Design, synthesis, and evaluation of 2β-alkenyl penam sulfone acids as inhibitors of β-lactamases

Article abstract of DOI:10.1021/jm9601967

A general method for synthesis of 2β-alkenyl penam sulfones has been developed. The new compounds inhibited most of the common types of β- lactamase. The level of activity depended very strongly on the nature of the substituent in the 2β-alkenyl group. Th

Full text of DOI:10.1021/jm9601967

3712
J . Med. Chem. 1996, 39, 3712-3722
Design , Syn th esis, a n d Eva lu a tion of 2â-Alk en yl P en a m Su lfon e Acid s a s
In h ibitor s of â-La cta m a ses†
Hans G. F. Richter,* Peter Angehrn, Christian Hubschwerlen, Malgosia Kania, Malcolm G. P. Page,
J ean-Luc Specklin, and Fritz K. Winkler
Preclinical Research, F. Hoffmann-La Roche Ltd., CH-4070 Basel, Switzerland
Received March 11, 1996X
A general method for synthesis of 2â-alkenyl penam sulfones has been developed. The new
compounds inhibited most of the common types of â-lactamase. The level of activity depended
very strongly on the nature of the substituent in the 2â-alkenyl group. The inhibited species
formed with the â-lactamase from Citrobacter freundii 1205 was sufficiently stable for X-ray
crystallographic studies. These, together with UV absorption spectroscopy and studies of
chemical degradation, suggested a novel reaction mechanism for the new inhibitors that might
account for their broad spectrum of action. The (Z)-2â-acrylonitrile penam sulfone Ro 48-
1220 was the most active inhibitor from this class of compound. The inhibitor enhanced the
action of, for example, ceftriaxone against a broad selection of organisms producing â-lacta-
mases. The organisms included strains of Enterobacteriaceae that produce cephalosporinases,
which is an exceptional activity for penam sulfones.
In tr od u ction
The most widespread mechanism of bacterial resis-
tance to â-lactam antimicrobials is the expression of
â-lactamases that degrade these antibiotics.1 The â-lac-
tamases exhibit considerable diversity in the range of
antibiotics they attack and have been classified into four
classes, from A to D, on the basis of their substrate
specificity and genetics.2 Class C â-lactamases and
extended broad-spectrum class A â-lactamases are
currently of particular clinical importance because
strains of Enterobacteriaceae and Pseudomonas aerugi-
nosa producing these enzymes have been selected by the
extensive use of third-generation cephalosporins in
antibacterial chemotherapy.1-3 Thus, there is a growing
interest in the development of effective inhibitors for
these â-lactamases.4 Commercially available inhibitors
(clavulanic acid (1) or penam sulfones such as sulbactam
combination with ceftriaxone exhibited useful activity
(2a ) and tazobactam (2b)) have potent activity against
against bacteria that were producing â-lactamases.
the class A â-lactamases, but they all lack activity
against class C â-lactamases.4-6
Ch em istr y
The three-dimensional structures of several â-lacta-
Syn th esis of th e 2â-F or m yl P en a m Ester 10. 6â-
mases belonging to classes A and C have been solved
Aminopenicillanic acid was transformed into the bro-
by X-ray crystallography.7-11 However, these are of
mopenicillinate 4 by the method of Micetich16 (Scheme
limited use for the design, using computer-assisted
2). Tributylphosphine17 in methanol was found to be
molecular modeling, of new molecules based on the
superior to zinc in acetic acid for debromination to give
known types of inhibitor.12,13 The principal reason for
the â-sulfoxide 5. The disulfide 6 was obtained from 5
this is that the reactions of the inhibitors with â-lacta-
by reaction with 2-mercaptobenzothiazole according to
mases involve branched pathways with many putative
Kamiya.18 A 3:1 mixture of the penam 7 and the
intermediates (Scheme 1), few of which have been
corresponding cepham 8 was produced by reaction of 6
positively identified.6,14,15 Despite this uncertainty, we
with chloroacetic acid in the presence of silver acetate.19
reasoned that the introduction of a certain degree of
rigidity into the 2â-side chain of a penam sulfone (for
example, by incorporating a double bond, such as in 3)
might improve the binding and slow down hydrolysis.
The unseparated mixture was treated with thiourea in
the presence of pyridine to give a rather unstable penam
alcohol (9) that was crystallized from the reaction
mixture. Oxidation of 9 to the â-formyl penam ester
We report the synthesis and the properties of some
2â-alkenyl penam sulfone acid derivatives 3 that display
improved activity against class C â-lactamases. Their
The penam esters 11a -n were obtained as E/Z mixtures
10 was performed using the conditions of Swern.20
Syn th esis of 2â-Alk en yl P en a m Su lfon e Acid s 3.
by reaction of 10 with various ylides (Scheme 3). For
11g,h (R ) CN), the relative yields of Z isomer 11h and
E isomer 11g depended on the reaction conditions: In
† Dedicated to Dr. Ivan Kompis on the occasion of his retirement as
Head of Infectious Diseases Research at Hoffmann-La Roche.
* To whom correspondence should be addressed.
X Abstract published in Advance ACS Abstracts, August 1, 1996.
standard Wittig conditions the ratio of 11h to 11g was
S0022-2623(96)00196-3 CCC: $12.00 © 1996 American Chemical Society

2â-Alkenyl Penam Sulfones as â-Lactamase Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19 3713
Sch em e 1. Suggested Interactions of Penam Sulfones with â-Lactamases
Sch em e 2a
a
Reagents: (a) NaNO2, KBr, H+ then CH3CO3H, benzophenone hydrazone; (b) PBu3, MeOH; (c) mercaptobenzothiazole, ∆; (d)
ClCH2COOH, AgOAc; (e) thiourea, pyridine; (f) DMSO, oxalyl chloride. Bhd: diphenylmethyl.
1:4, in MeCN at 0 °C in the presence of LiClO4 it was
4:1, and in Wittig-Horner conditions it was 1:1. Peter-
son olefination gave only 11h but in low yield. The
allylic penam esters 11o-r were produced by the
reactions shown in Scheme 4. The unsaturated alde-
hyde 11f was reduced to the allylic alcohol 11o with
DIBAH. This alcohol was reacted with dihydropyran
to give the ether 11p . The pyridinium derivative 11q
and the protected carbamate 11r were obtained from
11o by treatment with respectively triflate anhydride
in pyridine and chloroacetyl isocyanate.
The corresponding penam sulfones 12 were synthe-
sized by oxidation with m-chloroperbenzoic acid,
KMnO4, or RuO2-NaIO4 in a two-phase system and
converted to the free acids 3 by removal of the benzhy-
dryl blocking group in m-cresol at 50 °C (Table 1).
â-lactamase. The exception was the nitrile analogue Ro
48-1220 (3h ), which had a broad spectrum of activity
encompassing most serine â-lactamases (Tables 2 and
3). Particularly striking is the relatively even coverage
by 3h of class C enzymes where other penam sulfone
acids, such as tazobactam, are ineffective (Table 3).
Although the overall affinity for high molecular
weight penicillin-binding proteins (transpeptidases) was
rather low, there was some activity against the proteins
from Gram-positive organisms (see Supporting Informa-
tion). The pyridinium derivative 3q showed relatively
good activity against transpeptidases from Gram-posi-
tive bacteria that included proteins that had mutated
to become resistant to cephalosporins (Table 4). The
activity is not sufficient to result in a useful antibacte-
rial against most pathogens. In a few organisms the
reaction with penicillin-binding proteins is sufficient to
result in intrinsic activity (data not shown), just as with
other penam sulfone acids.21
Biologica l Resu lts a n d Discu ssion
Sp ectr u m of Action of 2â-Alk en yl P en a m Su l-
fon e Acid s. The 2â-alkenyl derivatives 3 had potent
activity against most class A â-lactamases typical of
penam sulfones.5,6 Several compounds achieved a simi-
lar potency against the class C â-lactamase from Cit-
robacter freundii (Table 2) but not against more resis-
tant class C enzymes such as the Escherichia coli AmpC
The inhibitors were all able to afford protection to
â-lactamase sensitive antibiotics when the combination
was applied to organisms harboring class A â-lacta-
mases susceptible to sulbactam and tazobactam (Table
5). The activity against organisms harboring more
active class A â-lactamases like TEM-4 or relatively

3714 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19
Richter et al.
