Design and synthesis of potent and selective BACE-1 inhibitor

Article abstract of DOI:10.1021/jm901168f

Highly potent BACE-1 protease inhibitors have been developed from an inhibitors containing a hydroxyethylene (HE) core displaying aryloxymethyl or benzyloxymethyl P1 side chain and a methoxy P1′ side chain. The target molecules were synthesized in good overall yields from chiral carbohydrate starting materials. The inhibitors show high BACE-1 potency and good, selectivity against cathepsin D, where the most potent inhibitor furnishes BACE-1 K_i ? 1 nM and displays > 1000-fold selectivity over cathepsin D.

Full text of DOI:10.1021/jm901168f

1458 J. Med. Chem. 2010, 53, 1458–1464
DOI: 10.1021/jm901168f
Design and Synthesis of Potent and Selective BACE-1 Inhibitors†
Catarina Bjorklund, Stefan Oscarson,‡,§ Kurt Benkestock, Neera Borkakoti, Katarina Jansson, Jimmy Lindberg,
‡
€
Lotta Vrang, Anders Hallberg, Asa Rosenquist, and Bertil Samuelsson*
^
,‡,
˚
‡Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University, SE-106 91 Stockholm, Sweden, §School of Chemistry and
Chemical Biology, UCD, Dublin 4, Ireland, Medivir AB, P.O. Box 1086, SE-141 22 Huddinge, Sweden, and ^Department of Medicinal
Chemistry, BMC, Uppsala University, Box 574, SE-751 23 Uppsala, Sweden
Received May 4, 2009
Highly potent BACE-1 protease inhibitors have been developed from an inhibitors containing a
hydroxyethylene (HE) core displaying aryloxymethyl or benzyloxymethyl P1 side chain and a methoxy
P10 side chain. The target molecules were synthesized in good overall yields from chiral carbohydrate
starting materials. The inhibitors show high BACE-1 potency and good selectivity against cathepsin D,
where the most potent inhibitor furnishes BACE-1 Ki , 1 nM and displays >1000-fold selectivity over
cathepsin D.
1. Introduction
and a methoxy group as the P10 side chain, that furnishes
highly potent BACE-1 inhibitors and inhibitors displaying
good to excellent selectivity over the antitarget cathepsin D.17
These new HE scaffolds have been investigated using the
5-substituted isophthalamide moiety of inhibitor 2 or using
a di-N-propylisophthalamide in the P2-P3 position.16,18 The
most potent inhibitor discovered in this series, 1d (Figure 1),
exhibits an impressive BACE-1 IC50 of 0.3 nM and Ki ,1 nM,
hence, >1000-fold selectivity over cathepsin D. Insertion of
an oxygen atom at P10 improved activity versus BACE-1
and selectivity versus cathepsin D, compared with inhibitor
3 previously synthesized in our laboratory having a P10 methyl
group.
Alzheimer’s disease (ADa) is a disabling, progressive, and
ultimately fatal form of dementia afflicting approximately
40% of the population over 80 years, with over 30 million
people suffering from AD worldwide.1,2 The neuropatho-
logical features of AD are complex, but one of the major
factors associated with AD is the buildup of amyloid plaques
and neurofibrillary tangels in the brain tissue.3 The primary
component of the amyloid plaques is aggregated amyloid-β
(Aβ) which is the cleavage product of amyloid precursor
protein (APP), a transmembrane protein,3 by the action of
β- and γ-secretase. In vitro, the aggregation state of Aβ has
been shown to influence neurotoxicity,4,5 and in neuritic
plaque found in the brain of AD patients, amyloid-β42
(Aβ1-42) dominates.6
2. Result and Discussion
BACE-1 knockout mice, transgenicfor human APP, donot
show any apparent adverse phenotype or any Aβ buildup in
the brain.7-9 These data have helped to validate BACE-1 as a
suitable drug target for Alzheimer’s disease and have led to an
intense search to identify safe and efficacious drugs based on
selective BACE-1 inhibition.10-15
Several groups have reported on BACE-1 inhibitors con-
taining substituted isophthalamides as highly promising P2-
P3 groups.10 Recently Stachel et al. reported that 5-substi-
tuted isophthalamides coupled to HEA cores furnish potent
and selective BACE-1 inhibitors.16 One of these inhibitors, 2
(Figure 1), displayed an IC50 of 11 nM.
2.1. Chemistry. The target compounds 1a-l were prepared
as outlined in Schemes 1 and 2. The diol 419-21 was reacted
with dibutyltin oxide in refluxing toluene, furnishing the tin
acetal which was subsequently reacted with either 4-bromo-
benzyl bromide or 3,5-difluorobenzyl bromide in the pre-
sence of tetrabutylammonium bromide, yielding the two
corresponding benzylated products 5a and 5b in 87% and
78% yields, respectively (Scheme 1).22 For the synthesis of
5c-e, the diol 4 was treated with triphenylphosphine (Ph3P)
and diisopropyl azodicarboxylate (DIAD) in dichloroethane
(DCE), furnishing epoxide 6 in 49% yield.22 Epoxide 6 was
then reacted with 4-bromophenol, 3,5-difluorophenol, or
phenol in N,N0-dimethylformamide (DMF) in the presence
of K2CO3, giving 5c-e in 81%, 71%, and 75% yields,
respectively.23 In all the following reaction steps the same
synthetic procedures were employed irrespective of 6-O-
substituents. Compounds 5a-e were thus converted to the
corresponding 5-azides 7a-e with inversion of configuration
using Ph3P, DIAD, and diphenylphosphoryl azide (DPPA)
in tetrahydrofuran (THF) in 68%, 88%, 90%, 58%, and
68% yields, respectively.22 Hydrolysis of the 1,2-O-isopro-
pylidene acetal in 7a-e followed by glycosylation was per-
formed using HCl in methanol (1 M), formed by in situ
Here, we describe a novel hydroxethylene (HE) motif,
readily available from chiral carbohydrate precursors and
displaying aryloxymethyl or benzyloxymethyl P1 side chain
†The PDB deposition codes for the BACE-1 complex crystal struc-
tures with inhibitor 1d is 3IXJ.
