Metabolism of a potent neuroprotective hydrazide

Article abstract of DOI:10.1016/j.bmc.2013.03.020

Abstract Using a drug discovery scheme for Alzheimer's disease (AD) that is based upon multiple pathologies of old age, we identified a potent compound with efficacy in rodent memory and AD animal models. Since this compound, J147, is a phenyl hydrazide, there was concern that it can be metabolized to aromatic amines/hydrazines that are potentially carcinogenic. To explore this possibility, we examined the metabolites of J147 in human and mouse microsomes and mouse plasma. It is shown that J147 is not metabolized to aromatic amines or hydrazines, that the scaffold is exceptionally stable, and that the oxidative metabolites are also neuroprotective. It is concluded that the major metabolites of J147 may contribute to its biological activity in animals.

Full text of DOI:10.1016/j.bmc.2013.03.020

Bioorganic & Medicinal Chemistry 21 (2013) 2733–2741
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Metabolism of a potent neuroprotective hydrazide
Chandramouli Chiruta a, Yanrong Zhao b, Fangling Tang b, Tao Wang b, David Schubert a,
⇑
a The Salk Institute for Biological Studies, La Jolla, CA 92037-1002, USA
b Pharmaron-Beijing, Beijing 100176, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Using a drug discovery scheme for Alzheimer’s disease (AD) that is based upon multiple pathologies of
old age, we identified a potent compound with efficacy in rodent memory and AD animal models. Since
this compound, J147, is a phenyl hydrazide, there was concern that it can be metabolized to aromatic
amines/hydrazines that are potentially carcinogenic. To explore this possibility, we examined the metab-
olites of J147 in human and mouse microsomes and mouse plasma. It is shown that J147 is not metab-
olized to aromatic amines or hydrazines, that the scaffold is exceptionally stable, and that the oxidative
metabolites are also neuroprotective. It is concluded that the major metabolites of J147 may contribute to
its biological activity in animals.
Received 5 January 2013
Revised 23 February 2013
Accepted 3 March 2013
Available online 24 March 2013
Keywords:
Metabolites
Alzheimer’s
Neuroprotective
Microsomes
Hydrazines
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
There are currently no effective drugs that prevent the nerve
cell death associated with Alzheimer’s disease (AD), ischemic
stroke, or any other age-associated neurodegenerative disease. To
address this concern, we devised a drug discovery scheme based
upon cell culture models of age-associated pathologies, such as
the loss of nerve trophic factors, proteotoxicity, oxidative stress
and reduced energy metabolism. A recent result of this program
is a compound called J147 (Fig. 3A). J147 promotes memory in both
normal rodents and AD mice, and limits synapse loss and pathol-
ogy in AD animals.1 J147 is a hydrazide with the potential to be
metabolized to aromatic amines or hydrazines,2 a class of com-
pounds with well-defined carcinogenic and toxic properties.3–5
There are several FDA approved drugs that are aromatic hydra-
zones, such as the antihypertensive drug dihydralazine (for review,
see6). However, because of the concern about the intrinsic stability
of the hydrazone bridge and the production of aromatic amines/
hydrazines from J147, its metabolites from both human and mouse
microsomes were identified, synthesized, and assayed for biologi-
cal activity in several neuroprotection assays. It is shown that the
core structure of J147 is stable in vivo, that no aromatic amines
or hydrazines are produced, and that many of the metabolites have
biological activities similar to J147.
2.1. Microsomal metabolism of J147 in human and mouse liver
microsomes
To determine the stability of J147, 5 lM J147 was added to
mouse or human whole blood and plasma, or microsomes plus or
minus NADPH and the disappearance of J147 monitored by LC–
MS/MS. Figure 1 shows that J147 is relatively unstable in human
microsomes in the presence of NADPH. It has a half-life of
4.5 min in human microsomes and less than 4 min in mouse
microsomes (not shown). To determine the metabolites, the reac-
tion mixture was extracted after 5 min with ethyl acetate (J147
is not very soluble in acetonitrile and strongly adheres to plastic)
and run on a reverse-phase HPLC column. Figure 2 shows the HPLC
profile before and after exposure to mouse (A and B) and human (C
and D) microsomes.
The structures of the metabolites (M1, M2, M3, M4 and M5) of
J147 were identified using a AB API4000 Q Trap LC/MS/MS instru-
ment. To identify J147 metabolites, the chromatographic and MS
fragmentation behaviors of the parent compound were first inves-
tigated. The retention time of J147 was 9.49 min (mouse) or
10.97 min (human) because different HPLC columns were used.
In positive scan mode, J147 formed
a protonated molecule
[M+H]+ at m/z 351. Figure 3A shows the product ion spectrum of
J147 under the Enhanced Product Ion (EPI) scan. The fragments
of J147 were formed predominantly by the cleavage of the N–N
and the C–F bond, resulting in product ions at m/z 136, 216, 200,
333 and 109. The CID pathways of J147 were deduced and are
⇑
Corresponding author. Tel.: +1 8584534100; fax: +1 8585359062.
E-mail address: schubert@salk.edu (D. Schubert).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.03.020

2734
C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
shown in Figure 3A. Using this fragment ion pattern, the mass
spectra and chromatographic behaviors of the detected metabo-
lites were compared with those of the parent compound and avail-
able authentic standards to characterize the metabolic structural
modifications.
J147 metabolite (mouse) M1’s retention time is 8.36 min, exhib-
iting a protonated molecule m/z 337.1, with a mass shift of ꢀ14 Da
relative to J147 and metabolite (mouse) M2’s retention time is at
7.60 min, exhibiting a protonated molecule m/z 353, with a mass
shift of 2 Da from J147 and 16 Da from M1. Figure 3B and C shows
the product ion spectrum of M1 and M2, respectively under the EPI
scan. The fragment ions of M1 and M2 at m/z 122, 138, 200, 216
and 218 were formed by the cleavage of the N–N bond. The frag-
ment ions at m/z 319 and 335 were also observed by the cleavage
of the C–F bond from M1 and M2, respectively. The CID pathways
of M1 and M2 were deduced and are shown in Figure 3B and C.