Sch em e 3a
a
Reagents employed according to general methods described in the text: (a) phosphorane (method A) or phosphonium bromide, butylene
oxide (method B); (b) NaIO4, RuO2 (method C), or KMnO4 (method D), or m-chloroperbenzoic acid (method E); (c) m-cresol, 50 °C (method
F). Physical properties are given in the Supporting Information for intermediates 11 and 12 and in Table 1 for the penam sulfone acids
3.
Sch em e 4a
a
Reagents and conditions: (a) DIBAH; (b) dihydropyran, p-TSS; (c) Tf2O, pyridine; (d) chloroacetyl isocyanate; (e) NaIO4, RuO2 (method
C); (f) NaHCO3, MeOH.
susceptible class C â-lactamases (such as the C. freundii
â-lactamase) showed more differentiation, but it was
similar to that which can be obtained with sulbactam
or tazobactam. A number of compounds, particularly
3h ,c,j (R ) CN, CONH2, and Cl, respectively), had
significantly broader spectra of action. The biggest
improvement in activity was obtained against organisms
such as Enterobacter and Pseudomonas that have more
resistant class C â-lactamases and less permeable outer
membranes. The nitrile derivative Ro 48-1220 (3h ) had
the broadest spectrum of action in combination with
other â-lactam antibiotics, and the synergy with ceftri-
axone against strains of Citrobacter and Enterobacter
lowers the MIC values to a useful level.
en yl P en a m Su lfon es. The 2â-alkenyl penam sulfones
reacted rapidly with â-lactamases to form inhibited
enzyme species that had increased UV absorbance
typical for penam sulfones6,22 (Figure 1). The subse-
quent reaction, however, resulted in products with a
different UV spectrum, which suggested that the stable
acyl intermediate formed with 2â-alkenyl penam sul-
fones may have a different structure from that formed
with tazobactam.
The kinetic parameters describing the formation of
the inhibited enzyme complexes were not very different
from those obtained with tazobactam (Table 6). The
dissociation constants observed with class C â-lacta-
mases were slightly higher than those determined for
tazobactam with the same enzymes. Thus, recognition
Mod e of In h ibition of â-La cta m a ses by 2â-Alk -

2â-Alkenyl Penam Sulfones as â-Lactamase Inhibitors
Ta ble 1. Physical Properties of the Penam Sulfone Acids 3
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19 3715
1H NMR (D2O) δ (J , Hz)
yield
c
compd methoda
(%)
elemental anal.b
IR (cm-1
)
2-Hd
5-He
6-He
CHdCHf
3a
3b
3c
3d
3e
3g
3h
3i
3j
3k
3l
3m
3n
3o
3q
3s
F
F
F
F
F
G
G
F
F
G
G
F
G
F
F
F
50 NAg
80 NA
70 NA
1782, 1715
1782, 1684
3434, 1781, 1668 4.68 5.10 (1.2, 4)
1785, 1626 4.65 5.16 (1.2, 4)
1782, 1681, 1627 4.66 5.16 (2, 4)
2240, 1782, 1629 4.62 5.15 (2.0, 4)
2222, 1782, 1629 4.80 5.19 (1.6, 4.4) 3.47 (16, 1.6) 3.73 (16, 4.4) 6.15
1780, 1626
1779, 1626
1784, 1629
4.62 5.15 (2.0, 4)
4.62 5.14 (1.4, 4)
3.46 (16, 2)
3.71 (16, 4)
6.35
6.46
6.35
7.10 (16)
3.46 (17, 1.4) 3.71 (17, 4)
3.44 (16, 1.2) 3.69 (16, 4)
3.47 (17, 1.2) 3.78 (17, 4)
3.46 (17, 2)
3.47 (16, 2.0) 3.72 (16, 4)
6.88 (16)
7.10 (16)
67
C
12H15N2O7SNa
6.95d 6.95d
27 NA
82 NA
3.72 (17, 4)
6.56
6.07
7.03 (16)
7.08 (16)
6.68 (12)
80
C
10H9N2O5SNa
32 NA
65 C9H9ClNO5SNa
NA
3.70 (17, 1)
4.88 5.10 (1, 4)
4.66 5.15 (1, 4)
4.65 5.15 (1, 4)
5.19 5.04 (1.2, 4.1) 3.42 (16, 1.2) 3.58 (16, 4.1) 6.06
4.70 5.17 (1.4)
3.44 (17, 4)
3.47 (16, 1)
3.47 (16, 1)
5.96
6.80
6.80
6.71 (8.4)
7.05 (16)
7.16 (16)
7.14 (12.5)
7.10 (16)
6.24 (16)h
6.37 (16)i
6.19 (16)j
100
98
C
C
14H13N2O5SNa
12H11N2O5S2Na 1778, 1627
3.72 (16, 4)
3.72 (16, 4)
57 NA
1781, 1623
1787, 1627
3428, 1780, 1622 4.48 5.08 (2, 4)
1780, 1625
NA
91
62
C
C
11H10N3O6SNa
10H12NO6SNa
3.47 (16, 1)
3.42 (16, 2)
3.42 (17, 1.5) 3.57 (17, 4)
3.42 (17, 2) 3.69 (17, 4)
3.73 (14, 4)
3.68 (16, 4)
7.03
5.94
6.14
5.97
55 NA
62 NA
4.52 5.10 (1.5, 4)
4.48 5.08 (1.6, 4)
a
Methods F and G refer to general procedures described in the Experimental Section; all the derivatives were Na salt, except 3q,
obtained as inner salt. C, H, N, S. c KBr. Singlet. e Doublet of doublets. f Doublet. NA ) data not available. 5.94 (doublet of triplets,
b
d
g
h
i
j
J ) 16, 0.8 Hz), 6.24 (doublet of triplets, J ) 16, 4.5 Hz). 6.37 (doublet of triplets, J ) 16, 6 Hz). 6.19 (doublet of triplets, J ) 16, 4.5
Hz).
Ta ble 2. Inhibition of â-Lactamases by Penam Sulfones
IC50 (µM)
class A
B. licheniformis 749/C
class C
compd
TEM-1
TEM-3
E. coli AmpC
C. freundii 1928
P. aeruginosa 18SH
3a
3b
3c
3d
3e
3g
3h
3j
2.9
6.0
3.4
4.4
3.9
0.32
0.32
6.6
2.3
0.54
5.4
1.7
3.6
1.8
6.5
14
13
1.2
12
10
1.3
1.1
5.6
19
13
8.5
12
0.46
0.054
0.13
0.18
0.42
0.69
0.13
0.06
0.092
0.054
0.16
0.22
0.60
0.14
0.76
61
5.3
81
72
3.0
21
1.5
1.7
46
2.7
45
5.2
48
330
10
0.37
2.3
4.3
15
4.9
2.4
5.8
1.1
7.5
0.38
1.1
15
8.9
0.54
11
7.6
75
2.5
0.15
0.70
0.21
0.6
1.0
0.41
1.3
0.49
6.9
16
3k
3l
3m
3n
3o
3q
3s
6.5
3.6
9.1
0.48
1.2
Ta ble 3. Inhibition of â-Lactamases by 3h (Ro 48-1220)
enzyme
IC50 (µM)
class
Ambler
C
(Bush33
(1)
)
source
tazobactam
Ro 48-1220
Morganella morgani U1627
C. freundii 1982
Enterobacter cloacae 908R
Ent. cloacae 6300
E. coli SNO3
P. aeruginosa 18SH
P. aeruginosa GN10362
P. aeruginosa MK1184
Bacteroides fragilis 98
Bacillus licheniformis 749/C
S. aureus PC1
0.19
0.93
1.90
85.0
18.0
0.82
2.11
36.0
98.0
0.51
0.50
0.42
0.56
0.15
0.93
0.56
0.24
0.21
0.21
0.41
3.40
1.51
0.38
0.86
1.97
3.70
0.32
0.25
1.07
0.98
0.13
0.54
0.09
0.10
0.21
0.43
24
A
A
A
(2a)
(2b)
(2b′)
RTEM-1
RTEM-2
RTEM-3
SHV-2
PSE-1
OXA-1
OXA-2
A
D
(2c)
(2d)
0.83
4.8
4000
>10000
B. fragilis 36
Xanthomonas maltophila 328
B. fragilis BF101
B
(3)
200
is not altered by the introduction of the double bond in
the 2â-side chain. For both tazobactam and the 2â-
alkenyl penam sulfones, the affinities with class C
enzymes were often markedly less than the affinity
either type of penam sulfone had with class A enzymes
(Table 6). High affinity with class C enzymes is de-
pendent on fulfilling interactions in the binding pocket
provided for the acylamino side chain of penicillins and

3716 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19
Richter et al.