*To whom correspondence should be addressed. Phone: þ46-8-
54683104. Fax: þ46-8-54683199. E-mail: bertil.samuelsson@
medivir.com.
aAbbreviations: Aβ, amyloid-β; AD, Alzheimer’s disease; APP,
amyloid precursor protein; BACE-1, β-site amyloid precursor protein
cleaving enzyme; Cath D, cathepsin D; HE, hydroxyethylene.
r
pubs.acs.org/jmc
Published on Web 02/03/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1459
Figure 1. HEA (2) and HE (1d, 3) BACE-1 inhibitor Ki or IC50 and their cathepsin D selectivities.
Scheme 1a
a Reagents and conditions: (i) Ph3P, DIAD, DCE, reflux; (ii) Bu2SnO, toluene, reflux; (iii) tetrabutylammonium bromide, 4-bromobenzyl bromide, or
3,5-difluorobenzyl bromide, toluene, 90 ꢀC; (iv) 4-bromophenol or 3,5-difluorophenol or phenol, K2CO3, DMF, 120 ꢀC; (v) Ph3P, DIAD, DPPA, THF,
room temp; (vi) HCl, MeOH, room temp.
Scheme 2a
a Reagent and conditions: (i) MeI, Ag2O, DMF, room temp; (ii) H2SO4, 1,4-dioxane, reflux; (iii) PDC, DCM, room temp; (iv) R3NH2 (I, II, III, IV, V,
or VI), 2-hydroxypyridine, DIPEA, 70 ꢀC; (v) Ph3P, H2O, MeOH; (vi) R4COOH (A or B), Py-BOP, DIPEA, DCM, room temp.
addition of acetyl chloride to methanol at 0 ꢀC,19 delivering
the methyl glycosides 8a-e as anomeric R/β mixtures in
87%, 76%, 95%, 85%, and 98% yields, respectively.
respectively.25 Subsequent oxidation at the anomeric posi-
tion using pyridinium dichromate in dichloromethane
(DCM) gave the corresponding lactones 11a-e in 66%,
68%, 59%, 63%, and 72% yields, respectively.26 The lac-
tones 11a-e were ring-opened with (S)-2-amino-N-benzyl-3-
methylbutyramide (I) (see Figure 2) upon heating in diiso-
propylethylamine (DIPEA) using 2-hydroxypyridine as an
activator of the lactone to give the ring-opened amides 12a-e
in 85%, 82%, 96%, 97%, and 30% yields, respectively.27
The lactone 11d was also ring opened using amines II, III, IV,
V, and VI, employing the same protocol (vide supra) to
give product 12f-j in 82%, 80%, 79%, 39%, and 43%
yields, respectively. The amine I was synthesized from
The conversion of the methyl glycosides 8a-e into the
target compounds 1a-l was achieved over several steps as
shown in Scheme 2. Methylation at the 2-O-position in 8a-e,
using methyl iodide in the presence of Ag2O in DMF,
furnished the methyl glycosides 9a-e in 96%, 77%, 90%,
64%, and 90% yields, respectively.24 Hydrolysis of the
methylglycosides in refluxing 1,4-dioxane in the presence
of sulphuric acid for ∼1 h followed by quenching with
aqueous sodium carbonate afforded the desired pro-
ducts 10a-e in 69%, 63%, 81%, 97%, and 59% yields,

1460 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Bjo€rklund et al.
Figure 2. R3 carboxylic acids, R2 amine, and R1 ether library.
Table 1. Target Compounds and Inhibition Data
a ND = not determined. b Inhibition at an inhibitor concentration of 0.5 μM (5 μM). Ki values are the mean from two different experiments.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1461
Boc-Val-OH and benzylamine (see Supporting Information),
and the amines II-VI were commercially available. Reduction
of the azide group in 12a-jwas achieved using Ph3P in methanol
containing a few drops of water, providing the corresponding
amines which were coupled without prior purification with
carboxylic acids A and B16,18,28 using benzotriazole-1-yloxytris-
(pyrrolidino)phosphonium hexafluorphosphate (Py-BOP) and
DIPEA in DCM,22,29 furnishing the 12 target compounds 1a-l
in yields of 30-81%.
2.2. Structure-Activity Relationships. We have previously
disclosed that a novel HE motif incorporating a methylaryl-
oxy or methylbenzyloxy P1 substituent that coupled to the
previously reported 5-substituted isophthalamide (A)16 in
the P2-P3 position furnishes highly potent BACE-1 inhi-
bitors, e.g., inhibitor 3 with a Ki of 1 nM.30 A potential
drawback with this inhibitor series is the apparent low selec-
tivity toward cathepsin D, for which 3 displays a Ki of 59 nM,
furnishing a modest 60-fold shift in Ki over cathepsin D. To
optimize potency and cathepsin D selectivity for this inhi-
bitor series, we have examined the S10 pockets of BACE-1
and cathepsin D probing for potential discriminating
molecular features. The comparisons were made using the
1LYB31 3D structure of cathepsin D and the 3DM632 3D
structure for BACE-1. While the methyl group has fre-
quently been used as a P10 residue in HE template, the S10
pocket of BACE-1 can accommodate larger groups. More-
over, the S10 pocket of BACE-1 can also accommodate polar
interactions from the side chains of Thr72 and Asp228, in
contrast to the S10 pocket in cathepsin D which comprises
only hydrophobic residues.
