The exploration of aza-quinolines as hematopoietic prostaglandin D synthase (H-PGDS) inhibitors with low brain exposure

Article abstract of DOI:10.1016/j.bmc.2020.115791

GlaxoSmithKline and Astex Pharmaceuticals recently disclosed the discovery of the potent H-PGDS inhibitor GSK2894631A 1a (IC₅₀ = 9.9 nM) as part of a fragment-based drug discovery collaboration with Astex Pharmaceuticals. This molecule exhibited good murine pharmacokinetics, allowing it to be utilized to explore H-PGDS pharmacology in vivo. Yet, with prolonged dosing at higher concentrations, 1a induced CNS toxicity. Looking to attenuate brain penetration in this series, aza-quinolines, were prepared with the intent of increasing polar surface area. Nitrogen substitutions at the 6- and 8-positions of the quinoline were discovered to be tolerated by the enzyme. Subsequent structure activity studies in these aza-quinoline scaffolds led to the identification of 1,8-naphthyridine 1y (IC₅₀ = 9.4 nM) as a potent peripherally restricted H-PGDS inhibitor. Compound 1y is efficacious in four in vivo inflammatory models and exhibits no CNS toxicity.

Full text of DOI:10.1016/j.bmc.2020.115791

Bioorg. Med. Chem. 28 (2020) 115791
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The exploration of aza-quinolines as hematopoietic prostaglandin D
synthase (H-PGDS) inhibitors with low brain exposure
,
Rodolfo Cadillaa, David N. Deatona *, Young Doa, Patricia A. Elkinsb, Daniela Ennulatb,
Jeffrey H. Gussb, Jason Holta, Michael R. Jeunea, Andrew G. Kingb, Jan C. Klapwijkb,
H. Fritz Kramera, Nicholas J. Kramerb, Susan B. Laffanb, Paresh I. Masuriab,
Alan V. McDougala, Paul N. Mortensond, Caterina Musettib, Gregory E. Peckhama,
Beth L. Pietrakb, Chuck Poolea, Daniel J. Pricea, Alan R. Rendinab, Girish Satib, Gordon Saxtyd,
Barry G. Shearera, Lisa M. Shewchuka, Helen F. Sneddonc, Eugene L. Stewarta,
J. Darren Stuarta, Dean N. Thomasb, Stephen A. Thomsona, Paris Wardb, Joseph W. Wilsona,
Tiahshun Xua, Mark A. Youngmanb
a GlaxoSmithKline, 5 Moore Drive, P.O. Box 13398, Research Triangle Park, NC 27709, USA
b GlaxoSmithKline, 1250 South Collegeville Road, Collegeville, PA 19426, USA
c GlaxoSmithKline, Gunnels Wood Road, Stevenage, Hertfordshire SG1 2NY, UK
d Astex Pharmaceuticals, 436 Cambridge Science Park, Milton Road, Cambridge CB4 0QA, UK
A R T I C L E I N F O
A B S T R A C T
Keywords:
GlaxoSmithKline and Astex Pharmaceuticals recently disclosed the discovery of the potent H-PGDS inhibitor
GSK2894631A 1a (IC50 = 9.9 nM) as part of a fragment-based drug discovery collaboration with Astex Phar-
maceuticals. This molecule exhibited good murine pharmacokinetics, allowing it to be utilized to explore H-
PGDS pharmacology in vivo. Yet, with prolonged dosing at higher concentrations, 1a induced CNS toxicity.
Looking to attenuate brain penetration in this series, aza-quinolines, were prepared with the intent of increasing
polar surface area. Nitrogen substitutions at the 6- and 8-positions of the quinoline were discovered to be
tolerated by the enzyme. Subsequent structure activity studies in these aza-quinoline scaffolds led to the iden-
tification of 1,8-naphthyridine 1y (IC50 = 9.4 nM) as a potent peripherally restricted H-PGDS inhibitor. Com-
pound 1y is efficacious in four in vivo inflammatory models and exhibits no CNS toxicity.
Prostaglandin D2
PGD2
Hematopoietic prostaglandin D synthase
H-PGDS
H-PGDS inhibitor
CNS exposure
1. Introduction
progressive loss of muscle function, loss of ambulation, respiratory
insufficiency, and cardiomyopathy, leading to a life expectancy of ~ 30
Duchenne muscular dystrophy (DMD) is a degenerative skeletal
muscle disease, resulting from exon deletions, insertions, point muta-
tions, duplications, splicing errors, or small deletions in the dystrophin
gene on Xp21.1 The myocyte protein dystrophin helps connect the
sarcolemma and the basal lamina of the extracellular matrix to the actin
cytoskeleton contractile apparatus, supporting muscle strength and
function. Lack of dystrophin results in destabilization of the glycopro-
tein complex of the sarcolemma with calcium influx, inflammation, and
mitochondrial dysfunction, leading to muscle necrosis, fibrosis, and fat
deposition.2 As muscle repair and reconstruction fails to keep up with
muscle deterioration and loss, this debilitating orphan disease results in
years due to cardiorespiratory complications.3 Dystrophin is the third
largest gene in the human genome, containing 79 exons. With over
7,000 identified mutations, it is the most common cause of dystro-
phinopathy. Due to its X chromosomal linkage, it is primarily a male
disease, affecting approximately 1 in 5000 male births. There is no cure
for DMD, although treatments, including corticosteroids, cardiovascular
drugs (ACE inhibitors and β-blockers), osteoporosis drugs, ventilation,
and physical therapy, have delayed disease progression, prolonged life
expectancy, and improved quality of life.4–5 With its high unmet medical
need and associated costs, numerous therapies are under investigation
for DMD.6–7 Multiple gene therapies seek to target dystrophin directly,
* Corresponding author.
E-mail address: david.n.deaton@gsk.com (D.N. Deaton).
https://doi.org/10.1016/j.bmc.2020.115791
Received 25 April 2020; Received in revised form 18 September 2020; Accepted 24 September 2020
Available online 3 October 2020
0968-0896/© 2020 Elsevier Ltd. All rights reserved.

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
including AAV-delivered mini-dystrophins, stop codon readthrough
agents, CRISPR/CAS9 corrective vectors, and exon skipping agents.
Agents intended to affect muscle repair are also under consideration,
such as myoblast transplant, utrophin modulators targeting cytoskeletal
protein upregulation, and myostatin inhibitors. In addition, a host of
indirect mechanisms are being explored including histone deacetylase
inhibitors, anti-inflammatory agents, anti-fibrotic drugs, calcium reba-
lancing treatments, and supportive care drugs.
inhibitors and study their effects in animal models. A number of com-
panies, including Osaka (HQL-79),21 Evotec,22 AstraZeneca,23 Pfizer,24
and Sanofi (SAR191801B)25 have published their efforts, with some of
this work being recently reviewed.26 Furthermore, Taiho has entered the
clinic with TAS-205 to treat Duchenne muscular dystrophy patients.13
The structures of several representative H-PGDS tool compounds are
depicted in Fig. 1. Although these developments are promising, attempts
to translate efficacious tool compounds in animal models into safe and
effective approved drugs in humans encounter many difficulties. These
difficulties include interspecies alterations in pharmacokinetics related
to plasma protein binding differences,27–28 phase 1 metabolism in-
equalities, due to variations in enzyme expression, activity, and drug
affinity,29 phase 2 conjugation distinctions,30 gut microbiome coloni-
zation disparities,31 and drug metabolite discrepancies. Furthermore, in
humans, pharmacokinetics and pharmacodynamics can vary due to
age,32 sex,33–34 and ethnicity,35 as well as between healthy volunteers
and diseased patient populations.36 Moreover, animal disease models
may not translate to human efficacy in diseased patients.37 All of these
complications can make translations from preclinical species to humans
troublesome, potentially leading to poor human exposure,38–40 human
toxicity, and/or lack of efficacy in humans. Due to these high attrition
rates, maximizing the probability of clinical success in treating diseases
with aberrant PGD2 production requires the progression of multiple H-
PGDS chemotypes into human studies to ultimately discover a H-PGDS
medicine, necessitating continued efforts to discover and develop new
H-PGDS agents.
Several lines of evidence link the anti-inflammatory target hemato-
poietic prostaglandin D synthase (H-PGDS, EC 5.3.99.2) to DMD.8 H-
PGDS is induced in necrotic muscle fibers of DMD patients.9 Indeed a
metabolite of PGD2, 11,15-dioxo-9a-hydroxy-2,3,4,5-tetranorprosta-
1,20-dioic acid (tetranor-PGDM), is elevated in the urine of DMD pa-
tients and correlates with disease progression.10–11 Pharmacologically, a
H-PGDS inhibitor, HQL-79, suppresses necrotic muscle volume in the
mdx mouse model of DMD, as well as attenuating a loss in muscle
strength.12 As another H-PGDS inhibitor, TAS-205, dose-dependently
decreased the total excretion of tetranor-PGDM in DMD boys in a phase I
clinical trial, there is hope that the observation in rodents may translate
to clinical efficacy.13
H-PGDS is a σ-class glutathione-S-transferase that converts prosta-
glandin PGH2 to PGD2 in inflammatory cells.14–15 It is expressed in mast
cells, eosinophils, dendritic cells (DCs), megakaryocytes, and activated
type 2 helper T lymphocytes (Th2 cells), as well as tissue-resident
macrophages such as Langerhans cells, Kupffer cells, and microglia.
PGD2 is a cyclic, oxygenated lipid derivative of the polyunsaturated fatty
acid arachidonic acid that acts as an autocrine and paracrine factor to
induce chemotaxis and stimulate cytokine production in Th2 cells, mast
cells, eosinophils and DCs. PGD2 signaling is mediated through the seven
transmembrane receptors DP1 (Gs-coupled, stimulates cAMP production
and Ca+2 release) and DP2 (Chemoattractant Receptor-homologous
molecule expressed on Th2 cells receptor (CRTH2), Gi-coupled, mobi-
lizes Ca+2).16 Inhibiting H-PGDS attenuates PGD2 production/signaling
and reduces inflammation. H-PGDS has also been implicated in
asthma,17 allergic rhinitis,18 and Krabbe disease.19 PGD2 is also pro-
duced by the evolutionarily unrelated, but functionally convergent
enzyme, lipocalin prostaglandin D synthase (L-PGDS).20 Selective H-
PGDS inhibition is expected to provide anti-inflammatory benefits,
without side effects from ablation of all PGD2 signaling via pan-PGDS
inhibition or direct antagonism of DP1 and/or DP2.
These authors recently disclosed the discovery of the H-PGDS tool
compound GSK2894631A 1a, which was elaborated from a ligand effi-
cient fragment, quinoline-3-carbonitrile (IC50 = 220,000 nM, LE =
0.43), as part of a larger fragment-based drug discovery collabo-
ration41–42 between Astex Therapeutics and GlaxoSmithKline.43 The
tool compound 1a (IC50 = 9.9 nM, LE = 0.42) is a potent, competitive,
and reversible inhibitor of H-PGDS, as measured in a RapidFire™ mass
spectrometry assay determining the inhibition of the conversion of
PGH2, generated in situ by COX-2 from arachidonic acid, to PGD2. In
addition, compound 1a exhibited good H-PGDS cellular activity (RBL
IC50 = 160 nM) in a rat basophilic leukemia (RBL) cell PGD2 production
assay. An X-ray co-crystal structure of 1a bound to hH-PGDS and the
glutathione cofactor (PDB Code #6N4E) reveals several key contribu-
tions to this high potency, including hydrogen bonds between the basic
quinoline nitrogen and an enzyme-bound, structural water molecule and
between the carboxamide NH and thiol of the glutathione cofactor, as
As a consequence of the roles of PGD2 in the biology of inflammatory
diseases, researchers have exerted significant efforts to discover H-PGDS
Fig. 1. H-PGDS Tool Compounds.
2

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
well as a face-to-face π-π stacking interaction between the quinoline ring
and the indole of 104Trp. In addition, compound 1a had a low i.v.
clearance (Cl = 5.6 mL/min/kg), moderate i.v. steady state volume of
distribution (VSS = 1.7 l/kg), long i.v. terminal half-life (t1/2 = 3.8 hr),
and good oral exposure (p.o. DNAUC = 1800 ng/hr/mL, F = 61%) in
mice, making it suitable for exploring PGD2 in vivo biology. Indeed,
compound 1a attenuated PGD2 production in an acute in vivo murine
mast cell degranulation model (ED50 = 0.032 mg/kg, EC50 = 21 nM) of
inflammation.
Unfortunately, at higher doses (≥10 mg/kg) in a chronic eccentric
muscle damage model (vide infra Fig. 6) in mice, 1a-treated animals
exhibited seizure-like activity at day four of dosing, precluding its
further development. No adverse effects were seen in lower dose groups
(≤6 mg/kg) at up to 60 days of treatment. Furthermore, another quin-
oline H-PGDS inhibitor, with a higher brain to plasma ratio than 1a,
showed seizure symptoms earlier, at day two of dosing (10 mg/kg), in
this same muscle damage model. Even though 1a did not exhibit activity
in a broad panel of over 40 protein assays at concentrations of 10,000
nM, the seizure-like symptoms were likely off-target, as they were also
seen in the chronic eccentric muscle damage model dosed in H-PGDS
(-/-) mice. H-PGDS inhibitor 1a crosses the blood:brain barrier and has a
brain to blood ratio of 0.59. Thus, elevated CNS concentrations of 1a, at
higher doses and prolonged exposures, are presumed to induce these
adverse events. Under this hypothesis, one strategy to circumvent this
toxicity is to limit brain penetration of the H-PGDS inhibitor, thus pre-
venting interference with any putative CNS protein.
There are several chemical strategies for lowering brain exposure,44
including increasing hydrogen bond donors,45 lowering intrinsic
permeability/lipophilicity,46 adding negative charges,47 increasing
active brain transporter (e.g. P-glycoprotein) efflux,48 and increasing
polar surface area (PSA)49 in the inhibitor template. The latter tactic,
increasing the PSA of the quinoline template, was explored. Generally,
molecules with PSAs greater than 90 Å2 are poorly brain penetrant.50
Substituting a nitrogen atom for a carbon atom in the quinoline ring
results in 13 Å2 increase in calculated topological polar surface area
(tPSA) (e.g. 1c tPSA = ~71 Å2 → 1e tPSA = 84 Å2).
Scheme 1. Synthesis of aza-quinolines. Reagents and conditions: a) EtOH,
80 ◦C, 84%; b) POCl3, W, 160 ◦C, 55%; c) Pd(PPh3)2Cl2, Et3SiH, MeCN, 70 ◦C,
μ
46%; d) LiOH, THF, EtOH, H2O, 50 ◦C, 99%; e) T3P® or HATU, iPr2NEt,
DMF, 21–90%.
employing triethylsilane as the reductant, yielded the aza-quinoline
ester 6a. Some carboxylic acid precursors, such as 6-methoxy-2-naph-
thoic acid 7a, 7-methoxycinnoline-3-carboxylic acid 7b, or 1,7-naph-
thyridine-3-carboxylic acid 7f, were commercially available, while
others, such as 6-methoxyquinoxaline-2-carboxylic acid 7c, were pre-
pared via literature procedures, and the rest, such as 7d-7e and 7g-7aa
were synthesized according to Schemes 1, 3, or 4. Thus, lithium hy-
droxide catalyzed hydrolysis of ester 6a provided 7-methoxy-1,5-naph-
thyridine-3-carboxylic acid 7d. These carboxylic acids 7a-7aa were
coupled to various amines, either commercially available (8a-8e, 8j, and
8l) or known in the chemical literature (8f-8i and 8k),43,51-53 to give the
target compounds 1d-1e, and 1g-1bk.
It was recognized that the strategy of inflating tPSA, via introduction
of additional nitrogens in the quinoline heterocycle, has several rami-
fications on protein–ligand interactions, creating some challenges in
intentional design. Calculated pKA’s for N1 for several prospective het-
erocycles (data not shown) confirmed the supposition that N1 basicity
could be significantly modulated, and as a result, hydrogen-bond
strengths with both a structural, protein-bound water, and with water
in bulk solution would be affected. To what extent modulating these
interactions would cancel was unclear a priori. Desolvation of an addi-
tion hydrogen-bond acceptor was expected to be a factor in binding
potency as well. Furthermore, the heterocycle’s dipole, quadrupole, and
The 7-methoxyquinazoline-3(4H)-carboxamide 1f was prepared as
shown in Scheme 2. First, the amine 9 was coupled to the chloroformate
10 to afford the carbamate 11. Then, displacement of the para-nitro-
phenol moiety of activated carbamate 11, with amine 8a, gave the urea
12. Subsequent palladium catalyzed reduction of the nitro group of urea
12, employing hydrogen as the reductant, provided the aniline 13.
Finally, condensation of aniline 13 with triethylorthoformate, followed
by ring closure, and loss of ethanol afforded the quinazoline 1f.
The 7-methoxy-1,6-naphthyridine-3-carboxylic acid 7e was synthe-
sized as illustrated in Scheme 3. First, ytterbium catalyzed condensation
of aniline 14 with acetal 15, followed by cyclization, and dehydration
afforded the 3-bromo-1,6-naphthyridine 16a. Then, palladium cata-
lyzed carbonylation of bromide 16a, with carbon monoxide, afforded
ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a. Finally, methoxide
displacement of chloride 17a and subsequent ester hydrolysis gave 1,6-
naphthyridine-3-carboxylic acid 7e. Alternatively, 7-methoxy-1,8-naph-
thyridine-3-carboxylic acid 7g was prepared from pyridine-2,6-diamine
18. First, phosphoric acid catalyzed condensation of aniline 18 with 2-
bromomalonaldehyde 19, followed by Friedel-Craft cyclization and
dehydration provided 6-bromo-1,8-naphthyridin-2-amine 16b. Then,
carbonylation of bromide 16b, catalyzed by palladium, yielded the ester
inducible dipole moments that contribute to π-π stacking strength could
be affected and conceivably alter the strength of interactions with 104Trp
in the binding site. Finally, introducing electron density in proximity to
substitution, in particular the central amide, could impact conforma-
tional preference and therefore the energy required to adopt the bound
conformation. Given the complicated balance of energetics at play, a
deliberate and methodical exploration of aza-quinolines, derived from
compound 1c, was implemented to explore their effects on H-PGDS in-
hibition, CNS exposure, and other drug properties.
2. Chemistry
The aza-quinolines could be prepared as shown in Schemes 1–4. The
1,5-naphthyridine 6a was synthesized as depicted in Scheme 1. First,
thermal condensation of aniline 2 with diethyl 2-(ethoxymethylene)
malonate 3, via 1,4-aza-Michael addition, followed by β-elimination of
ethanol, afforded the aminomethylenemalonate 4. Then, phosphorus
oxychloride catalyzed Friedel-Craft acylation of 4 yielded the 4-chloro-
aza-quinoline 5. Subsequent palladium catalyzed dechlorination of 5,
3

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Scheme 2. Synthesis of the 7-methoxyquinazoline-3(4H)-carboxamide 1f. Reagents and conditions: a) iPr2NEt, THF, 88%; b) iPr2NEt, THF, 79%; c) H2/Pd-C, MeOH,
100%; d) (EtO)3CH, 102 ◦C, 31%.
17b. Subsequent diazotization of aniline 17b provided the 7-hydroxy-
1,8-naphthyridine, which was converted to the chloride 17c via phos-
phorus oxychloride. Finally, methoxide displacement of chloride 17c
and subsequent ester hydrolysis gave 1,8-naphthyridine-3-carboxylic
acid 7g. Furthermore, aniline 16b was chlorinated with concentrated
hydrochloric acid and hydrogen peroxide to afford 6-bromo-3-chloro-
1,8-naphthyridin-2-amine 16c. Then, carbonylation of 16c, in a
similar manner as above, gave the ester 17f. Conversion of the aniline
17f to the hydroxyl 17g, via in situ generated nitrous acid, then trans-
formation to the chloride provided dichloride 17h. Treatment of the
chloride 17h with sodium methoxide afforded the ester 6o, whereas,
treatment of the chloride 17h with sodium 2,2,2-trifluoroethan-1-ol led
to carboxylic acid 7m, via in situ hydrolysis. Moreover, chlorides 17a,
17c, and 17h were converted to 7-amino/oxy derivatives 6c-6e, 6g-6k,
and 6p-6q or 7-alkyl derivatives 6f, 6l-6n, and 6r, via SNAr displace-
ment with nucleophiles or Suzuki/Negishi couplings with boronic acids/
zinc reagents, respectively. Upon hydrolysis, these esters, as well as
esters 6b, 6s, and 6t prepared according to Scheme 4, afforded car-
boxylic acids 7h-7l and 7n-7aa.
cyclization, and elimination yielded the bromide 16d. Then, carbonyl-
ation of bromide 16d, under a carbon monoxide atmosphere, catalyzed
by palladium, yielded the ester 6t.
3. Results and discussion
3.1. Aza-quinoline template SAR
As mentioned above, a consequence of this strategy to inflate the
tPSA of H-PGDS inhibitors with the introduction of additional ring ni-
trogens is that these substitutions in the quinoline template lead to al-
terations in the basicity of the ring as well as the hydrogen bonding
acceptor capacity of the N1 nitrogen of these aza-analogs, compared
with their corresponding quinolines (e.g. 1c meas. pKA = 3.59, calc. pKA
= 3.38 versus 1i meas. pKA < 2.0, calc. pKA = 0.87). These property
changes could affect the strength of a key hydrogen bond between the
N1 nitrogen and the aforementioned structural water molecule that is
observed crystallographically in the H-PGDS active site. They may also
alter the
π
-
π
stacking interaction with 104Trp and increase the des-
The ethyl 2-methoxypyrido[2,3-d]pyrimidine-6-carboxylate 6b was
synthesized as shown in Scheme 4. First, aldol condensation of aldehyde
20 with malonate 21, followed by dehydration and cyclization, afforded
the 7-oxo-7,8-dihydropyrido[2,3-d]pyrimidine 22. Then, treatment
with phosphorus oxychloride provided the chloride 23. Subsequent
palladium catalyzed dechlorination of 23, employing triethylsilane as
the reductant, yielded the pyrido[2,3-d]pyrimidine 17d. Oxidation of
the sulfide 17d with varying amounts of meta-chloroperoxybenzoic acid
then afforded either the sulfoxide 17e or the sulfone 17i. Subsequent
displacement of the oxidized sulfur moiety with oxygen or nitrogen
nucleophiles, respectively, then gave the esters 6b and 6s. Furthermore,
the 7-alkyl derivative 6t was prepared from aniline 24. First, acylation of
aniline 24 with cyclopropanecarbonyl chloride gave the amide 25.
Then, condensation of aniline 25 with ammonia, followed by
olvation penalty for an aza-quinoline relative to its corresponding
quinoline. This initial structure/activity exploration was conducted with
the 7-methoxy group to ease the synthesis of aza-analogs and avoid
exposure to difluoromethanol which might be metabolically activated to
toxic fluorophosgene. Initially, the contribution of the N1 hydrogen
bond enzyme interaction to inhibitor binding was probed by replacing
the quinoline nitrogen with a carbon atom, eliminating hydrogen
bonding acceptor ability in the analog. As shown in Table 1, the non-
basic naphthalene analog 1d (IC50 = 1300 nM) is almost 2 orders of
magnitude less potent than quinoline 1c (IC50 = 25 nM) yet pays a lower
desolvation cost than the quinoline. Thus, the N1 hydrogen bond likely
affords at least 2.4 kcal/mol (ΔG = -RTlnKi) to inhibitor binding energy,
and any aza-quinoline analog that has reduced hydrogen bond acceptor
capacity at its N1 nitrogen, should form an attenuated hydrogen bond
4

