Design and synthesis of water soluble β-aminosulfone analogues of SCH 900229 as γ-secretase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.04.095

In this paper we describe our strategy to improve the aqueous solubility of SCH 900229, a potent PS1-selective γ-secretase inhibitor for the treatment of Alzheimer's disease. Incorporation of ionizable amino groups into the side chain terminal generates w

Full text of DOI:10.1016/j.bmcl.2016.04.095

Accepted Manuscript
Design and synthesis of water soluble β-aminosulfone analogues of SCH
900229 as γ-secretase inhibitors
Wen-Lian Wu, Duane A. Burnett, John Clader, William J. Greenlee, Qin Jiang,
Lynn A. Hyde, Robert A. Del Vecchio, Mary E. Cohen-Williams, Lixin Song,
Julie Lee, Giuseppe Terracina, Qi Zhang, Amin Nomeir, Eric M. Parker, Lili
Zhang
PII:
S0960-894X(16)30493-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2016.04.095
BMCL 23868
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
1 April 2016
27 April 2016
28 April 2016
Please cite this article as: Wu, W-L., Burnett, D.A., Clader, J., Greenlee, W.J., Jiang, Q., Hyde, L.A., Del Vecchio,
R.A., Cohen-Williams, M.E., Song, L., Lee, J., Terracina, G., Zhang, Q., Nomeir, A., Parker, E.M., Zhang, L.,
Design and synthesis of water soluble β-aminosulfone analogues of SCH 900229 as γ-secretase inhibitors,
Bioorganic & Medicinal Chemistry Letters (2016), doi: http://dx.doi.org/10.1016/j.bmcl.2016.04.095
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Leave this area blank for abstract info.
Design and synthesis of water soluble -
aminosulfone analogues of SCH 900229 as -
secretase inhibitors
Wen-Lian Wu, Duane A. Burnett, John Clader, William J. Greenlee, Qin Jiang, Lynn A. Hyde, Robert A. Del Vecchio,
Mary E. Cohen-Williams, Lixin Song, Julie Lee, Giuseppe Terracina, Qi Zhang, Amin Nomeir, Eric M. Parker, Lili Zhang

Bioorganic & Medicinal Chemistry Letters
Design and synthesis of water soluble -aminosulfone analogues of SCH 900229
as -secretase inhibitors
Wen-Lian Wua,*, Duane A. Burnetta, John Cladera, William J. Greenleea, Qin Jiangb, Lynn A. Hydea,
Robert A. Del Vecchioa, Mary E. Cohen-Williamsa, Lixin Songa, Julie Leea, Giuseppe Terracinaa, Qi
Zhanga, Amin Nomeira, Eric M. Parkera, Lili Zhanga
a
Merck Research Laboratories, 2015 Galloping Hill Rd, Kenilworth, NJ 07033, USA
b
Albany Molecular Research, Inc., 26 Corporate Cir, Albany, NY 12212, USA
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
In this paper we describe our strategy to improve the aqueous solubility of SCH 900229, a
potent PS1-selective -secretase inhibitor for the treatment of Alzheimer's disease.
Incorporation of ionizable amino groups into the side chain terminal generates water soluble-
aminosulfone analogues of SCH 900229 that maintain robust in vitro potency and in vivo
efficacy.
Keywords:
2009 Elsevier Ltd. All rights reserved.
Alzheimer's disease
-Secretase inhibitor
-aminosulfone
aqueous solubility
Alzheimer’s disease (AD) is a neurodegenerative disease
afflicting 36 million people worldwide.1 The search for a disease
modifying treatment for AD has been accelerated, particularly in
the last two decades, with newly found genetic evidence
supporting the so-called amyloid hypothesis.2,3 This hypothesis
postulates that the β-amyloid (Aβ) peptides play a causative role
in the disease. As a result, numerous programs targeting the A
pathway have been reported.4 The two enzymes involved in the
A production are β-site amyloid cleaving enzyme 1 (BACE1)
and -secretase. -Secretase is a complex of four distinct
proteins: presenilin 1 (PS1) or presenilin 2 (PS2), nicastrin, aph-
1 or aph1- and pen-2. 5 There are thus multiple isoforms of -
secretase as a result of different combinations of PS1 and PS2
with Aph-1 and Aph-1. Selective inhibition of -secretase
isoforms containing PS1 proved to be a promising strategy in the
design and development of -secretase inhibitors (GSIs) that
maintained A-lowering efficacy but reduced mechanism-based
Notch-related adverse effects.6,7 Toward that end, we have
designed a series of bis-sulfone analogues as PS1-selective GSIs
and advanced the lead compound SCH 900229, [(-)-1a, Figure 1]
into clinical trials.8 Extensive structure-activity relationship
(SAR) efforts around the frame of SCH 900229 revealed limited
tolerance of substituents on the two phenyl rings. By contrast,
analogues bearing various sulfone side chains showed robust in
vitro potency; however, only selected compounds demonstrated
good pharmacokinetics and in vivo efficacy. The two carbon
linker is optimal for the sulfone series of analogues.9
The clinical candidate SCH 900229 is a non-ionizable
molecule with low aqueous solubility (kinetic solubility, Ksol 50
M) with a BCS classification of II. It is therefore likely that
mproving the physicochemical properties of SCH 900229 is
necessary for optimum therapeutic performance.
Figure 1. -Secretase inhibitor SCH 900229 and SAR development
In our -secretase inhibitor backup program, we aimed to
improve upon the aqueous solubility of SCH 900229, while
maintaining the basic structure and the favorable in vitro and in
vivo properties. The wide tolerance for various side chains in
terms of in vitro potency suggested that this region may be a
‘sweet spot’ for modulating physicochemical properties without
compromising the key interactions with the enzyme. As reported
previously,9 analogues bearing either a polar hydroxyl (1b, table
1) or carboxylic acid group (1c) showed great aqueous solubility
(Ksol > 200 M) but poor in vivo efficacy. We investigated

whether side chains with basic moieties, such as the -
aminosulfone
as
shown
in Figure 2, would make a difference. We reasoned that the basic amine group can be protonated to form salts that generally have a
higher aqueous solubility and a more rapid dissolution rate than the free base, while the neutral portion can still allow cell membrane
and blood-brain barrier (BBB) permeability. On the other hand, a strong electron-withdrawing sulfone group will keep the basicity of
the amine between 7 and 9. This balanced effect is important because excessive basicity combined with large lipophilic groups may
cause adverse effects, such as inhibition of the hERG channel or phospholipidosis.10 Indeed, similar considerations have been
demonstrated in the design of launched drugs, including dalfopristin11 and lapatinib.12
Figure 2. Design of -aminosulfone analogues
We took advantage of SCH 900229 intermediates to further the SAR exploration. As shown in Scheme 1, the racemic aldehyde 2
was elongated by Wittig olefination to give racemic 3. Reduction of the terminal alkene with 9-BBN followed by oxidation with H2O2
provided the racemic alcohol,13 which was resolved by chiral HPLC (Chiralcel OJ column) to give the desired enantiomerically pure
intermediate (-)-4. Compound 4 was converted to its tosylate derivative 5, and the tosylate was then displaced with hydroxyethylthiol
followed by mCPBA oxidation to give the -hydroxysulfone 6. The alcohol was derivatized to its mesylate 7, which was treated with
various amines to generate -aminosulfone analogues 8a-8m, presumably through -elimination followed by aza-Michael addition of
amines to the vinyl sulfone intermediate.14
Scheme 1. Reagents and conditions: (a) Ph3P+CH3Br-, n-BuLi, THF, -50 oC to rt, 62%; (b) 9-BBN, THF; H2O2, NaOH, 60%; (c) Chiral HPLC separation; (d)
TsCl, CH2Cl2, 90%; (e) HOCH2CH2SH, KOH, EtOH; (f) mCPBA, CH2Cl2, 80% in 2-step; (g) MsCl, Et3N, CH2Cl2, 94%; (j) R1R2NH, DMF, rt, >90%.
