Synthesis of the Potent, Selective, and Efficacious β-Secretase (BACE1) Inhibitor NB-360

pubs.acs.org/jmc

Article

Synthesis of the Potent, Selective, and Eﬃcacious β‑Secretase

(BACE1) Inhibitor NB-360

Heinrich Rueeger, Rainer Lueoend, Rainer Machauer,* Siem J. Veenstra, Philipp Holzer,

Konstanze Hurth, Markus Voegtle, Mathias Frederiksen, Jean-Michel Rondeau,

Marina Tintelnot-Blomley, Laura H. Jacobson, Matthias Staufenbiel, Grit Laue, and Ulf Neumann

Cite This: J. Med. Chem. 2021, 64, 4677 −4696

Read Online

sı

*

Metrics & More

Article Recommendations

Supporting Information

ACCESS

ABSTRACT: Starting from lead compound 4, the 1,4-oxazine

headgroup was optimized to improve potency and brain

penetration. Focusing at the 6-position of the 5-amino-1,4-oxazine,

the insertion of a Me and a CF₃group delivered an excellent

pharmacological proﬁle with a pK_aof 7.1 and a very low P-gp eﬄux

ratio enabling high central nervous system (CNS) penetration and

exposure. Various synthetic routes to access BACE1 inhibitors

bearing a 5-amino-6-methyl-6-(triﬂuoromethyl)-1,4-oxazine head-

group were investigated. Subsequent optimization of the P3

fragment provided the highly potent N-(3-((3R,6R)-5-amino-3,6-

dimethyl-6-(triﬂuoromethyl)-3,6-dihydro-2H-1,4-oxazin-3-yl)-4-

ﬂuorophenyl)-5-cyano-3-methylpicolinamide 54 (NB-360), able to reduce signiﬁcantly Aβ levels in mice, rats, and dogs in acute and

chronic treatment regimens.

initiates the pathway by cleaving APP at position one of Aβ,

generating the secreted amino-terminal part of APP (sAPPβ)

as well as the carboxy-terminal fragment C99. This trans-

membrane fragment is further cleaved by γ-secretase leading to

Aβ. Alternatively, APP may also be processed via the non-

amyloidogenic pathway, which is initiated by the metal-

loprotease α-secretase, which cleaves in the center of Aβ

leading to sAPPα and C83.^8,9

Knock out of the BACE1 gene blocks the generation of Aβ

and C99, shifting the APP processing toward the α-secretase

pathway as indicated by an increase in sAPPα and C83.¹⁰

Interestingly, mice carrying only a single BACE1 allele showed

a 50% reduction in BACE1 enzyme activity but a much smaller

eﬀect on Aβ.¹¹Nevertheless, a moderate 12% Aβ decrease in

transgenic mice carrying a single BACE1 allele translated into a

pronounced (70−90%) long-term reduction of Aβ deposi-

tion.¹²

INTRODUCTION

■

Alzheimer’s disease (AD) is the world’s leading cause of

dementia and the most prevalent neurodegenerative disease.

AD currently constitutes a considerable unmet medical need

and is expected to become a major challenge for the healthcare

systems of both industrial and developing countries during the

coming decades. Accumulation of amyloid-β (Aβ) in

parenchymal plaques and the brain vasculature and hyper-

phosphorylated tau in neuroﬁbrillary tangles is a pathological

hallmark of the disease.¹Neuroinﬂammation, driven by plaque-

activated microglia and astrocytes, is a likely mediator between

amyloid pathology and synaptic and neuronal dysfunction and

neuron loss.²Currently available longitudinal data suggest that

abnormal deposition of Aβ starts about two decades before the

onset of clinical disease, while markers of neurodegeneration

(elevated tau protein in cerebrospinal ﬂuid (CSF)) can be

detected several years before dementia diagnosis.³⁻⁵Progress

to identify patients with elevated AD risk in the presympto-

matic stage has fueled therapeutic concepts that aim to start

antiamyloid treatments already in the “preclinical” (only Aβ

pathology present) or the “prodromal” (Aβ pathology plus

signs of early neurodegeneration) stage of the disease.⁶First

clinical data indicating improved eﬃcacy of the Aβ antibody

solenazumab in patients with early, compared to advanced, AD

further support the current ideas of preventive treatment.⁷

Aβ is generated in the β-secretase pathway from the large

transmembrane β-amyloid precursor protein (APP). The

membrane-bound aspartic protease BACE1 (EC 3.4.23.46)

BACE2 is a closely related protein, with high similarity in

sequence and three-dimensional structure, and its physiological

role is not well understood.¹³Current BACE1 inhibitors show

little or no selectivity over BACE2, and it was unknown

Received: December 11, 2020

Published: April 12, 2021

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

4677

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

e

Table 1. Examples of Early Disclosed BACE1 Inhibitors with Amidine Headgroups

c

d

e

1

2

3

a

hBACE1 IC₅₀[μM]

0.49

0.48

9.5

0.033

0.122

6.6

0.044

0.004

8.8

a

APP CHO IC₅₀[μM]

pK_a

b

ER

25

59

6.1

a

b

c

d

Values are means of at least three experiments. ER is the eﬄux ratio (P_BL‑AP/P_AP‑BL) in MDCK cells transfected with human MDR1. Ref 35. Ref

e

36. Ref 37.

whether such lack of selectivity would have consequences in

chronic pharmacological treatment. BACE1 has a number of

physiological substrates in addition to APP. Germ-line deletion

of the BACE1 gene aﬀected the processing of neuregulin1

(NRG1), resulting in hypomyelination,^14,15APP-like protein1

and 2 (APLP1/2),¹⁶the close homologue of L1 (CHL1),¹⁷

seizure-6-protein (SEZ6),¹⁸and others. Nonetheless, general

and consistent malfunction and sickness were not observed in

mice lacking BACE1 activity. Thus, it seems necessary to

elucidate in detail the relative importance of these BACE1

substrates during embryonal development and adult life.

Moreover, it was shown that some BACE1 physiological

substrates are processed in the trans Golgi network, while APP

is primarily cleaved in the more acidic endosomal compart-

ment. The intracellular inhibitor distribution may therefore

contribute to the eﬀects on diﬀerent BACE1 substrates. While

our understanding of BACE1 biology continues to evolve,

careful monitoring of the safety of BACE1 inhibitors during

clinical trials is critical. Recent clinical data show that strong

pharmacological inhibition of BACE1 is accompanied by signs

of worsening in some cognitive domains.¹⁹Therefore, safety

and eﬃcacy of doses leading to partial BACE1 inhibition need

to be investigated carefully.

BACE1 belongs to the aspartyl protease family with two

aspartic acids constituting the catalytic dyad in the center of a

large binding site extending over 6−8 substrate amino acid

residues. Standard high-throughput screening (HTS) methods

failed in many institutions to deliver suitable hits. Therefore,

the initial phase of the BACE1 inhibitor design was dominated

by peptidomimetic approaches, similar to those used for

inhibitors of renin and HIV protease. These eﬀorts produced

potent inhibitors, which lacked signiﬁcant brain penetration

and in vivo CNS Aβ lowering activity at pharmacological

doses.^20,21Consequently, the focus shifted to nonpeptidic

BACE1 inhibitors with physicochemical properties within the

CNS drug space. During lead optimization, special attention

has to be paid to the blood−brain barrier (BBB) penetration

capability, as well as cellular activity and metabolic

stability.^22,23A moderate-to-high permeability (>50−150

nm/s) and minimal P-glycoprotein (P-gp) eﬄux (eﬄux/ﬂux

ratio (ER) less than 3-fold) in the MDCK-MDR1 assay have to

be attained to achieve a good BBB penetration and is

commonly observed for CNS drugs.²⁴⁻²⁷Remaining within

the boundaries of the CNS drug property space during the

structure-based design (MW < 450 Da, PSA < 70 Å, clog P < 4,

HBD < 2, number of oxygen and nitrogen atoms ∑(N + O) <

5 and pK_a< 8 to minimize P-gp²⁸⁻³¹and hERG channel³²

interactions) should lead to drug candidates with acceptable

pharmacokinetics, brain exposure, and cardiosafety proﬁles. By

considering these constraints in the design of cyclic

hydroxyethylamine BACE1 inhibitors, we could demonstrate

brain exposure but were unable to overcome the high

metabolic clearance observed in the cyclic sulfone and cyclic

sulfoxide hydroxyethylamine scaﬀolds.^33,34

A key advancement in the BACE1 inhibitor design was the

discovery of amidine- or guanidine-containing heterocycles by

high-throughput or/and fragment-based screening.³⁵⁻³⁸An

amidine embedded in a cyclic framework, which can form an

optimal hydrogen-bonding network with both active site

aspartyl moieties, was combined with ideal substitution

vectors, derived from structural knowledge gained by X-ray

crystallography and molecular modeling.

This has led to a series of BACE1 inhibitors with high

potency and ligand eﬃciency. Selective substitution of the

amidine containing cyclic framework with moieties having a

low P-gp interaction potential and the capability to reduce the

pK_ato ∼7 has provided in vivo eﬃcacious BACE1 inhibitors.

Several compounds with slightly diﬀerent substituted cyclic

amidine headgroups with such a proﬁle have advanced into

clinical studies.³⁹⁻⁴⁶

Previously, various groups had disclosed frontrunner BACE1

inhibitors containing amidine headgroups (1−3, Table

1).³⁵⁻³⁷Compound 3, bearing an amino-thiazine headgroup,

an aryl moiety in P1, and an extension into P3 via an amide

linker, expanded the ﬁeld beyond 5- and 6-membered

acylguanidines 1−2 (nomenclature P1, P2, and P3 according

to Schechter and Berger⁴⁷) while, at the same time, showing

excellent cellular activity and a low eﬄux ratio (ER) compared

to the amino-dihydropyrimidinone (1) and amino-hydantoin

headgroup (2) containing inhibitors.

To assist in the design of novel 6-membered heterocyclic

amidine headgroups, we co-crystallized compound 3 with

BACE1. In Figure 1, the topology of the active site of BACE1

bound to 3 is highlighted. Tight hydrogen bonds are formed

between the protonated 2-amino-thiazine headgroup and the

catalytic aspartates Asp32 (2.7 and 2.9 Å) and Asp228 (2.8 and

3.0 Å). In addition, the amide N−H forms a hydrogen bond

with the backbone carbonyl of Gly230. The connecting phenyl

ring is engaged in a network of van der Waals contacts with

hydrophobic residues lining the S1 pocket (Ile118, Trp115,

Phe108, and Tyr71 of the enzyme ﬂap). The orientation of the

2-acyl-pyridine moiety projecting into the S3 pocket is

stabilized by a favorable intramolecular electrostatic interaction

between the NH of the amide bond and the pyridyl nitrogen.

An sp²nitrogen at this position is preferred over an sp²carbon

atom to minimize steric interaction with the backbone of

4678

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

To set the baseline for potency and physicochemical

properties of this new class of 5-amino-1,4-oxazine inhibitors,

the C6-unsubstituted prototype compound 4 was synthe-

sized.⁵¹At that time, the predictive power of our internal

computational tools for pK_avalues was not very strong and a

large diﬀerence between the calculated and experimentally

measured pK_awas observed (6.2 vs 9.5, Table 2). To reduce

the amidine pK_ato about 7, we favored small ﬂuoroalkyl

substitutions (CF₃, CHF₂, CH₂F) over polar substituents.⁵¹

The introduction of hydroxy, alkoxy, amino, amido, sulfone,

and sulfonamide groups was assumed to bear the risk to

increase P-gp-mediated eﬄux due to the increase in polarity

(HBD > 2, number of oxygen and nitrogen atoms ∑(N + O)

> 5). For estimation of the required ﬂuorination level of a

methyl group at C6, we tried to transfer pK_ashifts previously

described in the literature. The established pK_ashifts of

ﬂuoroalkyl groups on amines (ΔpK_aNHCH₂CH2CH₃to

NHCH₂CH₂CF₃−2.0) have been investigated,⁵²which

suggested the CF₃group to be a good option to lower the

amidine pK_ato around neutrality. Herein, we disclose our

eﬀorts to convert our initial 5-amino-1,4-oxazine lead

compound 4 into the 5-amino-6-methyl-6-triﬂuoromethyl-

1,4-oxazine containing inhibitors, a class of compounds (A,

Figure 2) with an improved physicochemical property proﬁle

(measured pK_a∼ 7 and low eﬄux ratio) resulting in a robust

oral eﬀect in CNS Aβ lowering activity. Related approaches

were also pursued by others.^39,42,53We present here, in more

detail, the synthetic access to this highly substituted 5-amino-

1,4-oxazine headgroup and structure−activity relationship

(SAR) of P3 modiﬁcations (64a−j).⁵⁴A detailed account of

the pharmacokinetic properties and the pharmacological

activity of 54 (NB-360) in preclinical studies has already

been published elsewhere.⁵⁵

Figure 1. Co-crystal structure of 2-amino-1,3-thiazine inhibitor 3

bound to human BACE1 (PDB ID: 7B1E).

Gly230. This bisarylamide has become a widely used fragment,

especially in substituted 6-membered heterocyclic amidine

BACE1 inhibitors, likely because of its optimal shape

complementary with the S1 and S3 pockets and its

discrimination against cathepsin D.^{42,43,45,48−50}

Based on the understanding of the binding mode of the 2-

amino-1,3-thiazine 3, the interactions of novel 6-membered

amidine heterocycles were assessed together with their

potential BBB penetration properties. From this evaluation,

we selected the 5-amino-1,4-oxazine headgroup as our starting

point.⁵¹The measured pK_avalue of the 6-unsubstituted 5-

amino-1,4-oxazine headgroup turned out to be too basic and

outside of the target range (Table 2). Therefore, subsequent

ﬁne-tuning with diﬀerent electron-withdrawing substituents at

the 6- and/or 3-position was envisioned. Even weak electron-

withdrawing substituents at these proximal positions would be

expected to have a considerable pK_alowering eﬀect. In

addition, steric shielding of the amidine headgroup was

postulated to have a positive impact on activity and permeation

by reduced solvation (Figure 2). Additional steric bulk at the

6-position was hypothesized to further increase the chemical

and metabolic stability of the 5-amino-1,4-oxazine headgroup.

Following this strategy, we were able to generate a series of

potent BACE1 inhibitors with druglike properties.

CHEMISTRY

■

Our retrosynthetic analysis is presented in Figure 3. An

obvious disconnection for the C6-unsubstituted 5-amino-1,4-

oxazine headgroup in 4 was to build the more accessible 1,6-

ether linkage in the 1,4-oxazine scaﬀold from the correspond-

ing N-chloroacetylated intermediate prepared from amino

alcohol B (Path A). The ﬁnal target compounds could then

arise via amidine formation from the lactam and generation of

the aniline for amide coupling with diﬀerent P3 fragments. The

same disconnection (Path A) for the 3,3,6,6-tetra-substituted

Figure 2. (A) Binding mode of 2-amino-1,3-thiazine 3 and (B) proposed binding mode of 5-amino-1,4-oxazine inhibitors 4 and compound series

A. Not shielded areas (in blue) decrease with increasing bulk at position 6.

4679

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 3. Retrosynthetic analysis of 4, rac-20, and compound series A.

a

Scheme 1. Preparation of 4 and rac-20Referring to Retrosynthetic Pathway A

a

Reagents and conditions: (a) NaBH₄, BF₃-Et₂O, tetrahydrofuran (THF), 0−25 °C, 81/84%; (b) chloroacetyl chloride, Na₂CO₃, CH₂Cl₂-H₂O, 0−

25 °C, 95/38%; (c) t-BuOK, t-BuOH, reﬂux, 32/52%; (d) Lawesson’s reagent, THF, reﬂux, 78/59%; (e) t-BuOOH, 7 N NH₃in MeOH, NH₃

conc., THF, 25 °C, 50/90%; (f) (Boc)₂O, N,N-diisopropylethylamine, CH₂Cl₂, 25 °C, 93/49%; (g) NaN₃, CuI, (1R,2R)-N1,N2-

dimethylcyclohexane-1,2-diamine, Na-ascorbate, EtOH-H₂O, 90 °C, 51/54%; (h) H₂, Lindlar catalyst, EtOAc, 25 °C, 99/94%; (i) Chiralpak

IA 250 mm × 4.6 mm, hexane/EtOH/MeOH 80:10:10, 1.5 mL/min, 39/41%; (j) 17, EDC, HOAt, N,N-diisopropylethylamine, N,N-

dimethylformamide (DMF), 25 °C, 93/79%; (k) 4 N HCl in dioxane, 25−40 °C, 99/65%.

4680

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 2. Preparation of Racemic 38a−eReferring to Retrosynthetic Pathway C

a

Reagents and conditions: (a) TMSCH₂N₂, NEt₃, THF, 0.25 °C, 73/88%; (b) 23, Rh₂(TFA)₄, CH₂Cl₂, 25 °C, 35/34%; (c) MeLi, AlMe₃, toluene,

−78 °C, 69/65%; (d) TMSN₃, BF₃-Et₂O, toluene, 35 °C, 78/63%; (e) In, 4 N HCl-THF, 25 °C, 76/71%; (f) AlMe₃, CH₂Cl₂, 40 °C, (47/50%)/

(38/44%); (g) P₂S₅, hexamethyldisiloxane, toluene, 100 °C, 95/85%; (h) t-BuOOH, NH₃, THF, 0−25 °C; (i) (Boc)₂O, NEt₃, acetonitrile, 25 °C,

two steps 56%/57%; (j) NaN₃, CuI, (1R,2R)-N1,N2-dimethylcyclohexane-1,2-diamine, Na-ascorbate, EtOH-H₂O, 70 °C, 82%/-; (k) H₂, 10% Pd-

C, EtOH, 25 °C, 61%/two steps 51%; (l) 34, 35, or 36, EDC, HOAt, N,N-diisopropylethylamine, DMF, 25 °C; (m) 4 N HCl in dioxane, 40 °C,

two steps 33−85%.

