Inhibition of DD -peptidases by a specific trifluoroketone: Crystal structure of a complex with the actinomadura R39 DD -peptidase

Article abstract of DOI:10.1021/bi400048s

Inhibitors of bacterial dd-peptidases represent potential antibiotics. In the search for alternatives to β-lactams, we have investigated a series of compounds designed to generate transition state analogue structures upon reaction with dd-peptidases. The

Full text of DOI:10.1021/bi400048s

Article
pubs.acs.org/biochemistry
Inhibition of DD-Peptidases by a Specific Trifluoroketone: Crystal
Structure of a Complex with the Actinomadura R39 DD-Peptidase
Liudmila Dzhekieva,† S. A. Adediran,† Raphael Herman,‡ Frederic Kerff,‡ Colette Duez,‡
́
́
Paulette Charlier,‡ Eric Sauvage,‡ and R. F. Pratt*,†
†Department of Chemistry, Wesleyan University, Lawn Avenue, Middletown, Connecticut 06459, United States
‡Centre d’Ingen
iere des Proteines, Universite de Liege, B-4000 Sart Tilman, Liege, Belgium
* Supporting Information
́
́
́
̀
̀
S
ABSTRACT: Inhibitors of bacterial DD-peptidases represent potential antibiotics.
In the search for alternatives to β-lactams, we have investigated a series of compounds
designed to generate transition state analogue structures upon reaction with DD-
peptidases. The compounds contain a combination of a peptidoglycan-mimetic
specificity handle and a warhead capable of delivering a tetrahedral anion to the
enzyme active site. The latter includes a boronic acid, two alcohols, an aldehyde, and
a trifluoroketone. The compounds were tested against two low-molecular mass class
C
DD-peptidases. As expected from previous observations, the boronic acid was
a potent inhibitor, but rather unexpectedly from precedent, the trifluoroketone
[D-α-aminopimelyl(1,1,1-trifluoro-3-amino)butan-2-one] was also very effective.
Taking into account competing hydration, we found the trifluoroketone was the
strongest inhibitor of the Actinomadura R39 DD-peptidase, with a subnanomolar (free
ketone) inhibition constant. A crystal structure of the complex between the trifluoroketone and the R39 enzyme showed that a
tetrahedral adduct had indeed formed with the active site serine nucleophile. The trifluoroketone moiety, therefore, should be
considered along with boronic acids and phosphonates as a warhead that can be incorporated into new and effective DD-peptidase
inhibitors and therefore, perhaps, antibiotics.
he bacterial DD-peptidases are of considerable impor-
Ttance in medical practice because they are the targets of
β-lactam antibiotics.1 These enzymes catalyze the final
transpeptidation reaction in the biosynthesis of bacterial cell
walls and are essential to bacterial survival. The β-lactams,
acting as mechanism-based, transition state analogue inhibitors,2−4
are precisely structured to inactivate DD-peptidases in a manner
that these enzymes have been unable to escape through
evolution of a hydrolytic pathway. Bacteria have, however, been
able to achieve resistance to β-lactams in a number of ways
unrelated to DD-peptidase active site structure and, in particular,
through evolution of β-lactamases from DD-peptidases.2,4 The
β-lactamases, unlike DD-peptidases, are able to catalyze rapid
β-lactam hydrolysis and thus destruction of their antibiotic
activity.5
The rapid evolution of β-lactamases in response to new
β-lactam antibiotics and also to β-lactam-based β-lactamase
inhibitors6−8 emphasizes the need for and stimulates the search
for DD-peptidase inhibitors that are not β-lactam-based.9−15
One obvious approach is through transition state analogues,
because such molecules should, in principle, inhibit any
enzyme.16−18 Because the reactions catalyzed by DD-peptidases
in vivo are acyl transfer reactions with a covalent acyl(serine)−
enzyme intermediate (Scheme 1), substrate-based tetrahedral
anions covalently bound to the active site serine should be good
analogues of the transition states of both acylation and deacyla-
tion steps. In principle, therefore, molecule 1, in which
“peptidoglycan” is a specific peptidoglycan or peptidoglycan-
mimetic fragment and X is a reactive moiety that generates a
tetrahedral anion upon reaction with the active site serine,
would be the inhibitor of choice.
Approaches to 1 require the optimization of both pep-
tidoglycan and X. One might expect that the former goal would
be informed by studies of the substrate specificity of these en-
zymes, where a minimal, consensual, and thus broad spectrum
peptidoglycan fragment would emerge. This goal has, however,
not yet been achieved. It is true, of course, that peptidoglycan
structure varies in detail between bacterial species, but even
given this point, no consensual structure or class of privileged
structures has emerged from consideration of DD-peptidase
substrate specificity.19,20
DD-Peptidases have been classified structurally into two
groups, the low-molecular mass (LMM) and high-molecular
mass (HMM) enzymes.21 The latter group is essential for
bacterial growth and contains the killing targets of β-lactams.
Members of the former group are also inhibited by β-lactams
Received: January 11, 2013
Revised: February 17, 2013
Published: March 13, 2013
© 2013 American Chemical Society
2128
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
Scheme 1
but are not essential for bacterial survival, in the short term at
least. Both groups of enzymes are believed to catalyze acyl
transfer reactions of peptidoglycan, viz., transpeptidase (HMM)
and carboxypeptidase and endopeptidase (LMM) reactions.
Certain members of the LMM group, subclasses B and C, have
been shown to have strong specificity for a free amine
N-terminus in peptidoglycan substrates,19,22−27 but no strong
specificity for any particular element of peptidoglycan has
been demonstrated for the HMM enzymes.13,19,24,28,29 The
design of transition state analogue inhibitors for the latter
enzymes is thus difficult.
With respect to element X of 1, some advances have been
made. In particular, boronic acids (Scheme 2) have been shown
Scheme 2
to be effective. Peptidoglycan-mimetic boronates have been
found to be potent inhibitors of LMMC enzymes.13,20
The crystal structure of a complex of one such inhibitor, 2,
with the LMMC Actinomadura R39 DD-peptidase demon-
strated the reason for its specificity.13 β-Lactam-mimetic
boronic acids (bearing the amido side chains of classical
β-lactam antibiotics rather than peptidoglycan) have also
recently been found to be quite effective, although studies
of them are limited.11,12,31
MATERIALS AND METHODS
■
Commercially available reagents and solvents were purchased
from Aldrich and Acros Organics and used without purification,
unless otherwise noted. D-Glutamic acid and N-(Z)-D-alanine
were purchased from ChemImpex. Trifluoroacetaldehyde
hydrate was purchased as a 75% solution in water from
AlfaAesar. Compounds 24 (AlfaAesar), 25, and 27 were
commercial products. Trifluoroketone 26 was synthesized
in this laboratory by A. Shilabin. (Phenylacetyl)glycyl-D-
thiolactate (14) and boronic acid 2 were synthesized as
previously described.30 The Actinomadura R39 DD-peptidase
and Bacillus subtilis PBP4a were expressed and purified as
described previously.34,35
Beyond boronates, other sources of X, employed with a
variety of serine hydrolases, have been alcohols, phosphonates,
aldehydes, and trifluoroketones, yielding, in principle, tetrahe-
dral anions 3−6, respectively.32 Some assessment of the poten-
tial of these warheads with DD-peptidases has been made pre-
viously,14,33 although no indications of great potency have been
observed. To date, the combination of motifs 3−6 with a spe-
cific peptidoglycan fragment (as in 1) has not been investigated.
