6-Amino-4-(pyrimidin-4-yl)pyridones: Novel glycogen synthase kinase-3β inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.01.017

The synthesis and structure-activity relationships for a novel series of 6-amino-4-(pyrimidin-4-yl)pyridones derived from a high throughput screening hit are discussed. Optimization of lead matter afforded compounds with good potency, selectivity and central nervous system (CNS) exposure.

Full text of DOI:10.1016/j.bmcl.2011.01.017

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1429–1433
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
6-Amino-4-(pyrimidin-4-yl)pyridones: Novel glycogen synthase
kinase-3b inhibitors
⇑
Karen Coffman , Michael Brodney, James Cook, Lorraine Lanyon, Jayvardhan Pandit, Subas Sakya,
Joel Schachter, Elaine Tseng-Lovering, Matthew Wessel
Pfizer Global Research and Development, Groton Laboratories, Eastern Point Road, Groton, CT 06340, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and structure–activity relationships for a novel series of 6-amino-4-(pyrimidin-4-yl)pyri-
dones derived from a high throughput screening hit are discussed. Optimization of lead matter afforded
compounds with good potency, selectivity and central nervous system (CNS) exposure.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 9 November 2010
Revised 4 January 2011
Accepted 6 January 2011
Available online 11 January 2011
Keywords:
Glycogen synthase kinase-3b inhibitors,
Alzheimer’s disease
Glycogen synthase kinase-3 (GSK-3) is a serine and threonine
kinase that regulates a diverse group of physiological functions
ranging from differentiation and development, to metabolism, cell
cycle regulation, and neuroprotection.1 A considerable body of lit-
erature implicates GSK-3 dysregulation in Alzheimer’s disease
(AD).2,3 A histopathological hallmark of AD, neurofibrillary tangles
(NFTs) are composed of a highly phosphorylated form of the tau
protein. Glycogen synthase kinase-3b (GSK-3b) has been shown
to phosphorylate tau at AD relevant epitopes, implicating it as a
potential mediator of NFT formation and suggesting that a GSK-
3b inhibitor could represent a promising therapeutic approach
for the treatment of AD and other neurodegenerative disorders.4
Additionally, GSK-3 inhibition of the activity of glycogen synthase
and insulin receptor substrate-1, key targets in the insulin signal-
ing pathway, offers support to the hypothesis that an inhibitor
may be an effective treatment in type 2 diabetes and insulin resis-
tance.5 Evidence has also suggested GSK-3 inhibition as a potential
therapeutic approach for several other conditions, including bipo-
lar disorder, schizophrenia, Huntington’s disease, stroke, cardiac
ischemia, age-related loss of bone and muscle, chronic inflamma-
tory conditions, and as an adjunct to facilitate cancer chemother-
apy.6–9 Although, modulation of GSK-3 clearly has the potential
to be an interesting pharmaceutical target, the likelihood that inhi-
bition of GSK-3 may affect multiple systems raises concerns that
the range of its functions may be too broad to safely and selectively
modulate a particular disease state.
While it is widely appreciated that kinase dysregulation may
underlie a variety of pathologic conditions, kinase inhibitors have
traditionally represented challenging pharmaceutical targets due
to the difficulties involved in designing agents that are selective
for an individual kinase. Historically, the use of kinase inhibitors
has largely been limited to the field of oncology, where there is a
greater tolerance for off-target actions and cytotoxicity is a specific
objective of treatment. As more selective kinase inhibitors have
been identified, interest in their use in additional therapeutic areas
has grown. To explore their suitability as therapeutic agents for
CNS disorders, we sought to develop a series of potent, selective
and brain penetrant GSK-3b inhibitors.
High throughput screening of the Pfizer compound collection
identified compound 1 (Fig. 1) as a potent and ligand efficient
(LE = 0.52)10 GSK-3b inhibitor with an IC50 of 17.2 nM in a cell free
enzyme inhibition assay.11 A co-crystal structure of compound 1
with GSK-3b shows the compound bound in the active-site cleft
between the N- and C-lobes of the kinase (Fig. 2).12 The phenol
oxygen makes a pair of hydrogen bonds with a backbone amide
OH
N
N
GSK-3 CFA IC
17.2 nM
210 nM
27.0 nM
0.83
β
50
CLK1 IC50
CK2 IC50
MDR BA/AB
CLint
HO
24.7 mL/min/kg
Me
1
⇑
Corresponding author.