Ta ble 4. Inhibition of Penicillin-Binding Proteins by Selected Penam Sulfones
IC50 (µM)
E. coli W3110a
S. aureus Schocha
St. pneumoniaeb
compd
PBP 1b
PBP 2
PBP 3
PBP 1
PBP 2
PBP 3
PBP 4
PBP 2×
503c
604c
3h
3k
3l
3m
3q
340
>10000
>10000
>10000
>10000
34
145
143
14
>340
>300
286
>300
>300
17
15
14
14
15
17
145
143
28
170
145
286
28
17
15
14
143
15
5
200
39
22
100
8.2
280
33
14.1
80
4.5
4.8
50
297
15
30
3.9
8.1
a
Penicillin binding assayed using membrane fragments and 14C-labeled benzylpenicillin.30 b Assayed using recombinant soluble protein.31
c Cephalosporin resistant mutants of PBP 2×.32
Ta ble 5. Antibacterial Activity of Ceftriaxone in Combination with Penam Sulfones
MIC of ceftriaxone (mg/L) in a 1:4 combination with the compound
E. coli
Ent. cloacae
C. freundii
P. aeruginosa
compd
908R
1928
18SH
pTEM-3
pTEM-4
none
tazobactam
128
8
8
8
4
16
2
4
1
4
64
4
4
NA
2
8
1
2
1
4
8
16
8
4
8
>128
16
>16
16
4
>16
32
16
4
4
>16
>16
16
16
16
16
64
2
NA
0.5
2
2
2
NA
1
1
4
4
4
2
2
0.5
4
<0.25
<0.25
<0.25
0.25
0.5
0.25
<0.25
<0.25
0.25
1
0.5
1
0.5
0.5
0.25
1
3a
3b
3c
3d
3e
3g
3h
3j
3k
3l
32
16
>16
8
3m
3n
3o
3q
3s
8
8
8
4
4
>16
8
Ta ble 6. Kinetic Parameters Describing the Reaction of Ro 48-1220 (3h ) with Representative â-Lactamasesa
Ro 48-1220
tazobactam
enzyme
kon (mM-1 s-1
)
KS (µM)
koff (h-1
)
kon (mM-1 s-1
)
KS (µM)
koff (h-1
)
C. freundii 1928
E. coli SNO3
TEM-3
64
56
170
139
125
377
0.56
2.3
<0.001
<0.001
0.02
76
80
230
490
110
325
0.74
5
0.05
0.2
0.03
6.1
SHV-2
0.5
a
The values for the rate of formation of the acyl-enzyme complex (kon), the half-saturation constant derived from concentration
dependence studies (KS), and the rate of breakdown of the acyl-enzyme complex (koff) were determined using published methods.28,29
cephalosporins. These comprise hydrogen bonds to
main chain NH and the side chain of asparagine 152
as well as van der Waals interactions with tyrosine 221.
Neither type of penam sulfone has a side chain that can
utilize these interactions in the orientation the molecule
must adopt for acylation to occur. Despite the disparate
dissociation constants, class A and C enzymes come to
similar apparent inhibition constants (Table 2) because
of the different rates of deacylation, class A enzymes
being significantly more rapid (Table 6).
The rate of recovery of the enzyme observed after all
the sulfone had been consumed was slower with the 2â-
alkenyl penam sulfones than with tazobactam. The
activity eventually returned to the control activity
((10%), although this could take over 2 days. The acyl-
enzyme complex formed with the 2â-alkenyl penam
sulfones appears to be considerably more stable than
those formed with sulbactam or tazobactam, particu-
larly among the class C â-lactamases. Thus, it seems
that the additional potency of these compounds stems
from a property of the acyl-enzyme.
sodium salt of 3h with a slight excess of sodium
methoxide produced a rapid change in the UV absorp-
tion spectrum (Figure 2). After long incubation (>12
h), or in the presence of a larger excess of sodium
methoxide (10 equiv), a new spectrum with increased
absorption between 260 and 320 nm was established.
Evaporation of the solvent after 15 h and gel chroma-
tography allowed the isolation of the enamine 13
together with its corresponding amino acid 14 and the
glyoxylate 15. The UV absorption spectrum of 13 in
methanol had a peak at 268 nm with a shoulder at 290
nm (Figure 2), and the absorption properties cor-
responded fairly well with the difference spectrum of
the acyl-enzyme complex formed with E. coli â-lacta-
mase. This suggests that the stable acyl intermediate
formed with â-lactamases arises through elimination of
SO2 and electronic rearrangement to give the linear
conjugated system found in 13 (Scheme 5).
The identity of the stable acyl-enzyme was confirmed
by determining the X-ray crystal structure of complexes
of 3h ,c with the class C â-lactamase from C. freundii
(Figure 3). With 3c the inhibitor molecule is well-
defined by the electron density, which is consistent with
the proposed trans-enamine structure 13. The carbonyl
The reaction of sodium methoxide with some of the
penam sulfones was studied as a model for the enzymic
reaction. Treatment of a methanolic solution of the

2â-Alkenyl Penam Sulfones as â-Lactamase Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19 3717
F igu r e 1. Changes in the UV absorption spectrum of Ro 48-
1220 during the reaction with E. coli AmpC â-lactamase: (a)
spectral changes after reaction for 0.1 (lowest curve), 0.5, 1,
1.5, and 2 (upper solid curve) min and (b) time courses of the
changes at 265 nm (thin curve) and 285 nm (thick curve).
Measurements were made using 1 µM protein and 100 µM Ro
48-1220 in 0.1 M sodium phosphate buffer, pH 7.0, according
to general methods described in ref 28.
F igu r e 2. Changes in the UV absorption spectrum of Ro 48-
1220 (3h ) during methanolysis. (a) Treatment of 100 nmol of
the sodium salt of Ro 48-1220 with 1.2 equiv of sodium
methoxide in 1 mL of methanol. Temperature was maintained
at 30 °C in a thermostated cuvette holder. Scans were taken
at the times (in min) shown. (b) Comparison of the UV
absorption spectrum of the isolated enamide 13 in methanol
(solid line) with the difference spectrum of the reaction product
of Ro 48-1220 with E. coli AmpC â-lactamase (dotted line).
The difference spectrum was obtained using split cuvettes (0.8
cm total path length). Both sample and reference cuvettes
contained 1 µM â-lactamase and 100 µM Ro 48-1200 (as for
the experiment shown in Figure 1) in separate compartments.
The reaction was initiated by mixing the contents of the
sample cuvette and allowed to continue until no further change
at 265 nm was observed.
oxygen of the acyl ester is located in the oxyanion
hole7-11 and forms two hydrogen bonds with the main
chain NH of Ser64 and Ser318. The carboxamide group
appears engaged in two hydrogen bonds with the
protein, one with its oxygen to the main chain NH of
Gly320 and the other with its NH2 group to the Oγ of
Ser212. Despite these apparently good interactions, the
electron density indicated a second conformation of this
terminal carboxamide, as illustrated in Figure 3a, top.
