Discovery of (R)-2-amino-6-borono-2-(2-(piperidin-1-yl)ethyl)hexanoic acid and congeners as highly potent inhibitors of human arginases i and II for treatment of myocardial reperfusion injury

Article abstract of DOI:10.1021/jm400014c

Recent efforts to identify treatments for myocardial ischemia reperfusion injury have resulted in the discovery of a novel series of highly potent α,α-disubstituted amino acid-based arginase inhibitors. The lead candidate, (R)-2-amino-6-borono-2-(2-(piperidin-1-yl)ethyl)hexanoic acid, compound 9, inhibits human arginases I and II with IC₅₀s of 223 and 509 nM, respectively, and is active in a recombinant cellular assay overexpressing human arginase I (CHO cells). It is 28% orally bioavailable and significantly reduces the infarct size in a rat model of myocardial ischemia/reperfusion injury. Herein, we report the design, synthesis, and structure-activity relationships (SAR) for this novel series of inhibitors along with pharmacokinetic and in vivo efficacy data for compound 9 and X-ray crystallography data for selected lead compounds cocrystallized with arginases I and II.

Full text of DOI:10.1021/jm400014c

Article
pubs.acs.org/jmc
Discovery of (R)‑2-Amino-6-borono-2-(2-(piperidin-1-
yl)ethyl)hexanoic Acid and Congeners As Highly Potent Inhibitors of
Human Arginases I and II for Treatment of Myocardial Reperfusion
Injury
Michael C. Van Zandt,*,† Darren L. Whitehouse,† Adam Golebiowski,† Min Koo Ji,† Mingbao Zhang,†
R. Paul Beckett,† G. Erik Jagdmann,† Todd R. Ryder,† Ryan Sheeler,† Monica Andreoli,† Bruce Conway,†
Keyvan Mahboubi,† Gerard D’Angelo,† Andre Mitschler,‡ Alexandra Cousido-Siah,‡ Francesc X. Ruiz,‡
Eduardo I. Howard,‡,§ Alberto D. Podjarny,‡ and Hagen Schroeter∥
†The Institutes for Pharmaceutical Discovery, LLC, 23 Business Park Drive, Branford, Connecticut 06405, United States
‡Department of Integrative Biology, IGBMC, CNRS, INSERM, Universite
́
de Strasbourg, 1 rue Laurent Fries, 67404 Illkirch, France
§IFLYSIB, Conicet, UNLP, Calle 59 N° 789, La Plata, Argentina
∥Mars, Incorporated, 6885 Elm Street, McLean,Virginia 22101, United States
S
* Supporting Information
ABSTRACT: Recent efforts to identify treatments for myocardial ischemia reperfusion injury have
resulted in the discovery of a novel series of highly potent α,α-disubstituted amino acid-based arginase
inhibitors. The lead candidate, (R)-2-amino-6-borono-2-(2-(piperidin-1-yl)ethyl)hexanoic acid, compound
9, inhibits human arginases I and II with IC50s of 223 and 509 nM, respectively, and is active in a
recombinant cellular assay overexpressing human arginase I (CHO cells). It is 28% orally bioavailable and
significantly reduces the infarct size in a rat model of myocardial ischemia/reperfusion injury. Herein, we
report the design, synthesis, and structure−activity relationships (SAR) for this novel series of inhibitors along with
pharmacokinetic and in vivo efficacy data for compound 9 and X-ray crystallography data for selected lead compounds
cocrystallized with arginases I and II.
urea and L-ornithine. Because arginase and nitric oxide
synthases (NOS) utilize arginine as a common substrate,
upregulation of arginase may result in reduced levels of arginine
and, consequently, decreased production of nitric oxide (NO),
a physiologically important regulator of cardiac function.9
In addition to its potential role in regulating NO levels,
arginase also affects production of critical polyamines such as
putrescine, spermidine, and spermine. As arginase consumes
arginine, it produces ornithine, which is subsequently converted
to putrescine, spermidine, and spermine via the enzymes
ornithine decarboxylase, spermidine synthase, and spermine
synthase, respectively. These polyamines are important
regulators of ion channels, protein synthesis, and cell viability.
Thus, the arginase enzymes may control physiological signaling
events by controlling the intracellular levels of NO and
polyamine signal transducers.10
INTRODUCTION
■
Heart disease that leads to impaired blood flow and eventually a
myocardial infarction or heart attack is the leading cause of
death in the United States. Loss of blood flow to the heart,
typically from a coronary arterial occlusion, leads to ischemia or
lack of oxygen and nutrients and eventually death of heart
tissue. Once circulation is restored, oxidative stress from
formation of oxygen free radicals,1,2 tissue damage from
activated neutrophils,3 apoptosis,4,5 and calcium ion overload-
induced myocyte hypercontracture6,7 all contribute to
reperfusion injury. The severity of tissue damage that results
from the myocardial ischemia/reperfusion injury (MI/RI) is
dependent on the duration of ischemia and the extent of
microvascular dysfunction, inflammation, and oxidative stress
sustained during injury. Significant damage to the myocardium
can lead to reduced conduction velocities, arrhythmias and
ventricular aneurysms, each with potentially catastrophic
consequences. Reperfusion injury is a significant problem that
affects millions of people each year. New treatments that
improve patient outcomes are sorely needed.
To date, much of the work in this area has been pioneered by
the Christianson lab.11 Their discovery of 2-(S)-amino-6-
boronohexanoic acid (ABH), one of the first significant
arginase inhibitors, has made possible our current under-
standing of the role of arginase in various pathological states
and its validation as an important therapeutic target.12,13 On the
The disease pathology of MI/RI is not completely
understood but appears to be, at least, partially dependent on
the metalloprotease arginase. Arginase expression levels and
activity are increased in animal models of MI/RI.8 Arginase
catalyzes the manganese-mediated hydrolysis of L-arginine to
Received: January 3, 2013
Published: March 8, 2013
© 2013 American Chemical Society
2568
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of α,α-Disubstituted Amino Acid Analogues
a
Reagents and conditions: (a) NaHMDS, HMPA, THF, −78 °C; (b) allyl iodide, KHMDS, DME, −78 °C; (c) Li, NH3; (d) TMSCH2N2, MeOH;
(e) i. ozone, CH2Cl2, −78 °C, ii. PPh3, 4 h; (f) piperidine, NaBH(OAc)3, AcOH, CH2Cl2; (g) 6 N HCl, 95 °C, 12 h; (h) 6 N HCl, AcOH, H2O then
9 N HCl, 170 °C, 40 min, microwave; (i) TFA, CH2Cl2; (j) CH2O, NaCNBH3.
basis of work from Christianson and others, we initiated a drug
discovery program to identify novel potent arginase inhibitors
that would be suitable for clinical development as treatments
for myocardial ischemia/reperfusion injury (MI/RI). We
directed our efforts to a structure-based design program using
ABH as a starting template. After studying the arginase I
binding interactions with the amino acid-based inhibitor, we
rapidly identified the α-position of the amino acid as an
excellent site for novel substituents that could provide
additional binding interactions with the protein.14 In particular,
Asp 181 and Asp 183 (ArgI) appeared to be ideally positioned
to form new ionic interactions with the inhibitor to improve
binding affinity. Herein we report our initial efforts to identify,
prepare, and evaluate this new class of α,α-disubstituted amino
acid-based arginase inhibitors, and the discovery of 9, a novel
second generation arginase inhibitor with significant activity in
a rat model of MI/RI.
amino)-2-(2-(piperidin-1-yl)ethyl)hexanoic acid, 14, are de-
scribed to exemplify the general synthetic methods.
To obtain final compounds with the necessary configuration,
we began with (4S,5S)-tert-butyl 6-oxo-4,5-diphenyl-1,3-oxazi-
nane-3-carboxylate, 1, which was prepared as described by
Williams.15 In an effort to make the synthesis more convergent,
the protected butane boronic acid moiety was introduced in a
single step via alkylation with (4-iodobutyl)boronic acid pinacol
ester (2) using NaHMDS and HMPA in THF. Iodide 2 was
prepared from 4-bromobutene in three steps: hydroboration
with catechol borane followed by transesterification with
pinacol and displacement of the bromide with sodium iodide
in acetone (details in experimental section). Introduction of the
second alkyl group to the oxazinone core proved challenging.
Initial attempts with a variety of electrophiles (e.g., methyl
iodide, benzyl iodide, and allyl iodide), using a variety of bases
(LDA, KHMDS, LiHMDS, and NaH) in a variety of polar
aprotic solvents (e.g., DMF, THF, DMA, and DMSO) with a
variety of additives (including 18-crown-6, HMPA, and
TMEDA) provided only modest yields and required many
equivalents of base and electrophile. Eventually, we discovered
that the use of dimethoxyethane (DME) as the solvent
dramatically enhanced the reaction rate and that good
electrophiles like allyl iodide could be introduced very
effectively, even at −78 °C. Although it is not entirely clear
why this second alkylation was so difficult or why the use of
DME resulted in such a dramatic acceleration of the reaction, it
may be due, at least in part, to the boron atom in the side chain.
Once the oxazinone moiety is deprotonated, a stable 6-
membered boronate complex may be formed that effectively
blocks the enolate from reaction with electrophiles. DME as a
solvent may be more effective at preventing formation of this
cyclic complex or reducing its stability.
INHIBITOR DESIGN AND SYNTHESIS
■
As we began investigating methods to prepare our α,α-
disubstituted amino acid target compounds, we wanted to
utilize a flexible late-stage intermediate that could be prepared
on large research scale with very high enantioselectivity (>98%
ee). We also wanted the ability to make either enantiomer and
have the asymmetric center fixed in an early intermediate to
simplify the synthesis and characterization of final compounds.
After evaluating many approaches, we found the Williams′
diphenyloxazinone to be a highly effective optically active
glycine equivalent that could meet our needs. With our
objective of identifying novel ABH analogues containing
substituents in the α-position to form new interactions with
the Asp 181 and Asp 183 residues, we decided to pursue
compounds with a substituted ethyl amine substituent. These
compounds could be conveniently prepared from an
intermediate aldehyde using a reductive amination reaction.
This method was developed and is illustrated in Scheme 1.
Here, the syntheses of (R)-2-amino-6-borono-2-(2-(piper-
idin-1-yl)ethyl)hexanoic acid, 9, and (R)-6-borono-2-(methyl-
Disubstituted oxazinone (4) and its enantiomer were
prepared as common intermediates to provide access to target
compounds with the necessary stereochemistry for activity
(e.g., 9) and their inactive enantiomers to serve as negative
controls. To establish optical purity of compounds derived
2569
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
from intermediate 4 or its enantiomer, compound 9 and its
inactive enantiomer 15 (Table 1) were analyzed using a
and inactive enantiomers, and in all cases, the inactive
compounds had no activity (<10% inhibition at 300 μM).