In this paper, we present the synthesis and assessment of
the inhibitory activity of 7−12 versus representative LMMC
DD-peptidases. We have discovered that trifluoroketone 12
is a potent inhibitor of the Actinomadura R39 enzyme
and present a crystal structure of its complex with this enzyme.
Varian Mercury 300 and 400 MHz NMR spectrometers
were used to collect 1H, 31P, and 19F NMR spectra. Absorption
spectra and spectrophotometric reaction rates were measured
with Hewlett-Packard 8452A and 8453 spectrophotometers.
High-resolution mass spectra (HRMS) were obtained from the
Mass Spectrometry Laboratory (School of Chemical Sciences,
University of Illinois, Urbana, IL). Routine ESI mass spectra
were recorded with a Thermo LCQ Advantage instrument.
High-performance liquid chromatography purification was
achieved by a Varian ProStar unit equipped with a Varian
ProStar model 340 UV−vis detector and a Nucleosil C-18
reverse phase column (Macherey-Nagel). Flash column
2129
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
a
Scheme 3
a
Reagents and conditions: (a) BnBr, NaOH, K2CO3, H2O, reflux, 60%; (b) DIBAL-H, Et2O, −78 °C, 65%; (c) 17, THF, 25 °C, 1.5 h; (d) TFA,
CH2Cl2, 0 °C, 1 h, 90%.
a
Scheme 4
a
Reagents and conditions: (a) nitroethane, K2CO3, 50−60 °C, 3 h, 62%; (b) H2, 40 psi, 10% (w/w) PtO2, MeOH/CHCl3 (16:1), 12 h, 86%; (c) 19,
HBTU, Et3N, 25 °C, 24 h, 80%; (d) Dess−Martin periodinane, TFA, CH2Cl2, 25 °C, 3 h, 90%; (e) H2, 40 psi, 10% (w/w) Pd/C, MeOH, 12 h, 35%.
chromatography was performed on Silicycle 60 Å, 32−63 μm
silica gel.
water (0.3 mL) was added. The reaction mixture was allowed
to warm to room temperature and stirred for an additional
30 min; it was then dried over magnesium sulfate and filtered,
and the volatiles were evaporated, leaving a colorless gum, 16
(2.2 g), which was employed in the next step without puri-
Syntheses. D-6-(N,N′-Dibenzylamino)-6-(benzyoxylcar-
bonyl)-trans-hex-2-enoic Acid 19 (Scheme 3). N,N-Dibenzyl-
D-glutamic Acid Dibenzyl Ester 15. This compound was syn-
thesized following a previously reported general method.36 It
was purified by chromatography on silica gel (7:1 hexane:ethyl
acetate) producing 15 as a colorless oil in 60% yield: 1H NMR
(CDCl3, 300 MHz) δ 2.1 (q, J = 7.5 Hz, 2H), 2.43, 2.56
(d quint, J = 9, 9 Hz, 2H), 3.47 (t, J = 7.5 Hz, 1H), 3.55, 3.93
(AB q, J = 13.5 Hz, 4H), 5.26 (AB q, J = 3 Hz, 2H), 5.20, 5.32
(AB q, J = 12 Hz, 2H), 7.25−7.46 (m, 20H).
D-4-(N,N-Dibenzylamino)-4-(benzyloxycarbonyl)butanal
16. To a stirred solution of 15 (3 g, 6 mmol, 1 equiv) in 30 mL
of anhydrous diethyl ether under a nitrogen atmosphere was
added DIBAL (1 M in hexane, 6.5 mL, 1.1 equiv) slowly at
−78 °C. The reaction mixture was stirred for 15 min, and then
1
fication: H NMR (CDCl3, 300 MHz) δ 1.21 (t, J = 7.8 Hz,
2H), 2.3−2.5 (m, 2H), 3.35 (t, J = 7.2 Hz, 1H), 3.52, 3.86
(AB q, J = 13.5 Hz, 4H), 5.20, 5.26 (AB q, J = 10.8 Hz, 2H),
7.2−7.5 (m, 15H), 9.58 (s, 1H).
D-6-(N,N-Dibenzylamino)-6-(benzyloxycarbonyl)-trans-
hex-2-enoic Acid tert-Butyl Ester 18. To a stirred solution
of 16 (2.0 g, 5 mmol, 1 equiv) in dry THF (20 mL) was
added (tert-butoxycarbonylmethylene)triphenylphosphorane
17 (Aldrich) (2.6 g, 6.8 mmol, 1.2 equiv) at room temperature.
The reaction mixture was stirred for 1.5 h, after which solvent
was removed by evaporation. The crude product was purified
by chromatography on silica gel (3:1 hexane:ethyl acetate),
2130
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
1
yielding product 18 as a colorless oil (1.5 g, 60% yield): H
NMR (CDCl3, 300 MHz) δ 1.47 (s, 9H), 1.86 (m, 2H), 2.07
(m, 1H), 2.32 (m, 1H), 3.34 (t, J = 7.5 Hz, 1H), 3.53, 3.87 (AB
q, J = 13.5 Hz, 4H), 5.19, 5.25 (AB q, J = 10.8 Hz, 2H), 5.59
(d, J = 16 Hz, 1H), 6.70 (quint, J = 7.2 Hz, 1H), 7.2−7.4
(m, 15H).
D-6-(N,N-Dibenzylamino)-6-(benzyloxycarbonyl)-trans-
hex-2-enoic Acid 19. Ester 18 (0.5 g, 1 mmol) was dissolved in
dichloromethane (5 mL). To this solution, stirred in an ice
bath, was slowly added trifluoroacetic acid (6 mL). The
reaction mixture was stirred for 1 h at room temperature, after
which the solvent was evaporated. After the residue was dried
under vacuum, product 19 was obtained as a gum (0.4 g, 90%
yield): 1H NMR (CDCl3, 300 MHz) δ 1.86 (m, 2H), 2.07 (m,
1H), 2.32 (m, 1H), 3.34 (t, J = 7.5 Hz, 1H), 3.53, 3.87
(AB q, J = 13.5 Hz, 4H), 5.19, 5.25 (AB q, J = 10.8 Hz, 2H),
5.59 (d, J = 16 Hz, 1H), 6.7 (quint, J = 7.2 Hz, 1H), 7.2−7.4
(m, 15H); ES(+)MS m/z 444.4 (M + 1).
silica gel column (1:1 ethyl acetate/hexane), and product 23
was isolated in 90% yield as a colorless oil: H NMR (400
1
MHz, CDCl3) δ 1.39 (d, J = 8 Hz, 3H), 1.46 (d, J = 8 Hz, 3H),
1.75−1.96 (m, 2H), 2.06−2.18 (m, 1H), 2.24−2.36 (m, 1H),
3.31 (m, 1H), 3.49, 3.86 (AB q, J = 16 Hz, 4H), 4.11 (m, 1H),
4.88−5.06 (dd, J = 8 Hz, 1H), 5.2 (AB q, J = 16, 24 Hz, 2H),
5.70 (d, J = 6 Hz, 1H), 7.69 (m, 1H), 7.23 (s, 10H), 7.4
(s, 5H).
