Cephalosporin-derived inhibitors of beta-lactamase. Part 4: The C3 substituent.

Article abstract of DOI:10.1016/S0960-894X(02)00205-6

New C3-substituted beta-lactamase inhibitors were prepared and evaluated against representative class A and class C enzymes. It was possible to improve simultaneous inhibitory activity of both classes of serine hydrolase. Other inhibitors showed high selectivity for either the class C cephalosporinases or the class A penicillinases. This represents the first time that cephalosporin-derived inhibitors have demonstrated selectivity for the class A beta-lactamases.

Full text of DOI:10.1016/S0960-894X(02)00205-6

Bioorganic& Medicinal Chemistry Letters 12 (2002) 1663–1666
Cephalosporin-Derived Inhibitors of ꢀ-Lactamase.
Part 4: The C3 Substituent
John D. Buynak,* Lakshminarayana Vogeti, Venkata Ramana Doppalapudi,
George Martin Solomon and Hansong Chen
Department of Chemistry, Southern Methodist University, Dallas, TX 75275-0314, USA
Received 3 January 2002; accepted 8 March 2002
Abstract—New C3-substituted b-lactamase inhibitors were prepared and evaluated against representative class A and class C
enzymes. It was possible to improve simultaneous inhibitory activity of both classes of serine hydrolase. Other inhibitors showed
high selectivity for either the class C cephalosporinases or the class A penicillinases. This represents the first time that cephalo-
sporin-derived inhibitors have demonstrated selectivity for the class A b-lactamases. # 2002 Elsevier Science Ltd. All rights
reserved.
While the b-lactam antibiotics remain among the most
commonly prescribed antimicrobial products,1 the inci-
dence of resistance to these safe and effective drugs is
increasing. The most common form of such resistance is
the bacterial ability to produce b-lactamase, an enzyme
that hydrolyzes the b-lactam. More than 300 different
b-lactamases have been identified.2 These enzymes are
divided into four groups (A, B, C, and D), based on
their structure.3 Both classes A and C are currently
clinically relevant.4 One partially effective method to
overcome such resistance is to coadminister a combina-
tion of the antibioticand a b-lactamase inhibitor.5
Augmentin, the most commercially successful product,
combines the inhibitor clavulanic acid (1), with the
antibiotic amoxicillin. Other inhibitor/antibiotic combi-
nations include sulbactam (2), which is coadministered
with ampicillin in the form of Unasyn1 (injectable) or
Sultamicillin1 (oral), and tazobactam (3), which is
coadministered with piperacillin in the form of Zosyn1.
However, these products have limitations. In particular,
they are only active against class A-producing micro-
organisms, historically the most clinically relevant.6
Recently the incidence of infections by class C produc-
ing organisms, in particular, has been rising (Fig. 1).7
Figure 1. Commercial b-lactamase inhibitors.
Figure 2. Recently reported inhibitors from our group.
penicillinases as well as the class C cephalosporinases.8
A few of these inhibitors are shown in Figure 2. We
have also demonstrated the synergy of these new com-
pounds in combination with existing antimicrobial pro-
ducts in the treatment of resistant microorganisms.8a
Recently, the crystal structure of the extended spectrum
class C b-lactamase GC1, inhibited by 6 was obtained.9
By contrast to the penicillin-derived inhibitors, the
cephalosporin-derived series allows facile synthetic
modification at C3. This permitted us to produce ana-
logues with altered properties, including expanded inhi-
bitory profile. In fact, our initial cephalosporin-derived
Our group has recently reported inhibitors that possess
the ability to simultaneously inactivate both class A
*Corresponding author. Tel.: +1-214-768-2484; fax: +1-214-768-
4089; e-mail: jbuynak@mail.smu.edu
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00205-6

