Second generation of BACE-1 inhibitors part 2: Optimisation of the non-prime side substituent

Article abstract of DOI:10.1016/j.bmcl.2009.03.150

Our first generation of hydroxyethylamine transition-state mimetic BACE-1 inhibitors allowed us to validate BACE-1 as a key target for Alzheimer's disease by demonstrating amyloid lowering in an animal model, albeit at rather high doses. Finding a molecul

Full text of DOI:10.1016/j.bmcl.2009.03.150

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3669–3673
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Second generation of BACE-1 inhibitors part 2: Optimisation of the non-prime
side substituent
*
Nicolas Charrier, Brian Clarke, Emmanuel Demont , Colin Dingwall, Rachel Dunsdon, Julie Hawkins,
Julia Hubbard, Ishrut Hussain, Graham Maile, Rosalie Matico, Julie Mosley, Alan Naylor, Alistair O’Brien,
Sally Redshaw, Paul Rowland, Virginie Soleil, Kathrine J. Smith, Sharon Sweitzer, Pam Theobald,
David Vesey, Daryl S. Walter, Gareth Wayne
Neurology and Gastrointestinal Centre of Excellence for Drug Discovery, GlaxoSmithKline R&D, New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AW, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Our first generation of hydroxyethylamine transition-state mimetic BACE-1 inhibitors allowed us to val-
idate BACE-1 as a key target for Alzheimer’s disease by demonstrating amyloid lowering in an animal
model, albeit at rather high doses. Finding a molecule from this series which was active at lower oral
doses proved elusive and demonstrated the need to find a novel series of inhibitors with improved phar-
macokinetics. This Letter describes the discovery of such inhibitors.
Received 25 February 2009
Revised 30 March 2009
Accepted 30 March 2009
Available online 5 April 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
Alzheimer
BACE-1
Hydroxyethylamine
Aspartic protease
In the preceding paper,1we demonstrated that the chemical
space occupied by our first generation of inhibitors was unlikely
to contain a BACE-1 inhibitor with the combination of in vitro po-
tency, solubility, permeability and metabolic stability that would
provide efficacy in an animal model at low oral dose (ꢀ10–20 mg/
kg). We needed to find a novel series with similar in vitro efficacy
but which occupied an area of physicochemical space which we rea-
soned would lead to improved in vivo pharmacokinetics.
fore we did not want to modify this part of the molecule at this
stage.2 The prime side meta substituted benzylic residue was one
From X-ray crystallography studies the hydroxyethylamine core
of inhibitors such as GSK188909 (compound 1, Fig. 1) binds
strongly in the active site to the catalytic aspartyl residues there-
O
S
N
O
F
O
CF3
N
N
O
N
1
Figure 1. Structure of GSK188909.
* Corresponding author. Tel.: +44 0 1438764319; fax: +44 0 1438768232.
E-mail address: emmanuel.h.demont@gsk.com (E. Demont).
Figure 2. Non-prime side of inhibitor 9 (blue) and 10 (magenta) bound to BACE-1.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.150

3670
N. Charrier et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3669–3673
of the only active substituents identified from a large library and
was unlikely to be easily replaced.3 On the other end, key interac-
tions made by the non-prime side residue of our inhibitors are (1) a
hydrogen bond between the sulfonamide and Asn294; (2) a lipo-
philic interaction between the other meta-substituent (NHEt in
this case) and the S3 pocket of the enzyme. Abolishing either of
these interactions led to roughly a 100-fold loss of potency.4
We reasoned that if we could identify a novel template that ful-
filled these requirements, but had improved ligand efficiency, this
might lead to compounds which would not have to interact with
the prime side pockets of the enzyme to maintain good levels of
potency. These putative derivatives would therefore have a smaller
volume (described by their calculated molar refractivity, CMR), and
hence would be more likely to be permeable according to our pre-
dictive model (see preceding paper). This would also potentially
remove the prime side benzylic position, one of the known meta-
bolic sites in our inhibitors. Moreover, the new non-prime side res-
idues could themselves provide increased metabolic stability
compared to the bis meta-substituted benzamide present in
GSK188909 and its analogues.
