in cephalo-
sporins has led to a new class of antibiotics, i.e. carba-
cephalosporins (2), which were shown to exhibit
remarkable enhancement of chemical and serum stability
over their parent compounds.9±11 These include several anti-
biotics and b-lactamase inhibitors. However, unlike their
sulfur counterparts which are usually obtained by partial
synthesis from either penicillin sulfoxide esters or side
chain modi®cation of natural cephalosporins, carba-
cephalosporins are almost exclusively obtained by de
novo total synthesis.
The inherent chiral centers at the ring juncture, strained
[4.2.0] bicyclic system and a high density of functional
groups make these molecules very challenging targets for
large scale manufacture. The dif®culty in the synthesis
contributed, in no small part, to the fact that despite their
advantageous biological activities demonstrated both in
vivo and in vitro, the only carbacephalosporin that has
Keywords: aldolase; antibacterials; biocatalysis; cephalosporins; car-
bacephalosporins; b-hydroxy-a-amino acid; b-lactam; loracarbef; serine
hydroxymethyltransferase; SHMT.
* Corresponding author.
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00416-6

5668
B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
Scheme 1.
been developed and approved for clinical usage is
loracarbef (2, RˆD-phenylglycinoyl, R1ˆCl), discovered
by scientists at Kyowa Hakko,12 developed and marketed
by Eli Lilly as an orally active and broad spectrum
antibiotic.13
Result and Discussions
We envisioned that a protected loracarbef nucleus 9
(PGˆamino protecting group) could be prepared from a
b-hydroxy-a-amino acid derivative 14 (Aˆlatent carboxyl-
ate synthon) as outlined in Scheme 1 in a retrosynthetic
sense. The choice of p-nitrobenzyl ester was based on its
ready removal and the need to control the quality of the ®nal
product by not introducing new impurities. Use of
b-hydroxy-a-amino acid derivatives for synthesis of mono-
cyclic b-lactams was pioneered by Miller21,27±30 and was
used thereafter by Squibb scientists for the synthesis of
monocyclic antibiotics.12 Applications of variously substi-
tuted b-hydroxy-a-amino acid derivatives for synthesis of
bicyclic b-lactams was again reported by Miller33,34,51 and
in preliminary publications from these laboratories.
However, to achieve the proposed synthesis of loracarbef
in an enantioselective manner, four major challenges
emerged: First, an enantiopure or reasonably enriched
source of compound 14 had to be secured. Secondly, an
ef®cient methodology for the construction of the b-lactam
ring needed to be achieved to afford the monocyclic
compound 12. Thirdly, a suitable carboxylate synthon had
to be incorporated at an appropriate stage of the synthesis to
allow for the Dieckmann cyclization. Last, but not least, a
proper amino protecting group (PG) would have to be
chosen to enable high chemoselectivity throughout the
process and to ensure facile removal at the end to reveal
the highly functionalized molecule.
The ®rst laboratory-scale synthesis was achieved by an
intramolecular Horner±Emmons±Wadsworth condensation
to give the racemic nucleus 5 (Rˆt-butyl), which was then
kinetically resolved via enzymatic acylation with phenyl-
glycine.14 An exploratory scale synthesis from these
laboratories involved a partial synthesis from 6, whose
genesis from penicillin sulfoxide was known.15 One of the
®rst kilogram-scale syntheses of loracarbef utilized the
oxazolidinone chiral auxiliary of Evans and Sjogren16±18
for synthesis of 8. Subsequent elaboration of 8 to nucleus
5 (Rˆp-nitrobenzyl) involved appending the six-membered
ring via rhodium-mediated carbenoid insertion. An
improved synthesis of 4 was developed later and utilized a
highly ef®cient enzymatic kinetic resolution of 3-amino-
azetidinone 7.19,20 A further improvement in converting 8
to 5 (Rˆp-nitrobenzyl) featured a regiospeci®c Dieckmann
cyclization for annulation of the tetrahydropyridine.22,50
Other published syntheses of loracarbef have utilized
penicillin sulfoxide ester derived intermediates23,24 and an
enantioselective synthesis starting from sodium erythor-
bate.25,26 Still, a more streamlined and economical synthetic
process that produces high quality loracarbef in a robust,
safe, and environmentally friendly manner is highly
desirable.
While a number of methods are available for the synthesis
of compounds represented by the general formula 14, most
of them suffer from disadvantages including long and
tedious synthesis, use of expensive chiral auxiliaries and
involve chromatographic puri®cation processes. As a
starting material for the synthesis of a drug substance in
Scheme 2.

B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5669
Scheme 3.
multi-ton quantities, ready availability is a prerequisite. We
anticipated being able to use the enzyme serine hydroxy-
methyltransferase (SHMT)54 for construction of the desired
stereoisomer of 14, as shown in Scheme 2. Serine hydroxy-
methyltransferase (SHMT) (E.C.2.1.2.1) is a widely occur-
ring enzyme that catalyzes the physiological conversion of
serine (Ser) to glycine (Gly), transferring the hydroxy-
methylene unit to tetrahydrofolate;35±42 however, it has
been employed in the reverse direction for synthesis of
chiral b-hydroxy-a-amino acids42,43 from glycine and
various aldehydes. To test the viability of the chemistry
shown in Scheme 2, we required developmental quantities
of the enzyme which was known to be available from a
variety of natural sources. Initial quantities of the enzyme
were isolated from rabbit liver using the methodology of
Martini et al.44±46 In later studies the enzyme was derived
via expression of the cloned gly A gene from Escherichia
coli.47 As the desired SHMT enzyme accounts for a large
portion of the crude protein mass isolated from the E. coli,
the fermentation broth of the crude enzyme can be used as
such without the need of further puri®cation.
tion of the amino acid with o-phthaldehyde and N-acetyl-l-
cysteine. This methodology, by providing a chromophore
(isoindole) and a second chiral carbon atom, was able to
convert a racemic mixture of amino acids into a mixture
of diastereomers which were separable via conventional
reverse phase HPLC and could be detected using a normal
UV detector. However, for unequivocal stereochemistry
assignment of the products from the initial enzyme screen-
ing, all four possible stereoisomers of 14 (AˆCO2R) were
needed.
Synthesis of the racemic erythro stereoisomer is shown in
Scheme 4. The selective hydrogenation of oxime 2049
afforded ketoamino acid 21, which was further reduced by
diastereoselective hydrogenation under platinum oxide to
give erythro b-hydroxy-a-amino acid 22 in racemic form.
The free diacid derived from 22 was found to be prone to
cyclization under acidic conditions to give lactone 24.