M3 (mouse) was identified as the oxidized metabolite of J147
by addition of one hydroxyl group and exhibiting a protonated
molecule m/z 367 with a retention time at 7.86 min. Metabolite
M4 (mouse) was identified as the demethylated and dioxidized
metabolite of J147 by removal of methyl and addition of two
hydroxyls, and exhibiting a protonated molecule m/z 369 with a
retention time at 5.96 min. M5 (mouse) was identified as the
dioxidized metabolite of J147 by addition of two hydroxyls and
exhibiting a protonated molecule m/z 383 with a retention time
at 6.82 min. Fragment ions m/z 136, 216, m/z 138, 216, and
m/z 152, 216 of M3, M4 and M5, respectively were observed, which
were the result of the N–N bond cleavage. The fragment ions m/z
337, 339 and 353, due to loss of CH2OH and the fragment ions
m/z 349, 351 and 365, due to loss of H2O were observed for all
three metabolites M3, M4 and M5, respectively. A common
fragment ion m/z 204 in all three metabolites M3–M5 was ob-
served by loss of CH2OH followed by N–N bond cleavage from
the corresponding parent molecules. The product ion spectra under
EPI scan and CID pathways of M3, M4 and M5 are shown in
Figure 4A–C.
Figure 1. Metabolic stability of J147 in human liver microsomes. Five micromolar
J147 was added to pooled human liver microsomes (0.5 mg/mL) in buffer plus or
minus 1 mM NADPH, and the concentration of J147 determined by LC/MS/MS. The
percent remaining J147 is plotted as a function of time.
The metabolites M1–M5 of J147 were also observed in the
HPLC-UV chromatographs with similar retention times. In addi-
tion, these metabolites also showed similar isotope patterns in
the Enhanced Resolution Scan (ER) compared with the parent com-
pound (not shown).
A major goal of these studies was to determine if J147 is metab-
olized to potentially toxic compounds such as 2,4-dimethylaniline
(m/z = 121), 2,4-dimethylhydrazine (m/z = 136) or trifluoroaceta-
mide hydrolysed product, 3-methoxy benzaldehyde (2,4-dimethyl-
phenyl) hydrazone (m/z = 254). These compounds were not seen in
either human or mouse microsomal preparations or mouse plasma
following gavage of the drug.
Figure 5 summarizes the metabolites of J147. All are oxidation
products of the aromatic rings; there was no indication that the
hydrazone was cleaved, confirming that the molecular scaffold of
J147 is quite stable when incubated with microsomes. Only M1
and M3 were found in reproducible amounts in the metabolites
of human microsomes (Fig. 2C and D), indicating a more limited
metabolism by the human enzymes. The compound eluting just
before J147 in the human preparation (Fig. 2D) had the same mass
as J147 and was probably a stereoisomer of J147.
2.2. Blood metabolites of J147
Finally, because the metabolic products of J147 from liver
microsomes contain aromatic hydroxyl groups, it is possible that
they become sulfated or glucuronidated in the blood. Therefore,
the structural identification of the plasma metabolites in mice
was examined. Figure 6 shows the blood and brain distribution
of J147 as a function of time following gavage of a 20 mg/kg in corn
Figure 2. Separation of microsomal metabolic products: Five micromolar J147 was
incubated with mouse (A and B) or human (C and D) liver microsomes and
extracted immediately at zero time (A and C) or after 5 min (B and D), followed by
separation on a C18 reverse phase HPLC column. The fractions were collected, dried
down, re-suspended in ethanol, and their biological activity determined in three
neuroprotection assays (Fig. 7). Different HPLC columns were used for mouse and
human material, therefore there are different elution times.

C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
2735
A
B
C
+EPI (351.12) CE (15): 9.305 min fro...
Max. 4.7e5 cps.
F
F
NH
O
350.8
4.7e5
4.0e5
F
F
F
N
O
N
136.0
O
F
m/z 136
N
m/z 216
3.0e5
2.0e5
1.0e5
0.0
J-147
O
216.2
200.2
m/z 351
F
F
O
O
F
F
HN
N
N
O
134.0
352.2
180.3
200
333.4
300
109.0
100
238.6
250
m/z 109
50
150
350
400
m/z 200
O
m/z 333
m/z, Da
+EPI (337.10) CE (15): 8.118 min fro...
Max. 4.7e5 cps.
F
F
NH
OH
337.1
O
4.7e5
4.0e5
F
F
F
N
O
OH
m/z 122
F
N
N
3.0e5
2.0e5
m/z 216
F
M1
m/z 337
F
O
O
F
1.0e5
216.1
F
122.0
200.1
HN
N
N
319.0
300
0.0
50
100
150
200
250
350
400
m/z 200
m/z, Da
m/z 319
OH
+EPI (353.10) CE (15): 7.208 min fro...
Max. 4.8e5 cps.
