Quinazolinones as γ-secretase modulators

Article abstract of DOI:10.1016/j.bmcl.2010.11.111

Synthesis, SAR and evaluation of styrenyl quinazolinones as novel gamma secretase modulators are presented in this communication. Starting from literature and in-house leads we evaluated a range of quinazolinones which showed good modulation of γ-secretas

Full text of DOI:10.1016/j.bmcl.2010.11.111

Bioorganic & Medicinal Chemistry Letters 21 (2011) 773–776
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Quinazolinones as c-secretase modulators
⇑
Christian Fischer , Sanjiv Shah, Bethany L. Hughes, George N. Nikov , Jamie L. Crispino,
Richard E. Middleton, Alexander A. Szewczak, Benito Munoz, Mark S. Shearman
Merck Research Laboratories Boston, 33 Avenue Louis Pasteur, Boston, MA 02115, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Synthesis, SAR and evaluation of styrenyl quinazolinones as novel gamma secretase modulators are pre-
sented in this communication. Starting from literature and in-house leads we evaluated a range of qui-
Received 9 October 2010
Revised 20 November 2010
Accepted 23 November 2010
Available online 26 November 2010
nazolinones which showed good modulation of
c
-secretase activity.
Ó 2010 Elsevier Ltd. All rights reserved.
This communication is dedicated to George
N. Nikov
Keywords:
Alzheimer
Gamma Secretase
Modulator
APP
Ab42 lowering
Alzheimer’s disease (AD) is a progressive neurodegenerative
disorder, which in 2009 affected over 5 million patients in the US
alone.1 Current standard of care relies on symptomatic treatment,
which only provides short-term stabilization and to date no dis-
ease modifying therapy has been clinically validated. Late-onset,
sporadic AD is a poorly understood disease and while many
hypotheses exist, no causative relationship has been established
yet. The most widely accepted hypothesis for AD describes oligo-
mers of the b-amyloid (Ab) peptides, specifically the longer frag-
ment Ab42, as a neurotoxic species which in vitro and in vivo
result in neurodegeneration, thus causing the symptoms of AD
such as cognitive impairment.2
efficacy of GSIs is still awaiting clinical validation, it is widely ac-
cepted that inhibition of -secretase is associated with mecha-
nism-based side-effects potentially limiting the long-term use of
GSIs. Most adverse events have been associated with inhibiting
the processing of Notch, one of several other substrates of c-secre-
tase in addition to C-99. Efforts to develop GSIs that dissociate inhi-
bition of C-99 and Notch cleavage are a field of intense research
and several promising leads have been documented.7
An alternative approach to interrogate the amyloid hypothesis
came from the observation that the longer Ab42 fragments elicit
significantly greater neurotoxicity than their shorter counterparts
Ab36–40.2 It has therefore been postulated that modulation of
c
The last step in the formation of Ab is the proteolytic cleavage of
the c-cleavage to increase production of shorter fragments at the
C-993 by the
c
-secretase complex, wherein presenilin acts as the
protease producing Ab fragments as well as an intracellular do-
main (AICD) of unknown function.4 The biology of
-secretase is
expense of longer fragments such as Ab42 might be a safer ap-
proach to a disease modifying therapy. Gamma secretase modula-
tors (GSMs) modulate the cleavage of C-99 to decrease Ab42,
increase Ab37/388 and leave total Ab levels unchanged whilst not
affecting cleavage of other substrates, such as Notch.9
c
complex and the mechanism of intra-membrane proteolysis by
this multi-component enzyme is to date poorly understood.5
We and others have investigated inhibitors of
the past decade and a number of patents and publications have de-
scribed the efforts in this area.6 Several
-secretase inhibitors
(GSIs) have progressed into clinical trials, but only limited pharma-
codynamic and no efficacy data have been released yet. While the
c
-secretase for
The discovery that (R)-flurbiprofen (Tarenflurbil, Flurizan™) (1,
Fig. 1)10, amongst other nonsteroidal anti-inflammatory drugs,
showed a GSM-like profile fueled the research in this area. Recently
Myriad reported disappointing Phase III data for Flurizan™ leading
to the discontinuation of clinical development.11 While this has
c
been a major setback for the field of c-secretase modulation, the
cellular potency of Flurizan™ and brain exposure is very low. Addi-
tionally, studies have recently suggested that Flurizan™ might
bind to amyloid precursor protein (APP) and not to c-secretase it-
⇑
Corresponding author.
