Discovery of brain-penetrant, irreversible kynurenine aminotransferase ii inhibitors for schizophrenia

Article abstract of DOI:10.1021/ml200204m

Kynurenine aminotransferase (KAT) II has been identified as a potential new target for the treatment of cognitive impairment associated with schizophrenia and other psychiatric disorders. Following a high-throughput screen, cyclic hydroxamic acid PF-04859989 was identified as a potent and selective inhibitor of human and rat KAT II. An X-ray crystal structure and ¹³C NMR studies of PF-04859989 bound to KAT II have demonstrated that this compound forms a covalent adduct with the enzyme cofactor, pyridoxal phosphate (PLP), in the active site. In vivo pharmacokinetic and efficacy studies in rat show that PF-04859989 is a brain-penetrant, irreversible inhibitor and is capable of reducing brain kynurenic acid by 50% at a dose of 10 mg/kg (sc). Preliminary structurea€"activity relationship investigations have been completed and have identified the positions on this scaffold best suited to modification for further optimization of this novel series of KAT II inhibitors.

Full text of DOI:10.1021/ml200204m

Letter
pubs.acs.org/acsmedchemlett
Discovery of Brain-Penetrant, Irreversible Kynurenine
Aminotransferase II Inhibitors for Schizophrenia
Amy B. Dounay,* Marie Anderson, Bruce M. Bechle, Brian M. Campbell, Michelle M. Claffey,
Artem Evdokimov, Edelweiss Evrard, Kari R. Fonseca, Xinmin Gan, Somraj Ghosh, Matthew M. Hayward,
Weldon Horner, Ji-Young Kim, Laura A. McAllister, Jayvardhan Pandit, Vanessa Paradis, Vinod D. Parikh,
Matthew R. Reese, SuoBao Rong, Michelle A. Salafia, Katherine Schuyten, Christine A. Strick,
Jamison B. Tuttle, James Valentine, Hong Wang, Laura E. Zawadzke, and Patrick R. Verhoest
Pfizer Worldwide Research and Development, Neuroscience Chemistry, Eastern Point Road, Groton, Connecticut 06340,
United States
S
* Supporting Information
ABSTRACT: Kynurenine aminotransferase (KAT) II has been identified as a
potential new target for the treatment of cognitive impairment associated with
schizophrenia and other psychiatric disorders. Following a high-throughput screen,
cyclic hydroxamic acid PF-04859989 was identified as a potent and selective
inhibitor of human and rat KAT II. An X-ray crystal structure and 13C NMR
studies of PF-04859989 bound to KAT II have demonstrated that this compound
forms a covalent adduct with the enzyme cofactor, pyridoxal phosphate (PLP), in
the active site. In vivo pharmacokinetic and efficacy studies in rat show that
PF-04859989 is a brain-penetrant, irreversible inhibitor and is capable of reducing
brain kynurenic acid by 50% at a dose of 10 mg/kg (sc). Preliminary structure−
activity relationship investigations have been completed and have identified the positions on this scaffold best suited to modification
for further optimization of this novel series of KAT II inhibitors.
KEYWORDS: KAT II, kynurenic acid, kynurenine aminotransferase, hydroxamic acid
for schizophrenia and other diseases of the central nervous
system (CNS).
ynurenic acid (KYNA, 3) is a neuroactive metabolite
Kproduced in the kynurenine pathway of
L-tryptophan
The synthesis of KYNA in the brain is catalyzed primarily by
astrocytic kynurenine aminotransferases KAT I and KAT II.11
KAT II is currently viewed as the major determinant of KYNA
biosynthesis in rat and human brain. An X-ray structure of the
homodimeric human KAT II complex bound to its endogenous
cofactor, pyridoxal-5′-phosphate (PLP), has been reported, and
unique structural features of this enzyme suggest the possibility
of structure-based design of isozyme-specific inhibitors.12 The
recent X-ray structure of human KAT II bound to the selective
inhibitor BFF-122 further confirms that structure-based design
strategies may be useful for the design and optimization of
selective KAT II inhibitors.13
Several recent studies have employed the selective tool com-
pounds (S)-ESBA (4) and BFF-122 (5, Figure 2) via central
administration for study of KAT II inhibition in preclinical
models.14−18 However, identification of brain-penetrant KAT II
inhibitors with druglike properties remains a challenge. Addi-
tionally, recent reports have established that sequence variants in
human and rat KAT II cause significant cross-species enzyme
potency shifts for inhibitor 4.19,20 Our primary objective was to
degradation (Figure 1). Interest in the pharmacological
Figure 1. KYNA biosynthesis.