F
F
F
NH
353.0
O
HO
4.8e5
4.0e5
F
F
HN
O
N
OH
F
m/z 138
N
HO
m/z 218
3.0e5
2.0e5
M2
OH
m/z 353
F
F
F
O
O
138.0
150
F
1.0e5
218.2
219.8
200
m/z, Da
N
N
HN
200.3
243.1
250
335.3
HO
0.0
50
100
300
350
400
m/z 200
m/z 335
OH
Figure 3. The product ion spectrums and proposed CID pathways of J147 (A), M1 (B), and M2 (C) under the Enhanced Product Ion (EPI) scan.
oil. The peak concentration occurred around 2 h in both tissues
with a half-life of 2.5 h. At all times, the tissue concentration was
well above the EC50 of J147 in our cell culture screening assays.1
The structures of the metabolites were determined using the same
methods as for the microsomal metabolites and are shown in Ta-
ble 1. The two major metabolites had gains of sulfate and of glucu-
ronic acid. A minor metabolite was a loss of a methyl group of J147.
anhydride and triethylamine in CH2Cl2 from the corresponding
hydrazones. Finally, methyl and ethyl deprotection of J147 and 3,
respectively was performed by iodocyclohexane in DMF to yield
M1 and M2. Metabolite M3 was prepared from hydrazide 6 by
reduction of methyl ester using DIBALH in dichloromethane.
Hydrazine 4 prepared from the commercially available amine by
NaNO2 and SnCl2 reaction (Scheme 1).
2.3. Synthesis of J147, M1, M2 and M3
2.4. Neuroprotective activity of metabolites
The synthesis of the metabolites was carried out in 2–3 steps
using simple chemistry as described in our previous paper.1 Hydra-
zones 1, 2 and 5 were synthesized by condensation of appropri-
ately substituted aldehydes and hydrazines in EtOH at room
temperature (Scheme 1). Synthesis of intermediate hydrazides 3,
6 and J147 has been carried out by acetylation using trifluoroacetic
To confirm the human microsomal metabolite structures by
chemical synthesis and to provide sufficient material to further
evaluate them in biological assays, human hepatic microsomal
metabolites M1, M2 and M3 of J147 were prepared and character-
ized by 1H, 13C NMR, LC/MS and ESI-MS analysis. There is a possi-
bility of other ortho, meta hydroxylated isomers of M2 and ortho

2736
C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
A
F
+EPI (367.10) CE (15): 7.460 min fro...
Max. 4.8e5 c.ps
F
NH
O
F
367.0
F
4.8e5
4.0e5
HN
F
F
F
O
O
N
O
N
F
+
F
m/z 136
CH2
N
N
m/z 216
OH
3.0e5
2.0e5
1.0e5
0.0
M3
O m/z 367
m/z 337
O
O
F
F
O
N
136.0
F
N
337.4
349.4
F
F
204.2
216.3
F
F
F
O
+
CH2
186.2
166.1
150
88.0
100
254.4
250 300
m/z, Da
HN
m/z 349
HN
O
50
200
350
400
Hydroxylation of this methyl is
m/z 186
another possible structure of M3
m/z 204
B
F
F
+EPI (369.10) CE (15): 5.363 min fro...
Max. 3.5e5 cps.
NH
O
HO
F
F
369.1
F
3.5e5
3.0e5
2.5e5
2.0e5
1.5e5
1.0e5
5.0e4
HN
O
N
OH
F
m/z 138
N
+
CH2
F
m/z 216
F
OH
O
N
OH
F
M4
N
OH
+
m/z 369
OH
Hydroxylation of this
methyl is another possible
structure of M4
CH2
m/z 351
OH
F
F
F
F
O
O
F
F
339.1
351.1
216.1
204.2
138.1
150
327.0
300
N
HN
0.0
N
HO
50
100
200
m/z, Da
250
350
400
m/z 339
m/z 204
OH
C
+EPI (383.10) CE (15): 6.168 min fro...
Max. 4.5e5 cps.
F
F
F
NH
NH
O
HN
383.0
4.5e5
4.0e5
O
HO
O+
m/z 150
O
F
F
F
m/z 152
O
N
+
CH2
216
N
m/z
3.0e5
2.0e5
F
F
F
HO
O
N
OH
M5
N
O
F
m/z 383
OH
F
F
1.0e5
150.1
F
F
m/z 353
O
O
O
F
216.1
204.1
353.2
365.1
N
HN
0.0
N
50 100 150 200 250 300 350 400 450
m/z, Da
HO
+
CH2
m/z 204
m/z 365
O
Figure 4. The product ion spectrums and proposed CID pathways of M3 (A), M4 (B) and M5 (C) under the Enhanced Product Ion (EPI) scan.
hydroxymethyl isomer of M3 (Figs. 3C and 4A). However, because
the enzymatic hydroxylation is likely to take place at the para po-
sition due to steric and electronic factors around the aromatic ring,
only the potential M2 and M3 metabolites with the para-hydroxy
groups were made.10,11 The exclusive formation of these M2 and
M3 metabolites was confirmed by comparing them with the
authentic samples isolated from microsomes by HPLC and TLC.
All of the tested synthetic compounds have purity above 95%.
To determine if any of the breakdown products have biological
activity, each of the three major peaks in the human microsome
preparation were initially collected and measured for biological
activity in three assays: (1) trophic factor withdrawal assay; (2)
the ability to block intracellular Ab toxicity; and (3) protection
from oxidative stress. M1, M2, and M3 isolated directly from
microsomes were all active in these assays. Once their structures
were determined and the compounds were synthesized, it was
determined that most of the J147 metabolites were biologically ac-
tive in the neuroprotection assays (Fig. 7 and Table 2).
Trophic factors are greatly reduced in AD brain.12 In our assay
for trophic factor withdrawal (TFW), primary embryonic cortical
cells are plated at low density in serum-free medium (Fig. 7A). At
low density, the cells die within 2 days but can be rescued by com-
binations of neurotrophic growth factors, but not by one alone.13 In
contrast, cell death is prevented by J147 alone with an EC50 of
33 nM. M1 and M2 were similarly active, but M3 was 10-fold less
active in this assay (Table 2).