With the asymmetric center firmly established, the chiral
auxiliary could be removed to facilitate new analogue synthesis.
Treatment of oxazinone 4 with lithium in ammonia gave N-Boc
protected amino acid 5, which, after allowing the ammonia to
evaporate, was redissolved in methanol and treated with TMS-
diazomethane to provide methyl ester 6 in 58% yield (2 steps).
Ozonolysis in dichloromethane followed by reduction of the
ozonide with triphenylphosphine gave aldehyde 7, which could
be easily purified using column chromatography and was stable
at room temperature for several days or could be stored in the
refrigerator without significant decomposition for several
months.
Table 1. Inhibition of Recombinant hARG I and hARG II
Enzymes by Aminoethylene ABH Analogues
Aldehyde 7 was prepared in gram quantities and served as a
key intermediate for new analogue synthesis. Reductive
amination with piperidine and sodium triacetoxyborohydride
in acetic acid-dichloromethane gave amine 8, which was
deprotected with 6 N HCl at 95 °C overnight to give target
amino acid 9. Although these compounds are very polar and
have no chromophore, they could be conveniently purified
using reverse phase preparative HPLC equipped with an ELSD
detector (details in experimental section).
To facilitate a larger scale synthesis of compound 9 and other
analogues, an alternative synthetic route was developed. As also
illustrated in Scheme 1, oxazinone 4 could be subjected to the
same ozonolysis and reductive amination conditions to provide
amine 11 in 69% yield (2 steps). Subsequent cleavage of the
chiral auxiliary via hydrogenolysis, however, proved to be
exceedingly slow and required nearly stoichiometric palladium.
Eventually, we developed a new method to remove the
oxazinone auxiliary. Treatment with 9 N HCl in a microwave
reactor for 40 min at 170 °C conveniently cleaved the auxiliary
resulting in efficient formation of target amino acid 9 in 87%
yield. On the basis of our experience, this method seems to be
quite general and effective for acid and thermally stable
oxazinones.
The N-methylated analogue 14 could also be prepared from
oxazinone 11 as illustrated in Scheme 1. In this case, selective
deprotection of the Boc-group with TFA in dichloromethane
followed by reductive amination with formaldehyde using
sodium cyanoborohydride gave N-methyl oxazinone 13 in 73%
yield (2 steps). Final deprotection using the aforementioned
microwave conditions provided the target N-methyl analogue,
compound 14, in 76% yield.
RESULTS AND DISCUSSION
■
a
b
Stereochemistry of quaternary amino acid center. Enzyme potency
The primary objective of this program was to identify novel,
highly potent and efficacious arginase inhibitors to treat
individuals with MI/RI. All program compounds were initially
tested for potency in human recombinant arginase I and II
enzyme assays (hARG I and II), measuring the inhibition of
arginase-induced urea production. Selected compounds were
subsequently evaluated in a cellular functional assay with stably
transfected human arginase I in CHO cells; the end point
assessed was again inhibition of urea production. Results from
these experiments are listed in Tables 1 and 2. A selected early
compound, 9, which was potent against hARG I in both in vitro
enzyme and cellular assays, was selected for pharmacokinetic
evaluation and subsequently was evaluated in a rat model of
MI/RI. Results from these studies are provided in Table 3 and
Figure 9.
data is generated in triplicate. Compounds 9 and 14 were tested on
multiple occasions. The combined data expressed as mean S.E.mean
and number of experiments (in parentheses) for these compounds are
as follows: compound 9, ArgI IC50 223 22.39 and Arg II IC50 509
85.1;10 and compound 14, ArgI IC50 59.7 9.13 and Arg II IC50 67 +
13.3.3
Supelco Chirobiotic T2 analytical HPLC column and were
found to have an optical purity of >98% ee (details provided as
Supporting Information). Since all analogues were derived from
the same intermediates and the quaternary chiral center is
nonepimerizable, optical purity measurements of final com-
pounds using chiral HPLC were not done on a routine basis.
Several compounds, however, were prepared as both the active
2570
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
Table 2. Inhibitory Effects of Select Compounds in CHO
Cells Over-Expressing hArgI
example #
inhibition of CHO-hARG I (IC50)
9
8 μM
14
3 μM
15
inactive
19 μM
ABH
Table 3. Pharmacokinetic Profile of Compound 9 in Rat
route dose (mg/kg)
i.v. 10
p.o. 10
AUC0‑inf (mg·h/L)
C0 or Cmax (mg/L)
t1/2 (h)
21.2
40.3
3.30
7.89
1.86
n/a
5.92
0.45
n/a
CL (mL/min/kg)
Vss (L/kg)
n/a
n/a
F (%)
28%
Figure 1. Structure of compound 9 (stick model) in the active site
pocket of arginase I (surface colored according to charge), is shown
superimposed with a difference map calculated without the inhibitor
model contoured at 3 RMS (white contours). Note that the map
shows very clearly the inhibitor structure and indicates that the
piperidine ring has two conformations, a major one (yellow sticks,
occupancy 0.66, ring in distorted boat conformation) and a minor one
(gold sticks, occupancy 0.33, ring in chair conformation).
The use of X-ray crystallography early in the program was
essential. The structures provided a clear understanding of the
important interactions that make up the enzyme−inhibitor
complexes. An iterative structure based design approach was
then used to drive analogue synthesis, focused on introducing
new interactions with residues at the mouth of the active site
pocket, particularly with Asp 181 and Asp 183.
In general, the structure−activity relationships associated
with the compound class (Table 1) are consistent with binding
information derived from the enzyme−inhibitor complexes. In
compound 9, the α substituent, which consists of a piperidine
linked to the α carbon by a two-carbon aliphatic chain, results
in formation of new through-water hydrogen bonding
interactions with Asp 181 and Asp 183 and provides a 6-fold
increase in potency relative to ABH. The magnitude of this
improvement decreases when larger substituents force the side
chain amine into a conformation where the interactions
between the basic amine and carboxylate of Asp 181 and Asp
183 are suboptimal. Figure 1 shows the binding of compound 9
to the active site pocket of arginase I and the corresponding
electron density. Two alternative conformations are observed
for the piperidine ring: a distorted boat (major occupancy) and
a chair (minor occupancy). While in the distorted boat
conformation, the amine nitrogen forms hydrogen bonding
interactions through water molecule W1 with Asp 181 (3.10 Å)
and Asp 183 (3.70 Å) (Figure 3). In the chair conformation,
amine nitrogen is pointing toward the solvent and does not
have strong contacts with the protein (Figure 1). Moreover, the
minor chair conformation of the piperidine ring (in gold)
places the piperidine C1 atom of 9 in a position that is too close
to water molecule W1 (1.89 Å, Figure 2), implying that W1 is
not present in the population of molecules containing the chair
conformer. Therefore, the water-mediated contacts with Asp
181 and Asp 183 likely do not exist for the chair conformation,
which may explain its lower occupancy despite the fact that it is
the more thermodynamically stable conformer.
Figure 2. Minor chair conformation of the piperidine ring (in gold)
where the inordinately short distance between the piperidine C1 atom
(1.89 Å) and W1 implies that W1 is not present in this conformation
in the arginase I: compound 9 complex.
(2.66 Å). The carboxylic acid forms direct interactions with Ser
137 (OH, 2.57 Å) and Asn 130 (3.14 Å) and hydrogen bonding
interactions (through three water molecules) with Thr 135
(2.91 Å), Asn 139 (2.99 Å), and Asp 183 (2.72 Å) (Figure 4).
The butylboronic acid chain of the inhibitor is buried in the
active site cleft and makes contacts similar to those of ABH.
The oxygen atoms linked to the boron form close contacts with
the arginase Mn ions, and these substituents are surrounded by
aspartic acid residues Asp 128 (2.63 Å) and Asp 232 (3.14 Å)
(Figure 3).
Because of the many important protein interactions with the
amino acid amine N1, only limited modifications are possible.
Interestingly, as seen with compound 14, incorporation of a
simple methyl substituent at position R1 results in a nearly 4-
fold increase in potency. As illustrated in Figure 5, the binding
mode is very similar to that of compound 9, except the
Like ABH, the amino acid and boronic acid moieties form
multiple hydrogen bonding or ionic interactions that hold the
inhibitor securely within the active site. Figure 3 shows the
contacts between the protein, the Mn ions, and the major boat
conformation of 9. Amine N1 (amino acid amine) of the
inhibitor forms a direct contact with Asp 183 (2.91 Å) and
water mediated contacts with Glu 186 and Gly 142 through
water molecule W3 (3.0 Å), and with Asp 181 through W2
2571
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
Figure 3. Contacts of compound 9 with the arginase 1 and the Mn
atoms in the major conformation (distorted boat).
Figure 5. Contacts of compound 14 (stick model) within the active
site pocket of arginase I (surface colored according to charge),
superimposed with a difference map calculated without the inhibitor
model contoured at 3 RMS (white contours). Note that the map
shows very clearly the inhibitor structure and that the piperidine ring
adopts only the chair conformation.
Figure 4. Contacts of the carboxylic acids, Asn 139, Thr 135, and Asp
183, in the arginase I: compound 9 complex.
orientation of the piperidine ring resides exclusively in the more
thermodynamically stable chair conformation.
As expected, the contacts for compound 14 (Figure 6) are
very similar to those of compound 9. In each case, the inhibitor
forms an almost identical network of hydrogen bonding
interactions with the key active site amino acids and waters
W1 and W3. The major difference is that water molecule W2 is
ejected by the methyl group. Although this results in the loss of
W2 hydrogen bonding interactions, the increased potency
(lower IC50) for compound 14 (60 nM for compound 14
versus 223 nM for compound 9) implies a net increase in
binding energy. This probably results from a gain in entropy
due to the liberation of water molecule W2 and the
strengthening of existing hydrogen bonds made possible by
minor conformational changes of the inhibitor and active site
amino acid side chains.
Figure 6. Contacts of compound 14 with the arginase I and Mn atoms.
piperidine ring nitrogen atom and the aspartic acid residues
(181 and 183 in arginase I; 200 and 202 in arginase II) are
maintained.
As shown in Figure 7, for the arginase II complex, the
piperidine ring (contacting both aspartic acid residues) is in the
chair conformation, and one of the distances, between water
molecule W1 and Asp 202 (4.27 Å), is beyond hydrogen bond
range, while it can make a weak hydrogen bond (3.95 Å) in the
arginase I complex. This difference may partially explain the
weaker binding of compound 9 to arginase II (IC50 509 nM)
relative to arginase I (IC50 223nM).