On the basis of this observation, we introduced a methoxy
P10 side chain, which is still a small group but more polar
than the P10 methyl side chain (inhibitor 3). We noted that
not only was the potency enhanced, delivering a BACE-1
inhibitor with Ki ,1 nM, but also that the selectivity against
cathepsin D was dramatically improved (Ki of 1100 nM for
cathepsin D). The natural substrate, APP, of BACE-1 has a
polar Asp in the P10 position,33 and while it has previously
been reported34 that selectivity over cathepsin D can be
enhanced by increasing the polarity of the P10 side chain,
the BACE-1 inhibitors reported on displayed only weak
inhibitory activities with Ki in the low micromolar range.
Following this result a broader SAR study of this HE
scaffold with a P10 methoxy side chain was undertaken. With
the Val-benzylamide P20 group (I) reported by Ghosh et al.,35
various P1 side chains were introduced into the HE scaffold
(Table 1). It was soon apparent that with increased size of the
P1 group from that of inhibitor 1d a substantial loss in
potency could be observed, i.e., inhibitors 1a-c furnishing
Ki of 60, 80, and 75 nM, respectively. Interestingly, when the
3,5-difluorophenoxymethyl P1 side chain of 1d was replaced
with the corresponding nonfluorinated and smaller P1
group, 1e, a more than 10-fold loss in activity was observed
(Table 1). Possibly the activity of the 3,5-difluorophenoxy-
methyl P1 side chain can be attributed to advantageous
lipophilic interactions of the fluorine group with the over-
lapping S1-S3 pockets of BACE-1.
displaying Ki of >10 and 0.22 μM, respectively. It has been
reported that each oxygen of the sulfonamide of the P2-P3
group A contributes with hydrogen bonds to amino acids of
the S2 pocket36 which in part could account for the loss in
potency observed for inhibitors 1k and 1l.
To reduce molecular weight and the number of amide
bonds, we then examined the influence of replacing the
P20 Val-benzylamide group (I) with smaller substituents
(Table 1). Of the selected lipophilic substituents in inhibitors
1f-j, all were substantially less potent than lead inhibitor 1d.
The BACE-1 Ki for inhibitor 1j (100 nM) and the slightly less
active 1i (600 nM) are, however, surprisingly good. The P20
phenyl groups of 1i and 1j are based on modeling in a
coplanar arrangement with the P10-P20 amide, and while
occupying the S20-pocket more efficiently with close contact
interactions compared to the valine of 1d, they also retain the
important backbone hydrogen bonds to the carbonyl of
Gly34 and the amide of Thr72. In contrast, the benzyl
moiety of 1h may disrupt these backbone hydrogen
bonds while adjusting the conformation to fit in the S20
pocket.
2.3. X-ray Crystallography Analysis. The high potency of
this series, exemplified by the 3D crystal structure of inhi-
bitor 1d, is to a large extent due to the numerous hydrogen
bonds formed between the inhibitor and the enzyme.
The 8 times improved potency of 1d over 3 is likely due to
improved interactions in the S10 pocket where the P10 methyl
group of 3 does not have any close interactions with any of
the S10 residues. The closest residues are the polar side chains
of Thr72 and Asp228. Introduction of a methoxy group in
the P10 position thus introduces close contact interactions
with Tyr198 and Ile226 and makes it more suitable for the
polar environment in the S10 pocket of BACE-1. There are
no direct hydrogen bonds; however, the ether oxygen is
involved in an extensive hydrogen bond network, including
water and the side chains of Thr72, Thr231, Arg235, and
Thr329 (Figures 3 and 4).
3. Conclusions
In summary, we have developed a novel HE central core,
incorporating a methoxy P10 substituent and aryloxymethyl
or benzyloxymethyl P1 substituents and delivering potent
and cathepsin D selective BACE-1 inhibitors. The excellent
potencies, together with the observations that inhi-
bitors from this class can display high BACE-1 selecti-
vity over cathepsin D, make further investigations highly
warranted.
Having identified a highly potent P1 group, we also
examined the P2-P3 group B18 (Table 1), which previously
has demonstrated modest nanomolar potency toward
BACE-1 enzyme when coupled to a HE scaffold. Not
surprisingly this P2-P3 group, lacking interactions with the
S2 pocket, furnished substantially less potent inhibitors also
with the present HE isostere, with inhibitors 1k and 1l
Figure 3. Overview of the hydrogen bonding network of inhibitor
1d.

1462 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Bjo€rklund et al.
for 5 h with a Dean-Stark trap. The temperature was then
lowered to 90 ꢀC, and tetrabutylammonium bromide (1.32
equiv) and the appropriate benzyl bromide (1.27 equiv) were
added. The mixture was stirred at 90 ꢀC overnight. Then the
solvent was removed under vacuum, and the residue was
purified by column chromatography (toluene/ethyl acetate
1
10:1). 5a: H NMR (400 MHz, CDCl3) δ 7.47 (d, JHH = 8.3
Hz, 2H), 7.20 (d, JHH=8.3 Hz, 2H), 5.79 (d, JHH=3.7 Hz, 1H),
4.73 (t, JHH = 4.2 Hz, 1H), 4.55-4.45 (m, 2H), 4.24-4.18
(m, 1H), 4.01-3.95 (m, 1H), 3.60-3.55 (m, 1H), 3.49-3.44
(m, 1H), 2.09-2.03 (m, 1H), 1.88-1.79 (m, 1H), 1.49 (s, 3H),
1.31 (s, 3H); MS (ESI) m/z 395.1 ([M þ Na]þ calcd for
C16H21BrNaO5þ 395.0).