The middle part of the inhibitor is in van der Waals
contact with the protein, but neither the carboxylate
substituent nor the NH of the enamine make hydrogen
bonds to the protein. The side chain of Gln120, a
candidate for such interactions at first sight, is not well-
defined and may exist in more than one conformation.
A number of favorable nonpolar contacts are made
between the middle part of the elongated inhibitor and
the protein, most noticeably between the methyl sub-
stituent of the inhibitor and the side chains of Thr319
and Tyr221.
location in the active site were very similar to those
observed with 3c (Figure 3).
Chen and Herzberg15 have described the X-ray crystal
structure of an intermediate complex formed during the
reaction of clavulanic acid with the class A â-lactamase
from Staphylococcus aureus. Clavulanic acid undergoes
a similar set of reactions as the penam sulfones,
resulting in the formation of the enamine.4 The acyl-
enzyme complex, trapped at low temperature, contained
electron density that could be interpreted as a mixture
of the cis- and trans-enamines. The elongated trans-
enamine was placed in the same orientation in the
active site as the acyl moieties of the penam sulfones
described here, while the cis-isomer was in the opposite,
more expected orientation. The point in the reaction
of the penam sulfones with class C â-lactamases at
For the complex with 3h , the difference density at the
oxyanion hole was as strong as that of 3c but with the
rest of the inhibitor poorly defined. Nevertheless, it was
apparent that the conformation adopted by 3h and its

3718 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19
Richter et al.
Sch em e 5
formed using Merck silica gel 60 (particle size 40-63 µm), the
fractions containing the substance of interest were pooled, and
the solvent was evaporated. Water soluble substances were
purified by gel filtration (Mitsubishi Kasei Corp. MCI gel
CHP20P 75-150 µm) using a gradient of MeCN in water. TLC
was performed on Merck TLC plates (silica gel 60 F254), and
spots were visualized with aqueous KMnO4. 1H NMR spectra
were recorded on a Bruker AC250 spectrophotometer. Chemi-
cal shifts (δ) are reported in ppm relative to Me4Si as internal
standard, and J values are given in Hz. IR spectra were
recorded from KBr pellets using a Nicolet FTIR spectrometer.
Mass spectra were determined on a MS9 spectrometer with
SS 300 Finnigan MAT data system. Elemental analyses are
indicated by the symbols of the elements; analytical results
were within 0.4% of the theoretical values. Melting points
were determined with a Buchi 510 melting point apparatus
and are uncorrected.
which the reorientation of the inhibitor moiety occurs
is not clear. Productive binding must occur with the
side chain in the opposite orientation (pointing toward
the top in Figure 3). Relaxation of cis-enamine to the
intrinsically more stable trans-isomer and the inter-
actions in the side chain binding pocket would both
provide a driving force for the reorientation.
Con clu sion
A series of 2â-alkenyl penam sulfones have been
synthesized and evaluated for their potency as inhibi-
tors of â-lactamases and as partners for a â-lactamase-
labile antibiotic. Several of these compounds, especially
Ro 48-1220, had a broad spectrum of action against
â-lactamases including both class A enzymes, normally
sensitive to such compounds, and class C enzymes,
normally resistant to such compounds. This activity
against the isolated enzymes could be realized as
potentiation of the antibacterial activity of a â-lactam
antibiotic susceptible to attack by â-lactamases. The
mechanism of inhibition underlying the extended spec-
trum of action of the 2â-alkenyl penam sulfones was
investigated by spectroscopy and X-ray crystallography.
It was demonstrated that the inhibited species of the
enzyme is a linear vinologous enamine formed by
elimination of SO2 and electronic rearrangement. Such
a species has not been hitherto suggested as an inter-
mediate in the reaction of penam sulfones with â-lac-
tamases.
1. P r ep a r a tion of th e 2â-F or m yl P en a m Ester 10.
(2S,4S,5R,6S)-6-Br om o-3,3-d im et h yl-4,7-d ioxo-4-t h ia -1-
a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zh yd r yl
Ester (4). 4 was prepared according to the method of
Micetich.16 After workup, the crystalline residue was washed
with cold AcOEt and dried, giving a yield of 44%: mp 151-
152 °C (lit.16 mp 65-70 °C); IR 1796, 1743, 1490, 1209 cm-1
;
MS 481.6 (M + NH4+). Anal. (C21H20BrNO4S) C, H, N, S.
(2S,4S,5R)-3,3-Dim eth yl-4,7-d ioxo-4-th ia -1-a za bicyclo-
[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zh yd r yl Ester (5).
To a suspension of the bromide 4 (131 g, 0.28 mol) in MeOH
(1.4 L) was added dropwise PBu3 (67.2 g, 85% grade, 0.28
mmol) while maintaining the temperature below 40 °C. The
reaction mixture was stirred at room temperature for 45 min
and then diluted with Et2O (0.3 L) and cooled to 0 °C. After
crystallization the total yield was 90%: mp 157-158 °C (lit.16
mp 145-148 °C); IR 1793, 1756, 1200 cm-1; MS 401.4 (M +
NH4+), 406.4 (M + Na+). Anal. (C21H21NO4S) C, H, N, S.
(R)-2-[(R)-2-(Ben zoth ia zol-2-yld ith io)-4-oxoa zetid in -1-
yl]-3-m eth ylbu t-3-en oic Acid Ben zh yd r yl Ester (6). 6 was
Exp er im en ta l Section
Gen er a l Syn th etic Meth od s. Solvents were dried by
filtration through Al2O3 (neutral, Brockmann, no. 1) when
necessary and stored over a bed of molecular sieves (3 Å). All
the organic solutions obtained after extraction were washed
with water and brine and dried over anhydrous MgSO4. The
solvent was evaporated under reduced pressure with a water
bath temperature below 35 °C. Chromatography was per-
prepared using
a minor modification of the method of
Micetich.16 The oily residue obtained after workup was diluted
with Et2O and crystallized by addition of Et2O/hexane (1:1).
The total yield was 84.2%: mp 76-77 °C; IR 1772, 1740, 1652,
1486, 1240 cm-1; MS 531 (M - H); NMR (CDCl3) 1.91 (s, 3H),
3.21 (dd, 1H, J ) 15, 2.3), 3.43 (dd, 1H, J ) 15, 5), 4.90 and

2â-Alkenyl Penam Sulfones as â-Lactamase Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19 3719
F igu r e 3. Structures of acyl-enzyme complexes of C. freundii class C â-lactamase with penam sulfones. (Top) Stereoview of an
omit-difference electron density map for 3c contoured at 0.15 e‚Å-3. The atoms of Ser64 and of the covalently attached inhibitor
together with water molecules in the active site area were omitted for the calculation of structure factors. The active site of only
one of the two crystallographically independent molecules is shown, but the density distribution in the other active site is
comparable. The inhibitor is emphasized as a ball-and stick-model in green and the alternative conformation of its terminal
carboxamide function in magenta. Hydrogen bonds between inhibitor and protein are indicated with red lines. Selected protein
residues are labeled at their C-R-position. (Bottom) Stereoview of an Fobs(inhibitor) - Fobs(native) difference electron density map
for 3h contoured at 3 σ. The phases of the difference map were computed with the model of the uncomplexed enzyme structure
omitting all water molecules. The manually placed model of the inhibitor is seen to account fairly well for the observed density
features.
5.00 (2s, 2 × 1H), 5.15 (br, s, 1H), 5.39 (dd, 1H, J ) 2.3, 5),
6.90 (s, 1H), 7.30-7.90 (m, 14H). Anal. (C28H24N2O3S3) C, H,
N, S.
4-th ia-1-azabicyclo[3.2.0]h eptan e-2-car boxylic Acid Ben z-
h yd r yl Ester (7) a n d (2R,3S,6R)-3-(Ch lor oa cetoxy)-3-
m e t h yl-8-oxo-5-t h ia -1-a za b icyclo[4.2.0]oct a n e -2-ca r -
boxylic Acid Ben zh yd r yl Ester (8). ClCH2CO2H (762 g,
(2S,3R,5R)-3-[(Ch lor oa cetoxy)m eth yl]-3-m eth yl-7-oxo-

3720 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19
Richter et al.