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Scheme 4. Synthesis of pyrido[2,3-d]pyrimidines. Reagents and conditions: a)
K2CO3, NEt3, DMF, 80 ◦C, 45%; b) POCl3, 100 ◦C, 94%; c) Pd(PPh3)3Cl Et3SiH,
MeCN, 80 ◦C, 49%; d) m-CPBA, CH2Cl2, 0 ◦C, 73–78%. e) NaOMe, MeOH, 37%;
f) azetidine, iPr2NEt, NMP, 100 ◦C,
μW, 11%; g) cPrCOCl, pyridine, CH2Cl2,
90%; h) NH3, MeOH, sealed tube, 80 ◦C, 67%; i) Pd(dppf)Cl2•CH2Cl2, CO,
Scheme 3. Synthesis of 1,6-naphthyridine acids. Reagents and conditions: a)
Yb(OTf)3, MeCN, 80 ◦C, 35%; b) H3PO4, 120 ◦C, 38%; c) HCl, H2O2, 25%; d) Pd
(dppf)Cl2•CH2Cl2, CO, NEt3, EtOH, 80 ◦C, 46–63%; e) NaNO2, H2SO4, r.t.;
iPr2NEt, POCl3, 80 ◦C, 55%; f) NaNO2, H2SO4, r.t.; 64%; g) iPr2NEt, POCl3,
iPr2NEt, EtOH, 80 ◦C, 65%.
N4 nitrogen forming an intramolecular hydrogen bond with the amide
NH. Indeed, the quinoxaline 1g (IC50 = 490 nM, tPSA = 84 Å2) is an
order of magnitude more potent than the cinnoline, though, it still re-
mains ~ 20-fold less potent than the original quinoline. The quinoxa-
line’s calculated N1 HBAS (13.6 kcal/mol) is over 2 kcal/mol less than
the quinoline (16.1 kcal/mol), while its basicity (meas. pKA = <2.0,
calc. pKA = 0.62) is similar to the cinnoline 1e. The quinoxaline’s
dioxane, 80 ◦C, 52%; h) R2R3NH2, iPr2NEt, DMF or NMP, 100 ◦C,
μW, 16–94%;
i) R2R3YB(OH)2 or R2R3YZnBr, Pd2(dba)3/SPhos or Pd(dppf)Cl2/CuI, Na2CO3,
PhMe or THF, 60–110 ◦C, 21–91%; j) LiOH or NaOH, MeOH or THF, H2O,
25–60 ◦C, 19–100%; k) NaOMe or NaOCH2CF3, MeOH, 60 ◦C, 98–100%.
with the protein, resulting in weaker binding of the aza-quinoline. For
other templates explored in these laboratories, this hydrogen bond
interaction can be less critical for potency.
weaker hydrogen acceptor capacity, poorer π-π stacking alignment of
104Trp indole and quinoxaline dipoles, and the increased desolvation
cost, arising from the additional ring nitrogen, likely account for its
reduced inhibitory activity relative to quinoline 1c.
Although the cinnoline analog 1e (IC50 7900 nM) accomplishes
design objectives of increasing tPSA (tPSA = 84 Å2), it is over two orders
In contrast to the cinnoline 1e, the 3,4-dihydroquinazoline analog 1f
(IC50 = 890 nM, meas. pKA = 4.31, calc. pKA = 3.99, tPSA = 74 Å2) is
slightly more basic than the parent quinoline 1c and should form a
stronger hydrogen bond with the H-PGDS protein (calc. HBAS = 15.7
kcal/mol). Its 35-fold reduced inhibition activity, despite its stronger
of magnitude less potent than quinoline 1c. The cinnoline (meas. pKA
=
<2.0, calc. pKA = 0.55) is substantially less basic than its quinoline
congener (meas. pKA = 3.59, calc. pKA = 3.38), yet its calculated N1
hydrogen bond acceptor strength (HBAS) (14.3 kcal/mol) is less than 2
kcal/mol less than the quinoline (16.1 kcal/mol). Thus, a weaker
hydrogen bond and altered desolvation penalties are probably only
partial contributors to the decreased binding. Another plausible expla-
nation to account for the observed activity relies on ground state versus
bound state energies. The cinnoline N2-nitrogen lone pair induces a
preferred unbound conformation that results in an intramolecular
hydrogen bond with the NH of the amide and this conformation differs
geometrically – based on the H-PGDS ligand bound crystal structures –
and energetically from the protein-bound conformation, which would
prefer a conformation that allows formation of a hydrogen bond be-
tween the glutathione cofactor and the amide NH (vide infra). The
preferred bound conformation would be energetically unfavorable
because of lone pair:lone pair repulsion between the N2 nitrogen and the
amide carbonyl oxygen, resulting in a lower observed binding affinity
from this significant preorganization penalty in the form of the elec-
trostatic clash, and likely accounts significantly for its > 300-fold loss in
potency relative to 1c.
hydrogen bonding potential, likely results from a reduced π-π stacking
interaction (vide infra) between the indole of 104Trp and the non-
aromatic partially saturated dihydroquinazoline ring with its reduced
planarity.
The N1 nitrogen of the 1,5-naphthyridine analog 1h (IC50 = 180 nM,
tPSA = 84 Å2) is predicted to be slightly more basic (meas. pKA = <2.0,
calc. pKA = 0.91) than the quinoxaline 1g with a similar N1 nitrogen
calculated HBAS (15.8 kcal/mol) to the quinoline 1c and only loses 7-
fold inhibitory potency relative to the parent quinoline 1c, even
though the rotamer of the amide is not conformationally biased.
Furthermore, the N1 nitrogen of the 1,6-naphthyridine analog 1i (IC50
= 45 nM, tPSA = 84 Å2) is similar in basicity (meas. pKA = <2.0, calc.
pKA = 0.87) to the 5-aza-quinoline analog 1h, with lower N1 nitrogen
calculated HBAS (14.2 kcal/mol) and yet is ~ 4-fold more active than
1h. Intriguingly, 14Arg occupies two different rotamers in the two copies
of the H-PGDS binding site in the asymmetric unit of another 1,6-naph-
thyridine 1m (vide infra), with one conformation placing the electron-
deficient guanidinium side chain in proximity to the N6 nitrogen.
Even partial occupancy of this conformation for 1i may minimize the
Consistent with the assertion that the cinnoline N2 nitrogen steers
the carbonyl oxygen away from the bound conformation, the quinoxa-
line 1g should conversely promote the bound amide orientation, with its
5

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
attenuate brain penetration through increased tPSA. The slightly lower
potency of this compound likely results from the increased binding
desolvation cost of an additional heteroatom in the 6,8-diazaquinoline
template.
Table 1
H-PGDS enzyme inhibition data for aza-quinoline templates.
Desiring to identify aza-quinoline inhibitors with similar or better
properties than previous tool compounds as potential development
candidates, structure/activity relationships (SAR) in the quinoline series
were used to set analog advancement criteria for the aza-quinoline
templates. These learnings revealed that compounds with H-PGDS
Rapid-Fire™ mass spectroscopy inhibition assay IC50 ≤ 100 nM, artifi-
cial membrane permeability assay permeabilities PApp ≥ 30 nm/sec, and
protein binding ≤ 95%, usually exhibited good cell potency in a rat
basophilic leukemia (RBL) cell H-PGDS inhibition assay. Furthermore,
analogs with RBL IC50 ≤ 300 nM, good solubility (FaS-SIF solubility ≥
0.025 mg/mL), and acceptable murine pharmacokinetics (p.o. DNAUC
≥ 1000 ng/hr/mL) usually were efficacious in vivo in the chronic murine
eccentric muscle damage model. The most potent of these aza-quinoline
analogs in the enzyme assay, the 1–6 and 1,8-naphthyridines 1i (RBL
IC50 = 190 nM) and 1k (RBL IC50 = 150 nM), were also potent inhibitors
of PGD2 production in the RBL cellular assay, so it was decided to focus
further research efforts on these templates.
#
R
IC50 ± S.
EC50 ± S.D.b
Meas.
Calc.
c
d
D.a nM
nM
pKA
pKA
1a
9.9 ± 2.7
160 ± 160
2.43
3.02
3.59
–
3.03
3.02
3.38
–
1b
1c
1d
1e
1f
1800 ±
30,000 ±
1100
2,900
25 ± 3.4
100 ± 60
>50,000
>50,000
1300 ±
840
7900e
2600
±
<2.0
4.31
0.55
3.99
890 ± 260
490 ± 260
180 ± 99
45 ± 38
4300 ±
3500
1g
1h
1i
3200 ± 620
740 ± 190
190 ± 69
<2.0
<2.0
<2.0
2.74
2.58
<2.0
0.62
0.91
0.87
2.5
An early read on metabolic stability, surprisingly revealed that the 7-
methoxy-1,6-naphthyridine 1i (t1/2 = >3 hr) and 7-methoxy-1,8-naph-
thyridine 1k (t1/2 = >3 hr) had long half-lives in an in vitro mouse
liver microsome assay, as shown in Table 2. However, and not unex-
pectedly, 1i (p.o. DNAUC = 104 ng/hr/mL) and 1k (p.o. DNAUC = 64.9
ng/hr/mL) exhibited poor oral exposure in a mouse pharmacokinetic
study, likely due to O-dealkylation of the 7-methoxy methyl group.
Because of this poor oral exposure, compound concentrations in the
brain were not determined. Since both ligand-bound protein crystal
structures and exhaustive SAR of the quinoline template revealed flex-
ibility in substitution at the 6 and 7-positions of that scaffold, those
learnings were applied to the 6- and 7-positions of the naphthyridine
scaffolds in an attempt to alter their chemical properties toward
improved pharmacokinetics.
1j
7700f ±
39,000g
16,000
±
2600
1k
1l
30 ± 30
150 ± 13
ꢀ 0.24
ꢀ 2.81
140 ± 69
2000 ±
1000
a
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 4
in the H-PGDS RapidFire™ High Throughput Mass Spectrometry Assay.
b
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 2
3.2. Aza-quinoline substitution SAR
in the H-PGDS Rat Basophilic Leukemia (RBL) cell inhibition assay.
c
Measured basic pKA.
d
Previous SAR revealed that small alkyl groups and small heterocycles
projecting from the seven position of the quinoline are tolerated, so
these modifications were incorporated into the 1,6-naphthyridine tem-
plate. As shown in Table 3, the 7-azetidine derivative 1m (IC50 = 3.9
nM, tPSA = 78 Å2) is ~ 11-fold more potent than its 7-methoxy
Calculated basic pKA. pKAs were calculated using the ChemAxon software
program prediction.
e
Individual replicate pIC50 = 5.16, 5.34, <5.0, <5.0. The mean was obtained
by removing the < qualifier and averaging the values.
f
Individual replicate pIC50 = 5.26, 5.27, <5.0, <5.0. The mean was obtained
by removing the < qualifier and averaging the values.
comparator 1i. The π-electron-rich azetidine ring nitrogen donates its
g
Individual replicate pIC50 = 4.57, <4.30. The mean was obtained by
electrons into the aromatic
π-system of the naphthyridine ring,
removing the < qualifier and averaging the values.
enhancing both its
π-
π
stacking with 104Trp and the hydrogen bond to
the structural water molecule in the active site of the protein via mod-
ulation of the N1 nitrogen basicity. These reinforced interactions, as well
as hydrophobic contributions of the propylene to the binding affinity,
likely account for its substantial increase in inhibitory activity. Like the
methoxy analog 1i, the azetidine derivative 1m was moderately stable
to mouse liver microsomes (t1/2 = 2.7 hr) but had poor oral exposure (p.
o. DNAUC = 41.0 ng/hr/mL).
desolvation penalty associated with removing the N6 nitrogen from bulk
solution and may distinguish this aza-regioisomer from the less potent
quinoxaline 1g and 1,5-naphthyridine 1h. The pKA of the N1 nitrogen of
1,8-naphthyridine 1k (IC50 = 30 nM, tPSA = 84 Å2) is ~ 1 order of
magnitude less basic (meas. pKA = 2.58, calc. pKA = -0.24) than the
quinoline 1c, with lower calculated N1 nitrogen HBAS (13.6 kcal/mol),
and yet these compounds have similar potency. Possibly, both the N1
and N8 nitrogens of the 1,8-naphthyridine 1k can interact with the
conserved water molecule and this dual polar interaction compensates
for its reduced N1 acceptor strength. Finally, the N1 nitrogen of 1,7-
The 3-fluoro and 2-methyl azetidine derivatives 1n (IC50 = 12 nM,
tPSA = 78 Å2), 1o (IC50 = 160 nM, tPSA = 78 Å2) and 1p (IC50 = 4.8 nM,
tPSA = 78 Å2) were prepared in an effort to eliminate potential meta-
bolism of the azetidine ring either through electronic or steric alter-
ations. The electron withdrawing nature of the fluorine substituent in 1n
attenuates the electron donation capabilities of the azetidine nitrogen,
lowering the compound’s potency relative to the naked azetidine 1m,
while the (S)-methyl analog 1p maintains similar activity to 1m. In
contrast, the (R)-methyl azetidine 1o is substantially less active than the
unsubstituted analog 1m. This (R)-enantiomer is likely sterically
encumbering to the protein, and thus substantially reduces its relative
potency. Unfortunately, these modifications did not improve drug
properties, as the (S)-methylazetidine 1p (t1/2 = 0.98 hr) was quickly
naphthyridine 1j is a weaker hydrogen bond acceptor (meas. pKA
=
2.74, calc. pKA = 2.5, calc. HBAS = 13.5 kcal/mol) than the N1 nitrogen
of its direct comparator quinoline 1b and 1j (IC50 = 7,700 nM, tPSA =
75 Å2) is less potent than quinoline 1b (IC50 = 1,800 nM, meas. pKA
=
3.02, calc. pKA = 3.02, HBAS = 14.8 kcal/mol, tPSA = 62 Å2).
Since the 6-aza and 8-aza-quinoline analogs 1i and 1k were both
potent H-PGDS inhibitors, the more polar pyrido[2,3-d]pyrimidine
analog 1l (IC50 = 140 nM, meas. pKA = <2.0, calc. pKA = -2.81, HBAS =
12.3 kcal/mol, tPSA = 97 Å2) was also prepared, in an effort to further
6

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Table 2
H-PGDS drug property data for aza-quinolines.
#
Structure
AMPa
PAPP
FaS-SIFb
mg/mL
mLMc
t1/2
PBd
%
DNAUCe
ng•kg•hr/mL/mg
nm/s
min
1a
240
0.070
>180f
88
1820
1i
61
0.085
0.024
0.011
0.013
0.027
0.003
0.270
>180
>180
161f
67
73
79
73
84
80
78
104
–
1k
1m
1n
1q
1s
64.9
41.0
19.3
1040
72.5
73.7
78
–
>180f
>180f
>180f
>180f
165
23
1v
25
1y
44
0.140
0.210
–
>180
146
80
54
64
52
1320
23.4
204
1af
1ag
1ah
<10
22
>180
>180f
38
0.460
46.9
1ai
1aj
<10
0.400
0.630
>180f
>180
59
56
250
715
14
1ak
130
0.280
>180
76
45.3
1an
1ao
30
0.430
0.100
147
43
262
693
180
>180
–
1ap
1at
270
>1.000
>180
>180
–
358
–
0.100
74
77.5
(continued on next page)
7

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Table 2 (continued)
#
Structure
AMPa
PAPP
FaS-SIFb
mg/mL
mLMc
t1/2
PBd
%
DNAUCe
ng•kg•hr/mL/mg
nm/s
min
1au
1av
1aw
1ba
45
0.035
0.034
0.40
>180g
>180g
>180
>180
86
86
69
82
477
441
285
681
54
34
170
>1.000
1bf
110
0.740
>180
>180
87
81
132
1bg
<3
>1.000
86.2
1bh
280
0.046
>180
88
452
1bi
1bj
1bk
150
220
30
0.700
0.031
0.120
>180
>180
>180
86
87
81
362
152
252
a
Artificial membrane permeability assay at pH = 7.4 (PAPP = apparent permeability).
Fasted state-simulated intestinal fluid solubility assay at pH = 6.5.
Mouse liver microsome metabolism assay.
b
c
d
e
f
Human serum albumin protein binding assay, % compound bound.
Dose normalized area under the curve observed for the blood concentration versus time plot for an oral dose of 3.0 mg/kg in mice.
Mouse S9 fraction assay.
Tested as the racemate.
g
metabolized by mouse liver microsomes, and, although the 2-fluoroaze-
tidine 1n (t1/2 = >3 hr) was more stable in mouse liver microsomes, it
had poor oral exposure in mice (p.o. DNAUC = 19.3 ng/hr/mL).
The 7-cyclopropyl analog 1q (IC50 = 14 nM, tPSA = 75 Å2) is less
electron donating compared to the 7-methoxy compound 1i and there-
fore provides a less basic compound that likely forms a weaker hydrogen
bond with the structural water molecule in the active site, yet it has
greater H-PGDS inhibitory activity than 1i. Presumably, the hydropho-
bic contribution of the cyclopropyl group to binding compensates for the
reduced hydrogen bonding interactions. In contrast to the methoxy and
azetidine derivatives, 1q had good in vitro metabolic stability and
excellent oral exposure in mice (mLM t1/2 = > 3.0 hr, p.o. DNAUC =
1040 ng/hr/mL).
Small substitutions at the 7-position of the 1,8-naphthyridine tem-
plate were also investigated. As shown in Table 3, the 2,2,2-trifluoroe-
thoxy analog 1r (IC50 = 350 nM, tPSA = 78 Å2), prepared to prevent
O-dealkylation, was an order of magnitude less potent that the parent 7-
methoxy-1,8-naphthyridine 1k. Although some of this activity loss could
result from the reduced hydrogen bond acceptor capacity of the N1-
nitrogen, likely steric clashes of the trifluoroethyl group with the pro-
tein prevent favorable binding to the H-PGDS protein.
In addition, the 7-azetidino analogs 1s (IC50 = 5.3 nM, meas. pKA
=
5.03, HBAS = 13.6 kcal/mol, tPSA = 78 Å2), 1t (IC50 = 13 nM, tPSA =
78 Å2), 1u (IC50 = 310 nM, tPSA = 78 Å2), and 1v (IC50 = 4.4 nM, tPSA
8