5-amino-1,4-oxazine headgroup in compound class A is not

feasible, and alternative pathways had to be identiﬁed. The

neopentylic 1,2-ether formation (Path B) was judged to be

very diﬃcult, and only an intramolecular bond forming process

was given some consideration initially. We expected an

intermolecular ether formation to C starting with the

commercially available tertiary alcohol D and a sterically less

demanding reactive electrophile (Path C) to be more

promising. Subsequent asymmetric addition of a methyl

group to a chiral sulfoximine intermediate derived from ketone

C would install the chiral amine for lactam formation.

Immediate addition of the C3 methyl group to the ketone

and transformation of the tertiary alcohol into an amine with

subsequent lactam formation would instead lead to the achiral

3,3,6,6-tetra-substituted 5-amino-1,4-oxazine headgroup, the

common central intermediate of the diﬀerent pathways.

Alternatively, the attractive but somewhat more speculative

5-amino-1,4-oxazine formation by an intramolecular iodo-

amidination reaction was considered as well (Path D), utilizing

the amidine fragment (E), generated from D via allylic ether

formation and subsequent transformation of the ester moiety

into the corresponding amidine fragment.

After identiﬁcation of the required absolute conﬁguration at

C6, the development of a more versatile synthetic route to the

chiral key intermediate G was required to facilitate the P3

optimization process. The formation of the neopentylic ether

linkage with an advanced chiral P1 fragment together with the

chiral tert-alcohol D was initially considered attractive, but no

precedence for such transformation was found in the literature

at the time this work was started. Determined to ﬁnd a more

eﬃcient synthesis, we hypothesized that an activated aziridine

might be suﬃciently reactive for a nucleophilic ring-opening

with the tert-alkoxyde D (Path E). By this unprecedented key

transformation, the common key amidine intermediate G

4681

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 3. Preparation of 54Referring to Retrosynthetic Pathway D

a

Reagents and conditions: (a) NaH, 2,3-dibromoprop-1-ene, DMF, 25 °C, 81%; (b) (2-ﬂuorophenyl)boronic acid, Pd(PPh₃)₄, CsF, DMF-H₂O

10:1, 65 °C, 64%; (c) 7 N NH₃in MeOH, 55 °C, 98%; (d) triﬂuoroacetic anhydride, NEt₃, CH₂Cl₂, 5−25 °C, 93%; (e) 7 N NH₃in MeOH, N-

acetyl-^L-cysteine, 80 °C, 96%; (f) I₂, Na₂CO₃, CHCl₃, 100 °C, 44%; (g) H₂, 10 Pd-C, NaOAc, MeOH, 25 °C, 97%; (h) KNO₃, H₂SO₄, 0 °C, 39/

46%; (i) CHIRALPAK AD-H 5, CO₂-i-PrOH 90:10, 48/48%; (j) (Boc)₂O, NEt₃, acetonitrile, 25 °C, 90%; (k) H₂, 5% Pd-C, i-PrOH-THF 2:1, 50

°C, 92%; (l) EDC, HOAt, N,N-diisopropylethylamine, DMF, 25 °C, 81%; (m) CF₃COOH, CH₂Cl₂, 25 °C, 83%.

would become readily available via the corresponding amino-

nitrile intermediate.

dimethylaminopropyl)carbodiimide (EDC)-mediated coupling

with the P3 fragment 17 followed by N-Boc deprotection.

The synthesis of the highly substituted 5-amino-3,6-

dimethyl-6-triﬂuoromethyl-1,4-oxazine containing inhibitors

rac-(3R*,6R*)-38a,e (R¹= H) and rac-(3R*,6R*)-38b−d

(R¹= F) following path C in Figure 3 is depicted in Scheme 2.

Starting from the acid chlorides 21a or 21b, the diazoketones

22a and 22b were prepared with trimethylsilyldiazomethane.

Subsequent Rh-catalyzed O−H insertion on the tert-alcohol

rac-23 provided the tert-ethers rac-24a and rac-24b. Chemo-

selective addition of the methyl group was achieved with

trimethylaluminium at low temperature. The resulting racemic

diastereomeric tertiary benzylic alcohol 25a or 25b was

converted into azide 26a or 26b with trimethylsilyl-azide and

boron triﬂuoride diethyl etherate in toluene at a slightly

elevated temperature. Indium-mediated reduction of the azides

under acidic conditions at ambient temperature provided the

tert-amino-ester 27a or 27b, which upon treatment with

trimethylaluminium delivered the lactams rac-(2S*,5R*)-28a

and rac-(2R*,5R*)-29a or rac-(2S*,5R*)-28b and rac-

(2R*,5R*)-29b, respectively. The diastereoisomeric pairs

could be easily separated by ﬂash chromatography, and the

The synthesis of the C6-unsubstituted 5-amino-1,4-oxazine

lead compounds 4 (R¹= H) and rac-20 (R¹= F) is depicted in

Scheme 1. The chloroacylated amino alcohols rac-7a and rac-

7b were easily obtained from the corresponding commercially

available amino acids rac-5a and rac-5b by borane reduction

and acylation with chloroacetyl chloride. Cyclization to the

3,3-disubstituted morpholin-5-one rac-8a and rac-8b with t-

BuOK followed by thioamide formation with Lawesson’s

reagent and oxidative-mediated transformation into the

amidine with tert-butyl hydroperoxide/ammonia and subse-

quent N-Boc protection with Boc-anhydride provided the

masked headgroup intermediates rac-11a and rac-11b. Trans-

formation of the corresponding aryl-bromide into the anilide

13, 14, and rac-16 was carried out via Cu-catalyzed azide

substitution and subsequent catalytic reduction with the

Lindlar catalyst followed by a separation of the enantiomers

13 and 14 on the chiral stationary phase (Chiralpak IA).

Finally, the target compound 4 and the C4-ﬂuorinated

analogue rac-20 were obtained via 1-ethyl-3-(3-

4682

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

a

Scheme 4. Preparation of 54 and 64a−jReferring to Retrosynthetic Pathway E

a

Reagents and conditions: (a) BH₃Me₂S, THF, 65 °C, 99%; (b) 4-nitrobenzene-1-sulfonyl chloride, KHCO₃, and Cs₂CO₃, acetonitrile, 80 °C,

50%; (c) 4-nitrobenzene-1-sulfonyl chloride, NMM, acetonitrile, 0−25 °C, 88%; (d) (i) mesyl chloride, 4-methylmorpholine, acetonitrile, 0−5 °C,

(ii) NMM, acetonitrile, 30 °C, 84%; (e) mesyl chloride, NMM, acetonitrile, 0−5 °C, 95%; (f) DBU, CH₂Cl₂, 5−30 °C, 73%; (g) (R)-ethyl 3,3,3-

triﬂuoro-2-hydroxy-2-methylpropanoate, t-BuOK, DMF/THF, 35 °C, 87%; (h) (R)-ethyl 3,3,3-triﬂuoro-2-hydroxy-2-methylpropanoate, 1 M t-

BuOK in THF, DMF, 20−35 °C, 72%; (i) 7 N NH₃in MeOH, 50 °C, 99%; (j) triﬂuoroacetic anhydride, NEt₃, CH₂Cl₂, 0−5 °C, 95%; (k) K₂CO₃,

N-acetyl-^L-cysteine, EtOH, 80 °C, 92%; (l) KNO₃, H₂SO₄, 0−5 °C, 81%; (m) (Boc)₂O, NEt₃, CH₂Cl₂, 0−25 °C, 90%; (n) H₂, 5% Pd-C, dioxane,

45 °C, 92%; (o) 63a−j, EDC, HOAt, N,N-diisopropylethylamine (DIPEA), DMF, 5−25 °C, 16−92%; (p) CF₃COOH, CH₂Cl₂, 25 °C, 62−95%.

relative conﬁguration of the individual diastereoisomers was

assigned by NMR. In analogy to the sequence for C6-

unsubstituted amino-oxazines (Scheme 1), the N-Boc

protected amidines rac-(2R*,5R*)-31a and rac-(2R*,5R*)-

31b were obtained via their thiolactams rac-(2R*,5R*)-30a

and rac-(2R*,5R*)-30b, respectively, prepared with phospho-

rus pentasulﬁde and hexamethyldisiloxane in toluene.⁵⁶

Subsequent oxidative-mediated transformation into the

amidine with tert-butyl hydroperoxide/ammonia followed by

N-Boc protection completed the sequence. Cu-catalyzed

substitution of the aryl-bromides by sodium azide followed

by catalytic hydrogenation of rac-(2R*,5R*)-32a and rac-

(2R*,5R*)-32b with palladium on carbon provided the anilines

rac-(2R*,5R*)-33a and rac-(2R*,5R*)-33b, respectively.

Amide formation with the P3 fragments 34, 35, and 36

followed by acid-catalyzed N-Boc deprotection provided rac-

(3R*,6R*)-38a,e and the ﬂuorinated analogues rac-(3R*,6R*)-

38b−d. Having determined the required relative conﬁguration

at C6 with the individual synthesis of rac-(3R*,6R*)-38a and

rac-(3R*,6S*)-38e, a more viable synthesis of the 5-amino-6-

methyl-6-triﬂuoromethyl-1,4-oxazine headgroup possessing the

desired pK_ahad to be identiﬁed.

Starting from an allylic ether bearing already the chiral C5−

C6 amidine fragment, an intramolecular iodo-amidination

reaction was explored to assemble the 3,3,6,6-tetra-substituted

5-amino-1,4-oxazine scaﬀold G (Path D, Figure 3). As shown

in Scheme 3, the allylic ether rac-41 was prepared initially from

racemic ethyl 3,3,3-triﬂuoro-2-hydroxy-2-methylpropanoate

(rac-39) and 2,3-dibromoprop-1-ene followed by a Suzuki

coupling of rac-40 with 2-ﬂuorophenylboronic acid. Con-

version of the ethyl ester rac-41 to the amidine bearing allylic

ether rac-44 was achieved by aminolysis with NH₃in

methanol, conversion of the amide rac-42 to the nitrile rac-

43 with TFAA-NEt₃, and ﬁnal amidine formation with 7 N

NH₃and N-acetyl-L-cysteine in methanol at 80 °C.⁵⁷All

attempts to achieve an asymmetric intramolecular iodo-

amidination with chiral oxazolidine and amino acid auxiliaries

attached to the amidine moiety, as reported for the related

iodo-lactamizations, failed.⁵⁸

On the other hand, an achiral iodo-amidination with the

unsubstituted amidine rac-44 using iodine and Na₂CO₃in

chloroform at 100 °C led to the formation of a 1:1 mixture of

racemic diastereoisomeric iodo-amidines rac-45 in moderate

yield.⁵⁹Hydrogenolytic removal of iodine with 5% Pd-C in

methanol containing NaOAc and subsequent nitration of rac-

46 with KNO₃in H₂SO₄followed by chromatographic

separation of the rac-(2S*,5R*)-47 and rac-(2R*,5R*)-48

diastereoisomers and separation of the enantiomers of rac-

(2R*,5R*)-48 by SFC provided (2S,5S)-49 and (2R,5R)-50.

The remaining transformation to the aniline (2R,5R)-33b, the

key intermediate for the further evaluation of P3 SAR,

consisted of N-Boc-protection of the nitro-amidine derivative

(2R,5R)-50 and catalytic reduction of the aryl-nitro group of

(2R,5R)-51. The ﬁnal conversion of aniline (2R,5R)-33b into

inhibitor (3R,6R)-54 was initiated by EDC-mediated coupling

with the P3 fragment 52 followed by acid-catalyzed N-Boc

deprotection of (2R,5R)-53 with TFA in dichloromethane. To

enable the quantiﬁcation of a potential metabolite, some 33b

was converted into aniline 33c.

The most practical route to the key intermediate 33b (Path

E, Figure 3), exploiting the nucleophilic opening of a chiral

aziridine with a chiral alkoxy C5−C6 fragment, is depicted in

Scheme 4. The chiral amino alcohol 56, prepared via borane

reduction of 55,⁶⁰was initially transformed by a one-step

4683

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Figure 4. Superimposition of cocrystal structures of compounds bound to BACE1. (A) 5-Amino-1,4-oxazine inhibitor 4 (blue) with 2-amino-1,3-

thiazine inhibitor 3 (white); PDB ID: 6FGY and 7B1E. (B) 4 (blue) with 38a (white); PDB ID: 6FGY and 7B1P.

process with excess Ns-Cl and KHCO₃in reﬂuxing acetonitrile

into the nosylated aziridine 59. However, this procedure was

not scalable and was replaced by a reproducible two- or three-

step process requiring no chromatographic separation. Initial

nosylation with Ns-Cl/NMM at 0−5 °C in acetonitrile

provided selectively the N-nosylated amino alcohol 57, which

upon O-mesylation with Ms-Cl/NMM in acetonitrile at 0−5

°C delivered crystalline 58 in excellent yield. Addition of excess

base and increasing the temperature to 30 °C led directly to

the formation of nosylated aziridine 59. Alternatively, 59 can

be generated from 58 with 1,8-diazabicycloundec-7-ene

(DBU) in dichloromethane at ambient temperature. The

opening of aziridine 59 with the chiral tert-alcohol 39 worked

best with t-BuOK as the base in anhydrous N,N-dimethylfor-

mamide at 35 °C. By a controlled acidic workup, the highly

functionalized ether intermediate 60 can be obtained in the

crystalline form. Alternatively, the ether intermediate 60 can be

obtained directly from the crystalline mesylate 58 and chiral

tert-alcohol 39 with excess t-BuOK in N,N-dimethylforma-

mide. Conversion of the ethyl ester 60 to the nitrile bearing

ether 62 was achieved by aminolysis with NH₃in methanol

and conversion of amide 61 to nitrile 62 with TFAA-NEt₃.

Removal of the nosyl protecting group and subsequent

formation of the 3,3,6,6-tetra-substituted 3-amino-1,4-oxazine

46 were achieved in one step with K₂CO₃and N-acetyl-L-

cysteine in ethanol at 80 °C. The transformation of 46 into the

key intermediate 33b followed the same 3-step reaction

sequence as previously described for the preparation of anilide

33b in Scheme 3. The ﬁnal conversion of 33b into 64a−j,

bearing the diﬀerent P3 fragments 63a−j, followed the same 2-

step reaction sequence previously described for (3R,6R)-54 in

Scheme 3.

withdrawing substituent like a CF₃group at C6, to reduce the

pK_afrom 9.5 to ∼7,⁵²could be tolerated. Since disubstitution

at C6 was considered to be essential to ascertain suﬃcient

conﬁgurational, chemical, and metabolic stability, the absolute

stereochemistry at C6 could not entirely be predicted by

modeling. Therefore, the diastereoisomeric pair rac-

(3R*,6R*)-38a and rac-(3R*,6S*)-38e was synthesized. The

inhibitor rac-(3R*,6R*)-38a was found to be 50-fold more

potent over rac-(3R*,6S*)-38e. Subsequently, the optimal

binding of the (6R)-conﬁguration was conﬁrmed in the co-

crystal structure with BACE1 (Figure 4B). In addition, the

insertion of ﬂuorine at C4 of the P1 phenyl ring was

considered to be beneﬁcial to reduce the formation of a

possibly toxic aniline metabolite (33c) by hydrolysis of the

P3−amide bond in vivo. As expected, this remote modiﬁcation

had only a minor eﬀect on the pK_aof the amidine (pK_a9.5 for

4 vs 9.3 for rac-20, respectively, and 7.3 for rac-(3R*,6R*)-38a

vs 7.0 for rac-(3R*,6R*)-38b) but led to a ∼2-fold increased in

vitro potency (IC₅₀20 nM for rac-(3R*,6R*)-38a vs 11 nM for

rac-(3R*,6R*)-38b, Table 2). The improvement in inhibitory

potency may be driven by electrostatic interactions of the

ﬂuorine with Tyr71 and Phe108, which are in close proximity

(3.0 Å), as observed in the co-crystal structure of 38a in Figure

4. Similar potentially beneﬁcial electrostatic interactions of the

ﬂuorine atoms of the CF₃group with the carbonyl of Asp228

(3.0 Å) and the surrounding water network interacting with

Thr72 of the ﬂap can be envisaged, besides shielding of the H-

bonding network between the polar amidine headgroup and

the two catalytic aspartates Asp228 and Asp32, providing a 5-

fold increase in potency (IC₅₀58 nM for rac-20 vs 11 nM for

rac-(3R*,6R*)-38b). The identiﬁcation of this highly potent 6-

methyl-6-triﬂuoromethyl-substituted 5-amino-1,4-oxazine

headgroup with an optimal pK_aof 7.0, still allowing

protonation in the acidic environment in which BACE1 is

active, however, was set oﬀ by its rather demanding synthetic

accessibility. Excellent BACE1 inhibition in its native cellular

environment could be demonstrated with inhibitor rac-

(3R*,6R*)-38b in wtAPP CHO cells (IC₅₀22 nM), as well

RESULTS AND DISCUSSION

■

The X-ray co-crystal structure of the initial 6-unsubstituted 5-

amino-1,4-oxazine derivative 4 with BACE1 (Figure 4A)

conﬁrmed an optimal binding arrangement of the amidine with

the catalytic aspartates Asp32 and Asp228 (2.7−2.9 Å).⁵¹In

addition, we learnt that the NH of the amide of 4 formed an

about equally strong hydrogen bond with the backbone

carbonyl of Gly230 as the 2-amino-thiazine inhibitor 3 (3.0 Å

for 3 vs 3.1 Å for 4), in agreement with the nearly equal

potency when the same P1−P3 fragment is used (IC₅₀44 nM

for 3 vs 70 nM for 4). Closer examination of the binding mode

of the prototype compound 4 conﬁrmed that a small electron-

as a very high selectivity against human cathepsin D (IC₅₀

>

250 μM), for this BACE inhibitor with an amide linker

between P1 and P3 substituents, compared to shorter biaryl

P1−P3 extensions.⁶¹Equally important, a good permeation

(P_app= 67 nm/s) and a very low P-gp-mediated eﬄux (ER =

1.4) in MDCK cells transfected with human MDR1 could be

observed, which could be pointing to good brain penetration.

