Carbamate-appended N-alkylsulfonamides as inhibitors of γ-secretase

Article abstract of DOI:10.1016/j.bmcl.2007.11.102

The synthesis and γ-secretase inhibition data for a series of carbamate-appended N-alkylsulfonamides are described. Carbamate 54 was found to significantly reduce brain Aβ in transgenic mice. 54 was also found to possess markedly improved brain levels in transgenic mice compared to previously disclosed 1 and 2.

Full text of DOI:10.1016/j.bmcl.2007.11.102

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 464–468
Carbamate-appended N-alkylsulfonamides
as inhibitors of c-secretase
Carl P. Bergstrom,a,* Charles P. Sloan,a Wai-Yu Lau,a David W. Smith,a Ming Zheng,b
Steven B. Hansel,b,ꢀ Craig T. Polson,c Jason A. Corsa,c Donna M. Barten,c
Kevin M. Felsensteinc,ꢁ and Susan B. Robertsc
aDepartment of Discovery Chemistry, Research and Development, Bristol-Myers Squibb,
5 Research Parkway, Wallingford, CT 06492, USA
bDepartment of Drug Metabolism and Pharmacokinetics, Research and Development,
Bristol-Myers Squibb, 5 Research Parkway, Wallingford, CT 06492, USA
cDepartment of Neuroscience Biology, Bristol-Myers Squibb, Research and Development,
5 Research Parkway, Wallingford, CT 06492, USA
Received 26 October 2007; revised 21 November 2007; accepted 27 November 2007
Available online 3 December 2007
Abstract—The synthesis and c-secretase inhibition data for a series of carbamate-appended N-alkylsulfonamides are described. Car-
bamate 54 was found to significantly reduce brain Ab in transgenic mice. 54 was also found to possess markedly improved brain
levels in transgenic mice compared to previously disclosed 1 and 2.
Ó 2007 Elsevier Ltd. All rights reserved.
Alzheimer’s Disease (AD) is a progressive dementing
neurodegenerative disorder characterized pathologically
by the presence of plaques composed of the 40–42 amino
acid peptide amyloid-b (Ab) and by increased levels of
soluble Ab.1 The evidence is mounting that it is the
soluble oligomeric forms of Ab that are the primary
neurotoxic agents in AD.2 The sequential action of
b- and c-secretases is responsible for the cleavage of
b-amyloid precursor protein (APP) to release Ab
peptides.3 We have been focusing on the identification
of small molecules that inhibit c-secretase cleavage of
the C-terminal fragments of b-APP. The resulting inhibi-
tion of Ab production has the potential to offer a therapy
that would slow or halt the progression of AD.4
and 2 as potent inhibitors of c-secretase, using a cell-
based assay.6 Compounds 1 and 2 were both selected
for oral administration in Tg2576 bAPP-Swedish trans-
genic mice at a single 200 lmol/kg dose.7 Three hours
after dosing methyl sulfonamide 1, a 27% reduction in
brain Ab was observed. Similarly, 3 h after dosing tetra-
zole 2, a 41% reduction in brain Ab was observed. It was
speculated that the improved Ab reduction demon-
strated by tetrazole 2 was directly related to its higher
mean brain concentration compared to methyl sulfon-
amide 1. For this reason we continued to examine this
sulfonamide chemotype in an attempt to identify ana-
logs possessing both improved absolute brain levels
and improved brain to plasma ratios (Fig. 1).
In a previous communication,5 we described the identi-
fication of nitrogen-appended N-alkylsulfonamides 1
OH
OH
Cl
Cl
N
S
O
Cl
Cl
N
S
O
Keywords: Alzheimer’s disease; Secretase; Amyloid-b; Sulfonamides.
O
O
*
Corresponding author. Tel.: +1 2036776295; fax: +1 2036777702;
e-mail: carl.bergstrom@bms.com
N
S
O
N
N
O
N
ꢀ
N
Present address: Pfizer Inc., 2800 Plymouth Road, Ann Arbor, MI
48105, USA.