However, both compound 23 and 24 proved useful as analy-
tical standards. Availability of these compounds not only
facilitated analytical aspects of the biocatalysis reaction
but also provided valuable intermediates, allowing the
evaluation of downstream chemistry concurrent with the
development of the enzymatic process.
We initially chose the esters of succinic semialdehyde 16
(AˆCO2Me, CO2Et, COt2Bu) as potential substrates for the
enzymic reaction. Synthesis of the various substrates was
achieved by esteri®cation of 4-pentenoic acid followed by
ozonolysis (Scheme 3). Miller42 had shown that the methyl
ester of succinic semialdehyde was a good substrate for
the enzymic synthesis. Additionally, evidence for the
product being the desired l-erythro stereoisomer was
circumstantial.
The corresponding racemic threo stereoisomers were
synthesized as shown in Scheme 5. Methyl hippurate was
condensed with succinic anhydride giving 27, which was
reduced with sodium borohydride to give threo 28. Attempts
to hydrolyze both the N-benzoyl blocking group and the
methyl ester of 28 by saponi®cation were not successful;
however, hydrolysis in aqueous hydrochloric acid did effect
the desired hydrolyses. The product of the acid-mediated
hydrolysis was lactone 29, which could be hydrolyzed to
give the racemic threo stereoisomer 30.
For analysis of reaction mixtures arising from the enzyme-
mediated reaction of glycine with various aldehydes, an
HPLC methodology was developed based on a procedure
reported by Aswad48 for the separation and quanti®cation of
d- and l-aspartic acid. The procedure involved derivatiza-
With these authentic isomers of a-amino-b-hydroxyadipic
Scheme 4.

5670
B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
Scheme 5.
Scheme 6.
acid derivatives in hand and analytical methodology in
place, compound 16 was subjected to condensation with
excess glycine in the presence of the enzyme SHMT.
Indeed, l-erythro acid 31 (2S, 3S) was obtained in good
yield, along with a small amount of l-threo (2S, 3R) isomer
32, and their corresponding lactones (33, 34) and lactam
(35, Scheme 6).
nucleophilic displacement of the hydroxyl group by the
primary amide nitrogen failed. In most cases the distal
amide oxygen proved to be more nucleophilic than the
intended primary amide nitrogen and as a consequence,
amidate salt 42 was isolated as a stable solid, which upon
hydrolysis yielded lactone 43. Efforts aimed at converting
compound 43 into lactam 44 met with little success as a poor
nucleophile (amide nitrogen) failed to displace an equally
poor leaving group (lactone oxygen, Scheme 8). Apparently
two issues had to be resolved: First, the distal carboxylate
needed to be prevented from participating in the cyclization
event. Secondly the nucleophilicity of the amide nitrogen
needed to be enhanced.
The lactones, however, can be converted into open chained
diacids. Isolation of these highly water soluble compounds
from the aqueous enzymatic reaction media was facilitated
by converting the free amino acids in situ into the corre-
sponding amide 37a or carbamate 37b, respectively
(Scheme 7).
To address the ®rst concern, compound 45, an aldehyde with
a furan ring attached as a more benign carboxylate synthon,
and glycine were subjected to the enzyme catalyzed aldol
condensation. The anticipated product, b-hydroxy-a-amino
acid, also had the added advantage of being convertible into
pivotal intermediates (7 and 8) of a previous large scale
Further transformation of these compounds into the desired
b-lactam 44 proved very problematic. While the distal
carboxylic group can be differentiated and selectively
converted into dimethyl amide 39 by way of lactone 38,
numerous attempts to effect b-lactam ring closure through
Scheme 7.

B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5671
Scheme 8.
Scheme 9.
synthesis. To our disappointment, 45 turned out to be a poor
substrate for SHMT, causing severe enzyme inhibition.
Moreover, diastereoselectivity of the reaction was very
low and the resulting epimeric mixture 46 proved dif®cult
to separate. Aldehyde 47, on the other, showed much more
promise. Besides being a readily available and inexpensive
starting material, its reaction with glycine catalyzed by
SHMT afforded cyanoamino acid 48 in high enantio-
selectivity and diastereoselectivity (Scheme 9). Since
scale-up of the enzymatic reaction and carrying out the
Scheme 10.

5672
B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
Scheme 11.
entire sequence was resource intensive, we prepared
compound 50 from known acid 49 as a model to check
the viability of using nitrile as a carboxylate equivalent in
the total synthesis. It was quickly established that although
the methylene protons adjacent to the carboxylate in 51 are
more acidic than those next to the cyano group, the p-nitro-
benzyl ester appeared to be a much better electrophile, and
consequently, the only product isolated from the condensa-
tion reaction was keto nitrile 52. This notion is further
supported by the fact that when dinitrile 54 was subjected
to the same reaction condition, iminonitrile 55 was formed
in good yield. While this compound has the same carbon
framework and functional group alignment as the much
desired enol ester 10, its conversion to the latter
compound proved to be a formidable challenge due to the
lability of the b-lactam ring under hydrolytic conditions
(Scheme 10).
57 in high yield and selectivity. In situ protection of the
amino acid was achieved in high yield to give 58 which
was converted into mesylate 60. Here again the nucleo-
philicity of the primary amide nitrogen was suf®cient
enough and most of the starting compound went on to
give oxazoline 61,51 the same compound produced when
alcohol 59 was subjected to Mitsunobu conditions.52
However, displacement of the mesylate by the amide nitro-
gen was indeed possible by converting 60 into N-sulfona-
mate 62 using the Squibb protocol,31,32 affording b-lactam
63 in high yield. Direct attachment of a p-nitrobenzyl
acetate as preferred for the downstream chemistry was
complicated by side products formation, primarily hydroly-
sis of the esters. However, the desired pNB ester 65 could be
conveniently obtained by an in situ ester exchange reaction
under phase transfer catalysis (Scheme 11).53
Several methods were explored for the conversion of the
terminal alkene into a carboxylic acid, with an in situ oxida-
tion of the aldehyde obtained by ozonolysis being the most
preferable.55±57 This protocol afforded carboxylic acid 66 in
excellent yield. This compound was successfully converted
into loracarbef in ®ve synthetic steps as previously
reported.50 This consisted of sequential phenyl ester forma-
tion (PhOH, DMAP), Dieckmann cyclization (tBuOLi),
chlorination±deacylation ((PhO)3P, Cl2), side-chain attach-
ment and reductive p-nitrobenzyl ester removal, thus
completing the total synthesis of this carbacephalosporin
antibiotic from an unnatural chiral amino acid of
biosynthetic origin (Scheme 12).
