Specificity of extended O-aryloxycarbonyl hydroxamates as inhibitors of a class C β-lactamase

Article abstract of DOI:10.1016/j.bmc.2019.02.023

Class C β-lactamases have previously been shown to be efficiently inactivated by O-aryloxycarbonyl hydroxamates. O-Phenoxycarbonyl-N-benzyloxycarbonylhydroxylamine (1) and O-phenoxycarbonyl-N-(R)-[(4-amino-4-carboxy-1-butyl)oxycarbonyl]hydroxylamine (2),

Full text of DOI:10.1016/j.bmc.2019.02.023

Accepted Manuscript
Specificity of Extended O-Aryloxycarbonyl Hydroxamates as Inhibitors of a
Class C β-Lactamase
Alexandra A. Malico, K. Dave, S.A. Adediran, R.F. Pratt
PII:
S0968-0896(19)30035-5
https://doi.org/10.1016/j.bmc.2019.02.023
BMC 14759
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
6 January 2019
6 February 2019
Please cite this article as: Malico, A.A., Dave, K., Adediran, S.A., Pratt, R.F., Specificity of Extended O-
Aryloxycarbonyl Hydroxamates as Inhibitors of a Class C β-Lactamase, Bioorganic & Medicinal Chemistry (2019),
doi: https://doi.org/10.1016/j.bmc.2019.02.023
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Specificity of Extended O-Aryloxycarbonyl Hydroxamates as Inhibitors of a Class C -
Lactamase
Alexandra A. Malico, K. Dave, S.A. Adediran and R.F. Pratt*
Department of Chemistry, Wesleyan University, Middletown, CT 06459, USA
Keywords
-Lactamase inhibitor
Irreversible inhibitor
Covalent inhibitor
O-Aryloxycarbonyl hydroxamate
Inhibition kinetics
*Corresponding author
Email address: rpratt@wesleyan.edu

Abstract
Class C -lactamases have previously been shown to be efficiently inactivated by O-
aryloxycarbonyl hydroxamates. O-Phenoxycarbonyl-N-benzyloxycarbonylhydroxylamine (1)
and O-phenoxycarbonyl-N-(R)-[(4-amino-4-carboxy-1-butyl)oxycarbonyl]hydroxylamine (2),
for example, were found to be effective inactivators. The present paper describes a structure-
activity study of these molecules to better define the important structural elements for high
inhibitory activity. The results show that a well-positioned hydrophobic element (which may
interact with the Tyr221 residue of the enzyme) and a negatively charged element, e.g. a
carboxylate group (which may interact with Arg204), are required for high reactivity with the
enzyme. The new compounds were found to inactivate by forming a carbonyl cross-linked
enzyme (probably Ser64OCONHLys 315) as for 1 rather than the inert hydroxamoyl derivative
observed with 2.
2

1. Introduction
Bacteria resist the antibiotic action of -lactams in several ways, most particularly by the
production of -lactamases.1 These enzymes catalyze the hydrolysis of -lactams and readily
evolve to do this most efficiently. To overcome the threat posed by -lactamases to both classical
and new -lactams, there has been much interest for some years now in -lactamase inhibitors.2-4
Such molecules can, in principle, inactivate the -lactamases and thus allow -lactams to inhibit
bacterial cell wall synthesis, incapacitating the bacteria. Mechanistically, there are two groups of
-lactamases, the serine -lactamases (classes A, C, and D) and the metallo--lactamases (class
B). In the former group, an active site serine residue provides a nucleophilic hydroxyl group that
catalyzes -lactam hydrolysis by a double displacement mechanism via a covalent acyl-enzyme
intermediate (Scheme 1).5 Most effective serine -lactamase inhibitors, including those that are
currently used in medicine, act by covalent modification of the enzyme, generally of the active
site serine hydroxyl group. The recently developed diazabicyclooctanes6 and cyclic boronic
acids7 are of this type. Although non-covalent inhibitors of serine -lactamases have been
discovered, they have not yet achieved a sufficiently broad spectrum of high activity to allow
clinical development.8-11
Scheme 1. The -lactamase reaction.
3

Some time ago, we reported a new class of covalent inhibitors of serine -lactamases, the
O-aryloxycarbonyl hydroxamates.12,13 The lead compound 1 (Figure 1) had major activity
against class C -lactamases. The activity of various close analogues of 1 has also been
described,13,14 as well as that of cyclic variants.15 These compounds have also been shown to be
very effective inhibitors of N-terminal hydrolases such as the proteasome16 and penicillin
acylase.17 Most recently, the synthesis and activity of the peptide analogue 2 against -
lactamases has been reported.18 This compound was designed to be an inhibitor of bacterial DD-
peptidases and thus an antibiotic. It had little DD-peptidase inhibitory activity, however, but was
surprisingly active, more so than 1, against the Enterobacter cloacae P99 class C -lactamase.
The latter activity was interpreted in terms of the modeled structure shown in Figure 2, where
interactions between the polar terminus of the inhibitor and the protein were proposed to provide
the specificity of 2 as an inhibitor.18 Hydrophobic interaction between the trimethylene chain of
the inhibitor and the aromatic ring of Tyr221 may also be important. To determine the relative
contributions of the interactions described above, a new series of compounds 3 – 7 (Figure 1)
was designed. The synthesis of these compounds and their activity as inhibitors of the class C
P99 -lactamase are described in this paper.
4

Fig. 1. Aryloxycarbonyl hydroxamate inactivators of Class C lactamases
5

Fig. 2. An energy-minimized model of the tetrahedral intermediate formed on reaction of
compound 2 with the P99 -lactamase. Only heavy atoms are shown. The preparation of this
model is described in reference 18.
6

