Iminoheterocycles as γ-secretase modulators

Article abstract of DOI:10.1016/j.bmcl.2010.07.121

The synthesis of a novel series of iminoheterocycles and their structure-activity relationship (SAR) as modulators of γ-secretase activity will be detailed. Encouraging SAR generated from a monocyclic core led to a structurally unique bicyclic core. Selected compounds exhibit good potency as γ-secretase modulators, excellent rat pharmacokinetics, and lowering of Aβ₄₂ levels in various in vivo models.

Full text of DOI:10.1016/j.bmcl.2010.07.121

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5380–5384
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Iminoheterocycles as c-secretase modulators
*
John P. Caldwell , Chad E. Bennett, Troy M. McCracken, Robert D. Mazzola, Thomas Bara, Alexei Buevich,
Duane A. Burnett, Inhou Chu, Mary Cohen-Williams, Hubert Josein, Lynn Hyde, Julie Lee, Brian McKittrick,
Lixin Song, Giuseppe Terracina, Johannes Voigt, Lili Zhang, Zhaoning Zhu
Merck Research Laboratories, 2015 Galloping Hill Road, Kenilworth, NJ 07033, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a novel series of iminoheterocycles and their structure–activity relationship (SAR) as
Received 15 June 2010
Revised 26 July 2010
Accepted 27 July 2010
Available online 2 August 2010
modulators of
c
-secretase activity will be detailed. Encouraging SAR generated from a monocyclic core
-secretase mod-
led to a structurally unique bicyclic core. Selected compounds exhibit good potency as
ulators, excellent rat pharmacokinetics, and lowering of Ab42 levels in various in vivo models.
c
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Alzheimer’s disease
c
-Secretase modulators
A widely pursued strategy for the treatment of Alzheimer’s
We focused our study on an imidazyl-phenyl series typified by
compounds disclosed by Neurogenetics and Eisai (Fig. 1). We sought
to replace the aminothiazolyl (Neurogenetics6) and lactam (Eisai7)
portions with iminoheterocycles which have been shown to be brain
penetrant structural motifs.8 Thereby, we could alter the overall
molecular properties while maintaining what were believed to be
disease is inhibiting the production of neurotoxic, insoluble amy-
loid-b (Ab) peptides, especially Ab42. The final step in the generation
of Ab from the amyloid precursor protein (APP) involves cleavage by
c
-secretase.1 Furthermore,
c
-secretase inhibitors (GSIs) have been
shown to reduce the amyloid burden in animal models of Alzhei-
mer’s disease. Unfortunately, GSIs also inhibit cleavage of other
c
-
key pharmocophores for
compound from Eisai is a potent GSM (Fb42 IC50 = 64 nM) with a
good rat PK profile (AUC0–6h = 60.4 M h, brain:plasma concentra-
c-secretase modulation. Moreover, the
secretase substrates, including Notch. It is this inhibition of Notch
processing which is thought to be responsible for adverse events in
clinical trials, particularly gastrointestinal and immunological
toxicities.2
l
tion ratio of 0.4 after oral administration). When dosed orally at
30 mg/kg in a transgenic (CRND8) mouse model9, this compound
lowered Ab42 levels in the cortex and CSF.10–12
In contrast to GSIs, compounds known as c-secretase modula-
tors (GSMs) inhibit Fb42 production without interfering in the pro-
cessing of Notch and other substrates. While the exact reason for
the selectivity of these agents remains unclear, they are thought
Modeling studies suggested integration of monocyclic imino-
heterocycles maintained good overlap with the pharmacophores
of the Neurogenetics and Eisai compounds (Fig. 1) and therefore
were reasonable targets as potential GSMs.
to intervene at an allosteric site on c-secretase and shift the APP
cleavage site so that shorter, soluble, non-toxic peptides (e.g.,
Starting from (S)-1-(4-fluorophenyl)-ethanamine, iminopyrimidi-
none derivatives were prepared as shown in Scheme 1. Compound 1
Ab38) are produced instead of the highly insoluble and neurotoxic
3
Fb42
.
displayed moderate potency and functioned as a modulator of c-
Early accounts of cyclooxygenase (COX) inhibitors showed
promise in the prevention of Alzheimer’s patients. It was later
noted that derivatives of COX inhibitors functioned as modulators
secretase.13 All attempts to form the exocyclic double bond provided
thethermodynamic endocyclicregioisomer.14 The geminal di-methyl
analog, 3, was prepared to avoid double bond migration; however;
predominantly the Z-exocyclic double bond formed and resulted in
of c
-secretase independent of their COX activity.4 Several groups
have since explored this strategy towards the development of no-
vel GSMs.3 However, Flurizan is a weak GSM that failed in clinical
trials testing its ability to treat Alzheimer’s disease.5 Therefore, we
directed our efforts to develop GSMs which were not derived from
the COX inhibitors.
a significant loss of potency (Ab42 IC50 >20 lM).