4684

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

f

Table 2. In Vitro Proﬁle of the Initial 5-Amino-1,4-oxazine Lead Compounds⁵¹

a

4

rac-20

rac-(3R*,6R*)-38a

rac-(3R*,6S*)-38e

rac-(3R*,6R*)-38b

b

hBACE1 IC₅₀[μM]

0.070

0.017

>10

6.2

9.5

190

30

0.058

0.003

>10

6.0

9.3

124

8.2

0.020

0.032

73

4.8

7.3

1.08

1.16

139

4.8

7.3

49

0.011

0.022

>250

4.5

7.0

67

b

wtAPP-CHO IC₅₀[μM]

b

hCathD IC₅₀[μM]

c

calculated pK_a

measured pK_a

d

e

MDCK P_app[nm/s]

103

1.9

f

ER

1.6

1.4

a

b

c

Racemic compounds. Values are means of at least three experiments. Calculated using Advanced Chemistry Development (ACD/Labs)

Software V9.06 (1994-2005 ACD/Labs). Measured pK_adetermined by potentiometric titration.⁶²P_appis the permeability through an MDR1−

MDCK cell monolayer. ER is the eﬄux ratio (P_BL‑AP/ P_AP‑BL) in MDCK cells transfected with human MDR1.

d

e

f

Table 3. P3 Structure−Activity Relation

BACE1

BACE2

CathD

d

IC

wtAPP CHO

b

CYP450 3A4/2D6/ brain Aβ40 metabolite 33c

d

⁵⁰b

c

cmpd

R²

R³

X

Y

[μM]

IC₅₀[μM]

ER

2C9 IC₅₀[μM]

red. [%]

[% of parent]

a

38b

64a

H

Br

CN

C

N

C

N

C

0.011

0.016

0.034

0.006

0.005

0.062

0.006

0.005

0.033

0.014

0.016

0.008

0.003

0.004

0.007

0.005

0.130

0.005

0.004

0.012

0.017

0.002

>250

180

0.022

0.005

0.024

0.009

0.003

0.105

0.004

0.017

0.006

0.009

0.022

1.4

3.4

17

17/0.9/3

8/>20/>20

>20/>20/>20

17/>20/>20

7/12/>20

>20/9/>20

20/4/>20

4/7/17

a

64b

>250

30

>250

64c

54

H

2

75

79

14.3

0.16

Me

Cl

NH₂

Me

Cl

1.9

3.5

3.7

2.0

nd

1.2

1.9

1.6

64d

64e

64f

64g

64h

64i

64j

e

71

<0.2

OMe

OCHF₂

e

56

66

<0.2

0.3

Cl

>20/11/>20

a

b

c

Racemic compounds. Values are means of at least three experiments. ER is the eﬄux ratio (P_BL‑AP/P_AP‑BL) in MDCK cells transfected with

d

e

human MDR1. Percent brain Aβ reduction in rats (n = 5) 4 h post dosing of 10 μmol/kg p.o. relative to vehicle-treated animals. Below LLOQ of

LC-MS quantiﬁcation.

During the course of our work, we as well as other

groups^39,49explored a wide variety of oxazine-derived scaﬀolds

to identify BACE1 inhibitor headgroups with a promising in

vitro activity and physicochemical proﬁle and better synthetic

accessibility. However, in view of the diﬃculties in achieving a

superior physicochemical proﬁle, the 5-amino-6-methyl-6-

triﬂuoromethyl-1,4-oxazine headgroup merited extra eﬀort to

identify a scalable synthesis. In parallel, the identiﬁcation of the

most eﬃcacious and druglike P3 substituent was performed.

Early SAR exploration and the co-crystal structures of 4 and

rac-38a (Figure 4) suggested that the important amide NH

hydrogen bond to the backbone carbonyl of Gly230 is

impeded by an sp²C−H group ortho to the carbonyl of the

amide linking the P3 to the P1 fragment. We therefore focused

our P3 exploration on various 2-substituted pyridine, pyrazine,

and pyrimidine fragments (Table 3). At the beginning of our

SAR exploration, the 5-bromopicoline fragment was used to

compare the physicochemical and activity proﬁles of the

diﬀerent headgroups. However, this P3 fragment was not

suitable for in vivo proﬁling due to the high in vitro clearance in

rat liver microsomes (CL_int154 μL/min·mg for rac-38a and 62

μL/min mg for rac-38b) and inhibition of CYP450 isoforms

2D6 (IC₅₀0.9 μM for rac-38b) and 2C9 (IC₅₀3 μM for rac-

38b). Only a partial improvement of the metabolic proﬁle was

observed by substituting the pyridine with the less lipophilic

pyrimidine as shown in 64a, preserving the potency but

increasing the ER to 3.4. Substitution of the bromine with a

cyano group in rac-64b produced a clean CYP450 enzyme

inhibition proﬁle, however at the expense of a large increase in

the ER (3.4 to 17). The combination of the cyano group in

rac-64b with the less eﬄux-prone pyridine fragment decreased

the ER from 17 to 2. In addition, the high selectivity against

human Cathepsin D (>40 000-fold) was retained and the

cellular potency of 64c was comparable (IC₅₀9 nM for 64c vs

24 nM for rac-64b). Furthermore, a clean CYP450 proﬁle was

obtained (all IC₅₀> 20 μM), making 64c the ﬁrst valid

candidate for a PK/PD experiment in rats. The introduction of

a methyl group at C3 adjacent to the carbonyl moiety in the

pyridine ring, to impede amide hydrolysis in vivo, was well-

tolerated and gratifyingly led to a further improvement of the

cellular potency (IC₅₀3 nM for 54 vs 9 nM for 64c),

preserving the clean CYP450 proﬁle. In an attempt to further

4685

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

lower the lipophilicity, the pyrazine compound 64d was

prepared. Only a moderate increase in the ER was observed

compared to the pyrimidine (3.5 for 64d vs 17 for rac-64b),

however, at the expense of a >5-fold lower cellular potency

(IC₅₀24 nM for rac-64b vs 105 nM for 64d). Interestingly,

64d showed as an only example a slight preference (2-fold) for

inhibition of BACE1 over BACE2, while for all other examples,

no or even 8-fold preference for inhibition of BACE2 (64b,g)

was observed. Substitution of the methyl group at C3 of the

pyridine by Cl (64e) and NH₂(64f) produced the equipotent

analogues 64e and 64f, however, with a less clean metabolic

CYP450 proﬁle. Replacement of the cyano group on the

pyridine fragment with OMe (64g) and OCHF₂(64h,i)

delivered analogues of slightly lower potency and inferior

metabolic proﬁles, especially for 64i. Substitution of the cyano

and the Me group with Cl produced a nearly equipotent

analogue (IC₅₀8 nM for 64j vs 6 nM for 54), however, with

lower cellular potency (IC₅₀3 nM for 54 vs 22 nM for 64j)

and only moderate metabolic stability for the rather lipophilic

derivative (ΔC log P of 1) compared to the best inhibitor 54 in

Table 3.

Table 4. Extended Proﬁle of 54 (NB-360)

hBACE1 IC₅₀μM

hBACE2 IC₅₀μM

0.005

>250

0.003

0.033

0.003

0.004

3.7

7.1

-3.6

141

263

pepsin, hCathD, hCathE IC₅₀μM

hAβ₄₀wtAPP CHO cell IC₅₀μM

hAβ₄₀SweAPP CHO cell IC₅₀μM

hAβ₄₂wtAPP CHO cell IC₅₀μM

sAPPβ wtAPP CHO cell IC₅₀μM

log P (octanol/water)

pK_a

log PAMPA (log Pe pH 6.8) [cm/s]

MDR1−MDCK apical-basolateral (A−B) [nm/s]

MDR1−MDCK basolateral-apical (B−A) [nm/s]

MDR1−MDCK eﬄux ratio (B−A/A−B)

CYP P450 (3A4, 2D6, 2D9) IC₅₀μM

hERG manual patch clamp IC₅₀μM

1.9

>20

>30

Only compounds with a promising cellular potency (IC₅₀

<

10 nM), minimal in vitro P-gp eﬄux (ER < 5), and no

metabolic liabilities were selected for in vivo PK/PD

evaluation. Brain Aβ40-lowering was determined after oral

administration in rats (Sprague-Dawley, n = 5) 4 h post dosing.

Blood and brain samples were taken to assess compound levels

as well as the percentage of Aβ reduction in the forebrain and

optionally also in CSF. Concerned about the potential

formation of the undesired aniline metabolite 33c by metabolic

cleavage of the amide moiety, the amounts of 33c in blood

were routinely coquantiﬁed with the parent compound.

Indeed, considerable amounts of this metabolite were observed

with our ﬁrst PK/PD candidate 64c (14.3%, Table 3). With

the insertion of a small substituent at C3 adjacent to the P3-P1

linking carboxamide of 64c, for example, a methyl, chlorine, or

NH₂group (R²in Table 3), formation of the aniline

metabolite was almost eliminated, as shown in Table 3 by

comparison of compound 64c with 54, 64e, 64i, and 64j. The

most robust brain Aβ-lowering eﬀect after 4 h was observed

with 54 (−79%), 64e (−71%), and 64j (−66%) after

application of an oral dose of 10 μmol/kg in rats (Table 3).

Compound 54 possessed the most favorable in vitro properties

(cellular IC₅₀3 nM, >50 000-fold selectivity over human

Cathepsin D, inhibition constants for CYPs 3A4, 2D6, and

2C9 above 20 μM, and a low P-gp eﬄux ratio of 1.9). An

extended overall proﬁle of 54 is shown in Table 4, and the

binding to BACE1 is shown in Figure 5.

Figure 5. Co-crystal structure of 54 bound to human BACE1 (PDB

ID: 7B1Q).

blood and 0.54 μM·h in rat brain, showing an almost equal

distribution of the compound in both tissues and the absence

of signiﬁcant eﬄux at the blood−brain barrier. A single 0.5

mg/kg oral dose of NB-360 in dogs carrying a ventricular port

for repeated CSF collection reduced Aβ peptide 1−40 by more

than 80% and remained signiﬁcantly inhibited for at least 72

h.⁶³

From the candidates selected for an in vivo eﬃcacy

experiment, compound 54 revealed also the best overall PK

parameters (highest area under the curve (AUC) of 973, C_max

of 128, medium CL, and low V_ss), as shown in Table 5. PK

studies of 54 in mice and dogs showed a lower clearance and

volume of distribution in mice than in rats (mice vs rats: CL 16

vs 26, V_ss4.6 vs 9.0) and even much lower values in dogs (dogs

vs rats: CL 0.4 vs 26, V_ss0.9 vs 9.0).

Chronic treatment of APP-transgenic mice with 100 μmol/

kg NB-360 q.d. for 6 weeks led to a signiﬁcant reduction of

APP fragments C99, sAPPβ, Aβ 1−38, 1−40, and 1−42.

Amyloid plaques were profoundly reduced compared to

vehicle-treated animals, staying at baseline levels. Furthermore,

treatment with NB-360 strongly reduced the number of

activated microglia cells and astrocytes, markers of neuro-

inﬂammation.⁶³

Having achieved moderate-to-high oral bioavailability in

mice, rats, and dogs (37, 77, and 62%, respectively),

compound 54 (NB-360) was further evaluated in animal

models including those displaying characteristics of amyloid

pathology.⁶³In summary, NB-360 reduced Aβ40 by >80% in

the rat brain and CSF 8 h after oral doses (0.3−30 μmol/kg).

The NB-360 unbound fraction (AUC) was 0.63 μM·h in rat

In mice chronically treated with NB-360, we observed the

appearance of gray patches in the otherwise black fur. Others

and we investigated this further, with data showing that

inhibition of BACE2 is causally involved in hair depigmenta-

tion, probably via inhibition of the correct processing of the

melanosome protein PMEL-17 and improper melanin

distribution in the hair follicle.^64,65

4686

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Table 5. Pharmacokinetic Parameters in Rats, Mice, and Dogs

dose

[μmol/kg]

AUC_(0‑∞)

C

Hep. CL_int

in vivo clearance

V_ss

[L/kg]

t_1/2

c

a

ab

,

^maa^xb

c

,

cmpd species

[nM·h]

[nM]

%F

[μL/min/10⁶cells]

[mL/min/kg]

[h]

d

64e

64i

64j

54

rat

mouse

dog

6

511

873

564

973

882

74

80

83

128

152

1880

61

70

77

37

62

98

97

74

17

n.d.

43

24

47

26

16

0.4

8.6

7.6

7.1

9.0

4.6

0.9

2.8

5.6

2.3

4.9

3.6

36

d

6

d

6

d

6

0.67

67 370

n.d.

a

b

c

Oral (6 μmol/kg for rats and mice, 0.3 mg/kg for dogs). Dose normalized. Intravenous (2 μmol/kg for rats and mice, 0.1 mg/kg for dogs).

Cassette of ﬁve compounds with 6 μmol/kg.

d

The risk of unknown physiological consequences due to

occurring at therapeutic doses. Involvement of BACE2 in

pancreatic β-cell function and melanoma cell proliferation is

documented, and BACE2 may have other, yet undetected

physiological functions.^70,71This was considered a safety risk,

in particular for long-term prevention trials. Another oxazine

derivative, CNP520 (umibecestat), with a superior safety

proﬁle was identiﬁed and entered long-term trials in AD.⁷²

However, we decided to make NB-360 available to a large

number of interested academic institutions for preclinical

studies. Using a potent tool compound in a standardized way,

in terms of source, doses, and formulations, contributed

signiﬁcantly to deepening our understanding of the physio-

logical function of BACE1. Furthermore, NB-360 treatment of

APP-transgenic mice led to the discovery of various aspects of

BACE1 inhibition on Alzheimer’s disease biomarkers, neuronal

activity and networks, and to the kinetics of amyloid plaque

growth.^73,74

inhibition of BACE2 led to a discontinuation of development

eﬀorts for NB-360, but the excellent PK/PD properties of this

compound have facilitated research eﬀorts in mechanistic

animal models related to BACE1 inhibition in the context of

AD. For example, a combined histology/magnetic resonance

imaging study in old APP-transgenic mice with vascular

amyloid pathology addressed the question of the possible

occurrence of microhemorrhages as a potential consequence of

treatment with BACE1 inhibitors.⁶⁶Antiamyloid-β antibody

treatment in AD patients frequently resulted in cerebral

microbleeds. We did not observe such bleedings in APP23

mice treated with NB-360, indicating that microhemorrhages

are a consequence of the speciﬁc mechanism of action of

antibodies, but not a direct consequence of blocking the

generation of amyloid-β peptides via direct inhibition of

BACE1. Furthermore, chronic studies performed with NB-360

in APP-transgenic mice showed that BACE1 inhibition

markedly reduces the increase of downstream neurodegenera-

tion markers Tau protein and the neuroﬁlament light chain, as

well as deposition of vascular amyloid-β.⁶⁷⁻⁶⁹

EXPERIMENTAL SECTION

■

In Vitro Potency and Proﬁling Assays. Assays for human and

mouse BACE1, human BACE2, porcine pepsin, and human cathepsin

D and E were performed as described.⁶³Cellular potency was

determined in Chinese hamster ovary cells stably transfected with

human APP751 as described.⁶³

CONCLUSIONS

■

The octanol/water distribution coeﬃcient was derived from the

apparent permeability across an artiﬁcial liquid membrane.⁷⁵The

PAMPA assay was performed as described.²⁴MDCK cells over-

expressing human p-glycoprotein (MDR1−MDCK cells, obtained

from Prof. A. Berns, Netherlands Cancer Institute) were grown on

multiwell plates, 1 μm pore size (Becton Dickinson). After formation

of a tight monolayer, test compounds were applied to either the upper

or the lower chamber. After 1 h, liquid was removed from the upper

and lower chambers and analyzed by high-performance liquid

chromatography-mass spectrometry (HPLC-MS). Apical-to-baso-

lateral and basolateral-to-apical transport rates were calculated.