2. Results and Discussion
2.1 Synthesis
The O-aryloxycarbonyl hydroxamate inhibitors 3 - 7 were prepared essentially as
described earlier,12-14 by acylation of the appropriate hydroxamic acid with phenyl
chloroformate in the presence of the base imidazole (Schemes 4-8). Synthesis of the relevant
hydroxamic acid for the functionalized inhibitors 5 – 7 (Schemes 6-8), was achieved by
activation of an alcohol or phenol by reaction with carbonyl diimidazole in the presence of
imidazole. The resulting acyl-imidazole was then treated with a hydroxylamine
hydrochloride in the presence of imidazole. Protecting groups were removed in the final step
by treatment with trifluoroacetic acid. The final products were characterized by their IR and
1H NMR spectra and by high resolution mass spectrometry; the acyl hydroxamate moiety has
a distinctive carbonyl stretching absorption around 1800 cm-1 in IR spectra and a highly
downfield NH resonance around 12 ppm in 1H NMR spectra.12-14
2.2 Kinetics of the inhibition reaction
Aryloxycarbonyl hydroxamates have been shown to be irreversible inactivators of class C -
lactamases.12-14 The current compounds 3 - 7 also irreversibly inactivated a representative
class C -lactamase, that from Enterobacter cloacae P99. In order to determine the kinetics
of reaction of these compounds with the -lactamase, solutions containing a fixed
concentration of the enzyme and varied concentrations of the inhibitor were incubated and
the activity of the enzyme monitored as a function of time in each case (see Experimental
section for details). No return of activity of the enzyme was observed after the activity
reached a minimum value, i.e. the inactivation reaction is irreversible, as previously
observed.12-14 Examples of two such experiments with compound 4 are shown in Figure 3A.
7

A titration of enzyme activity with various concentrations of 4 was also performed (see
experimental section for details) with results shown in Figure 3B. The data from these
experiments were fitted simultaneously to Scheme 2 as described in the experimental section
and yielded the rate parameters presented in Table 1. Similar experiments were carried out
with the other compounds giving rate parameters that are also presented in Table 1.
Comparable results, previously published,12,18 for compounds 1 and 2 are also presented in
Table 1.
Scheme 2. Kinetic mechanism for reaction of the inhibitors 3 – 7 with the P99 -lactamase.
Fig. 3. (A) Activity of the P99 -lactamase (0.15 M) as a function of time in the presence of
4 [5.0 M () and 10.0 M ()]. (B) Activity of the P99 -lactamase (0.15 M) after complete
reaction with 4 at various concentrations. In both A and B the points are experimental and the
lines represent a least squares fit to all of the data.
8

9

Table 1. Rate constants for reaction of inhibitors 1 – 7 with the P99 -lactamase.
Compound
k1 (M-1s-1)
k2/k3
ki (M-1s-1)a
1b
2c
3
4
5
(6.1  0.2) x 103
(3.50  0.04) x 104
(3.6  1.2) x 103
(3.7  1.2) x 104
(2.8  0.6) x 104
40  5
2.0  0.1
1.25  0.20
2.4  0.4
25  1
2.3  0.1
≤ 1
(2.0  0.1) x 103
(1.6  0.1) x 104
(1.1  0.4) x 103
(1.4  0.5) x 103
(8.5  1.8) x 103
20 - 40
6
7
≤ 1
180 - 360
360  40
a ki = k1/(1 + k2/k3)
b Data from reference 12.
c Data from reference 18.
2.3 Mass spectra of EI complexes
Electrospray mass spectra of complexes of 3 - 7 with the P99 -lactamase were obtained
as described in the Experimental section. The spectra of all five compounds displayed a
dominant peak at M + (26  5) Da, where M is the mass of the parent protein (39,192 Da).
This is the same result as obtained with 1.12 In that case it was shown by means of a crystal
structure that the enzyme inactivation was achieved by insertion of a CO moiety (mass 28
Da), derived from the inhibitor, between the side chain oxygen atom of Ser 64 and the side
chain nitrogen atom of Lys 315 (with loss of two protons). Such an insertion probably arose
by the mechanism of Scheme 3. In this reaction, both X and Y are displaced by nucleophilic
attack, probably first by the active site serine residue that is the nucleophile in normal
substrate turnover, and then, in an intramolecular reaction, by Lys 315. In the present cases,
the order of displacement of X and Y is not known, but, in the case of 1, OPh was first
displaced.13 The protons released in the reaction are taken up by X and Y. Hydrolysis of
EOCOX, competing with cyclization, leads to the observation of a small amount of turnover
10

accompanying inactivation (Schemes 2 and 3), as observed for 1 and its analogues previously
studied.12-14,18
Scheme 3. Reaction of the inhibitors 1 – 7 with the P99 -lactamase.
Compound 2, in contrast to 1, and now to 3 - 7, has been reported to yield an inert
complex that retained the hydroxamate moiety of the inhibitor, i.e M + COX.18 It was
suggested that this complex arose from a conformational arrangement of the initial acyl-
enzyme intermediate to one inert to nucleophilic attack catalyzed by the enzyme and also
different from that needed to position it for attack by Lys 315 and cyclization (Scheme 3), the
latter of which must occur with 1 and 3 - 7.13 It is interesting that the hydroxamate side
chains of 5 and 6, despite their similarity to that of 2, although lacking the ammonium and
carboxylate groups, respectively, cannot achieve an inert M + COX conformation. Either the
charge or bulk of the polar terminus of 2 must favor the inert conformation rather than the
one leading to the cyclization reaction.
It was also noticed that the mass spectrum of the enzyme complex of compound 3
showed a peak at M + 185, i.e. M + CO + COX. Apparently 3 is less specific than the other
11