The synthesis of the ,-dimethyl substituted analogue 12 is depicted in Scheme 2. The tosylate 5 was treated with potassium
thioacetate to give thioester 9. Hydrolysis of the thioester with KOMe in methanol followed by addition of 2,2-dimethyloxirane
generated alcohol 10 in one-pot. After oxidation with mCPBA, treatment of the tertiary alcohol with mesyl chloride in the presence of
triethylamine gave the ,-unsaturated sulfone 11 directly. Finally, aza-Michael addition reaction between methylamine and 11 gave
the -aminosulfone analogue 12.

Scheme 2. Reagents and conditions: (a) MeCOSK, EtOH, 80%; (b) KOMe, MeOH, 2,2-dimethylepoxide, 60%; (c) mCPBA, CH2Cl2, 91%; (d) MsCl, Et3N, rt;
(e) MeNH2, MeCN, rt.
To synthesize cyclic amine analogues 14a-14e, a different strategy was adopted as depicted in Scheme 3. The tosylate 5 was treated
with various thio intermediates that were generated in-situ from appropriate thioester precursors 17a-17e to provide sulfides 13a-13e.
Oxidation of the sulfides with mCPBA and deprotection of N-Boc with TFA or N-Ac with KOH furnished the targets 14a-14e.
Scheme 3. Reagents and conditions: (a) Thioesters 17a-17e, KOMe, MeOH, > 50%; (b) mCPBA, CH2Cl2; > 90% (c) TFA, CH2Cl2, > 90%; (d) KOH, EtOH,
reflux, 77%.
The required thioester precursors 17a-17d were prepared from commercially available primary and secondary alcohols, as
exemplified by compound 17a in Scheme 4. Compound 17e was prepared according to a literature procedure.15
Scheme 4. Reagents and conditions: (a) MeSO2Cl, Pyridine, CH2Cl2; (b) MeCOSK, DMF, 100 oC, 83% in 2-step.
The amidine analogue 20 was synthesized from the intermediate 9 as shown in Scheme 5. As described previously, hydrolysis
followed by in-situ alkylation provided nitrile 18. Treatment of the nitrile with ethylenediamine in the presence of sulfur gave amidine
19.16 Finally, mCPBA oxidation under acidic conditions furnished the sulfone 20 in good yield.

Scheme 5. Reagents and conditions: (a) BrCH2CN, KOMe, MeOH, ~ 70%; (b) S, H2NCH2CH2NH2; (c) mCPBA, TFA, CH2Cl2, 40% in 2-step.
Analogues 8a-8m, 12, 14a-14e, and 20 were all enantiomerically pure and converted to their corresponding HCl salt form for all the
tests. They were first screened in the membrane-based in vitro -secretase assay and in the PS2/PS1 ratio assay.17,6 Selected analogues
were then examined for their in vivo activity in mice by measuring A reduction in plasma and brain after a single 10 mg/kg oral
dose.18 As summarized in Table 1, all the -aminosulfone analogues retained the excellent in vitro potency of SCH 900229 [(-)-1a],
confirming the notion that this region of the molecule does not interact with the enzyme directly. The primary, secondary and tertiary
amine analogue 8a-8f also display selectivity for PS1 (PS2/PS1 ratios ranging from 18-35). Cyclic amines in various rings, such as
14a-14e, still maintained good potency and PS1 selectivity. Blocking the -position of 8b with gem-dimethyl was well tolerated (12).
Interestingly, the amidine analogue 20 also has good in vitro potency, with an IC50 of 5.9 nM. Several compounds also reduced plasma
and/or cortex Aβ40 to varying degrees.
Table 1 Right-hand side chain sulfone substitution SAR data.