E-mail address: christian_fischer@merck.com (C. Fischer).
Deceased.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.111

774
C. Fischer et al. / Bioorg. Med. Chem. Lett. 21 (2011) 773–776
Table 1
S
CO2H
H
N
Me
CO2H
Modulation of Ab40/42 processing by compounds 14–24
Ar
F
N
MeO
R
N
N
N
Ar
N
N
R1
R2
Me
FlurizanTM (1)
2
3
Me
Compound
R
Ab42 IC50
M)
Ab40 IC50
M)
hERG IC50
M)
R1
R2
R3
(l
(
l
(l
O
Me
R4
MeO
N
N
H
N
N
N
N
N
H
14
15
0.84
3.89
>10
0.68
>10
F
N
N
N
O
O
O
O
Me
Me
E-2012 (4)
5
Me
MeO
HN
N
4.95
Figure 1. Selected gamma secretase modulators.
HN
N
F
self.12 Therefore, despite negative Phase III data, the discovery of
potent modulators of - secretase which directly target the en-
16
17
4.87
4.28
>10
>10
n.d.
>10
c
zyme rather than APP is still worth pursuing.
HN
N
We have reported the discovery and SAR of exquisitely selective
(Ab40 IC50/Ab42 IC50 >100) carboxylic acid GSMs such as piperi-
dine carboxylic acids 213 and Torrey-Pines14 as well as Eisai15 dis-
closed aryl imidazoles as represented by 3 and 4 (Fig. 1).
In an effort to diversify our GSM portfolio we embarked on a
program to identify novel non-carboxylic acids as modulators of
F
HN
N
18
19
20
21
0.66
0.34
2.18
0.37
>10
>10
>10
>10
0.87
c
-secretase. Recent publications have detailed our lead optimiza-
Cl
Br
tion strategy in pyrimidine derivatives (e.g., 5) as GSMs.16,17 Herein
O
O
O
O
we present our work on olefinic quinazolinones.18
HN
N
Our efforts in this area started with lessons learned from liter-
ature examples 3 and 4 as well as our in-house compounds such
as 5. Initially we focused our efforts on maintaining the olefinic
spacer of 4 while investigating amide replacements such as benz-
imidazoles (14, Table 1).19 While we were able to identify several
4.82
6.74
3.74
HN
N
imidazole derivatives which behaved as potent c-secretase modu-
lators (data not shown), they all suffered from significant hERG
binding.20 We hypothesized that a ring-expansion from benzimi-
dazoles to quinazolinones would lower their basicity and attenuate
hERG binding while maintaining their salient properties.
SO2Me
HN
N
CF3
Me
O
O
HN
N
Na
OH
NH2
N
NH
R
R
22
23
24
1.29
1.65
0.37
>10
>10
3.89
>10
>10
4.73
R
Cl
MeOH
Cl
N
8
S
O
6
7
HN
N
O
Me
Me
O
P(OEt)3
NH
NH
O
P
OEt
S
R
Cl
OEt
N
N
O
HN
N
8
9
MeO
N
CHO
S
NH
MeO
CHO
K2CO3
DMF
N
N
F
Me
Me
Synthesis of the cinnamyl quinazolinones (13) commenced
with the preparation of phosphonates (9) from chloromethyl qui-
nazolinones (8), which were either commercially available or alter-
natively prepared from the corresponding anthranilic acids (6)
with chloroacetonitrile (7) and freshly prepared sodium methoxide
(Scheme 1). The solvent-free reaction of chloromethyl quinazoli-
nones (8) with triethylphosphite gave the corresponding phospho-
nates (9) in generally good yields and most analogues could be
readily precipitated from diethylether requiring no purification.
The left-hand piece of the molecule was readily prepared from
commercially available 4-fluoro-3-methoxybenzaldehyde (10)
10
11
12
9
O
LiOH.H2O
THF/EtOH
HN
R
MeO
N
N
N
13
Scheme 1. Representative synthesis of compounds 14–32.
Me

C. Fischer et al. / Bioorg. Med. Chem. Lett. 21 (2011) 773–776
775
Ab40 production, with some compounds having no selectivity
(e.g., 15) and others having at least 15-fold selectivity (e.g., 18).
Quinazolinone 18 was early on identified as a compound with
reasonable cell potency, Ab42 selectivity and attenuated hERG
binding.25 Compound 18 was further evaluated in vivo and dis-
played excellent rat pharmacokinetics with acceptable oral bio-
availability (Fig. 2).