compounds, perhaps because of its mobile n-butyl side chain, and is able to acylate a second
enzyme functional group, e.g. Lys 67.
2.4 Structure-function issues among O-aryoxycarbonyl hydroxamate -lactamase
inhibitors
A combination of the data of Table 1 and the mass spectrometry results allows several
interesting conclusions to be drawn. First, the relative k1 values (second order rate constants
for reaction of E and I – see Scheme 2) will be discussed. The results from previous
papers12,18 show that k1 for 2 is significantly larger than that for 1. This result was interpreted
in terms of energetically favorable interactions between the polar side chain terminus of 2
and protein residues, as shown in Figure 1, particularly Arg204 and Ser212. These
interactions were proposed to restrict the mobility of 2 and thus contribute to its high rate of
reaction with the enzyme. The former of these residues was thought to interact with the
carboxylate terminus of 2 and the latter with the ammonium terminus. In addition, the
aromatic ring of Tyr221 may promote the reaction by hydrophobic reaction with the side
chain aromatic ring of 1 and the aliphatic chain of 2.
That the k1 value of 3, with only an aliphatic side chain, is comparable to that of 1 may
also reflect its hydrophobic interaction with Tyr221, as described above. Of more interest,
however, is the finding that the k1 value of 4 is greater than those of 1 and 3 and, indeed,
comparable to that of 2. This indicates that specifically placed hydrophobicity in the
hydroxamate side chain is of great importance in achieving a rapidly reacting inhibitor of this
type. The insertion of an additional methylene group into the side chain of 1 to yield 4 may
allow better positioning of the aryl group with respect to Tyr221.
12

Compounds 5 and 6 are structural analogues of 2 designed for evaluation of the
importance of the terminal carboxylate and ammonium groups of the side chain, respectively;
comparison of the reactivity of 5 and 6 is therefore of considerable interest. It is clear from
the data of Table 1 that the side chain terminal carboxylate of 5, presumably because of its
interaction with Arg 204, is much more effective in inducing reactivity than the side chain
ammonium group of 6 with its putative interaction with Ser212. Whereas 5 has comparable
activity to 2, 6 has little activity at all. This indicates that the ammonium group contributes
very little to the reactivity of 2. It must, however, contribute to the formation of a more inert
form of the acyl-enzyme (k4, Scheme 3), not susceptible to hydrolysis. 18
Compound 7 was conceived of as an easily prepared combination of 1 and 2, containing
the hydrophobic ring of 1 and the polar side chain terminal elements of 2. Its lack of
significant activity (Table 1), however, indicates a less than optimal placement of both motifs
with respect to the respective enzyme functional groups. Indeed, its failure emphasizes the
specificity of 2, 4 and 5.
The real effectiveness of these compounds as enzyme inactivators, as opposed to enzyme
acylators, is provided by the rate parameter ki, which takes into account not only the
acylation rate (k1) but also the partition of the acyl-enzyme between hydrolytic turnover (k2)
and enzyme inactivation (k3) (Scheme 2); in terms of Scheme 2, ki = k1/(1 + k2/k3).
Calculated values of ki from the experimentally determined k1 and k2/k3 values are also
shown in Table 1. The ki parameter thus identifies 2 and 5 as the most effective inactivators.
Notably 4, although the most reactive compound with the enzyme (k1), is a less effective
inactivator because of the partitioning of its derived acyl-enzyme in favor of hydrolysis. This
may derive from a relatively large k2 value, a small k3 value, or a combination of both.
13

3. Conclusions
In this paper, we have identified two important side chain determinants for the class C -
lactamase inactivating power of O-aryloxycarbonyl hydroxamates. For high activity, the
hydroxamate side chain should incorporate a hydrophobic element, as seen in 4 and a
negatively charged element, as seen in 2 and 5, both of which should have optimal positions
with respect to the enzyme surface during the acylation and inactivation reactions. It is likely
that a combination of these elements, both optimally placed (unlike in 7), would produce a
more potent -lactamase inactivator than those discovered to date, both in O-aroyl
hydroxamates and also in other platforms. The specificity of the hydroxamate moiety itself
and the aryloxide leaving group have been previously examined.13,14
4. Experimental section
4.1. General procedures
The solvents ethyl acetate, toluene, and dichloromethane (DCM) were purchased from Pharmco.
Hexanes, sodium sulfate and magnesium sulfate were purchased from Macron Fine Chemicals,
hydrochloric acid from Pharmco, benzene from EMD chemicals, triethylamine from Fluka
Chemica, phenyl chloroformate and imidazole from Sigma Aldrich, pyridine from Spectrum,
1,1’-carbonyldiimidazole (CDI), 4-butyl chloroformate, hydroxylamine hydrochloride, and -
valerolactone from Acros Organics, 1-amino-4-butanol and D-tyrosine t-butyl ester from Chem-
Impex, and O-benzylhydroxylamine hydrochloride from Aesar. Thin layer chromatography
plates were purchased from AnalTech. Silica gel for flash chromatography (32 – 63 m) was
obtained from Silicycle.
1H NMR spectra were obtained from a Varian 300 MHz spectrometer and IR spectra from a
Perkin Elmer Spectrum BX FT-IR system. NMR chemical shifts are reported as  [ppm]. High
14

resolution mass spectra were obtained from the Mass Spectrometry Laboratory, University of
Illinois at Urbana-Champaign.
The class C β-lactamase from Enterobacter cloacae P99 was purchased as a pure protein from
the Centre for Applied Microbiology and Research, Porton Down, Wiltshire, U.K. and used as
received. Bovine serum albumin (BSA) and MOPS buffer were purchased from Sigma Aldrich.
Absorption spectra and spectrophotometric reaction rates were obtained from a Hewlett Packard
8453 UV spectrophotometer.
4.2. Chemical Syntheses
Synthesis of 3 (Scheme 4)
Scheme 4
O-Benzyl-N-(4-butyloxycarbonyl) hydroxylamine (A1) A solution of O-benzylhydroxylamine
hydrochloride (1.48 g, 12.1 mmol) and pyridine (2.22 g, 28.2 mmol) in DCM (60 mL) was added
drop-wise under a nitrogen atmosphere at 0°C to a solution of 4-butyl chloroformate (1.65 g,
12.1 mmol) in DCM (120 mL). The reaction mixture was stirred for three hours at 0°C and then
slowly brought back to room temperature. The organic layer was evaporated to dryness and the
residues were dissolved in 50:50 ethyl acetate:ether. The organic layer was washed with 1 M
HCl, dried over magnesium sulfate, and evaporated to dryness to obtain A1 as a colorless oil.
Yield: 1.41 g, 52%. The product thus obtained was used in the next step without further
15