Ab is thought to be one of the toxic entities in AD, and the intra-
cellular accumulation of Ab and C-terminal amyloid precursor

C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
2737
F
F
O
F
N
N
HO
M2
OH
m/z 353
F
F
F
F
F
F
F
F
F
O
N
O
N
O
N
N
N
N
M3
M1
J147
O
OH
OH
O
m/z 367
m/z 337
m/z 351
F
F
F
F
F
F
O
N
O
N
N
N
HO
HO
M4
M5
m/z 383
OH
OH
O
OH
m/z 369
Figure 5. Metabolic products of J147.
aggregates. Metabolites M1, M2, and M3 all protect cells from
intracellular Ab toxicity with EC50s below 100 nM (Fig. 7B and
Table 2).
Many models of programmed cell death have been established
to characterize the oxidative mechanisms underlying neuronal
degeneration. One of the most robust of these models is the gluta-
mate-induced programmed cell death of immature cortical neu-
rons and of hippocampal HT22 nerve cells.17,18 In this assay,
elevated levels of extracellular glutamate interfere with cystine
uptake through the cystine/glutamate antiporter, which normally
carries cystine into cells at the expense of the outflow of glutamate.
The decreased cystine uptake leads to the depletion of intracellular
GSH, a cysteine-containing tripeptide vital for cell survival because
of its ability to act as an enzymatic cofactor and antioxidant. This
form of cell death, called oxytosis,18 is distinct from excitotoxicity
but is important to the understanding of neuronal degeneration in
that it represents a form of oxidative stress common to all CNS dis-
eases. Figure 7C and Table 2 show that J147 metabolites are much
less neuroprotective in this assay than J147, perhaps because they
are more oxidized forms of J147.
3. Conclusions
The above data show that the microsomal metabolites of J147
are oxidation products of the aromatic rings, that the hydrazone
scaffold of the compound remains intact, and that the human
microsomal metabolites are biologically active in the cell culture
neuroprotection assays used to develop J147 from its natural prod-
uct parent molecule curcumin.9,1 Although there is concern that
hydrazones/hydrazides associated with aromatic rings can be
metabolized to toxic amines/hydrazines,10,11 none of the com-
pounds were identified as either microsomal or blood metabolites.
Hydrolysis of trifluoroacetamide was also not observed. While it is
clearly impossible to say that none were produced, based upon the
technology that was used, they must constitute less than 5% of the
total. If there is no hydrolysis of the amide there will be no forma-
tion of hydrazone and no further breakdown into amine/hydrazine
derivatives, then the microsomal metabolites of J147 are likely to
be simple oxidation products of the aromatic rings. This prediction
Figure 6. Pharmacokinetics of J147 following a single gavage at 20 mg/kg in corn
oil. Blood and brain distribution as a function of time. x–x, blood; o–o, brain. The
data are presented as mean plus or minus the standard error of the mean. N = 3 per
time point.
protein (APP) fragments are also toxic.14,15 To determine if J147
metabolites are able to inhibit intracellular Ab toxicity, a human
neuroblastoma cell line that conditionally expresses the C-terminal
99 amino acids (C99) derived from the b-secretase cleavage of APP
was used. This cell line, called MC-65,16 has recently been used to
identify drugs that inhibit amyloid aggregation.8 The cells are rou-
tinely grown in the presence of tetracycline and, following its re-
moval, the expression of C99 is induced and the cells die within
4 days due to the accumulation of intracellular, toxic protein

2738
C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
O
N
CF3
HCl
H. HCl
6
.H2NHN
1
N
N
5
a
b
N
O
+
4
2
3
R
R
R
J147
M1
: R = 3-OMe
: R = 3-OH
1
2
: R = 3-OMe
: R = 3, 4-diOEt
c
3
: R = 3, 4-diOEt
c
M2
: R = 3, 4-diOH
O
N
CF3
NHNH2HCl
NH2
H. HCl
N
N
N
d, e
f
b
CO2Me
R
CO2Me
OMe
OMe
CO2Me
5
4
6
M3
: R = CO2Me
: R = CH2OH
g
Scheme 1. Reagents and conditions: (a) EtOH, rt, 1–5 h, 80–90%; (b) Et3N, (CF3CO)2O, CH2Cl2, 0 °C, 1–5 h, 60–80%; (c) iodocyclohexane, DMF, reflux, 12 h, 60–70%; (d) NaNO2,
6 N HCl, <15 °C, 1.5 h; (e) SnCl2, 0–10 °C, 2 h, 80%; (f) 3-methoxybenzaldehyde, EtOH, rt, 5 h, 80%; (g) DIBALH, ꢀ78 °C, 2 h, 68%.
Figure 7. J147 metabolites are biologically active. Increasing concentrations of J147 and its synthetic metabolites were assayed in three different assays for neuroprotection.
In all cases, the assay outcome was protection from nerve cell death caused by the lack of trophic support (A), the toxicity of Ab accumulation within cells (B), or oxidative
stress caused by the loss of glutathione (C). The calculated EC50 values are shown in Table 2. ꢁ = M1; 4 = M2;
r
= M3; s = J147; h = inactive analogue of J147 in which both
nitrogens are replaced by carbons.
Table 1
Mouse blood metabolites
Peak No.
Sample name
rt (min)
Expected (m/z)
Mass shift
Proposed biotransformation
1
2
3
4
J147
BM1
BM2
M1
9.27
5.35
5.54
8.04
351.1
431.1
513.2
337.1
—
80
162
ꢀ14
—
Sulfonation (Gain of ‘SO3’)
Demethylation and glucuronidation (loss of ‘CH2’ and gain of ‘C6H8O6’)
Demethylation (loss of ‘CH2’)
Mice were given J147 at 20 mg/kg iv and the J147 metabolites in the blood assayed 20 min later by procedures described for microsomal metabolites. Phenomenex Synergi 4
Polar-RP (50 ꢁ 2.00 mm) column used for analysis.
l
was confirmed by the identification and synthesis of the human
microsomal metabolites, as well as the observation of only sulfo-
nated and glucuronidated products of the microsomes found in
blood. The fact that the microsomal metabolites have similar bio-
logical activities as the parent drug and that they become sulfo-
nated or glucoronidated in circulation argues that toxicity due to
drug metabolism is unlikely to be an immediate concern with
J147, but will require study during the Investigational New Drug
Allowance and Phase 1 clinical trials.