The quinazolinones in Table 1 proved to be poorly soluble even
after substantial formulation efforts, which led to variable expo-
sures at higher doses. We hypothesized that the pyrimidone group
was responsible for the poor physicochemical properties and be-
gan to investigate N-alkylated and arylated quinazolinones as well
as aminopyrimidines, which were prepared either from commer-
cially available building blocks or in a one-pot procedure from
the corresponding pyrimidones.26 Gratifyingly, both approaches
led to compounds with significantly improved solubility.27 How-
ever, a slight loss in potency and/or increased hERG binding offset
the gain in solubility (Table 2).
Dose (iv; po) (mg/kg)
Clp (mL/min/kg)
Vdss (L/Kg)
t½ (iv) (h)
%F
1; 2
6.8
1.8
5.8
59
3.7
AUCN 0–24 po (µM h kg/mg)
Figure 2. Pharmacokinetic profile of 18 in Sprague–Dawley rats.
and 4-methyl-imidazole (11) with potassium carbonate as the
inorganic base and heating over night in DMF. Separation of the
undesired regio-isomer was achieved through a wash procedure
and subsequent purification on silica gel.21
Horner–Wadsworth–Emmons reaction of 3-methoxy-4-(4-
methyl-1H-imidazol-1-yl)benzaldehyde (12) with quinazolinone
phosphonates (9) yielded the final products in good yields after
purification by normal or reversed-phase chromatopgraphy.22
Quinazolinones 14–24 were profiled in our hAPP-overexpress-
ing SH-SY5Y cell line and showed potent modulation of Ab forma-
tion (Table 1).23
Substitution on the 7-position of the quinazoline ring proved
optimal for potency (e.g., 18, 19, 21) and we were delighted to
see significant attenuation of hERG binding as measured by dis-
placement of MK-499 from hERG (e.g., 18–20).24 Compounds in
this series were variably selective for inhibition of Ab42 versus
In summary, we have reported the discovery and SAR of styre-
nyl quinazolinones as modulators of
c-secretase. Starting from
benzimidazole 14, which suffered from significant hERG binding,
we discovered that ring-expansion to quinazolinones attenuated
binding to the hERG channel. Compound 18, with a functional
hERG IC50 >10 lM, was a potent GSM with an excellent pharmaco-
kinetic profile. Future directions in this series and novel strategies
for non-acid GSMs will be reported in due course.
Table 2
References and notes
Ab40/42 data for compounds 25–32
1. Alzheimer’s Association: 2009 Alzheimer’s disease facts and figures http://
www.alz.org/national/documents/report_alzfactsfigures.2009.pdf.
MeO
R
2. See for example: (a) Hung, L. W.; Ciccotosto, G. D.; Giannakis, E.; Tew, D. J.;
Perez, K.; Masters, C. L.; Cappai, R.; Wade, J. D.; Barnham, K. J. J. Neurosci. 2008,
28, 11950; (b) Moreno, H.; Yu, E.; Pigino, G.; Hernandez, A. I.; Kim, N.; Moreira,
J. E.; Sugimori, M.; Llinas, R. R. Proc. Natl. Acad. Sci. 2009, 106, 5901.
3. C-99 is the C-terminal fragment of the amyloid precursor protein (APP)
produced by b-secretase mediated proteolysis of APP.
4. (a) Li, H.; Wolfe, M. S.; Selkoe, D. J. Structure 2009, 17, 326; (b) Osenkowski, P.;
Li, H.; Ye, W.; Li, D.; Aeschbach, L.; Fraering, P. C.; Wolfe, M. S.; Selkoe, D. J.; Li,
H. J. Mol. Biol. 2009, 385, 642.
5. See for example: (a) Selkoe, D. J.; Wolfe, M. S. Cell 2007, 131, 215; (b) Wolfe, M.
S. Curr. Top. Med. Chem. 2008, 8, 2; (c) Steiner, H.; Fluhrer, R.; Haass, C. J. Biol.
Chem. 2008, 283, 29627; (d) Wolfe, M. S. Semin. Cell Dev. Biol. 2009, 20, 219.
6. For selected reviews, see: (a) Wu, W.-L.; Zhang, L. Drug Dev. Res. 2009, 70, 94;
(b) Olson, R. E.; Albright, C. F. Curr. Top. Med. Chem. 2008, 8, 17; (c) Imbimbo, B.