Concurrently, an iminohydantoin scaffold was developed
originating from (S)-1-(4-fluorophenyl)-ethanamine as shown in
Scheme 2. The chemistry was further improved at the oxidative ami-
nation stage through a two-step POCl3 conversion, followed by
amine displacement. Initial efforts were promising as shown in Ta-
ble 1. Several compounds with sub-micromolar potency were iden-
* Corresponding author. Tel.: +1 908 740 5199.
E-mail address: john.caldwell@merck.com (J.P. Caldwell).
tified exhibiting a c-secretase modulator profile. Furthermore, the
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.07.121

J. P. Caldwell et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5380–5384
5381
O
HO
F
Myriad
(R)-Flurbiprofen
Flurizan (tarenfluribil)
O
S
O
N
NH
N
N
F
N
N
S
N
Neurogenetics
Eisai
Figure 1. Known
c
-secretase modulators.
Table 1
Cellular IC50 of iminohydantoin analogs11
NH2
NBoc
HN
NHBoc
a, b
c, d
HN
N
O
N
O
F
O
F
N
F
N
N
NBoc
N
NH
N
N
R
R
HN
HN
OMe
4 - 17
OMe
18 - 24
N
N
N
O
F
F
e
f, g
R
O
R
OH
Compound
R
Ab42 IC50 (nM)
Abtotal/Ab42
F
F
1
4
5
6
7
8
OH
OCH3
NH2
NHCH3
NHEt
NHcPr
163
864
322
411
220
287
250
258
671
955
218
301
163
287
3502
94
141
23
55
25
97
136
71
78
32
21
79
66
141
70
nd
65
159
68
42
27
153
g
IC50 = 446 nM
Αβ 42
NH
HN
where R =
N
O
N
O
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
NHCH2cPr
NH(CH2)3CH3
Pyrrolidine
Piperidine
NH(CH2)2OH
NH(CH2)2OCH3
NH(CH2)3OH
NH(CH2)3OCH3
OH
NH2
NHEt
NHCH2cPr
NH(CH2)3CH3
NH(CH2)2OH
NH(CH2)3OH
N
R
OH
F
2
IC50 > 20
Αβ42
μM
(Z)
R
NH
NH
O
H2N
O
N
OMe
88
158
92
63
111
F
3
Αβ42 IC50 > 20 μM
Scheme 1. Reagents and conditions: (a) Cl2CS, K2CO3, DCM/H2O, rt, 90%; (b) NaH,
BocNH2 THF, 0 °C to rt, 76%; (c) EDCI, DIEA, NH2(CH2)2CO2 CH3, 70 °C; (d) NaH, THF,
3% over 2 steps; (e) RCHO, LHMDS, 94%; (f) MsCl, DBU, NaH, CH2Cl2, rt, 50%; (g) TFA,
CH2Cl2, rt, 50% for 1, 100% for 2.
An initial venture into preparation of a bicyclic system led to 32
(Scheme 3) with a poor cellular IC50, where Ab42 IC50 = 7.1 M.
l
Discouraged by this initial result, we focused our efforts upon a
different cyclization strategy that introduced conformational re-
straint between the benzylic moiety and the iminoheterocyclic
core. In this manner, 5-5, 5-6 and 5-7 bicyclic frameworks were
investigated as shown in Scheme 4. The key synthetic transforma-
tion relied on a latent amine being revealed either by deprotection
in the case of the 5-5 system or functional group manipulation in
the case of the 5-6 and 5-7 systems. Thus, when subjected to con-
ditions described in the preparation of the monocyclic series, the
amine readily displaced the activated thiohydantoin precursor.
The compounds were tested as a racemic mixture unless other-
wise specified (Table 3). In general, the ethyl substituent showed
better potency than the unsubstituted bicycles in direct compari-
sons (35 to 36, and 37 to 38). Additionally, the potency appears
to drop off slightly with an increase in ring size (33 vs 37).
