Inhibition of the RTEM-1 β-lactamase by boronic acids

Bioorgonic & Medicinal

Chemistry Leners, Vol. 4, No. 10, pp. 1229-1234. 1994 Elsevier Science Ltd Printed in Great Britain 0960-894X04 $6.00+0.00

0960-894X(94)EO133-Y

INHIBITION

OF THE RTEM-1 B_LACTAMASE

BY BORONIC

ACIDS

Richard Martin,# Marvin Gold1 and J. Bryan Jones#* Departments of Chemist# and Molecular and Mediml Genetic&, University of Toronto, 80 St. George Street, Toronto, Ontario, Canada M5S lA1

Abstract:

Inhibition constants for substituted phenyl determined. The trends and factors involved

and phenylethyl boronic acids in their binding are discussed.

have

been

Introduction The antibiotic bacterial

growth

properties

by acylation

cell wall synthesis.’ antibiotics

8-Lactamases

by catalyzing

of j3-lactamase

of penicillins

of these enzymes.2

defend

the hydrolysis

inhibitors

represents

one strategy

The most effective

continuing

for clinical reasons.

to increase,

Furthermore,

site binding requirements Boronic information

acid

lactamase inhibit

bacteria

8-lactamase

inhibitors

acid, is employed

the additional

information

represent

one

inhibitor

to be reversible

transition

was reported’,

state-analog

* Recently,

based on the 1.7

A

RTEM-1

8-lactamase,

beginning

with

group 43

inhibitors

P_ to P_

is of interest

that

drugs. has

already

provided

inhibitors that form tetrahedral in understanding

X-ray structure

compounds

1229

resistance

useful

Boronic acids have been shown by 1lB

us to examine

inhibitor studies.ss

capacities

that inhibitor studies provide about active

a breakthrough

resolution

Application

so far are themselves

With bacterial

classes of

in

effect of P_lactam

them inactive.

described

can identify new lead structures for potential

derivatives

the lethal

to inhibit involved

the &la&am-deactivating

clinically.3

of new structural

from Escherichia co/i. This prompted

the

against

for combating

identification

the active site serine of plactamases. mechanism

are due to their abilities of DD-carboxypeptidases

ring, thus rendering

in this regard, on class A and C 8-lactamases.

NMR spectroscopy

protease

growing

of the P-la&am

lactams, but only one of these, clavulanic lactamases

and cephalosporins

of the active site serine residues

the 84actamase

of the class A RTEM-1 P_

the capacities 1-22

adducts with

prepared

of boronic for previous

acids to serine

R. MARTIN et al.

1230

Phenylethyl

series

l,R=

Phenyl series

r 13, R= Ph 14, R= B

15, R= 4,R=

m Br

,,,JQ-

5D R=

o0

6, R=

16, R=

c

17, R=

N02

AMe

7, R= 18, R=

Mer

C Cl

8, R= Me

9, R=

0-

F Cl

0

0

IO, R=

19, R=

20, R =

21, R= Me0

11, R= 22, R = 12, R=

NH;!

1231

Inhibit&m of RTEM-1 P_lactamase

Results and Discussion Boronic acids l-22 were obtained as described previously@ and their inhibition constants determined using the method of Waley. 9 The assays were carried out with penicillin G (at X = 232 nm),lo 6aminopenicillanic

acid (at X = 220-262 nm)lo or nitrocefin (at X = 482 nm)ll

as substrates,

with the substrate selection being determined by the constraints imposed by the absorption spectra of the individual boronic acids. Competitive inhibition behavior was observed in each case, and the Kl’s, each determined in duplicate, were calculated using the Grafit program. The boronic acid structures l-22 are conveniently identifiable as being of the phenylethyl (I-12) or phenyl (13-22) family and the Kl data are recorded on this basis in Tables 1 and 2 respectively. Table 1: Inhibition Constants for Phenylethyl Boronic Acids l-12

2-Phenylethyl boronic acid (1) 2-(4Chlorophenyl)ethyl

29.8 f 0.7 b

boronic acid (2)

2-(4Trifluoromethylphenyl)ethyl

boronic acid (3)

41*3= 43t3=

2-(2-Naphthyl)ethyl boronic acid (4)

44f3C

2-(4Carboxyphenyl)ethyI

boronic acid (5)

4sflb

2-(3-Methoxyphenyl)ethyl

boronic acid (6)

50f3C

2+Methylphenyl)ethyl

boronic acid (7)

2-(3-Methylphenyl)ethyl

boronic acid (6)

58f3d 83f4d

2-(1-Naphthyl)ethyl boronic acid (0)

101 k6=

2-(2Chlorophenyl)ethyl

boronic acid (10)

106i7c

2-(2-Methylphenyl)ethyl

boronic acid (11)

2Cyclohexylethyl

boronic acid (12)