Example 9 was also cocrystallized with arginase II (see details
in Supporting Information). The two structures are very
similar; in particular, the through-water contacts between the
We have also solved the complex of arginase II cocrystallized
with compound 14 (see Supporting Information). In this case,
the inhibitor positions are very close for both Arg I and Arg II
2572
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
at 300 μM. Modification of the piperidine ring generally did not
result in improved potency. Analogues with a sulfur atom (16),
a 4′-hydroxyl substituent (17), or a benzo-moiety (18) were all
less active. Similar results were observed with the pyrrolidine
series. The addition of a hydroxymethyl (20) or methox-
ymethyl substituent (21, 22) to the parent pyrrolidine (19) also
did not improve potency. It is interesting to note the difference
in activity between the analogues with the methoxymethyl
substituents in the R- and S-configurations. In compound 21
(R-stereochemistry), the ether oxygen is able to form an
additional weak hydrogen bonding interaction with Ser 155
(OH, 3.78 Å, structural data not shown). This likely contributes
to the increased potency observed for compound 21 relative to
compound 22. Benzo-modification of the pyrrolidine side chain
(23) results in about a 2-fold decrease in potency relative to the
unsubstituted compound (19). This is consistent with what is
observed for the piperidines (compounds 9 and 18). Potency
for compounds 24−27, with acyclic side chains (ethyl,
hydroxyethyl; diethyl; methyl, propyl; or methyl, isopropyl)
varied depending on the size of the amine substituents. Those
with a small substituent (methyl 26 or hydrogen 27) were more
potent than those with two larger substituents (24 and 25).
This presumably results from a more optimal binding
interaction with Asp 181 and Asp 183 when the amine is less
sterically hindered.
To better characterize these compounds as arginase
inhibitors, they were evaluated in a cellular functional assay
with stably transfected human arginase I in CHO cells. As
illustrated in Table 2, the rank-order potencies for compounds
9 and 14 and ABH observed in the cell assay are consistent
with those obtained in the recombinant hARG I enzyme assay
(Table 1). The IC50 values for compounds 9 and 14 are 8 and 3
μM, respectively, each about 50-fold less active than that
observed in the isolated enzyme assay. For example, 15, the
inactive enantiomer of 9, was completely inactive in the cellular
assay. ABH has an IC50 of 19 μM in this assay. The differences
in activity between the isolated enzyme assay and the cellular
assay is not fully understood at present. Presumably, the
charged polar nature of these compounds precludes passive
diffusion through the cell membrane and an active transport
mechanism may be involved. Further work is required to
investigate the mechanism of cellular absorption.
Figure 7. Superimposition of the arginase I−compound 9 complex
(green carbons, red water W1, and green distances) and the arginase
II−compound 9 complex (yellow carbons, magenta water W1, and
yellow distances).
(see Figure 8). For these compounds, the distances between
the water molecule W1 and the aspartic acids are all within
A pharmacokinetic evaluation of compound 9 was conducted
after intravenous (i.v.) and oral (p.o.) dosing in male Sprague−
Dawley rats (n = 3 per dose route). The compound was
formulated in 0.9% saline and administered intravenously at 10
mg/kg by bolus through a preimplanted cannula at a dosing
volume of 1 mL/kg, and orally at 10 mg/kg via gavage at a
dosing volume of 2 mL/kg. As illustrated in Table 3, following
i.v. dosing with 10 mg/kg in fasted animals, compound 9 has a
terminal elimination half-life (t1/2) of 3.3 h with a volume of
distribution and total body clearance of 1.86 L/kg and 7.89
mL/min/kg, respectively. The oral bioavailability of compound
9 (10 mg/kg, p.o.) was 28% with a Cmax of 0.45 mg/L.
Figure 8. Superimposition of the arginase I−compound 14 complex
(green carbons, red water W1, and green distances) and the arginase
II−compound 14 complex (yellow carbons, magenta water W1, and
yellow distances).
To determine if arginase inhibition following myocardial
ischemia/reperfusion injury (MI/RI) would limit tissue
damage, compound 9 and its inactive enantiomer 15 were
evaluated in a rodent model of MI/RI, involving occlusion of
the left-main coronary artery for 45 min. Also included in the
study were sham (no occlusion), vehicle (occlusion, no drug),
and groups receiving intermittent preconditioning (IPC) as a
positive control. The IPC group was implemented as 3 × 5 min
of intermittent occlusion prior to the 45 min of sustained
hydrogen bonding range (Figure 8). This is consistent with the
statistically equivalent IC50 values for Arg I and Arg II (60 nM
and 67 nM, respectively).
As expected, the small, highly polar enzyme active site
accommodates only amino acids with the natural R-stereo
configuration. Unnatural amino acids with the S-configuration
like compound 15 are completely inactive with <10% inhibition
2573
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
Figure 9. Effect of compounds 9 and 15 on infarct size expressed as a percent of the area at risk (AAR; left panel) and percent of the left ventricle
(right panel) 2 days postreperfusion in a rat model of MI/RI. Compounds 9 (100 mg/kg, i.v.) and 15 (100 mg/kg, i.v.) were given as bolus
administration 15 min before occlusion of the left main coronary artery. Intermittent preconditioning (IPC) was implemented as 3 × 5 min of
intermittent occlusion prior to the 45 min sustained ischemia. *p < 0.05 vs vehicle; the n values for each group are indicated in the individual
columns.
on a Gilson preparative LC system consisting of a robotic liquid
handler 215, a syringe pump 402, an injection module 845z, and a
binary pump 322, using a Phenomenex C18 amino acid preparative
column with a Gilson ELSD detector. Using these methods, purity was
determined to be >95% except where noted. Analytical data for all
compounds is provided as Supporting Information. Thin-layer
chromatography using glass-backed silica plates containing a
fluorescent indicator (0.25 mm, Whatman, Merck) was used to
monitor reactions. The chromatograms were visualized using ultra-
violet illumination, exposure to iodine vapors, or dipped in an aqueous
potassium permanganate solution. All starting materials were used
without further purification. All reactions were carried out under an
atmosphere of dried nitrogen or argon. Compound names are based
on IUPAC convention and are derived from the structures using
ChemDraw Ultra 12.0.
ischemia. After 2 days, infarct size was measured and expressed
as a percent of the area at risk (AAR) and as a function of left
ventricle size. Results of the study are illustrated in Figure 9.
Infarct size in the control group averaged 51.8% of the AAR,
while IPC produced a reduction of infarct size to 33.1% of the
AAR. Treatment with compound 9 (100 mg/kg, i.v.) reduced
infarct size to 35.2% of AAR, similar to the IPC group, while
using compound 15, the inactive enantiomer of compound 9,
had no effect on infarct size.
In conclusion, a novel second generation series of α,α-
disubstituted amino acid ABH analogues has been identified
and optimized to be highly potent inhibitors of both human
arginase I and II. Significant improvements in potency were
achieved by introducing an aminoethylene substituent to form
new hydrogen bonding interactions with Asp 181 and Asp 183
near the mouth of the active site. In vivo evaluation of
compound 9, (R)-2-amino-6-borono-2-(2-(piperidin-1-yl)-
ethyl)hexanoic acid, in a rat model of MI/RI resulted in a
significant reduction in infarct size, which was similar to the
IPC. Continued development of this new class of compounds
may result in an important new treatment for ischemia
reperfusion injury.
General Procedures for the Synthesis of Substituted 2-Amino-2-
(2-aminoethyl)-6-boronohexanoic Acids. The 2-amino-2-(2-amino-
ethyl)-6-boronohexanoic acids in Table 1 were prepared from (4S,5S)-
tert-butyl-6-oxo-4,5-diphenyl-1,3-oxazinane-3-carboxylate 1 or its
enantiomer (compound 15) and 2-(4-iodobutyl)-4,4,5,5-tetramethyl-
1,3,2-dioxaborolane (2) as outlined in Scheme 1 via the general
methods described for compounds 9 and 14 below. (4S,5S)-tert-Butyl-
6-oxo-4,5-diphenyl-1,3-oxazinane-3-carboxylate (1) and its enantiomer
were prepared according to the methods described by Williams.15
Synthesis of 2-(4-Iodobutyl)-4,4,5,5-tetramethyl-1,3,2-dioxabor-
olane (2). Step 1: Synthesis of 2-(4-Bromobutyl)-4,4,5,5-tetrameth-
yl-1,3,2-dioxaborolane. While under nitrogen, neat catecholborane
(100 g, 833 mmol) in a 1 L round-bottomed flask was cooled in an ice
bath and carefully treated with 4-bromobutene (124.2 g, 920 mmol)
via syringe at a rate such that the temperature remained below 20 °C.
After the addition was complete, the ice bath was removed, and the
solution was allowed to warm to room temperature over 30 min, then
warmed to 95 °C for 6 h. After recooling to room temperature, the
reaction mixture was diluted with anhydrous THF (300 mL), treated
with pinacol (217.5 g, 1.84 mol) in one portion, and stirred for an
additional 16 h, while still under a nitrogen atmosphere. The resulting
solution was concentrated to give the crude product as a pale yellow
oil that was diluted with heptane (800 mL) and cooled to −40 °C in a
dry ice/acetone bath. After stirring for 1 h, the precipitated catechol
and excess pinacol were removed by filtration to give 2-(4-
bromobutyl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane as a colorless oil
(195 g, 89%), which generally contained 2−6% free catechol (this
material could be further purified by repeated precipitation from
chilled heptane, although this level of catechol contamination was not
Samples of compounds 9 and 15 will be provided, when
possible, to facilitate further preclinical research into the role of
arginases in MI/RI and other relevant diseases. Please contact
the corresponding author for details.
EXPERIMENTAL SECTION
■
Chemistry. General Methods. Melting points were determined in
open capillary tubes on a Thomas-Hoover apparatus and are
uncorrected. Proton nuclear magnetic resonance (1H NMR) spectra
were determined on a Bruker Avance III 400 MHz NMR
spectrometer. Chemical shifts are provided in parts per million
(ppm) downfield from tetramethylsilane (internal standard) with
coupling constants in hertz (Hz). Multiplicity is indicated by the
following abbreviations: singlet (s), doublet (d), triplet (t), quartet
(q), multiplet (m), and broad (br). Mass spectra were recorded on a
PE Sciex API 100 LC/MS system equipped with a Perkin-Elmer 200
Series autosampler and LC pump. Elemental analyses (C, H, N, S, Cl,
and Br) were performed by Atlantic Microlab, Inc. (Norcross,
Georgia) and are within 0.4% of theory unless otherwise noted.
Intermediate products from all reactions (nonpolar compounds) were
purified by either flash column chromatography or medium pressure
liquid chromatography using a Biotage Isolera One with SNAP
cartridges unless otherwise indicated. All final products were purified
1
detrimental to the subsequent step); H NMR (CDCl3, 400 MHz) δ
3.38 (t, J = 6.6 Hz, 2 H), 1.90−1.78 (m, 2 H), 1.58−1.44 (m, 2 H),
1.26 (s, 12 H), 0.78 (t, J = 7.5 Hz, 2 H).