R-enantiomer of the a-methyl benzyl was explored. Interestingly,
the R-enantiomer displayed higher potency when compared to its
corresponding S-enantiomer. Particularly noteworthy were com-
pounds 19, 20, 22, and 23 which displayed double-digit nanomolar
potency while still retaining the -secretase modulator profile.
c
Next, the effect of substitution of the benzylic position was ex-
plored as shown in Table 2. Compounds 25–31 were prepared
using similar intermediates detailed in Scheme 4. In the parent
iminohydantoin (R1 = NH2), the potency appears better with
methyl analog, 19, compared to 25, 28, and 30. Yet, in the ethyl
substituted iminohydantoin (R1 = NHEt), the unsubstituted ben-
zylic variant, 26, shows slightly increased potency compared to
the substituted analogs, 20, 29, and 31.

5382
J. P. Caldwell et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5380–5384
MeO
O
H2N
MeO
HN
O
a
b
+
HN
N
S
NCS
F
F
CHO
c
O
N
OMe
N
HN
S
F
O
O
e
N
S
N
HN
d
N
N
N
Cl
OMe
OMe
N
N
F
F
O
f
N
N
N
R
OMe
4 -17
N
F
Scheme 2. Reagents and conditions: (a) THF, rt, 100%; (b) NaH, THF, 100%; (c) piperidine, EtOH reflux, then PS-TsNHNH2, rt 95%; (d) Base, t-BuO2H, MeOH, rt, 10–20%; (e)
POCl3, microwave 170 °C; (f) amine, Et3N, THF, microwave 140 °C, 10–62% over 2 steps (e and f).
Table 2
Table 3
Cellular IC50 of iminohydantoin analogs11
Cellular IC50 of bicyclic analogs11
O
F
R2
O
N
N
N
N
R1
N
OMe
25 - 31
N
n
N
N
N
R
F
OMe
33 - 39
Compound
R1
R2
Ab42 IC50 (nM)
Abtotal/Ab42
25
26
27
28
29
30
31
NH2
NHEt
NHBu
NH2
NHEt
NH2
H
H
H
144
72
85
139
91
229
171
139
104
70
113
200
87
Compound
n
R
Ab42 IC50 (nM)
Abtotal/Ab42
33
34
35
36
37
38
39
1
1
2
2
3
3
3
H
Et
H
Et
H
Et
Et
rac
(-)
135
56
106
85
503
117
73
117
159
188
71
40
169
215
CH2OH
CH2OH
(CH2)2OH
(CH2)2OH
(+)
(+)
rac
rac
(-)
NHEt
117
O
MsCl
16
N
N
plasma Ab42 levels 80% while showing minimal (À10%) effect on
plasma Abtotal levels. It also had a minimal effect (À12%) on cortical
Ab42 levels. Despite significant exposure in both plasma and brain,
compound 27 had minimal effect on Ab42 in the CNS in rats, low-
ering CSF Ab42 levels 10% and 38% at 30 and 100 mg/kg orally,
respectively. Echoing these results, compound 39 achieved a very
low 8% reduction of CSF Ab42 despite high brain concentrations.
The Ab42 samples were taken 3 h after dosing in all of the studies.
Correlation of the PK and efficacy experiments was confirmed
by determination of compound exposure in the plasma and the
brain in the actual animals used in the efficacy experiments. For
compound 27 dosed at 30 mg/kg, plasma and brain levels at 3 h
N
N
N
OMe
F
32
Scheme 3. Preparation of a bicyclic analog via N-1 to N-2.
Compounds 20, 27, and 39 from both the monocyclic and bicy-
clic series were selected for further profiling. The selected com-
pounds displayed favorable pharmacokinetic profiles,15 with high
exposure in the plasma and acceptable levels distributed in the
brain at 6 h as shown in Table 4. In an in vivo transgenic (CRND8)
mouse study11 dosed at 30 mg/kg orally, compound 20 lowered
were 7.1
lM and 5.9 lM, respectively. Furthermore, compound
27 was shown not to be a Pgp substrate.