E-mail address: karen.j.coffman@pfizer.com (K. Coffman).
Figure 1. HTS hit, compound 1.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.017

1430
K. Coffman et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1429–1433
Figure 2. X-ray crystal structure structure of HTS hit, compound 1 bound to GSK-
3b.12 A co-crystal structure of compound 1 with GSK3-b shows the compound
bound in the active-site cleft between the N- and C-lobes of the kinase.
Figure 4. Overlap of X-ray crystal structure of HTS lead, compound 1 (in blue) with
proposed template 2 (in purple).
probability of CNS penetration as evidenced by favorable CNS Multi
Parameter Optimization (MPO) scores.18
and the carbonyl groups of residues Val135 and Asp133 in the
hinge region. Another hydrogen bond is formed between Lys85,
which is part of the catalytic triad, and the keto-oxygen on the pyr-
imidone ring. Although, its c log P is high (4.5), the lead compound
demonstrated reasonable human microsomal clearance in vitro13
(CLint 24.7 mL/min/kg). Despite bearing two hydrogen bond do-
nors, compound 1 was permeable, as demonstrated by an apical
to basal measurement in an MDCK MDR cell line (MDR AB
13.9 Â 10À6 cm/sec) and showed no P-gp mediated efflux liability
(MDR BA/AB 0.83).14 In order to evaluate kinase selectivity, com-
pound 1 was screened against a panel of 50 kinases.15 Notable
among the kinases screened, compound 1 inhibited CDC-like ki-
nase-1 (CLK1) and casein kinase 2 (CK2) with IC50’s of 210 nM
and 27 nM, respectively.
Structural modifications were sought to address two potential
concerns with lead compound 1: the metabolic liabilities associ-
ated with a phenol moiety16 and the number of hydrogen bond do-
nors in the HTS hit. Analysis of the kinase literature suggested that
4-substituted pyridines and pyrimidines might serve as viable phe-
nol replacements, interacting as single point hinge binders to the
active-site cleft.17 Furthermore, the X-ray structure of 1 showed
that the pyrimidone nitrogens were not forming direct hydrogen
bond interactions to the protein back-bone suggesting modifica-
tions would be tolerated in this region and could potentially lead
to improved kinase selectivity. Therefore, several replacements of
the pyrimidone ring that lacked additional hydrogen bond donors
or acceptors while maintaining the carbonyl group interaction
with Lys85 were proposed.
The synthesis of a diversity set of N-Me pyridones was accom-
plished using intermediate 10 (Scheme 1). Starting material
2,6-dihydroxyisonicotinic acid was converted to 2,6-dichloroisoni-
cotinic acid by treatment with POCl3. Displacement with NaOMe
afforded compound 5 and subsequent treatment with methyl
magnesium chloride yielded methyl ketone 6. Reaction with
Bredereck’s reagent19 afforded vinylogous ketone 7, which was
cyclo-condensed with formamidine acetate to generate 8. Treat-
ment with concd HCl followed by methylation with methyl iodide
afforded convergent intermediate 10. Synthesis of the final analogs
was accomplished by treatment of 10 with 1–5 equiv of the appro-
priate amine in the presence of 2–3 equiv of Hunig’s base at
temperatures ranging from 0 to 100 °C in DMF or NMP.20
Initially we investigated analogs where R = aryl. These com-
pounds exhibited reduced potency (see example 2, Table 1), per-
haps as a result of the increased distance between the hinge
OH
Cl
OMe
N
a
b
N
N
HO
HO
HO
OH
Cl
Cl
N
O
O
O
3
4
5
OMe
N
OMe
N
c
d
e
Docking studies of a series of N-Me pyridones, 2 (R = aryl) sug-
gested good overlap with compound 1 (Figs. 3 and 4) and con-
firmed the potential of the series as inhibitors of GSK-3b. The
carbonyl group of the pyridone ring maintains a reasonable hydro-
gen bonding distance (2.9 Å) to Lys85 and, in addition, the pyrim-
idine ring of 2 can accept a hydrogen bond from Val135. The R
group occupies the solvent exposed region of the kinase suggesting
that a range of substituents would likely be tolerated in the region.