Compound
1a
R
A40 IC50 (nM)a,b
membrane
1.3
PS2/PS1c
In vivo A40 (%)d
plasma
cortex
Me
25
-89
-78
(CH2)2OH
CH2COOH
1.0
2.7
0.8
4.1
0.9
0.5
2.0
1.7
1.6
1.1
0.6
-42
+47
-54
-44
-46
1b
1c
8a
8b
8c
8d
8e
8f
22
35
25
18
30
CH2CH2NH2
CH2CH2NHMe
CH2CH2NHEt
CH2CH2NHiPr
CH2CH2NHcBu
CH2CH2NMe2
-52
-61
20
17
13
12
-21
-45
8g
8h
8i
0.8
12
8j
1.3
3.7
0.8
2.2
9
8k
8l
10
10
-95
-51
8m
12
-33

1.0
0.7
1.4
23
14
30
-71
-57
-82
-42
-48
-56
14a
14b
14c
5.1
21
-90
-73
14d
1.3
5.9
-15
-86
14e
20
a The IC50 data are an average of at least two measurements. The standard deviations were within 20%. b Determined in HEKSwe-Lon 293 cells, see ref. 17.
c Selectivity for PS1 vs. PS2 was determined as a ratio of IC50 values obtained in an in vitro reconstituted PS1 and PS2 -secretase system (n =2), see ref. 6.
d Measured at 3 h after an oral dose of 10 mg/kg administered to,B6C3F1 mice; (-) indicates lowering of Aβ40, (+) indicates an increase of Aβ40, see ref.
18.
As exhibited in Table 2, HCl salts of most of the primary and secondary amine analogues, such as 8a-8c and 14a-14e, displayed
improved aqueous solubility (Ksol > 200 M) compared to SCH 900229 (50 M). Tertiary amine analogues generally did not improve
the kinetic solubility, as exemplified by analogue 8f. Compound 12 had diminished kinetic solubility (20 M), probably due to its
reduced hydration capacity. Surprisingly, the amidine analogue 20 also had poor kinetic solubility despite its high nitrogen count. The
pKa of compound 8b was determined to be 7.7. At physiological pH of 7.4, it was estimated that two-thirds of 8b will exist in
protonated form and the other one-third as the free base.
Table 2. in vitro and in vivo ADME profiles
Compound
Ksola
CYP3A4b
IC50 (M)
Co, Prec
15.1/2.8
>20/ 8.4
>20/ 1.4
>20/ 9.0
>20/>20
>20/>20
>20/ 1.0
>20/ 1.0
2.0/ 1.0
Hep Cld
AUC6h
B/Pf
(M)
L/min/106 cells
(nM.h)e
human
8.5
rat
dog
6.4
monkey
20.3
40.5
44.6
65.0
12.7
55.1
28.0
26.7
60.5
38.8
32.0
8a
8b
200
200
200
50
12.1
22.3
29.1
66.7
6.1
582
532
1355
blq
17.3
15.1
57.2
5.3
11.5
10.8
56.5
7.7
0.4
0.4
8c
8f
8m
12
200
>20
200
200
200
200
20
blq
20.4
25.8
12.8
15.2
18.9
13.9
25.9
23.5
13.6
12.5
21.1
17.9
20.1
477
2523
674
1021
1174
57
1.0
0.4
0.6
0.3
2.0
14a
14b
14c
14d
20
5.6
34.2
14.5
13.7
6.1/ 3.7
13.2/1.0
a The upper limit of the assay for kinetic solubility (Ksol) is 200 mM. bThere was no detectable inhibition of CYP2C9 and CYP 2D6 (IC50>20 μM). cData
for the preincubation (Pre) experiment was collected 30 min after addition of the compound to the assay system, for details on the co-incubation (Co)
experiment, see ref. 19. dSee ref. 20. eAUC was calculated over a 6h period after an oral dose of 10 mg/kg (n = 2), blq: below the limit of quantification,
see ref. 20. f B/P, brain/plasma ratio at the 6h time point.
