Design and synthesis of bicyclic heterocycles as potent γ-secretase modulators

Article abstract of DOI:10.1016/j.bmcl.2013.06.100

The evolution of amide 3 into conformationally restricted bicyclic triazolo-piperidine 14-S as a γ-secretase modulator is described. This is a potential disease modifying anti-Alzheimer's drug which demonstrated high in vitro and in vivo potency against Aβ42 peptide, reduced lipophilicity and enhanced brain free fraction compared to the previous series.

Full text of DOI:10.1016/j.bmcl.2013.06.100

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of bicyclic heterocycles as potent
modulators
c-secretase
a
a
a
Daniel Oehlrich a, , Frederik J. R. Rombouts , Didier Berthelot , François P. Bischoff ,
Michel A. J. De Cleyn a, Libuse Jaroskova a, Gregor Macdonald a, Marc Mercken a, Michel Surkyn a,
Andrés A. Trabanco b, Gary Tresadern a, Sven Van Brandt a, Adriana I. Velter a, Tongfei Wu a,
⇑
Harrie J. M. Gijsen a,
⇑
a Neuroscience, Janssen Research & Development, Turnhoutseweg 30, 2340 Beerse, Belgium
b Neuroscience, Janssen Research & Development, Jarama 75, 45007 Toledo, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The evolution of amide 3 into conformationally restricted bicyclic triazolo-piperidine 14-S as a c-secre-
Received 18 June 2013
Revised 28 June 2013
Accepted 29 June 2013
Available online xxxx
tase modulator is described. This is a potential disease modifying anti-Alzheimer’s drug which demon-
strated high in vitro and in vivo potency against Ab42 peptide, reduced lipophilicity and enhanced
brain free fraction compared to the previous series.
Ó 2013 Published by Elsevier Ltd.
Keywords:
Alzheimer’s disease
c
-Secretase modulators
O
Alzheimer’s disease (AD) represents one of the principal unmet
medical needs, being the sixth highest cause of death in the US.1
Presently 30 million people worldwide suffer from AD and this
population is predicted to quadruple by 2050. As such this repre-
sents a huge social and economic burden.2 Patients suffering from
the disease usually show a gradual decline in their cognitive capa-
bilities, motor and vocal control, which continues with a loss in
function of the body and concludes in death. Genetic evidence sup-
ports a hypothesis that AD is the result of the formation of Ab pla-
ques. These are created by the oligomerisation of the Ab peptides,
which then instigate the formation of neurofibrillary tangles and
ultimately lead to neuronal death.3
N
N
N
N
F
O
O
N
N
E-2012
F
O
H
N
N
H
N
N
F
N
N
O
N
2, IC50 = 0.016
M
1, IC50= 0.55
M
μ
μ
LE=0.2
The formation of Ab is a result of the sequential enzymatic
cleavage of amyloid precursor protein (APP) initially by aspartyl
Scheme 1. Ref. 11.
protease b-secretase 1 (BACE1) and then by c-secretase, which re-
sults in the formation of amyloid peptides of differing lengths. Of
these, Ab42 is the most prone to oligomerisation, and is reported
to initiate plaque formation.4 Post-mortem studies have shown
Ab42 is disproportionately abundant in the amyloid plaques.4
Due to the biochemical relationship between these enzymes and
and
c-secretase inhibitors (GSIs) and more recently, c-secretase
modulators (GSMs).6–8 GSMs act by reducing the production of
the toxic Ab42 peptides and biasing the APP processing towards
shorter length Ab peptides. Furthermore, GSMs function in such a
way that they do not impede the processing of the other c-secre-
tase substrates, for example Notch, thus resulting in a more opti-
mal phenotype when compared to the traditional GSIs.