Table 2
Neuroprotective activities of J147 and metabolites
Compound
name
Trophic factor
withdrawal
Intracellular Ab
toxicity
Oxidative
stress
M1
M2
M3
J147
0.012
0.031
0.320
0.033
0.078
0.034
0.092
0.011
0.185
1.0
0.180
0.020
EC50 values (in lM) of neuroprotective activities of J147 and metabolites.

C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
2739
mouse microsomes. In some cases heat-activated (98 °C, 10 min)
microsomes were used as controls and verapamil was used as a
control. At 0.1, 2, 5, 10 and 20 min at 37 °C the reaction was
stopped with three volumes of ethyl acetate, centrifuged at
16,000 rpm for 15 min and an aliquot of the supernatant used for
LC/MS/MS analysis. In some cases the individual metabolites were
directly isolated from the HPLC run and assayed for biological
activity.
4. Experimental section
4.1. Materials
Compounds J147, M1, M2 and M3 were synthesized in our lab-
oratory at Salk Institute. All starting materials, chemicals, reagents
were obtained from Sigma Aldrich, (Milwaukee WI), and used as
received. Solvents used for synthesis and chromatographic analysis
were HPLC or ACS reagent grade and were purchased from Fisher
Scientific Co (Pittsburg, PA). Thin layer chromatography (TLC) used
EMD silica gel F-254 plates (thickness of 0.25 mm). Flash chroma-
tography used EMD silica gel 60, 230–400 mesh and were pur-
chased from EMD Chemicals (San Diego, CA). MTT (3-(4,5
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) for the
bioassays was purchased from Sigma (St. Louis, MO). Liver micro-
somes from mouse and humans, DMEM/F12 medium N2 supple-
ment and the live-dead assay kit (#L3224) were purchased from
In Vitrogen (Carlsbad, CA).
4.5. Animal studies
Male Balb C mice were given a single dose by body weight intra-
venously at a volume of 5 mL/kg (1 mg/kg) or via oral gavage at a
volume of 10 mL/kg (5 mg/kg). Blood samples were collected into
Na-Heparin microtainer tubes for plasma separation. Blood sam-
ples were centrifuged at 4 °C at 6000 rpm for 5 min. Decanted plas-
ma samples were stored at ꢀ80 °C until used. Brains were perfused
and collected for the brain penetration. The tissue samples were
stored at ꢀ80 °C. Plasma and brain samples were thawed on ice
and kept at 4 °C during processing. Brain tissues were homoge-
nized in PBS, pH 7.4. An aliquot of plasma or brain homogenate
sample or calibration sample were mixed with three volumes of
methanol containing internal standard, incubated on ice for
10 min, and centrifuged. The protein free supernatant was used
for LC/MS/MS analysis.
4.2. Analytical methods
1H NMR and 13C NMR were recorded at 500 and 125 MHz, respec-
tively, on a Varian VNMRS-500 spectrometer at the Salk institute (La
Jolla, CA) using the indicated solvents. Chemical shift (d) is given in
parts per million (ppm) relative to tetramethylsilane (TMS) as an
internal standard. Coupling constants (J) are expressed in hertz
(Hz), and conventional abbreviations used for signal shape are as fol-
lows: s = singlet; d = doublet; t = triplet; m = multiplet; dd = dou-
blet of doublets; br s = broad singlet. Liquid chromatography mass
spectrometry (LCMS) was carried out using a Shimadzu LC-20AD
spectrometer at The Scripps Research Institute (La Jolla, CA), and
electrospray ionization (ESI) mass analysis with a Thermo Scientific
LTQ Orbitrap-XL spectrometer at the Salk institute (La Jolla, CA).
Melting points were determined with a Thomas-Hoover capillary
melting point apparatus at the Salk institute (La Jolla, CA), and are
uncorrected. All final compounds were characterized by LCMS and
1H NMR and gave satisfactory results in agreement with the pro-
posed structure. All of the tested compounds have a purity of at least
95% which was determined by analysis on a C18 reverse phase HPLC
column [Phenomenex Luna (50 mm ꢁ 4.60 mm, 3 m)] at The
Scripps Research Institute (La Jolla, CA), using 10–90% CH3CN/H2O
containing a 0.02% AcOH with a flow rate of 1 mL/min (5 min gradi-
ent) and monitoring by a UV detector operating at 254 nm.
4.6. Trophic factor withdrawal
Primary cortical neurons were prepared from 18-day-old rat
embryos according to published procedures7 and cultured at a
low cell density of 1 ꢁ 106/35 mm dish in DMEM/F12 (2 mL) con-
taining N2 supplement and the compounds to be assayed. Viability
was assayed 2 days later using a fluorescent live-dead assay and
the data presented as the percent of input cells surviving.
4.7. Intracellular amyloid toxicity
The induction of intracellular amyloid toxicity in MC65 cells
was done exactly as described.8 Briefly, cells from confluent cul-
tures were dissociated and plated at 4 ꢁ 105 per 35 mm tissue cul-
ture dish in the presence (no induction) or absence (APP-C99
induced) of 1 g/mL tetracycline and in the presence or absence of
the indicated compounds. At day 4 the control cells in the absence
of tetracycline were dead, and cell viability was determined by the
MTT assay.9 The data are presented as viability relative to controls
plus tetracycline.