The proposed series has properties associated with an increased
Cl
Cl
O
O
6
7
OH
OMe
O
N
N
f
g
N
N
N
N
N
Cl
Cl
Cl
N
N
8
9
10
O
N
h
OH
O
N
NR1R2
N
N
N
N
N
11
R
N
HO
Scheme 1. Reagents and conditions: (a) POCl3, (CH3)4 NCl, 130 °C, 48 h; (b) NaOMe/
MeOH, 80 °C, 48 h; (c) (1) MeMgCl/THF, À45 °C to 0 °C, 1 h, (2) methyl formate,
À25 °C, 30 min; (d) Bredereck’s reagent, 130 °C, 1 h; (e) K2CO3, formamidine
acetate/DMF, 100 °C, 5 h; (f) concd HCl, 80 °C, 4 h; (g) LiOt-Bu, MeI/acetone, 80 °C,
24 h; (h) HNR1R2, Hunig’s base/NMP, 0–150 °C, 1–48 h.
Me
1
2; R = aryl or NR1R2
Figure 3. Proposed template, compound 2.

K. Coffman et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1429–1433
1431
Table 1
Table 3
GSK-3b enzyme inhibition, human liver microsomal stability and CNS MPO scores
GSK-3b enzyme inhibition, human liver microsomal stability and P-gp efflux ratio
O
O
N
N
N
N
R
R
N
N
c
Compd
R
CFAa
IC50
(nM)
(LE)
WCAb
IC50
(nM)
CLint
(mL/
MDR
BA/
CNS
c
Compd
R
CFAa IC50
,
WCAb IC50
(nM)
,
CLint (mL/
min/kg)
CNS MPO
scored
MPO
(nM) (LE)
min/kg) ABd
scoree
1
2
HTS Hit
17.2 (.52)
>5280
24.7
4.7
>894
>10,000
<7.80
6.0
N
115
(.38)
21
22
4640
2320
149
1.23
1.35
5.8
5.8
12
13
50.1 (.55)
62.0 (.50)
2210
>910
15.2
32.0
6.0
6.0
N(Et)2
N
N
N
14
233 (.46)
>5040
<7.60
6.0
84.1
(.39)
61.4
O
MeO
N
GSK-3b cell free enzyme inhibition assay;11 IC50 values are the mean of at least
a
three experiments. LE = ligand efficiency.10
GSK-3b whole cell enzyme inhibition assay;21 IC50 values are the mean of at
b
OMe
51.0
(.37)
least three experiments.
23
24
3210
610
116
1.56
1.01
5.8
5.8
Human Liver Microsome Incubation assay.13
c
d
CNS Multi Parameter Optimization score.18 Provides a holistic assessment of a
compound’s attributes with respect to PK, safety and drug-likeness. Score >4.5
indicates an increased probability of CNS penetration.
N
OH
8.50
(.42)
<7.60
Table 2
N
N
GSK-3b enzyme inhibition, human liver microsomal stability and CNS MPO scores
O
O
6.10
(.39)
O
25
26
840
16.3
27.6
1.25
5.8
5.8
MeO
MeO
N
N
R
N
CFAa IC50
(nM) (LE) IC50, (nM) (mL/
WCAb
CLint
CNS
MPO
c
Compd
R
62.8
(.37)
1530
nt
min/kg) scored
H
N
a
15
16
>592
>10,000
800
nt
5.8
5.8
GSK-3bcell free enzyme inhibition assay;11 IC50 values are the mean of at least
three experiments. LE = ligand efficiency.10
GSK-3b whole cell enzyme inhibition assay;21 IC50 values are the mean of at
b
H
N
least three experiments.