D-α-Aminopimelyl(1,1,1-trifluoro-3-amino)butan-2-one
12. The required final product was prepared from 23 by
catalytic hydrogenation overnight at 40 psi in 10 mL of MeOH
with 10% Pd on carbon. After reaction, the solution was
filtered, and the filtrate was evaporated to dryness to give a
colorless solid. The crude product was dissolved in water and
washed with ethyl acetate three times. The aqueous solution
was separated and freeze-dried to give a colorless solid. This
was purified by two passages through a Sephadex G10 size-
exclusion chromatography column (water eluent). Product
1
D-α-Aminopimelyl(1,1,1-trifluoro-3-amino)butan-2-one
(Scheme 4). 3-Nitro-1,1,1-trifluoro-2-butanol 20. This com-
pound was prepared following the procedure reported by
McBee et al.37
12 was isolated in 35% yield as a colorless solid: H NMR
(400 MHz, D2O) δ 1.06 (d, J = 6.8 Hz, 3H), 1.22 (m, 2H),
1.46 (m, 2H), 1.70 (m, 2H), 2.12 (t, J = 7.5 Hz, 2H), 3.56
(t, J = 6 Hz, 1H), 4.15 (q, J = 7.2 Hz, 1H); 19F NMR (300
MHz, D2O) δ −82 (CF3); HRMS (ESI) [M + H]+ found
299.1220, calcd 299.1219.
1
Product 20 was obtained as a yellow oil in 62% yield: H
NMR (300 MHz, CDCl3) δ 1.72 (d, 3H, J = 6 Hz), 3.2−3.6
(br s, 1H), 4.76−4.92 (m, 2H).
Syntheses of 7−11 are described in detail in the Supporting
Information.
3-Amino-1,1,1-trifluoro-2-butanol Hydrochloride 21. This
compound was prepared as a mixture of diastereoisomers, fol-
lowing the procedure reported by Shao et al.38 The colorless
Enzyme Kinetics Studies. Inhibition of the Actinomadura
R39 DD-Peptidase by Compounds 7−12. In view of the nature
of the inhibitors and all precedent,20,24−26 the inhibition was
interpreted in all cases as purely competitive. Equilibrium
constants for inhibition of the R39 DD-peptidase by compounds
7 (0−3.0 mM), 10 (0−2.0 mM), 11 (0−1.0 mM), 8 (0−
400 μM), 9 (0−2.0 mM), and 12 (0−40 μM) were obtained
from steady state competition experiments in which 14 was
employed as a chromogenic (245 nm; Δε = 2500 cm−1 M−1)
substrate (0.5 mM). The reaction conditions were 20 mM
MOPS buffer, pH 7.50, 25 °C, and an enzyme concentration of
210 nM. Under these conditions, the Km value of the substrate
was 38.4 μM.25 Measurements of the initial velocity versus
concentrations of 7−11 were fit to Scheme 5 and eq 1 to obtain
1
solid hydrochloride salt was obtained in 86% yield: H NMR
(300 MHz, d6-DMSO) δ 1.20 (d, J = 4.8 Hz, 3H), 1.27 (d, 3H,
J = 4.8 Hz), 3.40 (m, 1H), 3.53 (m, 1H), 4.14 (m, 1H), 4.38
(m, 1H), 8.06 (br s, 3H), 8.22 (br s, 3H).
D-1-[6-(N,N-Dibenzylamino)-6-(benzyloxycarbonyl)]-trans-
hex-2-enoyl-3-amino-1,1,1-trifluoro-2-butanol 22. To a mix-
ture of D-1-[6-(N,N-dibenzylamino)-6-(benzyloxycarbonyl)]-
trans-hex-2-enoic acid 19 (488 mg, 1.1 mmol, 1 equiv) and
3-amino-1,1,1-trifluoro-2-butanol hydrochloride 21, dissolved
in 10 mL of DMF, was added triethylamine (767 μL, 2.75
mmol, 2.5 equiv) followed by HBTU (1046 mg, 2.75 mmol,
2.5 equiv) at 0−5 °C. The reaction mixture was stirred at room
temperature for 24 h. The solvent was removed in vacuo and
the residue dissolved in ethyl acetate. The organic solution was
washed with 0.1 M hydrochloric acid, a saturated aqueous solu-
tion of sodium bicarbonate, and brine and dried over anhydrous
sodium sulfate. The solvent was removed in vacuo, and the
residue was purified by silica gel column chromatography (ethyl
acetate/hexane) to give product 22 in 80% yield as a colorless
Scheme 5
1
oil: H NMR (400 MHz, CDCl3) δ 1.32 (d, 3H, J = 7.5 Hz),
1.76 (m, 2H), 2.06−2.16 (m, 1H), 2.25−2.36 (m, 1H), 3.32
(dd, J = 6 Hz, 1H), 3.46, 3.85 (AB q, J = 16 Hz, 4H), 4.07 (m, 1H),
4.28 (m, 1H), 4.83 (d, J = 7.5 Hz, 1H), 4.91 (d, J = 7.5 Hz, 1H),
5.21 (AB q, J = 16, 24 Hz, 2H), 5.40 (d, J = 5 Hz, 1H), 7.67
(m, 1H), 7.2−7.4 (m, 15 H).
the Ki value. Total progress curves of inhibition of R39 DD-
peptidase by compound 12 were fit to Scheme 8 (see below)
with Dynafit39 to obtain the Ki value directly.
D-1-[6-(N,N-Dibenzylamino)-6-(benzyloxycarbonyl)]-trans-
hex-2-enoyl-3-amino-1,1,1-trifluorobutan-2-one 23. To a
stirred solution of 22 (380 mg, 0.67 mmol, 1 equiv) in
dichloromethane (25 mL) at room temperature was added
Dess−Martin periodinane (850 mg, 2 mmol, 3 equiv), followed
by trifluoroacetic acid (150 μL, 2 mmol, 3 equiv) at room
temperature. The reaction mixture was stirred for 3 h, after
which it was quenched with a sodium bicarbonate solution and
extracted with ethyl acetate. After the solution had been dried
and the solvent evaporated, the crude product was purified on a
v0 = V(Km + s)/[Km(1 + i/Ki) + s]
(1)
Inhibition of B. subtilis PBP4a by Compounds 7−12.
Equilibrium constants for inhibition of B. subtilis PBP4a by
compounds 7−12 were obtained in a manner similar to that
used for the constants for the R39 enzyme. The reaction con-
ditions were 20 mM MOPS buffer, pH 7.50, 25 °C, and an
enzyme concentration of 164 nM. Under these conditions, Km
was 0.44 0.08 mM for 14 as a substrate. The total progress curves
for inhibition by compounds 8 (0−400 μM), 11 (0−1.0 mM),
2131
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
and 12 (0−40 μM) were fit to Scheme 5 with Dynafit39 to
obtain Ki values.
The Ki values for 24−27 with both enzymes were obtained
from the measurement of initial rates at several concentrations
of inhibitor in each case and application of eq 1.