modulation of this pathway has been prompted by reports that
KYNA is an inhibitor of α7 nicotinic acetylcholine receptor
function1 and an antagonist at the glycine site of the N-methyl-D-
aspartate (NMDA) receptor.2 KYNA has also recently been
identified as a potent agonist at the aryl hydrocarbon receptor
(AHR).3 A growing body of preclinical evidence suggests that
KYNA is directly involved in the pathophysiology of schizo-
phrenia,4 among other psychiatric and neurological disorders.5,6
Elevated KYNA levels have been observed in the cerebrospinal
fluid7,8 and postmortem prefrontal cortex9 of schizophrenia
patients. Studies have also shown an elevation of KYNA in
the cerebrospinal fluid of patients with bipolar disorder.10
Thus, the reduction of KYNA levels in the brain via modulation
of the kynurenine pathway may provide a novel treatment
Received: August 18, 2011
Accepted: January 16, 2012
Published: January 25, 2012
© 2012 American Chemical Society
187
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192

ACS Medicinal Chemistry Letters
Letter
suggests that compounds of this type could serve as useful tools in
preclinical rodent models.
An X-ray crystal structure (3.2 Å) of 7 bound to human KAT
II reveals several key interactions that account for the remarkable
potency and selectivity of this inhibitor (Figure 4).30 First, the
Figure 2. Selective KAT II inhibitors.
identify a selective, brain-penetrant tool compound for further
exploration of KAT II inhibition in preclinical models of
schizophrenia and cognition. Identification of a series of KAT II
inhibitors with submicromolar binding affinity at both the rat and
the human isoforms of KAT II would facilitate preclinical studies
and clinical candidate identification. To enable rapid series
optimization toward a clinical therapeutic agent, our criteria for
novel lead compounds included high ligand efficiency21 (LE >
0.35) and excellent drug-like physicochemical properties (MW,
log P, polar surface area).22−24
In an effort to identify novel KAT II inhibitors, a high-
throughput screen of the Pfizer compound file was conducted.
Aminodihydroquinolone 6 (Figure 3) initially emerged as an
Figure 4. X-ray structure of KAT II-PLP adduct with 7 (carbon
scaffold from 7 in orange and PLP in magenta). A cutaway section of
the molecular surface forming the catalytic site of KAT II is shown in
pale blue. The substrate-binding pocket is formed by residues from
both subunits of the KAT II dimer, one colored in green and the other
in orange. Hydrogen bond interactions between the ligand and the
protein are indicated as dotted lines, with distances indicated in Å.
X-ray structure indicates that 7 is covalently attached to PLP in
the active site of KAT II, via a linkage between the primary
amine of 7 and the aldehyde functionality of PLP. At the resolu-
tion of this structure, the two most likely adduct structures, a
Schiff base I and enamine II (Scheme 1), are indistinguishable,
but additional NMR studies have confirmed the enamine
structure (vide infra). A series of dialysis and kinetic studies
were carried out to further characterize this adduct, confirming
that it is formed irreversibly.31 This covalent adduct bears a
number of similarities to the recently reported KAT II-bound
adduct formed between PLP and inhibitor BFF-122 (5).13 As in
the case of 5, the formation of a covalent adduct between 7 and
PLP is crucial to the observed potency of this inhibitor. The PLP
portion of the adduct, although not covalently linked to KAT II,
maintains numerous H-bond donor−acceptor interactions with
the enzyme, as detailed in the previously reported human KAT
II-PLP structure.12 For example, the nitrogen atom of the
pyridine ring of PLP interacts with the highly conserved Asp-230
residue, whose side chain is held in place by a hydrogen bond to
the OH group of Tyr-195. Likewise, the PLP O3′ atom forms
hydrogen bonds with Tyr-233 and Asn-202. The PLP phos-
phate group participates in a network of hydrogen-bonding
interactions, including contacts with Ser-262, Arg-270, Ser-117,
Gln-118, and Ser-260 of one subunit and Tyr-74 of the other
Figure 3. Properties of HTS hit PF-04859989 (7).
interesting hit, but follow-up testing of an authentic sample of 6
demonstrated that this compound had no significant inhibitory
activity toward hKAT II. Chromatographic separation,
characterization, resynthesis, and screening of minor impurities
from the original HTS sample batch led to the identification
of (3S)-3-amino-1-hydroxy-3,4-dihydroquinolin-2(1H)-one
(PF-04859989, 7) as the active constituent (hKAT II IC50
=
23 nM) from the HTS sample. Because of its high potency and
low molecular weight (178 Da), compound 7 is an attractive
hit from the standpoint of ligand efficiency (LE = 0.82).