4.3. Metabolic determination
Mass spectra were acquired in the positive mode scanning over
the mass range of 50–1000. LCMS M+H signals were consistent with
the predicted molecular weights for the reported compounds.
Metabolite analyses were performed on AB API4000 Q Trap LC/
MS/MS spectrometer and HPLC Shimadzu (DGV-20A3), LC-20AD
and CTC Analytics HTC PAL System at Pharmaron (Beijing, China).
4.8. Oxytosis assay
HT22 cells were plated at 2 ꢁ 103 cells per well in 96-well tissue
culture dishes in DMEM plus 10% fetal calf serum. The following
day the test compounds were added in triplicate at indicated con-
centrations. Thirty minutes after compound addition, 5 mM gluta-
mate was added to initiate the cell death cascade. Twenty hour
later, the MTT assay was performed. The results are presented as
the percentage of survival relative to the controls with vehicle
alone.
The HPLC column used was Phenomenex Synergi 4
l Polar-RP
(50 ꢁ 2.00 mm) or a Phenomenex Luna 5 C18 (50 ꢁ 2.00 mm). HPLC
resolution was achieved with a gradient consisting of solvent A
(water–acetonitrile–formic acid, 95:5:0.1) and solvent B (water–
acetonitrile–formic acid, 5:95:0.1) with 0.3 mL/min flow rate. The
Luna column was used for the human microsomes (Fig. 2C and D)
and the Polar-RP for mouse (Fig. 2A and B). Blood metabolites were
4.9. Synthesis of (E)-N-(2,4-dimethylphenyl)-2,2,2-trifluoro-N0-
(3methoxybenzyli-dene) acetohydrazide (J147)
analyzed on the Phenomenex Synergi 4
l
Polar-RP (50 ꢁ 2.00 mm).
A mixture of 3-methoxybenzaldehyde (500 mg, 3.67 mmol)
and (2,4-dimethylphenyl) hydrazine hydrochloride (632 mg,
3.67 mmol) in EtOH (10 mL) was stirred at room temperature for
1 h, obtained solid was filtered off, washed with ethanol and dried
4.4. Microsomal stability
Microsomal stability assays were done in 5 mM phosphate buf-
fer, pH 7.3, with or without 1 mM NADPH and 0.5 mg/mL human or

2740
C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
under vacuum to afford hydrazone hydrochloride 1 (960 mg) in
90% yield as light brown solid. This unstable hydrazone
1H), 7.15 (s, 1H), 7.19 (d, J = 8.5 Hz, 1H), 7.23 (s, 1H), 7.30 (s,
1H); 13C NMR (CDCl3, 125 MHz) d ppm 16.99, 21.29, 113.11,
115.18, 115.85, 117.52, 122.94, 127.20, 128.54, 128.70, 129.98,
132.55, 136.21, 140.88, 144.44, 144.56, 147.26; MS (ESI): m/z calcd
for C17H15F3N2O3 ([M+H]+) 353.1107; found 353.1176 ([M+H]+).
a
(570 mg, 1.96 mmol) was dissolved in CH2Cl2 (10 mL), Et3N
(0.54 mL, 3.93 mmol) followed by (CF3CO)2O (0.27 mL, 1.96 mmol),
was added at 0 °C and the mixture was stirred at room tempera-
ture for 1 h. Reaction mixture was diluted with aq satd NaHCO3
solution (20 mL), extracted with CH2Cl2 (2 ꢁ 30 mL), dried
(Na2SO4) and evaporated, resulting solid was recrystalized from
ethanol to give J147 (550 mg, 80%) as a white solid: mp 70–
72 °C; LCMS purity 98%; 1H NMR (CDCl3, 500 MHz) d ppm 2.09
(s, 3H), 2.42 (s, 3H), 3.83 (s, 3H), 6.95 (dd, J = 8.5, 2.0 Hz, 1H),
7.05 (d, J = 8.0 Hz, 1H), 7.13 (d, J = 7.5 Hz, 1H), 7.21 (d, J = 8.5 Hz,
1H) 7.24–7.30 (m, 4H); 13C NMR (CDCl3, 125 MHz) d ppm 17.20,
21.50, 55.47, 111.97, 116.03, 117.40, 118.32, 121.23, 128.78,
128.97, 129.90, 129.99, 132.81, 134.95, 136.49, 141.16, 144.10,
160.22; MS (ESI): m/z calcd for C18H17F3N2O2 ([M+H]+) 351.1314;
found 351.1366 ([M+H]+).
4.12. Synthesis of methyl 4-hydrazinyl-3-methylbenzoate
hydrochloride (4)
To a stirred and cooled (0 °C) slurry of commercially available
methyl 4-amino-3-methylbenzoate (2 g, 12.12 mmol) in 6 N HCl
(12 mL) was added NaNO2 (0.92 g, 13.33 mmol) in water (2 mL)
drop wise maintaining a temperature of <15 °C during the addition.
The mixture was stirred an additional 1.5 h affording a light yel-
low, homogeneous solution. To the mixture was carefully added
4.58 g (24.24 mmol) of anhydrous SnCl2. The temperature during
the addition was kept <10 °C. The mixture was stirred at 0 °C for
1 h. The precipitated solid was collected by filtration, dried under
vacuum to afford methyl 4-hydrazinyl-3-methylbenzoate hydro-
chloride 4 (1.75 g 81%) as a brown solid: mp 224–226 °C; 1H
NMR (CDCl3, 500 MHz) d ppm 2.22 (s, 3H), 3.80 (s, 3H), 6.99 (d,
J = 8.5 Hz, 1H), 7.70 (s, 1H), 7.77 (d, J = 8.5 Hz, 1H); MS (ESI): m/z
calcd for C9H12N2O2 ([M+H]+) 181.0932; found 181.0911 ([M+H]+).