Step 2: Synthesis of 2-(4-Iodobutyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (2). While under a nitrogen atmosphere, a solution of
2574
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
crude 2-(4-bromobutyl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (195
g, 0.741 mol) and sodium iodide (333.6 g, 2.24 mol) in acetone (500
mL) was heated to 55 °C overnight. After cooling to room
temperature, the solution was concentrated, diluted with heptane (1
L), cooled to 0 °C with stirring for 30 min, and filtered. The filtrate
was concentrated to give 2-(4-iodobutyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (2) as a colorless oil (189 g, 83%). H NMR (CDCl3,
400 MHz) δ 3.18 (t, J = 7.2 Hz, 2 H), 1.90−1.78 (m, 2 H), 1.58−1.44
(m, 2 H), 1.24 (s, 12 H), 0.79 (t, J = 7.5 Hz, 2 H).
butyl)morpholine-4-carboxylate (10). A solution of (3R,5R,6S)-tert-
butyl-3-allyl-2-oxo-5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxa-
borolan-2-yl)butyl)morpholine-4-carboxylate (4, 65.0 g, 113 mmol) in
DCM (400 mL) was cooled to −78 °C and treated with ozone until a
pale blue−gray color persisted. After thin-layer chromatography
indicated the absence of starting material, the ozone inlet tube was
replaced with a nitrogen inlet tube, and nitrogen was bubbled through
the solution for 20 min to remove any excess ozone. Triphenylphos-
phine (44.5 g, 169.5 mmol) was added in one portion, the cooling
bath was removed, and the mixture was stirred for 16 h. The resulting
solution was concentrated and purified by suction filtration through a
short pad of silica gel (10−50% ethyl acetate in heptane) to give
(3R,5R,6S)-tert-butyl-2-oxo-3-(2-oxoethyl)-5,6-diphenyl-3-(4-(4,4,5,5-
tetramethyl-1,3,2-dioxaborolan-2-yl)-butyl)-morpholine-4-carboxylate,
10, as a colorless foam (48.1 g, 73%), which was used without further
1
Large Scale Synthesis of (R)-2-Amino-6-borono-2-[2-piperidin-1-
yl)-ethyl]-hexanoic Acid (9). Step 1: Synthesis of (3R,5R,6S)-tert-
Butyl 2-Oxo-5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxabor-
olan-2-yl)butyl)morpholine-4-carboxylate (3). A solution of
(2S,3R)-tert-butyl-6-oxo-2,3-diphenylmorpholine-4-carboxylate (1,
68.0 g, 192.8 mmol) and 2-(4-iodobutyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (2, 74.7 g, 241.0 mmol, 1.25 equiv) in THF (1000 mL)
and HMPA (100 mL) was cooled to −78 °C and treated with sodium
bis(trimethylsilyl)amide (250.6 mL, 1.0 M, 250.6 mmol) dropwise
over 60 min. After stirring for 2 h at −78 °C, the solution was warmed
to 0 °C with stirring for an additional 4 h and then warmed slowly to
room temperature overnight. The solution was diluted with ethyl
acetate (1000 mL) and quenched with saturated aqueous NH4Cl. The
organic solution was washed successively with water and saturated
aqueous sodium chloride, dried over anhydrous MgSO4, filtered, and
concentrated to give the product as a sticky white gum. Heptane (800
mL) was added, and the slurry was stirred at room temperature for 2 h
to precipitate the product. The product was vacuum filtered to give a
white solid that was dried in a vacuum desiccator overnight, to afford
(3R,5R,6S)-tert-butyl-2-oxo-5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)butyl)morpholine-4-carboxylate (3) as a white
solid (85.8 g, 83%). [α]D +39.64° (c 3.91, CHCl3); 1H NMR (CDCl3,
400 MHz) δ 7.25 (m, 3 H), 7.12 (m, 1 H), 7.06 (m, 2 H), 6.98 (m, 2
H), 6.57 (t, J = 7.5 Hz, 2 H), 5.93 (d, J = 3 Hz, 1 H), 5.20 and 4.99 (d,
J = 3 Hz, ∼ 1:2 rotamers, combined 1 H), 5.01 and 4.81 (dd, J1 = 10.5,
J2 = 4.5 Hz, ∼ 1:2 rotamers, combined 1 H) 2.15 (m, 1 H), 1.98 (m, 1
H), 1.52−1.66 (m, 4 H), [1.44 (s, 3 H) and 1.08 (s, 6 H); ∼1:2 Boc
rotamers], 1.24 (s, 12 H) 0.85 (m, 2 H); LC-MS 558.3 (M + Na)+,
536.3 (M + H)+, 480.3 (M + H-iBu)+, 436.2 (M + H-Boc)+; Anal.
Calcd for C31H42BNO6, C, 69.53; H, 7.91; N, 2.62. Found C, 69.36; H,
7.80; N, 2.55.
1
purification. [α]D −60.21° (c 7.23, CHCl3); H NMR (CDCl3, 400
MHz) δ 9.91(s, 1 H), 7.49−7.27 (m, 10 H), 7.18 (s, 1 H), 6.35 (s, 1
H), 3.25 (d, J = 1 3 Hz, 1 H), 2.20 (br s, 1 H), 1.93 (br s, 1 H), 1.55
(m, 1 H), 1.50 (s, 9 H), 1.14 (s, 12 H), 1.03 (m, 2 H), 0.97 (m, 2 H),
0.68 (t, J = 7.6 Hz, 2 H); LC-MS 578.3 (M + H)+, 522.2 (M + H-
iBu)+, 478.2 (M + H-Boc)+; Anal. Calcd for C33H44BNO7, C, 68.63;
H, 7.68; N, 2.43. Found C, 68.53; H, 7.64; N, 2.42.
Step 4: Synthesis of (3R,5R,6S)-tert-Butyl 2-Oxo-5,6-diphenyl-3-
(2-(piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lan-2-yl)butyl)morpholine-4-carboxylate (11). A solution of alde-
hyde 10 (38.0 g, 65.86 mmol) and piperidine (13.15 mL, 131.7 mmol)
in 1,2-dichloroethane (250 mL) and acetic acid (7.8 mL, 131.7 mmol)
was stirred at room temperature for 60 min and then treated with
sodium triacetoxyborohydride (34.91 g, 164.65 mmol) in 4 portions.
After stirring for 16 h at room temperature, saturated aqueous
NaHCO3 was added until the solution reached pH 8−9, and stirring
was continued for an additional 5 min. The resulting mixture was
added to a separatory funnel, diluted with saturated aqueous sodium
chloride (50 mL), and extracted with DCM (3 × 250 mL). The
organic layer was dried over anhydrous MgSO4, filtered, and
concentrated under reduced pressure. Purification by flash column
chromatography, eluting with 30−100% ethyl acetate in heptane
(+0.5% NEt3), gave a ∼ 20:1 mixture of (3R,5R,6S)-tert-butyl-2-oxo-
5,6-diphenyl-3-(2-piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)-butyl)-morpholine-4-carboxylate (LC-MS 647.3
(M + H)+, 591.2 (M + H-iBu)+, 547.2 (M + H-Boc)+) and
(3R,5R,6S)-4-(tert-butoxycarbony)-2-oxo-5,6-diphenyl-3-(2-piperidin-
1-yl)ethyl)morpholin-3-yl)butyl boronic acid (formed by silica-
promoted deprotection of the pinacol protecting group) as a colorless
foam (40.3 g, 95%). [α]D −32.4° (c 8.81, CHCl3); 1H NMR (CDCl3,
400 MHz) δ 7.41−7.20 (m, 10 H), 7.18 (s, 1 H), 6.22 (s, 1 H), 2.60−
2.09 (m, 6 H), 2.06 (m, 1 H), 1.52 (s, 9 H), 1.52−1.30 (m, 7 H), 1.10
(s, 12 H), 1.10−0.96 (m, 4 H), 0.78 (m, 2 H), 0.72 (m, 2 H); LC-MS
565.3 (M + H)+, 509.2 (M + H-iBu)+, 465.2 (M + H-Boc)+. This
mixture could be used in the subsequent deprotection steps, without
separation of the boronic acid from the boronate.
Step 5: Synthesis of (R)-2-Amino-6-borono-2-(2-(piperidin-1-
yl)ethyl)hexanoic Acid (9). A solution of (3R,5R,6S)-tert-butyl-2-
oxo-5,6-diphenyl-3-(2-piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)butyl)morpholine-4-carboxylate 11 (40.3 g) in
6 N hydrochloric acid (300 mL) and acetic acid (20 mL) was warmed
to 70 °C with stirring overnight. After cooling to room temperature,
the reaction mixture was transferred to a separatory funnel, diluted
with deionized water (50 mL), and washed with DCM (3×). The
aqueous layer was concentrated to give the crude boronic acid
intermediate as a white foam (34.2 g) that was redissolved in 9 N
hydrochloric acid (200 mL) and transferred into 20 mL Biotage
microwave reactor vials (20 tubes). The vials were capped and then
heated, with stirring, in a Biotage Initiator microwave reactor to 170
°C for 40 min. After all 20 vials had been processed, the combined
aqueous phase was washed with DCM (2 × 50 mL) and concentrated
to give (R)-2-amino-6-borono-2-(2-piperidin-1-yl)ethyl)hexanoic acid,
dihydrochloride monohydrate salt as a very hygroscopic white foam
(24.1 g). 1H NMR (D2O, 300 MHz) δ 3.34 (d, J = 11.5 Hz, 2 H), 3.14
(m, 1 H), 2.97 (m, 1 H), 2.77 (t, J = 12 Hz, 2 H), 2.19 (t, J = 8.5 Hz, 2
Step 2: Synthesis of (3R,5R,6S)-tert-Butyl 3-Allyl-2-oxo-5,6-
diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)-
morpholine-4-carboxylate (4). A solution of (3R,5R,6S)-tert-butyl-2-
oxo-5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-
butyl)morpholine-4-carboxylate (3, 70.0 g, 123.2 mmol) and TMEDA
(100 mL) in DME (1,200 mL) was cooled to −78 °C, charged with
allyl iodide (310.5 g, 1.85 mol, 15 equiv), and treated with potassium
bis(trimethylsilyl)amide (962 mL, 1.0 M in THF, 862 mmol, 7 equiv)
dropwise over 60 min. After stirring for 2 h, a second portion of allyl-
iodide (150 g, 0.89 mol) was added, followed by the dropwise addition
of additional KHMDS (400 mL, 1.0 M in THF). After stirring an
additional 3 h at −78 °C, the reaction mixture was slowly warmed to
room temperature overnight. Once complete as indicated by thin-layer
chromatography, the reaction mixture was quenched with saturated
aqueous sodium chloride and extracted with diethyl ether (3 × 1.5 L).