Human CYP450 Enzymes Inhibition Assays. The following

substrates were used to assess the reversible inhibition of the CYP450

enzymes: CYP3A4, midazolam; CYP2D6, Bufuralol; and CYP2C9,

Diclofenac. Assays were performed in human liver microsomes

supplemented with NADPH. Test compounds were diluted to 0.5−20

μM ﬁnal concentrations. The incubation time was 10 min in 50 mM

sodium phosphate buﬀer (pH 7.5). After stopping and centrifugation,

analysis was performed using a solid-phase-extraction (SPE) mass

spectrometry system consisting of a TSQ Quantum Discovery MAX

mass spectrometer controlled by Xcalibur 3.0 and equipped with an

electrospray ion source (Ion Max electrospray interface) from

Thermo Fisher Scientiﬁc (Reinach, Switzerland) and a RapidFire

high-throughput SPE system from Agilent Technologies (Waldbronn,

Germany).

In summary, we have reported the discovery of a novel 3-aryl

substituted (3R,6R)-5-amino-3,6-dimethyl-6-triﬂuoromethyl-

3,6-dihydro-2H-1,4-oxazine headgroup. Selective structural

modiﬁcations at C6 of the 1,4-oxazine-based prototype

compound 4 aimed at ﬁne-tuning the pK_aand at shielding

the amidine and ether oxygen as much as possible from

undesired chemical and metabolic transformations. This

resulted in the identiﬁcation of a 3-aryl substituted (3R,6R)-

5-amino-3,6-dimethyl-6-triﬂuoromethyl-3,6-dihydro-2H-1,4-

oxazine headgroup bearing an optimal pK_aof ∼7.0. While

retaining the good brain penetration properties of the initial

lead compound 38a, the potency, pharmacokinetics, and

metabolic proﬁle were further optimized by structure-based

modiﬁcations of the P3 fragment. From the candidates with

the best in vitro proﬁle (cellular potency, minimal CYP and

hERG liabilities, and low P-gp eﬄux ratio) and the most robust

brain Aβ peptide reduction in rats (Table 3), compound 54

(NB-360) demonstrated an excellent overall proﬁle (Table 4)

and was selected for a more comprehensive PK/PD assessment

in rats, dogs, and APP-transgenic mice. The lowering of pK_aof

1,4-oxazine with CF₃substitution had a profound eﬀect on

activity and P-gp eﬄux. Further optimization of P3 improved

metabolic stability, delivering with NB-360 an orally active

BACE1 inhibitor with an excellent PD proﬁle in mice, rats, and

dogs. Compound NB-360 did not enter clinical trials in

Alzheimer’s disease since hair discoloration observed in mice

was considered a sign for oﬀ-target inhibition of BACE2

Animal Experiments. All animals were maintained under

standard housing conditions with access to standard pelleted food

and water ad libitum, throughout the experiments. Animal experi-

ments were conducted in accordance with Swiss national animal

welfare regulations, under the ethically approved animal experimenta-

4687

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

tion licenses authorized by the Cantonal Veterinary Authority of Basel

City and the Federal Veterinary Oﬃce of Switzerland.

C18 column, 50 mm × 2.1 mm, 1.8 μm, 50 °C; mobile phase, A =

0.05% formic acid + 3.75 mM ammonium acetate in H₂O, B = 0.04%

formic acid in AcCN; gradient, 0.0−1.4 min, 2−100% B; ﬂow rate 1.2

mL/min; 218 and 254 nM; 1 min post time; 1−5 μL injection. All

assay compounds had a measured purity of ≥95% (by thin-layer

chromatography (TLC) and UV) as determined using this analytical

UPLC or UPLC-MS system. Low-resolution mass spectral (MS) data

were determined on an Agilent 1100 mass spectrometer using ES

ionization modes (positive and/or negative) and water-MeOH 3:7 +

2 of 25% ammonium hydroxide solution. High-resolution mass

spectral (HRMS) analyses were performed using electrospray

ionization in the positive ion mode after separation by liquid

chromatography (Nexera from Shimadzu). The elemental composi-

tion was derived from the mass spectra acquired at the high resolution

of about 30 000 on an LTQ Orbitrap XL mass spectrometer (Thermo

Scientiﬁc). The high mass accuracy below 1 ppm was obtained using a

lock mass.

The procedures for the syntheses of compounds 4 and rac-20

(Scheme 1, retrosynthetic pathway A), compounds rac-(3R*,6R*)-

38a−d and rac-(3R*,6S*)-38e (Scheme 2, retrosynthetic pathway C),

and compound 54 (Scheme 3, retrosynthetic pathway D) are available

in the supplementary part including the separation of (rac-49) and 50.

The ﬁnal steps to 51, 33b, 53, and 54 are identical, as described in

procedure E.

Pharmacokinetic studies used male Sprague-Dawley rats, C57BL6

mice, and Beagle dogs, obtained from Charles River (France, rodents)

and Marshall (Italy, dogs). Compounds were dosed intravenously as a

solution (NMP/PEG200-based mixtures), whereas for oral dosing,

compounds were suspended in methylcellulose 0.5% (w/v)/99.5%

water/0.1% Tween 80 and applied via oral gavage. Mouse blood was

collected by sublingual bleeding or at sacriﬁce, whereas a serial blood

sampling allowed obtaining individual PK proﬁles in rats and dogs.

Blood samples were taken between 5 min (intravenous) and 15 min

(oral) intervals and then longer, subsequently up to 24 h (mice), 48 h

(rat), and 408 h (dog). Brain exposures were determined in the

pharmacological studies by collecting the brain after decapitation at

selected time points after dosing.

For bioanalytics, brain samples were homogenized with 2 mM

KH₂PO₄buﬀer (approx. 1/2, w/v). A structurally similar internal

standard was added to the blood or brain homogenate samples, and

compounds were extracted by quenching with a fourfold volume of

acetonitrile and an aliquot of the supernatant was directly injected

into the LC/MS/MS system. Separation from endogenous

components was achieved using a reverse-phase column and a

gradient elution with water/1% formic acid and acetonitrile/1%

formic acid. BACE1 inhibitors were ionized with the positive

electrospray mode and quantiﬁed in the MRM mode.

Procedure for the Synthesis of Compounds 54 and 64a−64j

For rat pharmacokinetic/pharmacodynamics studies, male Charles

River Sprague-Dawley rats of approx. 300 g body weight were used,

and experiments were done as described before.⁶³

(Scheme 4, Retrosynthetic Pathway E).

X-ray Analysis. The catalytic domain of human BACE1 was

expressed in Escherichia coli and refolded, as already described in

Hanessian et al. (2015), Supporting Information.⁷⁶The crystal

structures of the complexes with 3, 38a, and 54 were obtained by

soaking orthorhombic crystals of unliganded BACE1, grown at 19 °C

from 15% PEG 1500 in water by the method of vapor diﬀusion in

hanging drop using 15.0−16.4 mg/mL BACE1 (residues 46−447 of

Uniprot entry P56817) in 10 mM Tris-HCl pH 7.4, 25 mM NaCl.

The soaking buﬀer was 28.5−30.0% PEG 1500, 50 mM sodium

citrate pH 5.2−5.5, with 1.0 mM compound and 0.9−1.8% DMSO.

Crystals were directly ﬂash-frozen, and diﬀraction data were collected

either at the Swiss Light Source (SLS, in Villigen, Switzerland)

beamline X10SA, with a MAR CCD 225 (compound 3) or a Pilatus

(compound 54) detector, or in the laboratory (compound 38a), using

an FR-E superbright rotating anode X-ray generator and a

SATURN92 CCD detector. All diﬀraction data were processed with

XDS⁷⁷as implemented in APRV.⁷⁸The diﬀraction data extended to

between 1.62 and 1.94 Å. All structures were determined by diﬀerence

Fourier and reﬁned using CNX⁷⁹and Coot,⁸⁰to crystallographic R-

factors ranging from 18.6 to 18.9% (free R-factors ranging between

21.1 and 21.8%).

(R)-2-Amino-2-(2-ﬂ uorophenyl)propan-1-ol Hydrochloride (56 ).

To a suspension of (R )-2-amino-2-(2-ﬂ uorophenyl)-propanoic acid

hydrochloride (55 ) (34.0 g, 150 mmol) in THF (350 mL) was added

under argon the BH₃dimethylsulﬁ de complex (42.7 mL, 450 mmol)

at 50 °C over a period of 1 h. The reaction mixture was heated at

re ﬂ ux for 16 h. The excess borane was carefully destroyed by slow

addition of MeOH at 25 −35 °C. After multiple evaporations with

MeOH, the residue was dissolved in 2 N aq. HCl (90 mL) and stirred

at 80 −90 °C for 0.5 h. The cold reaction mixture was extracted with

tert-butylmethyl ether, and the aqueous phase was evaporated to

dryness. Residual water was removed azeotropically with AcCN/

toluene. The residual hydrochloride salt was dried under reduced

pressure at 50 °C for 24 h to provide 56 (30.8 g, 99%) as a white

1

solid, which was used without further puri ﬁ cation. H NMR (360

MHz, DMSO-d₆) δ 8.68 (s, 3H), 7.51 −7.39 (m, 2H), 7.34 −7.22 (m,

2H), 3.87 −3.71 (m, 2H), 1.62 (s, 3H). MS m /z = 170 [M + H]⁺.

Synthesis. Unless otherwise mentioned, all reagents were

obtained from commercial suppliers and used without further

puriﬁcation unless noted otherwise. Anhydrous solvents were

obtained from Aldrich and used directly. All reactions involving air-

or moisture-sensitive reagents were performed under a nitrogen or

argon atmosphere. All microwave-assisted reactions were conducted

with a Smith synthesizer from Personal Chemistry, Uppsala, Sweden.

Silica gel chromatography was performed using either glass columns

packed with silica gel (230−400 mesh) or prepacked silica gel

cartridges from Isco. All NMR spectra were collected on a Bruker 360,

400, or 600 MHz or on a Varian 300 or 400 MHz spectrometer. The

chemical shifts were expressed as ppm (δ units) with tetramethylsi-

lane or residual protonated solvent used as the reference. All tested

compounds were puriﬁed to ≥95% purity as determined by reverse-

phase ultrahigh-pressure liquid chromatography (UPLC). UPLC

analysis was obtained on a Waters ACQUITY UPLC and UPLC-MS.

UPLC method A (2.0 min UPLC run): Acquity UPLC HSS T3 C18

column, 50 mm × 2.1 mm, 1.7 μm, 35 °C; mobile phase, A = 0.1%

TFA in H₂O, B = 0.1% TFA in AcCN; gradient, 0.0−1.5 min, 5−

100% B; ﬂow rate 1 mL/min; 218 and 254 nM; 1 min post time; 1 μL

injection. UPLC-MS (2.0 min UPLC run): Acquity UPLC HSS T3

(R)-N-(2-(2-Fluorophenyl)-1-hydroxypropan-2-yl)-4-nitrobenze-

nesulfonamide (57). To a suspension of (R)-2-amino-2-(2-

ﬂuorophenyl)propan-1-ol hydrochloride (56) (4.05 g,19.6 mmol) in

AcCN (25 mL) were added 4-methylmorpholine (5.4 mL, 49.1

mmol) and 4-nitrobenzene-1-sulfonyl chloride (4.87 g, 21.56 mmol)

at 0−5 °C. The reaction mixture was stirred for 0.5 h at 0−5 °C and 2

h at 25 °C. The reaction was diluted with water (100 mL), and the

AcCN was slowly removed under reduced pressure. The remaining

water mixture was stirred for 1 h in an ice bath before the precipitated

product was ﬁltered oﬀ and washed with cold water and a small

amount of Et₂O. The title compound 56 was obtained as a white solid

(6.15 g, 88%) after drying under reduced pressure at 60 °C for 24 h

and was used without further puriﬁcation. ¹H NMR (400 MHz,

CDCl₃-d) δ 8.12 (d, J = 8.6 Hz, 2H), 7.77 (d, J = 8.6 Hz, 2H), 7.34

(dd, J = 8.1, 1.5 Hz, 1H), 7.22−7.15 (m, 1H), 7.10 (td, J = 7.6, 1.4

Hz, 1H), 6.65 (ddd, J = 12.8, 8.1, 1.4 Hz, 1H), 5.57 (s, 1H), 4.26 (dd,

4688

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

J = 11.3, 5.4 Hz, 1H), 3.78 (dd, J = 11.3, 7.8 Hz, 1H), 2.17 (dd, J =

7.9, 5.6 Hz, 1H), 1.60 (s, 3 H). MS m/z = 372 [M + NH₄]⁺.

was extracted with EtOAc. Combined extracts were washed with

water and brine, dried over MgSO₄, ﬁltered, and concentrated. The

crude product was crystallized from EtOAc/hexane to provide 59

(0.74 g, 73%) as a yellow solid. TLC (hexane/EtOAc 1:1) R_f= 0.66.

¹H NMR (360 MHz, CDCl₃) δ 8.33−8.28 (m, 2H), 8.14−8.09 (m,

2H), 7.38 (td, J = 7.6, 1.8 Hz, 1H), 7.26−7.19 (m, 1H), 7.07 (td, J =

7.5, 1.2 Hz, 1H), 6.96 (ddd, J = 10.5, 8.2, 1.2 Hz, 1H), 3.00 (s, 1H),

2.57 (s, 1H), 1.98 (s, 3H). MS m/z = 337 [M + H]⁺.

(R)-2-(2-Fluorophenyl)-2-(4-nitrophenylsulfonamido)propyl

Methanesulfonate (58). To a suspension of (R)-N-(2-(2-ﬂuoro-

phenyl)-1-hydroxypropan-2-yl)-4-nitrobenzenesulfonamide (57) (5.7

g, 16 mmol) in AcCN (60 mL) were added 4-methylmorpholine (2.5

mL, 22.4 mmol) and a solution of methanesulfonyl chloride (1.62 mL,

20.8 mmol) in AcCN (5 mL) at 0−5 °C over a period of 20 min.

After stirring for 0.5 h at 20 °C, the reaction mixture was poured onto

ice-water and the AcCN was removed under reduced pressure. The

aqueous phase was stirred for 1 h at 0−5 °C, and the crystallized

product was ﬁltered oﬀ. The residue was washed with Et₂O and dried

under reduced pressure at 50 °C for 24 h to provide 58 (6.62 g, 95%)

(R)-Ethyl 3,3,3-Triﬂuoro-2-hydroxy-2-methylpropanoate (39). To

a solution of (R)-3,3,3-triﬂuoro-2-hydroxy-2-methylpropanoic acid

(130.0 g, 822 mmol) in EtOH (420 mL) were added trimethyl-

orthoformate (244.0 g, 1645 mmol) and H₂SO₄(9.2 mL, 173 mmol),

and the reaction mixture was heated at reﬂux for 50 h. The reaction

mixture was reduced to half of its volume, and the remaining solution

was added to sat. aq. NaHCO₃solution (300 mL), and the product

was extracted with tert-butylmethyl ether. Combined extracts were

washed with brine, ﬁltered, and concentrated under normal pressure.

The residual oil was puriﬁed by vacuum distillation at 200 mbar (b.p.

1

as a light-yellow solid. H NMR (400 MHz, DMSO-d₆) δ 8.99 (s,

1H), 8.23−8.14 (m, 2H), 7.66−7.58 (m, 2H), 7.37 (td, J = 8.1, 1.8

Hz, 1H), 7.26−7.18 (m, 1H), 7.13 (td, J = 7.6, 1.3 Hz, 1H), 6.69

(ddd, J = 12.9, 8.1, 1.3 Hz, 1H), 4.67 (d, J = 9.7 Hz, 1H), 4.61 (d, J =

9.7 Hz, 1H), 3.24 (s, 3H), 1.67 (s, 3H). MS m/z = 337 [M −

MsOH]⁺.

1

88−90 °C) to provide 39 (131.4 g, 86%) as a colorless oil. H NMR

(400 MHz, CDCl₃) δ 4.44−4.26 (m, 2H), 3.82 (s, 1H), 1.58 (s, 3H),

1.34 (td, J = 7.1, 1.3 Hz, 3H).

(R)-2-(2-Fluorophenyl)-2-methyl-1-((4-nitrophenyl)sulfonyl)-

aziridine (59). Method A. To a solution of (R)-2-amino-2-(2-

ﬂuorophenyl)propan-1-ol hydrochloride (56) (200 mg, 0.875 mmol)

in AcCN (5 mL) were added under argon 2-nitrobenzene-sulfonyl

chloride (480 mg, 2.1 mmol) and KHCO₃(442 mg, 4.38 mmol), and

the reaction mixture was heated at reﬂux for 16 h. To accelerate the

aziridine formation, Cs₂CO₃(285 mg, 0.875 mmol) was added and

the reaction mixture was heated for another 6 h at reﬂux. The reaction

mixture was added to cold sat. aq. NaHCO₃solution, and the product

was extracted with EtOAc. Combined extracts were washed with

water and brine, dried over MgSO₄, ﬁltered, and concentrated. The

residual oil was puriﬁed by ﬂash chromatography on silica gel

(hexane/EtOAc 10) to provide 59 (152 mg, 50%) as a yellow oil. ¹H

NMR (360 MHz, CDCl₃) δ 8.33−8.28 (m, 2H), 8.14−8.09 (m, 2H),

7.38 (td, J = 7.6, 1.8 Hz, 1H), 7.26−7.19 (m, 1H), 7.07 (td, J = 7.5,

1.2 Hz, 1H), 6.96 (ddd, J = 10.5, 8.2, 1.2 Hz, 1H), 3.00 (s, 1H), 2.57

(s, 1H), 1.98 (s, 3H). MS m/z = 337 [M + H]⁺.