These early learning experiences underscored the need and
importance of strategically addressing all four elements of
the synthesis: (1) availability of the chiral synthon; (2) an
ef®cient b-lactam forming step, a selective ring annulation;
(3) a judicious choice of protecting groups; and (4) an
ef®cient annulation protocol. This also led us to examine
the use of an ethylene unit as a latent carboxyl group and to
exploit the higher nucleophility of the amide nitrogen in its
anionic form. To our delight, 4-pentenal (56) proved to be
an excellent substrate for SHMT, and intense optimization
efforts on both the enzyme expression and biocatalysis
fronts yielded a set of condition that afforded amino acid
Scheme 12.

B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5673
In summary, we have developed a new synthesis of loracar-
bef based on the feasibility of using serine hydroxylmethyl
transferase (SHMT) as a catalyst for the enantioselective
condensation between glycine and 4-pentenal. The synthe-
sis entails a novel enantioselective biocatalysis step, and a
chemoselective b-lactam forming reaction. As our goal was
to develop an industrial process for the synthesis of this
valuable antibiotic in large scale, much effort has been
focused on improving the practical aspects of the chemistry
and biology. It is evident that in order to realize the
envisioned chemistry on large scale utilizing an enzyme
which has rarely been used in synthesis, a collaborative
effort between molecular biologists, biochemists, analytical
chemists, organic chemists and chemical engineers must be
coordinated to address various concerns in a coherent
fashion. We have demonstrated the ef®ciency and potential
of using isolated enzymes for enantioselectively establish-
ing chiral centers via carbon±carbon formation, a process
that would have involved a tedious protection±deprotection
protocol, chiral auxiliary attachment±detachment or resolu-
tions if more traditional approaches were employed.
Applications of a SHMT catalyzed condensation reac-
tions to the synthesis of other natural and synthetic
molecules of biological importance as well as improved
syntheses of loracarbef of similar genesis will be reported in
due time.
was then rinsed with Et2O (50 mL). The ®ltrate was washed
with brine (50 mL), dried over MgSO4 and concentrated in
vacuo to afford the crude 16 (19.9 g, 94%). The crude
material was puri®ed by distillation (61±638C at 4 mmHg)
to afford 16 as a clear oil (13.2 g, 62.5%). 1H NMR (CDCl3,
300 MHz): d 2.62±2.66 (t, 2H), 2.79±2.83 (t, 2H), 3.70 (s,
3H), 9.82 (s, 1H).
20. Dimethyl 3-oxo-adipate (19) (5.0 g, 26.6 mmol) was
dissolved in AcOH (10.0 mL), treated with H2O (5 mL)
and cooled to 08C. The material was diazotized with
NaNO2 (2.8 g, 39.8 mmol) in H2O (12 mL) by stirring at
0±58C for 4 h. The reaction mixture was diluted with
CH2Cl2 (50 mL), extracted with 5% NaHCO3 (3£25 mL),
and dried over MgSO4. Concentration in vacuo afforded 20
as a yellow oil (5.1 g, 88.4%). 1H NMR (CDCl3, 300 MHz):
d 2.67±2.71 (t, 2H), 3.15±3.19 (t, 2H), 3.71 (s, 3H), 3.90 (s,
3H). 13C NMR (CDCl3, 75 MHz): d 27.3, 32.4, 52.2, 52.3,
150.0, 161.9, 174.1, 193.7.
21. Compound 20 (1.0 g, 46.5 mmol) was dissolved in a
mixture of MeOH (21 mL) and 1N HCl (4.9 mL,
49.0 mmol) and added to a Parr bottle containing 10%
palladium on carbon (200 mg) under a nitrogen atmosphere.
The resulting mixture was hydrogenated at 50 psi for 21 h
using a Parr shaker. The reaction mixture was ®ltered
through a pad of celite and concentrated in vacuo to afford
21 as a white solid (1.1 g, 100%). 1H NMR (D2O,
300 MHz): d 1.86±2.07 (m, 2H), 2.25±2.40 (m, 1H),
2.50±2.65 (m, 1H), 3.16 (s, 1H), 3.18 (s, 3H). 13C NMR
(D2O, 75 MHz): 30.3, 37.6, 55.1, 57.1, 166.9, 177.9,
179.4, 201.2.
Experimental
All reactions were carried out under a nitrogen atmosphere.
1
Reactions were monitored by HPLC, TLC or GC. H and
13C NMR spectra were recorded on Bruker AC300 or GE
QE300 instruments. IR-spectra were recorded on a Perkin±
Elmer 281. Elemental analyses were performed by Eli Lilly
Physical Chemistry Laboratories.
22. Compound 21 (0.46 g, 19.2 mmol) was dissolved in a
mixture of MeOH (15 mL) and H2O (10 mL) and added to a
Parr bottle containing platinum oxide hydrate (30 mg). The
resulting mixture was hydrogenated at 50 psi for 15 h using
a Parr apparatus. The reaction mixture was ®ltered through a
pad of celite and concentrated in vacuo to afford 22 as a
thick oil (0.46 g, 98%). 1H NMR (D2O, 300 MHz): d 1.17±
1.28 (m, 2H), 1.75±1.88 (m, 2H), 2.96 (s, 3H), 3.13 (s, 3H),
3.35±3.43 (m, 1H), 3.51 (d, 1H). 13C NMR (D2O, 75 MHz):
d 30.3, 32.8, 54.9, 56.2, 59.9, 71.6, 118.7, 170.7, 178.7,
180.1.
18. 4-Pentenoic acid (17, 25.0 g, 250 mmol) was cooled to
08C and treated with oxalyl chloride (24.0 mL, 275 mmol)
over a period of 0.75 h. The resulting mixture was stirred at
0±58C for 1 h and then warmed to 35±408C, during which
time gas evolution was evident. After evolution of gas had
ceased, the mixture was cooled to 08C and treated with
MeOH (40.6 mL, 999 mmol). This mixture was stirred at
0±58C for 1 h and then warmed to room temperature. After
15 h, a distillation apparatus was put in place and the excess
methanol was removed via distillation while keeping the pot
temperature below 708C. The resulting residue was
dissolved in Et2O (150 mL), washed with 5% NaHCO3
(50 mL), brine (50 mL), and dried over MgSO4. The solvent
was removed in vacuo with no external heating of the bath
23. Compound 22 (3.07 g, 12.7 mmol) was dissolved in H2O
(20 mL) and cooled to 108C. 1N NaOH (38.1 mL, 38.1
mmol) was added dropwise over 0.5 h and the resulting
mixture stirred at 10±158C for 6 h. The reaction mixture
was concentrated in vacuo to a thick oil (4.25 g). The
crude material was slurried in MeOH (75 mL) for 4 h at
room temperature, ®ltered, and the resulting white solids
were rinsed with MeOH (25 mL). Vacuum drying afforded
3.02 g of methanolate solvate of 23 (93.9%). 1H NMR (D2O,
300 MHz): d 1.65±1.78 (m, 2H), 2.20±2.43 (m, 2H), 3.35
(s, 3H), 3.40±3.89 (d, 1H), 3.78±3.88 (m, 1H). 13C NMR
(D2O, 75 MHz): d 185.7, 181.6, 75.5, 63.1, 51.5, 36.9, 30.9.