8.07 mol) was added to a suspension of AcOAg (82.1 g, 0.49
mol) in CH2Cl2 (3 L). The suspension was treated with a
solution of the disulfide 6 (172 g, 0.323 mol) in CH2Cl2 (0.3 L).
After 6 h, the suspension was filtered through Dicalite and
the solvent was evaporated. The residue was taken up in
AcOEt and filtered through Dicalite. The filtrate was cooled
to 0 °C, and the pH of the solution was brought to pH 6.5 with
NaHCO3 (a mixture of solid and saturated aqueous solution).
The organic phase was separated, and the aqueous layer was
extracted with AcOEt. The combined organic layers were
worked up and chromatographed (AcOEt/hexane, 3:7), giving
131 g (88.2%) of a foam (as a 3:1 penam/cepham mixture). This
mixture was used for the next step.
4-t h ia -1-a za b icyclo[3.2.0]h ep t a n e-2-ca r b oxylic
Acid
Ben zh yd r yl Ester (11o). To a solution of the aldehyde 11f
(5.80 g, 14.2 mmol) in toluene (0.24 L) was added dropwise,
under argon and at 0 °C, a solution of DIBAH in toluene (14.2
mL, 20%, 21.3 mmol). After stirring at room temperature for
6 h, the reaction was quenched with saturated aqueous NH4-
Cl (0.15 L) and the mixture extracted with CH2Cl2. The
organic layer was worked up and chomatographed (AcOEt/
hexane, 9:16) yielding 1.90 g (32%) of 11o: IR (film) 3480,
1778, 1750, 1202 cm-1; MS 427.6 (M + NH4)+.
(E)-(2S,3S,5R)-3-Meth yl-7-oxo-3-[3-[(R)- a n d (S)-(tet-
r a h yd r op yr a n -2-yl)oxy]p r op en yl]-4-t h ia -1-a za b icyclo-
[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zh yd r yl Ester (11p ).
To a solution of the alcohol 11o (620 mg, 1.50 mmol) in CH2-
Cl2 (30 mL) was added p-toluenesulfonic acid (4.5 mg, 0.024
mmol) followed by 3,4-dihydro-2H-pyran (0.25 mL, 2.70 mmol).
The solution was stirred for 1 h; then the solvent was
evaporated. The residue was chromatographed (AcOEt/hex-
ane, 9:16) yielding 700 mg (93%) of 11p : IR (film) 3031, 1782,
1749, 1257 cm-1; MS 511.6 (M + NH4)+.
(E)-(2S,3S,5R)-1-[3-[2-[(Ben zh yd r yloxy)ca r b on yl]-3-
m eth yl-7-oxo-4-th ia -1-a za bicyclo[3.2.0]h ep ta n -3-yl]a llyl]-
p yr id in iu m Tr iflu or om eth a n su lfon a te (11q). A solution
of the alcohol 11o (410 mg, 1.0 mmol) in CH2Cl2 (6 mL) was
cooled to -40 °C, and (CF3SO2)2O (0.25 mL, 1.5 mmol) was
added followed by pyridine (0.2 mL, 2.5 mmol). After 1 h at
-40 °C, the solvent was evaporated and the residue taken up
in CH2Cl2 for workup yielding 600 mg (97%) of 11q: IR 1777,
1743, 1630, 1160 cm-1; MS 471.6 (M+).
(E)-(2S,3S,5R)-3-[3-[[[(Ch lor oa cetyl)a m in o]ca r bon yl]-
oxy]p r op en yl]-3-m eth yl-7-oxo-4-th ia -1-a za bicyclo[3.2.0]-
h ep ta n e-2-ca r boxylic Acid Ben zh yd r yl Ester (11r ). Chlo-
roacetyl isocyanate (0.09 mL, 1.05 mmol) was added under
argon to a solution of the alcohol 11o (300 mg, 0.73 mmol) in
THF (10 mL). After 3 h, the solvent was evaporated and the
residue was chromatographed (AcOEt/hexane, 9:16) affording
300 mg (77%) of 11r : IR 3300, 1779, 1753, 1730, 1495, 1203
cm-1; MS 546.4 (M + NH4)+.
(2S,3R,5R)-3-(Hyd r oxym eth yl)-3-m eth yl-7-oxo-4-th ia -
1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zyd r yl
Ester (9). The mixture of 7 and 8 (total 131 g, 0.29 mol) was
dissolved in DMF (0.27 L) and cooled to 0 °C, and pyridine
(135 mL) was added. Thiourea (68.6 g, 0.9 mol) was added to
the solution, which was stirred at 0 °C until the thiourea was
completely dissolved. The reaction mixture was then allowed
to reach room temperature, and the solvent was evaporated.
The residue was dissolved in AcOEt and, after workup,
concentrated to about 0.5 L and diluted with hexane (0.5 L).
The resulting crystals were recrystallized from AcOEt/hexane.
The mother liquor was evaporated and the residue quickly
chromatographed (AcOEt/hexane, 3:7). The total yield was
74%: mp 129-130 °C; IR 3430, 1770, 1738, 1496, 1200 cm-1
;
MS 401.3 (M + NH4)+, 406.2 (M + Na)+; NMR (CDCl3) 1.24
(s, 3H), 2.22 (pseudo-t, 1H, J ) 6), 3.08 (dd, 1H, J ) 16, 1.7),
3.53-3.61 (pseudo-m, 3H), 5.37 (dd, 1H, J ) 16, 1.7), 6.94 (s,
1H), 7.30-7.38 (m, 10H). Anal. (C21H21NO4S) C, H, N, S.
(2S,3R,5R)-3-For m yl-3-m eth yl-7-oxo-4-th ia-1-azabicyclo-
[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zh yd r yl Ester (10).
A solution of oxalyl chloride (31.8 g, 0.25 mol) in CH2Cl2 (1 L)
was cooled under argon to -70 °C, and anhydrous DMSO (22.2
g, 284 mmol) was added dropwise. The solution was stirred
at this temperature for 15 min; then a solution of the alcohol
9 (64 g, 167 mmol) in CH2Cl2 (0.2 L) was added dropwise while
maintaining the temperature below -60 °C. The reaction
mixture was stirred for 3 h at -70 °C, before addition of TEA
(59.1 g, 584 mmol). The mixture was then allowed to reach
-10 °C over 20 min. The reaction was quenched with 1 N
HCl (0.8 L), and the organic layer was separated. The aqueous
layer was washed with CH2Cl2, and the solvent from the
combined organic layers was evaporated. The residue was
crystallized from Et2O/hexane affording 51.5 g of 10 (80.8%):
mp 111.8-112.8 °C; IR 1786, 1740, 1707, 1204 cm-1; MS 339.0
(M - CH2CO); NMR (CDCl3) 1.26 (s, 3H), 3.06 (dd, 1H, J )
16.2, 2), 3.54 (dd, 1H, J ) 16.2, 4), 5.34 (s, 1H), 5.42 (dd, 1H,
J ) 2, 4), 6.95 (s, 1H), 7.30-7.38 (m, 10H), 9.20 (s, 1H). Anal.
(C21H19NO4S) C, H, N, S.