141 i7d 276 f 13 =

a Determined9 in duplicate in NaH2P04 (50mM) buffer, pH 7.0, 25OC Substrates: b nitrocefin 0.1 mM, 0.5% DMSO, KM = 25.5?0.5 pM C 8-aminopenicillanic acid 2.5-5 mM, KM = 160flO pM d penicillin G potassium salt 1 mM, KM = 2021 pM As Table 1 shows, all of the phenylethyl boronic acids I-12 are good inhibitors, with the best being the parent member of the.series, phenylethyl boronic acid (1) itself. Competitive inhibition was observed for all the Table 1 inhibitors with the exception of 2-(4trifluoromethylphenyl)ethyl acid (3). for which slowbinding

boronic

kinetics were manifest. For substituted phenylethyl boronic acids, a

general trend in the potencies of inhibition is that, for a given function, the Kl increases for para, meta, and ortho substitution respectively. This is most clearly seen for the methyl-substituted inhibitors 7,6, and 11. This trend is clearly structurally dictated, but graphic analyses did not permit the specific interactions responsible to be identified.

1232

R.

The

inhibition

shown in Table 2. their phenylethyl

constants,

MARTIN

all competitive,

et al.

observed

for the phenyl

The range of Kl values for these inhibitors

counterparts,

with this series exhibiting

boronic

acids 13-22 are

is seen to be much broader

than for

both the best (13) and worst (22) inhibitors of

the current study. The wider range of Kt values reflects the greater sensitivity

of the inhibition

potency

of phenyl boronic acids to the nature of the ring substituent. Table 2: Inhibition Constants

for Phenyl Boronic Acids 13-22

Inhibitor

3-Biphenyl

Ki (PM)

boronic acid (14)

28f2b

3-Bromophenyl

boronic acid (15)

83f2a

2,CDichlorophenyl 3-Nitrophenyl

152 f lob

4Methoxyphenyl

670f20a

boronic acid (21)

a nitrocefin,

subtilisin

b baminopenicillanic

hence its susceptibility

studies

with the Table

on Kl’s were proportional

to nucleophilic

to the electrophilicity

for RTEM-1 8-lactamase

The Hammett-type

inhibitors there is a linear correlation to the electrophilicity

2 inhibitors,*a

attack by the active site serine hydroxyl

validity of this concept was also analyzed 13-22.

acid, C penicillin G potassium

salt

as for Table 1.

Carlsberg

effects of the ring substituents

3700flooa

boronic acid (22)

Assay conditions

proportional

boronic acid (19)

491 f 27 =

Substrate:

However,

143*!:9b

Phenyl boronic acid (20) 3-Aminophenyl

inhibitors

121 f: 3 a

boronic acid (18)

3-Chloro-4-fluorophenyl

series

94*6b

boronic acid (16)

boronic acid (17)

3,bDichlorophenyl

In previous

13.9 f 0.4 a

boronic acid (13)

4-Bromophenyl

between

plot of Kl versus

from linearity

of the boron atom, and group.

Accordingly,

in Figure

1.

inhibitor

boronic acid (14) and 3-biphenyl cases,

the

rationalizable

favorable

program’*

the contribution

the points for the excellent

analysis

to

binding

conferred

by

their

inhibitors,

respective

with the Insight II program 12 (version

(version

2.4) identified

a favorable

binding

2.2).

since 22, factors

of the electrophilicity

boronic acid (13) lie well above the Hammett-plot

contributions

by graphics

using the Delphi

In contrast,

also.

The point for 3-aminophenyl

of the current study, other unfavorable

that we cannot identify at this time are clearly strongly overriding of its boron atom to El binding.

For most

boron atoms for the RTEM 1 t3-lactamase in the Figure 1 plot.

the

phenyl

that their potency is generally

boronic acid (22) falls well below the line while we are unable to account for this deviation, with its Kt of 3700 pM, is by far the poorest

that the

by the substituted

o is shown

the Kl’s with o, confirming

of their respective

there are three deviations

inhibition

it was noted

Cbromophenyl

line.

In both these

substituents

are

For 14, calculations

contribution

between

the

1233

Inhibition of RTEM-I P-lactamase

electronegative para-Br-substituent and the strongly electroposkive region of the g~nidinium

residue

of Arg 244 while molecular mechanics and dynamics calculations with the Discover program’* (version 2.9) revealed phenyl stacking between the 3phenyl

substituent of the inhibitor and the

hydroxyphenyl residue of Tyr 105 as a source of additional binding power. Figure 1: Hammer Plot of Phenyt Boronlc acids 13-22 s 4d 4.6