purification. 1H NMR (CDCl3, 300 MHz): 0.95 (t, 3H, J = 6.0 Hz), 1.41 (quint, 2H), 1.66 (hext,
2H), 4.14 (m, 2H), 4.85 (s, 2H), 7.18 (s, 1H, -NH), 7.38 (s, 5H).
N-(4-butyloxycarbonyl) hydroxylamine (A2) Compound A1 (1.41 g, 6.3 mmol) and 10% Pd/C
catalyst (0.2 g) were added to ethanol (20 mL). The reaction was hydrogenated in a Parr
apparatus at 45 psi for 4 hours. The catalyst was removed by filtration and the filtrate was
concentrated by rotary evaporation to obtain A2 as a colorless oil. Yield 0.55 g, 64%.
1H NMR (DMSO, 300 MHz): 0.88 (t, 3H, J = 6.2 Hz), 1.32 (m, 2H), 1.54 (m, 2H), 4.00 (m, 2H),
8.65 (s, 1H, -NH), 9.54 (s, 1H, -OH).
O-(Phenoxycarbonyl)-N-(4-butyloxycarbonyl) hydroxylamine (3) Imidazole (0.206 g, 4.1 mmol)
in DCM (3.3 mL) was added to A2 (0.546 g, 4.1 mmol) in DCM (8 mL) under a nitrogen
atmosphere at 0°C. The mixture was stirred for 20 minutes. A solution of phenyl chloroformate
(0.46 g, 4.1 mmol) in DCM (1.6 mL) was added drop-wise to the reaction mixture over one
minute. The resulting mixture was stirred for 45 minutes under nitrogen atmosphere at 0°C and
then filtered. Ethyl acetate (10 mL) was added to the filtrate to produce a colorless precipitate.
The precipitate was removed by filtration and the filtrate rotary evaporated to dryness and then
dried further overnight by means of an oil pump. The crude product was purified by
chromatography on silica gel (90:10 hexane:ethyl acetate), yielding 3 as a yellow oil (0.164 g,
14%). 1H NMR (D6-DMSO, 300 MHz): 1.18 (t, 3H, J = 6.2 Hz), 1.33 (m, 2H), 1.69 (m, 2H),
4.00 (m, 2H), 7.40 (m, 5H), 11.56 (s, 1H, -NH). FTIR (cm-1) 1734, 1799. High resolution
ES(+)MS m/z = 276.0840, calcd for C12H15H5NO5 276.0848.
Synthesis of 4 (Scheme 5)
Scheme 5
16

2-Phenylethyl chloroformate (B1) 2-Phenylethanol (2.06 g, 16.8 mmol) in DCM (62 mL) was
added to a pre-cooled solution of triphosgene (5.00 g, 16.8 mmol) in DCM (31 mL) under a
nitrogen atmosphere at -20°C and stirred for 10 minutes. Triethylamine (2.04 g, 20.2 mmol) in
DCM (9.3 mL) was added to the reaction mixture and stirred to room temperature over 4 hours.
The reaction mixture was evaporated to dryness and the resulting residue was stirred in ethyl
acetate (60 mL). The resulting suspension was then filtered, the filtrate evaporated to dryness,
and further dried overnight by means of an oil pump. The product thus obtained was used for the
next step without further purification. 1H NMR (CDCl3, 300 MHz): 3.05 (t, 2H, J = 6.9 Hz), 4.51
(t, 2H, J = 7.2 Hz), 7.30 (m, 5H). FTIR (cm-1) 1778.
O-Benzyl-N-(2-phenylethyloxycarbonyl) hydroxylamine (B2) A solution of O-
benzylhydroxylamine hydrochloride (2.8 g, 16.8 mmol) and pyridine (3.06 g, 39.1 mmol) in
DCM (83 mL) was added drop-wise under a nitrogen atmosphere to a solution of B1 (3.1 g, 16.8
mmol) in DCM (167 mL) at 0°C. The reaction mixture was stirred for three hours at 0°C and
then slowly brought back to room temperature. The reaction mixture was evaporated to dryness
and the resulting residue was dissolved in 50:50 ethyl acetate:ether. The organic layer was
washed with 1 M HCl, dried over magnesium sulfate, and evaporated to dryness to obtain the
17