Method B. To a suspension of (R)-N-(2-(2-ﬂuorophenyl)-1-

hydroxypropan-2-yl)-4-nitrobenzenesulfonamide (57) (5.0 g, 13.4

mmol) in CH₂Cl₂(150 mL) were added at 0−5 °C NEt₃(7.5 mL,

53.6 mmol) and at 0−5 °C a solution of methanesulfonyl chloride

(2.1 mL, 26.8 mmol) in CH₂Cl₂(10 mL). The reaction mixture was

stirred for 0.5 h at 10−15 °C and for 2 h at 25−30 °C to complete

aziridine formation and then poured onto cold 10% Na₂H₂PO₄

solution, and the product was extracted with EtOAc. Combined

extracts were washed with water and brine, dried over MgSO₄,

ﬁltered, and concentrated. The crude product was crystallized from

EtOAc/hexane to provide 58 (3.8 g, 84%) as a yellow solid. TLC

(hexane/EtOAc 1:1) R_f= 0.66. ¹H NMR (360 MHz, CDCl₃) δ 8.33−

8.28 (m, 2H), 8.14−8.09 (m, 2H), 7.38 (td, J = 7.6, 1.8 Hz, 1H),

7.26−7.19 (m, 1H), 7.07 (td, J = 7.5, 1.2 Hz, 1H), 6.96 (ddd, J = 10.5,

8.2, 1.2 Hz, 1H), 3.00 (s, 1H), 2.57 (s, 1H), 1.98 (s, 3H). MS m/z =

337 [M + H]⁺.

(R)-Ethyl 3,3,3-Triﬂuoro-2-((R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propoxy)-2-methylpropanoate (60).

Method A. To a solution of (R)-ethyl 3,3,3-triﬂuoro-2-hydroxy-2-

methylpropanoate (39) (2.89 g, 15.5 mmol) in anhydrous DMF (20

mL) was added a 1 M solution of tert-butoxide in THF (15 mL, 15

mmol), and after 10 min of stirring, (R)-2-(2-ﬂuorophenyl)-2-methyl-

1-((4-nitrophenyl)sulfonyl)aziridine (59) (3.37 g, 10 mmol) was

added at 35 °C and the reaction mixture was stirred for 6−10 h at 35

°C (monitored by HPLC). The reaction mixture was neutralized with

1 N aq. HCl at 0−5 °C to pH 7 and diluted with water (30 mL), and

after stirring at 0−5 °C for 1 h, the precipitated product was collected

and dried. Recrystallization from EtOAc-hexane gave 60 (4.54 g,

1

87%) as yellow crystals. H NMR (400 MHz, CDCl₃) δ 8.24−8.19

(m, 2H), 7.92−7.83 (m, 2H), 7.66 (td, J = 8.1, 1.8 Hz, 1H), 7.29−

7.23 (m, 1H), 7.19 (td, J = 7.6, 1.4 Hz, 1H), 6.86 (s, 1H), 6.76 (ddd, J

= 12.7, 8.1, 1.4 Hz, 1H), 4.42 (qq, J = 10.7, 7.1 Hz, 2H), 3.82 (d, J =

8.9 Hz, 1H), 3.67 (d, J = 8.9 Hz, 1H), 1.62 (s, 3H), 1.60 (s, 3H), 1.40

(t, J = 7.1 Hz, 3H). MS m/z = 540 [M + NH₄]⁺.

Method B. To a solution of (R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propyl methanesulfonate (58) (4.0 g, 9.1

mmol) in anhydrous DMF (60 mL) were added under argon (R)-

ethyl 3,3,3-triﬂuoro-2-hydroxy-2-methylpropanoate (39) (3.05 g,

16.38 mmol) and at 20−25 °C a 1 M solution of potassium tert-

butoxide in THF (27.3 mL, 15 mmol). The reaction mixture was

stirred for 30 min at 35 °C and then quenched by slow addition to

cold 1 N aq. 1 N HCl, and the product was extracted with tert-

butylmethyl ether. Combined extracts were washed with sat. aq.

NaHCO₃solution and brine, dried over MgSO₄, ﬁltered, and

concentrated. The crude product was recrystallized from EtOAc/

diisopropylether/hexane to provide 69 (4.73 g, 72%) as a crystalline

yellow solid. TLC (hexane/EtOAc 1:1) R_f=0.59. ¹H NMR (400

MHz, CDCl₃) δ 8.24−8.19 (m, 2H), 7.92−7.83 (m, 2H), 7.66 (td, J =

8.1, 1.8 Hz, 1H), 7.29−7.23 (m, 1H), 7.19 (td, J = 7.6, 1.4 Hz, 1H),

6.86 (s, 1H), 6.76 (ddd, J = 12.7, 8.1, 1.4 Hz, 1H), 4.42 (qq, J = 10.7,

7.1 Hz, 2H), 3.82 (d, J = 8.9 Hz, 1H), 3.67 (d, J = 8.9 Hz, 1H), 1.62

Method C. To a solution of (R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propyl methanesulfonate (58) (1.3 g, 3

mmol) in CH₂Cl₂(30 mL) was added DBU (0.7 mL, 4.5 mmol) at

0−5 °C, and the reaction mixture was stirred for 2 h at 25 °C. The

mixture was added to cold 10% Na₂H₂PO₄solution, and the product

4689

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(s, 3H), 1.60 (s, 3H), 1.40 (t, J = 7.1 Hz, 3H). MS m/z = 540 [M +

title compound 46 (6.05 g, 92%) as a yellow amorphous solid, which

was used without further puriﬁcation. ¹H NMR (400 MHz, CDCl₃) δ

7.46 (td, J = 8.1, 1.9 Hz, 1H), 7.23 (ddt, J = 9.7, 4.9, 2.4 Hz, 1H), 7.10

(td, J = 7.6, 1.3 Hz, 1H), 7.02 (ddd, J = 12.4, 8.1, 1.3 Hz, 1H), 4.36

(br s, 2H), 4.07 (d, J = 11.7 Hz, 1H), 3.98−3.90 (m, 1H), 1.54 (d, J =

1.3 Hz, 3H), 1.46 (s, 3H). MS m/z = 291 [M + H]⁺.

NH₄]⁺.

(R)-3,3,3-Triﬂuoro-2-((R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propoxy)-2-methylpropanamide (61). A

solution of (R)-ethyl 3,3,3-triﬂuoro-2-((R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propoxy)-2-methylpropanoate (60) (4.5 g,

8.6 mmol) in 7 N NH₃in MeOH was stirred in a sealed glass vessel at

50 °C for 36 h. The reaction mixture was concentrated, and the

residual foam was dried under reduced pressure to aﬀord the title

compound 61 (4.4 g, 99%) as a yellow amorphous solid, which was

(2R,5R)-5-(2-Fluoro-5-nitrophenyl)-2,5-dimethyl-2-(triﬂuoro-

methyl)-5,6-dihydro-2H-1,4-oxazin-3-amine (50). To a solution of

(2R,5R)-5-(2-ﬂuorophenyl)-2,5-dimethyl-2-(triﬂuoromethyl)-5,6-di-

hydro-2H-1,4-oxazin-3-amine (46) (6.0 g, 20.6 mmol) in 98% sulfuric

acid (40 mL) was added in portions KNO₃(2.19 g, 21.65 mmol) at

0−5 °C over a period of 0.5 h, and the reaction mixture was stirred for

3 h at 22 °C. The reaction mixture was added to ice-water and was

basiﬁed to pH 8−9 at 15−25 °C with 30% aq. NaOH. The product

was extracted with EtOAc. Combined extracts were washed with

brine, dried over MgSO₄, ﬁltered, and concentrated. The crude

product was puriﬁed by ﬂash chromatography on silica gel (hexane/

EtOAc 10:1 to 1:1) to provide the title compound 50 (5.62 g, 81%)

as a light-yellow solid. TLC (toluene/EtOAc 3:1) R_f= 0.34. ¹H NMR

(400 MHz, CDCl₃) δ 8.52 (dd, J = 6.7, 3.0 Hz, 1H), 8.15 (dt, J = 8.9,

3.6 Hz, 1H), 7.17 (dd, J = 10.6, 8.9 Hz, 1H), 4.76 (s, 2H), 4.06 (d, J =

11.6 Hz, 1H), 3.92 (d, J = 11.7 Hz, 1H), 1.56 (s, 3H), 1.49 (s, 3H).

MS m/z = 336 [M + H]⁺.

1

used without further puriﬁcation. H NMR (400 MHz, DMSO-d₆) δ

8.85(s, 1H), 8.20 (d, J = 8.8 Hz, 2H), 7.91 (s, 1H), 7.73 (d, J = 8.8

Hz, 2H), 7.63 (s, 1H), 7.44 (td, J = 8.0, 1.8 Hz, 1H), 7.23 (m, 1H),

7.14 (td, J = 7.6, 1.4 Hz, 1H), 6.78 (m, 1H), 3.98 (d, J = 8.9 Hz, 1H),

3.72 (d, J = 8.9 Hz, 1H), 1.63 (s, 3H), 1.48 (s, 3H). MS m/z = 511

[M + NH₄]⁺.

N-((R)-1-(((R)-2-Cyano-1,1,1-triﬂuoropropan-2-yl)oxy)-2-(2-

ﬂuorophenyl)propan-2-yl)-4-nitrobenzenesulfonamide (62). To a

solution of (R)-3,3,3-triﬂuoro-2-((R)-2-(2-ﬂuorophenyl)-2-(4-

nitrophenylsulfonamido)propoxy)-2-methylpropanamide (61) (4.28

g, 8.5 mmol) and NEt₃(2.96 mL, 21.25 mmol) in CH₂Cl₂(50 mL)

was added dropwise triﬂuoroacetic anhydride (1.45 mL, 10.2 mmol)

over a period of 30 min. After stirring for 2 h at 0−5 °C, the reaction

mixture was added to cold 5% aq. NaHCO₃solution and the product

was extracted with tert-butylmethyl ether. Combined extracts were

washed with water, 0.1N aq. HCl, sat. aq. NaHCO₃solution, and

brine, dried over MgSO₄, ﬁltered, and concentrated. The crude

product was crystallized from EtOAc/heptane to provide the title

tert-Butyl ((2R,5R)-5-(2-Fluoro-5-nitrophenyl)-2,5-dimethyl-2-

(triﬂuoromethyl)-5,6-dihydro-2H-1,4-oxazin-3-yl)carbamate (51).

To a solution of (2R,5R)-5-(2-ﬂuoro-5-nitrophenyl)-2,5-dimethyl-2-

(triﬂuoromethyl)-5,6-dihydro-2H-1,4-oxazin-3-amine (50) (5.4 g,

15.9 mmol) in CH₂Cl₂(100 mL) were added Boc₂O (4.5 g, 20.63

mmol) and NEt₃(4.42 mL, 31.7 mmol), and the reaction mixture was

stirred overnight at 22 °C. The reaction mixture was added to 10% aq.

NaH₂PO₄solution, and the product was extracted with CH₂Cl₂.

Combined extracts were washed with water, dried over MgSO₄,

ﬁltered, and concentrated. The residue was crystallized from EtOAc/

diisopropylether to provide the title compound 51 (6.38 g, 90%) as a

light-yellow solid. TLC (hexane/EtOAc 3:1) R_f= 0.37. ¹H NMR (400

MHz, CDCl₃) δ 11.11 (s, 1H), 8.33−8.25 (m, 2H), 7.32 (t, J = 9.9

Hz, 1H), 4.43 (d, J = 12.4 Hz, 1H), 4.14 (d, J = 12.4 Hz, 1H), 1.73 (s,

3H), 1.57 (d, J = 1.9 Hz, 9H), 1.55 (s, 3H). MS m/z = 436 [M + H]⁺.

1

compound 62 (3.85 g, 95%) as a white solid. H NMR (400 MHz,

DMSO-d₆) δ 8.87(s, 1H), 8.18 (d, J = 8.9 Hz, 2H), 7.65 (d, J = 8.8

Hz, 2H), 7.36 (td, J = 8.0, 1.8 Hz, 1H), 7.20 (m, 1H), 7.12 (td, J =

7.6, 1.4 Hz, 1H), 6.69 (m, 1H), 4.29 (d, J = 8.9 Hz, 1H), 4.09 (d, J =

8.9 Hz, 1H), 1.78 (s, 3H), 1.62 (s, 3H). MS m/z = 493 [M + NH₄]⁺.

tert-Butyl ((2R,5R)-5-(5-Amino-2-ﬂuorophenyl)-2,5-dimethyl-2-

(triﬂuoromethyl)-5,6-dihydro-2H-1,4-oxazin-3-yl)carbamate (33b).

A solution of tert-butyl ((2R,5R)-5-(2-ﬂuoro-5-nitrophenyl)-2,5-

dimethyl-2-(triﬂuoromethyl)-5,6-dihydro-2H-1,4-oxazin-3-yl)-

carbamate (51) (6.2 g, 14.18 mmol) in dioxane (100 mL) was

hydrogenated at atmospheric pressure with 5% Pd-C (1.5 g) for 3 h at

45 °C. The reaction mixture was ﬁltered through Celite, and the

ﬁltrate was lyophilized. The white residue was recrystallized from tert-

butylmethylether/hexane to provide the title compound 33b (5.28 g,

92%) as a white solid. ¹H NMR (400 MHz, CDCl₃) δ 10.85 (s, 1H),

6.84 (dd, J = 11.8, 8.5 Hz, 1H), 6.60−6.48 (m, 2H), 4.32 (d, J = 12.1

Hz, 1H), 4.00 (d, J = 12.1 Hz, 1H), 3.62 (s, 2H), 1.61 (s, 3H), 1.51

(s, 3H), 1.49 (s, 9H). MS m/z = 406 [M + H]⁺.

(2R,5R)-5-(2-Fluorophenyl)-2,5-dimethyl-2-(triﬂuoromethyl)-5,6-

dihydro-2H-1,4-oxazin-3-amine (46). To a solution of N-((R)-1-

(((R)-2-cyano-1,1,1-triﬂuoropropan-2-yl)oxy)-2-(2-ﬂuorophenyl)-

propan-2-yl)-4-nitrobenzenesulfonamide (62) (10.7 g, 22.5 mmol) in

EtOH (70 mL) were added 2-acetamido-3-mercaptopropanoic acid

(7.34 g, 45 mmol) and K₂CO₃(6.84 g, 49.5 mmol), and the reaction

mixture was heated at reﬂux for 72 h. The reaction mixture was

concentrated to one-third of its volume and added to 10% aq. K₂CO₃

solution, and the product was extracted with tert-butylmethyl ether.

Combined extracts were washed with brine, dried over MgSO₄,

ﬁltered through a plug of silica gel, and concentrated. The residual

foam was dried under reduced pressure at 50 °C for 16 h to aﬀord the

4690

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

[M + H]⁺. HRMS (ESI+): m/z calcd for C₁₈H₁₇N₅O₂BrF₄[M + H]⁺

490.04963, found 490.04965.

t e r t - B u t y l ( ( 2 R , 5 R ) - 5 - ( 2 - F l u o r o - 5 - ( 5 - c y a n o - 3 -

methylpicolinamido)phenyl)-2,5-dimethyl-2-(triﬂuoromethyl)-5,6-

dihydro-2H-1,4-oxazin-3-yl)carbamate (53). To a solution of tert-

butyl ((2R,5R)-5-(5-amino-2-ﬂuorophenyl)-2,5-dimethyl-2-(triﬂuoro-

methyl)-5,6-dihydro-2H-1,4-oxazin-3-yl)carbamate (33b) (3.49 g, 8.6

mmol) in DMF (30 mL) were added at 0−5 °C 5-cyano-3-

methylpicolinic acid (1.54 g, 9.46 mmol), EDC (1.74 g, 11.18 mmol),

and HOAt (1.52 g, 11.18 mmol), and the reaction mixture was stirred

overnight at 20 °C. The reaction mixture was added to sat. aq.

NaHCO₃solution, and the product was extracted with tert-

butylmethylether. Combined extracts were washed with brine, dried

over MgSO₄, ﬁltered, and concentrated. The crude product was 2×

recrystallized with diisopropylether to provide the title compound 53

(3.88 g, 81%) as a white crystalline solid. ¹H NMR (400 MHz,

CDCl₃) δ 11.08−11.03 (m, 1H), 10.03 (s, 1H), 8.72 (d, J = 1.9 Hz,

1H), 7.95 (d, J = 1.8 Hz, 1H), 7.77 (dt, J = 8.5, 3.4 Hz, 1H), 7.61 (dd,

J = 7.1, 2.6 Hz, 1H), 7.13 (dd, J = 11.5, 8.8 Hz, 1H), 4.42 (d, J = 12.2

Hz, 1H), 4.09 (d, J = 12.1 Hz, 1H), 2.85 (s, 3H), 1.70 (s, 3H), 1.59−

1.53 (m, 12H). MS m/z = 550 [M + H]⁺.

rac-N-(3-((3R*,6R*)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-

3,6-dihydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-cyanopyrimi-

dine-2-carboxamide Triﬂuoroacetate (rac-64b). rac-64b was

obtained according to the procedure described for compounds 53

and 54 starting from rac-33b and 5-cyano-3-methylpyrazine-2-

carboxylic acid (63b) as an oﬀ-white solid (16, 94%). ¹H NMR

(400 MHz, CDCl₃) δ 11.16 (s, 1H), 9.95 (s, 1H), 9.18 (s, 2H), 8.96

(s, 2H), 7.95−7.87 (m, 1H), 7.57−7.49 (m, 1H), 7.10 (dd, J = 11.5,

8.7 Hz, 1H), 4.45 (d, J = 12.5 Hz, 1H), 4.01 (d, J = 12.5 Hz, 1H),

1.73 (s, 3H), 1.67 (s, 3H). MS m/z = 437 [M + H]⁺.