The combined organic phase was washed successively with water and
saturated aqueous NH4Cl, dried over anhydrous MgSO4, filtered, and
concentrated. Purification by suction filtration through silica gel (0−
20% ethyl acetate in heptane) gave (3R,5R,6S)-tert-butyl-3-allyl-2-oxo-
5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)-
morpholine-4-carboxylate as a pale yellow oil (4, 65.5 g, 92.5%). [α]D
1
−43.13° (c 6.38, CHCl3); H NMR (CDCl3, 400 MHz) δ 7.02−7.18
(m, 11 H), 6.09 (d, J = 3 Hz, 1 H), 5.86 (m, 1 H), 5.22 (m, 2 H), 3.21
(m, 1 H), 2.84 (m, 1 H), 2.17 (m, 2 H), 1.52 (s, 9 H), 1.10−1.45 (m, 9
H), 1.18 (s, 12 H), 0.48 (t, J = 7 Hz, 2H); LC-MS 598.3 (M + Na)+,
576.3 (M + H)+, 510.3 (M + H-iBu)+, 476.3 (M + H-Boc)+; Anal.
Calcd for C34H46BNO6, C, 70.95; H, 8.06; N, 2.43. Found C, 71.02; H,
8.27; N, 2.55.
Step 3: Synthesis of (3R,5R,6S)-tert-Butyl 2-Oxo-3-(2-oxoethyl)-
5,6-diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-
2575
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
H), 1.76 (m, 4 H), 1.55 (m, 3 H), 1.23 (m, 4 H), 1.06 (m, 1 H), 0.59
(t, J = 7.5 Hz, 2 H); LC-MS 287.2 (M + H)+, 269.3 (M + H-H2O)+,
251.4 (M + H-2H2O)+; Anal. Calcd for C15H33BN2O4·2HCl, 1.25
H2O salt, expected C 40.92, H 8.32, Cl 18.58, N 7.34; measured C
40.90, H 7.95, Cl 18.45, N 6.97.
Dowex-50WX8−400 ion resin (Aldrich cat# 217514, 600 g) was
washed by stirring successively with MeOH (2×) and water (2×), then
briefly suction dried. The crude amino acid was dissolved in water
(100 mL) and stirred with the resin at room temperature for 2 h (until
LC-MS analysis of the aqueous phase failed to detect the amino acid
MH+ ion). The resin was filtered and then transferred to a sintered
chromatography column. Water (∼300 mL) was run through the resin
column until an aliquot of the collected water fraction failed to give a
positive AgNO3 test (i.e., once all of the chloride had been successfully
eluted from the column).
room temperature, the reaction mixture was transferred to a separatory
funnel, diluted with deionized water, and washed with DCM. The
aqueous layer was concentrated to give the crude boronic acid
intermediate as a white foam that was redissolved in 9 N hydrochloric
acid and transferred into 20 mL Biotage microwave reactor vials. The
vials were capped and then heated, with stirring, in a Biotage Initiator
microwave reactor to 170 °C for 40 min. After all vials had been
processed, the combined aqueous phase was washed with DCM and
concentrated to give (R)-6-borono-2-(methylamino)-2-(2-(piperidin-
1-yl)ethyl)hexanoic acid, dihydrochloride monohydrate salt as a very
hygroscopic white foam.
Dowex-50WX8-400 ion resin (Aldrich cat# 217514) was washed by
stirring successively with MeOH (2×) and water (2×), then briefly
suction dried. The crude amino acid was dissolved in water (15 mL)
and stirred with the resin at room temperature for 2 h (until LC-MS
analysis of the aqueous phase failed to detect the amino acid MH+
ion). The resin was filtered and then transferred to a sintered
chromatography column. Water (∼50 mL) was run through the resin
column until an aliquot of the collected water fraction failed to give a
positive AgNO3 test (i.e., once all of the chloride had been successfully
eluted from the column).
The free-based amino acid was then recovered from the resin by
eluting with aqueous 4 N NH4OH solution. Fractions were collected
until the aqueous solutions were negative for ninhydrin staining and
showed no MH+ ion in LC-MS analysis. The combined product
containing fractions were concentrated to give the purified amino acid
as a nonhygroscopic free base hydrate (1.32 g, 76%). 1H NMR (D2O,
400 MHz) δ 3.57 (t, J = 10.4 Hz, 2 H), 3.40−3.28 (m, 1 H), 3.21−
3.12 (m, 1 H), 3.04−2.91 (m, 2 H), 2.71 (s, 3 H), 2.45−2.28 (m, 2 H),
2.05−1.86 (m, 4 H), 1.85−1.66 (m, 3 H), 1.53−1.40 (m, 3 H), 1.39−
1.28 (1 H), 1.28−1.15 (m, 1 H), 0.80 (t, J = 7.6 Hz, 2 H); 13C NMR
(D2O, 400 MHz) δ 171.88, 66.75, 53.58, 53.35, 51.52, 31.82, 27.31,
26.61, 24.99, 23.30, 22.88, 22.83, 21.00; ESI-LCMS m/z calcd for
C14H29BNO4, expected 300.2; found 301.3 (M + H)+.
The salt-free amino acid was then recovered from the resin by
eluting with aqueous 4 N NH4OH solution. Fractions were collected
until the aqueous solutions were negative for ninhydrin staining and
showed no MH+ ion in LC-MS analysis. The combined product-
containing fractions were concentrated to give (R)-2-amino-6-borono-
2-(2-(piperidin-1-yl)ethyl)hexanoic acid (9) as a nonhygroscopic free
1
base hydrate (17.9 g, 87% from 11). [α]D +15.4° (c 2.27, H2O); H
NMR (D2O, 400 MHz) δ 2.84−3.05 (m, 6 H), 1.89 (m, 2 H), 1.68 (br
s, 5 H), 1.51 (m, 3 H), 1.31 (m, 3 H), 1.08 (m, 1 H), 0.70 (t, J = 7.0
Hz, 2 H); 13C NMR (D2O, 400 MHz) δ 180.04, 62.22, 53.99, 53.20
(2× C), 38.98, 32.13, 25.86, 23.85, 23.53 (2× C), 21.72, 14.18; 11B-
NMR (D2O, 400 MHz) singlet at δ 32.55; LC-MS 309.4 (M + Na)+,
287.3 (M + H)+, 269.2 (M + H-H2O)+, 251.1 (M + H-2H2O)+; Anal.
Calcd for C15H33BN2O4·1.5 H2O salt, expected C 49.85, H 9.65, Cl
0.00, N 8.94; measured C 49.95, H 9.58, Cl trace < 0.25, N 8.77.
Synthesis of (R)-6-Borono-2-(methylamino)-2-(2-(piperidin-1-yl)-
ethyl)hexanoic Acid (14). Step 1: Synthesis of (3R,5R,6S)-5,6-
Diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)butyl)morpholin-2-one (12). A solution of
(3R,5R,6S)-tert-butyl 2-oxo-5,6-diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-
(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)morpholine-4-
carboxylate, 11 (5.16 g, 7.98 mmol), in DCM (50 mL, 0.16 M) was
treated with TFA (10 mL) and stirred at room temperature. After
stirring overnight, the solution was concentrated under reduced
pressure, dissolved in water, and treated with saturated aqueous
NaHCO3 to pH 9−10. The aqueous solution was extracted with ethyl
acetate (3×), dried over anhydrous MgSO4, filtered, and concentrated
to give (3R,5R,6S)-5,6-diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-(4-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)morpholin-2-one
(12) as a colorless oil (3.86 g, 89%) that was used in the next step
without further purification.
Synthesis of (R)-2-Amino-6-borono-2-[2-piperidin-1-yl)-ethyl]-
hexanoic acid (9) via Lithium in Ammonia Reduction. Step 1:
Synthesis of (R)-Methyl 2-Allyl-2-(tert-butoxycarbonylamino)-6-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)hexanoate (6). A
three-necked RBF equipped with nitrogen inlet tube and dry ice
condenser was charged with (3R,5R,6S)-tert-butyl 3-allyl-2-oxo-5,6-
diphenyl-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)-
morpholine-4-carboxylate (4.60 g, 8.00 mmol) and THF (10 mL).
After cooling the condenser to −78 °C and the flask to −45 °C (CO2
(s), CH3CN), ammonia (80 mL) was condensed into the flask. Once
complete, lithium metal (0.55 g, 80 mmol, small pieces) was carefully
added over 10 min. After stirring an additional 40 min, the reaction
mixture was carefully quenched with NH4Cl (s) until the solution
became clear. The bath was removed, and the ammonia was allowed to
evaporate overnight. The resulting residue was diluted with ethyl
acetate and washed successively with 0.5 N HCl and saturated aqueous
NaCl, dried over anhydrous MgSO4, filtered, and concentrated.
Without further purification, the crude product was dissolved in 50%
methanol in toluene (80 mL, 0.1 M) and treated with TMS-
diazomethane (2.0 M in hexanes) until the pale yellow color persisted.
With TLC indicating the reaction complete, the excess TMS-
diazomethane was quenched with acetic acid until the solution
became clear. The solution was concentrated, diluted with ethyl
acetate, washed successively with saturated aqueous NaHCO3 and
saturated aqueous NaCl, dried over anhydrous MgSO4, filtered, and
concentrated. Purification by MPLC (1−60% ethyl acetate in heptane
over 6 CV) gave (R)-methyl 2-allyl-2-(tert-butoxycarbonylamino)-6-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)hexanoate as colorless oil
(1.9 g, 58%). Rf 0.46 (30% ethyl acetate in heptane); [α]D +6.51° (c
Step 2: Synthesis of (3R,5R,6S)-4-Methyl-5,6-diphenyl-3-(2-
(piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-
2-yl)butyl)morpholin-2-one (13). A solution of (3R,5R,6S)-5,6-
diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetramethyl-1,3,2-di-
oxaborolan-2-yl)butyl)morpholin-2-one, 12 (3.86 g, 7.06 mmol), in
THF (50 mL) and acetonitrile (150 mL) was treated with aqueous
formaldehyde (5 mL, 37%), acetic acid (1 mL), and NaBH3CN (1.01
g) and stirred at room temperature. After stirring overnight, the
solution was diluted with saturated aqueous NaHCO3, extracted with
ethyl acetate (3×), dried over anhydrous MgSO4, filtered, and
concentrated. Purification by MPLC (0−10% methanol in DCM)
gave (3R,5R,6S)-4-methyl-5,6-diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-
(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)butyl)morpholin-2-
1
one (13) as a colorless foam (3.26 g, 82%). H NMR (CDCl3, 400
MHz) δ 7.10−6.98 (m, 6 H), 6.96−6.90 (m, 4 H), 5.92 (d, J = 6.8 Hz,
1 H), 4.06 (d, J = 6.8 Hz, 1 H), 2.58−2.32 (m, 7 H), 2.36 (s, 3 H),
2.06−1.98 (m, 1 H), 1.68−1.36 (m 12 H), 1.20 (s, 12 H), 0.82 (m, 2
H).