17.4 (.44)
119
c
Human Liver Microsome Incubation assay.13
MDR-MDCK assay.14
d
MeO
CNS Multi Parameter Optimization score.18 Provides a holistic assessment of a
e
H
N
compound’s attributes with respect to PK, safety and drug-likeness. Score >4.5
indicates an increased probability of CNS penetration.
17
18
>401
>731
>9000
nt
5.8
5.8
OMe
H
N
binding nitrogen of the pyrimidine ring and Val135 (3.8 Å) as com-
pared to the 2.8 Å hinge interaction with the phenol moiety in the
original hit, compound 1. Additionally, the phenol in compound 1
makes two hydrogen bonds with the hinge, acting as both hydro-
gen bond acceptor and donor, whereas the pyrimidine in com-
pound 2 maintains only hydrogen bond acceptor capabilities.
Replacement of the aryl moiety on the new core with a variety of
simple amine substituents, however, afforded a group of analogs
which inhibit GSK-3b activity in the cell free enzyme assay while
maintaining good ligand efficiency (see Table 1, compounds 12–
14).
>10,000
60.7
OMe
N
H
MeO
19
20
>252
>260
>9000
nt
5.8
5.8
N
H
>10,000
211
MeO
Me
N
21
133 (.38)
1690
>304
6.0
MeO
A cohort of secondary amine substituted analogs (Table 2) was
prepared as part of a larger library of compounds. Preliminary SAR
revealed an ortho-methoxy substituent on an appended phenyl
ring affords at least a 30-fold increase in activity in the cell free en-
zyme inhibition assay compared to the meta or para methoxy
substituted analogs (see example 16 vs examples 17 and 18). Addi-
tionally, this set of compounds illustrated the importance of chain
a
GSK-3b cell free enzyme inhibition assay;11 IC50 values are the mean of at least
three experiments. LE = ligand efficiency.10
b
GSK-3b whole cell enzyme inhibition assay;21 IC50 values are the mean of at
least three experiments.
c
Human Liver Microsome Incubation assay.13
CNS Multi Parameter Optimization score.18 Provides a holistic assessment of a
d
compound’s attributes with respect to PK, safety and drug-likeness. Score >4.5
indicates an increased probability of CNS penetration.

1432
K. Coffman et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1429–1433
length to an appended phenyl group; a two carbon linker is opti-
mal for activity in the cell free assay (see example 16 vs examples
19 and 20). Example 21 confirmed that removal of the hydrogen
bond donor and formation of a tertiary amine is tolerated. CNS
MPO scores, a holistic assessment of a compound’s attributes with
respect to pharmacokinetics, safety and drug-likeness, were favor-
able (>5) for each analog indicating the compounds have an in-
creased probability of CNS penetration.18 Compound 16
demonstrated potent activity (17.4 nM) in the cell free assay. Fur-
ther in vitro evaluation showed no potential for P-gp liability, but a
CLint of 119 mL/min/kg in the in vitro human liver microsome assay
indicated that rapid clearance would be a liability for this analog.
The low clearance values for examples 12 and 13 suggested that
constraining example 16 into a piperidine or morpholine ring had
the potential to reduce clearance and maintain potency. In an at-
tempt to exploit this SAR, morpholine and piperidine analogs
substituted with an aryl group in the 2 or 3 position were gener-
ated according to the procedure in Scheme 1 (Table 3). The most
potent of this group, examples 24 and 25, maintained reasonable
ligand efficiency, exhibited improved in vitro human liver micro-
some intrinsic clearance values, demonstrated inhibition at
16, 24 and 25 were negative in this preliminary genetic toxicology
assay. These data, along with CK2 and CLK1 selectivity data are
summarized in Table 4.
In conclusion, a series of 6-amino-4-(pyrimidin-4-yl)pyridines
has been developed as novel glycogen synthase kinase-3b inhibi-
tors. A single point hinge binding motif for the series helped to pro-
vide improved selectivity over a range of kinases. Removal of
hydrogen bond donors from the core increased the probability of
CNS penetration and reduced the potential for metabolic liability.
Although none of the compounds demonstrates activity in the
whole cell enzyme inhibition assay of better than 500 nM, several
analogs exhibit good pharmacokinetic properties and offer
improvements in potency and selectivity over the original HTS lead
compound.