Furthermore, other physicochemical properties of 7 (clog
P = −0.1, tPSA25 = 66.6, and CNS MPO score26 = 5.2),
combined with its clean profile in the standard CEREP panel,27
position 7 in excellent chemical space for a lead compound
targeting the CNS.28 Compound 7 is also selective for KAT II
over human KAT I, KAT III, and KAT IV (IC50 values of 22, 11,
and >50 μM, respectively).29 Finally, the modest potency shift
(∼10×) between human and rat KAT II (rKAT II IC50 = 263 nM)
Scheme 1. Postulated Imine−Enamine Isomerization of the PF-04859989-PLP Adduct
188
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192

ACS Medicinal Chemistry Letters
Letter
subunit. The other two elements of the functional triad of the
3-aminohydroxamic acid moiety of 7 also contribute key binding
interactions: the carbonyl oxygen of 7 serves as an H-bond
acceptor in an interaction with the side chain of Asn-202, and the
other oxygen atom in the hydroxamic acid functionality of 7
accepts a hydrogen bond from Arg-399. The fused phenyl group
of 7 contributes to van der Waals interactions with the
hydrophobic pocket formed by Leu-40, Tyr-74, and Leu-293.
To definitively characterize the KAT II-PLP adduct observed
in the X-ray structure, a series of protein NMR experiments
were designed to distinguish between the Schiff base I and the
enamine adduct II, the two most likely covalent inhibitor
structures (Scheme 1). The detailed studies by Silverman and
co-workers on cycloserine inhibition of γ-amino butyric acid
aminotransferase (GABA-AT), another PLP-dependent enzyme,
demonstrated that initial Schiff base formation between
cycloserine and PLP in the enzyme active site is followed by a
series of proton transfer events, which ultimately lead to
aromatization of the cycloserine adduct.32,33 By analogy, we
anticipated that the initially formed Schiff base adduct between
7 and PLP in the KAT II active site may undergo a series of
proton transfers, leading to the formation of a more stable
isomeric α,β-unsaturated (aromatized) product. The incorpo-
ration of a 13C label into our inhibitor, affording the isotopic
analogue 7a, provided a dramatic signal enhancement of the
labeled carbon atom in the 13C NMR spectra. Predicted 13C
chemical shifts for the labeled carbon in 7a were significantly
different for the Schiff base I (62 ppm) as compared to the α,β-
unsaturated adduct II (139 ppm).34 NMR experiments
examining 7a in a buffer solution (not shown) and in a solution
containing KAT II without PLP (Figure S1A in the Supporting
Information)35 revealed a distinctive 13C signal at 50 ppm. Upon
addition of PLP to the KAT II inhibitor solution (Figure S1B in
the Supporting Information), the signal at 50 ppm disappeared
and a new signal appeared at 137 ppm, consistent with the
formation of the α,β-unsaturated adduct II.36
Guided by the X-ray structure of 7 bound to KAT II, we
sought to build an understanding of the structural requirements
for potent and selective inhibition of KAT II within this series of
irreversible inhibitors. Although irreversible inhibition was not
one of our lead criteria at the outset of the program, maintaining
this attribute of 7 was a high priority through our optimiza-
tion efforts. The potential advantages of irreversible inhibitors
include low dose requirements and reduced off-target
toxicity.37,38 Furthermore, this irreversible inhibitor does not
covalently bind to the protein, which should significantly reduce
the potential for an immune response. To fully explore the
structure−activity relationship (SAR) around the PF-04859989
scaffold, several synthetic approaches were developed (see
Scheme 2); detailed descriptions of these methods have recently
been disclosed.39,40 In general, the synthesis of 7 and its
analogues requires access to α-amino acids or α-amino esters,
which undergo reductive cyclization to afford 3-amino-1-
hydroxy-3,4-dihydroquinolin-2(1H)-ones. In the case of 7 itself,
direct catalytic reductive cyclization using Pt/C and H2
(method A) provided clean conversion to the desired product
without competitive over-reduction to the corresponding
lactam. However, for substituted analogues, alternate cyclization
conditions were generally required to minimize over-reduction
of the nitro functionality. Cyclization with SnCl2 in the presence
of NaOAc (method C) generally provided 5-, 6-, or 7-
substituted 3-amino-1-hydroxy-3,4-dihydroquinolin-2(1H)-ones
in good yield. However, efficient conversion to 8-substituted
Scheme 2. Synthesis of Hydroxamic Acids 7−27
Reagents and conditions: Method A: (i) Pt/C, pyridine, H2. (ii) HCl/
MeOH. Method B: Pt/C, MeOH, HCl, H2. Method C: (i) SnCl2,
NaOAc, THF. (ii) Boc2O, Et3N. (iii) HCl. Method D: (i) Boc2O,
Et3N. (ii) CF3CH2OTf. (iii) Pt/C, pyridine, H2. (iv) HCl. Method E:
(i) phenyliodine(III) bis(trifluoroacetate) (PIFA), CH2Cl2. (ii) BF3
diethyl etherate, THF.