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Table 3
Table 3 (continued)
H-PGDS
enzyme
inhibition
data
for
aza-quinoline
#
R
IC50 ± S.D.a, nM
6.4 ± 7.1
EC50 ± S.D.b, nM
170 ± 140
substitution.
1ag
#
R
IC50 ± S.D.a, nM
3.9 ± 0.7
EC50 ± S.D.b, nM
24 ± 2.7
1m
a
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 4
in the H-PGDS RapidFire™ High Throughput Mass Spectrometry Assay.
1n
12 ± 2.9
170 ± 36
b
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 2
in the H-PGDS Rat Basophilic Leukemia (RBL) cell inhibition assay.
c
Individual replicate pIC50 = 8.24, >8.6, 8.40, 8.47. The mean was obtained
1o
1p
1q
1r
160 ± 110
4.8 ± 2.3
14 ± 5.8
450 ± 58
50 ± 8.2
94 ± 20
by removing the > qualifier and averaging the values.
= 78 Å2) exhibited similar SAR trends as their corresponding congeners
in the 1,6-naphthyridine template. Moreover, the 2,2,2-trifluoroethyla-
mino- and 2,2-difluoroethylamino- derivatives 1w (IC50 = 690 nM,
tPSA = 87 Å2) and 1x (IC50 = 150 nM, tPSA = 87 Å2) lost inhibitory
potency, similar to 2,2,2-trifluoroethoxy analog 1r, with the smaller and
less electron withdrawing analog 1x showing the least reduction in ac-
tivity. In addition, the two analogs 1s (p.o. DNAUC = 72.5 ng/hr/mL)
and 1v (p.o. DNAUC = 73.7 ng/hr/mL) that met the progression criteria
to murine pharmacokinetic studies exhibited poor oral exposure.
Similar to the cyclopropyl 1,6-naphthyridine analog 1q, small 7-
alkyl substitutions in the 1,8-naphthyridine series, like the cyclopropyl
1y (IC50 = 9.4 nM, meas. pKA = 3.10, HBAS = 13.6 kcal/mol, tPSA = 75
Å2), the isopropyl 1z (IC50 = 8.5 nM, tPSA = 75 Å2), and the cyclobutyl
1aa (IC50 = 10 nM, tPSA = 75 Å2) are potent H-PGDS inhibitors.
Although the cyclobutyl analog 1aa (mLM t1/2 = 0.63 hr) is rapidly
metabolized in vitro, like 1q, the cyclopropyl derivative 1y has good oral
exposure (mLM t1/2 = > 3.0 hr, p.o. DNAUC = 1320 ng/hr/mL).
Quinoline SAR further revealed that halogens, such as chloro- and
bromo-, appended to the 6-position of the quinoline template formed a
halogen bond with the side chain carboxylate of 96Asp to enhance H-
PGDS inhibitory potency. Thus, 6-chloro derivatives in the 1,8-naph-
thyridine were prepared to explore their effect on enzyme inhibition.
As shown in Table 3, the combination 6-chloro-7-methoxy analog 1ab
(IC50 = 6.6 nM, meas. pKA = <2.0, HBAS = 13.0 kcal/mol, tPSA = 84
Å2) is a 6-fold better H-PGDS inhibitor than its deschloro derivative 1i.
In contrast, the 6-chloro-7-azetidino derivatives 1ac (IC50 = 2.4 nM,
tPSA = 78 Å2) and 1ad (IC50 = 3.8 nM, tPSA = 78 Å2), as well as the 6-
chloro-7-cyclopropyl analog 1ae (IC50 = 4.7 nM, tPSA = 75 Å2), do not
have as dramatic increases in potency, relative to 1s, 1v, and 1y,
respectively. Possibly the chloro substituent prevents the bulkier 7-sub-
stituents from accessing the optimal bound conformation. Furthermore,
the combination analogs, involving the addition of the chlorine to the
1,8-naphthyridine scaffold, exhibit reduced FASSIF solubility of < 15
350 ± 230
1100 ± 260
1s
1t
5.3 ± 1.8
13 ± 12
75 ± 12
1000 ± 120
1u
1v
1w
310 ± 170
4.4 ± 1.6
690 ± 91
3000 ± 1200
14 ± 4.8
9600 ± 2600
1x
150 ± 49
12,000 ± 8200
1y
9.4 ± 2.9
8.5 ± 5.6
10 ± 9.3
42 ± 18
64 ± 5.2
120 ± 23
1z
1aa
μ
g/mL in all cases, thereby limiting progression.
7-Substituted pyrido[2,3-d]pyrimidine analogs were also prepared,
with both the 7-azetidino analog 1af (IC50 = 12 nM, tPSA = 97 Å2) and
the 7-cyclopropyl analog 1ag (IC50 = 6.4 nM, tPSA = 97 Å2) being more
potent H-PGDS inhibitors than the corresponding methoxy derivative 1l.
Yet, despite good in vitro metabolic stability (1af mLM t1/2 = 2.4 hr; 1ag
mLM t1/2 = > 3 hr), neither analog had good oral exposure (1af p.o.
DNAUC = 23.4 ng/hr/mL; 1ag p.o. DNAUC = 204 ng/hr/mL) in mice.
1ab
1ac
6.6 ± 3.1
2.4 ± 0.4
22 ± 1.8
3.9c ± 1.4
1ad
1ae
1af
3.8 ± 1.6
4.7 ± 0.9
12 ± 5.0
3.9 ± 0.3
5.2 ± 0.8
340 ± 500
3.3. Aza-quinoline amide SAR
After the exploration of 6- and 7-ring substitutions, the best aza-
quinoline scaffolds were combined with optimal amines from the
quinoline template, in an effort to produce 3-carboxamide aza-
quinolines with improved drug properties. These amines when inte-
grated into the quinoline series had demonstrated improvements in
potency (e.g. 2-(6-aminospiro[3.3]heptan-2-yl)propan-2-ol included in
9

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
1au and 2-(3-aminobicyclo[1.1.1]pentan-1-yl)propan-2-ol included in
1aw), solubility (e.g. (R)-2-((trans-4-aminocyclohexyl)amino)-3,3,3-
trifluoropropan-1-ol included in 1ao, trans-4-(3-fluoroazetidin-1-yl)
cyclohexan-1-amine included in 1ap, and trans-N1-(1,1-difluoropropan-
2-yl)cyclohexane-1,4-diamine included in 1ba), and size (e.g. 3-amino-
1-methylcyclobutanol included in 1ay, (S)-3-aminopyrrolidin-2-one
included in 1az and 2-(trans-3-aminocyclobutyl)propan-2-ol included in
1at). Although the azetidino analogs, such as 1m, 1s, and 1af, are potent
H-PGDS inhibitors, these compounds exhibit low oral exposure in mice
relative to quinoline 1a. Other amines that might provide better oral
exposures were combined with these templates. As shown in Table 4, the
7-azetidino-1,6-naphthyridine derivatives, 3-methyl-3-hydroxycyclobu-
tyl 1ah (IC50 = 56 nM), and γ-lactams 1ai (IC50 = 12 nM), and 1aj (IC50
= 37 nM), are less potent H-PGDS inhibitors than 1m. Although 1ai and
1aj have better oral exposure in mice (1ah p.o. DNAUC = 46.9 ng/hr/
mL; 1ai p.o. DNAUC = 250 ng/hr/mL; 1aj p.o. DNAUC = 715 ng/hr/mL)
than 1m, these compounds do not meet the progression criteria of ≥
1000 ng/hr/mL. Furthermore, the corresponding 1,8-naphthyridine
analogs 1ar (IC50 = 140 nM) and 1as (IC50 = 71 nM) are less potent
than 1s, and their permeability (PAPP = < 10 nm/sec) is too low for
passive oral absorption, precluding pharmacokinetic studies. In
contrast, the more potent 6-chloro-7-(2-methylazetidinyl)-1,8-naph-
thyridine analogs, (cyclobutyl)propan-2-ol 1bc (IC50 = 3.1 nM), 3-
Table 4
H-PGDS enzyme inhibition data for amide substitution.
#
R1
R2
IC50 ± S.
EC50 ± S.
D.a, nM
D.b, nM
1ah
56c
±
450 ± 15
110 ± 87
260 ± 140
280 ± 63
3.7
1ai
1aj
1ak
12c
2.9
±
37 ± 6.8
65 ± 36
1al
38 ± 19
210 ± 60
1am
1an
1ao
290 ±
910 ± 59
260 ± 17
310 ± 61
100
110 ±
methyl-3-hydroxycyclobutyls 1bd (IC50 = 5.0 nM) and 1be (IC50
=
64
5.0 nM), 4-methyl-4-hydroxycyclohexyl 1bf (IC50 = 4.6 nM), and
γ-lactam 1bg (IC50 = 5.1 nM), have higher permeabilities, possibly due
31 ± 9.7
35 ± 11
14 ± 14
to their increased lipophilicities. Yet, 1bc (t1/2 = 16 min), 1bd (t1/2
=
24 min), and 1be (t1/2 = 63 min) have low in vitro metabolic stability,
and 1bf (t1/2 > 180 min, p.o. DNAUC = 132 ng/hr/mL) and 1bg (t1/2
>
1ap
1aq
280 ± 0.0
88 ± 4.3
180 min, p.o. DNAUC = 86.2 ng/hr/mL) have poor oral exposure. Thus,
no acceptable improvements in oral exposure were realized in the 7-
azetidinyl-aza-quinolines.
The amine replacements in the 7-cyclopropyl aza-quinolines were
also explored. The 1,6-naphthyridine 1q has good oral exposure (p.o.
DNAUC = 1040 ng/hr/mL), but limited solubility (FaS-SIF = 0.027 mg/
mL). Thus, several analogs with solubility-enhancing amines - identified
from the quinoline series - were synthesized. As shown in Table 4, these
6-aza-quinoline compounds, (cyclobutyl)propan-2-ol 1ak (IC50 = 65
nM, FaS-SIF = 0.283 mg/mL), bicycle 1al (IC50 = 38 nM, FaS-SIF = 0.43
mg/mL), 3-methyl-3-hydroxycyclobutyl 1am (IC50 = 290 nM, FaS-SIF
= 0.097 mg/mL), γ-lactam 1an (IC50 = 110 nM, FaS-SIF = 0.420 mg/
mL), and 4-aminocyclohexyls 1ao (IC50 = 31 nM, FaS-SIF = 0.104 mg/
mL), 1ap (IC50 = 35 nM, FaS-SIF > 1.000 mg/mL), and 1aq (IC50 = 14
nM, FaS-SIF = 0.011 mg/mL), are slightly less potent than 1q, but are
generally more soluble. Yet, the four compounds that were progressed
into PK studies have lower oral exposures than 1q (1ak (p.o. DNAUC =
45.3 ng/hr/mL), 1an (p.o. DNAUC = 262 ng/hr/mL), 1ao (p.o. DNAUC
= 693 ng/hr/mL), 1ap (p.o. DNAUC = 358 ng/hr/mL)) in mice.
Since the 1,8-naphthyridine 1y also has good oral exposure (p.o.
DNAUC = 1,320 ng/hr/mL), additional amine analogs were prepared to
further probe the pharmacokinetic properties of this template. Of the
tertiary alcohol derivatives synthesized, the spiro enantiomers 1au
(IC50 = 3.9 nM, FaS-SIF = 0.035 mg/mL) and 1av (IC50 = 4.9 nM, FaS-
SIF = 0.034 mg/mL) are the most potent, while the shorter (cyclobutyl)
propan-2-ol 1at (IC50 = 36 nM, FaS-SIF = 0.100 mg/mL), and bicyclic
1aw (IC50 = 17 nM, FaS-SIF = 0.401 mg/mL) are less potent. The
shortest 3-methyl-3-hydroxycyclobutyl diastereomers 1ax (IC50 = 300
nM, FaS-SIF = 0.88 mg/mL) and 1ay (IC50 = 160 nM, FaS-SIF = 0.13
mg/mL) are the least potent. The SAR within this series of amines are
consistent with the general understanding that potency can be achieved
by increasing hydrophobic surface area, driven by the entropic benefits
of transferring the ligand from bulk solution. The oral exposures of the
four most potent tertiary alcohols were determined in mice (1at (FaS-
SIF = 0.100 mg/mL, p.o. DNAUC = 77.5 ng/hr/mL), 1au (FaS-SIF =
0.035 mg/mL, p.o. DNAUC = 477 ng/hr/mL), 1av (FaS-SIF = 0.034 mg/
mL, p.o. DNAUC = 441 ng/hr/mL), 1aw (FaS-SIF = 0.401 mg/mL, p.o.
1ar
1as
1at
1au
140 ±
17,000 ±
58
11,000
71 ± 42
36 ± 14
6000 ±
3300
280 ± 88
32 ± 16
3.9 ±
1.0
1av
4.9 ±
29 ± 5.5
0.3
1aw
17 ± 9.3
230 ± 0.0
1ax
1ay
1az
1ba
300 ±
7400 ±
180
1800
160 ±
1300 ± 41
640 ± 220
28 ± 7.6
66
44 ± 10
11 ± 7.3
(continued on next page)
10

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Table 4 (continued)
#
Table 5
H-PGDS inhibitor pharmacokinetic data.a
R1
R2
IC50 ± S.
EC50 ± S.
D.a, nM
D.b, nM
#
Species
t1/2, h
Cl, mL/min/kg
VSS, L/kg
F, %
B:Bb
1a
1q
1y
1y
1y
mouse
mouse
mouse
rat
3.8
1.1
2.9
5.1
6.2
5.6
22
1.7
1.6
1.6
1.6
1.0
61
0.59
0.17
0.06
–
130
71
9.0
4.5
1.9
100
92
dog
–
1bb
14 ± 6.6
50 ± 7.1
a
Mice were dosed at 1 mg/kg i.v. and 3 mg/kg p.o. with an N = 2; rats were
dosed at 0.4 mg/kg i.v. and 2.4 mg/kg p.o. with an N = 3; dogs were dosed at 0.5
mg/kg i.v. and 1 mg/kg p.o. with an N = 3.
b
Brain:blood ratio (B:B).
1bc
1bd
1be
1bf
3.1 ±
<2.5
1.2
distribution (VSS = 1.6 l/kg), and moderate terminal half-life (t1/2 = 1.1
h), with excellent oral bioavailability (F = 130%). The 1,8-naphthyri-
dine 1y has a lower i.v. clearance (Cl = 9.0 mL/min/kg), similar
steady state volume of distribution (VSS = 1.6 l/kg), and longer terminal
half-life (t1/2 = 2.9 h), as well as high oral bioavailability (F = 71%).
More importantly, both 1q and 1y have lower brain penetration with
blood:brain ratios of 0.17 for 1q and 0.06 for 1y, supporting the hy-
pothesis that increasing the PSA of these inhibitors would attenuate CNS
exposure. As 1y has very low brain exposures, it’s pharmacokinetics was
also profiled in rat and dog, where the compound also exhibited low
clearances (rat Cl = 4.5 mL/min/kg, dog Cl = 1.9 mL/min/kg) and high
oral bioavailabilities (rat F = 100%, dog F = 92%).
5.0 ±
4.7 ± 0.2
<2.5
2.3
5.0 ±
1.5
4.6 ±
3.7 ± 0.2
3.8 ± 0.2
11 ± 2.3
2.2
1bg
1bh
5.1 ±
2.7
3.9 ±
3.4. 1,8-Naphthyridine 1y binding kinetics
2.2
As 1,8-naphthyridine 1y had the best pharmacokinetics in these aza-
quinoline inhibitor series, its binding kinetics were explored further. The
equilibrium dissociation constant Kd for 1y (Kd = 0.24 ± 0.051 nM) was
measured in the presence of 1 mM glutathione by titration of the
quenching of the indole intrinsic fluorescence of 104Trp in the active site
with 1y excited at 285 nm.54 Furthermore, the kon of 1y (kon = 3.3 ±
0.07 X 107 Mꢀ 1 secꢀ 1) was measured by stopped-flow spectrofluorimetry
from the single exponential fits of intrinsic tryptophan quench time
courses, and the koff of 1y (calculated koff = 0.0079 ± 0.017 sꢀ 1) was
calculated from the values of Kd and kon (Kd = koff/kon). Inhibitor 1y
rapidly binds to and releases from the enzyme. Because the IC50 of 1y is
elevated due to prostaglandin substrate competition and active site
titration of hH-PGDS in the activity assay, the Kd is a better reflection of
the intrinsic potency of 1y.
1bi
1bj
1bk
6.4 ±
190 ± 6.2
54 ± 6.7
14 ± 2.3
6.3
7.1 ±
4.9
4.6 ±
1.4
a
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 4
in the H-PGDS RapidFire™ High Throughput Mass Spectrometry Assay.
b
S.D. = standard deviation; mean ± S.D. All inhibitors were tested with N ≥ 2
in the H-PGDS Rat Basophilic Leukemia (RBL) cell inhibition assay.
c
N = 2.
3.5. Aza-quinoline X-ray co-crystal structures
DNAUC = 285 ng/hr/mL)), but dose normalized areas under the curve
are lower than 1y. Like the 3-methyl-3-hydroxycyclobutyl di-
astereomers 1ax and 1ay, the γ-lactam 1az (IC50 = 44 nM) does not
meet the cell potency criteria for further progression, likely due to its
limited permeability (PAPP < 10 nm/sec). In contrast, the amino de-
rivatives 1ba (IC50 = 11 nM) and 1bb (IC50 = 14 nM) are potent H-PGDS
inhibitors, yet the one analog progressed to a PK study in mice, 1ba (FaS-
SIF > 1.00 mg/mL, p.o. DNAUC = 681 ng/hr/mL), has a DNAUC < 1000
ng/hr/mL.
To gain further understanding of the binding mode of aza-quinoline
H-PGDS inhibitors, protein co-crystal structures of inhibitors 1m and 1y
bound with hH-PGDS and GSH were obtained. The co-crystal structure
of the 1,6-naphthyridine 1m (PDB #6W58) is shown in Fig. 2, overlaid
with the co-crystal structure of the original 3-cyanoquinoline fragment
(PDB #6N69). Not surprisingly, 1,6-naphthyridine 1m binds in a similar
manner to the 3-cyanoquinoline fragment, with both inhibitors forming
face to face
π
-
π
stacking interactions with the indole side chain of 104Trp
Furthermore, 6-chloro derivatives of the less potent and more polar
derivatives 1at, 1ax, 1ay, and 1az were prepared, with the aim of
increasing inhibitory activity and permeability. These analogs, 1bh
(IC50 = 3.9 nM, PAPP = 280 nm/sec), 1bi (IC50 = 6.4 nM, PAPP = 150 nm/
sec), 1bj (IC50 = 7.1 nM, PAPP = 220 nm/sec), and 1bk (IC50 = 4.6 nM,
PAPP = 30 nm/sec), were substantially more potent and had higher PApp
values, yet none of them (1bh (p.o. DNAUC = 452 ng/hr/mL), 1bi (p.o.
DNAUC = 362 ng/hr/mL), 1bj (p.o. DNAUC = 152 ng/hr/mL), 1bk (p.o.
DNAUC = 252 ng/hr/mL)) have oral exposures in mice better than 1y.
As only aza-quinolines 1q and 1y met the oral DNAUC progression
criteria, these compounds were subjected to more complete pharmaco-
kinetic profiling. As shown in Table 5, the 1,6-naphthyridine 1q has a
low i.v. clearance (Cl = 22 mL/min/kg), moderate steady state volume of
(face of indole 104Trp to face of quinoline 1m distance = 3.6 Å). To
accommodate the 7-azetidine moiety into a small hydrophobic pocket
formed by 99Met, 155Ile, 152Tyr, 156Cys, and 13Gly, the 1,6-naphthyridine
ring is shifted slightly towards the solvent exposed portion of the pro-
tein. This displacement also slightly alters the interaction of the N1-
nitrogen of the 1,6-naphthyridine ring with the structural water mole-
cule that - in turn - interacts with the protein via hydrogen bonding to
the C-terminal carboxylic acid of 199Leu, the hydroxyl side chain of
159Thr, and another bound water molecule. This structural water
molecule is key to maintaining the integrity of the protein fold and the
formation of the binding site.55 The increased basicity of the 7-azetidine
derivative 1m strengthens the hydrogen bonding interaction with the
structural water molecule and, along with the increased hydrophobic
11

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
inhibitors 1a and 1y have shifted slightly relative to the 3-cyanoquino-
line fragment to accommodate the 7-difluoromethoxy and 7-cyclopropyl
moieties, respectively. However, the 1,8-naphthyridine is shifted
slightly more than the quinoline and 1,6-naphthyridine rings, and one
might speculate that this displacement allows both the N1 (structural
water oxygen to N1 naphthyridine nitrogen distance = 2.9 Å) and N8
nitrogen (structural water oxygen to N8 naphthyridine nitrogen dis-
tance = 2.9 Å) of 1y to maximally hydrogen bond to the structural water
molecule, thus strengthening this interaction. As the 7-cyclopropyl de-
rivative 1y is less basic than the 7-methoxy-1,8-naphthyridine 1k and
therefore forms a comparably weaker hydrogen bond with the structural
water, the increased potency of 1y likely arises purely from hydrophobic
interactions with the hH-PGDS protein. Moreover, none of these shifts,
relative to the 3-cyanoquinoline fragment, dramatically alter the face to
face
π
-
π
stacking interaction between the indole side chain of 104Trp and
the various aromatic rings (face of indole 104Trp to face of quinoline/
1,6-naphthyridine/1,8-naphthyridine distances all equal 3.6 Å); yet
electronic perturbations by the 7-position substituents, 3-carboxamide
group, and heteroatomic rearrangements in the ring systems, likely
Fig. 2. Co-crystal structure of inhibitor 1m bound to hH-PGDS with GSH
superimposed on the co-crystal structure of 3-cyanoquinoline complexed with
rH-PGDS and GSH. Ligand binding site of the X-ray co-crystal structure of 1m
complexed with hH-PGDS and GSH superimposed on the co-crystal structure of
3-cyanoquinoline complexed with rH-PGDS and GSH (PDB code 6N69). The
hH-PGDS carbons are colored magenta while the rH-PGDS carbons are colored
orange, inhibitor 1m carbons colored dark green, the 3-cyanoquinoline carbons
are colored light green, and both GSH carbons colored cyan. Hydrogen bonds
are depicted as yellow dashed lines. The coordinates of hH-PGDS and GSH
bound with 1m have been deposited in the Brookhaven Protein Data Bank (PDB
code 6W58). This figure was generated using PyMOL version 1.7.64.6 (The
PyMOL Molecular Graphics System, Version 1.7.64 Schrodinger, LLC).
subtlely adjust the strength of the π-π cloud interactions between each of
these rings and the 104Trp indole.
Like quinoline 1a, inhibitors 1m and 1y also form hydrogen bonds
between their 3-carboxamide NHs and the sulfur of the GSH. Further-
more, hydrophobic interactions of the chair conformation of the cyclo-
hexyl moieties with the side chains of 105Ala, 11Met, and 9Phe enhance
binding interactions with the enzyme. Thus, amide substituents with
increased hydrophobic surface area, such as the spiro[3.3]heptane 1au,
exhibit enhanced potency relative to more polar and smaller amides,
such as the 3-methyl-3-hydroxycyclobutane 1ax. Although mostly sol-
vent exposed and thus limiting its contributions to binding affinity, the
tertiary alcohol increases water solubility of these analogs. These human
H-PGDS co-crystal structures bound with ligand and cofactor help
rationalize much of the SAR observed in the aza-quinoline carboxamide
inhibitor series, as well as providing structural insight into the binding
requirements of the enzyme and a blueprint for aiding design of other H-
PGDS inhibitors.
interactions of the azetidine group with the hH-PGDS protein, ratio-
nalizes the increased potency of 1m relative to the 7-methoxy-1,6-naph-
thyridine 1i.
The co-crystal structure of the 1,8-naphthyridine 1y (PDB #6W8H) is
shown in Fig. 3, with the co-crystal structure of quinoline 1a (PDB
#6N4E) overlaid with it. As with the 1,6-naphthyridine 1m, the
3.6. 1,8-Naphthyridine 1y cross screening
Since, in addition to H-PGDS, several other enzymes also utilize the
substrate PGH2 for production of prostanoids, the potency of 1,8-naph-
thyridine 1y was determined against two other synthase enzymes in the
prostanoid pathway, lipocalin prostaglandin D synthase (L-PGDS)20 and
microsomal prostaglandin E synthase (m-PGES),56 to ensure that inhi-
bition of these enzymes was not a factor in any pharmacodynamic study
outcomes of inflammatory disease models. Because 1y did not inhibit L-
PGDS (IC50 > 10,000 nM) or m-PGES (IC50 > 100,000 nM) at the highest
concentrations tested, the compound was deemed suitable for use in vivo
to probe the protein specific effects of H-PGDS inhibition in inflamma-
tory models.
Moreover, the general off-target liabilities of 1y were also evaluated.
Against a broad panel of proteins, including 7-TM G-protein coupled
receptors (5HT1B, 5HT2A, 5HT2B, 5HT2C, α1B, α2C, β2, κ, μ, A2A, CB1, D1,
D2, H1, M1, M2, MRGPRX2, NK1, V1A), ion channels (5HT3, CaV1.2,
GABAA, KV1.5, KV7.1, NaV1.5), enzymes (AChE, COX-2, MAO-A, PDE3A
Fig. 3. Co-crystal structure of inhibitor 1y bound to hH-PGDS with GSH
superimposed on the co-crystal structure of 1a complexed with hH-PGDS and
GSH. Ligand binding site of the X-ray co-crystal structure of 1y complexed with
hH-PGDS and GSH superimposed on the co-crystal structure of 1a complexed
with hH-PGDS and GSH (PDB code 6N4E). The 1y hH-PGDS carbons are colored
green, while the 1a hH-PGDS carbons are colored gray with inhibitor 1y car-
bons colored gold, inhibitor 1a carbons are colored magenta and both GSH
carbons are colored cyan. Hydrogen bonds are depicted as yellow dashed lines.
The coordinates of hH-PGDS and GSH bound with 1y have been deposited in
the Brookhaven Protein Data Bank (PDB code 6W8H). This figure was gener-
ated using PyMOL version 1.7.64.6 (The PyMOL Molecular Graphics System,
Version 1.7.64 Schrodinger, LLC).
,
PDE4B), transcription factors (AhR, PXR), kinases (Aurora B, LCK,
PI3Kγ), and transporters (BSEP, NET, OATP1B1, SERT), 1,8-naphthyri-
dine 1y did not exhibit any activity in these assays at concentrations
up to 10,000 nM. Compound 1y also did not inhibit cytochrome p450
3A4 monooxygenase enzyme (IC50 > 100,000 nM) or block the human
ether-a-go-go-related gene potassium ion channel (Kv11.1, IC50
>
80,000 nM)57 at the highest concentrations assayed. This selectivity
profile, coupled with its high H-PGDS inhibitory potency and good
pharmacokinetic characteristics, makes 1y a suitable tool compound for
studying the pharmacodynamic effects of H-PGDS inhibition.
12