We recently reported the discovery of novel bicyclic heterocy-
cles as potent GSMs derived from conformationally restricted ami-
no-pyridones 1 which in turn were derived from rigidified amides
as for example E-2012 (Scheme 1).9 This investigation culminated
in the discovery of benzimidazole 2, which showed robust in vivo
the onset of AD, BACE1 and
c-secretase have been the target of
concentrated research efforts to inhibit the APP processing path-
way and thereby prevent pathogenesis and disease progression.5
This has manifested in increasing reports of optimized BACE1
⇑
Corresponding authors.
E-mail addresses: doehlric@its.jnj.com (D. Oehlrich), hgijsen@its.jnj.com (H.J.M.
Gijsen).
0960-894X/$ - see front matter Ó 2013 Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.bmcl.2013.06.100
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

2
D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
peptide was observed 4 h after administration,14 despite levels of
2.7 M being present in the brain.
This result prompted further optimization, with the aim to first
O
O
S
l
N
H
N
H
N
N
N
F
N
increase potency and ultimately brain penetration. In a first explo-
O
O
ration, amide 3 was cyclized into a triazole ring 5. This led to a two-
fold increase in potency (Scheme 3, Compound 5, IC50 = 1.29 lM).
3, IC50 = 2.6
LE = 0.2
M
μ
4, IC50 = 0.158
M
μ
Assuming the H-bond donor in triazole 5 could limit central
penetration, we methylated the triazole nitrogen and obtained a
mixture of regioisomers 6 and 7 (Scheme 4). Regio-isomer 6, was
clearly more potent than 7, and threefold more potent than the
non-methylated parent compound 5. Conformational analysis
(Fig. 1 and Supplementary data) suggested the position of the
methyl group in 7 was detrimental for activity compared to 6 for
two possible reasons. Firstly, the methyl in 7 disturbs the planarity
between the triazole and the 4-imidazophenyl ring, the dihedral
angle being 45 degrees in 7 compared to 0.5 degrees in 6. Other
literature accounts around similar GSMs suggest torsion and
reduction in planarity in this region of the molecule is detrimental
for activity.15,5 It is expected that the acceptor atom (N-4 of the tri-
azole in this case) has to be co-planar with the 4-imidazophenyl
ring. Secondly, we have previously discussed the importance of
having a twisted or out-of-plane orientation for the distal aromatic
group.9 Indeed, molecule 6 exhibits a twisted orientation of this 4-
fluorophenyl ring, whereas for 7 it is planar with respect to the tri-
azole. An analog of 6 having the position of the 4-fluorophenyl and
methyl group on the triazole reversed also showed similar potency
(Scheme 4, compound 8).
Given the perceived importance of conformational effects, we
targeted molecules with structural motifs which induced low en-
ergy conformations with the aforementioned characteristics:
namely co-planarity between triazole and 4-imidazophenyl ring
whilst a non-planar arrangement between triazole and distal phe-
nyl. Hence, a methylene spacer was introduced between the tria-
zole and the 4-fluorophenyl group (Scheme 5, compound 9).
Unfortunately, this did not result in the anticipated improvement
of potency, which was attributed to the conformational freedom
of the benzyl group in 9. To rigidify the structure and optimize
the orientation of the 4-fluorophenyl substituent, a 6-membered
ring was formed by interconnecting the N-methyl group and ben-
zylic position. The resulting compound 10 displayed a significant
cLogP = 5.2
cLogP = 3.8
Mouse 30 mpk PO t = 4 h
Plasma: 11.9
M
μ
Brain: 2.7 M (B/P 0.22)
μ
Reduction A 42: 12 %
β
Reduction total A : 3 %
β
Scheme 2.
O
N
N
S
N
H
F
N
H
N
F
N
N
N
O
O
3, IC50 = 2.6
M
μ
5, IC50 = 1.29 M
μ
Scheme 3.
lowering of Ab42 and Ab40, with the characteristic increase in con-
centration of the shorter peptides, for example, Ab37 and Ab38.