analogues (e.g., 13−16) required activation of the carboxylic
acid to the trifluoroethyl ester prior to cyclization (method D).
An oxidative cyclization approach using phenyliodine(III)
bis(trifluoroacetate) (PIFA, method E) also provided ready
access to select analogues in the series (e.g., 9 and 11).
Our preliminary SAR studies on the 3-amino-1-hydroxy-3,4-
dihydroquinolin-2(1H)-one template explored the role of stereo-
chemistry of the 3-amino group and the effects of substituents at the
1-, 3-, and 4-positions. Comparison of 7 [(3S)-amino-1-hydroxy-3,4-
dihydroquinolin-2(1H)-one] with its enantiomer 8 demonstrated
that the (S)-enantiomer is ∼10-fold more potent, with similar shifts
against both human and rat enzymes. This enantiopre-
ference in inhibition of KAT II correlates with the stereochemistry
of the enzyme's substrate, (L)-kynurenine (2) [(S)-2-amino-4-(2-
aminophenyl)-4-oxobutanoic acid]. Methyl hydroxamate 9 displayed
a complete loss of activity at KAT II, indicating that a free
hydroxamic acid group is crucial in this series. Likewise, introduction
of a methyl group at the C3-position (compound 10) resulted in
complete loss of activity. This finding may be explained by the
possibility of unfavorable steric interactions between this methyl
group and Leu 293 and/or Tyr 142 in this region of the binding
pocket. Furthermore, this SAR observation is consistent with our
NMR studies of 7a, which demonstrated that a hydrogen substituent
at C3 is mechanistically required for potent, irreversible inhibition.
The addition of a methyl group at the C4 position (compound 11)
also led to significant loss of activity. In brief, the preliminary SAR
work around positions 1−4 of PF-04859989 demonstrate that the
(S)-configuration is preferred and that even minor modifications
within this domain lead to dramatic loss of KAT II potency. These
SAR findings are consistent with our X-ray structure and mechanistic
studies, which suggest that each moiety in this aminohydroxamic acid
functional group triad forms a key interaction with KAT II.
Furthermore, introduction of any adjacent substituents that interfere
with the formation of the inhibitor−PLP enamine adduct in the
KAT II active site also leads to significant potency loss.
Next, the effects of introducing substituents at positions 5
through 8 on the 3-amino-1-hydroxy-3,4-dihydroquinolin-
2(1H)-one template were explored (Table 2). The relative
potency of analogues was initially assessed by evaluating IC50
values from our human and rat KAT II enzyme assays. These
studies demonstrated that only small R1 substituents (R1 = F,
compound 12) are tolerated by human KAT II. Significant loss
of potency is observed with even slightly larger substituents at
189
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192

ACS Medicinal Chemistry Letters
Letter
substitution at this position, with complete loss of activity
observed for compounds 13−16. The evaluation of R4
substituents likewise demonstrated that only small groups
(R4 = F, compound 23) are tolerated at this position. All
compounds with larger R4 substituents (compounds 24−27) are
significantly less potent than 7. This potency loss is likely due to
unfavorable steric interactions between the R4 substituents and
the Tyr74 and Leu293 in hKAT II. These larger R4 substituents,
like the R1 substituents, also bring about complete loss of rat
KAT II potency. Evaluation of R2 and R3 substituents revealed
that both the human and the rat KAT II enzymes are less
sensitive to small changes at these positions. The addition of
small R2 (Cl, Me, OMe) or R3 (Cl, Me, CF3) groups41 to 7 does
not appear to dramatically affect the human or rat KAT II
potency: human and rat KAT II IC50 values for analogues
17−22 are within 2-fold of the respective values for 7. Our
assessment of steric and electronic effects of R1−R4 in this series
(e.g., MeO vs Cl, CF3) suggests that electronic effects do not
drive the SAR. These preliminary SAR data suggest that the R2
and R3 positions hold the most promise for further SAR work to
optimize properties and potency for this series.