J. P. Caldwell et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5380–5384
5383
F
F
F
H2N
BocHN
c, d
e,f,g
BnHN
H2N
a, b
HN
OHC
O
NHBoc
F
N
S
h,i, j
k
33
34
N
OMe
N
OTBDPS
n=1,2
O
F
H2N
l,e,f,g
N
HN
N
HO
n=1,2
S
OMe
OH
N
F
O
n=1
n=1,2
35 or 36
n,o
N
h,m
N
n=2
37, 38, or 39
N
SCH3
OMe
N
F
Scheme 4. Reagents and conditions: (a) BnNH2–HCl, NaCN, H2O MeOH, 94%; (b) LAH, Et2O, 56%; (c) Boc2O, CH2Cl2, 86%; (d) Pd(OH)2/C, THF, MeOH 94%; (e) methyl
2-isothiocyanatoacetate, THF, 0 °C, 100%; (f) NaH, THF, 89%; (g) 3-methoxy-4-(4-methyl-1H-imidazol-1-yl)benzaldehyde, piperidine, EtOH reflux, rt 87%; (h) NaH, MeI,
DMF, 86%; (i) TFA, CH2Cl2; (j) Et3N, THF, microwave, 68–72%; (k) NaH, EtI, DMF, 43%; (l) TBDMSCl, imid, DMF, 99%; (m) TBAF, HOAc, THF; (n) MsCl, DMAP, Et3N, DCE; (o) amine,
80 °C, 17–62% over 2 steps.
4. (a) Weggen, S.; Eriksen, J. L.; Das, P.; Sagi, S. A.; Wang, R.; Pietrzik, C. U.; Findlay,
K. A.; Smith, T. E.; Murphy, M. P.; Bulter, T.; Kang, D. E.; Marquez-Sterling, N.;
Golde, T. E.; Koo, E. H. Nature 2001, 414, 212; (b) Weggen, S.; Eriksen, J. L.; Sagi,
S. A.; Pietrzik, C. U.; Ozols, V.; Fauq, A.; Golde, T. E.; Koo, E. H. J. Biol. Chem. 2003,
278, 31831; (c) Takahashi, Y.; Hayashi, I.; Tominari, Y.; Rikimaru, K.; Morohashi,
Y.; Kan, T.; Natsugari, H.; Fukuyama, T.; Tomita, T.; Iwatsubo, T. J. Biol. Chem.
2003, 278, 18664.
Table 4
Profiles of selected compounds, dosed at 10 mg/kg orally
Compound
AUC0–6h
M h)
Plasma
Cmax
Brain:plasma
concentration ratio @6 h
BrainC6h
(ng/g)
(
l
(l
M)
20
27
39
10.1
10.0
16.9
2.3 @ 2 h
2.3 @ 2 h
4.2 @ 1 h
1.3
0.7
0.3
914
449
275
5. (a) Myriad Genetics Inc. press release: June 30, 2008.; (b) Imbimbo, B. P. J.
Alzheimers Dis. 2009, 17, 757.
6. Cheng, S.; Comer, Daniel D.; Mao, L.; Balow, G. P.; Pleynet, D. U.S. Patent
7244,739 B2, July 17, 2007.
7. (a) Kimura, T.; Kawano, K.; Doi, E.; Kitazawa, N.; Takaishi, M.; Ito, K.; Kaneko, T.;
Sasaki, T.; Sato, N.; Miyagawa, T.; Hagiwara, H.; WO 2005115990, 2005.; (b)
Kushida, I.; Eriko, I WO 2006046575 A1, 2006.
8. Zhu, Z.; Sun, Z.-Y.; Ye, Y.; Voigt, J.; Strickland, C.; Smith, E. M.; Cumming, J.;
Wang, L.; Wong, J.; Wang, Y-S.; Wyss, D. F.; Chen, X.; Kuvelkar, R.; Kennedy, M.
E.; Favreau, L.; Parker, E.; McKittrick, B. A.; Stamford, A.; Czarniecki, M.;
Greenlee, W.; Hunter, J. C. J. Med. Chem. 2010, 53, 951.
9. Janus, C.; Pearson, J.; McLaurin, J.; Mathews, P. M.; Jiang, Y.; Schmidt, S. D.;
Chishti, M. A.; Horne, P.; Heslin, D.; French, J.; Mount, H. T. J.; Nixon, R. A.;
Mercken, M.; Bergeron, C.; Fraser, P. E.; St. George-Hyslop, P.; Westaway, D.
Nature 2000, 408, 979.
In summary, the use of the structurally distinct iminoheterocy-
cle core permitted rapid SAR exploration and provided a novel ser-
ies of small-molecule c-secretase modulators. Several compounds
in this series possess double-digit nanomolar potency in the cell-
based assay. Surprisingly, although several compounds exhibited
excellent pharmacokinetic profiles with high brain levels, only a
modest decrease in Ab42 levels in the cortex and CSF was realized.