Table 1. Data Collection and Refinement Statistics
a
Data Collection
space group
P21
cell dimensions
a = 103.6 Å, b = 91.7 Å, c = 106.9 Å,
β = 94.30°
Kinetics of Aminolysis of 14 by 7−10, Catalyzed by the
R39 DD-Peptidase. Aminolysis of substrate 14 by amines 7−12,
catalyzed by the R39 DD-peptidase, involves amine attack on the
acyl−enzyme E−S intermediate (Scheme 6). The initial rates of
resolution range (Å)
48.9−2.6 (2.74−2.6)
61653
no. of unique reflections
b
Rmerge (%)
9.4 (63.2)
6.9 (6.9)
redundancy
completeness (%)
⟨I⟩/⟨σI⟩
100 (100)
13.5 (3.3)
Scheme 6
Refinement
resolution range (Å)
no. of non-hydrogen protein atoms
no. of water molecules
Rcryst (%)
47.9−2.6
13891
322
19.3
Rfree (%)
23.7
root-mean-square deviation from ideal
stereochemistry
bond lengths (Å)
bond angles (deg)
0.007
1.00
60.7
45.0
reaction of 14, catalyzed by this enzyme, as a function of
concentrations of 7−10 were fit to Scheme 6 with Dynafit.39
These experiments were conducted in 20 mM MOPS buffer at
pH 7.50 and 25 °C with an enzyme concentration of 210 nM.
Crystallization, Data Collection, and Structure Deter-
mination. Crystals of the enzyme were obtained by the hang-
ing drop vapor diffusion method by mixing 4 μL of a 25 mg/mL
protein solution [also containing 5 mM MgCl2 and 20 mM
Tris (pH 8.0)], 2 μL of a well solution [2.0 M ammonium
sulfate and 0.1 M MES (pH 6)], and 0.5 μL of 0.1 M CoCl2.
Crystals were soaked in 6 μL of a solution containing 3.0 M
ammonium sulfate, 0.1 M MES (pH 6.0), and 0.5 μL of
0.5 M 12.
mean B factor (all atoms) (Å2)
mean B factor (ligands) (Å2)
c
Ramachandran plot
favored regions (%)
allowed regions (%)
outlier regions (%)
PDB entry
96.9
3.1
0.0
3zcz
a
Statistics for the highest-resolution shell are given in parentheses.
b
Rmerge = ∑|Ii − Im |/∑Ii, where Ii is the intensity of the measured
reflection and Im is the mean intensity of all symmetry-related
reflections. Values in parentheses are for the outer resolution shell.
Using rampage.63
c
Data for the R39−12 crystals were measured on beamline
PROXIMA at SOLEIL (Paris, France) and processed using
XDS. Molecular replacement processed with the native struc-
ture produced an interpretable density map, which clearly shows
12 in the four monomers of the asymmetric unit. Refinement
was conducted with Refmac,40 Coot,41 and TLS42 with final
R and Rfree values of 19.3 and 23.7%, respectively. Data col-
lection and refinement statistics are summarized in Table 1.
Table 2. Inhibition of DD-Peptidases by Specific Inhibitors
R39
PBP4a
Ki (μM)
inhibitor
Ki (μM)
k4 (s−1 M−1
)
a
b
d
2
0.032 (0.5)
NO
0.007 0.001
c
0.057 0.001
c
7
6700 100 (3000) (6.1 0.4) × 103 >1000 (1000)
ce
e
,
8
1250 50 (500)
cf
(2.1 0.1) × 104 (110 25) (300)
g
,
9
NI (2000)
(2.9 0.1) × 102 ND
cf
,
g
10
NI (2000)
≤1.0 × 102
ND
RESULTS AND DISCUSSION
■
h
d
11
(76 13) (1000)
NO
(500 150) (1000)
Synthesis of the new compounds 7−12 was achieved in a
straightforward fashion (Schemes 3 and 4 and Schemes S1−S5
of the Supporting Information) via the versatile intermediate
19. They were all characterized by NMR and MS methods. All,
along with the carboxylate 13 and simple functional group
analogues 24−27, were treated as rapid equilibrium competitive
inhibitors of two LMMC DD-peptidases, the Actinomadura R39
DD-peptidase and B. subtilis PBP4a. No indications of slow
binding were observed in any case. Values of the inhibition
constant, Ki, obtained for 7−12, along with those for 2 and 13,
previously determined,30 are as reported in Table 2.
h
i
i
12
(0.37 0.04)
(2.2 0.1) × 103 (13.5 0.3) (300)
ci
,
(0.17 0.02)
j
j
g
13
480
3.0 × 103
ND
a
b
From ref 30. Highest inhibitor concentration (micromolar)
c
employed in parentheses. Fluorescence experiment (see the text).
d
No rate acceleration observed at the highest concentration employed.
e
Observed value divided by 4, because it is assumed that only one of
the four diastereomers present has significant activity (see the text).
f
g
No inhibition observed at the highest concentration employed. Not
h
determined. Ki values not corrected for hydration; see the text for
corrected values. Observed value divided by 2, because it is assumed
i
that only one of the two diastereomers present has significant activity
j
(see the text). From ref 25.
Alcohols 7 and 8 had very little inhibitory activity against
either enzyme. Even the trifluoromethyl substituent of 8, which
should reduce the alcohol pKa into a range where its anion
would be accessible to the active site oxyanion hole, produced
only a weak inhibitor of PBP4a. Configuration 3 of alcohols at
the active site is therefore unfavorable. It should be noted in
passing that alcohol 8 was obtained as a mixture of four
diastereoisomers in roughly equal amounts (see the 19F NMR
data in the Supporting Information), only one of which would
be expected to be an inhibitor (2R,3S). Serine hydrolases, in
2132
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
general, are not efficiently inhibited by alcohols, even when
specific (e.g., refs 43 and 44). Phosphonates 9 and 10 were also
ineffective against the DD-peptidases. Presumably, configuration
28 is not favorable either. A series of arylamidophosphonic
acids was also found to lack significant inhibitory activity
against the R39 DD-peptidase.12 Specific aryl phosphonate esters
are known to be time-dependent covalent inhibitors of
β-lactamases45 and (weakly) of a LMMB DD-peptidase.46 No
time-dependent inhibition of the present LMMC DD-peptidases
by 9, however, was observed.
we are unaware of any report in the literature of NMR ob-
servation of the carbonyl form of a α-trifluoro-α′-amidoketone
in aqueous solution (e.g., refs 51 and 52). Thus, Ki values for
12, corrected for hydration, can be estimated to be 0.37 and
13.5 nM for the R39 DD-peptidase and PBP4a, respectively.
These estimates represent upper limits and may be an order
of magnitude larger than the real values. These carbonyl
compounds are certainly powerful inhibitors of the LMMC DD-
peptidases and, upon reaction with the enzymes, may well
generate structures resembling tetrahedral high-energy inter-
mediates or transition states (see below).
The lack of specificity of the trifluoromethyl moiety itself and
the importance of the specific N-terminus are confirmed by the
data listed in Table 3, where the effectiveness of generic
Carboxylate 13 is also a weak inhibitor of the R39 enzyme
(Table 2). The trigonal carboxylate apparently does not sit well
in the oxyanion hole. This is, in fact, seen in the crystal struc-
ture of a complex of the carba analogue of 13 with the R39
DD-peptidase,26 where the carboxylate is oriented away from the
oxyanion hole. It is certainly reasonable that 13, the product of
a DD-carboxypeptidase or endopeptidase reaction, would be a
poor inhibitor.
Table 3. Inhibition of DD-Peptidases by Generic Inhibitors
Ki (mM)
inhibitor
R39
PBP4a
a
b
24
25
26
27
NI (3.0)
(0.23 0.03) (3.0)
a
c
NI (2.5)
ND
(0.27 0.06) (2.0)
5.0 1.6 (3.0)
a
c
NI (2.5)
ND
Both of the specific carbonyl compounds 11 and 12 are
inhibitors of both LMMC enzymes. Trifluoroketone 12 is
particularly effective (it is assumed that only one of the two
diasteromers of 12, present in approximately equal amounts,
has significant activity; this isomer, seen in the crystal structure
described below, has D-stereochemistry at the ketone terminus).