4.4.2. Synthesis of 5c-e. General Epoxide Nucleophilic Open-
ing Procedure. The epoxide 6 (1 equiv) and the appropriate
phenol (2 equiv) were dissolved in DMF. K2CO3 (0.25 equiv)
was added, and the mixture was heated to 110 ꢀC and stirred
at that temperature overnight. The solvent was removed by
coevaporation with toluene. The residue was purified by column
chromatography (toluene/ethyl acetate 20:1). 5c: 1H NMR (400
MHz, CDCl3) δ 7.36 (d, JHH=9.1 Hz, 2H), 6.79 (d, JHH=9.1
Hz, 2H), 5.82 (d, JHH=3.6 Hz, 1H), 4.77 (t, JHH=4.2 Hz, 1H),
4.34-4.27 (m, 1H), 4.16-4.11 (m, 1H), 4.07-4.02 (m, 1H),
3.96-3.91 (m, 1H), 2.50 (bs, 1H), 2.18-2.13 (m, 1H), 1.94-1.85
(m, 1H), 1.51 (s, 3H), 1.32 (s, 3H); MS (ESI) m/z 381.0 ([M þ
Na]þ calcd for C15H19BrNaO5þ 381.0).
4.4.3. Synthesis of 7a-e. General Azide Substitution Procedure.
Compounds 5a-e (1 equiv) and triphenylphosphine (1.3 equiv)
were dissolved in THF. The mixture was cooled to -15 ꢀC
(ice/acetone). Diisopropyl azodicarboxylate (2.1 equiv) was
added dropwise to the solution. The temperature was then
increased to 0 ꢀC, and diphenylphosphoryl azide (1.4 equiv)
was added. The temperature was kept at 0 ꢀC for an additional
30 min and then at room temperature overnight. The mixture
was concentrated, and the residue was purified by column
chromatography (toluene/ethyl acetate 30:1). 7a: 1H NMR
(300 MHz, CDCl3) δ 7.48 (d, JHH = 8.3 Hz, 2H), 7.22 (d,
Figure 4. X-ray crystal structure of inhibitor 1d in the active site of
BACE-1 and its interactions with BACE-1 are depicted.
4. Experimental Section
4.1. Protease Enzyme Assays. The BACE-1 and cathepsin D
assays were performed as previously described.32
4.2. Crystallography. The details of the crystallization and
structure determination procedures have been published else-
where.32 Briefly, the complex of BACE-1 and 1d was crystallized
˚
in the monoclinic space group P21 with cell dimensions (in A) of
a=81.9, b=102.7, c=100.4 and β =103.4ꢀ. The strucuture was
determined to 2.2 A resolution with an R of 0.21 (Rfree=0.24)
and deposited in RCSB PDB (3IXJ).
˚
4.3. General Methods. All glassware was dried over an open
flame before use in connection with an inert atmosphere. Con-
centrations were performed under reduced pressure at <40 ꢀC
(bath temperature). Thin layer chromatography was performed
using Merck silica gel 60 F-254 plates with detection by UV,
charring with 8% sulfuric acid or ammonium molybdate (100 g),
Ce(IV) sulfate (2 g), sulfuric acid (10%, 2 L). Column chroma-
tography was performed on silica (0.035-0.070 mm). NMR
spectra were recorded at 25 ꢀC on a Varian (300 or 400 MHz) or
a Bruker (400 or 500 MHz) instrument using the solvent residual
peak (CDCl3 1H δ 7.26 and 13C δ 77.17 or CD3OD-d4 1H δ 3.31
and 13C 49.0) as standard. Unless stated otherwise, all materials
were obtained from commercial suppliers and used without
further purification. DCM was refluxed over CaH2 and distilled
J
HH=8.3 Hz, 2H), 5.81 (d, JHH=3.6 Hz, 1H), 4.73 (t, JHH=
4.2 Hz, 1H), 4.52 (s, 2H), 4.33-4.26 (m, 1H), 3.74-3.69 (m, 2H),
3.54-3.50 (m, 1H), 2.07-2.01 (m, 1H), 1.92-1.84 (m, 1H), 1.49
(s, 3H), 1.31 (s, 3H); MS (ESI) m/z 420.1 ([M þ Na]þ calcd for
C16H20BrN3NaO4þ 420.1).