1: Aβ 40 IC50 = 0.32 nM
2 : Aβ 40 IC50 = 0.51 nM
ꢁ
Present address: Johnson and Johnson PRD, Spring House, PA
19477, USA.
Figure 1.
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.102

C. P. Bergstrom et al. / Bioorg. Med. Chem. Lett. 18 (2008) 464–468
465
Table 1. c-Secretase inhibition for carbamates 12–22
One of many strategies explored to achieve this goal was
to study carbamates that are structurally closely related
to the nitrogen-appended N-alkylsulfonamides 1 and 2.
Herein we report the SAR for this series of c-secretase
inhibitors.
OH
O
Cl
Cl
N
S
O
O
Aryl carbonate 11 was selected as the key synthon for
the synthesis of a targeted set of carbamates. Mitsun-
obu8 conditions were employed to effect the coupling
of arylsulfonamide 8 with (S)-4-[[dimethyl(1,1-dimethyl-
ethyl)-silyl]oxy]-2-propanol.9 The resulting silyl ether 9
was deprotected with TBAF to produce alcohol 10,
which was then acylated with 4-nitrophenyl chlorofor-
mate to produce 11 (Scheme 1).
NR2R3
O
Compound
Ab40 IC50 (nM)10
NR2R3
12
13
15
10
HN
A small collection of commercially available amines
along with 3-(1H-imidazol-1-yl)propan-1-amine 5 and
HN
HN
3-(2H-tetrazol-2-yl)propan-1-amine
7 were acylated
with aryl carbonate 11. The THP protecting group on
the benzyl alcohol was then removed as shown in
Scheme 1 to produce carbamates 12–22. The steric
requirements for c-secretase inhibition did not appear
to be stringent as methyl carbamate 12 was nearly equi-
potent to tert-butyl carbamate 20 as shown in Table 1.
The most potent carbamates in this set were imidaz-
ole-appended 15 and tetrazole-appended 16.
14
15
16
17
18
19
20
21
22
10
N
2.3
2.1
12
HN
HN
N
N
N
N
N
Imidazole 15 was selected for further analog studies due
to its basic properties that allowed for salt formulations.
HN
HN
HN
N
Br
15
O
N
N
O
N
c
a
b
N
N
H2N
O
5
3
3
4
6
O
5.6
12
N
N
c
N
N
Br
O
N
O
N
N
N
N
O
HN
N
N
H2N
7
O
18
OTHP
OTHP
OTHP
N
S O
O
d
f
N
1.8
O
Cl
Cl
NH
S O
O
Cl
Cl
N
S O
O
Cl
Cl
OR1
O
O
11
8
9 : R1 = TBS
10 : R1 = H
e
Initially, chain length analogs of 15 were examined to
determine whether its potency could be improved. As
part of this objective, imidazole 26 was synthesized from
2-(1H-imidazol-1-yl)ethanan-1-amine 25 as shown in
Scheme 2. 26 was found to be modestly more potent
than 15 as shown in Table 2.11 Chain length analogs 9
and 30 were also synthesized (Scheme 2) and found to
be no more potent than 15 (Table 2).
NO2
OR4
N
S O
O
h
R4 = THP
g
O
O
Cl
12-22 : R4 = H
NR2R3
Cl
Scheme 1. Reagents and conditions: (a) Imidazole, K2CO3, CH3CN,
reflux, 16 h, 74%; (b) Tetrazole, K2CO3, CH3CN, reflux, 12 h, 47%; (c)
Hydrazine, aq EtOH, reflux, 16 h, 83%; (d) (S)-1-[[Dimethyl(1,1-
dimethylethyl)-silyl]oxy]-2-propanol9, DIAD, PPh3, THF, rt, 16 h,
68%; (e) TBAF, THF, rt, 3 h, 94%; (f) 4-Nitrophenyl chloroformate,
py, THF, rt, 6 h, 85%; (g) HNR2R3, DMF, rt, 8 h; (h) 1 M HCl, THF,
rt, 6 h.