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
3.7. 1,8-Naphthyridine 1y pharmacodynamics
Inhibitor 1y was profiled in several in vivo inflammation models.
First, to evaluate the acute effects of a H-PGDS inhibitor on mast cell
release of PGD2, 1,8-naphthyridine 1y was evaluated in a murine mast
cell degranulation model of inflammation.58 Male C57BL/6J mice were
administered p.o. with vehicle or H-PGDS inhibitor 1y at various doses
(0.003, 0.01, 0.03, 0.1, 0.3, and 1 mg/kg), then anesthetized and chal-
lenged one hour later with vehicle (PBS) or the N-methyl-p-methox-
yphenethylamine/formaldehyde synthetic polymer 48/80 (i.p. 0.75
mg/mL) with abdomen massage. After seven minutes, blood was
collected (1y analysis), the mice were euthanized, and lavage fluid was
collected from the abdominal cavity (PGD2 analysis). As shown in Fig 4,
the mast cell secretagogue 48/80 induces production and release of
PGD2 into peritoneal lavage fluid versus phosphate buffered saline
(PBS). H-PGDS inhibitor 1y attenuates this PGD2 release to baseline
levels in a dose-dependent manner with an ED50 = 0.009 mg/kg (1y
blood EC50 = 3.4 nM) in this acute inflammation model. Levels of the
related prostaglandins PGE2 and PGF1α were not significantly different
between 1y treated and untreated animals after challenge with synthetic
polymer 48/80.
Fig. 5. H-PGDS inhibition blocks lipopolysaccharide challenge-induced PGD2
production in vivo. Young adult male C57BL/6J mice were administered (p.o.)
vehicle or the H-PGDS inhibitor 1y at varying doses (from 0.003 mg/kg to 1.0
mg/kg) one hour prior to LPS or PBS injection. Blood samples were withdrawn
via cardiac puncture 30 min post LPS injection and animals were then eutha-
nized and skeletal muscle tissues were isolated. PGD2 levels in plasma and
skeletal muscle were determined using LC/MS/MS. Data are depicted as means
± S.E.M and analyzed using ANOVA followed by Dunnett’s comparing vehicle
+ LPS to respective treatment group (*p < 0.05; **p < 0.01; &p < 0.001 vs.
vehicle + LPS).
Secondly, the acute effects of H-PGDS inhibition on lipopolysac-
charide (LPS)-triggered PGD2 production in mice were evaluated in vivo.
The Gram-negative bacterial endotoxin LPS binds 7-TM pattern recog-
nition receptors and activates NF-kB. These events initiate an inflam-
matory cascade that results in cytokine production and enzyme
synthesis, including COX-mediated prostaglandin synthesis. Twelve-
week-old C57BL6/N mice (n = 6/group) were dosed orally with
vehicle, or 1y at 0.003, 0.01, 0.03, 0.1, 0.3, or 1.0 mg/kg. One hour
later, animals received intraperitoneal injection of PBS or 20 ng/kg LPS.
Blood samples were withdrawn via cardiac puncture 30 min post LPS
injection and animals were then euthanized and skeletal muscle tissues
were isolated. Plasma and muscle prostaglandins and 1y levels were
determined using LC/MS/MS. In contrast to plasma which exhibits very
low (90 pg/mL) prostanoid concentrations in the basal state, skeletal
muscle exhibited a substantial amount of PGD2 (0.6 ng/g) before LPS
stimulation (Fig. 5). LPS induced a significant increase in PGD2 levels in
plasma (16-fold, p ≤ 0.001) and a modest increase in skeletal muscle
(1.7-fold, p = 0.09) at 30 min after intraperitoneal administration. In-
hibitor 1y treatment inhibited LPS-induced PGD2 increase in plasma and
skeletal muscle in a dose-dependent manner. Significant PGD2 inhibition
were achieved at 0.003 mg/kg and 0.03 mg/kg for plasma and muscle,
respectively. Plasma PGD2 levels returned to baseline at 0.3 mg/kg.
Muscle PGD2 concentrations dropped below basal levels in the two high
dose groups, suggesting both L-PGDS and H-PGDS contribute to basal
PGD2 production in skeletal muscle. There is a linear relationship be-
tween 1y dose and drug levels in blood and skeletal muscle (data not
shown). The estimated 1y ED50 for systemic and muscle specific H-PGDS
inhibition of LPS-induced PGD2 synthesis is 0.0031 mg/kg and 0.0138
mg/kg with an estimated EC50 of 2.6 nM and 1.4 ng/g in plasma and
skeletal muscle, respectively. Taken together, these results demonstrate
that 1y is a potent inhibitor of LPS-induced PGD2 synthesis in mice.
PGE2 and PGF1α were not significantly different between 1y treated and
untreated animals after challenge with LPS. The EC50 values for 1y in
both the acute murine inflammatory models (MCD EC50 = 3.4 nM, LPS
EC50 = 2.6 nM) are similar, but ~ 10-fold lower that the RBL cell EC50
value (EC50 = 42 nM). This difference could be due to intrinsic species-
related differences in compound potency between mouse and rat H-
PGDS inhibition and/or differences in species plasma protein binding.
H-PGDS inhibitor 1y was also studied in two chronic eccentric
contraction-induced muscle injury models. Male C57Bl/6 mice (10–12
weeks-old, n = 7–8/group) were administered vehicle or 1y at 1, 3, or
10 mg/kg orally before being anesthetized and challenged by unilateral
eccentric contractions of the right hind limb for 10 min (Day 0), fol-
lowed by 16 days of QD dosing with longitudinal measurements of
isometric limb torque determined for each animal at Days 1, 9, and 16.
Inhibitor 1y significantly enhanced functional recovery of injured limbs,
following eccentric contraction-induced muscle injury, at the top two
doses as shown in Fig. 6. At least part of this response was due to dose-
dependent protective effects of 1y, as determined by the initially milder
force deficit observed at Day 1 post-damage. More importantly, 1y
significantly hastened the time to full functional recovery of injured limb
muscles, with maximal efficacy observed at ≥ 10 mg/kg QD. Using an
Fig. 4. H-PGDS inhibition blocks 48/80 challenge-induced PGD2 production in
vivo. Young adult male C57BL/6J mice were administered (p.o.) vehicle or the
H-PGDS inhibitor 1y at varying doses (from 0.003 mg/kg to 1.0 mg/kg). One
hour later, mice were anesthetized, injected with (i.p.) 0.2 mL PBS or 48/80
(0.75 mg/mL), and peritoneal lavage fluid was collected after 7 min. (A) Data
are presented as means (n = 7–8/group) ± S.E.M. and analyzed using ANOVA
followed by Dunnett’s comparing vehicle + 48/80 to respective treatment
group (*p < 0.05; **p < 0.01 vs 48/80 + vehicle).
13

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
integrated measurement of limb recovery over time, the ED50 for 1y was
determined to be 3 mg/kg, with an estimated AUC0-last of 5300 hr•ng/
mL, a CMax of 1700 ng/mL, an estimated free AUC0-last of 1200 hr•ng/
mL, and a free CMax of 370 ng/mL. Free blood concentrations reached or
exceeded the cellular IC50 at doses tested, although maximal efficacy
was observed at the top (10 mg/kg) dose.
Duchenne muscular dystrophy (DMD) is a rare degenerative muscle
disorder characterized by progressive muscle wasting, weakness,
mobility loss, and premature death, resulting from mutations of the
dystrophin gene. A commonly used preclinically model of DMD is the
mdx mouse, which is deficient in dystrophin. Although the pathological
severity of this model is milder than human DMD, by aging these mice (6
to 8 month-old) and stressing them with bouts of eccentric contractions
(1 or more), its translatability to humans can be enhanced.59 In a 5-point
dose–response evaluation of 1y in 8 month-old male mdx mice, animals
(n = 7–8/group) received an initial oral dose of 0 (vehicle), 0.1, 0.3, 1, 3,
or 10 mg/kg 1y 10 min prior to damage-inducing eccentric contractions,
and then daily (QD) thereafter for 43 days. Compared to vehicle alone,
1y, at 1, 3, and 10 mg/kg, significantly improved functional recovery,
following eccentric contraction-induced muscle injury in mdx mice as
shown in Fig. 7. Whereas vehicle-treated mdx mice exhibited a ~ 25%
loss of limb function (compared to pre-injury values) at study end, 1y
treated mice displayed near-complete (~90% to 100%) restoration of
pre-injury function. An integrated measure of limb recovery over time
(AUC) demonstrated significant reductions in total recovery time by 1y
treatment at doses ≥ 1.0 mg/kg and was used to calculate an ED50 value
of 1.1 mg/kg. The AUC0-last associated with the ED50 was estimated to be
1100 hr•ng/mL with an estimated CMax of 260 ng/mL, an estimated free
AUC0-last of 200 hr•ng/mL, and a free CMax of 49 ng/mL. Free blood
concentrations reached or exceeded the cellular IC50 at doses ≥ 0.3 mg/
kg, while maximal efficacy occurred at 10 mg/kg. Steady state PGD2
levels in gastrocnemius muscle were suppressed relative to vehicle
treated animals, at all doses, and generally decreased with dose. More-
over, no seizure symptoms were observed in any of the mouse models
upon dosing of H-PGDS inhibitor 1y.
Fig. 7. H-PGDS inhibition attenuates muscle function loss and enhances re-
covery after eccentric muscle damage in mdx mice. Effect of different doses of
an H-PGDS inhibitor, 1y, on limb force following eccentric (lengthening)
contraction-induced muscle injury in mdx mice. Doses were administered 10
min prior to damage and QD thereafter.
sufficient inhibitory exposures over many days/weeks for the resolution
of muscle injury. The higher doses necessary for efficacy in the chronic
eccentric contraction-induced muscle injury model might also reflect the
lower muscle to plasma exposure of compound 1y (muscle:plasma =
0.56) in mice.
3.8. 1,8-Naphthyridine 1y safety assessment
To gain a better understanding of the development potential of 1y, it
was evaluated in a battery of non-clinical safety studies. H-PGDS in-
hibitor 1y was negative in genetic toxicity in vitro screening tests; the
bacterial reverse mutation assay (Ames test), both with and without
metabolic activation and the mouse lymphoma assay. These results
suggest that 1y is unlikely to be carcinogenic as a result of drug-induced
gene mutations or chromosomal aberrations.
The diversity of these multiple inflammatory challenges models is
reflected (to an extent) in the distinct 1y treatment responses. MCD and
LPS experiments are acute in nature and likely reflect CMax-driven
response profiles. In contrast, eccentric contraction-induced muscle
injury is comprised of an acute phase of alarm/inflammation and
cellular debris removal, followed by a chronic phase of myofibrillar
remodeling and gross remodeling, depending on multiple, temporally-
distinct interactions of the immune-, vascular- and neuromuscular sys-
tems. Compound 1y efficacies in the more physiological and less
contrived chronic model are likely driven by AUC values and require
After an oral pharmacokinetic rat study, demonstrated that a single
300 mg/kg dose of 1y was well tolerated, 1y was evaluated in a male
Wistar Han rat oral, once daily, seven day repeat dose range finding
study at dose levels of 10, 30, and 100 mg/kg/day in 0.5% hydrox-
ypropylmethylcellulose with 0.1% Tween™-80 (N = 4/dose). Systemic
exposure (mean AUC and CMax values) increased dose proportionally
with increasing dose, and there was no marked change in systemic
exposure after repeat dosing. Inhibitor 1y was well tolerated at doses of
30 mg/kg/day, with no abnormal microscopic findings. Administration
of 100 mg/kg/day was not tolerated, with rats found dead or terminated
on day 7 prior to the scheduled necropsy. Rats given 100 mg/kg/day had
microscopic findings in the heart (necrosis/inflammatory cell in-
filtrates), kidney (proximal tubular vacuolation and hyaline cast for-
mation), adrenal gland (cortical hypertrophy) and thymus (increased
lymphocytolysis). Day 7 AUC values at 10, 30, and 100 mg/kg/day were
120, 410, and 820 µg•hr/mL, respectively; respective CMax values were
8.7, 24, and 57 μg/mL. Notably, there was no seizure-like activity or
other CNS signs noted in this study, thus limiting brain exposure pre-
vented manifestation of any off-target, adverse CNS events.
Furthermore, H-PGDS inhibitor 1y was evaluated in a 4 day dog oral,
once daily, dose range finding study at doses of 10, 30, and 75 mg/kg/
day (n = 1/sex). Compound 1y AUC and CMax increased approximately
proportionally on Day 1, but both AUC and CMax values increased
greater than proportionally on day 4, suggesting potential accumulation
of 1y. Inhibitor 1y was well tolerated at dose levels up to 30 mg/kg/day
with no abnormal microscopic findings; however, daily dosing of 75
mg/kg/day was not tolerated for the duration of the study based on
deterioration in clinical condition/decreased body temperature and
body weight loss. Macroscopically, dogs given 75 mg/kg/day had
Fig. 6. H-PGDS inhibition attenuates muscle function loss and enhances re-
covery after eccentric muscle damage in wild-type mice. Effect of different
doses of an H-PGDS inhibitor, 1y, on limb force following eccentric (length-
ening) contraction-induced muscle injury in normal C57Bl6/N mice. Doses
were administered 10 min prior to damage and QD thereafter.
14

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
discoloration in the small intestine and esophagus (female), which
correlated microscopically with diffuse degeneration/regeneration or
multifocal erosion/ulceration. Additional microscopic findings were
identified in the adrenal glands (degeneration of the zona glomerulosa,
cytoplasmic alteration in the zona fasciculata, and hemorrhage of the
zona fasciculata), and thymus (decreased cortical lymphoid cellularity
consistent with stress). Clinical pathology changes included increased
red cell mass due to hemoconcentration and increases in hepatobiliary
parameters and cardiac troponin I concentrations. Day 4 AUC values in
males and females given 10, 30, or 75 mg/kg/day were 120, 430, and
3600 µg•hr/mL, respectively; respective CMax values were 13, 34, and
(dimethylamino)-methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium
3-
oxid hexafluorophosphate (HATU), azobis(isobutyronitrile) (AIBN), 4-
(2-hydroxyethyl)-piperazine-1-ethanesulfonic acid (HEPES), and
dithiothreitol (DTT). All reactions except those in aqueous media were
carried out with the use of standard techniques for the exclusion of
moisture. Reactions were monitored by thin-layer chromatography
(TLC) on 0.25 mm silica gel plates (60F-254, E. Merck) and visualized
with UV light, iodine, iodoplatinate, potassium permanganate, cerium
molybdate, or 5% phosphomolybdic acid in 95% ethanol. Final com-
pounds were typically purified either by flash chromatography on silica
gel (E. Merck 40–63 mm), radial chromatography on a Chromatotron™
using prepared silica gel plates, or on a Biotage Horizon or ISCO Com-
biflash® pump and fraction collection system utilizing prepacked silica
gel. Analytical purity was assessed either by reversed-phase high per-
formance liquid chromatography (RP-HPLC) using an Agilent 1100
system equipped with a diode array spectrometer (λ range 190–400 nm)
or by the LC-MS method detailed below. The stationary phase was a
210 μg/mL.
Although there was a margin of safety between the efficacious dose
(ED50, 3 mg/kg) and the non-tolerated dose (75 mg/kg), other H-PGDS
inhibitors were discovered with improved safety profiles; therefore, 1y
was not advanced into GLP safety studies.
Keystone Scientific BDS Hypersil™ C-18 column (5 μm, 4.6 mm × 200
4. Conclusions
mm). The mobile phase employed 0.1% aqueous TFA with MeCN as the
organic modifier and a flow rate of 1.0 mL/min. Analytical data are
reported as retention time (tR) in minutes and percent purity. All com-
Desiring to explore the selective alteration of prostaglandin signaling
for human disease intervention, GSK, in collaboration with Astex, used a
fragment-based approach to discover H-PGDS inhibitors. These efforts
resulted in the discovery of quinoline H-PGDS inhibitors, including 1a.
Quinoline 1a is a potent H-PGDS inhibitor with good murine pharma-
cokinetic parameters. It was used to explore PGD2 biology and attenu-
ated mast cell degranulation in an acute inflammatory model, helping
validate H-PGDS as a viable target for inflammatory diseases. Unfortu-
nately, quinoline 1a on repeat dosing at high concentrations caused CNS
toxicity, precluding its further development. As the adverse effects were
deemed to be off-target, a strategy was devised to limit CNS exposure in
future H-PGDS inhibitors, and 1a was utilized as a starting point to
discover safer H-PGDS inhibitors. Tactically, a plan to limit brain
exposure by increasing the PSA of inhibitors was devised. Specifically,
PSA was increased by designing aza-quinolines. The preparation and
testing of the aza-quinolines revealed that 1,6- and 1,8-naphthyridine
inhibitors, such as 1i and 1k, were potent H-PGDS inhibitors. Further
modifications to the 6- and 7- ring positions, as well as carboxamide
amine substitutions, to improve H-PGDS inhibition and/or pharmaco-
kinetic parameters resulted in the discovery of 1,8-naphthyridine 1y.
1,8-Naphthyridine 1y is a potent H-PGDS inhibitor with good selectivity
versus other proteins, good pharmacokinetic parameters in mouse, rat,
and dog, and minimal CNS exposure. Moreover, compound 1y attenu-
ated inflammatory exacerbations in two acute rodent models of
inflammation and preserved muscle function in two chronic models of
muscle injury. Finally, in rat and dog dose range finding safety studies,
compound 1y did not cause any CNS toxicity at doses significantly
higher that its efficacious range, validating the strategy of limiting brain
penetration to avoid adverse CNS effects. Thus, 1,8-naphthyridine 1y is
an excellent tool to explore PGD2 biology and the potential therapeutic
possibilities of H-PGDS inhibition.
1
pounds were found to be ≥ 95% pure unless otherwise indicated. H
NMR spectra were recorded on either a Varian Unityplus™ 400 MHz or a
Bruker Avance™ III 400 MHz NMR spectrometer. Chemical shifts are
reported in parts per million (ppm, δ units). Coupling constants are re-
ported in units of hertz (Hz). Splitting patterns are designated as s,
singlet; d, doublet; t, triplet; q, quartet; p, pentet; h, hextet; m, multiplet;
or br, broad. Low-resolution mass spectra (MS) were recorded on a
Waters SQD. High-resolution MS were recorded on a Waters (Micro-
mass®) LCT time-of-flight mass spectrometer. Low-resolution mass
spectra were obtained under electrospray ionization (ESI), atmospheric
pressure chemical ionization (APCI), or fast atom bombardment (FAB)
methods. All studies were conducted in accordance with the GSK Policy
on the Care, Welfare and Treatment of Laboratory Animals and were
reviewed the Institutional Animal Care and Use Committee either at GSK
or by the ethical review process at the institution where the work was
performed.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-6-methoxy-2-naph-
thamide 1d
N,N-Diisopropylethylamine (0.095 mL, 0.544 mmol) was added to a
solution of 6-methoxy-2-naphthoic acid 7a (50 mg, 0.247 mmol) in N,N-
dimethylformamide (3 mL), followed by 1-[bis(dimethylamino)methy-
lene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate
(103 mg, 0.272 mmol) in portions. The reaction mixture was stirred for
~ 3 min and then 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (42.8 mg,
0.272 mmol, PharmaBlock) was added. After 40 min, LC-MS showed ~
50% product. Additional 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(15.6 mg, 0.099 mmol) was added, and the reaction mixture was stirred
for another 45 min. The reaction mixture is diluted with ethyl acetate
and washed with saturated potassium carbonate solution (2X), 0.1 M
hydrochloric acid (2X), and saturated sodium chloride (1X), dried over
sodium sulfate, filtered, and concentrated. The residue was purified by
silica gel chromatography, eluting with ethyl acetate:hexanes (0:1 to
7:3), then crystalized from dichloromethane:hexanes to give N-(trans-4-
(2-hydroxypropan-2-yl)cyclohexyl)-6-methoxy-2-naphthamide 1d (58
mg, 0.16 mol, 65% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H),
1.06–1.24 (m, 3H), 1.33 (q, J = 10 Hz, 2H), 1.84 (br d, J = 12 Hz, 2H),
1.92 (br d, J = 10 Hz, 2H), 3.68–3.80 (m, 1H), 3.89 (s, 3H), 4.03 (s, 1H),
5. Experimental section
All commercial chemicals and solvents were reagent grade and were
used without further purification unless otherwise specified. The
following abbreviations are utilized in the manuscript: tetrahydrofuran
(THF), diethyl ether (Et2O), dimethyl sulfoxide (DMSO), ethyl acetate
(EtOAc), dichloromethane (CH2Cl2), trifluoroacetic acid (TFA), N,N-
dimethylformamide (DMF), methanol (MeOH), dimethoxyethane
(DME), N-methylpyrrolidine (NMP), acetonitrile (MeCN), chloroform
(CHCl3), phosphorus oxychloride (POCl3), magnesium sulfate (MgSO4),
triethylamine (Et3N), 2-propanol (iPrOH), N,N-diisopropylethylamine
(iPr2NEt), sodium hydroxide (NaOH), t-butylmethyl ether (TBME),
acetic acid (AcOH), ethanol (EtOH), di-tert-butyldicarbonate (BOC2O),
sodium sulfate (Na2SO4), N,N-dimethylacetamide (DMA), sodium bi-
carbonate (NaHCO3), potassium carbonate (K2CO3), 1-[bis
15