Unfortunately, this benzimidazole series suffered from high LogP
and low solubility, which manifested in signs of liver toxicity.10
In parallel we had explored alternative chemical classes, includ-
ing benzamides represented by 3.12 Herein, we disclose the evolu-
tion of the amide series into more sophisticated and optimized
bicyclic triazoles represented by 14-S.13
In an effort to identify novel hit series, we performed a
Hit-Hopping exercise starting with compound 1 with the specific
aim to remove unwanted hydrogen bond donors (HBDs) and
acceptors (HBAs); this resulted in the identification of simplified
amides represented by 3 (Scheme 2). The prototype 3 showed a
ꢀ5 fold decrease in activity when compared to 1 (Scheme 1), while
maintaining the ligand efficiency (LE = approx 0.2). Further optimi-
zation of 3 by the introduction of the 3-phenylbenzylamine substi-
improvement in potency (IC50 = 0.22
hypothesis.
lM), consistent with our
tuent led to the potent compound 4 (IC50 = 0.158 lM), however
this potency increase was achieved at the expense of increasing
lipophilicity (4, cLogP = 5.2 vs 3, cLogP = 3.8). In mice, after an oral
dose of 30 mg/kg, the brain/plasma (B/P) ratio for compound 4 re-
mained limited (B/P 0.22) and no significant reduction of Ab42
Based on our previous experience with GSMs of similar moder-
ate potency, we anticipated that the in vitro activity of 10 would
not translate into a significant reduction of Ab42 in vivo. Hence
F
F
4
N
N
N
N
NH
N
N
F
N
N
N
N
N
N
N
N
O
O
O
6, IC50 = 0.62 M
μ
8, IC50 = 0.77
M
μ
5, IC50 = 1.65
M
μ
F
N
N
N
N
N
O
7, IC50 = 5.75
M
μ
Scheme 4.
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
Table 1
R'
R
X
N
N
N
Y
N
N
O
Compound
R
R0
X
Y
Potency IC50 (lM)
10
11
H
F
F
F
F
H
H
H
CH2
CH2
CH2
CH2
CH2
O
C
C
C
C
C
C
N
0.244
0.093
0.040
0.447
0.024
0.071
0.028
Me
Me
Me
CF3
CF3
CF3
11-S
11-R
12-S
13-S
14-S
CH2
Figure 1. Low energy 3D conformation of molecules 6 and 7. The lowest energy
conformations16 are shown which maintain an analogous orientation of the
triazole. Compound 6 is only 0.001 kcal/mol above the global minimum; compound
7 is 0.2 kcal/mol above global minimum. See Supplementary data for more details.
Figure 2. Lowest energy 3D conformation of molecule 11-S. See Supplementary
data for more details.
we continued our research efforts toward more potent analogs.
Gratifyingly, a boost in potency was achieved with the introduction
of an ortho-substituent on the 4-fluorophenyl group giving 11
(IC50 = 0.093 lM, racemate). Again, this was possibly due to subtle
the distal phenyl group, in this case promoted by the ortho methyl
substituent.
Conversion of the ortho-methyl to a trifluoromethyl group gave
compound 12-S which resulted in a further increase in potency.
This also resulted in an unfavorable rise in cLogP and considerable
conformational effects in this region of the molecule. The enantio-
mers of 11 were separated by chiral SFC to provide 11-R and 11-S.
Their absolute configuration was determined via VCD (see Supple-
mentary data). Interestingly, 11-R and 11-S display a clear differ-
ence in potency of 0.447 and 0.040
Advantageously, the hERG binding opposed the activity of the
two enantiomers, with 11-S having an IC50 of 7.9 M while 11-R
had an IC50 0.794 M. The minimum energy conformer of 11-S is
shown in Figure 2 and displays many of the features anticipated
to be beneficial for GSM activity, such as planarity between the tri-
azole and the 4-imidazophenyl ring, and the out-of-plane twist for
lM, respectively (Table 1).
inhibition of the CYP 3A4 isoform (Table 2, IC50 = 0.1 lM). To fur-
ther improve the physicochemical properties, in particular the
lipophilicity and remove potential sites of metabolism, compound
13-S was synthesized, in which the bicyclic piperidine ring was
converted to a morpholine. This resulted in a slight drop in po-
tency, but did have the desired effect on cLogP. (Tables 1 and 2).