It is likely (also see below) that these carbonyl compounds
form tetrahedral adducts 4 and 5, respectively, at the active site.
The greater effectiveness of the trifluoroketone is likely to be
due in part to the higher alcohol acidity of the adduct,47
allowing thermodynamically more favorable formation of 6
than 5 at pH 7.5 (pKa values of trifluoroethanol and ethanol are
12.4 and 16,48 respectively, favoring anion formation by the
former by a factor of 103.6). This factor is essentially sufficient
to rationalize the higher inhibitory potency of 12 compared to
that of 11.
a
b
No inhibition observed. Highest inhibitor concentration (millimolar)
c
employed in parentheses. Not determined.
inhibitors 24−27, bearing the “warheads” of 2, 7, 11, and 12,
respectively, is displayed. Only 26 shows weak inhibition of the
R39 DD-peptidase as do 24 and 26 of PBP4a.
It was striking that the weaker “inhibitors” of the R39 DD-
peptidase, 7−9, actually produced an increase in the initial rate
of substrate turnover, as shown for 7 in Figure 1, for example.
If it is assumed that the reacting species of 11 and 12 are the
free carbonyls,47 then the observed or apparent Ki values of
Table 2 should be corrected for the competing hydration in
aqueous solution (Scheme 7) via eq 2
KH = [IOH2]/[I] Ki = Kiobs/(1 + KH)
(2)
Scheme 7
Figure 1. Effect of alcohol 7 on initial rates of turnover of thiolester
substrate 14 (0.5 mM) catalyzed by the R39 DD-peptidase. The points
are experimental, and the line represents the fit of the data to Scheme
6 (see the text).
1
where Kiobs is the observed value. From H NMR spectra (not
shown), KH for 11 is 30, and therefore, from eq 2, the “real” Ki
values for the carbonyl form of 11 are 2.5 and 16.1 μM for
the R39 DD-peptidase and PBP4a, respectively. The effect of
hydration is even more dramatic with 12. Although estimates of
KH for trifluoroacetone in the literature are curiously wide-
ranging,47,49,50 the lower limit estimate is 35.49 If the additional
effect of the amido substituent is included as a factor of 30
(estimated from the KH values of acetaldehyde49 and 11), then
a lower limit estimate of KH for 12 would be 1000. Certainly,
The likely explanation for this, with established precedent,25
is that, because of their free D-amino acid termini, these com-
pounds may also act as acyl acceptors and, in cases where
enzyme deacylation is rate-determining, as is presumably true
for thiolester 14, accelerate the observed reaction. Data like
2133
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
those shown in Figure 2 were fit to Scheme 6 to obtain values
for k4 (Table 2). Obviously, the k4 values for 7 and 8 are much
higher than for the anionic 9 and 10 and are comparable to that
(1.4 × 104 s−1 M−1) of the best acceptor yet discovered for this
enzyme, the good substrate 29.25 It is, however, striking that
anionic terminal carboxylate 13 is a much better acceptor than
the equally anionic phosphonates 9 and 10. The ability of
enzymes to bind carboxylates and phosph(on)ates differently
has been noticed previously,53 although the functional
significance of the distinction here is not evident.
reported in Table 2 derives from the fits of these data. This
phenomenon was not observed in the experiments with 12 and
PBP4a or with 11 and both enzymes probably because of the
higher concentrations of inhibitors required for the experi-
ments, leading to smaller fractional depletions of the inhibitors
in these cases.
Because of the complication created by the ability of 7 and 8
to act as strong acyl group acceptors, their activity as inhibitors,
although weaker than that of 2 and 12, could not be readily
determined from experiments in which turnover of the sub-
strate was monitored. Therefore, the binding strengths of 7 and
8 were determined from the inhibition of their irreversible
inactivation of the R39 enzyme with cefotaxime, monitored
fluorimetrically.54 These measurements showed that 7 and 8
did in fact bind very weakly [at millimolar concentrations
(Table 2)]. The fluorescence inhibition method was checked
with 2 and 12, with acceptable results (Table 2).
The acyl acceptor ability of these inhibitors probably also led
to the initially unexpected observation of an apparent decrease
in the effectiveness of 12 as an inhibitor of the R39 DD-
peptidase as a function of time in experiments with substrate 14
(Figure 2). This phenomenon can be interpreted in terms of
It is also striking that 11 and 12 are more effective inhibitors
of the R39 enzyme than of PBP4a. This may relate to the same
order of reactivity of specific substrates with these enzymes.20
On the other hand, the analogous boronic acid 2 is a stronger
inhibitor of PBP4a than of the R39 enzyme. The latter dif-
ference, however, may relate to the interesting observation that
the binding of 2 to PBP4a is a slow (minutes; two-step?)
process,20 whereas its binding to the R39 enzyme is fast (with
respect to manual mixing times). No time dependence of
the inhibition of either of these enzymes by 11 and 12 was
observed.
Crystal Structure of the Complex of the R39 DD-
Peptidase with 12. An X-ray diffraction study of the title
complex was performed. Initial electron density maps resulting
from the molecular replacement were excellent for the four
protein monomers present in the asymmetric unit, allowing
a precise localization of 12 in the active site of each monomer
(Figure 3). The inhibitor 12 is covalently bonded to the active
serine Ser49; the carbonyl carbon is now tetrahedral, and the
ketone oxygen occupies the oxyanion hole, defined by the
backbone NH groups of Ser49 and Thr413. The D-methyl
group is directed into a hydrophobic pocket comprised of
residues Gly148, Leu349, and Met414. As previously observed
Figure 2. Progress curves showing turnover of substrate 14 (0.5 mM)
by the R39 DD-peptidase (0.21 μM) in the absence (−−−) and
presence () of 12 (20 μM). The empty circles are experimental, and
the lines represent fits of the data to Scheme 8 (see the text).
Scheme 8, which was fit to data like those shown in Figure 1
(fit shown). The product, Q, of the reaction of 12 with substrate
in the structures of the R39 enzyme with boronic acid 2,30
a
cephalosporin or a peptide,26 each bearing an aminopimelic
side chain, the aminopimelyl side chain runs along a hydro-
phobic groove and is fixed by a series of hydrogen bonds
between the amido group and Asn300 Nδ2 and Thr413 O
atoms, and between the terminal D-α-amino acid moiety and
the Arg351 guanidinium, the Asp142 carboxylate, and the
Ser415 hydroxyl.
Scheme 8
The most novel elements of the structure are the position of
the trifluoromethyl group and its interaction with the enzyme
active site. Branching from the tetrahedral carbon bonded to
Ser49 Oγ, the CF3 group occupies the likely position of a
leaving group in the tetrahedral intermediate of a peptidase
substrate (and thus of one of the boronate hydroxyl groups
in the analogous complex with 230). An overlap of the active
site elements of the boronate 2 complex and that of the
trifluoroketone 12, both bearing a free D-α-aminopimelyl
14, which presumably is 30, must be a weak inhibitor of the
R39 enzyme. Acylation of 12 by 14 would lead to a gradual
depletion of the inhibitor in solution and thus to a decreasing
level of inhibition with time (Figure 1). The Ki value for 12
2134
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
Figure 3. Crystal structure of the R39 DD-peptidase in complex with the specific trifluoroketone 12. In this stereoview, the electron density is an
|Fo| − |Fc| difference map calculated from the final coordinates of the model refined in the absence of ligand. The resulting positive density is shown
with green hatching and is contoured at 2.5σ. The protein backbone and side chains are colored orange, and the trifluoroketone is colored
gray. Heteroatoms are colored red (oxygen), blue (nitrogen), orange (sulfur), and yellow (fluorine). This figure was generated using PYMOL
(http://www.pymol.sourceforge.net).