In order to prioritize the analogues, other in vitro and in vivo ADME characteristics were also taken into account, including human
CYP inhibition, hepatocyte clearance in several species, and rat pharmacokinetics. As shown in Table 2, primary amine 8a showed
some CYP3A4 inhibition; while the secondary amine 8b had less CYP3A4 inhibition. Other amines 8c, 8f, 14a-14d and amidine 20
also showed more or less CYP3A4 liability when the compounds were preincubated. Compared to primary and secondary amine
analogues 8a-8c, tertiary amine such as 8f showed higher hepatocyte clearance across species and no plasma exposure in rats. The
pyrazine analogue 8m displayed low hepatocyte clearance but poor pharmacokinetics in rats, presumably due to its high polarity.
Compound 12, with ,-dimethyl groups, did not show improvement of hepatocyte clearance compared to compound 8b. Cyclic
secondary amine analogues 14a-14d displayed reasonably good plasma exposure (AUC6h) and brain penetration (B/P), consistent with
their robust A lowering in mice.
On the basis of overall profiles, analogue 8b was thus selected for further studies. In a pharmacokinetic study in rats, 8b showed
neither N-demethylation nor -elimination of the amine. It showed weak activity in the IonWorks assay for hERG activity: 1%
inhibition at 1 M and 21% inhibition at 10 M.21 Compound 8b demonstrated robust in vivo potency in a time course study in mice,
lowering A in plasma by -93% (6h), -35% (12h), and +50% (24h) after oral administration of a dose of 30 mg/kg. In brain, A levels
were also reduced by -73% (6h), -66% (12h) and -37% (24h), indicating good brain penetration.

In summary, we have designed a series of -aminosulfone analogues to improve the aqueous solubility of SCH 900229. The
syntheses are straightforward and efficient from the readily available SCH 900229 intermediates. All the analogues showed robust in
vitro potency and selected compounds also demonstrated good to excellent in vivo efficacy. Acceptable CYP inhibition, hepatocyte
clearance and pharmacokinetic profiles were identified with selected compounds. As expected, the -aminosulfone analogues have
higher aqueous solubility, probably due to salt formation. The design strategy described here may well find applications in other
medicinal chemistry optimization programs.
Acknowledgments
The authors would like to Dr. Jennifer Albaneze-Walker for providing a large quantity of key intermediate 2 and Dr. Malcolm
MacCoss for his support of the GSI backup program.
Abbreviations
AD, Alzheimer’s disease; A, -amyloid; PS1, presenilin-1; PS2, presenilin-2; APP, amyloid precursor protein; GSI, -secretase
inhibitor; SAR, structure-activity relationship, ADME, absorption, distribution, metabolism, and excretion; AUC, area under curve;
mCPBA, m-chloroperbenzoic acid.
References
1.
Alzheimer’s Disease Facts and Figures; Alzheimer’s Association:
Chicago, 2015; http://www.alz.org.
21. Bridal, TR.; Margulis, M.; Wang, X.; Donio, M.; Sorota. S. Assay Drug
Dev. Technol. 2010, 8, 755.
2.
3.
Hardy, J. A.; Selkoe, D. J. Science 2002, 297, 353.
Jonsson, T.; Atwal, J. K.; Steinberg, S.; Snaedal, J.; Jonsson, P. V.;
Bjornsson, S.; Stefansson, H.; Sulem, P.; Gudbjartsson, D.; Maloney, J.;
Hoyte, K.; Gustafson, A.; Liu, Y.; Lu, Y.; Bhangale, T.; Graham, R. R.;
Huttenlocher, J.; Bjornsdottir, G.; Andreassen, O. A.; Jonsson, E. G.;
Palotie, A.; Behrens, T. W.; Magnusson, O. T.; Kong, A.;
Thorsteinsdottir, U.; Watts, R. J.; Stefansson, K. Nature, 2012, 488, 96.
For recent review, see: D'Onofrio, G.; Panza, F.; Frisardi, V.; Solfrizzi,
F.; Imbimbo, B. P.; Paroni, G.; Cascavilla, L.; Seripa, D.; Pilotto, A.
Expert Opin. Drug Discov. 2012, 7, 19.
4.
5.
6.