The decrease in potency suggests an added value created by the
l
l
F
F
F
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
O
O
O
10, IC50 = 0.22 M
μ
6, IC50 = 0.62
M
μ
9, IC50 = 0.85
M
μ
Scheme 5.
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

4
D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Table 2
Profile of compounds 11-S to 14-S
11-S
12-S
13-S
14-S
IC50
hLMa
mLMa
(
lM)
0.040
41
9
0.024
13
7
0.071
5
6
0.028
10
6
CYP450 IC50. (lM)
3A4
2C19
2C9
2D6
1A2
hHerg IC50
6.94
9.24
5.66
>10
>10
7.94
28%
0.1
1.10
1.52
8.43
>10
>10
>10
12%
ND
>10
>10
8.51
>10
>10
>10
25%
8.5%
2.2%
>10
4.64
>10
>10
>10
11%
ND
b
(
lM)
hERG Patch-Xpress inh. at 3
l
Mc
Fu hPlasmad
2.7%
0.92%
Fu Braine
ND
ND
Concd Mouse Plasma 30 mg/kg, 4 h
Concd Mouse Brain 30 mg/kg, 4 h
B/P ratio mouse 4 h
9.8
4.8
0.50
49%
59%
4.6
l
l
M
M
9.7
l
l
M
M
3.6
l
l
M
M
12.3 lM
2.8 lM
0.23
62%
66%
4.5
12.7
1.29
46%
44%
4.9
3.36
0.93
22%
17%
3.4
Mouse Ab42 lowering, 30 mg/kg, 4 hf
Mouse Ab38 increase, 30 mg/kg, 4 hf
cLogP
a
b
c
d
e
f
% Metabolized after 15 min upon incubation with human (hLM) and mouse (mLM) liver microsomes at 1
Receptor binding on the human Herg ion channel.
% inhibition of IKr current at 3 lM.
lM concentration.
Unbound fraction to human plasma proteins.
Unbound fraction to rat brain tissue.
% Inhibition of Ab peptides as determined by ELISA.14
Table 3
Dog profile of compound 14-S
hydrogens on the phenethylic carbon, perhaps promoting more of
a twisted conformation of the distal phenyl. In an attempt to
positively modulate the lipophilicity while maintaining potency,
we envisaged that the replacement of the methoxy-phenyl ring
by a methoxy-pyridyl would have beneficial effects. Thus, as we
previously observed during our optimization towards
Dog plasma 20 mg/kg @ 8 h
CSF Ab42 lowering 20 mg/kg @ 8 h
CSF Ab38 increase 20 mg/kg @ 8 h
8670 ng/mL, 19 lM
50%
28%
"CO" 20 bar,
O
O
Pd(OAc)2,
NH2R,
HBTU
Br
R
OH
N
H
Ph2P(CH2)3PPh2
N
N
N
N
N
N
THF/H2O, 150 °C,
24 h
O
O
O
DMF, rt, 1 h
15
16
87 %
3: R =
4: R =
42 %
15 %
F
Ph
Scheme 6. Synthesis of benzamides 3 and 4.
O
NH2
N
H
N
NH
H
N
N
N
N
F
19
F
N
K2CO3, n-BuOH,
uW 120 °C, 12 h
K2CO3, DMSO,
120 °C, 1 h
F
N
N
N
N
O
O
O
17
18
5
62 %
50 %
MeI, NaH,
DMF, rt, 2 h
N
N
N
N
F
F
N
N
+
N
N
N
O
N
O
7 (22%)
Scheme 7. Synthesis of triazoles 5–7.