P. Drug Discovery Today 2008, 5, 169.
N
N
Me
Compound
R
Ab42 IC50
M)
Ab40 IC50
M)
hERG IC50
M)
(l
(l
(l
NH2
N
N
25
26
27
1.69
2.08
>10
6.10
>10
>10
0.14
0.99
0.31
Ph
O
N
N
7. For a recient reivew on APP processing, see: Thinakaran, G.; Koo, E. H. J. Biol.
Chem. 2008, 283, 29615; For a recent review on secretase biology and modes of
Me
O
c
-secretase inhibition/modulation, see: De Strooper, B.; Vassar, R.; Golde, T.
N
N
Nat. Rev. Neurol. 2010, 6, 99.
8. All GSMs lower Ab42 selectively over Ab40, however, some raise the levels of
Ab38 and/or Ab37. The consequence of this difference is unknown.
9. For selected reviews on GSMs, see: (a) Peretto, I.; La Porta, E. Curr. Top. Med.
Chem. 2008, 8, 38; (b) Breher, D. Curr. Top. Med. Chem. 2008, 8, 34; (c) Imbimbo,
B. P. Curr. Top. Med. Chem. 2008, 8, 54; (d) Tomita, T. Expert Rev. Neurother. 2009,
9, 661.
NH2
N
N
N
28
1.07
3.33
0.30
HN Ph
NH2
10. Flurizan is the cyclooxygenase (COX1
Flurbiprofen.
& 2)-inactive enantiomer of (S)-
N
N
11. Evin, G. Expert Rev. Neurother. 2008, 8, 1611.
Me
29
30
31
32
0.54
0.32
1.86
0.20
3.72
2.96
>10
0.98
0.51
1.12
0.39
12. Kukar, T. L.; Ladd, T. B.; Bann, M. A.; Fraering, P. C.; Narlawar, R.; Maharvi, G.
M.; Haely, B.; Chapman, R.; Welzel, A. T.; Price, R. W.; Moore, B.; Rangachari,
V.; Cusack, B.; Eriksen, J.; Jansen-West, K.; Verbeeck, C.; Yager, D.; Eckman,
C.; Ye, W.; Sagi, S.; Cottrell, B. A.; Torpey, J.; Rosenberry, T. L.; Fauq, A.;
Wolfe, M. S.; Schmidt, B.; Walsh, D. M.; Koo, E. H.; Golde, T. E. Nature 2008,
453, 925.
13. (a) For patents in this area, see: Hannam, J. C.; Hartmann, S.; M., Andrew; R.,
Mark, P. WO2007110667.; (b) Madin, A.; Ridgill, M. P.; Kulagowski, J. J. WO
2007116228.; (c) Garcia, Y.; Hannam, J. C.; Harrison, T.; Hamblett, C. L.; Hubbs,
J. L.; Kulagowski, J. J.; Madin, A.; Ridgill, M. P.; Seward, E. WO 2007125364.; (d)
Stanton, M. G.; Munoz, B.; Sloman, D. L.; Hubbs, J. L.; Hamblett, C. L. WO
2008030391.; (e) Munoz, B.; Hubbs, J. L.; Hamblett, C. L.; Zhou, H.; Martinez, M.
WO 2008085301.
S
CO2Et
NH2
N
N
O
NHBn
N
N
S
NH2
14. For a recent publication, see: Kounnas, M. Z.; Danks, A. M.; Cheng, S.; Tyree,
C.; Ackerman, E.; Zhang, X.; Ahn, K.; Nguyen, P.; Comer, D.; Mao, L.; Yu, C.;
Pleynet, D.; Digregorio, P. J.; Velicelebi, G.; Stauderman, K. A.; Comer, W. T.;
Mobley, W. C.; Li, Y.-M.; Sisodia, S. S.; Tanzi, R. E.; Wagner, S. L. Neuron
2010, 67, 769.
N
N
2.03
S

776
C. Fischer et al. / Bioorg. Med. Chem. Lett. 21 (2011) 773–776
15. For a key patent, see: Kimura, T.; Kawano, K.; Doi, E.; Kitazawa, N.; Shin, K.;
Miyagawa, T.; Kaneko, T.; Ito, K.; Takaishi, M.; Sasaki, T.; Hagiwara, H. WO
2005115990.