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
7.21 (dd, J = 9, 2 Hz, 1H), 7.36 (d, J = 2 Hz, 1H), 7.85 (d, J = 9 Hz, 1H),
7.87 (d, J = 2 Hz, 1H), 7.91 (d, J = 9 Hz, 1H), 8.24 (d, J = 8 Hz, 1H), 9.34
(s, 1H); LC-MS (LC-ES) C21H27NO3 M + H = 342; tR = 0.75 min, 100%
purity.
N,N-Diisopropylethylamine (0.238 mL, 1.363 mmol) was added to 4-
nitrophenyl 4-methoxy-2-nitrobenzylcarbamate 11 (0.1578 g, 0.454
mmol) in tetrahydrofuran (4.54 mL) at room temperature. Then, 2-
(trans-4-aminocyclohexyl)propan-2-ol 8a (0.093 g, 0.591 mmol, Phar-
maBlock) was added and the reaction mixture was stirred for sixteen
hours. 10% Citric acid was added to the reaction mixture and it was
extracted with diethyl ether, washed with saturated sodium bicarbon-
ate, dried over magnesium sulfate, filtered, and concentrated. The re-
action mixture was purified by silica gel chromatography, eluting with
methanol:ethyl acetate (0:1 to 1:19) to give 1-(trans-4-(2-hydrox-
ypropan-2-yl)cyclohexyl)-3-(4-methoxy-2-nitrobenzyl)urea 12 (0.1373
g, 0.357 mmol, 79% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.00 (s,
6H), 0.94–1.16 (m, 5H), 1.70–1.78 (m, 2H), 1.78–1.86 (m, 2H),
3.14–3.28 (m, 1H), 3.82 (s, 3H), 3.98 (br s, 1H), 4.34 (d, J = 6 Hz, 2H),
5.91 (d, J = 8 Hz, 1H), 6.20 (br t, J = 6 Hz, 1H), 7.30 (dd, J = 9, 3 Hz,
1H), 7.43 (d, J = 9 Hz, 1H), 7.50 (d, J = 3 Hz, 1H); LC-MS (LC-ES) M + H
= 366.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-methoxycinnoline-
3-carboxamide 1e
N,N-Diisopropylethylamine (0.474 mL, 2.72 mmol) was added to 7-
methoxycinnoline-3-carboxylic acid 7b (0.0924 g, 0.453 mmol) in 1,4-
dioxane (2.263 mL) at room temperature. Then, 2-(trans-4-amino-
cyclohexyl)propan-2-ol 8a (0.078 g, 0.498 mmol, PharmaBlock) was
added and the reaction mixture was stirred for five minutes. Then, n-
propylphosphonic acid anhydride (0.539 mL, 0.905 mmol) was added
and the reaction mixture was stirred for sixty-four hours. The reaction
mixture was poured into saturated sodium bicarbonate, extracted with
ethyl acetate (3X), dried over magnesium sulfate, filtered, and concen-
trated. The residue was purified by silica gel chromatography, eluting
with ethyl acetate to give N-(trans-4-(2-hydroxypropan-2-yl)cyclo-
hexyl)-7-methoxycinnoline-3-carboxamide (0.0677 g, 0.187 mmol,
41.4% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.05 (s, 6H), 1.06–1.26
(m, 3H), 1.50 (q, J = 12 Hz, 2H), 1.84 (br d, J = 12 Hz, 2H), 1.91 (br d, J
= 10 Hz, 2H), 3.86 (qt, J = 8, 4 Hz, 1H), 4.04 (s, 3H), 4.04 (s, 1H), 7.59
(dd, J = 9, 3 Hz, 1H), 7.87 (d, J = 2 Hz, 1H), 8.20 (d, J = 9 Hz, 1H), 8.72
(s, 1H), 9.02 (d, J = 8 Hz, 1H); LC-MS (LC-ES) C19H25N3O3 M + H = 344;
tR = 0.69 min, 100% purity.
1-(2-Amino-4-methoxybenzyl)-3-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)urea 13
Palladium on carbon (3.75 mg, 0.035 mmol) was added to 1-(trans-4-
(2-hydroxypropan-2-yl)cyclohexyl)-3-(4-methoxy-2-nitrobenzyl)urea
12 (0.1287 g, 0.352 mmol) in methanol (7.04 mL) at room temperature
under nitrogen atmosphere. Then, the reaction vessel was fitted with a
hydrogen balloon and the vessel was repeatedly evacuated and purged
with hydrogen, then stirred for two hours. Then, the vessel was
repeatedly evacuated and purged with nitrogen, filtered through Cel-
ite®, and concentrated. The residue was purified by silica gel chroma-
tography, eluting with methanol:ethyl acetate (1:19) to give 1-(2-amino-
4-methoxybenzyl)-3-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)urea
13 (0.1373 g, 0.389 mmol, 110% yield). 1H NMR (400 MHz, CD3SOCD3)
δ 1.00 (s, 6H), 0.92–1.16 (m, 5H), 1.70–1.78 (m, 2H), 1.78–1.86 (m,
2H), 3.14–3.28 (m, 1H), 3.62 (s, 3H), 3.96 (d, J = 6 Hz, 2H), 3.98 (s, 1H),
5.21 (s, 2H), 5.70 (d, J = 8 Hz, 1H), 5.93 (br t, J = 6 Hz, 1H), 6.03 (dd, J
= 8, 3 Hz, 1H), 6.14 (d, J = 3 Hz, 1H), 6.82 (d, J = 8 Hz, 1H); LC-MS (LC-
ES) M + H = 336.
4-Nitrophenyl 4-methoxy-2-nitrobenzylcarbamate 11
N,N-Diisopropylethylamine (1.220 mL, 6.99 mmol) was added to (4-
methoxy-2-nitrophenyl)methanamine hydrochloride 9 (0.5092 g, 2.329
mmol) in tetrahydrofuran (11.64 mL) at room temperature. Then, 4-
nitrophenyl chloroformate 10 (0.516 g, 2.56 mmol) was added and
the reaction mixture was stirred for sixteen hours. 10% Citric acid was
added to the reaction mixture and it was extracted with diethyl ether,
washed with saturated sodium bicarbonate, dried over magnesium sul-
fate, filtered, and concentrated. The reaction mixture was purified by
silica gel chromatography, eluting with ethyl acetate:hexanes (3:7) to
give 4-nitrophenyl 4-methoxy-2-nitrobenzylcarbamate 11 (0.7470 g,
2.043 mmol, 88% yield). 1H NMR (400 MHz, CD3SOCD3) δ 3.85 (s, 3H),
4.52 (d, J = 6 Hz, 2H), 7.36 (dd, J = 9, 3 Hz, 1H), 7.41 (d, J = 9 Hz, 2H),
7.56 (d, J = 6 Hz, 1H), 7.57 (s, 1H), 8.26 (d, J = 9 Hz, 2H), 8.60 (br t, J =
6 Hz, 1H); LC-MS (LC-ES) M + H = 348.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-methoxyquinazo-
line-3(4H)-carboxamide 1f
1-(2-Amino-4-methoxybenzyl)-3-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)urea 13 (0.0531 g, 0.158 mmol) was taken up in trimethyl
orthoformate (1.732 mL, 15.83 mmol) at room temperature under ni-
trogen atmosphere and heated at 102 ◦C for two hours. Then, the vessel
was cooled and concentrated. The residue was purified by silica gel
chromatography, eluting with methanol:ethyl acetate (1:19) to give N-
(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-7-methoxyquinazoline-3
(4H)-carboxamide 1f (0.0178 g, 0.049 mmol, 30.9% yield). 1H NMR
(400 MHz, CD3SOCD3) δ 1.02 (s, 6H), 1.05 (q, J = 10 Hz, 2H), 1.08–1.20
(m, 1H), 1.24 (q, J = 12 Hz, 2H), 1.80 (br d, J = 12 Hz, 2H), 1.87 (br d, J
= 10 Hz, 2H), 3.45 (qt, J = 8, 4 Hz, 1H), 3.72 (s, 3H), 4.02 (br s, 1H),
4.63 (s, 2H), 6.63 (d, J = 3 Hz, 1H), 6.70 (dd, J = 8, 3 Hz, 1H), 6.97 (d, J
= 8 Hz, 1H), 7.21 (d, J = 8 Hz, 1H), 7.85 (s, 1H); LC-MS (LC-ES)
1-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-3-(4-methoxy-2-
nitrobenzyl)urea 12
C19H27N3O3 M + H = 346; tR = 0.49 min, 96% purity.
16

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
N-(trans)-4-(2-Hydroxypropan-2-yl)cyclohexyl)-6-methoxyquinoxa-
Ethyl 7-methoxy-1,5-naphthyridine-3-carboxylate 6a
line-2-carboxamide 1g
To a nitrogen degassed solution of ethyl 4-chloro-7-methoxy-1,5-
naphthyridine-3-carboxylate 5 (250 mg, 0.937 mmol) in acetonitrile
(10 mL) was added bis(triphenylphosphine)palladium(II) chloride
(65.8 mg, 0.094 mmol) followed by triethylsilane (0.299 mL, 1.875
mmol). After heating 3 h at 70 ◦C, the reaction mixture was adsorbed to
silica gel and purified by silica gel chromatography, eluting with
methanol:dichloromethane (0:1 to 1:19) to give ethyl 7-methoxy-1,5-
naphthyridine-3-carboxylate 6a (100 mg, 46% yield). LC-MS (LC-ES)
M + H = 233.
N,N-Diisopropylethylamine (0.066 mL, 0.377 mmol) was added to a
solution of 6-methoxyquinoxaline-2-carboxylic acid60 7c (35 mg, 0.171
mmol) in N,N-dimethylformamide (3 mL), followed by 1-[bis(dimethy-
lamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexa-
fluorophosphate (71.7 mg, 0.189 mmol). After stirring for ~ 2 min, 2-
(trans-4-aminocyclohexyl)propan-2-ol 8a (29.7 mg, 0.189 mmol, Phar-
maBlock) was added in one portion. After 3 h, the reaction mixture is
diluted with ethyl acetate and washed with diluted aqueous hydro-
chloric acid (1X), saturated potassium carbonate solution (2X), and
saturated aqueous sodium chloride (1X). The organic phase was dried
over sodium sulfate, filtered and concentrated. The residue was chro-
matographed over silica gel, eluting with ethyl acetate:hexanes (1:4 to
4:1), then repurified by silica gel chromatography, eluting with ethyl
acetate:hexanes (1:4 to 4:1). The resulting material was dissolved in
ethyl acetate with a little methanol and washed with water (4X) and
saturated sodium chloride, dried over sodium sulfate, filtered, concen-
trated, and crystalized from dichloromethane:hexanes to give N-(trans)-
4-(2-hydroxypropan-2-yl)cyclohexyl)-6-methoxyquinoxaline-2-carbox-
amide 1g (34 mg, 0.094 mmol, 55% yield) as a yellow solid. 1H NMR
(400 MHz, CD3SOCD3) δ 1.05 (s, 6H), 1.06–1.26 (m, 3H), 1.45 (q, J = 12
Hz, 2H), 1.85 (br d, J = 16 Hz, 2H), 1.90 (br d, J = 16 Hz, 2H), 3.72–3.86
(m, 1H), 4.00 (s, 3H), 4.06 (br s, 1H), 7.54 (d, J = 2 Hz, 1H), 7.61 (dd, J
= 9, 2 Hz, 1H), 8.11 (d, J = 9 Hz, 1H), 8.60 (d, J = 8 Hz, 1H), 9.37 (s,
1H); LC-MS (LC-ES) C19H25N3O3 M + H = 344; tR = 0.71 min, 100%
purity.
7-Methoxy-1,5-naphthyridine-3-carboxylic acid 7d
A solution of ethyl 7-methoxy-1,5-naphthyridine-3-carboxylate 6a
(50 mg, 0.215 mmol) in tetrahydrofuran (5 mL), and ethanol (5.00 mL)
was treated with a solution of lithium hydroxide (25.8 mg, 1.076 mmol)
in water (5 mL). The reaction mixture was heated at 50 ◦C for 45 min.
The pH of the reaction was adjusted to ~ 6 and the solvent was removed
under vacuum to give crude 7-methoxy-1,5-naphthyridine-3-carboxylic
acid 7d (44 mg, 100% yield). LC-MS (LC-ES) M + H = 205.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-methoxy-1,5-
naphthyridine-3-carboxamide 1h
Diethyl 2-(((5-methoxypyridin-3-yl)amino)methylene)malonate 4
N,N-Diisopropylethylamine (0.171 mL, 0.980 mmol) was added to a
solution of 7-methoxy-1,5-naphthyridine-3-carboxylic acid 7d (50 mg,
0.245 mmol) in N,N-dimethylformamide (10 mL) at room temperature,
followed by 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (38.5 mg,
0.245 mmol, PharmaBlock). Next, 2,4,6-tripropyl-1,3,5,2,4,6-trioxatri-
phosphinane 2,4,6-trioxide (50% in ethyl acetate) (312 mg, 0.490
mmol) was added by a slow dropwise addition and the reaction was
allowed to stir at room temperature overnight. Saturated sodium bi-
carbonate solution was added to adjust the pH to ~ 8. The reaction
mixture was repeatedly extracted with ethyl acetate until no product
was in the organic layer as observed by thin layer chromatography (1:9
methanol:dichloromethane). The combined organic extracts were
concentrated, adsorbed to silica gel, and purified by silica gel chroma-
tography, eluting with methanol:dichloromethane (0:1 to 1:9) to give N-
(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-7-methoxy-1,5-naphthyr-
idine-3-carboxamide 1h (35 mg, 0.102 mmol, 42% yield). 1H NMR (400
MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.34 (q, J = 12 Hz,
2H), 1.85 (br d, J = 12 Hz, 2H), 1.94 (br d, J = 10 Hz, 2H), 3.72–3.82 (m,
1H), 4.00 (s, 3H), 4.04 (s, 1H), 7.83 (d, J = 3 Hz, 1H), 8.61 (br d, J = 8
Hz, 1H), 8.79 (d, J = 2 Hz, 1H), 8.82 (d, J = 3 Hz, 1H), 9.29 (d, J = 2 Hz,
1H); LC-MS (LC-ES) C19H25N3O3 M + H = 344; tR = 0.59 min, 100%
purity.
To a solution of 5-methoxypyridin-3-amine 2 (3 g, 24.17 mmol) in
ethanol (150 mL) was added diethyl 2-(ethoxymethylene)malonate 3
(5.49 g, 25.4 mmol) and the reaction was heated at 80 ◦C for 24 h. The
reaction mixture was concentrated under reduced pressure and the solid
was placed under vacuum overnight to give diethyl 2-(((5-methox-
ypyridin-3-yl)amino)methylene)malonate 4 (6.0 g, 84% yield). LC-MS
(LC-ES) M + H = 295.
Ethyl 4-chloro-7-methoxy-1,5-naphthyridine-3-carboxylate 5
Diethyl 2-(((5-methoxypyridin-3-yl)amino)methylene)malonate 4
(500 mg, 1.699 mmol) was added to phosphorus oxychloride (3167 µL,
34.0 mmol). The reaction mixture was heated at 160 ◦C in a microwave
for 3 h. The solution was diluted with ethyl acetate (150 mL), cooled to
0 ◦C in an ice bath, and pH adjusted to ~ 7 with saturated sodium bi-
carbonate. The organic layer was separated, the aqueous layer was
reextracted with ethyl acetate (2X, 100 mL), and the combined organic
layers were adsorbed to silica gel and purified by silica gel chromatog-
raphy, eluting with ethyl acetate:hexanes (1:19 to 2:3) to give ethyl 4-
chloro-7-methoxy-1,5-naphthyridine-3-carboxylate 5 (250 mg, 55%
yield). LC-MS (LC-ES) M + H = 267.
3-Bromo-7-chloro-1,6-naphthyridine 16a
17