6 (24%)
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
5
O
O
O
(1) HCl/dioxane, rt, 30 min
(2) (COCl)2, DMF, DCM,
N2H4.H2O, EtOH,
130 °C, 32 h
(sealed tube)
NH2
O H
O
N
H
reflux, 1h
N
N
N
N
N
N
O
O
O
89 %
99 %
16
20
21
CH3CN,
n-BuOH,
150 °C, 3 h
sealed tube
56 %
H
4-fluoroiodobenzene,
CuI,
F
H2N
H2N
N
N
N
N
K3PO4, DMF,
150 °C, 72 h
N
N
N
N
N
14 %
N
O
O
8
22
Scheme 8. Synthesis of triazole 8.
F
F
N
F
N
O
N
N
NH
N
NH2
K2CO3, n-BuOH
150 °C, 14 h
MeI, NaH,
N
N
H
DMF, rt, 2 h
N
N
N
N
N
N
O
O
O
Quant.
26%
21
23
Scheme 9. Synthesis of triazole 9.
9
O
S
C
25
O
N
Br
NH2OH.HCl, DIPEA,
MeOH.HCl
Br
Br
H2
dioxane, 12 h, rt
N
N
O
S
H2
N
N
O
N
H
N
H
N
N
N
99 %
96 %
R
24
27
26
OH
B
OH
50-70 %
Dioxane,
NaHCO3 aq,
Pd(PPh3)4,
150 °C, 10 min, MW
MeOH, Pd/C,
R
R
R
NaNO2, H2O,
40 % HBr, CuBr
6 N HCl/iPrOH,
50 °C, 16 h
N
N
N
N
N
H2
N
H2
N
35-70 %
30-80 %
Br
N
N
N
N
28a R = 4-F
28b R = 4F, 2Me
28c R = 2CF3
29a R = 4-F
29b R = 4F, 2Me
29c R = 2CF3
30a R = 4-F
30b R = 4F, 2Me
30c R = 2CF3
Scheme 10. General synthesis of 5,6,7,8-tetrahydro-[1,2,4]-triazolo[1,5-a]pyridine 30a–c.
benzimidazole 2,9 compound 14-S was synthesized, which was
found to be equipotent to 12-S while slightly decreasing the cLogP
(Table 2, Compound 12-S, cLogP = 4.9; 14-S, cLogP = 4.5).
exhibited an increased free fraction in brain of 2.2%. In addition
to the improved in vitro profile, compound 14-S demonstrated en-
hanced in vivo efficacy with a Ab42 reduction of 62%. This was ob-
When tested orally in mice at 30 mg/kg, 11-S displayed the ex-
pected profile of a GSM with a robust reduction of 49% in Ab42 and
59% increase in Ab38 and no effect on total concentration of Ab at
4 h post dosing (Table 2). Bioanalysis showed this effect occurred
served
administration of 30 mg/kg dose. The concentration in the brain
at this 4 h time point was 2.8 M. Unfortunately the B/P ratio
4
hours post dosing in mice following an oral
l
had been negatively affected being 0.23.
at a brain concentration of 4.8
l
M (B/P = 0.5) and a compound free
Compound 14-S was further profiled in the dog (Table 3). An
oral dose of 20 mg/kg resulted in a reassuring reduction of 50% in
Ab42 in the cerebrospinal fluid (CSF). The canine model was also
used to screen for early effects on liver function.10 Compound
fraction in brain of 0.92%. The in vitro profile of 11-S showed no
major flags apart from mild inhibition of the CYP 2C9 and 3A4 iso-
forms. These issues were resolved in compound 14-S, which also
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

6
D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
O
32
CF3
OH
O
O
nBuLi, THF,
Br
Br
Br
N
O
O
O
-78 °C, 20 min
N
N
NH
CH3CN, DIPEA,
90 °C, 3 hrs
CF3
O
N
N
N
N
Br
Br
N
34
Br
31
33
77 %
H
1, MeOH, TsOH
2, TsOH, Toluene
Dean-Stark, reflux
58 %
67% over 2 steps
CF3
O
N
N
N
Br
35
Scheme 11. Synthesis of [1,2,4]triazolo[5,1-c][1,4]oxazine 35.