Steiner, H.; Fluhrer, R.; Haass, C. J. Biol. Chem. 2008, 283, 29627
Lee, J.; Song, L.; Terracina, G.; Bara, T.; Josien, H.; Asberom, T.;
Sasikumar, T. K.; Burnett, D. A.; Clader, J.; Parker, E. M.; Zhang, L.
Biochemistry 2011 , 50, 4973.
7.
Borgegard, T.; Gustavsson, S.; Nilsson, C.; Parpal, S.; Klintenberg, R.;
Berg, A.-L.; Rosqvist, S.; Serneels, L.; Svensson, S.; Olsson, F.; Jin, S.;
Yan, H.; Wanngren, J.; Jureus, A.; Ridderstad-Wollberg, A.; Wollberg,
P.; Stockling, K.; Karlstrom, H.; Malmberg, A.; Lund, J.; Arvidsson, P.
I.; De Strooper, B.; Lendahl, U.; Lundkvist, J. J. Neurosci. 2012, 32,
17297.
8.
9.
Wu, W.-L.; Domalski, D.; Burnett, D. A.; Josien, H.; Bara, T.;
Rajagopalan, M.; Ruo Xu, R.; Clader, J.; Greenlee, W. J.; Brunskill, A.;
Hyde, L. A.; Del Vecchio, R. A.; Cohen-Williams, M. E.; Song, L.; Lee,
J.; Terracina, G.; Zhang, Q.; Nomeir, A.; Parker, E. M.; Zhang L.
ACS Med. Chem. Lett. 2012, 3, 892..
Wu, W.-L.; Domalski, D.; Burnett, D. A.; Josien, H.; Bara, T.;
Rajagopalan, M.; Ruo Xu, R.; Clader, J.; Greenlee, W. J.; Brunskill, A.;
Hyde, L. A.; Del Vecchio, R. A.; Cohen-Williams, M. E.; Song, L.; Lee,
J.; Terracina, G.; Zhang, Q.; Nomeir, A.; Parker, E. M.; Zhang L.
Bioorg. Med. Chem. Lett. 2013, 23, 844.
10. Charifson, P. S.; Walters, W. P. J. Med. Chem. 2014, 57, 9701.
11. Barriere, J. C.; Berthaud, N.; Beyer, D.; Dutka-Malen, S.; Paris, J. M.;
Desnottes, J. F. Curr. Pharm. Des. 1998, 4, 155.
12. Petrov, K. G.; Zhang, Y.-M.; Carter, M.; Cockerill, G. S.; Dickerson, S.;
Gauthier, C. A.; Guo, Y.; Mook, R. A., Jr.; Rusuak, D. W.; Walker, A.
L.; Wood, E. R.; Lackey, K. E. Bioorg. Med. Chem. Lett. 2006, 16,
4686.
13. Brown, H. C.; Knights, E. F.; Scouten, C. G. J. Am. Chem. Soc. 1974,
96, 7765.
14. Esteves, A. P.; Silva, M. E.; Rodrigues, L. M.; Oliveira-Campos, A. M.
F.; Hrdina, R. Tetrahedron Lett. 2007, 48, 9040.
15. Hayashi, K.; Hiki, S.; Kumagai, T.; Nagao, Y. Heterocycles 2002, 56,
433.
16. Mohammadpoor-Baltork, I.; Abdollahi-Alibeik, M. Bulletin of the
Korean Chemical Scoiety 2003, 24, 1354.
17. Zhang, L.; Song, L.; Terracina, G.; Liu, Y.; Pramanik, B.; Parker, E. M.
Biochemistry 2001, 40, 5049.
18. 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, 319, 1133.
19. Chu, I.; Favreau, L.; Soares, T.; lin, C-D.; Nomeir, A. A.; Rapid
Commun. Mass. Spectrom. 2000, 14, 207.
20. Lau, Y. Y.; Sapidou, E.; Cui, X.; White, R. E.; Cheng, K.-C.; Drug
Metab. Disp. 2002, 30, 1446.