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
Ytterbium(III) trifluoromethanesulfonate (43.1 g, 69.5 mmol) was
mmol) was added to 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (0.079
g, 0.502 mmol, PharmaBlock) in N,N-dimethylformamide (5 mL) at
room temperature. Then, N,N-diisopropylethylamine (0.22 mL, 1.263
mmol) was added, followed by 1-[bis(dimethylamino)methylene]-1H-
1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate (0.232 g,
0.610 mmol) and the reaction mixture was stirred for 150 min. The
reaction mixture was concentrated and the resulting residue was puri-
fied by silica gel chromatography, eluting with (ethyl acetate:ethanol
(3:1)):hexanes (0:1 to 1:0) to give a residue which was triturated/soni-
cated with ethyl acetate and filtered to give N-(trans-4-(2-hydrox-
ypropan-2-yl)cyclohexyl)-7-methoxy-1,6-naphthyridine-3-carboxamide
1i (0.123 g, 0.358 mmol, 71.3% yield) as a pale yellow powder. 1H NMR
(400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 3H), 1.31 (q, J = 10
Hz, 2H), 1.84 (br d, J = 12 Hz, 2H), 1.94 (br d, J = 10 Hz, 2H), 3.68–3.80
(m, 1H), 4.01 (s, 3H), 4.06 (s, 1H), 7.27 (s, 1H), 8.58 (d, J = 8 Hz, 1H),
8.90 (dd, J = 2, 1 Hz, 1H), 9.25 (d, J = 1 Hz, 1H), 9.34 (d, J = 2 Hz, 1H);
LC-MS (LC-ES) C19H25N3O3 M + H = 344; tR = 0.68 min, 100% purity.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-1,7-naphthyridine-3-
carboxamide 1j
added to a solution of 4-amino-6-chloronicotinaldehyde 14 (50 g, 278
mmol, AstaTech) in acetonitrile (500 mL) at 0 ◦C under argon. Then, 2-
bromo-1,1-dimethoxyethane 15 (98 mL, 833 mmol) was added and the
◦
resulting reaction mixture was heated to 80 C and stirred for sixteen
hours. On completion, the reaction mixture was filtered through a Cel-
ite® pad, the residue was washed with acetonitrile (500 mL) and the
filtrate was evaporated under reduced pressure to give an impure ma-
terial, which was purified via silica gel column chromatography, eluting
with ethyl acetate:hexanes (1:4) to afford 3-bromo-7-chloro-1,6-naph-
thyridine 16a (30 g, 98 mmol, 35% yield) as light yellow color solid.
1H NMR (400 MHz, CDCl3) δ 7.99 (s, 1H), 8.43 (m, 1H), 9.05 (s, 1H),
9.09 (d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H = 243.
Ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a
Triethylamine (17.69 mL, 127 mmol) was added to a solution of 3-
bromo-7-chloro-1,6-naphthyridine61 16a (20 g, 63.4 mmol) in ethanol
(400 mL) in an autoclave under carbon monoxide atmosphere (80 psi) at
room temperature. Then, [1,1′-bis(diphenylphosphino)ferrocene]
dichloropalladium(II) dichloromethane adduct (5.18 g, 6.34 mmol) was
added and the reaction mixture was heated to 80 ◦C and stirred for 90
min. On completion, the reaction mixture was filtered through a Celite®
pad, the residue was washed with ethanol (200 mL), and the solvent
evaporated under reduced pressure to give an impure material, which
was purified via silica gel column chromatography, eluting with ethyl
acetate:hexanes (1:9) to afford ethyl 7-chloro-1,6-naphthyridine-3-
carboxylate 17a (10.7 g, 38.6 mmol, 61% yield) as an off white solid.
1H NMR (400 MHz, CDCl3) δ 1.48 (t, J = 7 Hz, 3H), 4.51 (q, J = 7 Hz,
2H), 8.06 (s, 1H), 8.88–9.03 (m, 1H), 9.21 (s, 1H), 9.62 (d, J = 2 Hz, 1H);
LC-MS (LC-ES) M + H = 237.
N,N-Diisopropylethylamine (0.604 mL, 3.46 mmol) was added to
1,7-naphthyridine-3-carboxylic acid 7f (0.1003 g, 0.576 mmol,
Enamine) in 1,4-dioxane (2.88 mL) at room temperature. Then, 2-(trans-
4-aminocyclohexyl)propan-2-ol 8a (0.109 g, 0.691 mmol, Pharma-
Block) was added and the reaction mixture was stirred for five minutes.
Then, n-propylphosphonic acid anhydride (0.686 mL, 1.152 mmol) was
added and the reaction mixture was stirred for sixteen hours. The re-
action mixture was poured into saturated sodium bicarbonate, extracted
with ethyl acetate (3X), dried over magnesium sulfate, filtered, and
concentrated. The residue was purified by silica gel chromatography,
eluting with methanol:ethyl acetate (1:8) to give N-(trans-4-(2-hydrox-
ypropan-2-yl)cyclohexyl)-1,7-naphthyridine-3-carboxamide 1j (0.0552
g, 0.167 mmol, 29.1% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.05 (s,
6H), 1.06–1.24 (m, 3H), 1.32 (q, J = 12 Hz, 2H), 1.84 (br d, J = 12 Hz,
2H), 1.96 (br d, J = 10 Hz, 2H), 3.76 (qt, J = 8, 4 Hz, 1H), 4.05 (s, 1H),
8.02 (d, J = 6 Hz, 1H), 8.69 (d, J = 6 Hz, 1H), 8.71 (d, J = 8 Hz, 1H), 8.83
(d, J = 2 Hz, 1H), 9.36 (d, J = 2 Hz, 1H), 9.46 (s, 1H); LC-MS (LC-ES)
7-Methoxy-1,6-naphthyridine-3-carboxylic acid 7e
Sodium methoxide (25% in methanol, 1.15 mL, 5.03 mmol) was
added to ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a (0.238 g,
1.006 mmol) in methanol (10 mL) at room temperature and the reaction
mixture was heated to 60 ◦C for seven hours. The reaction mixture was
concentrated, then water (10 mL) was added and the reaction was
stirred for 90 min. The reaction mixture was filtered through a pad of
Celite® and the filter cake rinsed with water. The pH was adjusted to ~ 5
with 1 N hydrochloric acid (4 mL). A fine, milky precipitate formed. An
additional 1 N hydrochloric acid (1 mL) was added (pH = ~2). The
solids were filtered off and rinsed with water (2X), air-dried, and then
dried under vacuum to give 7-methoxy-1,6-naphthyridine-3-carboxylic
acid 7e (0.206 g, 1.009 mmol, 100% yield) as a pale yellow powder.
1H NMR (400 MHz, CD3SOCD3) δ 4.02 (s, 3H), 7.29 (s, 1H), 9.07 (d, J =
2 Hz, 1H), 9.35 (s, 1H), 9.36 (d, J = 2 Hz, 1H), 13.54 (br s, 1H); LC-MS
(LC-ES) M + H = 205.
C18H23N3O2 M + H = 314; tR = 0.50 min, 100% purity.
6-Bromo-1,8-naphthyridin-2-amine 16b
Phosphoric acid (60 mL, 183 mmol) was added to a mixture of
pyridine-2,6-diamine 18 (20 g, 183 mmol) and 2-bromomalonaldehyde
◦
19 (27.7 g, 183 mmol) at 0 C under nitrogen. The resulting reaction
mixture was heated to 120 ◦C and stirred for sixteen hours. On
completion, the reaction mixture was quenched with 2 M aqueous so-
dium hydroxide solution (150 mL). The precipitate was filtered, washed
with water (1000 mL), and dried to give an impure material. This ma-
terial was purified via neutral alumina column chromatography, eluting
with methanol:dichloromethane (1:9) to afford 6-bromo-1,8-naphthyri-
din-2-amine 16b (20 g, 69.4 mmol, 38% yield) as a yellow solid. 1H
NMR (400 MHz, CD3SOCD3) δ 6.85 (d, J = 9 Hz, 1H), 6.98 (br s, 2H),
7.90 (d, J = 9 Hz, 1H), 8.32 (s, 1H), 8.68 (s, 1H); LC-MS (LC-ES) M + H
= 224.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-methoxy-1,6-
naphthyridine-3-carboxamide 1i
Ethyl 7-amino-1,8-naphthyridine-3-carboxylate 17b
7-Methoxy-1,6-naphthyridine-3-carboxylic acid 7e (0.110 g, 0.539
18

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
stirred for 75 min, then the reaction mixture was filtered through a pad
of Celite® and the filter cake rinsed with water. The pH was adjusted to
~ 4–5 with 1 N hydrochloric acid (4 mL). A fine, milky precipitate
formed. An additional 1 N hydrochloric acid (1 mL) was added (pH =
~2), and the solids were filtered off and rinsed twice with water, air
dried, then dried under vacuum to give 7-methoxy-1,8-naphthyridine-3-
carboxylic acid 7g (0.202 g, 0.989 mmol, 98% yield) as a tan powder. 1H
NMR (400 MHz, CD3SOCD3) δ 4.05 (s, 3H), 7.23 (d, J = 9 Hz, 1H), 8.50
(d, J = 9 Hz, 1H), 8.95 (d, J = 2 Hz, 1H), 9.31 (d, J = 2 Hz, 1H), 13.46 (br
s, 1H); LC-MS (LC-ES) M + H = 205.
Triethylamine (19.33 mL, 139 mmol) was added to a solution of 6-
bromo-1,8-naphthyridin-2-amine62 16b (20 g, 69.4 mmol) in ethanol
(200 mL) in an autoclave under carbon monoxide atmosphere (100 psi).
Then,
[1,1′-bis(diphenylphosphino)ferrocene]dichloropalladium(II)
(5.07 g, 6.94 mmol) was added and the reaction mixture was heated to
100 ◦C and stirred for five hours. On completion, the reaction mixture
was filtered through a Celite® pad that was washed with ethanol (500
mL). The filtrate was evaporated under reduced pressure to give a res-
idue, which was purified via neutral alumina column chromatography,
eluting with methanol:dichloromethane (1:9) to afford ethyl 7-amino-
1,8-naphthyridine-3-carboxylate 17b (10 g, 43.8 mmol, 63% yield) as
an off white solid. 1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz,
3H), 4.36 (q, J = 7 Hz, 2H), 6.89 (d, J = 9 Hz, 1H), 7.29 (s, 2H), 8.09 (d,
J = 9 Hz, 1H), 8.60 (d, J = 2 Hz, 1H), 9.12 (d, J = 2 Hz, 1H); LC-MS (LC-
ES) M + H = 218.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-methoxy-1,8-
naphthyridine-3-carboxamide 1k
7-Methoxy-1,8-naphthyridine-3-carboxylic acid 7g (0.110 g, 0.539
mmol) was added to 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (0.081
g, 0.515 mmol, PharmaBlock) in N,N-dimethylformamide (5 mL) at
room temperature. Then, N,N-diisopropylethylamine (0.23 mL, 1.320
mmol) was added, followed by 1-[bis(dimethylamino)methylene]-1H-
1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate (0.235 g,
0.618 mmol) and the reaction mixture was stirred for 150 min. The
reaction mixture was concentrated and the resulting residue was puri-
fied by silica gel chromatography, eluting with (ethyl acetate:ethanol
(3:1)):hexanes (0:1 to 1:0) to give a residue which was triturated/soni-
cated with ethyl acetate and filtered to give N-(trans-4-(2-hydrox-
ypropan-2-yl)cyclohexyl)-7-methoxy-1,8-naphthyridine-3-carboxamide
1j (0.127 g, 0.370 mmol, 71.8% yield) as an off-white powder. 1H NMR
(400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 3H), 1.31 (q, J = 12
Hz, 2H), 1.83 (br d, J = 12 Hz, 2H), 1.93 (br d, J = 10 Hz, 2H), 3.68–3.80
(m, 1H), 4.03 (s, 3H), 4.06 (s, 1H), 7.19 (d, J = 9 Hz, 1H), 8.40 (d, J = 9
Hz, 1H), 8.54 (d, J = 8 Hz, 1H), 8.78 (d, J = 2 Hz, 1H), 9.27 (d, J = 2 Hz,
1H); LC-MS (LC-ES) C19H25N3O3 M + H = 344; tR = 0.66 min, 100%
purity.
Ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c
Sodium nitrite (1.210 g, 17.53 mmol) was added to ethyl 7-amino-
1,8-naphthyridine-3-carboxylate 17b (2 g, 8.77 mmol) at room tem-
perature. Then, concentrated sulfuric acid (4.77 mL, 88 mmol) was
added dropwise and the reaction mixture was stirred for sixteen hours.
On completion, the reaction mixture was diluted with ice water (50 mL)
and stirred for 10 min. The precipitate was filtered, washed with
pentane (10 mL) and diethyl ether (10 mL), and dried under vacuum to
afford ethyl 7-hydroxy-1,8-naphthyridine-3-carboxylate (1.8 g, 63%
yield) as off white solid (1H NMR (400 MHz, CD3SOCD3) δ 1.22–1.52 (t,
J = 7 Hz, 3H), 4.37 (q, J = 7 Hz, 2H), 6.44–6.79 (m, 1H), 7.86–8.17 (m,
1H), 8.52–8.76 (m, 1H), 8.92–9.09 (m, 1H), 12.39–12.71 (br s, 1H); LC-
MS (LC-ES) M + H = 219). N,N-Diisopropylethylamine (2.3 mL, 13.33
mmol) was added to a solution of ethyl 7-hydroxy-1,8-naphthyridine-3-
Ethyl 2-(methylthio)-7-oxo-7,8-dihydropyrido[2,3-d]pyrimidine-6-
carboxylate 22
◦
carboxylate (1.8 g, 6.67 mmol) in 1,4-dioxane (18 mL) at 0 C. Then,
phosphorus oxychloride (2.48 mL, 26.7 mmol) was added and the
◦
resulting reaction mixture was heated to 80 C and stirred for sixteen
hours. On completion, the reaction mixture was diluted with ice cold
water (50 mL) and extracted with ethyl acetate (2 X 50 mL). The com-
bined organic layers were washed with brine (50 mL), dried over
anhydrous sodium sulfate, filtered, and evaporated under reduced
pressure to give an impure material, which was purified via neutral
alumina column chromatography, eluting with ethyl acetate:petroleum
ether (2:3) to afford ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c
Diethyl malonate 20 (2.40 mL, 15.73 mmol) was added to a stirred
solution of 4-amino-2-(methylthio)pyrimidine-5-carbaldehyde 21 (2.00
g, 11.82 mmol, AstaTech) in N,N-dimethylformamide (40 mL). Then,
potassium carbonate (2.000 g, 14.47 mmol) was added and the reaction
mixture was heated to 85 ◦C and stirred overnight. As starting material
was still present, triethylamine (1.00 mL, 7.17 mmol) was added to the
mixture with stirring continued overnight. Additional triethylamine
(1.00 mL, 7.17 mmol) was added to the mixture and stirring was
continued for 8 h. Additional diethyl malonate (0.5 mL, 525 mg, 3.28
mmol) was added to the mixture and stirring was continued overnight.
The reaction temperature was increased to 100 ◦C and stirring was
continued for 1 h. Additional diethyl malonate (0.5 mL, 525 mg, 3.28
mmol) was added, followed by triethylamine (1.00 mL, 7.17 mmol) and
stirring was continued for 5 h. Then the mixture was cooled to room
temperature, poured into water (400 mL) and acidified with acetic acid
(6 mL) to ~ pH = 4. Some solid precipitated and was collected via
vacuum filtration. The filtrate was extracted with ethyl acetate (3X) and
the combined organic layers were washed with brine, dried over sodium
sulfate, filtered, and concentrated to give an orange solid. This material
1
(900 mg, 3.66 mmol, 55% yield) as an off white solid. H NMR (400
MHz, CDCl3) δ 1.37–1.52 (t, J = 7 Hz, 3H), 4.48–4.59 (q, J = 7 Hz, 2H),
7.58 (d, J = 9 Hz, 1H), 8.26 (d, J = 9 Hz, 1H), 8.88 (d, J = 2 Hz, 1H), 9.66
(d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H = 237.
7-Methoxy-1,8-naphthyridine-3-carboxylic acid 7g
Sodium methoxide (25% in methanol, 1.15 mL, 5.03 mmol) was
added to ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.238 g,
1.006 mmol) in methanol (10 mL) at room temperature and the reaction
◦
mixture was heated at 60 C for two hours. The reaction mixture was
concentrated and water (10 mL) was added. The reaction mixture was
19

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
was combined with the previously collected tan solid and recrystallized
from ethanol to give ethyl 2-(methylthio)-7-oxo-7,8-dihydropyrido[2,3-
d]pyrimidine-6-carboxylate 22 (903 mg, 3.40 mmol, 29% yield) as a tan
solid. The mother liquor from the recrystallization was evaporated
under reduced pressure. The remaining material was dissolved in a
minimal amount of dichloromethane and purified via silica gel chro-
matography, eluting with ethyl acetate:hexanes (1:19 to 1:1) to give
recovered starting material (230 mg) as a yellow solid. The aqueous
layer from the previous workup contained solid material. This solid was
collected via vacuum filtration, washed with water and dried in vacuo to
give ethyl 2-(methylthio)-7-oxo-7,8-dihydropyrido[2,3-d]pyrimidine-6-
carboxylate 22 (496 mg, 1.87 mmol, 16% yield) as a tan solid. 1H NMR
(400 MHz, CD3SOCD3) δ 1.29 (t, J = 7 Hz, 3H), 2.57 (s, 3H), 4.26 (q, J =
7 Hz, 2H), 8.51 (s, 1H), 8.99 (s, 1H), 12.66 (br s, 1H); LC-MS (LC-ES) M
+ H = 266.
1H); LC-MS (LC-ES) M + H = 250.
Ethyl 2-(methylsulfinyl)pyrido[2,3-d]pyrimidine-6-carboxylate 17e
To a stirred, cooled (0 ◦C) solution of ethyl 2-(methylthio)pyrido
[2,3-d]pyrimidine-6-carboxylate 17d (200 mg, 0.802 mmol) in
dichloromethane (10 mL) was added 3-chloroperoxybenzoic acid (370
mg, 1.651 mmol). The mixture was stirred for 30 min. The mixture was
poured into saturated aqueous sodium bicarbonate and extracted with
dichloromethane (2X). The combined organic layers were washed with
brine, dried over sodium sulfate, filtered, and concentrated to give crude
ethyl 2-(methylsulfinyl)pyrido[2,3-d]pyrimidine-6-carboxylate, con-
taining some ethyl 2-(methylsulfonyl)pyrido[2,3-d]pyrimidine-6-
carboxylate 17e (155 mg, 73% yield) which was carried forward to the
next reaction. 1H NMR (400 MHz, CD3SOCD3) δ 1.40 (t, J = 7 Hz, 3H),
3.00 (s, 3H), 4.46 (q, J = 7 Hz, 2H), 9.36 (d, J = 2 Hz, 1H), 9.73 (d, J = 2
Hz, 1H), 10.04 (s, 1H); LC-MS (LC-ES) M + H = 266.
Ethyl
7-chloro-2-(methylthio)pyrido[2,3-d]pyrimidine-6-carbox-
ylate 23
Methyl 2-methoxypyrido[2,3-d]pyrimidine-6-carboxylate 6b
A slurry of ethyl 2-(methylthio)-7-oxo-7,8-dihydropyrido[2,3-d]py-
rimidine-6-carboxylate 22 (0.795 g, 3.00 mmol) in phosphorus oxy-
chloride (6 mL, 64.4 mmol) was heated to 100 ◦C and stirred for 4 h. The
mixture was still not homogeneous. Stirring was continued for 1 h and
the mixture became homogeneous. After cooling to room temperature,
the mixture was carefully pipetted into rapidly stirring ice cold saturated
aqueous sodium bicarbonate and carefully and slowly adjusted to ~ pH
= 5 with saturated aqueous sodium bicarbonate. The mixture was stirred
for ~ 5 min and the resulting precipitate was collected via vacuum
filtration, washed with water and dried in vacuo to give ethyl 7-chloro-
2-(methylthio)pyrido[2,3-d]pyrimidine-6-carboxylate (681 mg, 2.40
mmol, 80% yield) as a tan solid. The filtrate was extracted with ethyl
acetate (2X), washed with brine, dried over sodium sulfate, filtered, and
concentrated to give ethyl 7-chloro-2-(methylthio)pyrido[2,3-d]pyrim-
idine-6-carboxylate 23 (117 mg, 0.412 mmol, 14% yield) as a yellow
solid. 1H NMR (400 MHz, CD3SOCD3) δ 1.36 (t, J = 7 Hz, 3H), 2.65 (s,
3H), 4.40 (q, J = 7 Hz, 2H), 9.10 (s, 1H), 9.51 (s, 1H); LC-MS (LC-ES) M
+ H = 284.
To a stirred solution of crude ethyl 2-(methylsulfinyl)pyrido[2,3-d]
pyrimidine-6-carboxylate 17e (154 mg, 0.581 mmol) in methanol (5
mL) was added 25% sodium methoxide (1 mL, 4.37 mmol) in methanol.
A precipitate formed immediately upon addition of the sodium meth-
oxide. The mixture was stirred for 20 min, then filtered and the collected
solid was washed with a small amount of methanol and dried in vacuo to
give methyl 2-methoxypyrido[2,3-d]pyrimidine-6-carboxylate 6b (47
mg, 0.214 mmol, 37% yield) as a white solid. 1H NMR (400 MHz,
CD3SOCD3) δ 3.10 (s, 3H), 3.25 (s, 3H), 8.32 (d, J = 2 Hz, 1H), 8.66 (d, J
= 2 Hz, 1H), 8.88 (s, 1H); LC-MS (LC-ES) M + H = 220.
2-Methoxypyrido[2,3-d]pyrimidine-6-carboxylic acid 7h
Ethyl 2-(methylthio)pyrido[2,3-d]pyrimidine-6-carboxylate 17d
To a stirred suspension of methyl 2-methoxypyrido[2,3-d]pyrimi-
dine-6-carboxylate 6b (45 mg, 0.205 mmol) in tetrahydrofuran (1.5 mL)
and methanol (0.5 mL) was added 1 M aqueous lithium hydroxide (0.25
mL, 0.250 mmol). The mixture eventually became homogeneous and
was stirred for 1 h. The solvent was removed under reduced pressure.
The remaining material was suspended in water, acidified with 1 N
aqueous hydrochloric acid and extracted with ethyl acetate (2X). The
combined organic layers were washed with brine, dried over sodium
sulfate, filtered, and concentrated to give 2-methoxypyrido[2,3-d]py-
rimidine-6-carboxylic acid 7h (18 mg, 0.088 mmol, 43% yield) as a
white solid. 1H NMR (400 MHz, CD3SOCD3) δ 4.09 (s, 3H), 9.10 (d, J = 2
Hz, 1H), 9.49 (d, J = 2 Hz, 1H), 9.71 (s, 1H), 13.68 (br s, 1H); LC-MS (LC-
ES) M + H = 206.
Acetonitrile (10 mL) was added to ethyl 7-chloro-2-(methylthio)
pyrido[2,3-d]pyrimidine-6-carboxylate 23 (678 mg, 2.390 mmol) and
bis(triphenylphosphine)palladium(II) chloride (85 mg, 0.121 mmol)
and the mixture was degassed by sparging with nitrogen for ~ 15 min.
Then, triethylsilane (0.50 mL, 3.13 mmol) was added to the mixture and
it was heated to 80 ◦C and stirred overnight. After stirring at 80 ◦C for 19
h, the mixture was cooled to room temperature and a solid precipitated.
The solid was collected by vacuum filtration, washed with hexane and
dried in vacuo to give ethyl 2-(methylthio)pyrido[2,3-d]pyrimidine-6-
carboxylate (237 mg, 0.951 mmol, 39.8% yield) as a tan solid. The
filtrate was evaporated to dryness, dissolved in a minimal amount of
dichloromethane, containing enough methanol for complete solubili-
zation, and purified by silica gel chromatography, eluting with ethyl
acetate:hexanes (1:19 to 2:3) to give ethyl 2,7-bis(methylthio)pyrido
[2,3-d]pyrimidine-6-carboxylate (101 mg, 0.342 mmol, 14% yield) as a
yellow solid and ethyl 2-(methylthio)pyrido[2,3-d]pyrimidine-6-
carboxylate 17d (52 mg, 0.209 mmol, 9% yield) as a yellow solid. 1H
NMR (400 MHz, CD3SOCD3) δ 1.38 (t, J = 7 Hz, 3H), 2.66 (s, 3H), 4.41
(q, J = 7 Hz, 2H), 9.13 (d, J = 2 Hz, 1H), 9.51 (d, J = 2 Hz, 1H), 9.65 (s,
N-((trans)-4-(2-Hydroxypropan-2-yl)cyclohexyl)-2-methoxypyrido
[2,3-d]pyrimidine-6-carboxamide 1l
20