3. P r ep a r a tion of th e 2â-Alk en yl P en a m Su lfon e
Ester s 12. (Z)-(2S,3S,5R)-3-(2-Cya n oeth en yl)-3-m eth yl-
4,4,7-tr ioxo-4-th ia -1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxy-
lic Acid Ben zh yd r yl E st er (12h ) (Met h od C). NaHCO3
(47.3 g, 0.56 mol) was added to a solution of NaIO4 (151.1 g,
0.71 mol) in water (1.8 L) at 0 °C. The solution was diluted
with MeCN (1 L) and CH2Cl2 (1.6 L), and RuO2 hydrate (364
mg, 0.75 mmol) was added. A solution of the penam esters
11g,h (57.1 g, 0.141 mol) in CH2Cl2 (0.45 L) was added in one
portion to the mixture which was then vigorously stirred at
room temperature for 30 min. Charcoal was added, and the
reaction mixture was diluted with brine and filtered. The
filtrate was extracted with CH2Cl2, and the combined organic
layers were worked up, chromatographed (AcOEt/hexane, 1:1),
crystallized from tert-butyl methyl ether, and recrystallized
from CH2Cl2/hexane giving 10.2 g of 12h (Z isomer). The
solvents from combined mother liquors were evaporated; the
residue was chromatographed (AcOEt/hexane, 3:7) and crys-
tallized from tert-butyl methyl ether affording 4.73 g of pure
12g (E isomer). Complete separation of the two isomers was
achieved by repeated chromatography and crystallization,
giving total yields of 11.0 g (17.8%) of 12g and 43.8 g (71.2%)
of 12h . 12g: TLC Rf ) 0.48 (AcOEt/hexane, 3:7); mp 145.5-
146.1 °C; IR (KBr) 2228, 1802, 1758, 1334, 1192 cm-1; MS
454.2 (M + NH4)+. Anal. (C23H20N2O5S) C, H, N, S. 12h :
2. P r ep a r a tion of th e 2â-Alk en yl P en a m Ester s 11. (E/
Z)-(2S,3S,5R)-3-(2-Cya n oeth en yl)-3-m eth yl-7-oxo-4-th ia -
1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid Ben zh yd r -
yl Ester (11g,h ) (Meth od A). A solution of the aldehyde 10
(33 g, 87 mmol) in MeCN (0.3 L) was added dropwise at -20
°C to a suspension of (cyanomethylene)triphenylphosphorane
(28.8 g, 95.6 mmol) and 0.4 M LiClO4 in MeCN (0.24 L). After
4 h, the solvent was evaporated. The residue was dissolved
in AcOEt, worked up, and chromatographed (CH2Cl2) giving
33.4 g of an oil (yield 95%). The two isomers were separated
by chromatography (AcOEt/hexane, 3:7) for analytical pur-
poses. 11g: TLC Rf ) 0.42 (AcOEt/hexane, 3:7); IR (KBr)
2220, 1781, 1743 cm-1; MS 422.2 (M + NH4+). 11h : TLC Rf
TLC Rf ) 0.53 (AcOEt/hexane, 3:7); mp 169-170 °C; [R]D
)
) 0.40 (AcOEt/hexane, 3:7); IR (Film) 2219, 1779, 1745 cm-1
;
+43.7° (c ) 1, CHCl3); IR 2220, 1797, 1753, 1334 cm-1; MS
MS 422.4 (M + NH4+).
435.3 (M - H)-. Anal. (C23H20N2O5S) C, H, N, S.
(E)-(2S,3S,5R)-3-Meth yl-3-[2-(1,2,4-oxadiazol-3-yl)vin yl]-
7-oxo-4-th ia-1-azabicyclo[3.2.0]h eptan e-2-car boxylic Acid
Ben zh yd r yl Ester (11n ) (Meth od B). A suspension of
[(1,2,4-oxadiazol-3-yl)methyl]triphenylphosphonium chloride23
(1.38 g, 3.6 mmol) and the aldehyde 10 (1.14 g, 3.0 mmol) in
1,2-epoxybutane (15 mL) was refluxed for 10 h. After filtra-
tion, the solvent was evaporated and the residue was chro-
matographed (CH2Cl2) giving 0.28 g (21%) of 11n : IR 1792,
1753, 1658, 1492, 1200 cm-1; MS 465.3 (M + NH4)+.
(Z)-(2S,3S,5R)-3-(2-Ca r ba m oylvin yl)-3-m eth yl-4,4,7-tr i-
oxo-4-th ia -1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid
Ben zh yd r yl Ester (12c) (Meth od D). To a solution of the
penam ester 11c (0.37 g, 0.87 mmol) in CH2Cl2 (10 mL) were
added AcOH (8 mL) and then, dropwise, a solution of KMnO4
(414 mg, 2.63 mmol) in water (30 mL). The reaction mixture
was stirred at room temperature for 30 min; then 30% H2O2
was added until the mixture was decolorized. The organic
layer was separated, and the aqueous layer was extracted with
CH2Cl2. The combined organic layers were worked up yielding
(E)-(2S,3S,5R)-3-(3-Hyd r oxyp r op en yl)-3-m eth yl-7-oxo-

2â-Alkenyl Penam Sulfones as â-Lactamase Inhibitors
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19 3721
225 mg (57%) of 12c: IR 1799, 1750, 1678, 1326 cm-1; MS
455.2 (M + H)+.
(E)-2-Am in o-5-cya n o-3-m eth ylp en t-3-en oic a cid (14):
NMR (DMSO-d6) 1.65 (s, 3H), 3.3 (d, 2H, J ) 7), 3.61 (s, 1H),
5.42 (t, 1H, J ) 7).
3-Hyd r oxy-3-su lfoxyp r op ion ic a cid m eth yl ester (15,
ten ta tively X ) OSO3Na ): NMR (DMSO-d6) 2.35 (dd, 1H, J
) 9, 14), 2.73 (dd, 1H, J ) 3, 14), 3.54 (s, 3H), 4.22 (dd, 1H, J
) 3, 9), 5.63 (br exchangeable, 1H).
6. X-r a y Cr ysta llogr a p h y. Crystals of the class C â-lac-
tamase from C. freundii were grown in 40% phosphate buffer
at pH 8.0 as described previously.8 For inhibitor binding
studies, crystals were soaked in 0.3 mL of 50% phosphate
buffer at pH 8.0 containing the respective inhibitor at a
concentration of 3-5 mM for 3 h.
(E)-(2S,3S,5R)-3-(2-Ca r ba m oylvin yl)-3-m eth yl-4,4,7-tr i-
oxo-4-th ia -1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid
Ben zh yd r yl Ester (12b) (Meth od E). A solution of m-
chloroperbenzoic acid (1.86 g, 7.55 mmol) in CH2Cl2 (30 mL)
was added dropwise at 0 °C to a solution of the penam ester
11b (0.64 g, 1.51 mmol) in CH2Cl2 (30 mL). After stirring at
room temperature for 6 h, the solution was washed with a 3%
Na2SO3 solution in saturated aqueous NaHCO3, worked up,
and crystallized from CH2Cl2/hexane affording 210 mg (30%)
of 12b: mp 186-188 °C; IR 1799, 1756, 1686, 1329 cm-1; MS
455.3 (M + H+).
(E)-(2S,3S,5R)-3-[3-(Ca r ba m oyloxy)p r op en yl]-3-m eth -
yl-4,4,7-t r ioxo-4-t h ia -1-a za b icyclo[3.2.0]h ep t a n e-2-ca r -
boxylic Acid Ben zh yd r yl Ester (12s). A solution of NaH-
CO3 (135 mg) in water (3 mL) was added under argon to a
solution of the protected carbamate 12r (450 mg, 0.80 mmol)
in THF (5 mL) and MeOH (1.7 mL). After 24 h the solvents
were evaporated, and the residue was taken up in AcOEt and
brine. The aqueous solution was extracted twice with AcOEt,
and the combined organic layers were worked up and chro-
matographed (AcOEt/hexane, 2:1) yielding 200 mg (78%) of
12s: IR 3481, 1799, 1749, 1731, 1601, 1329 cm-1; MS 484 (M
+ NH4)+.
4. P r ep a r a tion of th e 2â-Alk en yl P en a m Su lfon e Acid s
3. (E)-(2S,3S,5R)-3-(3-Hyd r oxyp r op en yl)-3-m eth yl-7-oxo-
4-th ia -1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid So-
d iu m Sa lt (3o) (Meth od F ). A solution of the penam sulfone
ester 12r (0.5 g, 0.95 mmol) in m-cresol (15 mL) was stirred
for 6 h at 50 °C. After cooling to room temperature, isobutyl
methyl ketone (50 mL) and 2 N sodium 2-ethylcaproate (0.55
mL, 1.1 equiv) were added, and the reaction mixture was
extracted twice with water (10 mL). The combined aqueous
solutions were washed with isobutyl methyl ketone and
lyophilized. The residue was dissolved in water (3 mL) and
purified by gel filtration giving 165 mg (62%) of 3o: IR 3428,
1780, 1622, 1398, 1311 cm-1; MS 274.3 (M - Na)-. Anal.