References and notes
1. Grimes, C. A.; Jope, R. S. Prog. Neurobiol. 2001, 65, 391.
2. Avila, J.; Hernandez, F. Expert Rev. Neurother. 2007, 7, 1527.
3. Hooper, C.; Killick, R.; Lovestone, S. J. Neurochem. 2008, 104, 1433.
4. Planel, E.; Sun, X.; Takashima, A. Drug Dev. Res. 2002, 56, 491.
5. Lee, J.; Kim, M.-S. Diabetes Res. Clin. Pract. 2007, 77S, S49.
6. Cohen, P.; Goedert, M. Nat. Rev. Drug Disc. 2004, 3, 479.
7. Meijer, L.; Flajolet, M.; Greengard, P. Trends Pharmacol. Sci. 2004, 25, 471.
8. Eldar-Finkelman, H. Trends Mol. Med. 2002, 8, 126.
<1 lM in a GSK-3b whole cell enzyme inhibition assay, and showed
no P-gp liability in an in vitro assay.
9. Ougolkov, A. V.; Billadeau, D. D. Future Oncol. 2006, 2, 91.
10. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
11. GSK-3b cell free enzyme assay:
Compounds 16, 24 and 25 were selected for further study which
included screening against a broad panel of kinases. Unlike the ori-
ginal hit, Compound 1, compounds 16 and 25 were devoid of activ-
ity at CK1 and CLK2, while compound 24, although inactive at CK1,
still inhibited CLK2.
Compounds 16, 24 and 25 were also submitted to an in vivo as-
say to assess CNS penetration.22 Compounds 24 and 25 demon-
strated good free brain to free plasma ratios in the in vivo assay,
however the study indicated a somewhat less favorable ratio for
compound 16.
Recombinant human GSK-3b was expressed in SF9/Baculo virus cells. His-tag
protein was purified by affinity chromatography to
column. Enzyme activity was assayed as the incorporation of [33P] from the
gamma phosphate of
33P]ATP (PerkinElmer) into biotinylated peptide
substrate bio-LC-S-R-H-S-S-P-H-Q-pS-E-D-E-E-E-OH (Anaspec). Reactions
were carried out in buffer containing 8 mM MOPS (pH 7.0), 10 mM
Magnesium acetate, 0.2 mM EDTA, 1 mM DTT and 2
M cold ATP. The 33P-
ATP was added for 0.025 Ci/well (120 L) and the final concentration of
substrate was 1.0 M. Enzyme was pre-incubated with test agent for 30 min at
a Ni-NTA Superflow
[
a
l
l
l
l
room temperature followed by initiation of the reaction by the addition of
substrate mix. Incubations were carried out at RT for 60 min. Reactions were
stopped by addition of 0.66 volume of buffer containing: 12.5 mM EDTA,
Additionally, the compounds of interest were screened in an
in vitro micronucleus assay for an early risk assessment of poten-
tial genetic toxicology hazards within the series.23 Compounds
0.25%Triton-X 100, 125
lM ATP, and 6.2 mg/ml streptavidin coated SPA beads
(Amersham) in PBS without Ca or Mg. Radioactivity associated with the beads
was quantified by scintillation counting of CPM in
(PerkinElmer).
a Trilux counter
12. PDB Id. 3Q3B. This structure was originally solved by Dr. Jay Bertrand and his
team at the Pharmacia Discovery Research Labs in Nerviano, Italy.
13. Assay method adapted from published protocols: (a) Riley, R. J.; McGinnity, D.
F.; Austin, R. P. Drug Metab. Dispos. 2005, 33, 1304; (b) Obach, R. S. Drug Metab.
Dispos. 1999, 27, 1350.
Table 4
GSK-3b enzyme activity, selectivity, human liver microsomal stability and in vivo
brain: plasma ratio for 16, 24 and 25
O
14. Assay method adapted from published protocol: Permeability is assessed as an
apical to basal measurement in the MDCK MDR cell line. Venkatakrishnan, K.;
Tseng, E.; Nelson, F.; Rollema, H.; French, J. L.; Horner, W.; Gibbs, M. A. Drug
Metab. Dispos. 2007, 35, 1341.