1
to afford 20.9 g (73.3%) of 18 as a clear liquid. H NMR
(CDCl3, 300 MHz): d 2.33±2.45 (m, 4H), 3.67 (s, 3H),
4.98±5.09 (m, 1H), 5.76±5.89 (m, 2H).
16. Compound 18 (20.8 g, 182 mmol) was dissolved in
CH2Cl2 (65 mL) and treated with Sudan III (5 mg). The
solution was ozonized at 2788C until the disappearance
of red color. Dimethyl sul®de (40.1 mL, 547 mmol) was
added and the mixture allowed to warm to room temperature
while stirring overnight. The solvent was removed in vacuo
and the crude material dissolved in Et2O (150 mL). This
solution was ®ltered through a pad of ¯orisil (50 g), which
24. Compound 23 (1.78 g, 7.0 mmol) was dissolved in 1N
HCl (21.0 mL, 21 mmol) and heated at re¯ux for 7 h. The
mixture was concentrated in vacuo to afford 2.3 g of a thick
oil. The crude material was slurried in 10:1 acetone/H2O
(25 mL), ®ltered, and dried in vacuo to afford 1.28 g

5674
B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
1
1
(93.4%) of a white solid. H NMR (D2O, 300 MHz): d
2.20±2.49 (m, 2H), 2.67±2.78 (m, 2H), 4.35±4.37 (d,
1H), 5.08±5.14 (m, 1H),
yield). H NMR (300 MHz DMSO-d6) d 1.64±1.77 (m,
2H), 2.38±2.45 (m, 2H), 3.65±3.72 (m, 1H), 4.57 (s, 2H),
5.13 (dd, 1H, Jˆ4.7 Hz), 6.93±6.98 (m, 3H), 7.27±7.32 (m,
2H), 8.35 (s,1H), 8.89 (d,1H, Jˆ8.9 Hz).
37b. Anal. Calcd for C14H17NO7: C 54.02, H 5.50, N 4.50.
Found C 57.47, H 5.15, N 4.74; 1H NMR (300 MHz,
DMSO-d6) d 1.45±1.75 (m, 2H), 2.10±2.40 (m, 2H),
3.60±3.70 (m, 1H), 3.95 (dd, 1H), 5.00 (s, 2H), 7.23±7.50
(m, 5H); 13C NMR (75 MHz, DMSO-d6) d 28.3, 30.2, 59.3,
65.5, 69.7, 127.6, 127.7, 128.3, 136.9, 156.0, 171.9, 174.4.
57. A solution of glycine (262.5 g, 3.5 mol), pyridoxal phos-
phate (0.92 g, 3.5 mmol), SHMT (35 g, ,0.35 mmol) in 3 L
of water was prepared, the pH was adjusted to 7 with 5%
ammonium hydroxide and then diluted to 3.5 L total volume
with water. 4-Pentenal (81.6 g, 0.97 mol) was added over
5 h at 158C and a pH of 6.9±7.1. After stirring an additional
2 h, the enzyme was removed from the reaction mixture by
concentration on a 30,000 MW ultra®lter followed by dia-
®lteration with 2.8 L of 0.01 M phosphate buffer (pHˆ6).
The resulting product ®ltrate was evaporated under vacuum
to a volume of ,5 L to remove unreacted 4-pentenal.
Excess glycine was removed from the product solution
which contained 57 (124 g) and the corresponding threo
isomer (11 g), by adsorption onto 2.5 L of SP207 resin
followed by washing with 1.5 L of water and elution with
5 L of 15% methanol. The SP207 eluent was concentrated
under vacuum to a volume of 1.25 L. 2-Methoxyethanol
(2.5 L) was added and the solution again concentrated to
1.25 L. The solid was ®ltered and washed with 2-methoxy
ethanol (600 mL), and dried overnight at 608C to yield the
desired product (127 g, 82% yield, 93:7 allo:threo ratio). 1H
NMR (300 MHz DMSO-d6) d 1.50±1.80 (m, 2H), 2.00±
2.40 (m, 2H), 3.92 (d, 1H), 4.00±4.20 (m, 1H), 5.00±5.20
(m, 2H), 5.80±6.00 (m, 1H).
37a. Anal. Calcd for C14H17NO7: C 54.02, H 5.50, N 4.50.
Found C 53.97, H 5.51, N 4.51. 1H NMR (300 MHz,
DMSO-d6) d 1.60±1.80 (m, 2H), 2.20±2.43 (m, 2H),
3.65±3.82 (m, 1H), 4.35 (dd, 1H), 4.58 (s, 2H), 6.85±7.00
(m, 3H), 7.22±7.28 (m, 2H), 8.10 (d, 1H); 13C NMR
(75 MHz, DMSO-d6) d 28.4, 30.2, 56.9, 66.6, 69.9, 114.6,
114.7, 121.1, 129.4, 157.6, 167.6, 171.3, 174.3.
39. Anal. Calcd for C16H22N2O6: C 56.80, H 6.55, N 8.28, O
28.37. Found: C 51.92, H 6.04, N 8.59, O 26.12; H NMR
1
(300 MHz, DMSO-d6) d 1.60±1.80 (m, 2H), 2.20±2.50 (m,
2H), 2.80 (s, 3H), 2.95 (s, 3H), 3.65±3.80 (m, 1H), 4.0 (dd,
1H), 5.05 (s, 2H), 7.28±7.40 (m, 5H), 13C NMR (75 MHz,
DMSO-d6) d 27.8, 28.6, 29.0, 334.0, 34.7, 36.6, 59.3, 65.4,
65.5, 70.1, 78.5, 127.6, 127.7, 128.3, 136.8, 136.9, 156.0,
172.0, 172.3.
40. Anal. Calcd for C16H23N3O5: C 56.96, H 6.87, N 12.45,
O 23.71. Found C 56.59, H 7.05, N 11.62, O 21.86; 1H NMR
(300 MHz, DMSO-d6) d 1.40±1.75 (m, 2H), 2.17±2.40 (m,
2H), 2.75 (s, 3H), 2.90 (s, 3H), 3.60 (m, 1H), 3.90 (t, 1H),
5.00 (s, 2H), 7.02 (s, 1H), 7.17 (d, 1H), 7.20±7.40 (m, 6H).