4.4.4. Synthesis of 8a-e. General Hydrolysis and Glycolsyla-
tion Procedure. Compounds 7a-e (1 equiv) were dissolved in
1 M HCl in MeOH (3 mL/mmol) and stirred at room tempera-
ture for 2 h. The mixture was then neutralized with NaHCO3
(aq). The volatile solvents were removed under vacuum, and the
residue was dissolved in DCM and washed with water (ꢀ2). The
organic layer was dried over MgSO4 and concentrated, and the
residue was purified by column chromatography (toluene/ethyl
˚
and stored over 4 A molecular sieves before use. Compounds
1a-l were further purified before biological testings using
preparative RP-HPLC, consisting of Waters 2767 autoinjector
and fraction collector, Waters 996 photodiode array detector,
and Micromass ZQ2000 mass detector (operated in þESI). The
preparative reversed phase column was an ACE C8, 21 mm ꢀ
100 mm, 5 μm, 100A from ACE (U.K.), and the mobile phases
were based on water/acetonitrile containing 0.1%TFA. The
purity of final compounds was assessed by two different ana-
lytical LC-MS methods and found to be g95% in all cases. The
LC-MS analyses were perfomed using a Waters LC-MS
instrument equipped with a UV/vis detector, Waters 996 and a
MS detector, Waters ZQ, using one of the following two
methods. (1) Method A: column, ACE C8, 50 mm ꢀ 3 mm,
3 μm particles; pump, Waters Alliance 2695; mobil phase A,
10 mM NH4OAc in water; mobil phase B, 10 mM NH4OAc in
90% acetonitrile; gradient, 20-100% B in 5 min followed by
2 min at 100% B. (2) Method B: as chromatography system A
except for the following: mobil phase A, 0.2% HCOOH in
water; mobil phase B, 0.18% HCOOH in 90% acetonitrile.
Optical rotations were measured at room temperature on a
Perkin-Elmer 341 polarimeter using a 10 cm, 1 mL cell.
acetate 15:1). 8a: 1H NMR (400 MHz, CDCl3) δ 7.48 (d, JHH
=
8.3 Hz, 2H), 7.21 (d, JHH=8.3 Hz, 2H), 4.70 (s, 1H), 4.50 (s, 2H),
4.46-4.38 (m, 1H), 4.24 (t, JHH=4.5 Hz, 1H), 3.62-3.54 (m,
2H), 3.48-3.44 (m, 1H), 3.37 (s, 3H), 2.06-1.98 (m, 1H),
1.93-1.86 (m, 1H), 1.80 (d, JHH=4.9 Hz, 1H); MS (ESI) m/z
394.0 ([M þ Na]þ calcd for C14H18BrN3NaO4þ 394.0).
4.4.5. Synthesis of 9a-e. General O-Methylation Procedure.
Compounds 8a-e (1 equiv) were dissolved in DMF. Methyl
iodide (8 equiv) and Ag2O (2 equiv) were added. The mixture
was stirred in room temperature overnight. The reaction was
quenched with CHCl3, and the solids were filtered off. The
filtrate was concentrated under vacuum, and the residue was
purified by column chromatography (toluene/ethyl acetate
1
15:1). 9a: H NMR (400 MHz, CDCl3) δ 7.50-7.45 (m, 2H),
7.23-7.18 (m, 2H), 4.99-4.96 (m, 1H), 4.50 (s, 2H), 3.79 (s, 3H),
3.60-3.56 (m, 2H), 3.50-3.44 (m, 1H), 3.39 (s, 3H), 3.33 (s, 1H),
2.20-2.12 (m, 1H), 2.09-2.03 (m, 1H); MS (ESI) m/z 452.1 ([M
- CH3 þ ACN þ H2O þ Na]þ calcd for C14H17BrN3O4þ 370.1).
4.4.6. Synthesis of Final Compounds 10a-e. General Hydro-
lysis Procedure. Compounds 9a-e (1 equiv) were dissolved in
4.4. Synthetic Procedures. 4.4.1. Synthesis of 5a,b. General
Benzylation Procedure. The diol 4 (1 equiv) and dibutyltin oxide
(1.32 equiv) were dissolved in toluene. The mixture was refluxed

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1463
1,4-dioxane/0.5 M H2SO4 1:1 and heated to reflux. After
complete reaction (∼1 h according to TLC), the mixture was
cooled to room temperature and then neutralized with Na2CO3
(aq). The volatile solvents were concentrated under vacuum.
The residue was dissolved in DCM and washed with H2O (ꢀ2).
The organic phase was dried over Na2SO4 and concentrated.
The residue was purified by column chromatography (toluene/
ethyl acetate 10:1). 10a: 1H NMR (300 MHz, CDCl3) δ 7.47 (d,
We also thank the following for excellent technical assistance:
˚
Stefan Roneus, Cathrine Ahgren, Sara Appelgren, Elisabet
Lilja, Elizabeth Hamelink, and Dr. Ian Henderson. Finally,
we acknowledge Medivir AB for performing the X-ray crystal
studies and for financial support.
Supporting Information Available: Experimental details and
structural characterization of all compounds and procedures for
the synthesis of amine I and carboxylic acid B. This material is
available free of charge via the Internet at http://pubs.acs.org.
JHH=8.2 Hz, 2H), 7.21 (d, JHH=8.2 Hz, 2H), 5.30 (d, JHH=5.6
Hz 1H), 4.51 (s, 2H), 4.38-4.28 (m, 1H), 3.78-3.75 (m, 1H),
3.72-3.68 (m, 2H), 3.61-3.54 (m, 1H), 3.35 (s, 3H), 3.14 (d,
J
HH = 5.9 Hz, 1H), 2.14-1.96 (m, 2H); MS (ESI) m/z 394.0
([M þ Na]þ calcd for C14H18BrN3NaO4þ 394.0).
References
4.4.7. Synthesis of 11a-e. General Oxidation Procedure.
Compounds 10a-e (1 equiv) was dissolved in DCM. At 0 ꢀC
pyridinium dichromate (1.5 equiv) and 4 A molecular sieves
(1) Blennow, K.; de Leon, M. J.; Zetterberg, H. Alzheimer’s disease.
Lancet 2006, 368, 387–403.