Table 1. Human and Rat KAT II Potency for Analogues of 7
To ensure a rigorous evaluation of SAR within this series of
irreversible inhibitors, potency was also assessed in terms of the
second order rate constant kinact/Ki (Table 2). Unlike IC50
values, kinact/Ki values are independent of substrate concen-
tration and preincubation time and are considered the best
measure of potency for irreversible inhibitors.42 In the case of 7
and its analogues 12−27, the IC50 potency trends are paralleled
by the kinact/Ki values. The kinact/Ki values indicate a subtle
differentiation of compounds 17−19 from 7, suggesting that
small R2 substituents may provide a boost in potency. This trend
was not observed in the IC50 SAR, perhaps due in part to the
relatively high enzyme concentration (30 nM) employed in that
assay format.
a
Values represent the geometric mean of at least three independent
b
c
determinations. n = 1. n = 2.
this position (compounds 13−16). Interestingly, the rat KAT II
enzyme is even more sensitive than the human enzyme to
Table 2. Human and Rat KAT II Potency for Analogues of 7
a
b
IC50 (nM)
kinact/Ki (M−1 s−1)
compd
R1
R2
R3
R4
hKAT II
rKAT II
hKAT II
rKAT II
7
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
H
H
H
H
23
40
263
631
18 500
5310
573
57
F
H
H
H
OMe
Cl
Me
CF3
H
H
H
H
572
252
>10 000
>10 000
21.3
42.0
H
H
H
c
c
H
H
H
1050
>10 000
H
H
H
174
29
>6810
118
541
Cl
Me
OMe
H
H
H
28 300
33 400
31 700
19 000
11 900
15 300
6680
1170
590
H
H
H
37
368
H
H
H
22
137
1840
465
H
Cl
Me
CF3
H
H
36
258
H
H
H
30
402
167
H
H
H
29
488
102
H
H
F
45
2060
>7970
>10 000
>4920
>10 000
H
H
H
Cl
Me
OMe
349
319
179
>10 000
9.5
H
H
H
H
H
H
1940
H
H
H
CF3
a
b
Values represent the geometric mean of at least three experiments. Values represent the arithmetic mean of at least three experiments. Additional
c
statistical details are provided in the Supporting Information. n = 2.
190
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192

ACS Medicinal Chemistry Letters
Letter
Compound 7 was selected as a protypical tool compound for
evaluation of the in vivo characteristics of this series of KAT II
inhibitors. In vivo pharmacokinetic experiments were carried
out to evaluate the utility of 7 as a systemically dosed tool
compound. Compound 7 displayed excellent CNS exposure in
rat after subcutaneous administration (Table 3). This
AUTHOR INFORMATION
Corresponding Author
*Tel: (860)-686-9208. E-mail: amy.dounay@pfizer.com.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Table 3. Rat Fractions Unbound and Plasma, Brain, and CSF
Exposure for 7
We gratefully acknowledge Larry James for KAT III screening,
Kevin Parris for help with the final x-ray refinement, and
Katherine Brighty for careful editing of this manuscript.
a
fraction unbound
concn (nM)
brain
39.3
plasma
89.3
free plasma
free brain
CSF
REFERENCES
10200 403
3760 776
4060 969
■
a
(1) Hilmas, C.; Pereira, E. F.; Alkondon, M.; Rassoulpour, A.;
Schwarcz, R.; Albuquerque, E. X. The Brain Metabolite Kynurenic
Acid Inhibits α7 Nicotinic Receptor Activity and Increases Non-α7
Nicotinic Receptor Expression: Physiopathological Implications.