13C NMR (75 MHz, DMSO-d6) d 28.5, 28.9, 34.7, 36.6,
59.3, 65.4, 70.1, 127.5, 127.7, 128.2, 137.0, 155.8, 172.1,
172.2.
58. A slurry of compound 57 (50 g, 0.314 mole, 8% threo
isomer) in 600 mL of H2O was generated in a 2 L beaker
with a magnetic stirrer, a thermometer, a pH meter, and an
addition funnel. Phenoxyacetyl chloride (47.78 mL,
1.10 equiv.) was added to the addition funnel. The amino
acid slurry was cooled to 108C and adjusted to pH 9.0±9.5
with 5N NaOH. The resulting solution was vigorously stir-
red while the acid chloride was added dropwise over
30 min, maintaining the pH at 9.0±9.5 and the temperature
at 5±108C. The reaction was stirred until the pH stabilized
without the need for additional base. The reaction mixture
was warmed to 258C, then extracted with 250 mL of EtOAc,
which was discarded (contains trace components and any
unreacted acid chloride). A solution of 1 M H2SO4 was
slowly added over 2 h to bring the pH to 2.0 (Note: seed
the solution at the point when it turns cloudy, typically
around pH 4.6). The product slurry was stirred for an addi-
tional h, and then ®ltered. The ®lter cake was rinsed with
water and dried in a vacuum oven at 708C to afford 84.6 g of
58 (HPLC: 97%, 4% threo isomer by 1H NMR, 93%
corrected yield). Anal. Calcd for C14H17NO7: C 54.02, H
42. Anal. Calcd for C17H25N3O7S: C 49.15, H 6.06, N 10.11,
O 26.96, S 7.72. Found C 48.40, H 6.23, N 9.54, O 29.08, S
7.00; 1H NMR (300 MHz, DMSO-d6) d 2.00±2.20 (m, 2H),
2.30 (s, 3H), 2.4 (t, 2H), 3.10 (s, 3H), 3.20 (s, 3H), 4.37 (dd,
1H), 5.05 (s, 2H), 5.50 (m, 1H), 7.20±7.40 (m, 6H), 7.60 (s,
1H), 7.80 (d, 1H); 13C NMR (75 MHz, DMSO-d6) d 23.94,
30.20, 34.23, 40.95, 56.29, 65.95, 91.03, 127.83, 128.28,
136.59, 156.50, 169.73, 179.89.
1
43. H NMR (300 MHz, DMSO-d6) d 1.85±2.00 (m, 1H),
2.05±2.20 (m, 1H), 2.30±2.60 (m, 4H), 4.17 (dd, 1H), 4.78
(dd, 1H), 5.05 (s, 2H), 7.20±7.50 (m, 7H), 7.65 (d, 1H).
55. A 100 mL three necked round-bottomed ¯ask equipped
with a magnetic stirring bar, addition funnel, thermometer
and a positive nitrogen pressure was charged with 1 M
tBuOK THF solution and anhydrous THF. To this at
2788C (dry ice/acetone bath) was added a solution of 54
dissolved in THF (10.0 mL) over a period of 4 min, while
maintaining the temperature below 2508C for 75 min. The
mixture was then quenched with tri¯uoroacetic acid
(0.15 mL) and brought to room temperature. The reaction
mixture was then concentrated to an off yellow precipitate.
To the precipitate was then added methylene chloride
(40 mL). The resulting precipitate was ®ltered and the
®ltrate concentrated to a yellow foam (120 mg, 60%
1
5.50, N 4.50. Found C 53.97, H 5.51, N 4.51; H NMR
(300 MHz, DMSO-d6) d 1.60±1.80 (m, 2H), 2.20±2.43
(m, 2H), 3.65±3.82 (m, 1H), 4.35 (dd, 1H), 4.58 (s, 2H),
6.85±7.00 (m, 3H), 7.22±7.28 (m, 2H), 8.10 (d, 1H). 13C
NMR (75 MHz, DMSO) d 28.4, 30.2, 56.9, 66.6, 69.9,
114.6, 114.7, 121.1, 129.4, 157.6, 167.6, 171.2, 174.3.
59. A solution of 58 (26.6 g; 90 mmol) in 300 mL of THF
cooled to 2508C was treated with N-methylmorpholine
(NMM, 10.2 mL, 92.8 mmol, 1 min), isobutylchloroformate
(IBCF, 12.4 mL, 95.6 mmol, 1.05 equiv., 25 min), and cold
ammonia (21 mL, 28%). The cooling bath was removed

B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5675
from the rapidly stirring slurry (exotherm observed at 258C)
which was allowed to warm to room temperature. After
concentrating under vacuum to a 95 g slurry, 300 mL of
H2O were added and the thick slurry was stirred for an
additional 2 h. Filtration, rinsing, and drying at 408C for
two days yielded 22.7 g of 59 (HPLC: 99.7%; no threo
isomer was present as indicated by 1H NMR analysis;
95% corrected yield). Anal. Calcd for C15H20N2O4: C
61.63, H 6.90, N 9.58, O 21.89. Found C 61.45, H 6.98, N
9.48, O 21.91; 1H NMR (300 MHz, DMSO-d6) d 1.20±1.50
(m, 2H), 1.88±2.20 (m, 2H), 3.62 (m, 1H), 4.23 (dd, 1H),
4.55 (s, 2H), 4.80±5.02 (m, 3H), 5.65±5.80 (m, 1H), 6.80±
7.00 (m, 3H), 7.10 (s, 1H), 7.25 (t, 2H), 7.40 (s, 1H), 7.83 (d,
1H); 13C NMR (75 MHz, DMSO-d6) d 29.4, 31.9, 56.8,
66.6, 70.1, 114.6, 121.1, 129.4, 138.6, 157.6, 167.4, 171.7,
204.3.
obtained. An analytical sample of its tetrabutylammonium
salt was made and puri®ed: Anal. Calcd for C16H36N.
C16H21N2O9S2: C 55.55, H 8.30, N 6.07, O 20.81, S
9.27. Found C 55.54, H 8.09, N 5.96, O 20.83, S
1
9.01; H NMR (300 MHz, DMSO-d6) d 0.80 (t, 12H),
1.23 (q, 8H), 1.50 (m, 10H), 1.90±2.20 (m, 2H), 2.95±
3.20 (m, 11H), 4.55 (s, 2H), 4.75 (m, 2H), 4.95 (m, 2H), 5.70
(m, 1H), 6.80±7.00 (m, 3H), 7.24 (t, 2H), 8.20 (d, 2H),
10.23 (s, 1H).