Step 3: Synthesis of (R)-6-Borono-2-(methylamino)-2-(2-(piper-
idin-1-yl)ethyl)hexanoic Acid (14). A solution of (3R,5R,6S)-4-
methyl-5,6-diphenyl-3-(2-(piperidin-1-yl)ethyl)-3-(4-(4,4,5,5-tetra-
methyl-1,3,2-dioxaborolan-2-yl)butyl)morpholin-2-one 13 (3.26 g,
5.82 mmol) in 6 N hydrochloric acid (25 mL) and acetic acid (3
mL) was warmed to 95 °C with stirring overnight. After cooling to
1
5.38, CHCl3); H NMR (CDCl3, 400 MHz) δ 5.70−5.52 (m, 1 H),
5.49−5.36 (m, 1 H), 5.05 (dd, J1 = 13.8 Hz, J2 = 1.2 Hz, 1 H), 3.73 (s,
3 H), 3.09−2.96 (m, 1 H), 2.50 (dd, J1 = 13.8 Hz, J2 = 7.8 Hz, 1 H),
2.29−2.10 (m, 1 H), 1.78−1.65 (m, 1 H), 1.43 (s, 9 H), 1.42−1.26
(m, 4 H), 1.23 (s, 12 H), 0.74 (t, J = 7.5 Hz, 2 H); ESI-LCMS m/z
calcd for C21H38BNO6, expected 411.3; found 412.3 (M + H)+.
2576
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
Step 2: (R)-Methyl 2-(tert-Butoxycarbonylamino)-2-(2-oxoethyl)-
6-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)hexanoate (7). A
solution of (R)-methyl 2-allyl-2-(tert-butoxycarbonylamino)-6-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)hexanoate (1.90 g, 4.62
mmol) in DCM (90 mL, 0.05 M) was cooled to −78 °C and treated
with ozone until a pale blue−gray color appeared. After TLC indicated
the absence of starting material, the ozone inlet tube was replaced with
nitrogen, and nitrogen was bubbled through the solution for 20 min to
remove any excess ozone. Triphenylphosphine (3.6 g, 13.8 mmol, 3
equiv) was added in one portion, the cooling bath was removed, and
the mixture was stirred for 4 h. The solution was concentrated and
purified by MPLC (1−50% ethyl acetate in heptane over 6 CV) and
gave (R)-methyl 2-(tert-butoxycarbonylamino)-2-(2-oxoethyl)-6-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)hexanoate as a colorless
oil (1.28 g, 67%). Rf 0.55 (30% ethyl acetate in heptane); [α]D +12.67°
A recombinant, fully active, truncated form of human arginase II
was constructed to circumvent aggregation problems observed with
the wild-type enzyme.19,20 The truncation variant ΔM1-V23/ΔH331-
I354 (114 kDa trimer) was created by site-directed mutagenesis using
mature human arginase II cDNA. The following primers were used to
generate the variant: sense mutagenic primer, TACTAACAT-
ATGGTCCACTCCGTTGCTGTGATAGGAGCC, and antisense
mutagenic primer, ACTGTACTCGAGTCAATGCCCTCCTTCT-
CTTGTCTGACCAAAGCTTGAAGC. The truncated variant, cloned
in pET23b, was overexpressed as described for arginase I. The pellet
from a 6 L culture was disrupted by sonication in a buffer of 10 mM
Tris·HCl pH 8.0, 10 mM NaCl, 1 mM β-mercaptoethanol, and 1 mM
MnSO4, and centrifuged at 4 °C. As for arginase I, partial purification
was performed by heating at 60 °C for 20 min. After heating, the
protein was clarified by centrifugation at 45 000 rpm for 30 min. Then
ammonium sulfate was added to the supernatant to 60% saturation,
and after centrifugation, the pellet was resuspended in a buffer of 10
mM Tris·HCl pH 8.0, 10 mM NaCl, 1 mM β-mercaptoethanol, and 1
mM MnSO4. The resuspended protein was dialyzed in the same buffer
and then applied to an anion exchange, capto Q column of 25 mL (GE
Healthcare) and eluted with a KCl gradient.
Enzymatic Assays. Arginase converts L-arginine to L-ornithine and
urea, and relative arginase activity was determined by measuring urea
levels in a colorimetric assay according to published methods.21
Purified recombinant full length human arginase I protein and
recombinant, fully active, truncated form of human arginase II were
used to develop 96-well plate human arginase I and arginase II assays.
Inhibition of arginase I and arginase II by program compounds was
followed spectrophotometrically at 530 nm. The compound to be
tested was dissolved in DMSO at an initial concentration that was 50-
fold greater than its final concentration in the cuvette. Ten microliters
of the stock solution was diluted in 90 μL of the assay buffer that
comprised 0.1 M sodium phosphate buffer containing 130 mM NaCl,
pH 7.4, to which was added ovalbumin (OVA) at a concentration of 1
mg/mL. Solutions of arginase I and II were prepared in 100 mM
sodium phosphate buffer, pH 7.4, containing 1 mg/mL of OVA to give
an arginase stock solution at a final concentration of 100 ng/mL.
To each well of a 96-well microtiter plate was added 40 μL of
enzyme, 10 μL of the test compound solution, and 10 μL of enzyme
substrate solution (L-arginine + manganese sulfate). For wells that
were used as positive controls, only the enzyme and its substrate were
added, while wells used as negative controls contained only manganese
sulfate.
After incubating the microtiter plate at 37 °C for 60 min, 150 μL of
a urea reagent obtained by combining equal proportions (1:1) of
reagents A and B was added to each well of the microtiter plate to stop
the reaction. The urea reagent was made just before use by combining
Reagent A (10 mM o-phthaldialdehyde and 0.4% polyoxyethylene
(23) lauryl ether (w/v) in 1.8 M sulfuric acid) with Reagent B (1.3
mM primaquine diphosphate, 0.4% polyoxyethylene (23) lauryl ether
(w/v), and 130 mM boric acid in 3.6 M sulfuric acid). After quenching
the reaction mixture, the microtiter plate was allowed to stand for an
additional 10 min at room temperature in order to allow color
development. The inhibition of arginase was computed by measuring
the optical density (OD) of the reaction mixture at 530 nm and
normalizing the OD value to percent inhibition observed in the
control. The normalized OD was then used to generate a
concentation−response curve by plotting the normalized OD values
against log[concentration] and using regression analysis to compute
the IC50 values.
Functional Cell Assay. Generation of CHO Cells Stably
Expressing Human Arginase I. The human arginase I (hARGI)
expression plasmid was purchased from Origene and subcloned into
pcDNA3.1(+) expression vector. Chinese hamster ovary cells (CHO)
cells were transfected with hARGI -pcDNA3.1(+) construct using
Lipofectamine 2000 according to the manufacturer’s instructions
(Invitrogen). After 2 days, cells were subjected to selective media
containing 800 μg/mL of Geneticin (Invitrogen). Single colonies of
cells expressing hARGI were obtained by using limiting dilutions.
Clones were first screened for hARGI expression by Western blotting
1
(c 7.50, CHCl3); H NMR (CDCl3, 300 MHz) δ 9.66 (s, 1 H), 5.62
(br s, 1 H), 3.75 (s, 3 H), 3.60 (br d, J = 17.4 Hz, 1 H), 2.95 (d, J =
17.4 Hz, 1 H), 2.30−2.15 (m, 1 H), 1.70−1.54 (m, 1 H), 1.40 (s, 9 H),
1.39−1.24 (m, 4 H), 0.74 (t, J = 7.8 Hz, 2 H); ESI-LCMS m/z calcd
for C20H36BNO7, expected 413.3; found 414.3 (M + H)+; Anal. Calcd
for C20H36BNO7, C, 58.12; H, 8.78; N, 3.39. Found C, 58.29; H, 8.94;
N, 3.35.
Step 3: Synthesis of (R)-Methyl 2-((tert-Butoxycarbonyl)amino)-
2-(2-(piperidin-1-yl)ethyl)-6-(3,3,4,4-tetramethylborolan-1-yl)-
hexanoate (8). A solution of (R)-methyl 2-(tert-butoxycarbonylami-
no)-2-(2-oxoethyl)-6-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-
hexanoate (0.148 g, 0.358 mmol, 1.0 equiv) and piperidine (0.046 g,
0.54 mmol, 1.5 equiv) in 1,2-dichloroethane (0.34 mL, 0.5 M) was
treated with sodium triacetoxyborohydride (0.19 g, 0.90 mmol, 2.5
equiv) in one portion. After stirring for 1.5 h, the reaction mixture was
quenched with saturated aqueous NaHCO3 (1 mL) and stirred for an
additional 5 min. The resulting mixture was added to a separatory
funnel, diluted with saturated aqueous NaCl (5 mL), and extracted
with dichloromethane (2 × 10 mL). The organic layer was dried over
anhydrous MgSO4, filtered, and concentrated under reduced pressure.
Purification by flash column chromatography eluting with 5%
methanol in chloroform gave (R)-methyl 2-((tert-butoxycarbonyl)
amino)-2-(2-(piperidin-1-yl)ethyl)-6-(3,3,4,4-tetramethylborolan-1-
yl)hexanoate (8) as a pale yellow oil (0.161 g, 93%). Rf 0.45 (10%
methanol in DCM); [α]D −3.37° (c 12.05, CHCl3); 1H NMR
(CDCl3, 300 MHz) δ 6.10−6.03 (br s, 1 H), 3.63 (s, 3 H), 2.48−2.02
(br s, 6 H), 1.89−1.61(br s, 4 H), 1.52−1.41 (br s, 4 H), 1.27 (s, 9 H),
1.25−1.22 (br s, 5 H), 1.16 (s, 12 H), 0.98 (br s, 1 H), 0.75 (t, J = 7.8
Hz, 2 H); ESI-LCMS m/z calcd for C25H47BN2O6, expected 482.4;
found 483.4 (M + H)+.
Step 4: (R)-2-Amino-6-borono-2-(2-(piperidin-1-yl)ethyl)hexanoic
Acid (9). A solution of ester 8 (0.187 g, 0.388 mmol) in 6 N HCl (5
mL) was heated to a gentle reflux for 16 h. After cooling to room
temperature, the reaction mixture was transferred to a separatory
funnel, diluted with deionized water (5 mL), and washed with DCM
(3 × 5 mL). The aqueous layer was frozen in liquid nitrogen and
lyophilized to give amino acid 9 as a dihydrochloride salt (0.124 g,
89%). Analytical data as previously described.