After exploring the ether length of imidazole-appended
carbamate 15, a small set of tertiary carbamates were
synthesized in order to compare to secondary carbamate
15. Amine 5 was reductively alkylated to form second-
ary amines 32–34 according to Scheme 3. These amines
were acylated with carbonate 11 using standard condi-

466
C. P. Bergstrom et al. / Bioorg. Med. Chem. Lett. 18 (2008) 464–468
O
O
N
N
Br
N
a
31: R4=H
b
N
N
N
N
N
a
N
N
N
32: R4=CH3
H2N
R4
H2N
H2N
N
H
33: R4=cyclopropyl
O
O
23
24
25
3
3
N
N
OTHP
OH
N
b
c,d
O
Cl
Cl
N
S O
O
N
H
O
Cl
Cl
N
S O
O
O
34
O
HN
O
N
N
OTHP
N
11
26
35: R5=Et
OH
36: R5=CH2
NO2
c,d
37: R5=
i
Pr
O
Cl
Cl
O
S O
O
OTHP
OTHP
N
OH
O
Cl
Cl
N
S O
O
O
O
e
N
h,i,j
OR1
11
Cl
Cl
N
Cl
Cl
NH
S O
O
R5
Cl
O
N
S O
S O
NH
NO2
O
N
O
O
Cl
(CH2)n
N
N
27: R1 = TBS
28 : R1 = H
f
Scheme 3. Reagents and conditions: (a) RCH2C(O)H, Na(CN)BH3,
HCl/MeOH, reflux, 16 h; (b) Acetone, Na(OAc)3BH, AcOH, CH2Cl2,
rt, 16 h; 31%; (c) 32–34, DMF, rt, 8 h; (d) 1 M HCl, THF, rt, 6 h.
8
29: n=1
30: n=2
Scheme 2. Reagents and condtions: (a) Imidazole, K2CO3, CH3CN,
reflux, 16 h, 69%; (b) Hydrazine, aq EtOH, reflux, 16 h; 74%; (c) 25,
DMF, rt, 8 h, 69%; (d) 1 M HCl, THF, rt, 6 h, 76%; (e) (S)-4-
[[Dimethyl(1,1-dimethylethyl)-silyl]oxy]-2-butanol9, DIAD, PPh3,
THF, rt, 16 h, 67%; (f) TBAF, THF, rt, 3 h, 93%; (h) 4-Nitrophenyl
chloroformate, py, THF, rt, 6 h, 86%; (i) 5/25, DMF, rt, 8 h; (j) 1 M
HCl, THF, rt, 6 h.
Table 3. c-Secretase inhibition for imidazoles 15, 35–37
OH
O
N
N
Cl
Cl
N
S
O
O
O
N
Table 2. c-Secretase inhibition for imidazoles 15, 26, 29, 30
R5
Compound
Ab40
IC50 (nM)10
Compound
R5
Ab40 IC50 (nM)10
15
35
36
37
H
Et
2.3
OH
N
0.06
0.06
CH2
iPr
19
O
Cl
Cl
O
15
2.3
0.93
4.0
7.7
S
O
HN
O
N
tions to produce tertiary carbamates 35–37. The most
potent of these inhibitors were ethyl carbamate 35 and
methylene cyclopropyl carbamate 36 (Table 3).
N
OH
N
Carbamates 35 and 36 were selected for oral administra-
tion in Tg2576 mice at a single 200 lmol/kg dose. Three
hours after dosing 35, a 15% reduction in brain Ab was
observed. The plasma level of 35 was 730 296 nM and
the brain level was 227 3 nM. Similarly, 3 h after dos-
ing 36, no reduction in brain Ab was observed. The plas-
ma level of 36 was 6830 2898 nM and the brain level
was 859 49 nM. It was hypothesized that the combina-
tion of the polar imidazole and the benzyl alcohol func-
tionalities may have been responsible for the reduced
brain efficacies of 35 and 36 compared to 2. For that rea-
son, we sought to explore the removal of the benzyl
alcohol functionality in the general chemotype repre-
sented by 35 and 36.