J. Neurosci. 2001, 21, 7463−7473.
(2) Kessler, M.; Terramani, T.; Lynch, G.; Baudry, M. A Glycine Site
Associated with N-Methyl-D-Aspartic Acid Receptors: Character-
ization and Identification of a New Class of Antagonists. J. Neurochem.
1989, 52, 1319−1328.
(3) Natale, B. C.; Murray, I. A.; Schroeder, J. C.; Flaveny, C. A.;
Lahoti, T. S.; Laurenzana, E. M.; Omiecinski, C. J.; Perdew, G. H.
Kynurenic Acid Is a Potent Endogenous Aryl Hydrocarbon Receptor
Ligand that Synergistically Induces Interleukin-6 in the Presence of
Inflammatory Signaling. Toxicol. Sci. 2010, 115, 89−97.
(4) Erhardt, S.; Schwieler, L.; Nilsson, L.; Linderholm, K.; Engberg,
G. The Kynurenic Acid Hypothesis of Schizophrenia. Physiol. Behav.
2007, 92, 203−209.
(5) Erhardt, S.; Olsson, S. K.; Engberg, G. Pharmacological
Manipulation of Kynurenic Acid: Potential in the Treatment of
Psychiatric Disorders. CNS Drugs 2009, 23, 91−101.
Drug concentrations measured at 0.25 h time point, 10 mg/kg sc dose.
compound displays relatively high free fraction in rat brain
and plasma protein binding assays and is brain-penetrant, with a
free plasma:free brain:CSF exposure ratio of 2.7:1:1.1.
In vivo efficacy of 7 was confirmed by measuring KYNA in
dialysates from prefrontal cortex in freely moving rats. At the
drug exposures described in Table 3 (dose, 10 mg/kg sc),
KYNA concentrations decreased to approximately 50% of basal
values (Figure 5).43 This decrease was achieved approximately
(6) Potter, M. C.; Elmer, G. I.; Bergeron, R.; Albuquerque, E.;
Guidetti, P.; Wu, H.-Q.; Schwarcz, R. Reduction of Endogenous
Kynurenic Acid Formation Enhances Extracellular Glutamate, Hippo-
campal Plasticity, and Cognitive Behavior. Neuropsychopharmacology
2010, 35, 1734−1742.
(7) Erhardt, S.; Blennow, K.; Nordin, C.; Skogh, E.; Lindstrom, L. H.;
Engberg, G. Kynurenic Acid Levels Are Elevated in the Cerebrospinal
Fluid of Patients with Schizophrenia. Neurosci. Lett. 2001, 313, 96−98.
(8) Nilsson, L. K.; Linderholm, K. R.; Engberg, G.; Paulson, L.;
Blennow, K.; Lindstrom, L. H.; Nordin, C.; Karanti, A.; Persson, P.;
Erhardt, S. Elevated Levels of Kynurenic Acid in the Cerebrospinal
Fluid of Male Patients with Schizophrenia. Schizophr. Res. 2005, 80,
315−322.
(9) Schwarcz, R.; Rassoulpour, A.; Wu, H.-Q.; Medoff, D.;
Tamminga, C. A.; Roberts, R. C. Increased Cortical Kynurenate
Content in Schizophrenia. Biol. Psychiatry 2001, 50, 521−530.
(10) Olsson, S. K.; Samuelsson, M.; Saetre, P.; Lindstrom, L.;
Jonsson, E. G.; Nordin, C.; Engberg, G.; Erhardt, S.; Landen, M.
Elevated Levels of Kynurenic Acid in the Cerebrospinal Fluid of
Patients with Bipolar Disorder. J. Psychiatry Neurosci. 2010, 35,
195−199.
(11) For a recent review, see Han, Q.; Cai, T.; Tagle, D. A.; Li, J.
Structure, Expression, and Function of Kynurenine Aminotransferases
in Human and Rodent Brains. Cell. Mol. Life Sci. 2010, 67, 353−368.
(12) Rossi, F.; Garavaglia, S.; Montalbano, V.; Walsh, M. A.; Rizzi, M.
Crystal Structure of Human Kynurenine Aminotransferase II, a Drug
Target for the Treatment of Schizophrenia. J. Biol. Chem. 2008, 283,
3559−3566.
Figure 5. PF-04859989 (10 mg/kg, sc) was dosed at time = 0. The
KYNA concentration was monitored over time in rat prefrontal cortex
using microdialysis, as described in the Supporting Information. Data,
expressed as percent of basal KYNA of each individual, are plotted as
means SEMs, n = 11−14 animals per group.