63. Three-step (mesylation, sulfamation, and b-lactam
closure), one-pot procedure: A slurry of 59 (1 g, 34 mmol)
in 4 mL of CH2Cl2 and 3.80 mL of 2,6-lutidine (32.5 mmol,
9.5 equiv.) was generated in a 25 mL, 3-neck round-bottom
¯ask equipped with a magnetic stirrer, a nitrogen bubbler, a
thermometer, and an addition funnel containing 0.40 mL of
MsCl (50 mmol, 1.5 equiv.), and another addition funnel
containing 0.93 mL of ClSO3H (140 mmol, 4.1 equiv.).
The MsCl was added over 5 min and the mixture stirred
for 2.5 h. The resulting slurry was cooled to 2308C, then
the ClSO3H was very carefully added over 15 min, keeping
the temperature between 216 to 2208C during the addition.
The thick slurry was allowed to warm to 26 to 2108C,
stirred for 2 h, then placed into the freezer at 2258C and
left overnight. An HPLC pro®le of the reaction indicated
94.1% of 62 and 1% of 60. The thick slurry was added to
10 g ice water and stirred at 0±38C for 35 min (resulting in a
pH of 0.95), then placed into a pre-chilled separatory funnel
and the layers separated. The aqueous layer was extracted
with 2£2 mL of CH2Cl2 and the organic layers combined.
Water (10 mL) was added and the biphasic mixture was
concentrated under vacuum to give 11 g of a solution,
while keeping the mixture cold at all times (a room tempera-
ture water bath was used). The cold solution (pHˆ2.34) was
poured into a beaker and 1.9 mL of 5 N KOH was added
until the pH reached 10.0 (a solid precipitated at pH 8.8).
The cold bath was removed and the reaction mixture was
allowed to warm to room temperature, and the pH adjusted
to 10 with 5N KOH. A thick slurry formed upon stir-
ring for 2.5 h. After adjustment with conc. HCl to a pH
of 5.6, the mixture was cooled with an ice bath and
5 mL of saturated KCl was added. The slurry was stir-
red for 1 h, ®ltered, and dried under vacuum at 458C
overnight to give 1.02 g of compound 63 as a monohydrate
(HPLC: 94%, corrected yieldˆ68.5% over three steps for an
average of 88.2%/step): Anal. Calcd for C15H19N2O7S: C
45.90, H 4.37, N 7.14, S 8.27. Found C 45.68, H 4.40, N
6.73, S 8.27.
60. A slurry of 59 (20.0 g, 68 mmol) in 80 mL of CH2Cl2
and 66.1 mL of 2,6-lutidine (0.567 mmol, 8.3 equiv.)
formed in a 2 L three-neck round-bottom ¯ask equipped
with a mechanical stirrer, a nitrogen bubbler, and a thermo-
meter was stirred at room temperature. Methanesulfonyl
chloride (6.88 mL, 89 mmol, 1.3 equiv.) was added to the
slurry all at once, and the mixture was allowed to exotherm
to room temperature. After 3 h, 500 mL of additional
CH2Cl2 was added and the solution was placed into a separa-
tory funnel. The organic solution was extracted with
2£250 mL of 1.5 M H3PO4, then concentrated to 60 g of a
thick slurry, which was triturated with 250 mL of Et2O and
15 mL of MeOH, stirred for 2 h, and then ®ltered. The ®lter
cake was rinsed well with Et2O, and dried at 438C in a
vacuum oven overnight to give 21.6 g of 60 (HPLC:
99.3%, corrected yield 88.2%): Anal. Calcd for
C16H22N2O6S: C 51.88, H 5.99, N 7.56, O 25.92, S 8.66.
1
Found C 50.80, H 5.88, N7.23, O 23.91, S 8.53; H NMR
(300 MHz, DMSO-d6) d 1.52±1.75 (m, 2H), 1.90±2.20 (m,
2H), 3.10 (s, 3H), 4.58 (s, 2H), 4.78 (dd, 1H), 4.85 (m, 1H),
4.90±5.06 (m, 2H), 4.75 (m, 1H), 6.82±6.95 (m, 3H), 7.25
(t, 2H), 7.40 (s, 1H), 7.75 (s, 1H), 8.27 (d, 1H); 13C NMR
(75 MHz, DMSO-d6) d 28.5, 29.0, 38.0, 53.8, 66.36, 80.6,
114.4, 114.5, 115.4, 121.0, 129.4, 137.2, 157.6, 168.0,
169.5.
62. To a solution of 2,6-lutidine (2.33 mL, 20 mmol,
7.4 equiv.) in 15 mL of CH2Cl2 was added ClSO3H
(0.66 mL, 10 mmol, 3.7 equiv.) over 20 min at 2128C.
The cold mixture was stirred for 30 min then 1.00 g of 60
(HPLC: 98%, 27 mmol) was added all at once. The cold
reaction mixture was stirred for 5 h at 210 to 2158C,
then placed in the freezer at 2258C overnight. The resulting
cold slurry was added to 15 g of ice water and kept cold
while monitoring the pH. After 1 h, the pH had stabilized at
around 1.0. The two-phase mixture was placed into a pre-
chilled separatory funnel and the layers separated. The
aqueous layer was extracted with 2£2 mL of CH2Cl2, and
the combined organic layers were placed in the freezer (or it
could be carried directly into the next step of b-lactam ring
closure). This solution was found to be stable for at least one
week. Exact yields using this process were dif®cult to deter-
mine since the product was not isolated but left as a solution
and carried on into the next step. However, based on HPLC
analyses and yields typically obtained in the b-lactam
closure, an estimated sulfamation yield of 95% was
64. Compound 63 (10 g, 24.4 mmol) and 100 mL of THF
was stirred and cooled to ,58C. To this was added 0.10 mL
of conc. HCl (1.2 mmol; ,0.05 equiv.). The mixture was
allowed to warm to 5±78C and held at this temperature
range for 1 h. HPLC analysis showed 0.75% of the starting
material remained. The reaction was quenched after 70 min
by adding 100 mL of water. The pH of the quenched reac-
tion was 1.6. THF was evaporated and the resulting crystal-
lized product was ®ltered and washed with water, dried
under vacuum at room temperature to give 6.22 g of a nearly
white solid (96.9% yield): Anal. Calcd for C15H18N2O3: C
65.68, H 6.61, N 10.21. Found C 65.93, H 6.56, N 10.33. 1H
NMR (300 MHz, DMSO-d6) d 1.35±1.60 (m, 2H), 1.82±
2.10 (m, 2H), 3.63 (m, 1H), 4.58 (s, 2H), 4.90±5.08 (m, 2H),

5676
B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5.12 (m, 1H), 5.80 (m, 1H), 6.90±7.00 (m, 3H), 7.30 (t, 2H),
8.40 (s, 1H), 8.90 (d, 1H); 13C NMR (75 MHz, DMSO-d6) d
29.3, 29.6, 52.7, 57.6, 66.5, 114.5, 115.0, 121.0, 129.3,
137.8, 157.7, 167.0, 168.1.