(2) Klafh, H.-W.; Staufenbiel, M.; Kornhuber, J.; Wiltfang, J. Thera-
peutic approaches to Alzheimer’s disease. Brain 2006, 2840–2855.
(3) Ghosh, A. K.; Kumaragurubaran, N.; Tang, J. Recent develop-
ments of structure based β-secretase inhibitors for Alzheimer’s
disease. Curr. Top. Med. Chem. 2005, 5, 1609–1622.
(4) Pike, C. J.; Burdick, D.; Walencewicz, A. J.; Glabe, C. G.; Cotman,
C. W. Neurodegeneration induced by β-amyloid peptides in vitro:
the role of peptide assembly state. J. Neurosci. 1993, 13, 1676–
1687.
˚
powder were added. The mixture was stirred overnight in room
temperature. The solids were filtered off. The filtrate was
concentrated under vacuum, and the residue was purified by
1
column chromatography (toluene/ethyl acetate 10:1). 11a: H
NMR (300 MHz, CDCl3) δ 7.48 (d, JHH=8.4 Hz, 2H), 7.20 (d,
J
J
HH=8.4 Hz, 2H), 4.71-4.64 (m, 1H), 4.52 (s, 2H), 4.15 (dd,
HH=6.0, 8.1 Hz, 1H), 3.76 (d, JHH=1.8 Hz, 1H), 3.74 (s, 2H),
(5) Yankner, B. A.; Dawes, L. R.; Fisher, S.; Villa-Komaroff, L.;
Oster-Granite, M. L.; Neve, R. L. Neurotoxicity of a fragment of
the amyloid precursor associated with Alzheimer’s disease. Science
1989, 245, 417–420.
(6) Hardy, J.; Selkoe, D. J. The amyloid hypothesis of Alzheimer’s
disease: progress and problems on the road to therapeutics. Science
2002, 297, 353–356.
(7) Lin, X.; Koelsch, G.; Wu, S.; Downs, D.; Dashti, A.; Tang, J.
Human aspartic protease memapsin 2 cleaves the β-secretase site of
β-amyloid precursor protein. Proc. Nat. Acad. Sci. U.S.A. 2000, 97,
1456–1460.
(8) Luo, Y.; Bolon, B.; Kahn, S.; Bennett, B. D.; Babu-Khan, S.;
Denis, P.; Fan, W.; Kha, H.; Zhang, J.; Gong, Y.; Martin, L.;
Louis, J. C.; Yan, Q.; Richards, W. G.; Citron, M.; Vassar, R. Mice
deficient in BACE-1, the Alzheimer’s β-secretase, have normal
phenotype and abolished β-amyloid generation. Nat. Neurosci.
2001, 4, 231–232.
(9) Roberds, S. L.; Andersson, J.; Basi, G.; Bienkowski, M. J.; G.,
B. D.; Chen, K. S.; Freedman, S. B.; Frigon, N. L.; Games, D.; Hu,
K.; Johnson-Wood, K.; Kappenman, K. E.; Kawabe, T. T.; Kola,
I.; Kuehn, R.; Lee, M.; Liu, W.; Motter, R.; Nichols, N. F.; Power,
M.; Robertson, D. W.; Schenk, D.; Schoor, M.; Shopp, G. M.;
Shuck, M. E.; Sinha, S.; Svensson, K. A.; Tatsuno, G.; Tintrup, H.;
Wijsman, J.; Wright, S.; McConlogue, L. BACE knockout mice are
healthy despite lacking the primary β-secretase activity in brain:
implications for Alzhemier’s disease therapeutics. Hum. Mol.
Genet. 2001, 10, 1317–1324.
(10) Gou, T.; Hobbs, D. W. Developement of BACE1 inhibitors for
Alzheimer’s disease. Curr. Med. Chem. 2006, 13, 1811–1829.
(11) Schmidt, B.; Baumann, S.; Braun, H. A.; Larbig, G. Inhibitors and
modulators of β- and γ-secretase. Curr. Top. Med. Chem. 2006, 6,
377–392.
(12) Ziora, Z.; Kimura, T.; Kiso, Y. Small-sized BACE1 inhibitors.
Drugs Future 2006, 53–63.
(13) Eder, J.; Hommel, U.; Cumin, F.; Martoglio, B.; Gerhartz, B.
Aspartic proteases in drug discovery. Curr. Pharm. Des. 2007, 13,
271–285.
(14) Beher, D.; Graham, S. L. Protease inhibitors as potential disease-
modifying therapeutics for Alzheimer’s disease. Expert Opin. In-
vest. Drugs 2005, 14, 1385–1409.
(15) Schmidt, B.; Baumann, S.; Narlawar, R.; Braun, H. A.; Larbig, G.
Modulators and inhibitors of γ- and β-secretase. Neurodegener.
Dis. 2006, 3, 290–297.
(16) Stachel, S. J.;Coburn, C. A.;Steele, T. G.; Jones, K. G.;Loutzenhiser,
E. F.; Gregro, A. R.; Rajapakse, H. A.; Lai, M.-T.; Crouthamel,
M.-C.; Xu, M.; Tugusheva, K.; Lineberger, J. E.; Pietrak, B. L.;
Espeseth, A. S.; Shi, X.-P.; Chen-Dodson, E.; Holloway, M. K.;
Munshi, S.; Simon, A. J.; Kuo, L.; Vacca, J. P. Structure-based design
of potent and selective cell-permeable inhibitors of human β-secretase
(BACE-1). J. Med. Chem. 2004, 47, 6447–6450.