Figure 4. Superimposition of the active site residues of the complex of 12 with the R39 DD-peptidase on those of the analogous complex with
boronic acid 2.30 In this stereoview, the colors of the complex of 12 are the same as those in Figure 3. The boronate complex is colored blue except
for the heteroatoms, which are colored as in Figure 3, and boron, which is colored rose. This figure was generated using PYMOL (http://
www.pymol.sourceforge.net).
N-terminus, for which this enzyme is specific,20,24 is shown in
Figure 4. The inhibitors are essentially identically placed in the
active site, as would be expected for transition state analogues
with such similar structure, with the CF3 group of 12 over-
lapping directly with the “leaving group” boronate hydroxyl.
With respect to the position of the CF3 group, one O−C−C−F
dihedral angle is 57°, indicating an essentially precise staggering
of the fluorine atoms with respect to the oxyanion O−. This
conformation would be expected, of course, on steric and elec-
trostatic grounds and has been observed in serine protease−
trifluoroketone complexes (for example, in ref 55). One
fluorine atom is directed out of the active site into the solvent
where it might well interact with water molecules in the manner
observed in a variety of crystal structures.55−58 Such interac-
tions are probably more dipolar in nature than those involving
classical hydrogen bonding and are probably of ∼1 kcal/mol
strength.59,60 The other two fluorine atoms appear to interact in
a bifurcated fashion at a distance of 3.0 Å each with the
hydroxyl of Ser298. This kind of arrangement has been
previously observed, for example, in a trifluoroketone adduct
with chymotrypsin,55 where two fluorines interact with a water
molecule, and in a thrombin complex, where a similar interac-
tion with a water molecule has been observed.57 It is interesting
that the Lys52 terminal nitrogen has not moved within 4.0 Å
of a fluorine atom. There are many examples of close NH···F
interactions of <3.0 Å in complexes of enzymes with fluorinated
2135
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
inhibitors.55−58,61 In this case, the Lys52 amine would certainly
be protonated, in a negatively charged transition state analogue
structure as it is and where it is strongly interacting with Ser49
Oγ (2.9 Å) and Asn300 Oδ (2.8 Å) atoms.
A combination of the inhibition kinetics and the crystal
structure, both discussed above, suggests that the trifluor-
omethyl moiety alone does not interact specifically and tightly
with the DD-peptidase active site but acts only as part of an
electrophilic ketone that does react with the active site serine in
a cooperative fashion with the specifically binding D-α-aminopimelyl
N-terminus.26,27,30,62
3-morpholinopropanesulfonic acid; NMR, nuclear magnetic
resonance; PBP, penicillin-binding protein; THF, tetrahydofuran.
REFERENCES
■
(1) Wise, E. M., and Park, J. T. (1965) Penicillin: Its basic site of
action as an inhibitor of a peptide cross-linking reaction in cell wall
mucopeptide synthesis. Proc. Natl. Acad. Sci. U.S.A. 54, 75−81.
(2) Tipper, D. J., and Strominger, J. L. (1965) Mechanism of action
of penicillins: A proposal based on their structural similarity to acyl-D-
alanyl-D-alanine. Proc. Natl. Acad. Sci. U.S.A. 54, 1133−1141.
(3) Lee, B. (1971) Conformation of penicillin as a transition-state
analog of the substrate of peptidoglycan transpeptidase. J. Mol. Biol. 61,
463−469.
CONCLUSIONS
■
(4) Pratt, R. F. (2002) Functional evolution of the serine β-lactamase
active site. J. Chem. Soc., Perkin Trans. 2, 851−861.
Compound 12 represents the first nanomolar trifluoroketone
inhibitor of a DD-peptidase. It owes its potency to the trifluoroketone
moiety, elevating it over the less electrophilic aldehyde 11 and
other tetrahedral species derived from 7−10, which occupy the
active site less favorably. As shown by the very limited effec-
tiveness of the generic analogues 24−27, it is clear that 12 is
able to achieve its effectiveness only in combination with the
strong specificity produced by the D-α-aminopimelyl terminus,
whose affinity for certain LMMC DD-peptidases has been
established. This combination of the trifluoroketone, which
forms a tetrahedral adduct with the active site serine, and the D-
α-aminopimelyl terminus into a complete inhibitor of the R39
DD-peptidase is seen very clearly in the crystal structure of the
complex between the enzyme and 12. The trifluoroketone
moiety and other activated carbonyls, therefore, should be con-
sidered along with boronic acids and phosphonates as warheads
that can be incorporated into new and effective DD-peptidase
inhibitors and therefore, perhaps, antibiotics.
̀
(5) Frere, J.-M., Ed. (2012) β-Lactamases, Nova, New York.
(6) Drawz, S. M., and Bonomo, R. A. (2010) Three decades of β-
lactamase inhibitors. Clin. Microbiol. Rev. 23, 160−201.
(7) Bebrone, C., Lassaux, P., Vercheval, L., Sohier, J. S., Jehaes, A.,
Sauvage, E., and Galleni, M. (2010) Current challenges in
antimicrobial chemotherapy: Focus on β-lactamase inhibition. Drugs
70, 651−679.
(8) Bush, K., and Macielag, M. J. (2010) New β-lactam antibiotics
and β-lactamase inhibitors. Expert Opin. Ther. Pat. 20, 1277−1293.
(9) Miguet, L., Zervosen, A., Gerards, T., Pasha, F. A., Luxen, A.,
̀
Disteche-Nguyen, M., and Thomas, A. (2009) Discovery of new
inhibitors of Streptococcus pneumoniae penicillin-binding protein (PBP)
2x by structure-based virtual screening. J. Med. Chem. 52, 5926−5936.
(10) Turk, S., Verlaine, O., Gerards, T., Zivec, M., Humljan, J., Sosic,
I., Amoroso, A., Zervosen, A., Luxen, A., Joris, B., and Gobec, S.
(2011) New non-covalent inhibitors of penicillin-binding proteins
from penicillin-resistant bacteria. PLoS One 6, e19418.
(11) Contreras-Martell, C., Amoroso, A., Woon, E. C. Y., Zervosen,
A., Inglis, S., Martins, A., Verlaine, O., Rydzik, A. M., Job, V., Luxen, A.,
Joris, B., Schofield, C. J., and Dessen, A. (2011) Structure-guided
design of cell wall biosynthesis inhibitors that overcome β-lactam
resistance in Staphylococcus aureus (MRSA). ACS Chem. Biol. 6, 941−
951.
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic details for the preparation of compounds 7−11. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(12) Woon, E. C. Y., Zervosen, A., Sauvage, E., Simmonds, K. J.,
Zivec, M., Inglis, S. R., Fishwick, C. W. G., Gobec, S., Charlier, P.,
Luxen, A., and Schofield, C. J. (2011) Structure guided development of
potent reversibly binding penicillin binding protein inhibitors. ACS
Med. Chem. Lett. 2, 219−223.