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuoro-phenyl)-5-cyanopicolinamide

Hydrochloride (64c). 64c was obtained according to the procedure

described for compounds 53 and 54 starting from 33b and 5-cyano-3-

methylpyrazine-2-carboxylic acid (63c) as an oﬀ-white solid (86,

1

91%). H NMR (400 MHz, DMSO-d₆) δ 10.85 (s, 1H), 9.26−9.10

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-cyano-3-methylpicoli-

namide (54). To a solution of tert-butyl ((2R,5R)-5-(2-ﬂuoro-5-(5-

cyano-3-methylpicolinamido)phenyl)-2,5-dimethyl-2-(triﬂuorometh-

yl)-5,6-dihydro-2H-1,4-oxazin-3-yl)carbamate (53) (3.86 g, 7.0

mmol) in CH₂Cl₂(30 mL) was added triﬂuoroacetic acid (5.1 mL,

70 mmol), and the reaction mixture was stirred for 2.5 h at 22 °C.

The reaction mixture was added to excess 10% aq. NaHCO₃solution,

and the product was extracted with EtOAc. Combined extracts were

washed with brine, dried over MgSO₄, ﬁltered, and concentrated. The

crude product was 2× recrystallized from iPrOH and dried at 50 °C

for 24 h to provide the title compound 54 (2.64 g, 83%) as a colorless

(m, 1H), 8.56 (dd, J = 8.2, 2.2 Hz, 1H), 8.32−8.16 (m, 1H), 7.93 (dd,

J = 7.4, 2.7 Hz, 1H), 7.84−7.62 (m, 1H), 7.16 (dd, J = 11.8, 8.9 Hz,

1H), 6,25 (br s, 2H), 3.94 (d, J = 11.5 Hz, 1H), 4.81 (d, J = 11.6 Hz,

1H), 1.49 (s, 3H), 1.43 (s, 3H). HRMS (ESI+): m/z calcd for

C₂₀H₁₈N₅O₂F₄[M + H]⁺436.13911, found 436.13904.

1

solid. M.p. 101−102 °C. TLC (CH₂Cl₂/MeOH 10:1) R_f=0.16. H

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-cyano-3-methylpyra-

zine-2-carboxamide (64d). 64d was obtained according to the

procedure described for compounds 53 and 54 starting from 33b and

5-cyano-3-methylpyrazine-2-carboxylic acid (63d) as an oﬀ-white

solid (55, 79%). ¹H NMR (400 MHz, MeOD-d₄) δ 9.01 (s, 1H), 7.86

(dd, J = 7.4, 2.8 Hz, 1H), 7.74 (m, 1H), 7.20 (dd, J = 11.8, 7.7 Hz,

1H), 4.19 (d, J = 11.5 Hz, 1H), 3.35 (d, J = 11.4 Hz, 1H), 2.93 (s,

3H), 1.57 (s, 3H), 1.32 (s, 3H). HRMS (ESI+): m/z calcd for

C₂₀H₁₉N₆O₂F₄[M + H]⁺451.15001, found 451.15018.

NMR (600 MHz, DMSO-d₆) δ 10.73 (s, 1H), 8.99 (s, 1H), 8.41 (s,

1H), 7.79 (dd, J = 7.4, 2.8 Hz, 1H), 7.72 (m, 1H), 7.16 (dd, J = 11.8,

8.7 Hz, 1H), 6.09 (s, 2H), 3.91 (d, J = 11.5 Hz, 1H), 3.81 (d, J = 11.4

Hz, 1H), 2.54 (s, 3H), 1.47 (s, 3H), 1.42 (s, 3H). ¹³C NMR (151

1

MHz, DMSO-d₆) δ 163.74, 156.17 (d, J_CF= 241.9 Hz), 153.68,

2

152.31, 148.78, 143.31, 134.19, 133.29 (d, J_CF= 13.1 Hz), 133.14,

1

124.65 (q, J_CF= 287.6 Hz), 121.93, 120.66 (³J_CF= 8.5 Hz), 116.58,

2

115.99 (d, ²J_CF= 25.2 Hz), 110.03, 72.07 (q, J_CF= 27.2 Hz), 67.27,

54.97, 25.63, 18.53, 18.29. ¹⁹F NMR (376 MHz, DMSO-d₆) δ −74.67

(s, 3F), −117.10 (s, 1F). HRMS (ESI+): m/z calcd for

C₂₁H₁₉N₅O₂F₄[M + H]⁺450.15476, found 450.15485.

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-3-chloro-5-cyanopicoli-

namide Hydrochloride (64e). 64e was obtained according to the

procedure described for compounds 53 and 54 starting from 33b and

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-bromopyrimidine-2-

carboxamide Hydrochloride (64a). 64a was obtained according to

the procedure described for compounds 53 and 38a starting from 33b

and 5-bromopyrimidine-2-carboxylic acid (63a) as an oﬀ-white solid

1

3-chloro-5-cyanopicolinic acid (63e) as a white solid (81, 95%). H

NMR (600 MHz, DMSO-d₆) δ 10.92 (s, sH), 9.12 (d, J = 1.5 Hz,

1H), 8.82 (d, J = 1.5 Hz, 1H), 7.71 (d, J = 5.7 Hz, 2H), 7.25−7.08

(m, 1H), 6.11 (s, 2H), 3.93 (d, J = 11.5 Hz, 1H), 3.79 (d, J = 11.4 Hz,

1H), 1.46 (s, 3H), 1.42 (s, 3H). HRMS (ESI+): m/z calcd for

C₂₀H₁₇N₅O₂ClF₄[M + H]⁺470.10014, found 470.10043.

1

(85, 88%). H NMR (600 MHz, DMSO-d₆) δ 11.7 (s, 1H), 11.1 (s,

1H), 9.61 (s, 2H), 9.32 (s, 2H), 7.99 (m, 1H), 7.93 (d, J = 3.2 Hz,

1H), 7.36 (dd, J = 11.8, 8.7 Hz, 1H), 4.34 (d, J = 11.5 Hz, 1H), 4.08

(d, J = 11.4 Hz, 1H), 1.65 (s, 3H), 1.59 (s, 3H). MS m/z = 490/492

4691

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-3,5-dichloropicolina-

mide Hydrochloride (64j). 64j was obtained according to the

procedure described for compounds 53 and 54 starting from 33b and

3-Amino-N-(3-((3R,6R)-5-amino-3,6-dimethyl-6-(triﬂuorometh-

yl)-3,6-dihydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-cyanopicoli-

namide (64f). 64f was obtained according to the procedure described

for compounds 53 and 54 starting from 33b and 3-amino-5-

cyanopicolinic acid (63f) as a light-yellow foam (53, 77%). ¹H NMR

(600 MHz, DMSO-d₆) δ 10.51 (s, 1H), 8.21 (s, 1 H), 7.84 (m, 1H),

7.69 (m, 1H), 7.63 (s, 1H), 7.24 (s, 2H), 7.13 (dd, J = 11.2, 7.6 Hz,

1H), 6.08 (s, 1H), 3.80−3.92 (m, 2H), 1.48 (s, 3H), 1.39 (s, 3H).

HRMS (ESI+): m/z calcd for C₁₈H₁₇N₅O₂BrF₄[M + H]⁺451.15001,

found 451.15012.

1

3,5-dichloropicolinic acid (63j) as a white solid (90, 88%). H NMR

(400 MHz, DMSO-d₆) δ 10.72 (s, 1H), 8.87−8.64 (m, 1H), 8.43 (d, J

= 1.9 Hz, 1H), 7.84−7.59 (m, 2H), 7.29−7.04 (m, 1H), 6.05 (br s,

2H), 3.91 (d, J = 11.7 Hz, 1H), 3.77 (d, J = 11.3 Hz, 1H), 1.45 (s,

3H), 1.40 (s, 3H). HRMS (ESI+): m/z calcd for C₁₉H₁₇N₄O₂Cl₂F₄

[M + H]⁺479.06592, found 479.06598.

ASSOCIATED CONTENT

■

sı

* Supporting Information

The Supporting Information is available free of charge at

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c02143.

Preparation details and analytical data for compounds 4,

rac-20, rac-(3R*,6R*)-38a−d, rac-(3R*,6S*)-38e, rac-

49, and 50 (PDF)

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-methoxy-3-methylpi-

colinamide (64g). 64g was obtained according to the procedure

described for compounds 53 and 54 starting from 33b and 5-

methoxy-3-methylpicolinic acid (63g) as a light-yellow foam (62,

Molecular formula strings and some data (CSV)

1

94%). H NMR (400 MHz, DMSO-d₆) δ 11.54 (s, 1H), 10.67 (s,

Accession Codes

1H), 9.55 (m, 2H), 8.24 (d, J = 2.5 Hz, 1H), 7.98 (m, 1H), 7.89 (dd,

J = 7.7, 2.4 Hz, 1H), 7.43 (d, J = 3.3 Hz, 1H), 7.30 (dd, J = 12.1, 8.8

Hz, 1H), 4.31 (d, J = 12.6 Hz, 1H), 4.07 (d, J = 12.8 Hz, 1H), 3.91 (s,

3H), 2.62 (s, 3H), 1.76 (s, 3H), 1.72 (s, 3H). HRMS (ESI+): m/z

calcd for C₂₁H₂₃N₄O₃F₄[M + H]⁺455.17008, found 455.17020.

Crystal structure coordinates and structure factors have been

deposited in the Protein Data Bank (PDB) with the accession

codes 7B1E (3), 7B1P (38a), and 7B1Q (54). Authors will

release the atomic coordinates and experimental data upon

article publication.

AUTHOR INFORMATION

■

Corresponding Author

Rainer Machauer − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland; orcid.org/0000-0001-

5081-6268; Email: rainer.machauer@novartis.com

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-5-(diﬂuoromethoxy)-3-

methylpicolinamide Hydrochloride (64h). 64h was obtained

according to the procedure described for compounds 53 and 54

starting from 33b and 5-(diﬂuoromethoxy)-3-methylpicolinic acid

(63h) as a light-yellow foam (76, 62%). ¹H NMR (400 MHz, DMSO-

d₆) δ 11.53 (s, 1H), 10.78 (s, 1H), 9.54 (m, 2H), 8.45 (d, J = 2.5 Hz,

1H), 7.97 (m, 1H), 7.88 (dd, J = 7.5, 2.5 Hz, 1H), 7.75 (d, J = 2.3 Hz,

1H), 7.45 (t, 1H), 7.32 (dd, J = 12.1, 8.8 Hz, 1H), 4.32 (d, J = 12.6

Hz, 1H), 4.08 (d, J = 12.6 Hz, 1H), 2.60 (s, 3H), 1.75 (s, 3H), 1.72

(s, 3H). HRMS (ESI+): m/z calcd for C₂₁H₂₁N₄O₃F₆[M + H]⁺

491.15124, found 491.15131.

Authors

Heinrich Rueeger − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Rainer Lueoend − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Siem J. Veenstra − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Philipp Holzer − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Konstanze Hurth − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Markus Voegtle − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Mathias Frederiksen − Global Discovery Chemistry, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

N-(3-((3R,6R)-5-Amino-3,6-dimethyl-6-(triﬂuoromethyl)-3,6-di-

hydro-2H-1,4-oxazin-3-yl)-4-ﬂuorophenyl)-3-chloro-5-

(diﬂuoromethoxy)picolinamide (64i). 64i was obtained according to

the procedure described for compounds 53 and 54 starting from 33b

and 3-chloro-5-(diﬂuoromethoxy)picolinic acid (63i) as a white solid

(92, 86%). ¹H NMR (600 MHz, DMSO-d₆) δ 10.74 (s, 1H), 8.59 (d,

J = 2.3 Hz, 1H), 8.13 (d, J = 2.3 Hz, 1H), 7.74 (dd, J = 7.3, 2.7 Hz,

1H), 7.72−7.66 (m, 1H), 7.50 (t, J = 72.7 Hz, 1H), 7.17 (dd, J = 11.7,

8.8 Hz, 1H), 6.10 (s, 2H), 3.93 (d, J = 11.5 Hz, 1H), 3.79 (d, J = 11.5

Hz, 1H), 1.47 (s, 3H), 1.42 (s, 3H). HRMS (ESI+): m/z calcd for

C₂₀H₁₈N₄O₃ClF₆[M + H]⁺511.09661, found 511.09677.

Jean-Michel Rondeau − Center for Proteomic Chemistry,

Structural Biology Platform, Novartis Institutes for

4692

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

BioMedical Research, Novartis Pharma AG, CH-4057 Basel,

Switzerland

Marina Tintelnot-Blomley − Global Discovery Chemistry,

Novartis Institutes for BioMedical Research, Novartis

Pharma AG, CH-4057 Basel, Switzerland

PrOH, 2-propanol; PD, pharmacodynamics; PK, pharmacoki-

netic; SAR, structure−activity relationship; sat., saturated;

SFC, chiral supercritical ﬂuid chromatography; THF, tetrahy-

drofuran; TFA, triﬂuoroacetic acid; TFAA, triﬂuoroacetic

anhydride; V_ss, volume of distribution

Laura H. Jacobson − Department of Neuroscience, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Matthias Staufenbiel − Department of Neuroscience, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Grit Laue − Metabolism and Pharmacokinetics, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

Ulf Neumann − Department of Neuroscience, Novartis

Institutes for BioMedical Research, Novartis Pharma AG,

CH-4057 Basel, Switzerland

REFERENCES

■

(1) Ising, C.; Stanley, M.; Holtzman, D. M. Current thinking on the

mechanistic basis of Alzheimer ’s and implications for drug develop-

ment. Clin. Pharmacol. Ther. 2015, 98, 469−471.

(2) Heneka, M. T.; Carson, M. J.; Khoury, J. E.; Landreth, G. E.;

Brosseron, F.; Feinstein, D. L.; Jacobs, A. H.; Wyss-Coray, T.;

Vitorica, J.; Ransohoff, R. M.; Herrup, K.; Frautschy, S. A.; Finsen, B.;

Brown, G. C.; Verkhratsky, A.; Yamanaka, K.; Koistinaho, J.; Latz, E.;

Halle, A.; Petzold, G. C.; Town, T.; Morgan, D.; Shinohara, M. L.;

Perry, V. H.; Holmes, C.; Bazan, N. G.; Brooks, D. J.; Hunot, S.;

Joseph, B.; Deigendesch, N.; Garaschuk, O.; Boddeke, E.; Dinarello,

C. A.; Breitner, J. C.; Cole, G. M.; Golenbock, D. T.; Kummer, M. P.

Neuroinflammation in Alzheimer ’s disease. Lancet Neurol. 2015, 14,

388−405.

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c02143

(3) Villemagne, V. L.; Pike, K. E.; Chetelat, G.; Ellis, K. A.; Mulligan,

R. S.; Bourgeat, P.; Ackermann, U.; Jones, G.; Szoeke, C.; Salvado, O.;

Martins, R.; O’Keefe, G.; Mathis, C. A.; Klunk, W. E.; Ames, D.;

Masters, C. L.; Rowe, C. C. Longitudinal assessment of A β and

cognition in aging and Alzheimer disease. Ann. Neurol. 2011, 69,

181−192.

(4) Villemagne, V. L.; Burnham, S.; Bourgeat, P.; Brown, B.; Ellis, K.

A.; Salvado, O.; Szoeke, C.; Macaulay, S. L.; Martins, R.; Maruff, P.;

Ames, D.; Rowe, C. C.; Masters, C. L. Amyloid beta deposition,

neurodegeneration, and cognitive decline in sporadic Alzheimer ’s

disease: a prospective cohort study. Lancet Neurol. 2013, 12, 357−

367.

(5) Jack, C. R., Jr.; Knopman, D. S.; Jagust, W. J.; Shaw, L. M.; Aisen,

P. S.; Weiner, M. W.; Petersen, R. C.; Trojanowski, J. Q. Hypothetical

model of dynamic biomarkers of the Alzheimer ’s pathological cascade.

Lancet Neurol. 2010, 9, 119−128.

(6) Sperling, R.; Mormino, E.; Johnson, K. The evolution of

preclinical Alzheimer ’s disease: implications for prevention trials.

Neuron 2014, 84, 608−622.

Author Contributions

The manuscript was written through contributions of all

authors. All authors have given approval to the ﬁnal version of

the manuscript.

Notes

The authors declare the following competing ﬁnancial

interest(s): H.R., R.L., R.M., S.J.V., P.H., K.H., M.V., M.F.,

J.-M.R., M.T.-B., L.H.J., M.S., G.L. and U.N. are current or

former employees of Novartis Pharma AG, which has a

commercial interest to develop BACE1 inhibitors.

ACKNOWLEDGMENTS

We thank K. Barker, M. Furegati, C. Hartwieg, M. Hediger, S.

Jacquier, P. Koeninger, C. Rolando, T. Ruppen, H. Schu

■

̈

tz, U.

Stauss, M. Vogt, J. Zadrobilek, and T. Zimmermann for

synthetic support; T. Dittmar, V. Trappe, I. Brzak, L. Perrot,

and U. Furrer for in vitro and in vivo studies; K. Beltz and R.