Acknowledgements
We are indebted to Professor Marvin J. Miller for many
insightful suggestions, Jim Kelly, Tim Mitchell and Steve
Lawrence for E coli harvesting and lysis, Bruce McGathey
and Bill Muth for E coli fermentation development, Surya
Vangala for enzyme puri®cation, Frank Nealy and Joe
Turpin for Analytical assays. We also would like to thank
Drs Jerry Thompson, Jerry Bryant, William Vladuchick,
and Thomas Eckrich for support.
65. Compound 64 (10 g, 36.5 mmol) was stirred with
100 mL of DMF while 5.18 mL (54.8 mmol) of methyl
bromoacetate was added. This was followed by addition
of 8.58 g (62.1 mmol) of powered K2CO3 and the reaction
mixture was stirred at room temperature for 48 h when
HPLC analysis showed that the alkylation reaction was
complete. The reaction mixture was diluted with 100 mL
of CH2Cl2 and washed with 5£100 mL of 1N HCl to
remove the DMF. The organic layer was stirred with
100 mL of H2O, 1.77 g (5.48 mmol) of Bu4NBr follwed
by 5N NaOH to adjust the pH to 13. After having been
stirred for 5 h, the layers were separated, and an equal
volume of CH2Cl2 was added to the aqueous layer. The
pH was lowered to 6.8 with 5N HCl and 8.67 g
(40.2 mmol) of p-nitrobenzyl bromide was added. The
mixture was stirred at room temperature for 12 h and. the
layers separated. The organic layer was washed with
100 mL of H2O and evaporated. The resulting solid was
slurried in 375 mL of 80:20 MTBE/hexane for 1 h, ®ltered
and washed sequentially with 100 mL of methyl-t-butyl-
ether and 100 mL of isopropanol to give compound 65
in 48% yield. Anal. Calcd for C24H25N3O7: C 61.66, H
References
1. Garcia-Rodriguez, J. A.; Bellido, J. L. M.; Sanchez, J. E. G. Int.
J. Antimicrob. Agents 1995, 5, 231±243.
2. Garau, J. Diagn. Microbiol. Infect. Dis. 1998, 31, 479±480.
3. The Organic Chemistry of b-Lactams, Georg, G. I., Ed.; 1993.
4. Kaufman, A. H.; Baum, J. Textb. Ocul. Pharmacol. 1997, 473±
491.
5. Marshall, W. F.; Blair, J. E. Mayo Clin. Proc. 1999, 74, 187±
195.
6. Moellering, J. R. C., Ed. Oral Cephalosporins [In: Antibiot.
Chemother. (Basel), 1995; 47], 1995.
7. Wise, R. Antibiot. Chemother. (7th Ed.) 1997, 202±255.
8. Zhang, T. Y.; Hat®eld, L. D. Compr. Heterocycl. Chem. II 1996,
1B, 591±622.
1
3.37, N 8.99. Found C 61.58, H 3.61, N 8.87. H NMR
(300 MHz, DMSO-d6) d 1.40±1.60 (m, 1H), 1.60±1.90
(m, 3H), 3.78 (m, 1H), 4.20 (q, 2H), 4.50 (s, 2H), 4.80±5.00
(m 2H), 5.15 (dd, 1H), 5.30 (s, 2H), 5.67 (m, 1H), 6.80±7.00
(m, 3H), 7.23 (t, 2H), 7.62 (d, 2H), 8.20 (d, 2H), 9.00 (d,
1H).
9. Force, R. W.; Nahata, M. C. Ann. Pharmacother. 1993, 27,
321±410.
10. Zeckel, M. L.; Therasse, D. G. Drugs Today 1995, 31, 351±
355.
11. Fassbender, M.; Lode, H.; Schaberg, T.; Borner, K.; Koeppe,
P. Clin. Infect. Dis. 1993, 16, 646±653.
66. Compound 65 (0.92 g, 2 mmol) was charged under a
nitrogen purge into a 100 mL, 3-neck ¯ask equipped with
a thermometer, a subsurface gas inlet, and a gas outlet. To
this was added 20 mL of dry acetonitrile, 2 mL of H2O (10%
of volume) and 1 mL of a 0.05 wt% solution of Sudan IIIB
dye which turned the solution red. The solution was cooled
to 2108C and ozone was slowly introduced. The red color
began to lighten slowly, and after 2 min, into a slightly
yellow one, indicating the starting material had been
converted into ozonide intermediates. The ¯ask was
removed from the cold bath and was treated with 0.68 mL
(6 mmol) of 30% H2O2. This solution was added into a
separate ¯ask containing a pH 4 buffer solution made
from 7 mL of H2O and 0.64 g of NaH2PO4 and heated to
558C. A solution of NaO2Cl (80%, 0.40 g, 3.5 mmol) in
2.8 mL of H2O was added dropwise into the reaction
mixture over 2 h via a syringe pump. The reaction mixture
was cooled to room temperature, treated with 0.2 g of
Na2SO3 to quench excess oxidant, and then acidi®ed with
concentrated HCl to pH 1. Acetonitrile was evaporated and
2 mL of Et2O was added to crystallize the product, which
was obtained in 82% yield after being ®ltered, washed
sequentially with H2O and Et2O and dried in vacuo: Anal.
Calcd for C16H20N2O5: C 56.91, H 4.78, N 8.66. Found C
12. Yamashita, K.; Shimizu, M.; Motizuki, H.; Sato, K.
Chemotherapy (Tokyo) 1993, 41, 63±69.
13. Brogden, R. N.; McTavish, D. Drugs 1993, 45, 716±736.
14. Matsukuma, I.; Yoshiye, S.; Mochida, K.; Hashimoto, Y.;
Sato, K.; Okachi, R.; Hirata, T. Chem. Pharm. Bull. 1989, 37,
1239±1244.
15. Beeter, J. B.; Hall, D. A.; Jordan, C. L.; Justice, R. M.;
Kinnick, M. D.; Morin Jr, J. M.; Paschal, J. W.; Ternansky, R. J.
Tetrahedron Lett. 1993, 34, 3051±3054.
16. Bodurow, C. C.; Boyer, B. D.; Brennan, J.; Bunnell, C. A.;
Burks, J. E.; Carr, M. A.; Doecke, C. W.; Eckrich, T. M.; Fisher,
J. W., et al., Tetrahedron Lett. 1989, 30, 2321±2324.
17. Evans, D. A.; Sjogren, E. B. Tetrahedron Lett. 1985, 26,
3783±3786.