3.70-3.64 (m, 1H), 3.55 (s, 3H), 2.48-2.38 (m, 1H), 2.32-2.21
(m, 1H); MS (ESI) m/z 392.0 ([M þ Na]þ calcd for
C14H16BrN3NaO4þ 392.0).
4.4.8. Synthesis of 12a-j. General Lactone Opening Procedure.
Compounds 11a-e (1 equiv) and the amine (I-IV) (2 equiv)
were dissolved in diisopropylethylamine. 2-Hydroxypyridine
(2 equiv) was added, and the mixture was heated to 70 ꢀC
overnight. If needed, a few drops of DMF was added for
solubility. The mixture was concentrated, and the residue was
purified by column chromatography (toluene/ethyl acetate 6:1).
12a: 1H NMR (400 MHz, CDCl3) δ 7.47 (d, JHH=8.3 Hz, 2H),
7.31-7.14 (m, 6H),7.19 (d, JHH=8.3 Hz, 2H), 7.00 (d, JHH
=
8.9 Hz, 1H), 4.46 (s, 2H), 4.36 (d, JHH=5.7 Hz, 2H), 4.34-4.28
(m, 1H), 3.88-3.80 (m, 2H), 3.63-3.51 (m, 2H), 3.44-3.39 (m,
1H), 3.41 (s, 3H), 2.80 (bs, 1H), 2.44-2.34 (m, 1H), 2.10-2.02
(m, 1H), 1.90-1.83 (m, 1H), 0.96 (d, JHH=6.8 Hz, 3H), 0.91
(d, JHH=6.8 Hz, 3H); MS (ESI) m/z 576.2 ([M þ H]þ calcd for
C26H35BrN5O5þ 576.2).
4.4.9. Synthesis of Final Compounds 1a-l. General Procedure.
Azides 12a-j (1.0 equiv) were dissolved in MeOH and a few
drops of water and then treated with Ph3P (1.5 equiv). The
mixture was stirred at room temperature overnight and then
concentrated under vacuum. Without further purification the
formed amine was used in the next step. 5-(Methanesulfo-
nylmethylamino)-N0-(1-phenylethyl)isophthalic acid (A) or car-
boxylic acid B (1.0 equiv), Py-BOP (1.0 equiv), and DIPEA
(1.0 equiv) were dissolved in DCM. The mixture was stirred at
room temperature for 30 min before the amine (∼1.5 equiv)
from the previous reaction dissolved in DCM and DIPEA (1.0
equiv) was added. After complete reaction the mixture was
washed with NaHCO3 (ꢀ1) and brine (ꢀ1). The water phase
was washed with DCM (ꢀ2). The organic layers were combined
and dried over Na2SO4, concentrated under vacuum, and
purified by column chromatography (toluene/ethyl acetate
1:1). 1a: [R]2D0 -15.2 (c 0.92, CHCl3); 1H NMR (400 MHz,
CDCl3) δ 8.16 (m, 1H), 7.97 (t, JHH=1.4 Hz, 1H), 7.94 (t, JHH
=
1.4 Hz, 1H), 7.48-7.06 (m, 18H), 5.36-5.24 (m, 1H), 4.44 (s,
2H), 4.37-4.31 (m, 1H), 4.28-4.17 (m, 4H), 3.83-3.79 (m, 1H),
3.66-3.59 (m, 2H), 3.43 (s, 3H), 3.32 (s, 3H), 2.83 (s, 3H),
2.36-2.25 (m, 1H), 2.15-2.05 (m, 1H), 1.98-1.84 (m, 1H), 1.58
(d, JHH=6.8 Hz, 3H), 0.94 (d, JHH=6.8 Hz, 3H), 0.90 (d, JHH
=
6.8 Hz); HRMS (ESI) m/z 908.2898 ([M þ H]þ calcd for
C44H55BrN5O9Sþ 908.2872).
(17) Liaudet-Coopman, E.; Beaujouin, M.; Derocq, D.; Garcia, M.;
ꢀ
Glondu-Lassis, M.; Laurent-Matha, V.; Prebois, D.; Rochefort,
H.; Vignon, F. Cathepsin D: newly discovered functions of a long-
standing aspartic protease in cancer and apaptosis. Cancer Lett.
2006, 237, 167–179.
Acknowledgment. We gratefully thank the Dr. Tatiana
Agback at Medivir AB for performing 2D NMR experiments.

1464 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Bjo€rklund et al.
(18) Hom, R. K.; Gailunas, A. F.; Mamo, S.; Fang, L. Y.; Tung, J. S.;
Walker, D. E.; Davis, D.; Thorsett, E. D.; Jewett, N. E.; Moon,
J. B.; John, V. Design and synthesis of hydroxyethylene-based
peptidomimetic inhibitors of human β-secretase. J. Med. Chem.
2004, 47, 158–164.
synthesis of unsaturated 1,4-dicarbonyl compounds and R,β-bute-
nolids. Tetrahedron 1995, 51, 2763–2776.
(27) Openshaw, H. T.; Whittaker, N. J. The synthesis of emetine and
related compounds. Part VII. The utility of bi-functional catalysts
in amine-ester interaction. J. Chem. Soc. 1969, 1, 89–91.
(28) Supporting information.
(29) Oscarsson, K.; Oscarson, S.; Vrang, L.; Hamelink, E.; Hallberg,
A.; Samuelsson, B. New potent C2-symmetric malaria plasmepsin I
and II inhibitors. Bioorg. Med. Chem. 2003, 11, 1235–1246.
(30) Unpublished results.