crude product. This material was purified by flash chromatography on silica gel (90:10
hexane:ethyl acetate), yielding the product B2 as a yellow oil. Yield 1.00 g, 22%.
1H NMR (D6-DMSO, 300 MHz): 2.89 (t, 2H, J = 6.8 Hz), 4.22 (t, 2H, J= 5.1 Hz), 4.68 (s, 2H),
7.29 (m, 10H), 10.34 (s, 1H).
N-(2-phenylethoxycarbonyl) hydroxylamine (B3) Compound B2 (1.00 g, 3.69 mmol) was
dissolved in ethanol (20 mL), !0% Pd/C catalyst (0.49 g) was added, and the reaction was
hydrogenated at 45 psi for 4 hours. The catalyst was removed by filtration, the filtrate was
concentrated using rotary evaporation, and the residue was further dried overnight under an oil
pump vacuum. The residue was washed with ethyl acetate and the precipitate was dried under
vacuum to obtain B3 as a colorless solid. Yield 0.50 g, 75%. 1H NMR (D6-DMSO, 300 MHz):
2.89 (t, 2H, J = 6.6 Hz), 4.16 (t, 2H, J = 7.2 Hz), 7.24 (m, 5H), 8.64 (s, 1H, -NH), 9.62 (s, 1H, -
OH).
O-(Phenoxycarbonyl)-N-(2-phenylethoxycarbonyl) hydroxylamine (4) Imidazole (0.11 g, 2.2
mmol), dissolved in DCM (3 mL), was added to B3 (0.40 g, 2.2 mmol) in DCM (7 mL) under a
nitrogen atmosphere at 0°C, and the combined solution stirred for 20 minutes. A solution of
phenyl chloroformate (0.25 g, 2.2 mmol) in DCM (1.5 mL) was added drop-wise over one
minute. The mixture was stirred for a further 45 minutes under nitrogen atmosphere at 0°C and
then filtered. Ethyl acetate (10 mL) was added to the filtrate to form a colorless precipitate. The
precipitate was removed by filtration and the filtrate was rotary evaporated to dryness and then
dried by means of an oil pump for 4 hours. The crude product was purified by recrystallization
(98:2 cyclohexane:benzene), yielding the product 4 as a colorless solid. Yield 76 mg, 11%. 1H
NMR (D6-DMSO, 300 MHz): 2.89 (t, 2H, J = 6.7 Hz), 4.31(t, 2H, J = 7.1 Hz), 7.26 (m, 10H),
18

11.58 (s, 1H). FTIR (cm-1) 1722, 1786. High resolution ES(+)MS m/z = 324.0842, calcd for
C16H15NO5 324.0848.
Synthesis of 5 (Scheme 6)
Scheme 6
Potassium 5-hydroxypentanoate (C1) -Valerolactone (18.6 g, 189 mmol) and potassium
hydroxide (10.6 g, 189 mmol) were combined in water (30 mL). This mixture was heated under
reflux for three hours. Sufficient water was then to dissolve the salt and the resulting mixture
filtered. The resulting filtrate was rotary evaporated to dryness. The resulting salt was suspended
in toluene (40 mL) and evaporated to dryness. The crude product was recrystallized (95%
ethanol) and dried overnight under an oil pump vacuum to yield C1. Yield 27.6 g, 94%.
1H NMR (D2O, 300 MHz): 1.41 (br, 4H), 2.07 (t, 2H, J = 7.2 Hz), 3.45 (t, 2H, J = 5.7 Hz).
4 Methoxybenzyl 5-hydroxypentanoate (C2) C1 (18.1 g, 0.116 mol), 4-methoxybenzyl chloride
(15.8 g, 0.116 mol), and sodium iodide (333 mg, 2.22 mmol) were dissolved in
dimethylformamide (400 mL) and stirred at room temperature for 24 hours. The reaction mixture
19

was rotary evaporated to dryness and dried further overnight by means of an oil pump. The
resulting residue was dissolved in dichloromethane (125 mL) and washed with a saturated
solution of sodium bicarbonate and water. The organic layer was separated and dried over
magnesium sulfate. The solution was filtered and the filtrate rotary evaporated to obtain C2 as a
yellow oil. Yield 27.6 g, 23%.1H NMR (D6-DMSO, 300 MHz): 1.41 (m, 2H), 1.56 (m, 2H), 2.31
(t, 2H, J = 4.9 Hz), 3.37 (m, 2H), 3.77 (s, 3H), 4.41 (t, 1H, -OH, J = 3.6 Hz), 5.04 (s, 2H), 7.00
(d, 2H, J = 7.2 Hz), 7.36 (d, 2H, J = 9.1 Hz).
N-[4-(benzyloxycarbonyl)butoxycarbonyl] hydroxylamine (C3) To a solution of C2 (1.00 g, 4.2
mmol) in acetonitrile (20 mL), carbonyl diimidazole (1.05 g, 6.3 mmol) was added and the
resulting solution stirred at room temperature for 90 minutes. Imidazole (1.14 g, 16.8 mmol) and
hydroxylamine hydrochloride (1.45 g, 21.0 mmol) were then added and the mixture stirred for 4
hours. The imidazole hydrochloride precipitate was removed by filtration and the filtrate
partitioned between ethyl acetate and 0.2 M aqueous citric acid. The organic layer was washed
with a brine solution, dried over sodium sulfate, and evaporated to dryness to obtain C3 as a
colorless oil. Yield 0.74 g, 60%. 1H NMR (D6-DMSO, 300 MHz): 1.54 (m, 4H), 2.31 (t, 2H, J =
4.8 Hz), 3.73 (s, 3H), 3.95 (t, 2H, J = 3.6 Hz), 4.38 (q, 2H, J = 5.1 Hz), 4.98 (s, 2H), 6.92 (d, 2H,
J = 6.3 Hz), 7.28 (d, 2H, J = 9.0 Hz), 8.62 (s, 1H, -NH), 9.51 (s, 1H, -OH).
O-Phenoxycarbonyl-N-[(4-benzyloxycarbonyl)butoxycarbonyl] hydroxylamine (C4) Imidazole
(0.13 g, 2.50 mmol) was added to C3 (0.74 g, 2.5 mmol) in DCM (10 mL) under a nitrogen
atmosphere at 0°C and the mixture stirred for 20 minutes. A solution of phenyl chloroformate
(0.31 g, 2.5 mmol) in DCM (1.5 mL) was then added drop-wise over one minute and the ensuing
mixture was stirred for 45 minutes under the nitrogen atmosphere at 0°C. The resulting mixture
was filtered. Ethyl acetate (10 mL) was added to the filtrate which led to the formation of a
20