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
To a thick, stirred suspension of 2-methoxypyrido[2,3-d]pyrimidine-
(0.0597 g, 0.253 mmol) in N,N-dimethylformamide (0.843 mL) at room
temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (0.048
g, 0.303 mmol, PharmaBlock) was added and the reaction mixture was
stirred for five minutes. Then, n-propylphosphonic acid anhydride
(0.301 mL, 0.506 mmol) was added and the reaction mixture was stirred
for sixteen hours. The reaction mixture was concentrated. The resulting
residue was purified by RP HPLC, eluting with acetonitrile:water with
0.1% ammonium hydroxide (5:95 to 100:0), then further purified by
silica gel chromatography, eluting with methanol:ethyl acetate (0:1 to
3:7) to give 7-(azetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclo-
hexyl)-1,6-naphthyridine-3-carboxamide 1n (0.0514 g, 0.133 mmol,
52.4% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24
(m, 3H), 1.29 (q, J = 12 Hz, 2H), 1.83 (br d, J = 12 Hz, 2H), 1.91 (br d, J
= 12 Hz, 2H), 2.39 (p, J = 7 Hz, 2H), 3.64–3.78 (m, 1H), 4.03 (s, 1H),
4.08 (t, J = 7 Hz, 4H), 6.54 (s, 1H), 8.38 (d, J = 7 Hz, 1H), 8.69 (s, 1H),
6-carboxylic acid 7h (16 mg, 0.078 mmol) and 2-((trans)-4-amino-
cyclohexyl)propan-2-ol 8a (15 mg, 0.095 mmol, PharmaBlock) in N,N-
dimethylformamide (1 mL) was added N,N-diisopropylethylamine
(0.050 mL, 0.286 mmol), followed by n-propylphosphonic acid anhy-
dride (0.095 mL, 0.161 mmol). The mixture quickly became homoge-
neous and was allowed to stir overnight and concentrated. The mixture
was purified by reverse phase chromatography, eluting with acetoni-
trile–water with 0.1% ammonium hydroxide (0:1 to 7:3) to give N-
((trans)-4-(2-hydroxypropan-2-yl)cyclohexyl)-2-methoxypyrido[2,3-d]
pyrimidine-6-carboxamide 1k (10 mg, 0.029 mmol, 37%) as a white
solid. 1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H),
1.31 (q, J = 12 Hz, 2H), 1.84 (br d, J = 12 Hz, 2H), 1.94 (br d, J = 11 Hz,
2H), 3.68–3.80 (m, 1H), 4.08 (s, 3H), 8.65 (d, J = 8 Hz, 1H), 8.96 (d, J =
2 Hz, 1H), 9.47 (d, J = 2 Hz, 1H), 9.64 (s, 1H); LC-MS (LC-ES)
C
18H24N4O3 M + H = 345; tR = 0.59 min, 100% purity.
Ethyl 7-(azetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6c
9.04 (s, 1H), 9.17 (s, 1H); LC-MS (LC-ES) C21H28N4O2 M + H = 369; tR =
0.55 min, 100% purity.
Ethyl 7-(3-fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6d
and Ethyl 7-(dimethylamino)-1,6-naphthyridine-3-carboxylate
N,N-Diisopropylethylamine (1.502 mL, 8.62 mmol) was added to
ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a (0.5102 g, 2.156
mmol) in N-methyl-2-pyrrolidone (7.19 mL) at room temperature. Then,
azetidine hydrochloride (0.605 g, 6.47 mmol) was added and the reac-
tion mixture was heated at 100 ◦C in the microwave for five hours. The
reaction mixture was diluted in dichloromethane, washed with satu-
rated sodium bicarbonate, dried over magnesium sulfate, filtered, and
concentrated. The resulting residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to
100:0), then further purified by silica gel chromatography, eluting with
ethyl acetate:hexanes (3:7 to 4:1) to give ethyl 7-(azetidin-1-yl)-1,6-
naphthyridine-3-carboxylate 6c (0.2198 g, 0.812 mmol, 37.6% yield).
1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H), 2.40 (p, J = 7
Hz, 2H), 4.10 (t, J = 7 Hz, 4H), 4.35 (q, J = 7 Hz, 2H), 6.53 (s, 1H), 8.83
(s, 1H), 9.15 (s, 1H), 9.17 (s, 1H); LC-MS (LC-ES) M + H = 258.
7-(Azetidin-1-yl)-1,6-naphthyridine-3-carboxylic acid lithium salt 7i
N,N-Diisopropylethylamine (1.259 mL, 7.23 mmol) was added to
ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a (0.4278 g, 1.808
mmol) in N,N-dimethylformamide (6.03 mL) at room temperature. Then
3-fluoroazetidin-1-ium chloride (0.605 g, 5.42 mmol) was added and the
reaction mixture was heated at 100 ◦C in the microwave for three hours.
The reaction mixture was diluted in dichloromethane, washed with
saturated sodium bicarbonate, dried over magnesium sulfate, filtered,
and concentrated. The residue was purified by silica gel chromatog-
raphy, eluting with ethyl acetate:hexanes (2:3 to 4:1), then further pu-
rified by RP HPLC, eluting with acetonitrile:water with 0.1% ammonium
hydroxide (30:70 to 80:20) and shaving fractions to give some pure
ethyl 7-(3-fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6d
(0.0831 g, 0.287 mmol, 15.86% yield) as well as some impure material
and some pure ethyl 7-(dimethylamino)-1,6-naphthyridine-3-carbox-
ylate (0.0419 g, 0.162 mmol, 8.98% yield).
Ethyl 7-(3-fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6d
1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H), 4.18 (br dd,
J = 24, 11 Hz, 2H), 4.36 (q, J = 7 Hz, 2H), 4.38–4.50 (m, 2H), 5.44–5.70
(m, 1H), 6.69 (s, 1H), 8.91 (s, 1H), 9.21 (s, 1H), 9.23 (s, 1H); LC-MS (LC-
ES) M + H = 276.
Lithium hydroxide (0.061 g, 2.56 mmol) was added to ethyl 7-
(azetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6c (0.2198 g, 0.854
mmol) in methanol (3.42 mL) and water (0.854 mL) at room tempera-
ture and the reaction mixture was stirred sixteen hours at 60 ◦C. Then,
the reaction mixture was concentrated to give 7-(azetidin-1-yl)-1,6-
naphthyridine-3-carboxylic acid lithium salt 7i (0.2291 g, 0.854 mmol,
100% yield). 1H NMR (400 MHz, CD3SOCD3) δ 2.36 (p, J = 7 Hz, 2H),
4.02 (t, J = 7 Hz, 4H), 6.51 (s, 1H), 8.50 (s, 1H), 8.96 (s, 1H), 9.25 (s,
1H); LC-MS (LC-ES) Mꢀ H = 230.
Ethyl 7-(dimethylamino)-1,6-naphthyridine-3-carboxylate
1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H), 3.18 (s, 6H),
4.35 (q, J = 7 Hz, 2H), 6.81 (s, 1H), 8.82 (s, 1H), 9.17 (s, 2H); LC-MS
(LC-ES) M + H = 246.
7-(3-Fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylic
acid
ammonia salt 7j
7-(Azetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,6-
naphthyridine-3-carboxamide 1m
Lithium hydroxide (0.022 g, 0.906 mmol) was added to ethyl 7-(3-
fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6d (0.0831 g,
0.302 mmol) in methanol (1.20 mL) and water (0.30 mL) at room
temperature and the reaction mixture was stirred sixteen hours at 60 ◦C.
The reaction mixture was concentrated. The reaction mixture was pu-
rified by RP HPLC, eluting with acetonitrile:water with 0.1% ammonium
hydroxide (0:100 to 60:40) to give 7-(3-fluoroazetidin-1-yl)-1,6-naph-
thyridine-3-carboxylic acid ammonia salt 7j (0.0579 g, 0.208 mmol,
N,N-Diisopropylethylamine (0.265 mL, 1.517 mmol) was added to 7-
(azetidin-1-yl)-1,6-naphthyridine-3-carboxylic acid lithium salt 7i
21

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
69.0% yield). 1H NMR (400 MHz, CD3SOCD3) δ 4.09 (br dd, J = 25, 10
Hz, 2H), 4.02–4.44 (m, 2H), 5.42–5.66 (m, 1H), 6.65 (s, 1H), 8.54 (s,
1H), 9.01 (s, 1H), 9.28 (s, 1H); LC-MS (LC-ES) M + H = 248.
7-(3-Fluoroazetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclo-
hexyl)-1,6-naphthyridine-3-carboxamide 1n
methylazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6e (0.2731 g,
1.007 mmol) in methanol (4.03 mL) and water (1.007 mL) at room
temperature and the reaction mixture was stirred three hours at 60 ◦C.
The reaction mixture was concentrated to give lithium 7-(2-methyl-
azetidin-1-yl)-1,6-naphthyridine-3-carboxylic acid 7k (0.2551 g, 0.969
mmol, 96% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.49 (d, J = 6 Hz,
3H), 2.03 (p, J = 9 Hz, 1H), 2.36–2.48 (m, 1H), 3.80 (q, J = 8 Hz, 1H),
3.98 (q, J = 8 Hz, 1H), 4.36 (h, J = 7 Hz, 1H), 6.51 (s, 1H), 8.51 (s, 1H),
8.97 (s, 1H), 9.25 (s, 1H); LC-MS (LC-ES) M + H = 244.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((R)-2-methyl-
azetidin-1-yl)-1,6-naphthyridine-3-carboxamide 1o and N-(trans-4-(2-
Hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-methylazetidin-1-yl)-1,6-
naphthyridine-3-carboxamide 1p
N,N-Diisopropylethylamine (0.230 mL, 1.315 mmol) was added to 7-
(3-fluoroazetidin-1-yl)-1,6-naphthyridine-3-carboxylic acid ammonia
salt 7j (0.0579 g, 0.219 mmol) in N,N-dimethylformamide (0.730 mL) at
room temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(0.041 g, 0.263 mmol, PharmaBlock) was added and the reaction
mixture was stirred for five minutes. Then, n-propylphosphonic acid
anhydride (0.261 mL, 0.438 mmol) was added and the reaction mixture
was stirred for sixty-four hours. The reaction mixture was concentrated.
The resulting residue was purified by RP HPLC, eluting with acetonitrile:
water with 0.1% ammonium hydroxide (5:95 to 100:0), then further
purified by silica gel chromatography, eluting with methanol:ethyl ac-
etate (0:1 to 1:4) to give 7-(3-fluoroazetidin-1-yl)-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-1,6-naphthyridine-3-carboxamide 1n
(0.0622 g, 0.153 mmol, 69.8% yield). 1H NMR (400 MHz, CD3SOCD3) δ
1.04 (s, 6H), 1.06–1.24 (m, 3H), 1.30 (q, J = 13 Hz, 2H), 1.83 (br d, J =
12 Hz, 2H), 1.92 (br d, J = 12 Hz, 2H), 3.64–3.78 (m, 1H), 4.03 (s, 1H),
4.14 (br dd, J = 24, 10 Hz, 2H), 4.34–4.48 (m, 2H), 5.38–5.68 (m, 1H),
6.70 (s, 1H), 8.42 (d, J = 7 Hz, 1H), 8.73 (s, 1H), 9.09 (s, 1H), 9.21 (s,
1H); LC-MS (LC-ES) C21H27FN4O2 M + H = 387; tR = 0.58 min, 100%
purity.
N,N-Diisopropylethylamine (0.426 mL, 2.441 mmol) was added to 7-
(2-methylazetidin-1-yl)-1,6-naphthyridine-3-carboxylic acid lithium
salt 7k (0.1018 g, 0.407 mmol) in N,N-dimethylformamide (1.36 mL) at
room temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(0.077 g, 0.488 mmol, PharmaBlock) was added and the reaction
mixture was stirred for five minutes. Then, n-propylphosphonic acid
anhydride (0.484 mL, 0.814 mmol) was added and the reaction mixture
was stirred for sixteen hours. The reaction mixture was concentrated.
The resulting residue was purified by RP HPLC, eluting with acetonitrile:
water with 0.1% ammonium hydroxide (5:95 to 100:0), then further
purified by silica gel chromatography, eluting with methanol:ethyl ac-
etate (0:1 to 1:9) to give racemic N-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)-7-(2-methylazetidin-1-yl)-1,6-naphthyridine-3-carbox-
amide (0.1064 g, 0.264 mmol, 64.9% yield). 1H NMR (400 MHz,
CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.30 (q, J = 12 Hz, 2H),
1.50 (d, J = 6 Hz, 3H), 1.83 (br d, J = 12 Hz, 2H), 1.92 (br d, J = 11 Hz,
2H), 2.04 (p, J = 9 Hz, 1H), 2.42–2.54 (m, 1H), 3.66–3.78 (m, 1H), 3.86
(q, J = 8 Hz, 1H), 4.02 (s, 1H), 4.04 (q, J = 6 Hz, 1H), 4.44 (h, J = 7 Hz,
1H), 6.54 (s, 1H), 8.38 (d, J = 7 Hz, 1H), 8.69 (s, 1H), 9.05 (s, 1H), 9.17
(s, 1H); LC-MS (LC-ES) C22H30N4O2 M + H = 383. Racemic N-(trans-4-
(2-hydroxypropan-2-yl)cyclohexyl)-7-(2-methylazetidin-1-yl)-1,6-
naphthyridine-3-carboxamide (0.0971 g, 0.254 mmol) was separated
into its enantiomers on a chiral IC column eluting with methanol:hex-
anes (3:2) with 1% diethylamine to give N-(trans-4-(2-hydroxypropan-2-
yl)cyclohexyl)-7-((R)-2-methylazetidin-1-yl)-1,6-naphthyridine-3-car-
boxamide 1o (0.0201 g, 0.050 mmol, 19.67% yield) as the first diaste-
reomer (>99% ee) to elute and N-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)-7-((S)-2-methylazetidin-1-yl)-1,6-naphthyridine-3-carbox-
amide 1p (0.0230 g, 0.057 mmol, 22.50% yield) as the last diastereomer
to elute (96.6% ee). The structures were assigned by analogy.
Ethyl 7-(2-methylazetidin-1-yl)-1,6-naphthyridine-3-carboxylate 6e
N,N-Diisopropylethylamine (1.361 mL, 7.81 mmol) was added to
ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a (0.4622 g, 1.953
mmol) in N-methyl-2-pyrrolidone (6.51 mL) at room temperature. Then
2-methylazetidine hydrochloride (0.420 g, 3.91 mmol) was added and
◦
the reaction mixture was heated at 100 C in the microwave for four
hours. The reaction mixture was diluted in dichloromethane, washed
with saturated sodium bicarbonate, dried over magnesium sulfate,
filtered, and concentrated. The resulting residue was purified by RP
HPLC, eluting with acetonitrile:water with 0.1% ammonium hydroxide
(5:95:100:0), then further purified by silica gel chromatography, eluting
with ethyl acetate:hexanes (1:4 to 3:2) to give ethyl 7-(2-methylazetidin-
1-yl)-1,6-naphthyridine-3-carboxylate 6e (0.2731 g, 0.956 mmol,
49.0% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H),
1.51 (d, J = 6 Hz, 3H), 2.04 (p, J = 8 Hz, 1H), 2.50 (p, J = 8 Hz, 1H), 3.90
(q, J = 8 Hz, 1H), 4.06 (q, J = 5 Hz, 1H), 4.35 (q, J = 7 Hz, 2H), 4.48 (h,
J = 7 Hz, 1H), 6.53 (s, 1H), 8.83 (s, 1H), 9.15 (s, 1H), 9.17 (s, 1H); LC-MS
(LC-ES) M + H = 272.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((R)-2-methyl-
azetidin-1-yl)-1,6-naphthyridine-3-carboxamide 1o
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H),
1.30 (q, J = 12 Hz, 2H), 1.51 (d, J = 6 Hz, 3H), 1.83 (br d, J = 12 Hz,
2H), 1.92 (br d, J = 12 Hz, 2H), 2.04 (p, J = 8 Hz, 1H), 2.42–2.54 (m,
1H), 3.66–3.78 (m, 1H), 3.87 (q, J = 8 Hz, 1H), 4.01 (s, 1H), 4.04 (q, J =
5 Hz, 1H), 4.44 (h, J = 6 Hz, 1H), 6.54 (s, 1H), 8.38 (d, J = 7 Hz, 1H),
8.68 (s, 1H), 9.05 (s, 1H), 9.17 (s, 1H); LC-MS (LC-ES) C22H30N4O2 M +
H = 383; tR = 0.62 min, 100% purity.
Lithium 7-(2-methylazetidin-1-yl)-1,6-naphthyridine-3-carboxylic
acid k
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-methyl-
azetidin-1-yl)-1,6-naphthyridine-3-carboxamide 1p
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H),
1.30 (q, J = 12 Hz, 2H), 1.50 (d, J = 6 Hz, 3H), 1.83 (br d, J = 12 Hz,
2H), 1.92 (br d, J = 11 Hz, 2H), 2.04 (p, J = 9 Hz, 1H), 2.42–2.54 (m,
Lithium hydroxide (0.072 g, 3.02 mmol) was added to ethyl 7-(2-
22

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
1H), 3.66–3.78 (m, 1H), 3.87 (q, J = 8 Hz, 1H), 4.02 (s, 1H), 4.04 (q, J =
5 Hz, 1H), 4.44 (h, J = 6 Hz, 1H), 6.54 (s, 1H), 8.38 (d, J = 8 Hz, 1H),
8.68 (s, 1H), 9.05 (s, 1H), 9.17 (s, 1H); LC-MS (LC-ES) C22H30N4O2 M +
H = 383; tR = 0.62 min, 100% purity.
8a (236 mg, 1.498 mmol, PharmaBlock) and N,N-diisopropylethylamine
(0.262 mL, 1.498 mmol) were added. The reaction was stirred at room
temperature over the weekend. Water was added to the vessel in at-
tempts to crash out the product, to no avail. The solution was concen-
trated in vacuo to give a crude solid. The residue was purified using
silica gel chromatography, eluting with methanol:dichloromethane (0:1
to 1:9) to give 7-cyclopropyl-N-((trans)-4-(2-hydroxypropan-2-yl)
cyclohexyl)-1,6-naphthyridine-3-carboxamide 1q (318 mg, 0.900
mmol, 90% yield). 1H NMR (400 MHz, CD3OD) δ 1.09–1.53 (m, 15H),
1.92–2.19 (m, 4H), 2.26–2.43 (m, 1H), 3.85–3.99 (m, 1H), 7.80 (s, 1H),
8.65 (d, J = 8 Hz, 1H), 8.82–8.92 (m, 1H), 9.28 (s, 1H), 9.39 (d, J = 2 Hz,
1H); LC-MS (LC-ES) C21H27N3O2 M + H = 354; tR = 0.65 min, 100%
purity.
Ethyl 7-cyclopropyl-1,6-naphthyridine-3-carboxylate 6f
Cyclopropylboronic acid (17.42 g, 203 mmol) was added to a solu-
tion of ethyl 7-chloro-1,6-naphthyridine-3-carboxylate 17a (20 g, 67.6
mmol) in toluene (600 mL) at room temperature under nitrogen. Then,
2-dicyclohexylphosphino-2′,6′-dimethoxybiphenyl (SPhos) (0.638 g,
1.555 mmol) was added, followed by the addition of tris(dibenzylide-
neacetone)dipalladium(0) (3.65 g, 3.99 mmol) and 2 M aqueous sodium
carbonate (42.3 mL, 85 mmol) and the reaction was purged with argon
for 10 min. The reaction mixture was heated to 110 ◦C and stirred for
sixteen hours. On completion, the reaction mixture was filtered through
a Celite® pad, the residue was washed with ethyl acetate (200 mL), and
the filtrate was evaporated under reduced pressure to give an impure
material, which was purified via silica gel column chromatography,
eluting with ethyl acetate:hexanes (1:4) to afford ethyl 7-cyclopropyl-
1,6-naphthyridine-3-carboxylate 6f (15 g, 61.4 mmol, 91% yield) as
an off white solid. 1H NMR (400 MHz, CDCl3) δ 1.05–1.24 (m, 4H), 1.46
(t, J = 7 Hz, 3H), 2.26–2.36 (m, 1H), 4.48 (q, J = 7 Hz, 2H), 7.78 (s, 1H),
8.84–8.93 (m, 1H), 9.21 (s, 1H), 9.54 (d, J = 2 Hz, 1H); LC-MS (LC-ES)
M + H = 243.
7-(2,2,2-Trifluoroethoxy)-1,8-naphthyridine-3-carboxylic acid 7m
and 7-Ethoxy-1,8-naphthyridine-3-carboxylic acid
2,2,2-Trifluoroethan-1-ol (0.18 mL, 2.470 mmol) was added to so-
dium hydride (60% in mineral oil, 0.121 g, 3.03 mmol) in tetrahydro-
furan (10 mL) under nitrogen. It bubbled gently. After 30 min, ethyl 7-
chloro-1,8-naphthyridine-3-carboxylate 17c (0.241 g, 1.018 mmol) was
added and the reaction mixture was stirred for nineteen hours. Water (2
mL) was added to the reaction mixture and it was allowed to stir for six
hours. Then, the reaction mixture was partitioned between diethyl ether
(25 mL) and water (15 mL) and the layers were separated. The aqueous
layer was filtered through a pad of Celite®, gently concentrated to
remove any remaining organics, and acidified with 1 N hydrochloric
acid (2 mL, pH = ~3–4). The precipitate was collected by filtration,
rinsed with water (2X), and air-dried, then dried under vacuum to give a
mixture of 7-(2,2,2-trifluoroethoxy)-1,8-naphthyridine-3-carboxylic
acid 7m and 7-ethoxy-1,8-naphthyridine-3-carboxylic acid (0.231 g,
~65:35 ratio). LC-MS (LC-ES) M + H = 219 and LC-MS (LC-ES) M + H =
273.
7-Cyclopropyl-1,6-naphthyridine-3-carboxylic acid 7l
A solution of sodium hydroxide (5.76 g, 144 mmol) in water (75 mL)
was added to a solution of ethyl 7-cyclopropyl-1,6-naphthyridine-3-
carboxylate 6f (30 g, 120 mmol) in tetrahydrofuran (75 mL) at room
temperature and the resulting reaction mixture was stirred for fifteen
hours. On completion, the reaction mixture was concentrated (until
tetrahydrofuran was removed) under vacuum, and then the reaction
mixture was acidified with 1 M hydrochloric acid solution. The precip-
itate was filtered, washed with water (250 mL), pentane (500 mL) and
diethyl ether (500 mL), and dried to afford 7-cyclopropyl-1,6-naphthyr-
idine-3-carboxylic acid 7 l (18 g, 84 mmol, 70% yield) as an off white
solid. 1H NMR (400 MHz, CD3SOCD3) δ 0.96–1.22 (m, 4H), 2.25–2.40
(m, 1H), 7.89 (s, 1H), 9.06 (dd, J = 2, 1 Hz, 1H), 9.43 (d, J = 3 Hz, 2H),
13.57 (br s, 1H); LC-MS (LC-ES) M + H = 215.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-(2,2,2-tri-
fluoroethoxy)-1,8-naphthyridine-3-carboxamide 1r and 7-Ethoxy-N-
(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-naphthyridine-3-car-
boxamide
7-Cyclopropyl-N-((trans)-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,6-
naphthyridine-3-carboxamide 1q
2-(trans-4-Aminocyclohexyl)propan-2-ol 8a (0.144 g, 0.916 mmol,
PharmaBlock) was added to a mixture of 7-(2,2,2-trifluoroethoxy)-1,8-
naphthyridine-3-carboxylic acid 7m and 7-ethoxy-1,8-naphthyridine-3-
carboxylic acid (0.249 g, 0.913 mmol) in N,N-dimethylformamide (10
mL) at room temperature. Then, N,N-diisopropylethylamine (0.40 mL,
2.296 mmol) was added, followed by 1-[bis(dimethylamino)methy-
lene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate
(0.418 g, 1.099 mmol) and the reaction mixture was stirred for four
hours. The reaction mixture was concentrated, and the resulting residue
was purified by silica gel chromatography, eluting with (ethyl acetate:
ethanol (3:1)):hexanes (0:1 to 1:0). The mixed fractions were repurified
by silica gel chromatography, eluting with (ethyl acetate:ethanol (3:1)):
hexanes (0:1 to 3:1) and the mixed fractions were repurified by silica gel
chromatography, eluting with (ethyl acetate:ethanol (3:1)):hexanes (0:1
to 3:1), then combined with the appropriate fractions to give residues
which were triturated/sonicated with ethyl acetate and filtered to give
To a stirring suspension of 7-cyclopropyl-1,6-naphthyridine-3-car-
boxylic acid 7 l (214 mg, 0.999 mmol) in N,N-dimethylformamide
(13.3 mL) was added N,N-diisopropylethylamine (0.262 mL, 1.498
mmol) followed by 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo
[4,5-b]pyridinium 3-oxide hexafluorophosphate (475 mg, 1.249 mmol)
in one portion. After ~ 5 min, 2-((trans)-4-aminocyclohexyl)propan-2-ol
23