4.10. Synthesis of (E)-N-(2,4-dimethylphenyl)-2,2,2-trifluoro-N0-
(3-hydroxybenzyli-dene) acetohydrazide (M1)
J147 (410 mg, 1.17 mmol) and iodocyclohexane (0.30 mL,
2.34 mmol) in DMF (2 mL) was refluxed for 12 h. Reaction mixture
was cooled and diluted with water, (10 mL), extracted with EtOAc
(2 ꢁ 20 mL), washed with brine (20 mL), dried (Na2SO4) and evap-
orated, resulting solid was recrystalized from ethanol to give M1
(250 mg, 63%) as a white solid: mp 151–153 °C; LCMS purity
99%; 1H NMR (CDCl3, 500 MHz) d ppm 2.08 (s, 3H), 2.41 (s, 3H),
5.14 (s, 1H) 6.86 (dd, J = 7.5, 2.0 Hz, 1H), 7.03 (d, J = 8.0, Hz, 1H),
7.07 (d, J = 8.0 Hz, 1H), 7.21 (m, 5H); 13C NMR (CDCl3, 125 MHz)
d ppm 16.98, 21.31, 113.42, 118.04, 121.27, 128.50, 128.78,
130.04, 132.69, 134.81, 136.20, 141.01, 143.82, 156.02; MS (ESI):
m/z calcd for C17H15F3N2O2 ([M+H]+) 337.1158; found 337.1203
([M+H]+).
4.13. Synthesis of (E)-methyl 4-(2-(3-methoxybenzylidene)-1-
(2,2,2-trifluoroacetyl) hydrazinyl)-3-methylbenzoate (6)
Hydrazine 4 (1 g, 4.63 mmol) and 3-methoxybenzaldehyde
(629 mg, 4.63 mmol) in EtOH (15 mL) was stirred at room temper-
ature for 5 h, precipitated solid was filtered off, washed with etha-
nol and dried under vacuum to afford hydrazone hydrochloride 5
(1.32 g 95%) as a white solid. This unstable hydrazone 5 (793 mg,
2.37 mmol) was dissolved in CH2Cl2 (20 mL), then Et3N (1.32 mL,
9.49 mmol) followed by (CF3CO)2O (0.66 mL, 4.75 mmol), was
added at 0 °C and the mixture was stirred at room temperature
for 12 h. Reaction mixture was diluted with aq sat NaHCO3 solution
(50 mL), extracted with CH2Cl2 (2 ꢁ 50 mL), dried (Na2SO4) and
evaporated. Flash chromatography of the residue over silica gel
using 5ꢀ20% EtOAc/hexane gave (E)-methyl 4-(2-(3-methoxyben-
zylidene)-1-(2,2,2-trifluoroacetyl) hydrazinyl)-3-methylbenzoate
6 (560 mg, 60%) as a thick viscous colorless syrup: 1H NMR (CDCl3,
500 MHz) d ppm 2.20 (s, 3H), 3.84 (s, 3H), 3.98 (s, 3H), 6.98 (dd,
J = 8.5, 2.0 Hz, 1H), 7.12 (d, J = 7.5 Hz, 1H), 7.18 (s, 1H), 7.28 (m,
3H), 8.09 (d, J = 8.0 Hz, 1H), 8.15 (s, 1H); MS (ESI): m/z calcd for
4.11. Synthesis of (E)-N0-(3,4-dihydroxybenzylidene)-N-(2,4-
dimethylphenyl)-2,2,2-trifluoroacetohydrazide (M2)
A mixture of 3,4-diethoxybenzaldehyde (300 mg, 1.54 mmol)
and (2,4-dimethylphenyl) hydrazine hydrochloride (266 mg,
1.54 mmol) in EtOH (5 mL) was stirred at room temperature for
2 h, precipitated solid was filtered off, washed with ethanol and
dried under vacuum to afford hydrazone hydrochloride
2
(480 mg) in 89% yield as a white solid. The hydrazone (479 mg,
1.38 mmol) was dissolved in CH2Cl2 (10 mL), Et3N (0.76 mL,
5.51 mmol) followed by (CF3CO)2O (0.38 mL, 2.75 mmol), was
added at 0 °C and the mixture was stirred at room temperature
for 3 h. Reaction mixture was diluted with aq sat NaHCO3 solution
(20 mL), extracted with CH2Cl2 (2 ꢁ 30 mL), dried (Na2SO4) and
evaporated, resulting solid was recrystallized from ethanol to give
(E)-N0-(3,4-diethoxybenzylidene)-N-(2,4-dimethylphenyl)-2,2,2-
trifluoroacetohydrazide 3 (502 mg, 80%) as a white crystals: mp
110–112 °C; 1H NMR (CDCl3, 500 MHz) d ppm 1.57–1.45 (m, 6H),
2.09 (s, 3H), 2.41 (s, 3H), 4.16–4.09 (m, 4H), 6.81 (d, J = 8.5 Hz,
1H), 6.97 (dd, J = 8.5, 2.0 Hz, 1H), 7.04 (d, J = 8.0 Hz, 1H), 7.19 (m,
2H), 7.24 (s, 1H), 7.35 (s, 1H); MS (ESI): m/z calcd for C21H23F3N2O3
([M+H]+) 409.1733; found 409.1781 ([M+H]+).
C
19H17F3N2O4 ([M+H]+) 395.1213; found 395.1230 ([M+H]+).