O
26
29
30
Cl
Cl
O
S
O
HN
O
N
N
OH
Cl
Cl
N
O
S
O
NH
O
O
N
N
OH
N
Cl
Cl
Toward that end, the 2,5-difluoro, the 2-fluoro-5-chloro,
and the 2,5-dichloro analogs of the arylsulfamido carba-
mate 15 were synthesized as shown in Scheme 4. The
most potent of these new compounds was the 2,5-di-
chloro analog 52 (Table 4). It is worth noting that aryl
O
S
O
NH
O
N
O
N

C. P. Bergstrom et al. / Bioorg. Med. Chem. Lett. 18 (2008) 464–468
467
X
X
X
N
S O
O
a
b,c
d
Y
NH
S O
O
Y
NH2
Y
OH
38:X=Y=F
39:X=F, Y=Cl
40:X=Y=Cl
Cl
Cl
44:X=Y=F
45:X=F, Y=Cl
46:X=Y=Cl
41:X=Y=F
42:X=F, Y=Cl
43:X=Y=Cl
X
X
O
Y
N
S O
O
O
O
Y
N
S O
O
e
50 - 54
(see Tables 4 and 5)
O
O
N
Cl
R5
Cl
47:X=Y=F
48:X=F, Y=Cl
49:X=Y=Cl
N
N
NO2
Scheme 4. Reagents and conditions: (a) p-ClPhSO2Cl, py, CH2Cl2, rt, 8 h; (b) (S)-1-[[Dimethyl(1,1-dimethylethyl)-silyl]oxy]-2-propanol9, DIAD,
PPh3, THF, rt, 16 h; (c) TBAF, THF, rt, 3 h; (d) 4-Nitrophenyl chloroformate, py, THF, rt, 6 h; (e) 5, 31–32, DMF, rt, 8 h.
brain Ab was a significant improvement over what had
been produced by benzyl alcohol analogs 35 and 36.
Table 4. c-Secretase inhibition for imidazoles 50–52
X
In summary, we have described the SAR for a series of
sulfonamide inhibitors c-secretase exemplified by carba-
mate 12 and its analogs. Tertiary carbamates containing
a tethered imidazole such as compounds 35, 36, and 54
are the most potent c-secretase inhibitors identified thus
far in this carbamate series. Aryl chloride 54 generated a
more pronounced decrease in brain Ab in transgenic
mice relative to the benzyl alcohol analogs 35 and 36 de-
spite the superior potency of 35 and 36. This may be due
to improved brain concentrations of 54 over 35 and 36.
54 also achieved a significantly improved brain to plas-
ma ratio and a higher absolute brain concentration com-
pared to the previously disclosed sulfonamide 2;
accompanied by a similar reduction of brain Ab in mice.
Strategies for improved brain efficacy in this class are
being pursued and will be discussed in future
communications.
O
N
N
Y
N
S
O
O
O
HN
Cl
Compound
X
Y
Ab40 IC50 (nM)10
50
51
52
F
F
1.2
F
Cl
Cl
Cl
0.66
0.41
Table 5. c-Secretase inhibition for imidazoles 53–54
Cl
O
N
N
Cl
Cl
N
S
O
O
O
N
R5
References and notes
Compound
R5
Ab40 IC50 (nM)10
52
53
54
H
0.41
0.48
0.27
1. (a) Olson, R. E.; Copeland, R. A.; Seiffert, D. Curr. Opin.
Drug Disc. Dev. 2001, 4, 390; (b) Walsh, D. M.; Selkoe, D.
J. Neuron 2004, 44, 181; (c) Kobayashi, D. T.; Chen, K. S.
Genes Brain Behav. 2005, 4, 173.
Me
Et
2. (a) Gong, Y.; Chang, L.; Viola, K. L.; Lacor, P. N.;
Lambert, M. P.; Finch, C. E.; Krafft, G. A.; Klein, W. L.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10417; (b) Haas,
C.; Selkoe, D. J. Nat. Rev. Mol. Cell Biol. 2007, 8, 101.