1664
J. D. Buynak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1663–1666
Scheme 3. Synthesis of 24 to 27: (i) R-SnBu3, cat. Pd2(dba)3; (ii) TFA,
anisole; (iii) NaHCO3.
Scheme 4. Synthesis of 30 and 31: (i) 4 equiv mCPBA; (ii) TFA, anisole;
(iii) NaHCO3.
Scheme 1. Synthesis of amides 12a to 12l: (i) Ph3P¼CHCO2All,
CH2Cl2; (ii) p-TolSO2Na, cat. Pd(PPh3)4, MeOH; (iii) R1R2NH, BOP
reagent, TEA, CH3CN, rt, 1–2 h, 70–80%; (iv) 2.5 equiv mCPBA;
(v) TFA/anisole, 0 ꢀC to rt; (vi) NaHCO3.
Scheme 5. Synthesis of 35 and 36: (i) (Me3Sn)2, cat. Pd2 (dba)3, THF;
(ii) XeF2, AgOTf, 0 ꢀC, CH2Cl2, 5 min; (iii) TFA, anisole;
(iv) NaHCO3.
As shown in Scheme 3, iodosulfone 16 could be coupled
with a number of organostannanes to produce new C3-
substituted inhibitors. The sulfur side chain of inter-
mediates 20 and 21 was also further oxidized to the sul-
fone oxidation state as shown in Scheme 4.
Scheme 2. Synthesis of C3 halides 17, 18, and 19: (i) Tf2O, DIPEA,
DCM À78 ꢀC, 85%; (ii) LiX (2.5 equiv), THF, rt 36 h, 70–90%;
(iii) PCl5, py, DCM/MeOH, 0 ꢀC, 74%; (iv) i-prONO, cat. TFA,
EtOAc; (v) propylene oxide, cat. Rh2(OOct)4, C6H6; (vi) a-py-CH2–
PPh3Cl, KO-t-Bu, THF/DCM (30% overall for steps iv, v, and vi);
(vii) mCPBA (2.5 equiv), DCM, rt, 30 min, 90%; (viii) TFA, anisole;
(ix) NaHCO3.
The fluoride 35 and the C3 unsubstituted compound 36
were prepared from the stannane 32 as shown in
Scheme 5. The production of 34 is presumably produced
by the presence of adventitious amounts of HF in the
reaction.
inhibitors (e.g., 5) themselves were initially discovered
as selective inactivators of the class C enzymes.10 Only
through modification at C3, was it possible to achieve
the goal of dual inhibition of classes A and C.8b,c We
have recently developed new synthetic methodology
leading to the preparation of a wider range of sub-
stituted C3 derivatives in this series.11 We now report
the biological activity of these new materials.
Compound 38 was prepared from 37 as shown below in
Scheme 6. Steps ii and iii are performed in the same pot
to avoid formation of the lactone.
The biological activity of these new inhibitors against
the representative class A enzyme TEM-1 and the class
C enzyme P99 are reported in Table 1.13 A survey of the
data indicates that modification of the C3 substituent
within this cephalosporin-derived group of inhibitors
has a marked effect on the inhibitory activity. The ratios
of the IC50 values in the case of the class A TEM-1:
worst inhibitor/best inhibitor=2437, and in the case of
The target molecules were prepared in a number of
ways. As shown in Scheme 1, carboxylic acid 10 or 11
(prepared as described previously8b) could be coupled with
a variety of amines to produce amides 12a–12l.
C3 halides, 17, 18, and 19 were prepared from the com-
mercially available12 13 as shown in Scheme 2.