Modelling studies suggested that the non-prime side interac-
tions described above could be realised using sulfonamides cyc-
lised back onto the adjacent phenyl ring (Scheme 1). These
studies also suggested that these interactions might also be
achieved using the corresponding tricyclic derivatives, which
might be synthesized using the chemistry already developed in
this area (see synthesis of bicyclics non-prime side residues in
the preceding paper).
The synthesis of intermediates 3, 5 and 7 which were required
to test this hypothesis is described in Scheme 2, starting from com-
mercially available materials 2 and 6, or known intermediate 4.5 To
facilitate direct comparisons and to simplify the chemistry, the sul-
fonamide was generally incorporated into a seven-membered ring,
which could be readily obtained by Michael addition or amide cou-
pling as shown below. The sulfonamide nitrogen was alkylated in
order to remove an additional and likely unnecessary, according
O
R2
R2
O
O
O
S
O
S
N
O
O
R1
R1
N
N
O
O
S
R1
N
N
N
( )m
( )m
OR
X
X
( )n
Y
Y
N
Scheme 1. Potential replacement of the bis meta-substituted non-prime side residue. n = 1,2; m = 0–2; X, Y = C, N.
O
O
S
O
O
O
O
O
N
N
O
N
N
N
N
O
O2N
Cl
O2N
N
S
O
O
O
O
f
O
O
g-j
a-c
d,e
O
N
N
N
NO2
N
2
3
O
O
O
S
O
O
N
N
N
5
N
O
O
S
O
O
O
k,l
m,n
N
N
4
O
O
O
S
O
N
O
N
O
N
O
q,r
O
O
S
o,p
O
NO2
NO2
NO2
6
7
Scheme 2. Synthesis of key intermediates for the synthesis of novel non-prime side substituent. Reagents and conditions: (a) H2SO4, MeOH, reflux, 90%; (b) Cl(CH2)2NH2ꢁHCl,
NEt3, EtOH, reflux, 86%; (c) Fe(0), AcOH, T < 30 °C; (d) CH3CHO, NaHB(OAc)3, AcOH, (CH2Cl)2, 25 °C, 71% (two steps); (e) NH4COOH, Pd/C, MeOH, reflux, 98%; (f)
tBuOCOCH2SO2Cl, pyridine, DMAP, CH2Cl2, 25 °C then NEt3, EtOH, reflux, 20%; (g) CH3I, K2CO3, DMF, 25 °C, 100%; (h) 4 N HCl, dioxane, rt, 98%; (i) EDAC, DMF, 25 °C, 60%; (j)
BH3, THF, reflux, 98%; (k) Cl(CH2)2SO2Cl, pyridine, DMAP, CH2Cl2, 25 °C; (l) NEt3, CH2Cl2, 25 °C; (m) NaH, DMF, 25–100 °C, 85% (three steps); (n) NaH, MeI, DMF, 25 °C, 76%; (o)
Cl(CH2)2SO2Cl, NEt3, CH2Cl2, 25 °C then K2CO3, DMF, 50 °C, 30% (two steps); (p) K2CO3, CH3I, DMF, 25 °C, 92%; (q) K2CO3, DMF, 50 °C, 73%; (r) NH4COOH, Pd/C, MeOH, reflux;
98%.

N. Charrier et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3669–3673
3671
to modelling based on previous co-crystal structures of inhibitors
bound to BACE-1, hydrogen-bond donor.2
A comparison of potency and selectivity of a first set of inhibi-
tors, incorporating these novel bicyclic and tricyclic non-prime
side substituents is presented in Table 1.6
Inhibitors containing the tricyclic residue C or the bicyclic resi-
due D were typically less potent than those containing residues A
or B (compare 10–11, 14, 16, 19–20 with 8–9, 12–13, 15 and 17–
18, respectively). Inhibitors containing residue B were more potent
against BACE-1 than inhibitors containing residue A, and more
selective against Cat-D but less so against BACE-2 (compare 9, 13
and 18 with 8, 12 and 17, respectively). In contrast to what had
been observed in the bis meta-substituted series, it was also now
possible to achieve excellent BACE-1 inhibition without incorpo-
rating large meta-substituted benzyl prime side residues (com-
pounds 15 and 18). These findings focused our attention on this
new series of inhibitors bearing indole-like tricyclic non-prime
side residues.