Ramos for compound formulation support; S. Lehmann and E.

Wirth for X-ray support; and R. Hemmig for the preparation of

refolded BACE1. We are grateful to the machine and beamline

groups at the Paul Scherrer Institute, Villigen, Switzerland, for

the X-ray data collection of the BACE1 complexes with

compounds 3 and 54.

(7) Siemers, E. R.; Sundell, K. L.; Carlson, C.; Case, M.;

Sethuraman, G.; Liu-Seifert, H.; et al. Phase 3 solanezumab trials:

secondary outcomes in mild Alzheimer ’s disease patients. Alzheimer’s

Dementia 2016, 12, 110−120.

(8) Masters, C. L.; Selkoe, D. J. Biochemistry of amyloid beta-

protein and amyloid deposits in Alzheimer disease. Cold Spring

Harbor Perspect. Med. 2012, 2, No. a006262.

(9) Selkoe, D. J. Resolving controversies on the path to Alzheimer ’s

therapeutics. Nat. Med. 2011, 17, 1060−1065.

ABBREVIATIONS USED

(10) Abramowski, D.; Wiederhold, K. H.; Furrer, U.; Jaton, A. L.;

Neuenschwander, A.; Runser, M. J.; Danner, S.; Reichwald, J.;

Ammaturo, D.; Staab, D.; Stoeckli, M.; Rueeger, H.; Neumann, U.;

Staufenbiel, M. Dynamics of Abeta turnover and deposition in

different beta-amyloid precursor protein transgenic mouse models

following gamma-secretase inhibition. J. Pharmacol. Exp. Ther. 2008,

327, 411−424.

■

aq., aqueous; AUC, area under the curve; BACE, β-site APP-

cleaving enzyme; Boc, butyloxycarbonyl; t-BuOK, potassium

tert-butoxide; CL, clearance; CSF, cerebrospinal ﬂuid; CNS,

central nervous system; CYP, cytochrome P450 enzyme;

DME, 1,2-dimethoxyethane; DBU, 1,8-diazabicycloundec-7-

ene; DIPEA, N,N-diisopropylethylamine; DMF, N,N-dime-

thylformamide; DMSO, dimethyl sulfoxide; EDC, 1-ethyl-3-

(3-dimethylaminopropyl)carbodiimide; ee, enantiomeric ex-

cess; eq., equivalent; Et₂O, diethylether; EtOAc, ethyl acetate;

EtOH, ethanol; ER, eﬄux ratio; hERG, human ether-a-go-go

related gene; HOAt, 1-hydroxy-7-azabenzotriazole; iPrOH, iso-

propanol; LC, liquid chromatography; MDCK, Madin−Darby

canine kidney; MDR1, multidrug-resistance gene 1; MeOH,

methanol; MS, mass spectrospopy; Ms, methanesulfonyl; Ms-

Cl, methanesulfonyl chloride; NaOAc, sodium acetate; NMM,

N-methylmorpholine; NMP, N-methyl-2-pyrrolidone; Ns,

nosyl; Ns-Cl, nosyl chloride; PEG, poly(ethylene glycol); i-

(11) Rabe, S.; Reichwald, J.; Ammaturo, D.; de Strooper, B.; Saftig,

P.; Neumann, U.; Staufenbiel, M. The Swedish APP mutation alters

the effect of genetically reduced BACE1 expression on the APP

processing. J. Neurochem. 2011, 119, 231−239.

(12) McConlogue, L.; Buttini, M.; Anderson, J. P.; Brigham, E. F.;

Chen, K. S.; Freedman, S. B.; Games, D.; Johnson-Wood, K.; Lee, M.;

Zeller, M.; Liu, W.; Motter, R.; Sinha, S. Partial reduction of BACE1

has dramatic effects on Alzheimer plaque and synaptic pathology in

APP transgenic mice. J. Biol. Chem. 2007, 282, 26326−26334.

(13) Ostermann, N.; Eder, J.; Eidhoff, U.; Zink, F.; Hassiepen, U.;

Worpenberg, S.; Maibaum, J.; Simic, O.; Hommel, U.; Gerhartz, B.

Crystal structure of human BACE2 in complex with a hydroxyethyl-

amine transition-state inhibitor. J. Mol. Biol. 2006, 355, 249−261.

4693

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

(14) Willem, M.; Garratt, A. N.; Novak, B.; Citron, M.; Kaufmann,

S.; Rittger, A.; DeStrooper, B.; Saftig, P.; Birchmeier, C.; Haass, C.

Control of peripheral nerve myelination by the beta-secretase BACE1.

Science 2006, 314, 664−676.

(15) Velanac, V.; Unterbarnscheidt, T.; Hinrichs, W.; Gummert, M.

N.; Fischer, T. M.; Rossner, M. J.; Trimarco, A.; Brivio, V.; Taveggia,

C.; Willem, M.; Haass, C.; Mobius, W.; Nave, K. A.; Schwab, M. H.

Bace1 processing of NRG1 type III produces a myelin-inducing signal

but is not essential for the stimulation of myelination. Glia 2012, 60,

203−217.

central nervous system (CNS) and non-CNS marketed drugs. J.

Pharmacol. Exp. Ther. 2002, 303, 1029−1037.

(28) Hitchcock, S. A. Structural modifications that alter the P-

glycoprotein efflux properties of compounds. J. Med. Chem. 2012, 55,

4877−4895.

(29) Didziapetris, R.; Japertas, P.; Avdeef, A.; Petrauskas, A.

Classification analysis of P-glycoprotein substrate specificity. J. Drug

Targeting 2003, 11, 391−406.

(30) Norinder, U.; Haeberlein, M. Computational approaches to the

prediction of the blood-brain distribution. Adv. Drug Delivery Rev.

2002, 54, 291−313.

(31) Clark, D. E. Rapid calculation of polar molecular surface area

and its application to the prediction of transport phenomena. 1.

prediction of intestinal absorption. J. Pharm. Sci. 1999, 88, 807−814.

(32) Aronov, A. M. Ligand structural aspects of hERG channel

blockade. Curr. Top. Med. Chem. 2008, 8, 1113−1127.

(33) Rueeger, H.; Lueoend, R.; Rogel, O.; Rondeau, J. M.; Möbitz,

H.; Machauer, R.; Jacobson, L.; Staufenbiel, M.; Desrayaud, S.;

Neumann, U. Discovery of cyclic sulfone hydroxyethylamines as

potent and selective beta-site APP-cleaving enzyme 1 (BACE1)

inhibitors: structure-based design and in vivo reduction of amyloid

beta-peptides. J. Med. Chem. 2012, 55, 3364−3386.

(34) Rueeger, H.; Lueoend, R.; Machauer, R.; Veenstra, S. J.;

Jacobson, L. H.; Staufenbiel, M.; Desrayaud, S.; Rondeau, J.-M.;

Möbitz, H.; Neumann, U. Discovery of cyclic sulfoxide hydroxyethyl-

amines as potent and selective β -site APP-cleaving enzyme 1

(BACE1) inhibitors: structure based design and in vivo reduction

of amyloid β -peptides. Bioorg. Med. Chem. Lett. 2013, 23, 5300−5306.

(35) Edwards, P. D.; Albert, J. S.; Sylvester, M.; Aharony, D.;

Andisik, D.; Callaghan, O.; Campbell, J. B.; Carr, R. A.; Chessari, G.;

Congreve, M.; Frederickson, M.; Folmer, R. H. A.; Geschwindner, S.;

Koether, G.; Kolmodin, K.; Krumrine, J.; Mauger, R. C.; Murray, C.

W.; Olsson, L.-L.; Patel, S.; Spear, N.; Tian, G. Application of

fragment-based lead generation to the discovery of novel, cyclic

amidine β -secretase inhibitors with nanomolar potency, cellular

activity, and high ligand efficiency. J. Med. Chem. 2007, 50, 5912−

5925.

(36) Malamas, M. S.; Erdei, J.; Gunawan, I.; Barnes, K.; Johnson, M.;

Hui, Y.; Turner, J.; Hu, Y.; Wagner, E.; Fan, K.; Olland, A.; Bard, J.;

Robichaud, A. J. Aminoimidazoles as potent and selective human

beta-secretase (BACE1) inhibitors. J. Med. Chem. 2009, 52, 6314−

6323.

(37) Kobayashi, N.; Ueda, K.; Itoh, N.; Suzuki, S.; Sakaguchi, G.;

Kato, A.; Yukimasa, A.; Hori, A.; Koriyama, Y.; Haraguchi, H.; Yasui,

K.; Kanda, Y. Preparation of 2-aminodihydrothiazine derivatives as β-

secretase inhibitors. WO2007049532A1, 2007.

(38) Jennings, L. D.; Cole, D. C.; Stock, J. R.; Sukhdeo, M. N.;

Ellingboe, J. W.; Cowling, R.; Jin, G.; Manas, E. S.; Fan, K. Y.;

Malamas, M. S.; Harrison, B. L.; Jacobsen, S.; Chopra, R.; Lohse, P.

A.; Moore, W. J.; O’Donnell, M. M.; Hu, Y.; Robichaud, A. J.; Turner,

M. J.; Wagner, E.; Bard, J. Acylguanidine inhibitors of beta-secretase:

optimization of the pyrrole ring substituents extending into the S1 ′

substrate binding pocket. Bioorg. Med. Chem. Lett. 2008, 18, 767−771.

(39) Oehlrich, D.; Prokopcova, H.; Gijsen, H. J. The evolution of

amidine-based brain penetrant BACE1 inhibitors. Bioorg. Med. Chem.

Lett. 2014, 24, 2033−2045.

(40) May, P. C.; Willis, B. A.; Lowe, S. L.; Dean, R. A.; Monk, S. A.;

Cocke, P. J.; Audia, J. E.; Boggs, L. N.; Borders, A. R.; Brier, R. A.;

Calligaro, D. O.; Day, T. A.; Ereshefsky, L.; Erickson, J. A.;

Gevorkyan, H.; Gonzales, C. R.; James, D. E.; Jhee, S. S.;

Komjathy, S. F.; Li, L.; Lindstrom, T. D.; Mathes, B. M.; Martenyi,

F.; Sheehan, S. M.; Stout, S. L.; Timm, D. E.; Vaught, G. M.; Watson,

B. M.; Winneroski, L. L.; Yang, Z.; Mergott, D. J. The potent BACE1

inhibitor LY2886721 elicits robust central Abeta pharmacodynamic

responses in mice, dogs, and humans. J. Neurosci. 2015, 35, 1199−

1210.

(16) Eggert, S.; Paliga, K.; Soba, P.; Evin, G.; Masters, C. L.;

Weidemann, A.; Beyreuther, K. The proteolytic processing of the

amyloid precursor protein gene family members APLP-1 and APLP-2

involves alpha-, beta-, gamma-, and epsilon-like cleavages: modulation

of APLP-1 processing by n-glycosylation. J. Biol. Chem. 2004, 279,

18146−18156.

(17) Barao, S.; Gärtner, A.; Leyva-Díaz, E.; Demyanenko, G.;

̃

Munck, S.; Vanhoutvin, T.; Zhou, L.; Schachner, M.; López-Bendito,

G.; Maness, P. F.; De Strooper, B. , Patricia, F.;, B. Antagonistic effects

of BACE1 and APH1B-γ -secretase control axonal guidance by

regulating growth cone collapse. Cell Rep. 2015, 12, 1367−1376.

(18) Kuhn, P. H.; Koroniak, K.; Hogl, S.; Colombo, A.; Zeitschel,

U.; Willem, M.; Volbracht, C.; Schepers, U.; Imhof, A.; Hoffmeister,

A.; Haass, C.; Rossner, S.; Brase, S.; Lichtenthaler, S. F. Secretome

protein enrichment identifies physiological BACE1 protease sub-

strates in neurons. EMBO J. 2012, 31, 3157−3168.

(19) Wessels, A. M.; Lines, C.; Stern, R. A.; Kost, J.; Voss, T.;

Mozley, L. H.; Furtek, C.; Mukai, Y.; Aisen, P. S.; Cummings, J. L.;

Tariot, P. N.; Vellas, B.; Dupre, N.; Randolph, C.; Michelson, D.;

Andersen, S. W.; Shering, C.; Sims, J. R.; Egan, M. F. Cognitive

outcomes in trials of two BACE inhibitors in Alzheimer ’s disease.

Alzheimer’s Dementia 2020, 16, 1483−1492.

(20) Ghosh, A. K.; Osswald, H. L. BACE1 (beta-secretase)

inhibitors for the treatment of Alzheimer ’s disease. Chem. Soc. Rev.

2014, 43, 6765−6813.

(21) Yuan, J.; Venkatraman, S.; Zheng, Y.; McKeever, B. M.; Dillard,

L. W.; Singh, S. B. Structure-based design of beta-site APP cleaving

enzyme 1 (BACE1) inhibitors for the treatment of Alzheimer ’s

disease. J. Med. Chem. 2013, 56, 4156−4180.

(22) Wager, T. T.; Villalobos, A.; Verhoest, P. R.; Hou, X.; Shaffer,

C. L. Strategies to optimize the brain availability of central nervous

system drug candidates. Expert Opin. Drug Discovery 2011, 6, 371−

381.

(23) Wager, T. T.; Chandrasekaran, R. Y.; Hou, X.; Troutman, M.

D.; Verhoest, P. R.; Villalobos, A.; Will, Y. Defining desirable central

nervous system drug space through the alignment of molecular

properties, in vitro ADME, and safety attributes. ACS Chem. Neurosci.

2010, 1, 420−434.

(24) Feng, B.; Mills, J. B.; Davidson, R. E.; Mireles, R. J.; Janiszewski,

J. S.; Troutman, M. D.; de Morais, S. M. In vitro P-glycoprotein assays

to predict the in vivo interactions of P-glycoprotein with drugs in the

central nervous system. Drug Metab. Dispos. 2008, 36, 268−275.

(25) Doran, A.; Obach, R. S.; Smith, B. J.; Hosea, N. A.; Becker, S.;

Callegari, E.; Chen, C.; Chen, X.; Choo, E.; Cianfrogna, J.; Cox, L.

M.; Gibbs, J. P.; Gibbs, M. A.; Hatch, H.; Hop, C. E.; Kasman, I. N.;

Laperle, J.; Liu, J.; Liu, X.; Logman, M.; Maclin, D.; Nedza, F. M.;

Nelson, F.; Olson, E.; Rahematpura, S.; Raunig, D.; Rogers, S.;

Schmidt, K.; Spracklin, D. K.; Szewc, M.; Troutman, M.; Tseng, E.;

Tu, M.; Van Deusen, J. W.; Venkatakrishnan, K.; Walens, G.; Wang,

E. Q.; Wong, D.; Yasgar, A. S.; Zhang, C. The impact of P-

glycoprotein on the disposition of drugs targeted for indications of the

central nervous system: evaluation using the MDR1A/1B knockout

mouse model. Drug Metab. Dispos. 2005, 33, 165−174.

(26) Raub, T. J. P-glycoprotein recognition of substrates and

circumvention through rational drug design. Mol. Pharmceutics 2006,

3, 3−25.

(27) Mahar Doan, K. M.; Humphreys, J. E.; Webster, L. O.; Wring,

S. A.; Shampine, L. J.; Serabjit-Singh, C. J.; Adkison, K. K.; Polli, J. W.

Passive permeability and P-glycoprotein-mediated efflux differentiate

(41) May, P. C.; Dean, R. A.; Lowe, S. L.; Martenyi, F.; Sheehan, S.

M.; Boggs, L. N.; Monk, S. A.; Mathes, B. M.; Mergott, D. J.; Watson,

B. M.; Stout, S. L.; Timm, D. E.; Smith Labell, E.; Gonzales, C. R.;

4694

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

Nakano, M.; Jhee, S. S.; Yen, M.; Ereshefsky, L.; Lindstrom, T. D.;

Calligaro, D. O.; Cocke, P. J.; Greg Hall, D.; Friedrich, S.; Citron, M.;

Audia, J. E. Robust central reduction of amyloid-beta in humans with

an orally available, non-peptidic beta-secretase inhibitor. J. Neurosci.

2011, 31, 16507−16516.

(42) Rombouts, F. J.; Tresadern, G.; Delgado, O.; Martinez-

Lamenca, C.; Van Gool, M.; Garcia-Molina, A.; Alonso de Diego, S.

A.; Oehlrich, D.; Prokopcova, H.; Alonso, J. M.; Austin, N.; Borghys,

H.; Van Brandt, S.; Surkyn, M.; De Cleyn, M.; Vos, A.; Alexander, R.;

Macdonald, G.; Moechars, D.; Gijsen, H.; Trabanco, A. A. 1,4-

Oxazine beta-secretase 1 (BACE1) inhibitors: from hit generation to

orally bioavailable brain penetrant leads. J. Med. Chem. 2015, 58,

8216−8235.

(43) Scott, J. D.; Li, S. W.; Brunskill, A. P.; Chen, X.; Cox, K.;

Cumming, J. N.; Forman, M.; Gilbert, E. J.; Hodgson, R. A.; Hyde, L.

A.; Jiang, Q.; Iserloh, U.; Kazakevich, I.; Kuvelkar, R.; Mei, H.;

Meredith, J.; Misiaszek, J.; Orth, P.; Rossiter, L. M.; Slater, M.; Stone,

J.; Strickland, C. O.; Voigt, J. H.; Wang, G.; Wang, H.; Wu, Y.;

Greenlee, W. J.; Parker, E. M.; Kennedy, M. E.; Stamford, A. W.