J. D. Buynak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1663–1666
1665
Scheme 6. Synthesis of 38: (i) 2.5 equiv mCPBA, DCM, 30 min, 81%;
(ii) Bu3SnH, Pd, AcOH, DCM; (iii) Dess–Martin, DCM, NaHCO3
(72% steps ii and iii); (iv) NaClO2, H2NSO3H, MeOH/H2O=2/1,
78%; (v) TFA, anisole; (vi) NaHCO3.
Scheme 7. Proposed mechanistic pathway leading to inhibition.
the class C P99 worst/best=2172. While tazobactam is
a specific inhibitor of the class A TEM-1 enzyme, sev-
eral of these new inhibitors display outstanding inhi-
bition of both enzymes. Included in this group are 12d,
25, 30, and 31. Others display a more highly specific
inhibition of the class C P99 enzyme, including halides
17, 18, 19, and 35, the unsubstituted 36, and the
biscarboxylate 38. This series of amides also demon-
strates, for the first time, that it is possible to design
cephalosporin-derived inhibitors that have substantially
higher inhibitory activity against the class A penicillinases
than against the class C cephalosporinases (e.g., 12g).
In the case of inhibitor 6, we have crystallographically
observed the formation of stabilized acyl enzyme 39,
and proposed the mechanism shown in Scheme 7.9 It is
logical to assume that each of these compounds inacti-
vates the enzymes in a similar manner. The observed
individual IC50 values likely represent a complex inter-
play of factors. Included among these are: (1) recog-
nition in the active site (i.e., formation of the Michaelis
Complex), (2) the rate of initial acylation of the active
site serine, (3) the rate of rearrangement of the initial
acyl-enzyme to the stabilized acyl-enzyme, and (4) the
stability of this resultant acyl enzyme toward hydrolysis
(perhaps owing to its chemical structure, its placement
in the active site, and/or induced conformational chan-
ges in the enzyme).
Table 1. b-Lactamase inhibitory activity against representative class
A (TEM-1) and class C (P99) enzymes14
However, some generalizations regarding structure–
activity relationships can be made. In our first paper in
this series,15 we observed that the prototypical cephalo-
sporin C3 side chain (i.e., R=CH2OAc) produced an
inhibitor which solely targeted the class C P99 enzyme.
In the second paper, we observed a dramatic enhance-
ment of inhibitory activity (especially against the class A
enzyme, TEM-1) by placing an electronegative group at
C3 (e.g., R=CN or R=CH¼CH–COR0). The present
work illustrates that, while the trend toward improved
activity with electronegative groups is still operative
(e.g., the high activity of sulfones 30 and 31), the effect
of the C3 substituent is more complex. The unexpect-
edly high overall activity of 25, having a (relatively)
non-electronegative C3 side chain illustrates that, in
addition to electronegativity, inhibitor–enzyme recogni-
tion at C3 probably plays a role in determining activity.
Such recognition may be occurring during formation of
the initial Michaelis complex or may be further enhan-
cing binding of the chemically stabilized acyl-enzyme.
This is further illustrated by the C3 substituent of 12d,
which has roughly the same electronegativity as that of
the other (unsaturated) amides, but probably attains
improved binding due to its positively charged side
chain thereby leading to enhanced inhibition.
Compd
R
TEM-1 P99
Inhibition Inhibition
IC50 (mM) IC50 (mM)
Tazobactam
6
0.25
0.2615
0.078
0.0701
0.240
0.0083
7.69
101.6
0.022
1.18
0.212
0.824
0.0055
0.128
0.127
6.34
CH¼CH–CONH2
12a
12b
12c
12d
12e
12f
12g
12h
12i
12j
12k
12l
17
18
19
24
25
26
27
30
31
CH¼CH–CONHCH2CF3
CH¼CH–CONHCH2CH2OH
CH¼CH–CONHCH(CH2)2
CH¼CH–CONH–CH2CH2 (CN3H4)
CH¼CH–CONHOH
CH¼CH–CONHC6F5
4.28
CH¼CH–CON(CH2CH2)2NMe
0.053
1.4
0.39
CH¼CH–CONHCH2Ph
0.11
1.1
CH¼CH–CONHNH2
CH¼CH–CO-NHC6H4OH
0.11
4.2
1.59
1.82
0.791
1.28
1.065
0.012
3.54
0.035
0.31
4.2
CH¼CH–CONHCH2CO2Na
CH¼CH–CONH(CH2)3NH2
Cl
Br
0.003
0.0029
0.0047
0.059
0.0217
0.0403
0.038
0.0907
0.174
0.011
0.116
0.032
I
SMe
SPh
Thiophen-20-yl
Ph
9.14
SO2Me
SO2Ph
F
H
CO2Na
0.0076
0.0094
1.84
18.52
10.44
35
36
38