(13) Dzhekieva, L., Kumar, I., and Pratt, R. F. (2012) Inhibition of
bacterial DD-peptidases (penicillin-binding proteins) in membranes
and in vivo by peptidoglycan-mimetic boronic acids. Biochemistry 51,
2804−2811.
(14) Shilabin, A. G., Dzhekieva, L., Misra, P., Jayaram, B., and Pratt,
R. F. (2012) 4-Quinolones as non-covalent inhibitors of high
molecular mass penicillin-binding proteins. ACS Med. Chem. Lett. 3,
592−595.
(15) Fedarovich, A., Djordjevic, K. A., Swanson, S. M., Peterson, Y.
K., Nicholas, R. A., and Davies, C. (2012) High-throughput screening
for novel inhibitors of Neisseria gonorrhoeae penicillin-binding protein
2. PLoS One 7, e44918.
(16) Pauling, L. (1946) Molecular architecture and biological
reactions. Chem. Eng. News 24, 1375−1377.
(17) Wolfenden, R. (1969) Transition state analogues for enzyme
catalysis. Nature 223, 704−705.
(18) Schramm, V. L. (1998) Enzymatic transition states and
transition state analog design. Annu. Rev. Biochem. 67, 693−720.
(19) Pratt, R. F. (2008) Substrate specificity of bacterial DD-
peptidases (penicillin-binding proteins). Cell. Mol. Life Sci. 65, 2138−
2155.
(20) Nemmara, V. V., Dzhekieva, L., Sarkar, K. S., Adediran, S. A.,
Duez, C., Nicholas, R. A., and Pratt, R. F. (2011) Substrate specificity
of low-molecular mass bacterial DD-peptidases. Biochemistry 50,
10091−10101.
AUTHOR INFORMATION
■
Corresponding Author
*Telephone: (860) 685-2629. E-mail: rpratt@wesleyan.edu.
Fax: (860) 685-2211.
Funding
This research was supported by National Institutes of Health
Grant AI-17986 (R.F.P.) and in part by the Belgian Program on
Interuniversity Poles of Attraction initiated by the Belgian State,
Prime Minister’s Office, Science Policy programming (IAP P6/
19), the Fonds de la Recherche Scientifique (IISN 4.4505.09,
IISN 4.4509.11, and FRFC 2.4511.06F), and the University of
Liege (Fonds speciaux, Credit classique, C-06/19 and C-09/
̀ ́ ́
75). C.D. and F.K. are research associates of the FRS-FNRS,
Belgium.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS
■
DIBAL, diisobutylaluminum hydride; ESMS, electrospray ioniza-
tion mass spectrometry; HBTU, O-(benzotriazol-1-yl)-N,N,N′,N′-
tetramethyluronium hexafluorophosphate; HMM, high-molecular
mass; HRMS, high-resolution mass spectrometry; LMM, low-
molecular mass; MES, 4-morpholinoethanesulfonic acid; MOPS,
2136
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
(21) Ghuysen, J.-M. (1991) Serine β-lactamases and penicillin-
binding proteins. Annu. Rev. Microbiol. 45, 37−67.
(22) Anderson, J. W., and Pratt, R. F. (2000) Dipeptide binding to
the extended active site of the Streptomyces R61 D-alanyl-D-alanine-
peptidase: The path to a specific substrate. Biochemistry 39, 12200−
12209.
(40) Murshudov, G. N., Vagin, A. A., and Dodson, E. J. (1997)
Refinement of macromolecular structures by the maximum-likelihood
method. Acta Crystallogr. D53, 240−255.
(41) Emsley, P., and Cowtan, K. (2004) Coot: Model-building tools
for molecular graphics. Acta Crystallogr. D60, 2126−2132.
(42) Painter, J., and Merritt, E. A. (2006) Optimal description of a
protein structure in terms of multiple groups undergoing TLS motion.
Acta Crystallogr. D62, 439−450.
(23) McDonough, M. A., Anderson, J. W., Silvaggi, N. R., Pratt, R. F.,
Knox, J. R., and Kelly, J. A. (2002) Structures of two kinetic
intermediates reveal species specificity of penicillin-binding proteins. J.
Mol. Biol. 322, 111−122.
́ ́
(43) Begue, J.-P., Bonnet-Delpon, D., Fischer-Durand, N., Amour, A.,
and Reboud-Ravaux, M. (1994) Stereoselective synthesis and inhibitor
properties towards human leucocyte elastase of chiral β-peptidyl
trifluoromethyl alcohols. Tetrahedron: Asymmetry 5, 1099−1110.
(44) Potetinova, J. V., Voyushina, T. L., and Stepanov, V. M. (1997)
Enzymatic synthesis of peptidyl-amino aldehydes-serine proteinase
inhibitors. Bioorg. Med. Chem. Lett. 7, 705−710.
(24) Anderson, J. W., Adediran, S. A., Charlier, P., Nguyen-Distec
̀
he,
M., Frere, J.-M., Nicholas, R. A., and Pratt, R. F. (2003) On the
̀
substrate specificity of bacterial DD-peptidases: Evidence from two
series of peptidoglycan-mimetic peptides. Biochem. J. 373, 949−955.
(25) Adediran, S. A., Kumar, I., Nagarajan, R., Sauvage, E., and Pratt,
R. F. (2011) Kinetics of reactions of the Actinomadura R39 DD-
peptidase with specific substrates. Biochemistry 50, 376−387.
(45) Rahil, J., and Pratt, R. F. (1992) Mechanism of inhibition of the
class C β-lactamase of Enterobacter cloacae P99 by phosphonate
monoesters. Biochemistry 31, 5869−5878.
(46) Silvaggi, N. R., Anderson, J. W., Brinsmade, S. A., Pratt, R. F.,
and Kelly, J. A. (2003) The crystal structure of phosphonate-inhibited
D-Ala-D-Ala peptidase reveals an analogue of a tetrahedral transition
state. Biochemistry 42, 1199−1208.
(47) Allen, K. N., and Abeles, R. H. (1989) Inhibition kinetics of
acetylcholinesterase with fluoromethyl ketones. Biochemistry 28,
8466−8473.
(48) Jencks, W. P., and Regenstein, J. (1975) in Handbook of
Biochemistry and Molecular Biology (Fassman, G. D., Ed.) 3rd ed., Vol.
1, pp 305−351, CRC, Cleveland, OH.
(49) Guthrie, J. P. (1975) Carbonyl addition reactions: Factors
affecting the hydrate-hemiacetal and hemiacetal-acetal equilibrium
constants. Can. J. Chem. 53, 898−906.
(50) Smith, R. A., Copp, L. J., Donnelly, S. L., Spencer, R. W., and
Krantz, A. (1988) Inhibition of cathepsin B by peptidyl aldehydes and
ketones: Slow-binding behavior of a trifluoromethyl ketone.
Biochemistry 27, 6568−6573.
(51) Imperiali, B., and Abeles, R. H. (1986) Inhibition of serine
proteases by peptidyl fluoromethyl ketones. Biochemistry 25, 3760−
3767.
(26) Sauvage, E., Powell, A. J., Heilemann, J., Josephine, H. R.,
Charlier, P., Davies, C., and Pratt, R. F. (2008) Crystal structures of
complexes of bacterial DD-peptidases with peptidoglycan-mimetic
ligands; the substrate specificity puzzle. J. Mol. Biol. 381, 383−393.