Pharmacology. Enzyme Preparation. Recombinant human
arginase I cloned in pET-24a (Novagen) was overexpressed and
purified using the procedure described previously.16,17 Briefly, E. coli
BL21(DE3) strain (Novagen) transformed with the pET-24a plasmid
was grown in LB medium at 37 °C for 3 h. Protein expression was
induced by the addition of 1 mM isopropyl-1-thio-β-D-galactopyrano-
side (IPTG) (Euromedex). The pellet from a 6 L culture was
disrupted by sonication in 10 mM Tris·HCl pH 8.0, 10 mM NaCl, 1
mM β-mercaptoethanol, 1 mM MnSO4, and centrifuged at 4 °C.
Partial purification of arginase I was done by heating at 60 °C for 20
min, based on the heat stability of this protein.18 After heating, the
protein was clarified by centrifugation at 45 000 rpm for 30 min. The
supernatant was further purified in a hitrap SP FF column of 5 mL
(GE Healthcare) equilibrated with 10 mM Tris·HCl pH 8.0, 10 mM
NaCl, 1 mM β-mercaptoethanol, 1 mM MnSO4, and eluted with a KCl
gradient.
2577
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
and subsequently for arginase activity (see below). The highest
expressing clone was chosen to move forward and was maintained in
media containing 400 μg/mL of Geneticin.
After the completed reperfusion for 2 days, animals were
anesthetized with isoflurane (2.5−3%). Following a midline
thoracotomy, the heart was arrested in diastole by infusion of cold
cardioplegia solution into the left ventricle via a 25 gauge needle
inserted into the cardiac apex. The heart with coronary snare suture
was excised and mounted onto a modified Langendorff isolated-heart
perfusion system. The coronary artery was then reoccluded, and Evans
blue/Monastral blue dye was injected into the cannulated aorta for
delineation of the area-at-risk. The fresh heart was weighed and the
right ventricle removed, and the left ventricle was sectioned into 2 mm
thick slices from cardiac apex to base and incubated with
triphenyltetrazolium chloride (1% in PBS) at 37 °C for 15−30 min
followed by immersion in 10% formalin in phosphate buffered saline.
Fixed slices were digitally scanned for image analysis using Image J
software.
The study tested the effect of compound 9 (100 mg/kg, i.v.) and
compound 15 (100 mg/kg, i.v.), the inactive enantiomer of compound
9. For both studies, compounds were administered as a bolus 15 min
before occlusion of the coronary artery. Results are expressed as mean
SEM; statistical analysis was conducted using X with p < 0.05
regarded as significant. The MI/RI study was conducted in the
laboratory of Dr. Francisco Villarreal at the University of California,
San Diego, on a fee-for-service basis.
X-ray Crystallography. Crystallization of Human Arginase I
and the Truncation Variant ΔM1-V23/ΔH331-I354 of Human
Arginase II. Crystals of the native enzyme arginase I were obtained by
vapor diffusion method at 4 °C. Drops containing 4.0 μL of enzyme
solution [5 mg/mL human arginase I, 10 mM Tris·HCl pH 7.5, 10
mM NaCl, 1 mM β-mercaptoethanol, and 0.2 mM MnSO4] and 4 μL
of precipitant solution [100 mM malonic acid, imidazol, boric system
pH5, and 25% polyethylene glycol (PEG) 1500] were equilibrated
against a 1 mL reservoir of precipitant solution. Rod-like crystals
appeared in approximately one week and grew to typical dimensions of
0.1 mm × 0.1 mm × 0.4 mm. Soaking was used to obtain crystals of
the arginase I−inhibitor complexes.22−24 Native crystals of arginase I
were transferred in a drop containing 15 mM compounds 9 or 14
dissolved in the precipitant solution (100 mM malonic acid, imidazol,
boric system pH 5, and 25% PEG 1500). The crystals were soaked
during one week then cryoprotected by the precipitant solution
containing 30% ethylene glycol prior to flash cooling in liquid
nitrogen. Crystals of the human arginase II−inhibitor complexes were
obtained by cocrystallization of arginase II with 2 mM compound 9 or
14. Crystals were obtained by vapor diffusion method at 4 °C. Protein
was concentrated at 5 mg/mL in 10 mM Tris·HCl pH 8.0, 10 mM
NaCl, 1 mM β-mercaptoethanol, and 0.2 mM MnSO4. Crystals were
obtained under the H3 condition of a Silver Bullets screen (Hampton
Research) using Tacsimate in the reservoir. Crystals appeared in a few
days and grew to typical dimensions of 0.3 mm × 0.3 mm × 0.5 mm.
They were cryoprotected by the reservoir solution containing 30%
ethylene glycol prior to flash cooling in liquid nitrogen.
Arginase Activity Measurement in CHO Cells. Parental CHO and
CHO clone stably transfected with hARGI were seeded in a 96-well
plate. The next day, the media was replaced with new media
containing 5 mM arginine. The cells were incubated at 37 °C. After 24
h, the media was removed and the amount of urea in the media (which
serves as an indicator of arginase I activity), was measured using a
spectroMAX plate reader at 530 nm as described in the hrARG
screening assay. The concentration of urea in each sample was
calculated using urea standards with known concentrations of urea.
Inhibition of Arginase I by Arginase Inhibitors in CHOK Cells
Stably Transfected with hARGI. Parental CHOK and CHOK clones
stably transfected with hARGI were seeded in a 96-well plate. The next
day, test compounds are added to cells in media containing 5 mM
arginine. Each compound was dissolved in PBS and was tested at 0.1,
0.3, 1, 3, 10, 30, 100, and 300 μM. Five wells of a 96-well plate were
used for each concentration tested. The cells were incubated at 37 °C
for 24 h, after which the amount of urea was measured using a
spectroMAX plate reader at 530 nm as described in the hrARG
screening assay. The amount of urea generated from parental CHO
cells (background) were subtracted from all the values. The urea
production in the absence of any compound was considered the
maximum urea, and the % inhibition of urea production in the
presence of each compound was calculated. The Prism application was
used to depict an IC50 curve and to determine the IC50 values for each
compound.
Pharmacokinetic Evaluation of Select Inhibitors in Rats. Single
dose pharmacokinetics were evaluated in male Sprague−Dawley rats
by intravenous bolus (i.v.) and oral (p.o.) dosing. Three rats were
evaluated per dose group. Compound was freshly formulated in 0.9%
saline prior to dosing, at 10 mg/mL for i.v. dosing at a dose volume of
1 mL/kg animal body weight, and at 5 mg/mL for p.o. dosing at a
dose volume of 2 mL/kg. Animals were fasted overnight prior to
dosing, with water given ad libitum. Compound was administered i.v.
through a preimplanted cannula or orally by gavage. Food was
reintroduced to animals 4 h following dosing. Blood samples were
collected through preimplanted cannulae (dual cannulated animals
used for i.v. dosing; blood collection separate from dosing cannula) at
0.25 mL per draw, followed by volume replacement with 0.9% saline.
Samples were collected at predose 0.083 (i.v. only), 0.25, 0.5, 1, 2, 4, 6,
8, and 24 h following dosing. Blood samples were maintained on ice
and centrifuged at 10 000g to obtain plasma. Plasma was frozen at −20
°C prior to analysis. Analysis was performed by LC/MS/MS using
Agilent 1100 HPLC pumps with detection on a PESciex API 4000
Qtrap. Pharmacokinetic evaluation was performed using standard
noncompartmental analyses in Phoenix WinNonlin software.
Efficacy in a Rat Model of Myocardial Ischemia/Reperfusion
Injury (MI/RI). Studies were performed to determine the effect of
arginase inhibition on infarct size following MI/RI. Male Sprague−
Dawley rats (250−300 g; Charles River Laboratories) were
anesthetized with a ketamine/xylazine mixture (100 mg/kg:8 mg/kg,
i.p.) and orally intubated with a modified endotracheal tube for
mechanical ventilation. The right jugular vein was cannulated using
PE-50 tubing and the heart exposed via left thoracotomy in the fourth
intercostal space. For animals designated to receive ischemia/
reperfusion, a complete occlusion of the left-main coronary artery
was achieved by tightening a releasable snare. The pericardium was
incised, and a silk 6-O suture was placed around the vessel and a small
portion of the underlying epicardium. Complete coronary occlusion
was confirmed visually by noting the prompt and sustained pallor of
the anterior wall distal to the ligation site. Occlusion was maintained
for 45 min followed by release of the snare; reperfusion was confirmed
by prompt return of color to the myocardium distal to the occlusion
site. For animals designated as sham, the snare was placed but not
tightened. After the surgery, the animals were allowed to recover and
returned to their cages, given water and standard rat chow ad libitum
and monitored daily until the terminal procedure.
Data Collection. For both crystals of human arginase I− and II−
compound 9 complexes, X-ray diffraction data were collected,
respectively, at 1.30 and 1.60 Å resolution on X06DA beamline at
the synchrotron Swiss Light Source (SLS), Switzerland. The crystals
belong to space group P3 with unit cell parameters of a = b = 90.14 Å
and c = 69.33 Å and to space group P42212 with unit cell parameters of
a = b = 127.74 Å and c = 159.11 Å.
For both crystals of the arginase I− and II−compound 14
complexes, X-ray diffraction data were collected, respectively, at 1.45
and 1.80 Å resolution on X06SA beamline at the synchrotron SLS.
The crystals belong to space group P3 with unit cell parameters of a =
b = 90.32 Å and c = 69.53 Å and to space group P42212 with unit cell
parameters of a = b = 127.81 Å and c = 159.09 Å. All data were
processed with HKL2000.25
Both data collection and refinement statistics for the four complexes
are included as Supporting Information.
Structural Refinement. The structures of all complexes arginases I
and II with inhibitors 9 and 14 were solved by molecular replacement
with AMORE using the structure of a rat arginase protein complexed
2578
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
with a boronic inhibitor (Protein Data Bank ID: 1D3V) as a search
model, against 8−3 Å data. Clear solutions were always obtained.
Further crystallographic refinement involved several repeated cycles
of conjugate gradient energy minimization and temperature factor
refinement, performed with the CCP4 suite.26 Amino acid side-chains
were fitted into 2Fo−Fc and Fo−Fc electron density maps. The final
Fo−Fc maps indicated a clear electron density for each different
inhibitor. Water molecules were fitted into difference maps, and riding
H-atoms were introduced in the final cycles. The programs Phenix27
and Coot were used for refinement and fitting the models to the
electron density.28 Both arginase I complexes were solved from twin
crystals and were refined according to the twin law “-h,-k, l”.