X

4.4

hot

4.2 4 Jd

-

logKI

ad a.4 a2 s 2.8 od f4

4.a

0

M

a4

a0

QI

Q

In interpreting Kt values of competitive inhibitors, the contribution of desolvation energysblc+das an inhibitor is transferred from aqueous solution to the enzyme’s active site must be considered, with desolvatton- driven El formation being most favored for hydrophobic inhibitors. On this basis, binding of the cydohexyl derivative 12 should be fadlitated over that of phenylethyl boronic acid (11).whereas the Kl data show that the converse is the case. The reason for the superior inhi~ti~

power of i was

explained by graphics analysis, which revealed that, as for 13 above, a likely explanation was advantageous considering

phenyl stacking of the aromatic group of 1 with the side chain of Tyr 105.

desolvation

In

effects, it should be noted that the highest single desolvation cost for Normally

inhibitors 1-22 is incurred in removing the carboxylate function of 5 from aqueous solution, this barrier results in poor enzyme-inhibitor ~~o~phenyle~yl

inhi~tor

binding .ab However, with &r Kl value of 49 FM. the

5 is in fact one of the better inhibitor

of the current study. Delphi’*

calculations showed that this unexpectedly strong binding is attributable to a strong electrostatic interaction between the carboxylate ion and the guanidinium function of Arg 244 that overrides the unfavorable desolvation factor opposing El formation.

R. MARTIN et al.

1234

Acknowledgments We thank the Natural Sciences and Engineering Research Council of Canada for support, the University of Toronto for a Open Doctoral Fellowship (to RM), Dr. Susan E. Jensen (University of Alberta) for the 8-lactamase-containing E. co/i. strain, Dr. Valeri Martichonok for graphics assistance, and Dr. Natalie Strynadka and Professor Michael N.G. James (University of Alberta) for the enzyme’s X-ray coordinates and helpful discussion. References

and notes

1.

Waxman, D.J., Strominger, J.L., Ann. Rev. Biochem., 1983, 52, 825-869; F&e, J-M., Joris, B., CRC Crit. Rev. Microbial., 1985, 11, 299-396; Ghuysen, J-.M., Annu. Rev. Microbial., 1991, 45, 37-67.

2.

Waley, S.G., Cartwright, S.J., Med. Res. Rev., 1983, 3, 341-382; Fisher, J., in Antimicrobial Drug Resistance, 1984, Ed. L.E. Bryan, Academic press, Orlando, Florida, 33-79; Fink, AL., Pharm. Res., 1985, 55-61; Knowles, J.R., Act. Chem. Res., 1985, 18, 97-104; Pratt, R.F., in Design of Enzyme Inhibitors as Drugs, 1989, Eds. M. Sandler and H.J. Smith, Oxford University Press, Oxford, 178205.

3.

Brown, AG.,

4.

Beesley, T., Gascoyne, N., Knott-Hunziker, V., Gundstrom, T., Biochem. J., 1983, 209, 229-233.

5.

Cromptom,

6.

Baldwin,

et a/., J. Antibiot., 1976, 29, 666-669; Brown, A.G., Howarth, T.T., King, T.J., J. Chem. Sac. Chem. Commun., 1976, 266-267; Cole, M., Reading, C., Antimicrob. Agents Chemother., 1977, 7 7, 852-857; Knowles, J.R., Fisher, J., Charnas, R.L., Biochemistry, 1978, 77, 2180-2189; Knowles, J.R., Charnas, R.L., Biochemistry, 1981, 20, 3214-3219.

I.E., Cuthbert, J.E., Claridge,

Petursson,

S.,

Waley,

S.G.,

Jaurin,

B.,

B.K., Lowe, G., Waley, S.G., Biochem. J., 1988, 251, 453-459. T.D.W.,

Derome,

A.E., Bradley,

D.S., Twyman,

M., Waley,

S.G., J.

Chem. Sot. Chem. Commun., 1991,573-574. 7.

Strynadka, N.C.J., Hiroyuki, A., Jensen, S.E., Johns, James, M.N.G., Nature, 1992, 359,700-705.

8.

(a) Jones, J.B., Keller, T.H., Seufer-Wasserthal, P., Biochim. Biophys. Res. Commun., 1991, 776, 401-405. (b) Jones, J.B., Seufer-Wasserthal, P., Martichonok, V., Keller, T.H., Chin, B., Martin, R., Bioorg. Med. Chem., 1994, 2, 35-48. (c) Wolfenden, R., Science, 1983, 222, 10871093. (d) Wolfenden, R., Kati, W.M., Act. Chem. Res., 1991, 24, 209-215.

9.

Waley, S.G., Biochem. J., 1982, 205, 631-633.

10.

Waley, S.G., Biochem. J., 1974, 139, 789-790.

11.

O’Callaghan, 283-288.

12.

Biosym Technologies,

C.H., Morris, A., Kirby, S.M., Shingler,

Inc., San Diego, CA, USA.

(Received in USA 22 March 1994; accepted 11 April 1994)

K., Sielecki,

A., Betzel,

C., Sutoh,

K.,

A.H., Antimicrob. Ag. Chemother., 1972,