3. (a) Selkoe, D. J. Physiol. Rev. 2001, 81, 741; (b) Haass, C.
EMBO J. 2004, 23, 483.
4. Esler, W. P.; Wolfe, M. S. Science 2001, 293, 1449.
5. Bergstrom, C. P.; Sloan, C. P.; Wang, H. H.; Parker,
M. F.; Smith, D. W.; Zheng, M.; Hansel, S. B.; Polson,
C. T.; Guss, V. L.; Corsa, J. A.; Barten, D. M.;
Felsenstein, K. M.; Roberts, S. B. Bioorg. Med. Chem.
Lett. 2008, 18, 175.
6. H4 human neuroglioma cells expressing HPLAPÆ-
bAPP164SFAD were grown in high glucose (4.5 g/L)
DMEM (Invitrogen) supplemented with 10% FBS,
100 lg/mL pen-strep, 2 mM glutamine, and 100 lg/mL
geneticin. Cells were aliquoted into a 96-well plate, and
after attachment the medium was replaced with Ultracul-
chloride 52 was roughly five times more potent than the
benzyl alcohol analog 15.
Next, N-alkylated analogs of 52 were synthesized, as de-
scribed in Scheme 4, to determine whether a similar in-
crease in potency could be achieved as had been
previously observed with N-alkylated analogs of 15.
Table 5 shows that methyl carbamate 53 was roughly
equipotent to 52, while ethyl carbamate 54 was modestly
more potent than 52. Compound 54 was selected for
oral administration in Tg2576 mice at
a single
200 lmol/kg dose. Three hours after dosing, a 50%
reduction in brain Ab was observed. The plasma level
of this compound was 98,942 60,560 nM and the brain
level was 41,171 18,463 nM. This 50% reduction in

468
C. P. Bergstrom et al. / Bioorg. Med. Chem. Lett. 18 (2008) 464–468
ture (Whittaker Bioproducts) containing individual com-
9. (S)-3-[[Dimethyl(1,1-dimethylethyl)-silyl]oxy]-2-propanol
was synthesized as follows. Commercially available (S)-
1,2-propanediol (2.09 g, 27.5 mmol) was stirred with
TBSCl (4.55 g, 30.2 mmol), TEA (4.59 mL, 33.0 mmol),
and DMAP (0.335 g, 2.75 mmol) in CH2Cl2 (55 mL) at
0 °C for 4 h. The resulting mixture was concentrated and
purified by silica gel column chromatography eluting with
4:1 hexanes/ethyl acetate to isolate the title compound in
74% yield. (S)-4-[[dimethyl(1,1-dimethylethyl)-silyl]oxy]-2-
butanol was synthesized in an analogous fashion from (S)-
1,3-butanediol in 81% yield.
10. IC50s were determined using a cell-based assay (see Ref. 6).
Values are means of two experiments, with 12 drug
concentrations in each experiment; intra-assay variance
was <10%.
11. It was decided not to examine 26 further due to its
instability in solution.
pounds of interest (final DMSO concentration of 1%).
After an overnight incubation, the conditioned medium
was removed and evaluated for the presence of Ab in a
sandwich ELISA using a monoclonal C-terminal Ab40
specific capture antibody and an HRP labeled monoclonal
antibody to the N-terminus of Ab for detection. The
endpoint measurement of Ab1–40 level was developed
using TMB reagent followed by the addition of 1 M
phosphoric acid. The plates were read at 450 nm.
7. Barten, D. M.; Guss, V. L.; Corsa, J. A.; Loo, A.; Hansel,
S. B.; Zheng, M.; Munoz, B.; Srinivasan, K.; Wang, B.;
Robertson, B. J.; Polson, C. T.; Wang, J.; Roberts, S. B.;
Hendrick, J. P.; Anderson, J. J.; Loy, J. K.; Denton, R.;
Verdoorn, T. A.; Smith, D. W.; Felsenstein, K. M.
J. Pharmacol. Exp. Ther. 2005, 312, 635.
8. Mitsunobu, O. Synthesis 1981, 1.