colorless precipitate. The precipitate was removed by filtration and the filtrate was rotary
evaporated to dryness and further dried overnight by means of an oil pump. This procedure
yielded the crude product as a yellow oil that was purified by flash chromatography on silica gel
(90:10 hexane:ethyl acetate). The required product C4 was thus obtained as a colorless oil (57.6
mg, 59% yield). 1H NMR (D6-DMSO, 300 MHz): 1.58 (m, 4H), 2.34 (t, 2H, J = 6.7 Hz), 3.74 (s,
3H), 4.11 (t, 2H, J = 5.3 Hz), 4.99 (s, 2H), 7.35 (m, 9H), 11.54 (s, 1H).
O-Phenoxycarbonyl-N-(4-carboxybutyloxycarbonyl) hydroxylamine (5) TFA (1 mL) was added
drop-wise with stirring under a nitrogen atmosphere at room temperature to a solution of C4
(100 mg) in DCM (1 mL). The subsequent reaction was monitored by TLC until it reached
completion. Excess solvent was removed via rotary evaporation, yielding the required product 5
as a colorless solid (72 mg, 100%). 1H NMR (D6-DMSO, 300 MHz): 1.57 (m, 4H), 2.23 (t, 2H, J
= 6.9 Hz), 4.12 (t, 2H, J = 5.3 Hz), 7.35 (m, 5H), 11.55 (s, 1H). FTIR (cm-1) 1757, 1794. High
resolution ES(+)MS 320.0739, calcd for C13H15NO7 320.0746.
Synthesis of 6 (Scheme 7)
Scheme 7
21

4- t-Butoxycarbonylaminobutanol (D1) To a pre-cooled solution of 4-aminobutanol (5.00 g, 56.1
mmol) and triethylamine (8.50 g, 84 mmol) in 66% aqueous dioxane (225 mL), a solution of di-
t-butyl dicarbonate (14.0 mL, 61.6 mmol) in dioxane (75 mL) was added and the mixture was
stirred at room temperature for 4 hours. The solvent was removed by rotary evaporation and the
resulting residue was partitioned between ethyl acetate and 10% aqueous citric acid. The product
was extracted from the aqueous layer with ethyl acetate three times and the combined organic
layer was washed with a 5% sodium bicarbonate solution and dried over magnesium sulfate. The
crude product was purified by fractional distillation from a Kugelrohr apparatus at 130 oC and
0.1 torr, yielding D1 as a yellow oil (8.20 g, 77% yield). 1H NMR (D6-DMSO, 300 MHz): 1.35
(s, 9H), 2.87(q, 4H), 3.35 (q, 4H), 4.34 (t, 1H, -OH, J = 5.3 Hz), 6.75 (t, 1H, J = 3.4 Hz).
N-[(4-t-butoxycarbonylamino)butoxycarbonyl] hydroxylamine (D2) To D1 (1.00 g, 5.3 mmol) in
acetonitrile (20 mL), carbonyl diimidazole (1.28 g, 7.95 mmol) was added and the mixture
stirred at room temperature for 90 minutes. Imidazole (1.44 g, 21.2 mmol) and hydroxylamine
hydrochloride (1.82 g, 26.5 mmol) were then added and the resulting mixture stirred for 4 hours.
The imidazole hydrochloride precipitate was removed by filtration and the filtrate was
partitioned between ethyl acetate and 0.2 M citric acid. The organic layer was washed with a
brine solution, dried over sodium sulfate, and evaporated to dryness to obtain D2 as a colorless
oil. Yield 1.10 g, 83%. 1H NMR (D6-DMSO, 300 MHz): 1.35 (s, 9H), 1.37 (m, 2H) 1.46 (q, 2H,
J = 8.1 Hz), 2.90 (q, 2H, J = 5.9 Hz), 3.96 (t, 2H, J = 2.9 Hz), 6.76 (t, 1H, -NH), 8.62 (s, 1H, -
NH), 9.48 (s, 1H, -OH).
O-Phenoxycarbonyl-N-[4-(t-butoxycarbonyl)butoxycarbonyl] hydroxylamine (D3) Imidazole
(0.299 g, 4.4 mmol) was added to D2 (1.10 g, 4.4 mmol) in dichloromethane (10 mL) under a
nitrogen atmosphere at 0°C. The mixture was stirred for 20 minutes. To the reaction mixture, a
22

solution of phenyl chloroformate (0.50 g, 4.4 mmol) in dichloromethane (2 mL) was added drop-
wise over one minute. The mixture was stirred for 2 hours under a nitrogen atmosphere at 0°C.
The mixture was filtered. To the filtrate, ethyl acetate (10 mL) was added, giving rise to a
colorless precipitate. The precipitate was removed by filtration and the filtrate evaporated to
dryness. The residue was further dried overnight by means of an oil pump, yielding the crude
product as a colorless solid (1.34 g). The crude product was purified by flash chromatography on
silica gel (80:20 hexane:ethyl acetate), yielding D3 as a yellow oil. Yield 427 mg, 25% yield. 1H
NMR (D6-DMSO, 300 MHz): 1.35 (s, 9H), 1.40 (m, 4H), 1.58 (q, 2H, J = 5.4 Hz), 2.91 (q, 2H, J
= 5.8 Hz), 4.15 (t, 2H, J = 5.3 Hz), 6.85 (t, 1H, -NH), 7.26 (d, 2H), 7.32 (t, 1H), 7.47 (t, 2H),
11.53 (s, 1H, -NH). FTIR (cm-1) 1757, 1805.
O-Phenoxycarbonyl-N-(4-aminobutoxycarbonyl) hydroxylamine.TFA (6) TFA (1 mL) was added
drop-wise with stirring under nitrogen at room temperature to a solution of D3 (100 mg) in
dichloromethane (1 mL). The reaction was monitored by TLC to completion over 1 hour.
Solvent and excess TFA were removed by rotary evaporation. Further drying under an oil pump
vacuum yielded the trifluoroacetate salt of the required product 6 as a yellow oil. Yield 67 mg,
100%. 1H NMR (D6-DMSO, 300 MHz): 1.64 (m, 4H), 2.49 (q, 2H, J = 1.5 Hz), 4.18 (t, 2H, J =
6.3 Hz),7.26 (d, 2H, J = 6.2 Hz), 7.32 (t, 1H, J = 8.1 Hz), 7.47 (t, 2H, J = 7.8 Hz), 7.66 (br s, 3H,
NH3+), 11.58 (s, 1H, -NH). FTIR (cm-1) 1744, 1800. High resolution ES(+)MS m/z = 269.1136,
calcd for C12H17NO5 269.1137.
Synthesis of 7 (Scheme 8)
Scheme 8
23