Refinement statistics are presented in Table 1. The atomic coordinates
have been deposited in the Protein Data Bank (PDB IDs: 4HWW,
4HXQ, 4HZE and 4I06) and will be released upon publication. The
inhibitor binding sites were analyzed using Coot,28 while figures were
generated with the PyMOL Molecular Graphics System (Delano
Scientific LLC).
silazane; NO, nitric oxide; NOS, nitric oxide synthase; SAR,
structure−activity relationship; TFA, trifluoroacetic acid; THF,
tetrahydrofuran; TMEDA, tetramethylenediamine;
TMSCH2N2, trimethylsilyl-diazomethane
REFERENCES
■
(1) Zweier, J. L.; Talukder, M. A. H. The role of oxidants and free
radicals in reperfusion injury. Cardiovasc. Res. 2006, 70, 181−190.
(2) Alkaitis, M. S.; Crabtree, M. J. Recoupling the cardiac nitric oxide
synthases: tetrahydrobiopterin synthesis and recycling. Curr. Heart
Failure Rep. 2012, 9, 200−210.
(3) Vinten-Johansen, J. Involvement of neutrophils in the patho-
genesis of lethal myocardial reperfusion injury. Cardiovasc. Res. 2004,
61, 481−497.
(4) Eefting, F.; Rensing, B.; Wigman, J.; Pannekoek, W. J.; Liu, W.
M.; Cramer, M. J.; Lips, D. J.; Doevandans. Role of apoptosis in
reperfusion injury. Cardiovasc. Res. 2004, 61, 414−426.
(5) Liao, Y.-H.; Xia, N.; Zhou, S.-F.; Tang, T.-T.; Yan, X.-X.; Lv, B.-J.;
Nie, S.-F.; Wang, J.; Iwakura, Y.; Xiao, H.; Yuan, J.; Jevallee, H.; Wei,
F.; Shi, G.-P.; Cheng, X. Interleukin-17A contributes to myocardial
ischemia/reperfusion injury by regulating cardiomyocyte apoptosis
and neutrophil infiltration. J. Am. Coll. Cardiol. 2012, 59, 420−429.
(6) Ruiz-Meana, M.; Garcia-Dorado, D.; Hofstaetter, B.; Piper, H.
M.; Soler-Soler, J. Propagation of cardiomyocyte hypercontracture by
passage of Na+ through gap junctions. Circ. Res. 1999, 85, 280−287.
(7) Hoffman, J. W.; Gilbert, T. B.; Poston, R. S.; Silldorff, E. P.
Myocardial reperfusion injury: etiology, mechanisms, and therapies. J.
Extra-Corpor. Technol. 2004, 36, 391−411.
(8) Gonon, A. T.; Jung, C.; Katz, A.; Westerblad, H.; Shemyakin, A.;
Sjoquist, O.; Lundberg, J. O.; Pernow, J. Local arginase inhibition
during early reperfusion mediates cardioprotection via increased nitric
oxide production. PLoS One 2012, 7, e42038.
(9) Rastaldo, R.; Pagliaro, P.; Cappello, P.; Penna, C.; Mancardi, D.;
Westerhof, N.; Losano, G. Nitric oxide and cardiac function. Life Sci.
2007, 81, 779−793.
(10) Shantz, L. M.; Giordano, E. Polyamine Metabolism and the
Hypertrophic Heart. In Polyamine Cell Signaling: Physiology,
Pharmacology, and Cancer Research; Wang, J-Y; Casero, R. A., Jr.,
Eds; Humana Press: Totowa, NJ, 2006; pp 123−137.
(11) Ash, D. E.; Scolnick, L. R.; Kanyo, Z. F.; Vockley, J. G.;
Cederbaum, S. D.; Christianson, D. W. Molecular basis of hyper-
argininemia: structure-function consequences of mutations in human
liver arginase. Mol. Genet. Metab. 1998, 64, 243−249.
(12) Santhanam, L.; Christianson, D. W.; Nyhan, D.; Berkowitz, D.
E. Arginase and vascular aging. J. Appl. Physiol. 2008, 105, 1632−1642.
(13) Ryoo, S.; Gupta, G.; Benjo, A.; Lim, H. K.; Camara, A.; Sikka,
G.; Sohi, J.; Santhanam, L.; Soucy, K.; Tuday, E.; Baraban, E.; Lies, M.;
Gerstenblith, G.; Nyhan, D.; Shoukas, A.; Christianson, D. W.; Alp, N.
J.; Champion, H. C.; Huso, D. Berkowitz. endothelial arginase II: a
novel target for treatment of atherosclerosis. Circ. Res. 2008, 102, 923−
932.
(14) Golebiowski, A.; Beckett, R. P.; Van Zandt, M.; Ji, M. K.;
Whitehouse, D.; Ryder, T.; Jagdmann, E.; Andreoli, M.; Mazur, A.;
Padmanilayam, M.; Cousido-Siah, A.; Mitschler, A.; Ruiz, F. X.;
Podjarny, A.; Schroeter, H. 2-Substituted-2-amino-6-boronohexanoic
acids as arginase inhibitors. Biorg. Med. Chem. Lett. 2013, in press.
(15) Dastlik, K. A.; Sundermeier, U.; Johns, D. M.; Chen, Y.;
Williams, R. M. An improved synthesis of optically pure 4-Boc-5,6-
diphenylmorpholin-2-one and 4-Cbz-5,6-diphenylmorpholin-2-one.
Synlett 2005, 693−696.
ASSOCIATED CONTENT
* Supporting Information
■
S
1H NMR and 13C NMR spectra, LC MS/MS data, analytical
HPLC chromatograms, chiral HPLC chromatograms, and X-
ray crystallography data collection and refinement statistics.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Phone: 203-494-1672. E-mail: mvzandt@snet.net.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Mars, Incorporated, for their helpful discussions and
generous financial support of this program. We also thank the
IGBMC Structural genomics platform staff, in particular Pierre
Poussin-Courmontagne and Alastair McEwen. We thank Drs.
Di Constanzo, Centeno, and Velazquez for the clone of
arginase I and Drs. Cederbaum, Deignan, and Gotoh for the
clone of arginase II. The crystallographic experiments were
performed on the X06SA and X06DA beamlines at the Swiss
Light Source, Paul Scherrer Institut, Villigen, Switzerland. In
particular, we thank Takashi Tomizaki, Vincent Olieric,
Anuschka Pauluhn, Christian Stirnimann, and Meitian Wang
for their help in the beamline. E.I.H. is member of the Carrera
del Investigador, Conicet, Argentina. This work was supported
by the Centre National de la Recherche Scientifique (CNRS),
the Institut National de la Sante
the Hopital Universitaire de Strasbourg (H.U.S.), and the
Universite de Strasbourg. Finally, we thank Dr. Francisco
́ ́
et de la Recherche Medicale,
̂
́
Villarreal at the University of California, San Diego, for his
efforts in conducting the MI/RI experiments.
ABBREVIATIONS USED
■
AAR, area at risk; ABH, 2-(S)-amino-6-boronohexanoic acid;
Arg, arginase; CHO, Chinese hamster ovary; DCM, dichloro-
methane; DMA, dimethylacetamide; DME, 1,2-dimethoxy-
ethane; DMF, dimethylformamide; DMSO, dimethylsulfoxide;
ELSD, evaporative light scattering detector; HMPA, hexame-
thylphosphoramide; IPC, intermittent preconditioning;
KHMDS, potassium hexamethyldisilazane; LiHMDS, lithium
hexamethyldisilazane; MeOH, methanol; MI/RI, myocardial
ischemia/reperfusion injury; NaHMDS, sodium hexamethyldi-
(16) Mora, A.; Rangel, M. D. A.; Fuentes, J. M.; Soler, G.; Centeno,
F. Implications of the S-shaped domain in the quaternary structure of
human arginase. Biochim. Biophys. Acta 2000, 1476, 181−190.
(17) Di Costanzo, L.; Sabio, G.; Mora, A.; Rodriguez, P. C.; Ochoa,
A. C.; Centeno, F.; Christianson, D. W. Crystal structure of human
arginase I at 1.29 Å resolution and exploration of inhibition in the
immune response. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 13058−
13063.
2579
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580

Journal of Medicinal Chemistry
Article
(18) Scolnick, L. R.; Kanyo, Z. F.; Cavalli, R. C.; Ash, D. E.;
Christianson, D. W. Altering the binuclear manganese cluster of
arginase diminishes thermostability and catalytic function. Biochemistry
1997, 36, 10558−10565.
(19) Colleluori, D. M.; Morris, S. M.; Ash, D. E. Expression,
purification, and characterization of human type II arginase. Arch.
Biochem. Biophys. 2001, 389, 135−143.
(20) Cama, E.; Colleluori, D. M.; Emig, F. A.; Shin, H.; Kim, S. W.;
Kim, N. N.; Traish, A. M.; Ash, D. E.; Christianson, D. W. Human
arginase II: crystal structure and physiological role in male and female
sexual arousal. Biochemistry 2003, 42, 8445−8451.
(21) Jung, D.; Biggs, H.; Erikson, J.; Ledyard, P. U. Clin. Chem. 1975,
21, 1136.
(22) Cama, E.; Pethe, S.; Boucher, J. L.; Han, S.; Emig, F. A.; Ash, D.
E.; Viola, R. E.; Mansuy, D.; Christianson, D. W. Inhibitor
coordination interactions in the binuclear manganese cluster of
arginase. Biochemistry 2004, 43, 8987−8999.
(23) Cox, J. D.; Cama, E.; Colleluori, D. M.; Pethe, S.; Boucher, J. L.;
Mansuy, D.; Ash, D. E.; Christianson, D. W. Mechanistic and
metabolic inferences from the binding of substrate analogues and
products to arginase. Biochemistry 2001, 40, 2689−2701.
(24) Shishova, E. Y.; Di Costanzo, L.; Emig, F. A.; Ash, D. E.;
Christianson, D. W. Probing the specificity determinants of amino acid
recognition by arginase. Biochemistry 2009, 48, 121−131.
(25) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. Methods in Enzymology, Volume 276:
Macromolecular Crystallography, Part A, Carter, C. W., Jr., Sweet, R. M.,
Eds.; Academic Press: New York, 1997; Vol. 276, pp 307-326.
(26) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.;
McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 235−242.
(27) Adams, P. D.; Afonine, P. V.; Bunkoczi, G.; Chen, V. B.; Davis,
I. W.; Echols, N.; Headd, J. J.; Hung, L. W.; Kapral, G. J.; Grosse-
Kunstleve, R. W.; McCoy, A. J.; Moriarty, N. W.; Oeffner, R.; Read, R.
J.; Richardson, D. C.; Richardson, J. S.; Terwilliger, T. C.; Zwart, P. H.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010, 66, 213−221.
(28) Emsley, P.; Lohkamp, B.; Scott, W.; Cowtan, K. Features and
Development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66, 486−501.
2580
dx.doi.org/10.1021/jm400014c | J. Med. Chem. 2013, 56, 2568−2580