N
N
R
N
15. Card, A.; Caldwell, C.; Min, H.; Lokchander, B.; Xi, H.; Schiabola, S.; Kamath, A.
V.; Clugston, S. L.; Tschantz, W. R.; Wang, L.; Moshinsky, D. J. J. Biomol. Screen.
2009, 14, 31.
16. Miao, Z.; Obach, R. S. Metabolism Pharmacokinetics and Toxicity of Functional
Groups In Impact of Chemical Building Blocks on ADMET; Smith, D. A., Ed.; RSC
Publishing, 2010; Vol. 1, pp 460–482.
17. Backes, A.; Zech, B.; Felber, B.; Klebl, B.; Muller, G. Expert Opin. Drug Discov.
2008, 3, 1427.
18. Wager, T.; Hou, X.; Verhoest, P.; Villalobos, A. A. C. S. Chem. Neurosci. 2010, 1,
435.
b
Compd
R
CFAa
IC50
(nM)
(LE)
CK2%
inhib
@
CLK1% CLint
inhib
@
b/pc IVMN
pre-
(mL/
min/
kg)
screend
10
lM
10 lM
H
N
17.4
(.44)
16
24
24%
3%
28%
119
0.35 Neg
MeO
N
19. (a) Bredereck, H.; Simchen, G.; Rebsdat, S.; Kantlehner, W.; Horn, P.; Wahl, R.;
Hoffmann, H.; Greishaber, R. Chem. Ber. 1968, 101, 41; (b) Bolduc, A.; Dufresne,
S.; Hanan, G. S.; Skene, W. G. Can. J. Chem. 2010, 88, 236.
20. Synthetic examples and analytical data:
OH
O
8.50
(.42)
100%
<7.60 0.96 Neg
Compound 24 was synthesized by dissolving intermediate 13 (100 mg,
0.45 mmol) in 1 ml DMF and adding Hunig’s base (196 lL, 1.13 mmol) and
N
4-(4-phenyl)-4-hydroxypiperidine (200 mg, 1.13 mmol). The resulting mixture
was heated at 100 °C for 19 h. After cooling to room temperature, water was
added and the reaction mixture was extracted three times with ethyl acetate.
The combined organic extracts were dried (Na2SO4) and concentrated under
reduced pressure. The crude product was purified on a silica gel flash column
eluting with 4% MeOH in CH2Cl2 to afford 50 mg of the desired product as a
yellow oil which crystallizes on standing (30% yield).
16.1
(.39)
25
2%
15%
16.3
1.86 Neg
MeO
GSK-3b cell free enzyme inhibition assay;11 IC50 values are the mean of at least
a
three experiments. LE = ligand efficiency.10
Human Liver Microsome Incubation assay.13
b
1H NMR (400 MHz, chloroform-d) d ppm 1.90–2.00 (m, 2H) 2.26 (td, J = 12.98,
4.30 Hz, 2H) 3.08–3.18 (m, 2H) 3.32 (td, J = 12.01, 1.95 Hz, 2H) 3.59 (s, 3H) 6.67
(d, J = 1.76 Hz, 1H) 6.84 (d, J = 1.95 Hz, 1H) 7.26–7.32 (m, 1H) 7.33–7.41 (m, 2H)
7.49–7.56 (m, 2H) 7.64 (dd, J = 5.27, 1.37 Hz, 1H) 8.79 (d, J = 5.47 Hz, 1H) 9.25
(d, J = 1.17 Hz, 1H). 13C NMR (CDCl3) 31.76, 38.40, 47.89, 70.91, 93.63, 112.14,
In vivo brain to plasma measurements (rat).22 Calculations were based on the
average concentrations in brain and plasma at the 1 h time point.
c
d
In vitro micronucleus assay, high throughput screen in CHO cells, no
activation.23

K. Coffman et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1429–1433
1433
117.87, 124.70, 127.68, 128.78, 146.99, 157.23, 158.17, 159.24, 161.98, 164.58.
MS 363 (m+1).