Additional work in the development of small-molecule
modulators will be presented in due course.
c-secretase
10. Unpublished results.
11. Cell based assay: Human embryonic kidney (HEK) 293 cells stably transfected
with APP carrying both Swedish and London mutations in pcDNA3.1 vector
(Invitrogen) were treated with GSM compounds for 5 h. Ab in conditioned
media was measured using MesoScale Discovery (MSD) technology-based
sandwich immunoassays. Ab42 was measured using a pair of labeled antibodies
TAG-G2-11 and biotin-4G8; Ab40 was measured using an antibody pair of TAG-
G2-10 and biotin-4G8; total Ab was measured using TAG-W02 and biotin-4G8.
CSF and cortex Ab assays: Male CD rats (ꢀ120 g; Crl:CD(SD); Charles River
Laboratories, Kingston, NY) were used for all experiments. Compounds were
formulated in 20% hydroxypropyl-b-cyclodextrin and administered orally to
male rats at a dosing volume of 5 ml/kg. Three hours after drug administration,
cerebrospinal fluid (CSF) was collected from the cisterna magna and the brain
was removed immediately following euthanasia with excess CO2, quickly
frozen on dry ice and stored at À70 °C until Ab analysis. Only visibly clear CSF
samples were included. Rat CSF Ab40 and Ab42 was analyzed using AlphaLISA
Amyloid Assay kits (Perkin–Elmer) according to the manufacturer’s
instructions. For brain cortex Ab42 analysis, half of the cortex was
homogenized and extracted in 5 M guanidine–HCl/50 mM Tris–HCl, pH 8.
Acknowledgments
We thank Dr. William Greenlee, Dr. Eric Parker, and Dr. John
Hunter for their support and guidance of this work and Dr. Jesse
Wong for preparation of intermediates.
References and notes
1. Hardy, J.; Selkoe, D. J. Science 2002, 297, 353.
2. (a) Maillard, I.; Adler, S. H.; Pear, W. S. Immunity 2003, 19, 781; (b) Stanger, B.
Z.; Datar, R.; Murtaught, L. C.; Melton, D. A. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
12443; (c) Beher, D. Curr. Top. Med. Chem. 2008, 8, 34.
3. (a) Pissarnitski, D. Curr. Opin. Drug Discovery Dev. 2007, 10, 392; (b) Stanton, M.
G.; Hubbs, J.; Sloman, D.; Hamblett, C.; Andrade, P.; Angagaw, M.; Bi, G.; Black,
R. M.; Crispino, J.; Cruz, J. C.; Fan, E.; Farris, G.; Hughes, B. L.; Kenific, C. M.;
Middleton, R. E.; Nikov, G.; Sajonz, P.; Shah, S.; Shomer, N.; Szewczak, A. A.;
Tanga, F.; Tudge, M. T.; Shearman, M.; Munoz, B. Bioorg. Med. Chem. Lett. 2010,
20, 755.
The extracts were sonicated and partially purified using
a solid phase
extraction matrix in 96-well format, the HLB plate (Waters). The samples
eluted from the HLB plate were dried and resuspended in freshly prepared PBS/
0.5% Tween 20. Ab40 and Ab were measured using AlphaLISA amyloid assay
kits.

5384
J. P. Caldwell et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5380–5384
12. Hyde, L. A.; McHugh, N. A.; Chen, J.; Zhang, Q.; Manfra, D.; Nomeir, A. A.; Josien,
H.; Bara, T.; Clader, J. W.; Zhang, L.; Parker, E. M.; Higgins, G. A. J. Pharmacol.
Exp. Ther. 2006, 1133.
the production of longer species like Ab42, total Ab production is unchanged
while Ab42 production is reduced. Thus, GSMs increase the Abtotal/Ab42 ratio.
14. Conditions: (a) DBU, NaH, MsCl, CH2Cl2, rt; (b) DIEA, MsCl, rt then THF, LHMDS
À40 °C to 0 °C.
13. The ratio of Abtotal/Ab42 represents the extent of
ratio of Abtotal/Ab42 = 1 indicates an inhibitor of
indicates a modulator of -secretase. Since GSMs shift the cleavage specificity
of the enzyme, increasing the production of shorter Ab peptides while reducing
c
-secretase modulation. A
c
-secretase, while a ratio >5
15. Korfmacher, W. A.; Cox, K. A.; Ng, K. J.; Veals, J.; Hsien, Y.; Wainhaus, S.; Broske,
L.; Prelusky, D.; Nomeir, A.; White, R. E. Rapid Commun. Mass Spectrom. 2001,
15, 335.
c