O
O
B B
to be used as an imidazolium salt for the esterification, hence 16
was treated with HCl in 1,4-dioxane and evaporated prior to the
esterification. Ethyl ester 20 was treated with hydrazine hydrate
to provide hydrazide 21. Condensation of 21 with acetonitrile
yielded triazole 22. Finally a copper-catalyzed arylation of 22 with
4-fluoroiodobenzene provided 8 in 14% yield.
O
O
O
B
Br PdCl2(dppf), DMF
KOAc, 150 °C, MW
O
X
X
N
N
N
N
O
O
36a X = CH
36b X = N
37a X = CH
37b X = N
Benzyl analog 9, was synthesized by reaction of hydrazide inter-
mediate 21 with 2-(4-fluorophenyl)acetonitrile under basic condi-
tions (Scheme 9). The resulting triazole 23 was then alkylated with
MeI to afford 9.
Scheme 12. Formation of pinacol boronate ester 37.
The triazolopyridine core 27 (Scheme 10) was prepared by
adding 2-amino-3-bromo-pyridine 24 to isothiocyanate 25 to
give 26, followed by condensation with hydroxylamine to afford
bromo derivative 27. Suzuki–Miyaura cross coupling of 27 to 28
with an appropriately decorated boronic acid, followed by hydro-
genation afforded the corresponding bicyclic triazolo-piperidines
29a–c. Conversion of the amino-triazoles 29a–c to the corre-
sponding bromo analog 30a–c was achieved using Sandmeyer
conditions.
To form 35, 3,5-dibromotriazole 31 was alkylated under basic
conditions with 2-(2-bromoethoxy)tetrahydropyran 32 giving 33.
Regioselective lithium halogen exchange was achieved by treat-
ment of 33 with n-BuLi at ꢁ78 °C, which was then quenched by
addition of the appropriate aldehyde to afford secondary alcohol
34. Deprotection followed by condensation using a Dean–Stark
apparatus resulted in bicyclic triazolomorpholine 35 (Scheme 11).
Boronate esters 37a and 37b were prepared from the previously
published bromo-analogs 36a and 36b (Scheme 12).9
14-S demonstrated clear long-lasting activity, which was seen
without increases in the liver function biomarkers bilirubin and
alanine transaminase (ALT), despite high compound concentra-
tions/levels. This is considered a clear improvement over GSMs
from older series, such as 1 and 2, which in the same model had
shown elevations on bilirubin and/or liver enzymes already at
exposure levels below 10 l
M.10
Chemistry. Compounds 3 and 4 were synthesized from previ-
ously reported bromide 159 via Pd catalyzed CO insertion in THF/
water to provide acid 16 (Scheme 6). Peptide coupling of 16 with
(1S)-1-(4-fluorophenyl)ethanamine or 3-phenylbenzylamine gave
amides 3 and 4, respectively, in unoptimised yields.
For the synthesis of triazoles 5, 6 and 7, we first substituted 4-
fluoro-3-methoxybenzonitrile 17 with 4-methylimidazole to af-
ford intermediate 18 (Scheme 7). Nitrile 18 was condensed with
hydrazide 19 to yield triazole 5. Finally, alkylation of 5 with MeI
provided the regioisomeric triazoles 6 and 7 in a 1:1 ratio.
Triazole 8, was prepared starting from acid 16 (Scheme 8). First,
16 was esterified to 20. In order to suppress side reactions, 16 had
Suzuki–Miyaura cross coupling reactions between boronate
esters 37a and 37b and bromo-derivatives (30a–c, 35) under
R
R
O
N
N
B
Y
O
X
Y
MeO
N
N
N
N
N
Pd(PPh3)4,
N
Br
dioxane, NaHCO aq,
X
150 °C, 15 min,3MW
N
N
O
27-75 %
30a Y = CH2, R = 4-F
10 X = CH, Y = CH2, R = 4-F
11-S X = CH, Y = CH2, R = 4-F, 2-Me
12-S X= CH, Y = CH2, R = 2-CF3
13-S X = CH, Y = O, R = 2-CF3
14-S X = N, Y = CH2, R = 2-CF3
30b Y = CH2, R = 4-F, 2-Me
30c Y = CH2, R = 2-CF3
35 Y = O, R = 2-CF3
Scheme 13. Synthesis of final compounds 10–14-S.