1 h after dosing, and KYNA concentrations returned to baseline
levels approximately 20 h postdose. This prolonged duration of
action is attributable to the irreversible inhibition of KAT II by 7.
In conclusion, 7 is a potent, brain-penetrant KAT II inhibitor
that exhibits all of the necessary attributes to serve as a
pharmacological tool for building confidence in the biological
rationale of this target. As irreversible inhibitors of KAT II,
compounds in this series may have unique advantages as
potential clinical candidates. Preliminary SAR investigations
have identified that the R2 and R3 positions on this scaffold are
ideal for additional modifications to optimize properties and
potency for this series. Future publications will detail the
further elaboration of this series and advancement of our
understanding of structural−functional relationships for this
unique series of irreversible inhibitors of KAT II.
(13) Rossi, F.; Valentina, C.; Garavaglia, S.; Sathyasaikumar, K. V.;
Schwarz, R.; Kojima, S.; Okuwaki, K.; Ono, S.; Kajii, Y.; Rizzi, M.
Crystal Structure-Based Selective Targeting of the Pyridoxal 5′-
Phosphate Dependent Enzyme Kynurenine Aminotransferase II for
Cognitive Enhancement. J. Med. Chem. 2010, 53, 5684−5689.
(14) Pellicciari, R.; Rizzo, R. C.; Costantino, G.; Marinozzi, M.;
Amori, L.; Guidetti, P.; Wu, H.-Q.; Schwarcz, R. Modulators of the
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures for the synthesis of 7−27, assay
protocols, and CEREP screening data for 7. This material is
available free of charge via the Internet at http://pubs.acs.org.
191
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192

ACS Medicinal Chemistry Letters
Letter
Kynurenine Pathway of Tryptophan Metabolism: Synthesis and
Preliminary Biological Evaluation of (S)-4-(Ethylsulfonyl)-benzoylala-
nine, a Potent and Selective Kynurenine Aminotransferase II (KAT II)
Inhibitor. ChemMedChem 2006, 1, 528−531.
(15) Amori, L.; Wu, H.-Q.; Marinozzi, M.; Pellicciari, R.; Guidetti, P.;
Schwarcz, R. Specific Inhibition of Kynurenate Synthesis Enhances
Extracellular Dopamine Levels in the Rodent Striatum. Neuroscience
2009, 156, 196−203.
(16) Amori, L.; Guidetti, P.; Pellicciari, R.; Kajii, Y.; Schwarcz, R. On
the Relationship Between the Two Branches of the Kynurenine
Pathway in the Rat Brain In Vivo. J. Neurochem. 2009, 109, 316−325.
(17) Wu, H.-Q.; Pereira, E. F. R.; Bruno, J. P.; Pellicciari, R.;
Albuquerque, E. X.; Schwarcz, R. The Astrocyte-Derived α7 Nicotinic
Receptor Antagonist Kynurenic Acid Controls Extracellular Glutamate
Levels in the Prefrontal Cortex. J. Mol. Neurosci. 2010, 40, 204−210.
(18) Schwarcz, R.; Kajii, Y.; Ono, S. Preparation of amino-
piperazinylquinolonecarboxylates as kynurenine-amino-transferase in-
hibitors. WO 2009064836.
(19) Pellicciari, R.; Venturoni, F.; Bellocchi, D.; Carotti, A.;
Marinozzi, M.; Macchiarulo, A.; Amori, L.; Schwarcz, R. Sequence
Variants in Kynurenine Aminotransferase II (KAT II) Orthologs
Determine Different Potencies of the Inhibitor (S)-ESBA. Chem-
MedChem 2008, 3, 1199−1202.
(33) Cycloserine does not inhibit human or rat KAT II (IC50 > 10
000 nM in our assays).
(34) ACD Labs 9.0.
(35) Figure S1 is available in the Supporting Information.
(36) Control experiments have demonstrated that the nonenzymatic
reaction between 7 and PLP does not proceed to an appreciable extent
under these conditions.
(37) Johnson, D. S.; Weerapana, E.; Cravatt, B. F. Strategies for
discovering and derisking covalent, irreversible enzyme inhibitors.
Future Med. Chem. 2010, 2, 949−964.
(38) Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. The Resurgence
of Covalent Drugs. Nat. Rev. Drug Discovery 2011, 10, 307−317.