O-(N-Hydroxyaminocarbonyl)-D-tyrosine (E3) was prepared as shown in Scheme 7 from N-t-
butoxycarbonyl-D-tyrosine t-butyl ester using the same methodology as in the syntheses
described above.
O-[(N-Phenoxycarbonyloxy)aminocarbonyl]-N-t-butoxycarbonyl-D-tyrosine t-butyl ester (E4)
The hydroxamic acid E3 (100 mg, 0.252 mmol) and imidazole (17.2 mg, 252 mmol) were
dissolved in DCM (1.4 mL) and the solution stirred to 0 oC in an ice bath. Phenyl chloroformate
(39.6 mg, 0.252 mmol) dissolved in DCM (0.6 mL) was then added by syringe and the mixture
stirred at 0 oC for 3h. The mixture was filtered and the filtrate rotary evaporated to dryness. The
residue was stirred with ethyl acetate (4 mL) at room temperature. The precipitate was removed
by filtration and the filtrate evaporated to dryness. Purification of the product was achieved by
preparative thin layer chromatography on silica gel (30% ethyl acetate/hexane). Yield 34 mg,
44%. 1H NMR (D6-DMSO, 300 MHz): 2.96 (m, 2H), 4.00 (m, 1H), 7.08 (d, J = 7.7 Hz, 2H),
24

7.19 (d, J = 8.7 Hz, NH), 7.25 (d, J = 8.7 Hz, 2H), 7.26 (d, J = 8.7 Hz, 2H), 7.34 (t, J = 7.6 Hz,
1H), 7.48 (t, J = 7.7 Hz, 2H), 12.3 (br s, ONH). FTIR (cm-1) 1700, 1720, 1775, 1805.
O-[(N-Phenoxycarbonyloxy)aminocarbonyl]-D-tyrosine (7) The N-protected material E4 (140
mg) was dissolved in DCM (1 mL) at room temperature. Redistilled anisole (59 L, 2 eq) was
added with stirring, followed by TFA (1.0 mL). This mixture was stirred at room temperature for
5 h, after which benzene (2 mL) was added and the solvents and TFA were removed by rotary
evaporation. An additional benzene addition (1 mL) was followed by evaporation. The residue
was stirred with diethyl ether (4 mL) and the solid residue isolated by decantation. This
procedure was repeated twice more and the residue, an off-white powder, was dried under an oil
pump vacuum. Yield 38 mg, 42%. 1H NMR (D6-DMSO, 5% CF3CO2D): 3.09 (m, 2H), 4.12 (m,
1H), 7.12 (d, J = 8.7 Hz, 2H), 7.24 (d, J = 7.3 Hz, 2H), 7.28 (d, J = 8.5 Hz, 2H), 7.30 (t, J = 8.5
Hz, 1H), 7.44 (t, J = 7.5 Hz, 2H). FTIR (cm-1) 1740, 1805. High resolution ES(+)MS m/z =
361.1031, calcd for C17H17NO7 361.1036.
4.3 Analytical and Kinetics Methods
The concentration of the P99 β-lactamase in stock solutions was measured
spectrophotometrically, assuming the extinction coefficient of the enzyme to be 7.1 x 104 M-1cm-
1 at 280 nm.19 All kinetics experiments were performed at 25 oC in 20 mM MOPS buffer, pH =
7.5. Stock solutions of 3 - 7 (10 mM) were prepared in dry DMF. Spontaneous hydrolyses of 3,
5, and 6 (each 5.0 M) were monitored spectrophotometrically at 270 nm. The full progress
curves were then fitted to a first order rate equation and the rate constants k0 thus obtained. Rate
constants for hydrolysis of 4 and 7 could not be obtained in this way since no spectral change
was observed over the relevant time period. The rate constants for 4 and 7 were assumed to be
the same as that for 1 in fitting of the inactivation kinetics for these compounds.
25

Inactivation kinetics The P99 -lactamase (150 nM) was incubated with inhibitor (2.0 M and,
separately, 5.0 M 3; 5.0 M and, separately, 10.0 M 4; 0.5 M and, separately, 2.0 M 5; 20.0
M 6; and 2.5 M and, separately, 5.0 M and 10.0 M 7) in MOPS buffer containing 0.1%
bovine serum albumin. At suitable times, aliquots of the reaction mixture (20 μL) were then
added to cuvettes containing 980 μL of MOPS buffer containing cephalothin substrate (300 M,
Km = 15.4 M). Initial rates of hydrolysis of the substrate were monitored at 270 nm. The initial
rates, proportional to active enzyme concentrations, were plotted against time of reaction.
Titration experiments were also performed. Here, enzyme (150 nM) was incubated with various
concentrations of inhibitor (0 - 0.50 M 3, 0 - 2.0 M 4, 0 - 0.40 M 5, 0 - 2.0 M 6, and 0 - 2.0
M 7) and the inactivation reaction allowed to proceed until no further loss of enzyme activity
occurred. Initial rates of hydrolysis of cephalothin were then determined as described above and
the initial rates plotted as a function of initial inhibitor concentration. These data, from both
types of experiment, taken together, were fitted to Scheme 2 by means of the Dynafit program.20
To do this, since (k2 + k3) >> k1[I], k2 was fixed at 100 s-1 and the values of k1 and k3 allowed to
vary. The fit, therefore, allowed determination of k1 and k2/k3 values.
Mass spectra of inactivated enzymes Electrospray mass spectra of the P99 -lactamase before
and after inactivation by 3 – 7 were obtained exactly as previously described.13,18 This involved
acid precipitation of the enzyme by trichloroacetic acid followed by the washing and drying of
the precipitate formed.
Acknowledgments
This research did not receive any specific grant from funding agencies in the public, commercial,
or not-for-profit sectors. The authors do thank Wesleyan University, however, for institutional
support.
26