4.14. Synthesis of (E)-2,2,2-trifluoro-N-(4-(hydroxymethyl)-2-
methylphenyl)-N0-(3-methoxybenzylidene)acetohydrazide (M3)
To
a
stirred and cooled (ꢀ78 °C) solution of
6 (235 mg,
0.59 mmol) in PhMe (5 mL) was added drop wise DIBALH (1.0 M
in CH2Cl2, 1.19 mL, 1.19 mmol). The resulting solution was stirred
at this temperature for 1 h and then transferred to an ice bath
for 1 h. Na2SO4ꢂ10H2O (4 g) was then added, the cold bath was re-
moved and stirring was continued for 4 h. The resulting slurry was
filtered through a pad of Celite (2 ꢁ 3 cm), which was rinsed with
EtOAc (3 ꢁ 10 mL). Evaporation of the solvent and flash chroma-
tography of the residue over silica gel using EtOAc/hexane 20–
50% gave M3 (150 mg, 68%) as a dark yellow gummy syrup: 1H
NMR (CDCl3, 500 MHz) d ppm 2.14 (s, 3H), 3.83 (s, 3H), 4.78 (s,
2H), 6.96 (dd, J = 8.0, 2.0 Hz, 1H), 7.12 (d, J = 7.5 Hz, 1H), 7.17 (d,
J = 8.0 Hz, 1H), 7.27 (m, 3H), 7.41 (d, J = 8.0 Hz, 1H), 7.46 (s, 1H);
13C NMR (CDCl3, 125 MHz) d ppm 17.14, 55.26, 64.56, 111.67,
117.30, 121.00, 126.34, 129.00, 129.83, 130.17, 131.47, 134.55,
136.90, 143.56, 144.03, 159.94; MS (ESI): m/z calcd for
Hydrazide
3
(439 mg, 1.07 mmol) and iodocyclohexane
(0.55 mL, 4.30 mmol) in DMF (3 mL) was refluxed for 12 h. Reac-
tion mixture was cooled and diluted with water, (10 mL), extracted
with EtOAc (2 ꢁ 20 mL), washed with brine (20 mL), dried
(Na2SO4) and evaporated, resulting solid was recrystallized from
ethanol to give M2 (200 mg, 60%) as brown crystals: mp 181–
183 °C; LCMS purity 97%; 1H NMR (CDCl3, 500 MHz) d ppm 2.07
(s, 3H), 2.40 (s, 3H), 5.50 (br s, OH), 5.68 (br s, OH), 6.83 (d,
J = 8.0 Hz, 1H), 6.93 (dd, J = 8.0, 2.0 Hz, 1H), 7.03 (d, J = 8.0 Hz,
C
18H17F3N2O3 ([M+H]+) 367.1263; found 367.1286 ([M+H]+).

C. Chiruta et al. / Bioorg. Med. Chem. 21 (2013) 2733–2741
2741
8. Maezawa, I.; Hong, H. S.; Wu, H. C.; Battina, S. K.; Rana, S.; Iwamoto, T.; Radke,
G. A.; Pettersson, E.; Martin, G. M.; Hua, D. H.; Jin, L. W. J. Neurochem. 2006, 98,
57.
9. Liu, Y.; Dargusch, R.; Maher, P.; Schubert, D. J. Neurochem. 2008, 105, 1336.
10. Ulgen, M.; Ozer, U.; Kucukguzel, I.; Gorrod, J. W. Drug Metab. Drug Interact.
1997, 14, 83.
Acknowledgements
We thank Professor Edward Roberts at The Scripps Research
Institute for his advice and LCMS analysis. We also thank Wolfgang
Fischer and William Low at Salk Institute for Mass spectral analy-
sis. This work was supported by the Fritz B. Burns Foundation.
11. Ulgen, M.; Durgun, B. B.; Rollas, S.; Gorrod, J. W. Drug Metab. Drug Interact.
1997, 13, 285.
12. Tapia-Arancibia, L.; Aliaga, E.; Silhol, M.; Arancibia, S. Brain Res. Rev. 2008, 59,
201.
References and notes
13. Abe, K.; Takayanagi, M.; Saito, H. Jpn. J. Pharmacol. 1990, 53, 221.
14. McPhie, D. L.; Lee, R. K.; Eckman, C. B.; Olstein, D. H.; Durham, S. P.; Yager, D.;
Younkin, S. G.; Wurtman, R. J.; Neve, R. L. J. Biol. Chem. 1997, 272, 24743.
15. LaFerla, F. M.; Green, K. N.; Oddo, S. Nat. Rev. Neurosci. 2007, 8, 499.
16. Sopher, B. L.; Fukuchi, K.; Smith, A. C.; Leppig, K. A.; Furlong, C. E.; Martin, G. M.
Brain Res. Mol. Brain Res. 1994, 26, 207.
1. Chen, Q.; Prior, M.; Dargusch, R.; Roberts, A.; Riek, R.; Eichmann, C.; Chiruta, C.;
Akaishi, T.; Abe, K.; Maher, P.; Schubert, D. PLoS One 2011, 6, e27865.
2. Kucukguzel, S. G.; Kucukguzel, I.; Ulgen, M. Farmaco 2000, 55, 624.
3. Neumann, H. G. Crit. Rev. Toxicol. 2007, 37, 211.
4. Benigni, R.; Passerini, L. Mutat Res. 2002, 511, 191.
5. Malca-Mor, L.; Stark, A. A. Appl. Environ. Microbiol. 1982, 44, 801.
6. Rollas, S.; Kucukguzel, S. G. Molecules 1910, 2007, 12.
17. Davis, J. B.; Maher, P. Brain Res. 1994, 652, 169.
18. Tan, S.; Schubert, D.; Maher, P. Curr. Top. Med. Chem. 2001, 1, 497.
7. Behl, C.; Davis, J. B.; Lesley, R.; Schubert, D. Cell 1994, 77, 817.