16. (a) Blurton, P.; Fletcher, S.; Teall, M.; Harrison, T.; Munoz, B.; Rivkin, A.;
Hamblett, C.; Siliphaivanh, P.; Otte, K. WO 2008099210.; (b) Rivkin, A.; Ahearn,
S. P.; Chichetti, S. M.; Hamblett, C. L.; Garcia, Y.; Martinez, M.; Munoz, B. WO
2010019392.
17. (a) Rivkin, A.; Ahearn, S. P.; Chichetti, S. M.; Kim, Y. R.; Li, C.; Rosenau, A.;
Kattar, S. D.; Jung, J.; Shah, S.; Hughes, B. L.; Crispino, J. L.; Middleton, R. E.;
Szewczak, A. A.; Munoz, B.; Shearman, M. S. Bioorg. Med. Chem. Lett. 2010, 20,
1269; (b) Rivkin, A.; Ahearn, S. P.; Chichetti, S. M.; Hamblett, C. L.; Garcia, Y.;
Martinez, M.; Hubbs, J. L.; Reutershan, M. H.; Daniels, M. H.; Siliphaivanh, P.;
Otte, K. M.; Li, C.; Rosenau, A.; Surdi, L. M.; Jung, J.; Hughes, B. L.; Crispino, J. L.;
Nikov, G. N.; Middleton, R. E.; Moxham, C. M.; Szewczak, A. A.; Shah, S.; Moy, L.
Y.; Kenific, C. M.; Tanga, F.; Cruz, J. C.; Andrade, P.; Angagaw, M. H.; Shomer, N.
H.; Miller, T.; Munoz, B.; Shearman, M. S. Bioorg. Med. Chem. Lett. 2010, 20,
2279.
22. Other analogues, including amino quinazolines used in Table 2, were prepared
in a similar manner. For detailed synthetic procedures, see Ref. 18
23. IC50 measurements for inhibition of Ab40 and Ab42 production were
determined using electrochemiluminescent detection of these peptides
secreted by SH-SY5Y cells stably overexpressing the APP C-terminal
fragment SPA4CT. Consistent with the profiles of
c-secretase modulators,
total Ab peptide levels were constant. The GSM profile of selected compounds
was confirmed by mass spectrometry (SELDI), which confirmed the appearance
of shorter (Ab37) fragments while Ab42 formation was suppressed. (a) Best, J.
D.; Jay, M. T.; Otu, F.; Ma, J.; Nadin, A.; Ellis, S.; Lewis, H. D.; Pattison, C.; Reilly,
M.; Harrison, T.; Shearman, M. S.; Williamson, T. L.; Atack, J. R. J. Pharmacol. Exp.
Ther. 2005, 313, 902; (b) Clarke, E. E.; Shearman, M. S. J. Neurosci. Methods 2000,
102, 61; (c) Dyrks, T.; Dyrks, E.; Monning, U.; Urmoneit, B.; Turner, J.;
Beyreuther, K. FEBS Lett. 1993, 335, 89.
24. For a description of the hERG radioligand displacement assay, see: Butcher, J.
W.; Claremon, D. A.; Connolly, T. M.; Dean, D. C.; Karczewski, J.; Koblan, K. S.;
Kostura, M. J.; Liverton, N. J.; Melillo, D. G. WO 200205860.
18. Fischer, C.; Munoz, B.; Rivkin, A. WO 2008097538.
19. The olefin spacer is crucial; the more flexible ethylene spacer is significantly
25. In a functional evaluation of hERG (IKr) currents with PatchXpress, compound
18 could not be fully evaluated due to solubility limitations; at the maximum
less potent. The saturated analogues of 14 and 15 have Ab42 IC50s >10
lM.
soluble concentration (10 lM) a 39% (N = 4) mean blockade was observed.
20. We have also investigated several other heterocyclic amide replacements, such
as oxazoles, oxadiazoles, thiazoles and thiadiazoles. While all of them behaved
as GSMs, their binding to the hERG channel was generally equal or greater than
their cellular potency as GSMs.
26. Compound 31 was prepared in a one-pot procedure from the corresponding
fused pyrimidone 24 according to: Wan, Z.-K.; Binnun, E.; Wilson, D. P.; Lee, J.
Org. Lett. 2005, 7, 5877.
27. Solubility of compounds 15–24 was not quantifiable at pH 7 in phosphate
21. Careful work-up and purification was necessary to remove the unwanted
regio-isomer. See Ref. 18 for detailed synthetic procedures.
buffered saline (PBS) buffer. Compounds 25 and 27 had 6 and 33 lM solubility,
respectively, in PBS.