References
1. Frère J-M, Sauvage E, Kerff, F. From ”an enzyme able to destroy penicillin” to
carbapenemases: 70 years of -lactamase misbehavior. Current Drug Targets 2016; 17:
974-982.
2. Garber K. A -lactamase inhibitor revival provides new hope for old antibiotics. Nature
Rev Drug Discov. 2015; 14:445-447.
3. Bush K. A resurgence of -lactamase inhibitor combinations effective against multidrug-
resistant Gram-negative pathogens. Int J Antimicrob Agents. 2015; 46: 483-493.
4. Wright GD. Empowering older antibiotics. Cell. 2016; 167: 301.
5. Fink AL, Page, MI. The mechanisms of catalysis by -lactamases. In: Frère J-M, ed.
Beta-Lactamases. New York, NY: Nova Science; 2012: 161-198.
6. Lahiri SD, Mangani S, Durand-Reville T, Benvenuti, M, De Luca F, Sanyal G, Docquier
J-D. Structural insight into potent broad-spectrum inhibition with reversible recyclization
mechanism: avibactam in complex with CTX-M-15 and Pseudomonas aeruginosa AmpC
-lactamases. Antimicrob. Agents Chemother. 2013; 57: 2496-2505.
7. Hecker SJ, Reddy KR, Totrov M, Hirst, GC, Lomovskaya O, Griffith DC, King P,
Tsivkovski R, Sun D, Sabet M, Tarazi Z, Clifton MC, Atkins K, Raymond A, Potts KT,
Abendroth J, Boyer SH, Loutit JS, Morgan EE, Durso S, Dudley MN. Discovery of a
cyclic boronic acid -lactamase inhibitor (RPX7009) with utility vs class A serine
carbapenemases. J Med Chem. 2015; 58: 3682-3692.
8. Powers RA, Morandi F, Shoichet BK. Structure-based discovery of a novel, noncovalent
inhibitor of AmpC -lactamase. Structure. 2002; 10: 1013-1023.
27

9. Babaoglu K, Simeonov A, Irwin JJ, Nelson ME, Feng B, Thomas CJ, Cancian L, Costi
MP, Maltby DA, Jahav A, Inglese J, Austin CP, Shoichet BK. J Med Chem. 2008; 51:
2502-2511.
10. London N, Miller RM, Krishnan S, Uchida K, Irwin JJ, Eidam O, Gibold L,
Cimermancic P, Bonnet R, Shoichet BK, Taunton J. Covalent docking of large libraries
for the discovery of chemical probes. Nature Chem Biol. 2014; 10: 1066-1072.
11. Pemberton OA, Zhang X, Nichols DA, DeFrees K, Jaishankar P, Bonnet R, Adams J,
Shaw LN, Renslo AR, Chen Y. Antibacterial spectrum of a tetrazole-based reversible
inhibitor of serine -lactamases. Antimicrob Agents Chemother. 2018; 62: e02563-17.
12. Wyrembak PM, Babaoglu K, Pelto RB, Shoichet BK, Pratt RF. O-Aryloxycarbonyl
hydroxamates: New ß-lactamase inhibitors that cross-link the active site. J. Am. Chem.
Soc. 2007; 129: 9548-9549.
13. Pelto RB, Pratt RF. Kinetics and mechanism of inhibition of a serine ß-lactamase by O-
aryloxycarbonyl hydroxamates. Biochemistry. 2008; 47: 12037-12046.
14. Tilvawala R, Pratt RF. Covalent inhibition of serine ß-lactamases by novel hydroxamic
and acid derivatives. Biochemistry. 2013; 52: 3712-3720.
15. Tilvawala R, Pratt RF. Kinetics of a two-stage pro-inhibitor of serine -lactamases.
Biochemistry. 2013; 52: 7060-7070.
16. Adediran SA, Lin G, Pelto RB, Pratt RF. Crossover inhibition as an indicator of
convergent evolution of enzyme mechanisms: a -lactamase and a N-terminal
nucleophile hydrolase. FEBS Lett. 2012; 586: 4186-4189.
28

17. Adediran SA, Pratt RF. Penicillin acylase and O-aryloxycarbonyl hydroxamates: Two
acyl-enzymes, one leading to hydrolysis, the other to inactivation. Arch Biochem
Biophys. 2017; 614: 65-71.
18. Tilvawala R, Cammara M, Adediran SA, Brodbelt JS, Pratt RF. A new covalent inhibitor
of class C -lactamases reveals extended active site specificity. Biochemistry. 2015; 54:
7375-7384.
19. Joris, B., DeMeester, F., Galleni, M. Reckinger, G., Coyette, J. Frere, J.-M., Van
Beeumen, J. The -lactamase of Enterobacter cloacae P99. Chemical properties, N-
terminal sequence and interaction with 6--halogenopenicillinates. Biochem. J. 1985;
228: 241-248.
20. Kuzmic. P. Program DYNAFIT for the analysis of enzyme kinetic data: Application to
HIV proteinase. Anal. Biochem. 1996; 237: 260-272.
29