(27) Sauvage, E., Duez, C., Herman, R., Kerff, F., Petrella, S.,
̀
Anderson, J. W., Adediran, S. A., Pratt, R. F., Frere, J.-M., and Charlier,
P. (2007) Crystal structure of the Bacillus subtilis penicillin-binding
protein 4a and its complex with peptidoglycan-mimetic peptide. J. Mol.
Biol. 371, 528−539.
(28) Josephine, H. R., Charlier, P., Davies, C., Nicholas, R. A., and
Pratt, R. F. (2006) Reactivity of penicillin-binding proteins with
peptidoglycan-mimetic β-lactams: What’s wrong with these enzymes?
Biochemistry 45, 15873−15883.
(29) Kumar, I., Josephine, H. R., and Pratt, R. F. (2007) Reactions of
peptidoglycan-mimetic β-lactams with penicillin-binding proteins in
vivo and in membranes. ACS Chem. Biol. 2, 620−624.
(30) Dzhekieva, L., Rocaboy, M., Kerff, F., Charlier, P., Sauvage, E.,
and Pratt, R. F. (2010) Crystal structure of a complex between the
Actinomadura R39 DD-peptidase and a peptidoglycan-mimetic
boronate inhibitor: Interpretation of a transition state analogue in
terms of catalytic mechanism. Biochemistry 49, 6411−6419.
(31) Zervosen, A., Bouillez, A., Herman, R., Amoroso, A., Joris, B.,
Sauvage, E., Charlier, P., and Luxen, A. (2012) Synthesis and
evaluation of boronic acids as inhibitors of penicillin binding proteins
of classes A, B, and C. Bioorg. Med. Chem. 20, 3915−3924.
(32) Kraut, J. (1977) Serine proteases: Structure and mechanism of
catalysis. Annu. Rev. Biochem. 46, 331−358.
(52) Angelastro, M. R., Baugh, L. E., Bey, P., Burkhart, J. P., Chen, T.-
M., Durham, S. L., Hare, C. M., Huber, E. W., Janusz, M. J., Koehl, J.
R., Marquart, A. L., Mehdi, S., and Peet, N. P. (1994) Inhibition of
human neutrophil elastase with peptidyl electrophilic ketones 2. Orally
active PG-Val-Pro-Val pentafluoroethyl ketones. J. Med. Chem. 37,
4538−4554.
(33) Pechenov, A., Stefanova, M. E., Nicholas, R. A., Peddi, S., and
Gutheil, W. G. (2003) Potential transition state analogue inhibitors for
the penicillin-binding proteins. Biochemistry 42, 579−588.
(34) Granier, B., Duez, C., Lepage, S., Englebert, S., Dusart, J.,
(53) Stamper, C. G. F., Morello, A. A., and Ringe, D. (1998)
Reaction of alanine racemase with 1-aminoethylphosphonic acid forms
a stable external aldimine. Biochemistry 37, 10438−10445.
̀
Dideberg, O., Van Beeumen, J., Frere, J. M., and Ghuysen, J. M.
̀
(54) Fuad, N., Frere, J.-M., Ghuysen, J.-M., Duez, C., and Iwatsubo,
(1992) Primary and predicted secondary structures of the Actino-
madura R39 extracellular DD-peptidase, a penicillin-binding protein
(PBP) related to the Escherichia coli PBP4. Biochem. J. 282, 781−788.
(35) Duez, C., Van Hove, M., Gallet, X., Bouillene, F., Docquier, J.-
M. (1976) Mode of interaction between β-lactam antibiotics and the
exocellular DD-carboxypeptidase-transpeptidase from Streptomyces R39.
Biochem. J. 155, 623−629.
(55) Brady, K., Wei, A., Ringe, D., and Abeles, R. H. (1990) Structure
of chymotrypsin-trifluoromethyl ketone inhibitor complexes: Compar-
ison of slowly and rapidly equilibrating inhibitors. Biochemistry 29,
7600−7607.
(56) Bernstein, P. R., Gomes, B. C., Kosmider, B. J., Vacek, E. P., and
Williams, J. C. (1995) Nonpeptidic inhibitors of human leukocyte
elastase. 6. Design of a potent, intratracheally active, pyridone-based
trifluoromethyl ketone. J. Med. Chem. 38, 212−215.
(57) Salvagnini, C., Michaux, C., Remiche, J., Wouters, J., Charlier, P.,
and Marchand-Brynaert, J. (2005) Design, synthesis and evaluation of
graftable thrombin inhibitors for the preparation of blood-compatible
polymer materials. Org. Biomol. Chem. 3, 4209−4220.
(58) Nielsen, T. K., Hildmann, C., Riester, D., Wegener, D.,
Schwienhorst, A., and Ficner, R. (2007) Complex structure of a
bacterial class 2 histone deacetylase homologue with a trifluorome-
thylketone inhibitor. Acta Crystallogr. F63, 270−273.
̀
D., Brans, A., and Frere, J.-M. (2001) Purification and characterization
of PBP4a, a new low molecular-weight penicillin-binding protein from
Bacillus subtilis. J. Bacteriol. 183, 1595−1599.
(36) Rodriquez, M., and Taddei, M. (2005) A simple procedure for
the transformation of L-glutamic acid into the corresponding γ-
aldehyde. Synthesis, 493−495.
(37) McBee, E. T., Hathaway, C. E., and Roberts, C. W. (1956) The
ethanolysis of 1,1,1-trifluoro-2,3-epoxybutane and 2-methyl-1,1,1-
trifluoro-2,3-epoxypropane. J. Am. Chem. Soc. 78, 4053−4057.
(38) Shao, Y. M., Yang, W. B., Kuo, T. H., Tsai, K. C., Lin, C. H.,
Yang, A. S., Liang, P. H., and Wong, C. H. (2008) Design, synthesis,
and evaluation of trifluoromethyl ketones as inhibitors of SARS-CoV
3CL protease. Bioorg. Med. Chem. 16, 4652−4660.
(39) Kuzmic, P. (1996) Program DYNAFIT for the analysis of
enzyme kinetic data: Application to HIV proteinase. Anal. Biochem.
237, 260−273.
2137
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138

Biochemistry
Article
(59) Caminati, W., Melandri, S., Maris, A., and Ottaviani, P. (2006)
Relative strengths of the OH···Cl and OH···F hydrogen bonds. Angew.
Chem., Int. Ed. 45, 2438−2442.
(60) Muller, K., Faeh, C., and Diederich, F. (2007) Fluorine in
pharmaceuticals: Looking beyond intuition. Science 317, 1881−1886.
(61) Dalvit, C., and Vulpetti, A. (2012) Intermolecular and
intramolecular hydrogen bonds involving fluorine atoms: Implications
for recognition, selectivity, and chemical properties. ChemMedChem 7,
262−272.
(62) Majumdar, S., and Pratt, R. F. (2009) Intramolecular
cooperativity in the reaction of diacyl phosphates with serine β-
lactamases. Biochemistry 48, 8293−8298.
(63) Lovell, S. C., Davis, I. W., Arendall, W. B., III, de Bakker, P. I.,
Word, J. M., Prisant, M. G., Richardson, J. S., and Richardson, D. C.
(2003) Structure validation by Cα geometry: ϕ, ψ and Cβ deviation.
Proteins 50, 437−450.
2138
dx.doi.org/10.1021/bi400048s | Biochemistry 2013, 52, 2128−2138