Figure 3. Non-prime side of inhibitor 9 (blue) and 11 (green) bound to BACE-1.
In an attempt to understand the results presented above, a rep-
resentative of each novel series was co-crystallized with a BACE-1
construct and the structures were elucidated by X-ray crystallogra-
phy.7 The non-prime sides of these derivatives bound to BACE-1
are shown below (Figs. 2 and 3).
No significant displacements of enzyme residues or changes to
the inhibitor binding motifs were seen and it is difficult to rational-
ise the observed differences in potency on this basis.
to developing the SAR of the sulfonamide ring. A comparison of
activity between the six- and seven-membered rings is high-
lighted below (Table 2). Inhibitors bearing a 6,6,5 tricycle non-
prime side residue demonstrated nanomolar potency against
BACE-1, particularly when interacting with the S20 pocket (com-
pound 21). However, truncation of the prime side residue results
in a 10-fold drop in activity relative to the 7,6,5 tricyclic inhibitors
However, we were able to prioritise work on one of the novel
series and to develop this series further. We turned our attention
Table 1
Activity and selectivity of selected BACE-1 inhibitors
O
CF3
O
S
N
O
O
E
R2
F
R1
N
N
A
N
CF3
O
G
H
O
S
O
S
O
S
N
N
N
N
O
O
O
N
N
HN
D
C
B
R1
R2
BACE-1 IC50
(
lM)
BACE-2 IC50
(lM)
Cat-D IC50
(
lM)
Ab40 IC50 (lM)
a
b
b
c
Compd
8
9
A
B
C
D
A
B
D
B
D
A
B
C
D
E
0.007
0.002
0.026
0.019
0.091
0.015
0.19
0.007
0.026
0.12
4.26 (589)
0.059 (24)
3.8 (146)
3.59 (189)
11.19 (123)
0.49 (33)
18.71 (98)
0.028 (4)
1.23 (47)
19.5 (163)
0.263 (13)
16.22 (12)
13.55 (90)
6.76 (933)
0.019
0.006
0.084
0.056
—
0.090
1.185
0.008
0.180
—
1.259 (525)
2.24 (86)
1.21 (64)
19.95 (219)
6.61 (441)
4.1 (22)
0.89 (127)
1.07 (41)
30.2 (252)
7.93 (397)
26.3 (48)
8.15 (54)
10
11
12
13
14
15
16
17
18
19
20
F
G
H
0.020
0.55
0.15
0.016
—
1.574
a
In all tables, IC50S reported are means of the values of 3 different experiments. Each IC50 is within threefold of the mean value.
Data in parentheses indicates compound selectivity as ratio of enzyme and BACE-1 IC50S.
In SHSY5Y wild type cells. See previous paper for details.
b
c

3672
N. Charrier et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3669–3673
Table 2
Comparison of BACE-1 activity for 6,6,5 and 7,6,5 tricyclic non-prime side
O
O
O
R1
N
N
J
O
S
( )n
I
N
N
O
L
K
a
a
Compd
N
R1
I
BACE-1 IC50
(
lM)
BACE-2 IC50
(l
M)
Cat-D IC50
1.00 (125)
(l
M)
Ab40 IC50 (lM)
21
22
23
24
25
26
27
28
0
1
0
1
0
1
0
1
0.008
0.002
0.023
0.005
0.060
0.009
0.091
0.015
0.35 (44)
0.096 (48)
0.39 (17)
0.11 (23)
0.62 (10)
0.21 (23)
1.17 (13)
1.26 (84)
—
2.95 (1476)
0.78 (34)
1.00 (200)
0.30 (5)
0.66 (73)
6.61 (73)
0.012
0.129
0.072
—
0.093
0.868
0.106
J
K
L
58.88 (3925)
a
Data in parentheses indicates compound selectivity as ratio of enzyme and BACE-1 IC50S.