Discovery of the 3-imino-1,2,4-thiadiazinane 1,1-dioxide derivative

verubecestat (MK-8931) - a beta-site amyloid precursor protein

cleaving enzyme 1 inhibitor for the treatment of Alzheimer ’s disease. J.

Med. Chem. 2016, 59, 10435−10450.

(44) Butler, C. R.; Brodney, M. A.; Beck, E. M.; Barreiro, G.; Nolan,

C. E.; Pan, F.; Vajdos, F.; Parris, K.; Varghese, A. H.; Helal, C. J.; Lira,

R.; Doran, S. D.; Riddell, D. R.; Buzon, L. M.; Dutra, J. K.; Martinez-

Alsina, L. A.; Ogilvie, K.; Murray, J. C.; Young, J. M.; Atchison, K.;

Robshaw, A.; Gonzales, C.; Wang, J.; Zhang, Y.; O’Neill, B. T.

Discovery of a series of efficient, centrally efficacious BACE1

inhibitors through structure-based drug design. J. Med. Chem. 2015,

58, 2678−2702.

(45) Cheng, Y.; Brown, J.; Judd, T. C.; Lopez, P.; Qian, W.; Powers,

T. S.; Chen, J. J.; Bartberger, M. D.; Chen, K.; Dunn, R. T., 2nd;

Epstein, O.; Fremeau, R. T., Jr.; Harried, S.; Hickman, D.; Hitchcock,

S. A.; Luo, Y.; Minatti, A. E.; Patel, V. F.; Vargas, H. M.; Wahl, R. C.;

Weiss, M. M.; Wen, P. H.; White, R. D.; Whittington, D. A.; Zheng,

X. M.; Wood, S. An orally available BACE1 inhibitor that affords

robust CNS Abeta reduction without cardiovascular liabilities. ACS

Med. Chem. Lett. 2015, 6, 210−215.

(46) Jeppsson, F.; Eketjall, S.; Janson, J.; Karlstrom, S.; Gustavsson,

S.; Olsson, L. L.; Radesater, A. C.; Ploeger, B.; Cebers, G.; Kolmodin,

K.; Swahn, B. M.; von Berg, S.; Bueters, T.; Falting, J. Discovery of

AZD3839, a potent and selective BACE1 inhibitor clinical candidate

for the treatment of Alzheimer disease. J. Biol. Chem. 2012, 287,

41245−41257.

(47) Schechter, I.; Berger, A. On the size of the active site in

proteases. I. papain. Biochem. Biophys. Res. Commun. 1967, 27, 157−

162.

(51) Veenstra, S. J.; Rueeger, H.; Voegtle, M.; Lueoend, R.; Holzer,

P.; Hurth, K.; Tintelnot-Blomley, M.; Frederiksen, M.; Rondeau, J.

M.; Jacobson, L.; Staufenbiel, M.; Neumann, U.; Machauer, R.

Discovery of amino-1,4-oxazines as potent BACE-1 inhibitors. Bioorg.

Med. Chem. Lett. 2018, 28, 2195−2200.

(52) Morgenthaler, M.; Schweizer, E.; Hoffmann-Roder, A.; Benini,

F.; Martin, R. E.; Jaeschke, G.; Wagner, B.; Fischer, H.; Bendels, S.;

Zimmerli, D.; Schneider, J.; Diederich, F.; Kansy, M.; Muller, K.

Predicting and tuning physicochemical properties in lead optimiza-

tion: amine basicities. ChemMedChem 2007, 2, 1100−1115.

(53) Gijsen, H. J. M.; Alonso de Diego, S. A.; De Cleyn, M.; Garcia-

Molina, A.; Macdonald, G. J.; Martinez-Lamenca, C.; Oehlrich, D.;

Prokopcova, H.; Rombouts, F. J. R.; Surkyn, M.; Trabanco, A. A.; Van

Brandt, S.; Van den Bossche, D.; Van Gool, M.; Austin, N.; Borghys,

H.; Dhuyvetter, D.; Moechars, D. Optimization of 1,4-oxazine beta-

secretase 1 (BACE1) inhibitors toward a clinical candidate. J. Med.

Chem. 2018, 61, 5292−5303.

(54) Badiger, S.; Chebrolu, M.; Hurth, K.; Jacquier, S.; Lueoend, R.

M.; Machauer, R.; Rueeger, H.; Tintelnot-Blomley, M.; Veenstra, S. J.;

Voegtle, M. Novel heterocyclic derivatives as BACE inhibitors and

their preparation and use in the treatment of neurological disorders.

WO2012095469A1, 2012.

(55) Neumann, U.; Jacobson, L. H.; Perrot, L.; Beltz, K.; Vogg, B.;

Brzak, I.; Tanja, D.; Trappe, V.; Kolly, C.; Frieauff, W.; Dubost, V.;

Jivkov, M. N.; Theil, D.; Staufenbiel, M.; Shimshek, D. R. Preclinical

pharmacology of BACE inhibitor CNP520. Alzheimer’s Dementia

2016, 12, P433−P434.

(56) Curphey, T. J. Thionation with the reagent combination of

phosphorus pentasulfide and hexamethyldisiloxane. J. Org. Chem.

2002, 67, 6461−6473.

(57) Moser, H.; Fliri, A.; Steiger, A.; Costello, G.; Schreiber, J.;

Eschenmoser, A. Poly(dipeptamidinium)-Salze: Definition und

̈

Methoden zur praparativen Herstellung. Helv. Chim. Acta 1986, 69,

1224−1262.

(58) Shen, M.; Li, C. Asymmetric iodolactamization induced by

chiral oxazolidine auxiliary. J. Org. Chem. 2004, 69, 7906−7909.

(59) Gataullin, R. R.; Afon’kin, I. S.; Fatykhov, A. A.; Spirikhin, L.

V.; Abdrakhmanov, I. B. Iodocyclization of N-(2-nitrophenyl)- and N-

phenyl-N ′-[2-(alk-1-enyl)phenyl]ethanimidamides. Mendeleev Com-

mun. 2001, 11, 201−202.

(60) Brauer, D. J.; Schenk, S.; Roßenbach, S.; Tepper, M.; Stelzer,

O.; Häusler, T.; Sheldrick, W. S. Water soluble phosphines: part XIII.

chiral phosphine ligands with amino acid moieties. J. Organomet.

Chem. 2000, 598, 116−126.

(61) Huang, H.; La, D. S.; Cheng, A. C.; Whittington, D. A.; Patel,

V. F.; Chen, K.; Dineen, T. A.; Epstein, O.; Graceffa, R.; Hickman, D.;

Kiang, Y. H.; Louie, S.; Luo, Y.; Wahl, R. C.; Wen, P. H.; Wood, S.;

Fremeau, R. T., Jr. Structure- and property-based design of

aminooxazoline xanthenes as selective, orally efficacious, and CNS

penetrable BACE inhibitors for the treatment of Alzheimer ’s disease.

J. Med. Chem. 2012, 55, 9156−9169.

(62) Box, K.; Bevan, C.; Comer, J.; Hill, A.; Allen, R.; Reynolds, D.

High-throughput measurement of pKa values in a mixed-buffer linear

pH gradient system. Anal. Chem. 2003, 75, 883−892.

(63) Neumann, U.; Rueeger, H.; Machauer, R.; Veenstra, S. J.;

Lueoend, R. M.; Tintelnot-Blomley, M.; Laue, G.; Beltz, K.; Vogg, B.;

Schmid, P.; Frieauff, W.; Shimshek, D. R.; Staufenbiel, M.; Jacobson,

L. H. A novel BACE inhibitor NB-360 shows a superior

pharmacological profile and robust reduction of amyloid-beta and

neuroinflammation in APP transgenic mice. Mol. Neurodegener. 2015,

10, No. 44.

(48) Wu, Y.-J.; Guernon, J.; Yang, F.; Snyder, L.; Shi, J.; McClure,

A.; Rajamani, R.; Park, H.; Ng, A.; Lewis, H.; Chang, C. Y.; Camac,

D.; Toyn, J. H.; Ahlijanian, M. K.; Albright, C. F.; Macor, J. E.;

Thompson, L. A. Targeting the BACE1 active site flap leads to a

potent inhibitor that elicits robust brain A β reduction in rodents. ACS

Med. Chem. Lett. 2016, 7, 271−276.

(49) Woltering, T. J.; Wostl, W.; Hilpert, H.; Rogers-Evans, M.;

Pinard, E.; Mayweg, A.; Gobel, M.; Banner, D. W.; Benz, J.; Travagli,

M.; Pollastrini, M.; Marconi, G.; Gabellieri, E.; Guba, W.; Mauser, H.;

Andreini, M.; Jacobsen, H.; Power, E.; Narquizian, R. BACE1

inhibitors: a head group scan on a series of amides. Bioorg. Med. Chem.

Lett. 2013, 23, 4239−4243.

(50) Hilpert, H.; Guba, W.; Woltering, T. J.; Wostl, W.; Pinard, E.;

Mauser, H.; Mayweg, A. V.; Rogers-Evans, M.; Humm, R.;

Krummenacher, D.; Muser, T.; Schnider, C.; Jacobsen, H.; Ozmen,

L.; Bergadano, A.; Banner, D. W.; Hochstrasser, R.; Kuglstatter, A.;

David-Pierson, P.; Fischer, H.; Polara, A.; Narquizian, R. beta-

Secretase (BACE1) inhibitors with high in vivo efficacy suitable for

clinical evaluation in Alzheimer ’s disease. J. Med. Chem. 2013, 56,

3980−3995.

(64) Rochin, L.; Hurbain, I.; Serneels, L.; Fort, C.; Watt, B.; Leblanc,

P.; Marks, M. S.; De Strooper, B.; Raposo, G.; van Niel, G. BACE2

processes PMEL to form the melanosome amyloid matrix in pigment

cells. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 10658−10663.

(65) Shimshek, D. R.; Jacobson, L. H.; Kolly, C.; Zamurovic, N.;

Balavenkatraman, K. K.; Morawiec, L.; Kreutzer, R.; Schelle, J.; Jucker,

M.; Bertschi, B.; Theil, D.; Heier, A.; Bigot, K.; Beltz, K.; Machauer,

R.; Brzak, I.; Perrot, L.; Neumann, U. Pharmacological BACE1 and

4695

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Journal of Medicinal Chemistry

pubs.acs.org/jmc

Article

BACE2 inhibition induces hair depigmentation by inhibiting PMEL17

processing in mice. Sci. Rep. 2016, 6, No. 21917.

(78) Kroemer, M.; Dreyer, M. K.; Wendt, K. U. APRV - a program

for automated data processing, refinement and visualization. Acta

Crystallogr. D Biol. Crystallogr. 2004, 60, 1679−1682.

(66) Beckmann, N.; Doelemeyer, A.; Zurbrugg, S.; Bigot, K.; Theil,

D.; Frieauff, W.; Kolly, C.; Moulin, P.; Neddermann, D.; Kreutzer, R.;

Perrot, L.; Brzak, I.; Jacobson, L. H.; Staufenbiel, M.; Neumann, U.;

Shimshek, D. R. Longitudinal noninvasive magnetic resonance

imaging of brain microhemorrhages in BACE inhibitor-treated APP

transgenic mice. Neurobiol. Aging 2016, 45, 50−60.

(67) Schelle, J.; Hasler, L. M.; Gopfert, J. C.; Joos, T. O.;

Vanderstichele, H.; Stoops, E.; Mandelkow, E. M.; Neumann, U.;

Shimshek, D. R.; Staufenbiel, M.; Jucker, M.; Kaeser, S. A. Prevention

of tau increase in cerebrospinal fluid of APP transgenic mice suggests

downstream effect of BACE1 inhibition. Alzheimer’s Dementia 2017,

13, 701−709.

(79) Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.;

̈

Gros, P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges,

M.; Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G.

L. Crystallography & NMR system: A new software suite for

macromolecular structure determination. Acta Crystallogr., Sect. D:

Biol. Crystallogr. 1998, 54, 905−921.

(80) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features

and development of Coot. Acta Crystallogr. D Biol. Crystallogr. 2010,

66, 486−501.

(68) Bacioglu, M.; Maia, L. F.; Preische, O.; Schelle, J.; Apel, A.;

Kaeser, S. A.; Schweighauser, M.; Eninger, T.; Lambert, M.; Pilotto,

A.; Shimshek, D. R.; Neumann, U.; Kahle, P. J.; Staufenbiel, M.;

Neumann, M.; Maetzler, W.; Kuhle, J.; Jucker, M. Neurofilament light

chain in blood and CSF as marker of disease progression in mouse

models and in neurodegenerative diseases. Neuron 2016, 91, 56−66.

(69) Schelle, J.; Wegenast-Braun, B. M.; Fritschi, S. K.; Kaeser, S. A.;

Jährling, N.; Eicke, D.; Skodras, A.; Beschorner, N.; Obermueller, U.;

Häsler, L. M.; Wolfer, D. P.; Mueggler, T.; Shimshek, D. R.;

Neumann, U.; Dodt, H. U.; Staufenbiel, M.; Jucker, M. Early A β

reduction prevents progression of cerebral amyloid angiopathy. Ann.

Neurol. 2019, 86, 561−571.

(70) Esterházy, D.; Stutzer, I.; Wang, H.; Rechsteiner, M. P.;

Beauchamp, J.; Dobeli, H.; Hilpert, H.; Matile, H.; Prummer, M.;

Schmidt, A.; Lieske, N.; Boehm, B.; Marselli, L.; Bosco, D.; Kerr-

Conte, J.; Aebersold, R.; Spinas, G. A.; Moch, H.; Migliorini, C.;

Stoffel, M. Bace2 is a beta cell-enriched protease that regulates

pancreatic beta cell function and mass. Cell Metab. 2011, 14, 365−

377.

(71) Matafora, V.; Farris, F.; Restuccia, U.; Tamburri, S.; Martano,

G.; Bernardelli, C.; Sofia, A.; Pisati, F.; Casagrande, F.; Lazzari, L.;

Marsoni, S.; Bonoldi, E.; Bachi, A. Amyloid aggregates accumulate in

melanoma metastasis modulating YAP activity. EMBO Rep. 2020, 21,

No. e50446.

(72) Neumann, U.; Ufer, M.; Jacobson, L. H.; Rouzade-Dominguez,

̈

M. L.; Huledal, G.; Kolly, C.; Luönd, R. M.; Machauer, R.; Veenstra,

S. J.; Hurth, K.; Rueeger, H.; Tintelnot-Blomley, M.; Staufenbiel, M.;

Shimshek, D. R.; Perrot, L.; Frieauff, W.; Dubost, V.; Schiller, H.;

Vogg, B.; Beltz, K.; Avrameas, A.; Kretz, S.; Pezous, N.; Rondeau, J.

M.; Beckmann, N.; Hartmann, A.; Vormfelde, S.; David, O. J.; Galli,

B.; Ramos, R.; Graf, A.; Lopez Lopez, C. The BACE-1 inhibitor

CNP520 for prevention trials in Alzheimer ’s disease. EMBO Mol. Med.

2018, 10, No. e9316.

(73) Zhu, K.; Xiang, X.; Filser, S.; Marinkovic, P.; Dorostkar, M. M.;

Crux, S.; Neumann, U.; Shimshek, D. R.; Rammes, G.; Haass, C.;

Lichtenthaler, S. F.; Gunnersen, J. M.; Herms, J. Beta-site amyloid

precursor protein cleaving enzyme 1 inhibition impairs synaptic

plasticity via seizure protein 6. Biol. Psychiatry 2018, 83, 428−437.

(74) Neumann, U.; Machauer, R.; Shimshek, D. R. The β -secretase

(BACE) inhibitor NB-360 in preclinical models: from amyloid-β

reduction to downstream disease-relevant effects. Br. J. Pharmacol.

2019, 176, 3435−3446.

(75) Wohnsland, F.; Faller, B. High-throughput permeability pH

profile and high-throughput alkane/water log P with artificial

membranes. J. Med. Chem. 2001, 44, 923−930.

(76) Hanessian, S.; Yun, H.; Hou, Y.; Yang, G.; Bayrakdarian, M.;

Therrien, E.; Moitessier, N.; Roggo, S.; Veenstra, S.; Tintelnot-

Blomley, M.; Rondeau, J. M.; Ostermeier, C.; Strauss, A.; Ramage, P.;

Paganetti, P.; Neumann, U.; Betschart, C. Structure-based design,

synthesis, and memapsin 2 (BACE) inhibitory activity of carbocyclic

and heterocyclic peptidomimetics. J. Med. Chem. 2005, 48, 5175−

5190.

(77) Kabsch, W. Automatic processing of rotation diffraction data

from crystals of initially unknown symmetry and cell constants. J.

Appl. Crystallogr. 1993, 26, 795−800.

4696

https://doi.org/10.1021/acs.jmedchem.0c02143

J. Med. Chem. 2021, 64, 4677−4696

Article Doi

Synthesis of the Potent, Selective, and Efficacious β-Secretase (BACE1) Inhibitor NB-360

DOI: 10.1021/acs.jmedchem.0c02143

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.jmedchem.0c02143

Full text of DOI:10.1021/acs.jmedchem.0c02143

Products guided by the article

R&D Labs maybe for 1429869-71-5

Relevant to this article

Hot Product