50 mM Tris pH 7.5, 5 mM EDTA, 0.5% NP40, protease inhibitors (Roche
11873580001), Phosphorylated was measured in cellular lysates using
s
Compound 16 was prepared in a similar manner affording the desired product
in a 42% yield. 1H NMR (400 MHz, chloroform-d) d ppm 3.01–3.10 (m, 2H) 3.05
(t, J = 6.34 Hz, 2H) 3.41 (s, 3H) 3.48 (td, J = 6.34, 4.49 Hz, 2H) 3.89 (s, 3H) 5.03
(br s, 1H) 6.14 (d, J = 1.76 Hz, 1H) 6.44 (d, J = 1.95 Hz, 1H) 6.90–6.97 (m, 2H)
7.16 (dd, J = 7.42, 1.56 Hz, 1H) 7.22–7.27 (m, 2H) 7.64 (dd, J = 5.27, 1.37 Hz, 1H)
8.78 (d, J = 5.27 Hz, 1H) 9.26 (d, J = 1.37 Hz, 1H). 13C NMR (CDCl3) 27.49, 29.54,
44.95, 55.92, 82.88, 102.97, 111.00, 117.92, 121.64, 127.25, 128.73, 130.94,
147.67, 151.54, 157.44, 158.00, 159.13, 162.82, 163.07. MS 337 (m+1).
21. GSK-3b whole cell assay:
AlphaLISA (Perkin-Elmer) in 384 well Packard Optiplate (Perkin Elmer Cat.#
6007299). A capture antibody to total tau (HT7, Pierce) was conjugated to
Acceptor beads and added to lysate at 10
lg/mL final concentration in the
assay. Following 60 min incubation at room temperature, biotinylated
a
phospho-epitope specific detection antibody (AT8, Pierce) was used at 5 nM
final concentration and incubated for 30 min. Streptavidin donor beads
(Perkin-Elmer) were added at 40 lg/mL final concentration and incubated for
60 min in the dark. The plate was read on an Envision plate reader (Perkin
Elmer) using AlphaScreen settings.
A tetracycline-regulated expression system was used to produce a stable CHO
cell line (Clontech CHO AA8 #630904) carrying inducible expression of human
GSK3b and human Tau driven from modified pRevTRE Tet-Off vectors. The
human Tau gene encoding the 441aa full length protein (Accession
#NM_005910) was subcloned into vector pRevTre. GSK3b (Accession
#NM_002093) was subcloned into a modified pRevTre vector carrying zeocin
antibiotic resistance. A stable inducible cell line was created expressing both
the Tau gene and the GSK3b gene in the CHO Tet-Off cell line (CHO AA8). This
cell line was maintained in the uninduced state in MEM Alpha medium
(Invitrogen 12571-063) containing 10% FBS (Clontech # 8630-1), doxycycline
22. B/P Measurements:
Male Sprague–Dawley rats (n = 3/time point) were administered 5 mg/kg of
PF-X via subcutaneous administration. Blood samples were collected via
cardiac puncture after euthanization with CO2 at 0.5, 1, 2, and 4 h post-dose.
Samples were placed in to EDTA tubes and placed on ice. CSF was collected via
the cisterna magna and whole brain was collected via decapitation. CSF and
brain samples were immediately stored in dry ice. Blood samples were spun
down to collect plasma. Plasma, brain and CSF samples were stored at À20 °C
until analysis. LC/MS/MS was used to measure plasma, brain and CSF drug
levels. Brain to plasma ratios were calculated using the average concentrations
in brain and plasma measured at the 1 h time point.
(5 ng/mL Sigma D9891) zeocin (300
(300 g/mL Roche 10843555001), and Geneticin (300
10131-035). Transgene induction was initiated by removal of doxycycline
from the medium for 24 h. Following full induction of GSK3b and , cells were
incubated for 2 h with test compounds before lysis (lysis buffer: 250 mM NaCl,
l
g/mL Invitrogen R250-01), Hygromycin
l
l
g/mL Invitrogen
23. Aubrecht, J.; Xu, J. J. Assay method adapted from published protocols. In
Predictive Toxicology and Drug Safety; Xu, J. J., Urban, L., Eds.; Cambridge
University Press, 2010; pp 18–33.
s