O
O
O
O
O
S
O
N
O2N
S
O
O
N
N
O
O2N
h, i
d-g
a-c
O
O
O
N
N
N
29
30
31
Scheme 3. Synthesis of key intermediate for the synthesis of 8,6,5 tricyclic non-prime side residue. Reagents and conditions: (a) I2, AgSO4, MeOH, 25 °C, 95%; (b)
nC3H7CCSi(CH3)3, Pd(OAc)2, LiCl, K2CO3, DMF, 100 °C, 53%; (c) TBAF, THF, 0 °C, 89%; (d) Allyl bromide, NaH, DMF, 25 °C, 95%; (e) Fe, AcOH, 60 °C, 96%; (f) Cl(CH2)2SO2Cl,
pyridine, DMAP, CH2Cl2, 25 °C, 75%; (g) CH3I, K2CO3, DMF, 25 °C, 90%; (h) Grubbs II, toluene, reflux, 52%; (i) NH4COOH, Pd/C, MeOH, reflux; 73%.
(compare 23, 25 and 27 with 24, 26 and 28, respectively). The
6,6,5 derivatives also appeared to be less selective against BACE-
2 and Cat-D.
type are less potent and less selective that their 7,6,5 counter-
parts (compare 32 and 33 with 22 and 18, respectively). Overall,
the 7,6,5 ring system consistently provided the most potent
inhibitors.
The effect of the sulfonamide nitrogen substituent was also ex-
plored (Table 4). The presence of an acidic hydrogen is somewhat
detrimental to activity (compare 22 and 34). Methyl and ethyl sub-
stitution afforded a similar profile (Compare 13, 26 and 18 with 35,
38 and 39, respectively) whilst branched or more lipophilic alkyl
chains did not seem to provide any advantage in terms of potency
(compounds 36 and 37).
Comparison between the 8,6,5 and 7,6,5 tricyclic derivatives
proved less straightforward due to the chemical inaccessibility of
the former system (Scheme 3): Indole 29 was obtained by Larock
cyclisation.5 Sulfonamide 30 could be engaged in a ring-closing
metathesis reaction and reduction of the vinyl sulfonamide formed
led to intermediate 31. The chemistry to synthesise the desired
hydroxyethylamine BACE-1 inhibitors from 31 has already been
described.2
A comparison of activity between inhibitors bearing 8,6,5 and
7,6,5 tricyclic non-prime side residues is shown in Table 3. Again,
nanomolar potency could be achieved against BACE-1 with the
extended sulfonamide (compound 32). However, inhibitors of this
Overall, the data presented highlight the unique profile of this
novel series of hydroxyethylamine transition-state mimetics bear-
ing indolic 7,6,5 tricyclic non-prime side substituents, both in
terms of potency and selectivity, particularly against Cat-D. A de-
Table 3
Comparison of BACE-1 activity for 8,6,5 and 7,6,5 tricyclic non-prime side
a
b
b
Compd
Na
R1
I
BACE-1 IC50
(lM)
BACE-2 IC50
(l
M)
Cat-D IC50
(lM)
Ab40 IC50 (lM)
22
32
18
33
1
2
1
2
0.002
0.028
0.020
0.49
0.096 (48)
1.86 (66)
0.263 (13)
5.46 (11)
2.95 (1476)
1.57 (56)
7.93 (397)
2.9 (6)
0.012
0.339
0.016
—
H
a
See Table 2 for definition of n and R1.
Data in parentheses indicates compound selectivity as ratio of enzyme and BACE-1 IC50S.
b

N. Charrier et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3669–3673
3673
Table 4
Comparison of BACE-1 activity for N-substituted 7,6,5 tricyclic sulfonamides BACE-1 inhibitors
R1
O
O
S
R2
N
O
N
N
O
N
a
a
Compd
R1
R2
I
BACE-1 IC50
(l
M)
BACE-2 IC50
(l
M)
Cat-D IC50
(lM)
Ab40 IC50 (lM)
22
34
13
35
36
37
26
38
18
39
CH3
H
CH3
C2H5
CH(CH3)2
CH2CF3
CH3
C2H5
CH3
0.002
0.014
0.015
0.005
0.023
0.029
0.009
0.007
0.020
0.02
0.096 (48)
1.047 (76)
0.49 (33)
0.15 (30)
0.94 (41)
0.68 (23)
0.21 (23)
0.089 (13)
0.263 (13)
0.331 (17)
2.95 (1476)
14.12 (1023)
6.61 (441)
1.62 (324)
2.96 (129)
5.01 (173)
0.66 (73)
0.15 (21)
7.93 (397)
12.30 (617)
0.012
0.260
0.090
0.048
0.227
0.248
0.093
0.030
0.016
0.017
F
K
H
C2H5
a
Data in parentheses indicates compound selectivity as ratio of enzyme and BACE-1 IC50S.