(C10H12NO6SNa) C, H, N.
Data were collected at 15 °C with a Nicolet/Xentronics area
detector mounted on an Elliott GX21 rotating copper anode
X-ray generator operated at 30 kV, 90 mA with a 0.3 mm focal
spot and a graphite monochromator. The crystal-detector
distance was 120 mm; 0.15° frames were measured for 90 s
and processed using the XDS primary data reduction pro-
gram.26 Further data reduction was performed using the
CCP4 crystallographic package (Collaborative Computing
Project No. 4, A Suite of Programs for Protein Crystallography,
Daresbury Laboratory, Warrington, England, 1994). Analysis
and refinement of the structure were performed with X-Plor
version 3.124 and Moloc27 on Indigo 2 workstations (Silicon
Graphics Inc.). Data collection and refinement statistics are
given in the Supporting Information.
As reported,8 the uncomplexed enzyme structure has been
refined at 2.0 Å resolution, and the final model comprises 2
molecules/asymmetric unit and 250 water molecules. Refine-
ment of the acyl-enzyme complex with 3c started from this
model with the water molecules omitted. After rigid body
refinement of the two enzyme molecules, the inhibitor mol-
ecules were readily fitted into a Fobs - Fcalc difference electron
density map and added to the model. Several rounds of
positional and B-factor refinement interrupted by manual
corrections to the model and the addition of well-determined
water molecules led to the final model, documented in the
Supporting Information.
(Z)-(2S,3S,5R)-3-(2-Cyan oeth en yl)-3-m eth yl-4,4,7-tr ioxo-
4-th ia -1-a za bicyclo[3.2.0]h ep ta n e-2-ca r boxylic Acid (3h )
(Meth od G). A solution of the penam sulfone ester 12h (20.2
g, 46.3 mmol) in m-cresol (60 mL) was stirred at 50 °C for 6 h.
The reaction mixture was diluted with hexane and cooled to 0
°C. The crystals were collected by filtration and dissolved in
AcOEt. The solution was treated with Fuller’s earth, filtered,
and concentrated to 0.13 L. It was then diluted with hexane
(0.13 L) and crystallized overnight: total yield 10.8 g (86.4%);
mp 166 °C; [R]D ) +23.4° (c ) 1, H2O); IR 2220, 1808, 1762,
1712, 1331 cm-1; MS 161.2 (M - COOH + SO2); NMR (DMSO-
d6) 1.86 (s, 3H), 3.31 (dd, 1H, J ) 15, 2.5), 3.72 (dd, 1H, J )
15, 5), 4.98 (s, 1H), 5.36 (m, 1H), 6.36 and 6.77 (2d, 2 × 1H, J
) 7.5), 13.92 (1H, br). Anal. (C10H10N2O5S) C, H, N, S.
F or m a tion of th e Cor r esp on d in g Sod iu m Sa lt. The
free acid 3h (0.5 g, 1.85 mmol) was added in one portion to a
stirred solution of NaHCO3 (155 mg, 1.85 mmol) in water (20
mL). After 10 min the solution was lyophilized giving 541 mg
(100%) of the hygroscopic sodium salt: IR 2222, 1781, 1629,
1394, 1323 cm-1; MS 269.2 (M - Na)-. Anal. (C10H9N2O5-
SNa) C, H, N.
5. Meth a n olysis of th e P en a m Su lfon e Acid . A solution
of the acid 3h (100 mg, 0.371 mmol) in MeOH (10 mL) was
treated with 1 M NaOMe (1.1 mL). After 48 h, the solvent
was evaporated and the residue was purified by gel filtration
giving, after lyophilization, the enamides 13 (50 mg, 51%), the
amino acid 14 (20 mg, 30%), and the glyoxylate 15 (12 mg,
17%).
(3E)-5-Cya n o-2-[[2-(m eth oxyca r bon yl)vin yl]a m in o]-3-
m eth ylp en t-3-en oic a cid (13): E isomer NMR (DMSO-d6)
1.56 (s, 3H), 3.28 (d, 2H, J ) 7), 3.46 (s, 3H), 3.80 (d, 1H, J )
4), 4.25 (b, 1H), 5.30 (t, 1H, J ) 7), 7.06 (br, 1H), 7.40 (dd, 1H,
J ) 7, 13); Z isomer NMR (DMSO-d6) 1.64 (s, 3H), 3.25 (d,
2H, J ) 7), 3.58 (s, 3H), 3.95 (d, 1H, J ) 6), 4.30 (d, 1H, J )
10), 5.38 (t, 1H, J ) 7), 6.55 (dd, 1H, J ) 10, 17), 7.05 (dd, 1H,
J ) 6, 17); IR 2250, 1671, 1605 cm-1; MS(ISN) 237.2 (M -
Na).
For the complex with 3h , starting with the coordinates of
the refined structure of the complex with 3c produced a
significantly better fit to the data than starting with those of
the uncomplexed enzyme did. Fobs - Fcalc difference maps
calculated at this stage showed only very weak density in the
area expected, by analogy with the 3c complex, to be occupied
by 3h . Refinement of this complex was not pursued further
as first attempts showed that no improvement of the free
R-factor could be obtained25 and the B-factors of the inhibitor
atoms were refined to very high values except those of the
serine acyl ester moiety (see discussion section).
In subunit A of the noncrystallographic dimer, density
differences consistent with 3h occupying a position similar to
that occupied by 3c could be seen at rather low contour levels
(1.5σ), while nothing interpretable was seen in the active site
of subunit B. However, in both subunits the highest difference
density is seen as a peak extending from Oγ of Ser 64 into the
oxyanion hole. The peak height is in both cases consistent
with full occupancy of this site. A considerably clearer
difference density is obtained if the Fourier coefficients are
taken as the differences between the observed amplitudes of
inhibitor-soaked and native crystals. This is not surprising
in view of the fact that the two data sets scale with a mean
fractional difference of 0.107 as compared to 0.20 for calculated
and observed amplitudes.
7. In Vitr o An tiba cter ia l Activity. The minimal inhibi-
tory concentrations (MICs) of the test compound or combina-
tion were determined according to the standard method by a
serial 2-fold dilution method using Mu¨ller-Hinton broth (Difco
laboratories, Detroit, MI). The inocculum size was approxi-
mately 105 cfu/mL. The MIC of a compound or a combination
was defined as the lowest concentration that prevented visible
growth of bacteria after incubation at 37 °C for 18 h.
Ack n ow led gm en t. We gratefully acknowledge the
assistance given in various phases of this work by the

3722 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 19
Richter et al.
(15) Chen, C. C. H.; Herzberg, O. Inhibition of â-Lactamase by
Clavulanate. Trapped Intermediates in Cryocrystallographic
Studies. J . Mol. Biol. 1992, 224, 1103-1113.
(16) Micetich, R. G.; Maiti, S. N.; Spevak, P.; Tanaka, M.; Yamazaki,
T.; Ogawa, K. Synthesis of 2â-Azidomethylpenicillin-1,1-dioxides
and 3â-Azido-3R-methylcepham-1,1-dioxides. Synthesis 1986,
292-296.
(17) Chern, J .-W.; Huang, M.; Tien, J .-H.; Pai, S.-H. Efficient
Preparation of Penicillanate Ester: A Reductive Debromination
of Bromopenicillanate Ester and Bromopenicillanate-S,S-dioxide
Ester with Tri-n-butylphosphine. Heterocycles 1988, 27, 1349-
1351.
(18) Kamiya, T.; Teraji, T.; Saito, Y.; Hashimoto, M.; Nakaguchi, O.;
Oku, T. Studies on â-Lactam Antibiotics. I. A Novel Conversion
of Penicillins into Cephalosporins. Tetrahedron Lett. 1973, 32,
3001-3004.
following colleagues from Hoffmann-La Roche: Dr W.