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100

D. Oehlrich et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
7
2. (http://www.alzinfo.org/08/alzheimers/world-alzheimers-day)
alzheimers.about.com/od/financialissues/a/Costs_Alzheimer.htm).
(http://
microwave irradiation afforded compounds (10–14) as racemates.
The racemic compounds 11–14 were separated by chiral SFC
purification to provide the desired enantiomers (11-S–14-S)
(Scheme 13).
3. (a) Lahiri, D. K.; Ghosh, C.; Ge, Y. W. Ann N.Y. Acad. Sci. 2003, 1010, 643; (b)
Walsh, D. M.; Selkoe, D. J. Neuron 2004, 44, 181; (c) Tolia, A.; De Strooper, B.
Semin. Cell Dev. Biol. 2009, 20, 211; (d) Takami, M.; Nagashima, Y.; Sano, Y.;
Ishihara, S.; Morishima-Kawashima, M.; Funtamoto, S.; Yasuo, I. J. Neurosci.
2009, 29, 13042; (e) Hardy, J.; Selkoe, D. J. Science 2002, 297, 353.
4. (a) Dickson, D. W. J. Neuropathol. Exp. Neurol. 1997, 56, 321; (b) McKeith, I. G.;
Galasko, D.; Kosaka, K.; Perry, E. K.; Dickson, D. W.; Hansen, L. A.; Salmon, D. P.;
Lowe, J.; Mirra, S. S.; Byrne, E. J. L.; Lennox, G.; Quinn, N. P.; Edwardson, J. A.;
Ince, P. G.; Bergeron, C.; Burns, A.; Miller, B. L.; Lovestone, S.; Collerton, D.;
Jansen, E. N.; Ballard, C.; de Vos, R. A.; Wilcock, G. K.; Jellinger, K. A.; Perry, R. H.
Neurology 1996, 47, 1113; (c) Mandal, P. K.; Pettegrew, J. W.; Masliah, E.;
Hamilton, R. L.; Mandal, R. Neurochem. Res. 2006, 31, 1153.
In summary, conversion of amide GSMs into an isosteric
triazole, followed by conformational restriction to a bicyclic triaz-
olo-piperidine series resulted in the identification of 14-S. This
compound showed potent GSM activity in vitro and in vivo, in
combination with an improved drug-like profile compared to pre-
vious series. In the dog, at high exposure levels, no effects on liver
function were observed, which is most likely due to the reduced
lipophilicity of 14-S compared to previous potent GSMs such as 2
or reported NSAID derived GSMs.17 These results have strength-
ened our belief that liver toxicity findings with previous GSMs
were not target related and that it should be feasible to develop
GSMs without liver toxicity.18 To achieve significant and sustained
5. Oehlrich, D.; Berthelot, D. J. C.; Gijsen, H. J. M. J. Med. Chem. 2011, 54, 669.
6. Probst, G.; Xu, Y. Z. Expert Opin. Ther. Pat. 2012, 5, 511.
7. Hamada, Y.; Kiso, Y. Expert Opin. Drug Discov. 2009, 4, 391.
8. Gijsen, H. J. M.; Bishoff, F. P. Annu. Rep. Med. Chem. 2012, 47, 55.
9. Bischoff, F.; Berthelot, D.; De Cleyn, M.; Macdonald, G.; Minne, G.; Oehlrich, D.;
Pieters, S.; Surkyn, M.; Trabanco, A. A.; Tresadern, G.; Van Brandt, S.; Velter, I.;
Zaja, M.; Borghys, H.; Masungi, C.; Mercken, M.; Gijsen, H. J. M. J. Med. Chem.