O’Brien, A.; Redshaw, S.; Riddell, D.; Rowland, P.; Soleil, V.; Smith, K. J.; Stanway,
S.; Stemp, G.; Sweitzer, S.; Theobald, P.; Vesey, D.; Walter, D. W.; Ward, J.;
Wayne, G. Bioorg. Med. Chem. Lett. 2008, 18, 1022. 1017.
tailed account of our findings in this series will be the subject of
the following publication.
4. Clarke, B.; Demont, E.; Dingwall, C.; Dunsdon, R.; Faller, A.; Hawkins, J.; Hussain,
I.; MacPherson, D.; Maile, G.; Matico, R.; Milner, P.; Mosley, J.; Naylor, A.;
O’Brien, A.; Redshaw, S.; Riddell, D.; Rowland, P.; Soleil, V.; Smith, K. J.; Stanway,
S.; Stemp, G.; Sweitzer, S.; Theobald, P.; Vesey, D.; Walter, D. W.; Ward, J.;
Wayne, G. Bioorg. Med. Chem. Lett. 2008, 18, 1022. 1011.
5. Charrier, N.; Demont, E.; Dunsdon, R.; Maile, G.; Naylor, A.; O’Brien, A.; Redshaw,
S.; Theobald, P.; Vesey, D. S.; Walter, D. W. Synthesis 2006, 3467.
6. For IC50 < 10 nM, the FRET assay was considered as not reliable for SAR, which
was then driven by cellular activity; All compounds tested had purity > 95% by
LC–MS and/or NMR.
References and notes
1. Charrier, N.; Clarke, B.; Cutler, L.; Demont, E.; Dingwall, C.; Dunsdon, R.;
Hawkins, J.; Howes, C.; Hussain, I.; Maile, G.; Matico, R.; Mosley, J.; Naylor, A.;
O’Brien, A.; Redshaw, S.; Rowland, P.; Soleil, V.; Smith, K. J.; Sweitzer, S.;
Theobald, P.; Vesey, D.; Walter, D. W.; Wayne, G. Bioorg. Med. Chem. Lett. 2009.
doi:10.1016/j.bmcl.2009.03.165.
2. Beswick, P.; Charrier, N.; Clarke, B.; Demont, E.; Dingwall, C.; Dunsdon, R.; Faller,
A.; Gleave, R.; Hawkins, J.; Hussain, I.; Johnson, C. N.; MacPherson, D.; Maile, G.;
Milner, P.; Naylor, A.; O’Brien, A.; Redshaw, S.; Riddell, D.; Rowland, P.;
Skidmore, J.; Soleil, V.; Smith, K.; Stanway, S.; Stemp, G.; Stuart, A.; Theobald,
P.; Vesey, D.; Walter, D. S.; Ward, J.; Wayne, G. Bioorg. Med. Chem. Lett. 2008, 18,
1022.
7. The PDB deposition codes and refinement details for the BACE-1 complex crystal
structures are 9 (2wf1, 1.6 Å resolution, R = 0.184, Rfree = 0.207); 10 (2wf2, 1.8 Å
resolution, R = 0.208, Rfree = 0.247); 11 (2wf3, 2.1 Å resolution, R = 0.164,
Rfree = 0.217).
3. Clarke, B.; Demont, E.; Dingwall, C.; Dunsdon, R.; Faller, A.; Hawkins, J.; Hussain,
I.; MacPherson, D.; Maile, G.; Matico, R.; Milner, P.; Mosley, J.; Naylor, A.;