(39) Claffey, M. M.; Dounay, A. B.; Gan, X.; Hayward, M. M.; Rong,
S.; Tuttle, J. B.; Verhoest, P. R. Bicyclic and Tricyclic Compounds as
KAT II Inhibitors. WO2010146488.
(40) McAllister, L. A.; Bechle, B. M.; Dounay, A. B.; Evrard, E.; Gan,
X.; Ghosh, S.; Kim, J.-Y.; Parikh, V. D.; Tuttle, J. B.; Verhoest, P. R. A
General Strategy for the Synthesis of Cyclic N-Aryl Hydroxamic Acids
via Partial Nitro Group Reduction. J. Org. Chem. 2011, 76, 3484−
3497.
(41) Methoxy substitution was not evaluated at the R3 position due
to chemical instability of this compound and analogous alkoxy
analogues.
(42) Copeland, R. A. Evaluation of Enzyme Inhibitors in Drug
Discovery: A Guide for Medicinal Chemists and Pharmacologists; Wiley:
Hoboken, NJ, 2005; pp 1−265.
(43) KYNA concentrations can be maximally decreased by
approximately 80% from basal values at a dose of 32 mg/kg, sc. The
remaining 20% KYNA is likely biosynthesized by KAT I.
(20) Casazza, V.; Rossi, F.; Rizzi, M. Biochemical and structural
investigations on kynurenine aminotransferase II: an example of
conformation-driven species-specific inhibition? Curr. Top. Med. Chem.
2011, 11, 148−157.
(21) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand Efficiency: A
Useful Metric for Lead Selection. Drug Discovery Today 2004, 9,
430−431.
(22) Keseru, G. M.; Makara, G. M. The Influence of Lead Discovery
̈
Strategies on the Properties of Drug Candidates. Nat. Rev. Drug
Discovery 2009, 8, 203−212.
(23) Price, D. A.; Blagg, J.; Jones, L.; Greene, N.; Wager, T.
Physicochemical Drug Properties Associated with In Vivo Toxico-
logical Outcomes: A Review. Expert Opin. Drug Metab. Toxicol. 2009,
5, 921−931.
(24) Edwards, M. P.; Price, D. A. Role of physicochemical properties
and ligand lipophilicity efficiency in addressing drug safety risks. Annu.
Rep. Med. Chem. 2010, 45, 381−391.
(25) Ertl, P.; Rohde, B.; Selzer, P. Fast calculation of molecular polar
surface area as a sum of fragment-based contributions and its
applications to the prediction of drug transport properties. J. Med.
Chem. 2000, 43, 3714−3717.
(26) Wager, T. T.; Hou, X.; Verhoest, P. R.; Villalobos, A. Moving
beyond Rules: The Development of a Central Nervous System
Multiparameter Optimization (CNS MPO) Approach To Enable
Alignment of Druglike Properties. ACS Chem. Neurosci. 2010, 1, 435−
449.
(27) For details on the CEREP screening panel, see www.cerep.fr.
The PF-04859989 screening data set is provided in the Supporting
Information.
(28) Wager, T. T.; Chandrasekaran, R. Y.; Hou, X.; Troutman, M.
D.; Verhoest, P. R.; Villalobos, A.; Will, Y. Defining Desirable Central
Nervous System Drug Space through the Alignment of Molecular
Properties, in Vitro ADME, and Safety Attributes. ACS Chem. Neurosci.
2010, 1, 420−434.
(29) Compound 7 has also been screened against the following list of
PLP-dependent enzymes: human alanine aminotransferase (ALT)
(IC50 = 96 nM), rat γ-aminobutyric acid (GABA) transaminase (2%
inhibition at 10 μM), and human D-amino acid oxidase (DAAO)
(IC50 > 50 μM).
(30) The coordinates have been deposited in the Protein Data Bank,
with ID 3UE8.
(31) Anderson, M. Personal communication, Pfizer.
(32) Olson, G. T.; Fu, M.; Lau, S.; Rinehart, K. L.; Silverman, R. B.
An Aromatization Mechanism of Inactivation of Gamma-Aminobutyric
Acid Aminotransferase for the Antibiotic L-Cycloserine. J. Am. Chem.
Soc. 1998, 120, 2256−2267.
192
dx.doi.org/10.1021/ml200204m | ACS Med. Chem. Lett. 2012, 3, 187−192