18. Evans, D. A.; Sjogren, E. B. Tetrahedron Lett. 1985, 26,
3787±3790.
19. Briggs, B. S.; Wright, I. G.; Zmijewski, M. J.; Levy, J. N.;
Stukus, M. New J. Chem. 1994, 18, 425±434.
20. Zmijewski Jr, M. J.; Briggs, B. S.; Thompson, A. R.; Wright,
I. G. Tetrahedron Lett. 1991, 32, 1621±1622.
21. Miller, M. J.; Mattingly, P. G. Tetrahedron 1983, 39, 2563.
22. Bodurow, C. C.; Levy, J. N.; Wiitala, K. W. Tetrahedron Lett.
1992, 33, 4283±4286.
23. Aikins, J. A.; Blaszczak, L. C.; Lund, K. P.; Rizzo, J. R. EP
672656 CA:AN 1995: 987974.
1
57.11, H 5.00, N 8.60; H NMR (300 MHz, DMSO-d6) d
1.55±1.80 (m, 1H), 1.83±2.00 (m, 1H), 2.10 (t, 2H), 3.80
(m, 1H), 4.20 (q, 2H), 4.58 (s, 2H), 5.12 (dd, 1H), 5.30 (s,
2H), 6.80±7.00 (m, 2H), 7.23 (t, 2H), 7.62 (d, 2H), 8.20 (d,
2H), 9.03 (d, 1H).
24. Blaszczak, L. C.; Armour, H. K.; Halligan, N. G. Tetrahedron
Lett. 1990, 31, 5693±5696.
25. Dunigan, J.; Weigel, L. O. J. Org. Chem. 1991, 56, 6225±
6227.

B. G. Jackson et al. / Tetrahedron 56 (2000) 5667±5677
5677
26. Frazier, J. W.; Staszak, M. A.; Weigel, L. O. Tetrahedron Lett.
1992, 33, 857±860.
44. Martini, F.; Angelaccio, S.; Pascarella, S.; Barra, D.; Bossa, F.;
Schirch, V. J. Biol. Chem. 1987, 262, 5499±5509.
45. Bossa, F.; Barra, D.; Martini, F.; Schirch, L. V.; Fasella, P.
Eur. J. Biochem. 1976, 70, 397±401.
27. Ma, C.; Miller, M. J. Tetrahedron Lett. 1991, 32, 2577±2580.
28. Teng, M.; Miller, M. J. J. Am. Chem. Soc. 1993, 115, 548±554.
29. Teng, M.; Gasparski, C. M.; Williams, M. A.; Miller, M. J.
Bioorg. Med. Chem. Lett. 1993, 3, 2431±2436.
30. Teng, M.; Miller, M. J.; Nicas, T. I.; Grissom-Arnold, J.;
Cooper, R. D. G. Bioorg. Med. Chem. 1993, 1, 151±154.
31. Floyd, D. M.; Fritz, A. W.; Pluscec, J.; Weaver, E. R.;
Cimarusti, C. M. J. Org. Chem. 1982, 47, 5160±5167.
32. Cimarusti, C. M.; Bonner, D. P.; Breuer, H.; Chang, H. W.;
Fritz, A. W.; Floyd, D. M.; Kissick, T. P.; Koster, W. H.;
Kronenthal, D., et al., Tetrahedron 1983, 39, 2577±2589.
33. Miller, M. J. Tetrahedron 1996, 52, 4225±4238.
34. Guzzo, P. R.; Miller, M. J. J. Org. Chem. 1994, 59, 4862±
4867.
46. Martini, F.; Maras, B.; Tanci, P.; Angelaccio, S.; Pascarella,
S.; Barra, D.; Bossa, F.; Schirch, V. J. Biol. Chem. 1989, 264,
8509±8519.
47. Kreuzman, A. J.; Zock, J. M.; Dotzlaf, J. E.; Vicenzi, J. T.;
Queener, S. W.; Yeh, W. K. J. Ind. Microbiol. Biotechnol. 1997,
19, 369±377.
48. Aswad, D. W. Anal. Biochem. 1984, 137, 405±409.
49. Hoffman, C.; Tanke, R. S.; Miller, M. J. J. Org. Chem. 1989,
54, 3750±3751.
50. Jackson, B. G.; Gardner, J. P.; Heath, P. C. Tetrahedron Lett.
1990, 31, 6317±6320.
51. Miller, M. J.; Biswas, A.; Krook, M. A.; Woulfe, S. R.
Proceedings from the Third International Symposium on Recent
Advances in the Chemistry of b-Lactam Antibiotics; 1984; pp 305±
317.
35. Oliver, D. J.; Raman, R. J. Bioenerg. Biomembr. 1995, 27,
407±414.
36. Rao, N. A. New Trends Biol. Chem. 1991, 333±340.
37. Kisliuk, R. L. Chemtracts: Biochem. Mol. Biol. 1991, 2, 217±
220.
52. Bose, A. K.; Manhas, M. S.; Sahu, D. P.; Hegde, V. R. Can.
J. Chem. 1984, 62, 2498±2505.
38. Matthews, R. G.; Drummond, J. T. Chem. Rev. 1990, 90,
1275±1290.
39. Brody, T. Pteridines 1989, 1, 159±165.
53. Doecke, C. W.; Staszak, M. A.; Luke, W. D. Synthesis 1991,
985.
54. Schirch, L. In Advances in Enzymyology, Meister, A., Ed.;
Wiley: New York, 1982; Vol. 53, pp 83±112.
40. Rao, N. A.; Ramesh, K. S.; Ratnam, M.; Rao, D. N.;
Vijayalakshmi, D.; Bhaskaran, N. J. Sci. Ind. Res. 1987, 46,
248±260.
55. Haines, A. Methods for the Oxidation of Organic Compounds
Alkanes, Alkenes, Alkynes, and Arenes; Academic Press: London,
1985; pp 119±122, 146±148, 325±326.
41. Schirch, L. Folates Pterins 1984, 1, 399±431.
42. Lotz, B. T.; Gasparski, C. M.; Peterson, K.; Miller, M. J.
J. Chem. Soc., Chem. Commun. 1990, 1107±1109.
43. Miller, M. J. EP 460883 CA:116:233923.
56. Razumouski, S. D.; Zaikov, G. E. Ozone and Its Reactions
with Organic Compounds; Elsevier: Amsterdam, 1984; Vol. 15.
57. Dancanale, E.; Montanari, F. J. Org. Chem. 1986, 51, 567.