Arnold, A. Bubendorf, A. D’Arcy, I. Kohl, J . Nell, W.
Meister, B. Mensch, Dr. S. Mu¨ller, I. Pfister, M. Ritter,
and Dr. W. Vetter.
Su p p or tin g In for m a tion Ava ila ble: One table with the
physical properties of compounds 11 and 12, including 1H
NMR data, one table with the elemental analyses, one table
with IC50 values for inhibition of penicillin-binding proteins
from E. coli, S. aureus, and St. pneumoniae, and two tables
with statistical data from X-ray crystallographic analysis of
the acyl-enzyme complexes formed with compounds 3c,h (5
pages). Ordering information is given on any current mast-
head page.
(19) Baldwin, J . E.; Cobb, J . E.; Sheppard, L. N. Functionalization
of the 3â-Methyl Group of Penicillin. Tetrahedron 1987, 43,
1003-1012.
Refer en ces
(20) Tidwell, T. T. Oxidation of Alcohols by Activated Dimethyl
Sulfoxide and Related Reactions: An Update. Synthesis 1990,
857-870.
(21) Fass, R. J .; Prior, R. B. Comparative In Vitro Activities of
Piperacillin-Tazobactam and Ticarcillin-Clavulanate. Antimi-
crob. Agents Chemother. 1989, 33, 1268-1274.
(22) Kemal, C.; Knowles, J . R. Penicillanic Acid Sulfone: Interaction
with RTEM â-Lactamase from Escherichia coli at Different pH
Values. Biochemistry 1980, 20, 3688-3695.
(23) Claisse, J . A.; Foxton, M. W.; Gregory, G. I.; Sheppard, A. H.;
Tiley, E. P.; Warburton, W. K.; Wilson, M. J . 5-Unsubstituted
Acetylenic and Vinylic 1,2,4-Oxadiazoles. J . Chem. Soc., Perkin
Trans. 1 1973, 20, 2241-2249.
(1) Neu, H. C. The Crisis in Antibiotic Resistance. Science 1992,
257, 1064-1072.
(2) Ambler, R. P. The Structure of â-Lactamases. Philos. Trans. R.
Soc. London Ser. B 1980, 289, 321-331.
(3) Sanders, C. C.; Sanders, W. E. J . â-Lactam Resistance in Gram-
negative Bacteria: Global Trends and Clinical Impact. Clin.
Infect. Dis. 1992, 15, 824-839.
(4) Pratt, R. F. â-Lactamase Inhibition. In The Chemistry of
â-Lactams; Page, M. I., Ed.; Blackie Academic & Professional:
Glascow, 1992; pp 229-271.
(5) Livermore, D. M. Determinants of the Activity of â-Lactamase
Inhibitor Combinations. J . Antimicrob. Chemother. 1993, 31
(Suppl. A), 9-21.
(6) Bush, K.; Macalintal, C.; Rasmussen, B. A.; Lee, V. J .; Yang, Y.
Kinetic Interactions of Tazobactam with â-Lactamases from All
Major Structural Classes. Antimicrob. Agents Chemother. 1993,
37, 851-858.
(7) Moews, P. C.; Knox, J . R.; Dideberg, O.; Charlier, P.; Fre`re, J .-
M. â-Lactamase of Bacillus licheniformis 749/C at 2Å Resolution.
PROTEINS: Struct. Funct. Genet. 1990, 7, 156-171.
(8) Oefner, C.; D’Arcy, A.; Daly, J . J .; Gubernator, K.; Charnas, R.
L.; Heinze, I.; Hubschwerlen, C.; Winkler, F. K. Refined Crystal
Structure of â-Lactamase from Citrobacter freundii Indicates a
Mechanism for â-Lactam Hydrolysis. Nature 1990, 343, 284-
288.
(9) Lobkovsky, E.; Moews, P. C.; Lin, H.; Zhao, H.; Fre`re, J .-M.;
Knox, J . R. Evolution of an Enzyme Activity: Crystallographic
Structure at 2-Å Resolution of a Cephalosporinase from the
ampC Gene of Enterobacter cloacae P99 and Comparison with
a Class A Penicillinase. Proc. Natl Acad. Sci. U.S.A. 1993, 90,
11257-11261.
(10) Strynadka, N. C. J .; Adachi, H.; J ensen, S. E.; J ohns, K.; Sielecki,
A.; Betzel, C.; Sutoh, K.; J ames, M. N. G. Molecular Structure
of the Acyl-enzyme Intermediate in â-Lactam Hydrolysis at 1.7Å
Resolution. Nature 1992, 359, 700-705.
(11) J elsch, C.; Mourey, L.; Masson, J .-M.; Samama, J .-P. Crystal
Structure of Escherichia coli TEM-1 â-Lactamase at 1.8Å
Resolution. PROTEINS: Struct., Funct. Genet. 1993, 16, 364-
383.
(12) Toomer, C. T.; Schwalbe, C. H.; Ringan, N. S.; Lambert, P. A.;
Lowe, P. R.; Lee, V. J . Structural Studies on Tazobactam. J . Med.
Chem. 1991, 34, 1944-1947.
(13) Imtiaz, U.; Billings, E. M.; Knox, J . R.; Mobashery, S. A
Structure-Based Analysis of the Inhibition of Class A â-Lacta-
mases by Sulbactam. Biochemistry 1994, 33, 5728-5738.
(14) Brenner, D. G.; Knowles, J . R. Penicillanic Acid Sulfone: An
Unexpected Isotope Effect in the Interaction of 6R- and 6â-
Monodeuterio and of 6,6-Dideuterio Derivatives with RTEM
â-Lactamase from Escherichia coli. Biochemistry 1981, 20,
3680-3687.
(24) Bruenger, A. T.; Kuriyan, J .; Karplus, M. Crystallographic R
Factor Refinement by Molecular Dynamics. Science 1987, 235,
458-460.
(25) Bruenger, A. T. The R value: a Novel Statistical Quantity for
Assessing the Accuracy of Crystal Structures. Nature 1992, 355,
472-475.
(26) Kabsch, W. Evaluation of Single-Crystal X-ray Diffraction Data
from a Position-Sensitive Detector. J . Appl. Crystallogr. 1988,
21, 916-924.
(27) Gerber, P. R.; Mu¨ller, K. MAB: a Generally Applicable Molecular
Force Field for Structure Modelling in Medicinal Chemistry. J .
Comput.-Aided Mol. Des. 1995, in press.
(28) Page, M. G. P. The Kinetics of Non-stoichiometric Bursts of
â-Lactam Hydrolysis Catalysed by Class C â-Lactamases. Bio-
chem. J . 1993, 295, 295-304.
(29) Dubus, A.; Normark, S.; Kania, M.; Page, M. G. P. The Role of
Tyrosine 150 in Catalysis of â-Lactam Hydrolysis by AmpC
â-Lactamase from Escherichia coli Investigated by Site-directed
Mutagenesis. Biochemistry 1994, 33, 8577-8586.
(30) Then, R. L.; Kohl, I. Affinity of Carumonam for Penicillin-binding
Proteins. Chemotherapy 1985, 31, 246-254.
(31) Laible, G.; Hakenbeck, R.; Sicard, M. A.; J oris, B.; Ghuysen, J .-
M. Nucleotide sequences of the pbpx genes encoding the penicil-
lin-binding proteins 2× from Streptococcus pneumoniae R6 and
a cefotaxime-resistant mutant, C506. Mol. Microbiol. 1989, 3,
1337-1348.
(32) Laible, G.; Hakenbeck, R. Five Independent Combinations of
Mutations Can Result in Low-Affinity Penicillin-Binding Protein
2x of Streptococcus pneumoniae. J . Bacteriol. 1991, 173, 6986-
6990.
(33) Bush, K. Classification of â-Lactamases: Groups 1, 2a, 2b and
2b′. Antimicrob. Agents Chemother. 1989, 33, 264-270; Bush,
K. Classification of â-Lactamases: Groups 2c, 2d, 2e, 3 and 4.
Antimicrob. Agents Chemother. 1989, 33, 271-276.
J M9601967