2012, 55, 9089.
10. Borghys, H.; Tuefferd, M.; Van Broeck, B.; Clessens, E.; Dillen, L.; Cools, W.;
Vinken, P.; Straetemans, R.; de Ridder, F.; Gijsen, H.; Mercken, M. J. Alzheimer’s
Dis. 2012, 28, 809.
changes in Ab-levels, still high, >10 lM levels of 14-S were re-
quired. There is a growing consensus that amyloid targeting ther-
apies will need to be given early and chronically to have an
impact on the progression of AD.19 Therefore, a further leap in effi-
cacy will likely be required to make GSM treatment a viable option.
Considering further optimization of 14-S, an improvement in brain
penetration may help to achieve this.
11. IC50 represents the concentration of a compound that is required for reducing
the Ab42 level by 50%. The values are a mean of at least three determinations.
The confidence intervals for the IC50 values are 95%.
12. Work around
a similar series of benzamides was published recently:
Pettersson, M.; Johnson, D. S.; Subramanyam, C.; Bales, K. R.; am Ende, C.
W.; Fish, A. B.; Green, E. M.; Kauffman, G. W.; Lira, R.; Mullins, P. B.;
Navartnam, T.; Sakya, S. M.; Stiff, C. M.; Tran, T. P.; Vetelin, B. C.; Xie, L.;
Zhang, L.; Pustilnik, L. R.; Wood, K. M.; O’Donnell, C. J. Bioorg. Med. Chem.
Lett. 2012, 22, 2906.
Acknowledgments
13. Wu, T.; Gijsen, H. J. M.; Rombouts, F. J. R.; Bischoff, F. P.; Berthelot, D. J.-C.;
Oehlrich, D.; De Cleyn, M. A. J.; Pieters, S. M. A.; Minne, G. B.; Velter, A. I. PCT-
WO2011006903.
14. Mouse in vivo at 30 mg/kg oral administration after 4 h (n = 6) expressed as a
percentage of Ab42 lowering in brain compared with untreated animals.
Plasma and brain samples were analyzed for the tested compound using a
qualified research LC–MS/MS method. For a more detailed description of the
in vivo and in vitro assays, see Ref. 9
The authors would like to thank co-workers at Cellzome, the
members of the Janssen R&D Biology, ADME and screening depart-
ments and the members of the analysis and purification
department.
Supplementary data
15. Huang, X.; Aslanian, R.; Zhou, W.; Zhu, X.; Qin, J.; Greenlee, W.; Zhu, Z.; Zhang,
L.; Hyde, L.; Chu, L.; Cohen-Williams, M.; Palani, A. ACS J. Med. Chem. Lett. 2010,
1, 184.
Supplementary data (comparison of low energy conformations
for molecule 6, 7 and 11-S and the calculated and determined
VCD spectra of 11-S and 11-R) associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.bmcl.2013.06.100.
16. Conformational analysis was performed in Molecular Operating Environment
version 2011.10. Default MMFF94x force field settings were used with Born
solvation turned on (internal dielectric 1, external 80). Low mode MD approach
was used for the conformational searches, all settings were at default except
rejection limit was increased to 1000, inter-conformer RMSD acceptance was
set at 0.1 Å and energy window was increased to 10 kcal/mol.
17. Van Broeck, B.; Chen, J.-M.; Tréton, G.; Desmidt, M.; Hopf, C.;
Ramsden, N.; Karran, E.; Mercken, M.; Rowley, A. Br. J. Pharmacol.
2011, 163, 375.
References
1. Alzheimer’s Association, 2012 Alzheimer’s Disease Facts and Figures,
18. Gijsen, H. J. M.; Mercken, M. Int. J. Alzheimer’s Dis. 2012, 2012, 295207.
19. Miller, G. Science 2012, 337, 790.
Alzheimer’s & Dementia, Volume 8, Issue 2.
Please cite this article in press as: Oehlrich, D.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.06.100