1666
J. D. Buynak et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1663–1666
A close inspection of reported b-lactamase crystal
structures reveals that, when the two enzymes (TEM-1
and P99) are superimposed, the P99 structure is more
congested in the region of the cephalosporin C3 sub-
stituent.16 This may partially explain the better activity
of very small C3 side chain inhibitors like 36 against P99
and larger C3 side chain inhibitors like 12g against
TEM-1.
5. Maiti, S. N.; Phillips, O. A.; Micetich, R. G.; Livermore,
D. M. Curr. Med. Chem. 1998, 5, 441.
6. Bush, K.; Mobashery, S. Adv. Exp. Med. Biol. 1998, 456,
71.
7. Medeiros, A. A. Clin. Infect. Dis. 1997, 24 (Suppl. 1), S19.
8. (a) Buynak, J. D.; Rao, A. S.; Doppalapudi, V. R.; Adam,
G.; Petersen, P. J.; Nidamarthy, S. D. Bioorg. Med. Chem.
Lett. 1999, 9, 1997. (b) Buynak, J. D.; Doppalapudi, V. R.;
Rao, A. S.; Nidamarthy, S. D.; Adam, G. Bioorg. Med. Chem.
Lett. 2000, 10, 847. (c) Buynak, D. D.; Doppalapudi, V. R.;
Adam, G. Bioorg. Med. Chem. Lett. 2000, 10, 853.
9. Crichlow, G. V.; Nukaga, M.; Doppalapudi, V. R.; Buy-
nak, J. D.; Knox, J. R. Biochemistry 2001, 40, 6233.
10. Buynak, J. D.; Wu, K.; Bachmann, B.; Khasnis, D.; Hua,
L.; Nguyen, H. K.; Carver, C. L. J. Med. Chem. 1995, 38,
1022.
In summary, we have now demonstrated the profound
effect of the C3 substituent upon the ability of selected
cephalosporin sulfones to inhibit b-lactamase. Depend-
ing on the substituent, the inhibitor may select for inhi-
bition of class C b-lactamases, inhibition of class A
enzymes, or become a general inhibitor of both these
serine classes. Although increasing the electronegativity
at C3 seems to improve inhibition of both serine classes,
other factors, including enzyme-inhibitor recognition,
may also play a role.
11. Buynak, J. D.; Vogeti, L.; Chen, H. Org. Lett. 2001, 3,
2953.
12. Otsuka Chemical Co.
13. (a) The Class A TEM-1 b-lactamase, derived from
Escherichia coli, was kindly provided by Dr. Timothy Palzkill
(Baylor College of Medicine). It was purified to >90%
homogeneity by the reported procedure: Cantu, C.; Huang,
W.; Palzkill, T. J. Biol. Chem. 1997, 272, 29144. (b) The Class
C enzyme, derived from Enterobacter cloacae, strain P99, was
purchased from the Center for Applied Microbiology and
Research (CAMR), Porton Down, Salisbury, UK.
14. Assay method involves 4 min incubation of a solution of
inhibitor and enzyme (0.1–1 mM in enzyme), followed by
transfer of an aliquot into a dilute solution of the substrate
nitrocefin. Hydrolysis is monitored spectrophotometrically at
480 nm for 1 min. The rate is constant throughout this period.
Error is +10%, based on multiple experiments. The purity of
all compounds and intermediates was verified by NMR.
15. Buynak, J. D.; Wu, K.; Bachmann, B.; Khasnis, D.; Hua,
L.; Nguyen, H. K.; Carver, C. C. L. J. Med. Chem. 1995, 38,
1022.
Acknowledgements
We thank the Robert A. Welch Foundation, the Texas
Higher Education Coordinating Board, the National
Institutes of Health, and the Petroleum Research Fund,
administered by the American Chemical Society, for
support of this research. We also thank Otsuka Chemi-
cal Co. for a generous gift of compound 13.
References and Notes
1. Dax, S. L. Antibacterial Chemotherapeutic Agents; Chap-
mann and Hall: London, 1997; Chapter 1.
2. Bush, K. Clin. Infect. Dis. 2001, 32, 1085.
3. (a) Ambler, R. P. Philos. Trans. R. Soc. London, B Biol. Sci.
1980, 289, 321.
16. This is true despite the fact that the class C enzymes are
generally thought of as having a larger overall binding pocket
than the class A b-lactamases. Indeed the region of the pocket
in the vicinity of C7 is larger in class C enabling the enzymes
to more readily accommodate C7-modified antibiotics,
including the third and fourth generation cephalosporins.
4. Rice, L. B.; Bonomo, R. A. Drug Resist. Updates 2000, 3,
178.