(31) Baldwin, E. T.; Naryana Bath, T.; Gulnik, S.; Hosur, M. V.;
Sowder, M. C., II; Cachau, R. E.; Collins, J.; Silva, A. M.;
Erickson, J. W. Crystal structures of native and inhibited
forms of human cathepsin D: implications for lysosomal target-
ing and drug design. Proc. Nat. Acad. Sci. U.S.A. 1993, 90, 6796–
6800.
(19) Schmidt, O. T. In Methods in Carbohydrate Chemistry; Whistler,
R. L., Woffrom, M. L., BeMiller, J. N., Eds.; Academic Press:
New York, 1963; Vol. II, p 318.
€
(20) Oscarsson, K.; Lahmann, M.; Lindberg, J.; Kangasmetsa, T. U.;
Oscarson, S.; Hallberg, A.; Samuelsson, B. Design and synthesis of
HIV-1 protease inhibitors. Novel tetrahydrofuran P2/P20-groups
interacting with Asp29/30 of the HIV-1 protease. determination of
binding from X-ray crystal structure of inhibitor protease complex.
Bioorg. Med. Chem. 2003, 11, 1107–1115.
(21) Jeon, H. R.; Yoon, S.; Shin, Y.; Shin, J. E. N. Synthesis of a
precursor of bulgecinine, (4S,5R)-1-acetyl-2-formyl-5-benzyloxy-
methyl-4-pyrrolidinol. J. Korean Chem. Soc. 1997, 41, 150–156.
(22) Johansson, P.-O.; Chen, Y.; Belfrage, A. K.; Blackman, M. J.;
€
€
€
(32) Back, M.; Nyhlen, J.; Kvarnstrom, I.; Appelgren, S.; Borkakoti, N.;
˚
Jansson, K.; Lindberg, J.; Nystrom, S.; Hallberg, A.; Rosenquist, A.;
Samuelsson, B. Design, synthesis and SAR of potent statine-based
BACE-1 inhibitors: exploration of P1 phenoxy and benzyloxy
residues. Bioorg. Med. Chem. 2008, 16, 9471–9486.
€
Kvarnstrom, I.; Jansson, K.; Vrang, L.; Hamelink, E.; Hallberg,
A.; Rosenquist, A.; Samuelsson, B. Design and synthesis of potent
˚
inhibitors of the malaria aspartyl proteases plasmepsin I and II.
Use of solid-phase synthesis to explore novel statine mofits. J. Med.
Chem. 2004, 47, 3353–3366.
(33) Sauder, M. J.; Arthur, J. W.; Dunbrack, R. L., Jr. Modeling of
substrate specificity of the Alzheimer’s disease amyloid precursor
protein β-secretase. J. Mol. Biol. 2000, 300, 241–248.
€
(23) Zuccarello,G.;Bouzide, A.;Kvarnstrom, I.; Niklasson, G.; Svensson,
ꢀ
S. C. T.; Brisander, M.; Danielsson, H.; Nillroth, U.; Karlen, A.;
(34) Brady, S. F.; Singh, S.; Crouthamel, M.-C.; Holloway, M. K.; Coburn,
C. A.; Garsky, V. M.; Bogusky, M.; Pennington, M. W.; Vacca, J. P.;
Hazuda, D.; Lai, M.-T. Rational design and synthesis of selective
BACE-1 inhibitors. Bioorg. Med. Chem. Lett. 2004, 14, 601–604.
(35) Ghosh, A. K.; Bilcer, G.; Harwood, C.; Kawahama, R.; Shin, D.;
Hussain, K. A.; Hong, L.; Loy, J. A.; Nguyen, C.; Koelsch, G.;
Ermolieff, J.; Tang, J. Structure-based design: potent inhibitors of
human brain memapsin 2 (β-secretase). J. Med. Chem. 2001, 44,
2865–2868.
Hallberg, A.; Classon, B.; Samuelsson, B. HIV-1 protease inhibitors
based on acyclic carbohydrates. J. Org. Chem. 1998, 63, 4898–4906.
(24) Sureshan, K. M.; Das, T.; Shashidhar, M. S.; Gonnade, R. G.;
Bhadbhade, M. M. Sulfonate protecting groups: synthesis of
D- and L-myo-inositol-1,3,4,5-tetrakisphosphate precursors by a
novel silver(I) oxide- mediated O-alkylation of 2,4(6)-di-O-acyl-
6(4)-O-sulfonyl-myo-inositol
through intramolecular assistance of the sulfonyl group. Eur. J.
1,3,5-orthoformate
derivatives
Org. Chem. 2003, 1035–1041.
(36) Coburn, C. A.; Stachel, S. J.; Jones, K. G.; Steele, T. G.; Rush, D.
M.; DiMuzio, J.; Pietrak, B. L.; Lai, M.-T.; Huang, Q.; Lineberger,
J.; Jin, L.; Munshi, S.; Holloway, M. K.; Espeseth, A.; Simon, A.;
Hazuda, D.; Graham, S. L.; Vacca, J. P. BACE-1 inhibition by a
series of Ψ[CH2NH] reduced amide isosteres. Bioorg. Med. Chem.
Lett. 2006, 16, 3635–3638.
(25) Kinoshita, T.; Miwa, T.; Clardy, J. Efficient synthesis of methyl
β-D-fucofuranoside and D-fucofuranose derivatives. Carbohydr.
Res. 1985, 143, 249–255.
ꢀ
(26) Bonete, P.; Najera, C. Lithiated 3-tosylpropanal and 4-tosyl-2-
butanone dimethyl acetals as β-acylvinyl anion equivalents for the