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-7-(2,2,2-tri-
fluoroethoxy)-1,8-naphthyridine-3-carboxamide 1r (0.188 g, 0.457
mmol, 49.9% yield) as a white powder and 7-ethoxy-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-1,8-naphthyridine-3-carboxamide
(0.045 g, 0.126 mmol, 13.7% yield) as an off-white powder.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-(2,2,2-tri-
fluoroethoxy)-1,8-naphthyridine-3-carboxamide 1r
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 3H),
1.32 (q, J = 12 Hz, 2H), 1.84 (br d, J = 11 Hz, 2H), 1.94 (br d, J = 10 Hz,
2H), 3.68–3.80 (m, 1H), 4.06 (s, 1H), 5.21 (q, J = 9 Hz, 2H), 7.19 (d, J =
9 Hz, 1H), 8.40 (d, J = 9 Hz, 1H), 8.54 (d, J = 8 Hz, 1H), 8.78 (d, J = 2
Hz, 1H), 9.27 (d, J = 2 Hz, 1H); LC-MS (LC-ES) C20H24F3N3O3 M + H =
412; tR = 0.85 min, 97% purity.
N,N-Diisopropylethylamine (0.298 mL, 1.707 mmol) was added to 7-
(azetidin-1-yl)-1,8-naphthyridine-3-carboxylic acid 7n (0.0672 g, 0.285
mmol) in N,N-dimethylformamide (0.948 mL) at room temperature.
Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (0.054 g, 0.341 mmol,
PharmaBlock) was added and the reaction mixture was stirred for five
minutes. Then, n-propylphosphonic acid anhydride (0.339 mL, 0.569
mmol) was added and the reaction mixture was stirred for sixty-six
hours. The reaction mixture was concentrated. The resulting residue
was purified by RP HPLC, eluting with acetonitrile:water with 0.1%
ammonium hydroxide (5:95 to 100:0), then further purified by silica gel
chromatography, eluting with methanol:ethyl acetate (0:1 to 1:4) to give
7-(azetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-
naphthyridine-3-carboxamide 1s (0.0648 g, 0.167 mmol, 58.7% yield).
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.30
(q, J = 11 Hz, 2H), 1.83 (br d, J = 11 Hz, 2H), 1.92 (br d, J = 11 Hz, 2H),
2.39 (p, J = 7 Hz, 2H), 3.66–3.78 (m, 1H), 4.02 (s, 1H), 4.16 (t, J = 7 Hz,
4H), 6.78 (d, J = 9 Hz, 1H), 8.08 (d, J = 9 Hz, 1H), 8.31 (d, J = 7 Hz, 1H),
7-Ethoxy-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-naph-
thyridine-3-carboxamide
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 3H),
1.31 (q, J = 12 Hz, 2H), 1.40 (t, J = 7 Hz, 3H), 1.83 (br d, J = 12 Hz, 2H),
1.93 (br d, J = 10 Hz, 2H), 3.68–3.80 (m, 1H), 4.06 (s, 1H), 4.51 (q, J =
7 Hz, 2H), 7.16 (d, J = 9 Hz, 1H), 8.39 (d, J = 9 Hz, 1H), 8.53 (d, J = 8
Hz, 1H), 8.76 (d, J = 2 Hz, 1H), 9.26 (d, J = 2 Hz, 1H); LC-MS (LC-ES)
C
20H27N3O3 M + H = 358.
Ethyl 7-(azetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6g
8.51 (s, 1H), 9.10 (s, 1H); LC-MS (LC-ES) C21H28N4O2 M + H = 369; tR
=
0.47 min, 100% purity.
Ethyl 7-(3-fluoroazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6h
N,N-Diisopropylethylamine (1.35 mL, 7.75 mmol) was added to
ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.4588 g, 1.939
mmol) in N-methyl-2-pyrrolidone (6.46 mL) at room temperature. Then,
azetidine hydrochloride (0.544 g, 5.82 mmol) was added and the reac-
tion mixture was heated at 100 ◦C in the microwave for one hour. The
reaction mixture was diluted in dichloromethane, washed with satu-
rated sodium bicarbonate, dried over magnesium sulfate, filtered, and
concentrated. The resulting residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to 100:0)
to give ethyl 7-(azetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6g
(0.4469 g, 1.650 mmol, 85% yield). 1H NMR (400 MHz, CD3SOCD3) δ
1.34 (t, J = 7 Hz, 3H), 2.39 (p, J = 7 Hz, 2H), 4.18 (t, J = 7 Hz, 4H), 4.34
(q, J = 7 Hz, 2H), 6.81 (d, J = 9 Hz, 1H), 8.17 (d, J = 9 Hz, 1H), 8.65 (s,
1H), 9.12 (s, 1H); LC-MS (LC-ES) M + H = 258.
N,N-Diisopropylethylamine (1.23 mL, 7.09 mmol) was added to
ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.4194 g, 1.772
mmol) in N-methyl-2-pyrrolidone (5.91 mL) at room temperature. Then
3-fluoroazetidine hydrochloride (0.593 g, 5.32 mmol) was added and
◦
the reaction mixture was heated at 100 C in the microwave for one
hour. The reaction mixture was concentrated. The resulting residue was
purified by RP HPLC, eluting with acetonitrile:water with 0.1%
ammonium hydroxide (5:95 to 100:0), then further purified by silica gel
chromatography, eluting with methanol:ethyl acetate (0:1 to 1:9) to give
ethyl 7-(3-fluoroazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6h
(0.4561 g, 1.574 mmol, 89% yield).). 1H NMR (400 MHz, CD3SOCD3) δ
1.35 (t, J = 7 Hz, 3H), 4.25 (br dd, J = 24, 11 Hz, 2H), 4.35 (q, J = 7 Hz,
2H), 4.46–4.60 (m, 2H), 5.46–5.68 (m, 1H), 6.91 (d, J = 9 Hz, 1H), 8.25
(d, J = 9 Hz, 1H), 8.71 (s, 1H), 9.16 (s, 1H); LC-MS (LC-ES) + H = 276.
7-(Azetidin-1-yl)-1,8-naphthyridine-3-carboxylic acid 7n
7-(3-Fluoroazetidin-1-yl)-1,8-naphthyridine-3-carboxylic
acid
Lithium hydroxide (0.125 g, 5.21 mmol) was added to ethyl 7-
(azetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6g (0.4469 g, 1.737
mmol) in methanol (6.95 mL) and water (1.737 mL) at room tempera-
ture and the reaction mixture was stirred three hours at 60 ◦C. The re-
action mixture was concentrated. The reaction mixture was purified by
RP HPLC eluting with acetonitrile:water with 0.1% ammonium hy-
droxide (5:95 to 100:0), then further purified by silica gel chromatog-
raphy, eluting with methanol:ethyl acetate (1:9 to 4:1) to give 7-
(azetidin-1-yl)-1,8-naphthyridine-3-carboxylic acid 7n (0.4216 g, 1.696
mmol, 98% yield). 1H NMR (400 MHz, CD3SOCD3) δ 2.37 (p, J = 7 Hz,
2H), 4.11 (t, J = 7 Hz, 4H), 6.69 (d, J = 9 Hz, 1H), 8.03 (d, J = 9 Hz, 1H),
8.37 (s, 1H), 9.13 (s, 1H); LC-MS (LC-ES) Mꢀ H = 230.
lithium salt 7o
Lithium hydroxide (0.119 g, 4.97 mmol) was added to ethyl 7-(3-
fluoroazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6h (0.4561 g,
1.657 mmol) in methanol (6.6 mL) and water (1.7 mL) at room tem-
perature and the reaction mixture was stirred three hours at 60 ◦C. The
reaction mixture was concentrated to give 7-(3-fluoroazetidin-1-yl)-1,8-
naphthyridine-3-carboxylic acid lithium salt 7o (0.4332 g, 1.619 mmol,
98% yield). 1H NMR (400 MHz, CD3SOCD3) δ 4.17 (br dd, J = 24, 10 Hz,
2H), 4.38–4.54 (m, 2H), 5.44–5.66 (m, 1H), 6.79 (d, J = 9 Hz, 1H), 8.11
(d, J = 9 Hz, 1H), 8.44 (s, 1H), 9.18 (s, 1H); LC-MS (LC-ES) M + H = 248.
7-(Azetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-
naphthyridine-3-carboxamide 1s
24

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
7-(3-Fluoroazetidin-1-yl)-N-(trans-4-(2-hydroxypropan-2-yl)cyclo-
temperature and the reaction mixture was stirred three hours at 60 ◦C.
The reaction mixture was concentrated to give 7-(2-methylazetidin-1-
yl)-1,8-naphthyridine-3-carboxylic acid lithium salt 7p (0.3698 g,
1.404 mmol, 100% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.51 (d, J =
6 Hz, 3H), 1.94–2.06 (m, 1H), 2.42–2.54 (m, 1H), 3.94 (q, J = 8 Hz, 1H),
4.06 (q, J = 8 Hz, 1H), 4.50 (h, J = 6 Hz, 1H), 6.69 (d, J = 9 Hz, 1H), 8.03
(d, J = 9 Hz, 1H), 8.39 (s, 1H), 9.15 (s, 1H); LC-MS (LC-ES) M + H = 244.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((R)-2-methyl-
azetidin-1-yl)-1,8-naphthyridine-3-carboxamide 1u and N-(trans-4-(2-
Hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-methylazetidin-1-yl)-1,8-
naphthyridine-3-carboxamide 1v
hexyl)-1,8-naphthyridine-3-carboxamide 1t
N,N-Diisopropylethylamine (0.278 mL, 1.591 mmol) was added to 7-
(3-fluoroazetidin-1-yl)-1,8-naphthyridine-3-carboxylic acid lithium salt
7o (0.0674 g, 0.265 mmol) in N,N-dimethylformamide (0.88 mL) at
room temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(0.050 g, 0.318 mmol, PharmaBlock) was added and the reaction
mixture was stirred for five minutes. Then, n-propylphosphonic acid
anhydride (0.316 mL, 0.530 mmol) was added and the reaction mixture
was stirred for sixteen hours. The reaction mixture was concentrated.
The resulting residue was purified by RP HPLC, eluting with acetonitrile:
water with 0.1% ammonium hydroxide (5:95 to 100:0), then further
purified by silica gel chromatography, eluting with methanol:ethyl ac-
etate (0:1 to 1:4) to give 7-(3-fluoroazetidin-1-yl)-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-1,8-naphthyridine-3-carboxamide 1t
(0.0748 g, 0.184 mmol, 69.3% yield). 1H NMR (400 MHz, CD3SOCD3) δ
1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.30 (q, J = 12 Hz, 2H), 1.83 (br d, J =
11 Hz, 2H), 1.92 (br d, J = 11 Hz, 2H), 3.66–3.78 (m, 1H), 4.03 (s, 1H),
4.22 (br dd, J = 24, 11 Hz, 2H), 4.42–4.58 (m, 2H), 5.46–5.68 (m, 1H),
6.88 (d, J = 9 Hz, 1H), 8.14 (d, J = 9 Hz, 1H), 8.36 (d, J = 8 Hz, 1H), 8.56
N,N-Diisopropylethylamine (0.444 mL, 2.54 mmol) was added to 7-
(2-methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylic acid lithium
salt 7p (0.1061 g, 0.424 mmol) in N,N-dimethylformamide (1.4 mL) at
room temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(0.080 g, 0.509 mmol, PharmaBlock) was added and the reaction
mixture was stirred for five minutes. Then, n-propylphosphonic acid
anhydride (0.505 mL, 0.848 mmol) was added and the reaction mixture
was stirred for sixty-four hours. The reaction mixture was concentrated.
The resulting residue was purified by RP HPLC, eluting with acetonitrile:
water with 0.1% ammonium hydroxide (5:95 to 100:0), then further
purified by silica gel chromatography, eluting with methanol:ethyl ac-
etate (0:1 to 1:4) to give racemic N-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-carbox-
amide (0.0514 g, 0.128 mmol, 30.1% yield). 1H NMR (400 MHz,
CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.30 (q, J = 12 Hz, 2H),
1.52 (d, J = 6 Hz, 3H), 1.83 (br d, J = 12 Hz, 2H), 1.91 (br d, J = 11 Hz,
2H), 1.96–2.06 (m, 1H), 2.46–2.58 (m, 1H), 3.64–3.78 (m, 1H), 4.00 (q,
J = 8 Hz, 1H), 4.03 (s, 1H), 4.11 (q, J = 8 Hz, 1H), 4.56 (h, J = 7 Hz, 1H),
6.79 (d, J = 9 Hz, 1H), 8.07 (d, J = 9 Hz, 1H), 8.32 (d, J = 8 Hz, 1H), 8.52
(s, 1H), 9.09 (s, 1H); LC-MS (LC-ES) M + H = 383. Racemic N-(trans-4-
(2-hydroxypropan-2-yl)cyclohexyl)-7-(2-methylazetidin-1-yl)-1,8-
naphthyridine-3-carboxamide (0.1238 g, 0.324 mmol) was separated
into its enantiomers on a chiral IC column eluting with methanol:hex-
anes (3:2) with 1% diethylamine to give N-(trans-4-(2-hydroxypropan-2-
yl)cyclohexyl)-7-((R)-2-methylazetidin-1-yl)-1,8-naphthyridine-3-car-
boxamide 1u (0.028 g, 0.070 mmol, 21.49% yield) as the first diaste-
reomer (>99% ee) to elute and N-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)-7-((S)-2-methylazetidin-1-yl)-1,8-naphthyridine-3-carbox-
amide 1v (0.035 g, 0.087 mmol, 26.9% yield) as the last diastereomer to
elute (86.6% ee). The structures were assigned by vibrational circular
dichroism.
(s, 1H), 9.13 (s, 1H); LC-MS (LC-ES) C21H27FN4O2 M + H = 387; tR
=
0.48 min, 100% purity.
Ethyl 7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6i
N,N-Diisopropylethylamine (0.987 mL, 5.67 mmol) was added to
ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.3352 g, 1.416
mmol) in N-methyl-2-pyrrolidone (4.72 mL) at room temperature. Then
2-methylazetidine hydrochloride (0.457 g, 4.25 mmol) was added and
◦
the reaction mixture was heated at 100 C in the microwave for one
hour. The reaction mixture was diluted in dichloromethane, washed
with saturated sodium bicarbonate, dried over magnesium sulfate,
filtered, and concentrated. The resulting residue was purified by RP
HPLC, eluting with acetonitrile:water with 0.1% ammonium hydroxide
(5:95 to 100:0), then further purified by silica gel chromatography,
eluting with methanol:ethyl acetate (0:1 to 1:19) to give ethyl 7-(2-
methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6i (0.3815 g,
1.336 mmol, 94% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.34 (t, J = 7
Hz, 3H), 1.53 (d, J = 6 Hz, 3H), 1.94–2.06 (m, 1H), 2.46–2.60 (m, 1H),
4.03 (q, J = 7 Hz, 1H), 4.14 (q, J = 6 Hz, 1H), 4.34 (q, J = 7 Hz, 2H), 4.58
(h, J = 6 Hz, 1H), 6.81 (d, J = 9 Hz, 1H), 8.16 (d, J = 9 Hz, 1H), 8.65 (s,
1H), 9.12 (s, 1H); LC-MS (LC-ES) M + H = 272.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((R)-2-methyl-
azetidin-1-yl)-1,8-naphthyridine-3-carboxamide 1u
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H),
1.30 (q, J = 12 Hz, 2H), 1.52 (d, J = 6 Hz, 3H), 1.83 (br d, J = 12 Hz,
2H), 1.91 (br d, J = 11 Hz, 2H), 1.96–2.06 (m, 1H), 2.46–2.58 (m, 1H),
3.64–3.78 (m, 1H), 4.00 (q, J = 8 Hz, 1H), 4.02 (s, 1H), 4.11 (q, J = 8 Hz,
1H), 4.56 (h, J = 6 Hz, 1H), 6.79 (d, J = 9 Hz, 1H), 8.07 (d, J = 9 Hz, 1H),
8.31 (d, J = 8 Hz, 1H), 8.52 (s, 1H), 9.10 (s, 1H); LC-MS (LC-ES)
7-(2-Methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylic
acid
lithium salt 7p
C
22H30N4O2 M + H = 383; tR = 0.52 min, 100% purity.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-methyl-
azetidin-1-yl)-1,8-naphthyridine-3-carboxamide 1v
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H),
1.30 (q, J = 12 Hz, 2H), 1.52 (d, J = 6 Hz, 3H), 1.83 (br d, J = 11 Hz,
2H), 1.91 (br d, J = 12 Hz, 2H), 1.96–2.06 (m, 1H), 2.46–2.58 (m, 1H),
3.64–3.78 (m, 1H), 4.00 (q, J = 8 Hz, 1H), 4.02 (s, 1H), 4.11 (q, J = 8 Hz,
1H), 4.56 (h, J = 6 Hz, 1H), 6.78 (d, J = 9 Hz, 1H), 8.07 (d, J = 9 Hz, 1H),
Lithium hydroxide (0.101 g, 4.22 mmol) was added to ethyl 7-(2-
methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6i (0.3815 g,
1.406 mmol) in methanol (5.6 mL) and water (1.4 mL) at room
25

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
8.31 (d, J = 8 Hz, 1H), 8.51 (s, 1H), 9.09 (s, 1H); LC-MS (LC-ES)
2-yl)cyclohexyl)-7-((2,2,2-trifluoroethyl)amino)-1,8-naphthyridine-3-
carboxamide 1w (0.0219 g, 0.051 mmol, 24.65% yield). 1H NMR (400
MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.30 (q, J = 12 Hz,
2H), 1.83 (br d, J = 13 Hz, 2H), 1.92 (br d, J = 11 Hz, 2H), 3.66–3.78 (m,
1H), 4.02 (s, 1H), 4.38 (p, J = 8 Hz, 2H), 7.01 (d, J = 9 Hz, 1H), 8.09 (d,
J = 9 Hz, 1H), 8.17 (t, J = 7 Hz, 1H), 8.36 (d, J = 8 Hz, 1H), 8.55 (s, 1H),
9.12 (s, 1H); LC-MS (LC-ES) C20H25F3N4O2 M + H = 411; tR = 0.56 min,
98% purity.
C
22H30N4O2 M + H = 383; tR = 0.52 min, 99% purity.
Ethyl 7-((2,2,2-trifluoroethyl)amino)-1,8-naphthyridine-3-carbox-
ylate 6j
Ethyl 7-((2,2-difluoroethyl)amino)-1,8-naphthyridine-3-carboxylate
6k
N,N-Diisopropylethylamine (1.199 mL, 6.88 mmol) was added to
ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.4071 g, 1.720
mmol) in N-methyl-2-pyrrolidone (5.73 mL) at room temperature. Then
2,2,2-trifluoroethanamine (0.511 g, 5.16 mmol) was added and the re-
action mixture was heated at 100 ◦C in the microwave for five hours. The
reaction mixture was diluted in dichloromethane, washed with satu-
rated sodium bicarbonate, dried over magnesium sulfate, filtered, and
concentrated. The residue was purified by RP HPLC, eluting with
acetonitrile:water with 0.1% ammonium hydroxide (5:95 to 100:0),
then further purified by silica gel chromatography, eluting with ethyl
acetate:hexanes (2:3 to 1:0) to give ethyl 7-((2,2,2-trifluoroethyl)
amino)-1,8-naphthyridine-3-carboxylate 6j (0.1719 g, 0.546 mmol,
31.7% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.36 (t, J = 7 Hz, 3H),
4.36 (q, J = 7 Hz, 2H), 4.36–4.46 (m, 2H), 7.04 (d, J = 7 Hz, 1H), 8.20
(d, J = 9 Hz, 1H), 8.34 (br s, 1H), 8.70 (s, 1H), 9.16 (s, 1H); LC-MS (LC-
ES) M + H = 300.
N,N-Diisopropylethylamine (1.224 mL, 7.03 mmol) was added to
ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (0.4158 g, 1.757
mmol) in N-methyl-2-pyrrolidone (5.86 mL) at room temperature. Then
2,2-difluoroethanamine (0.427 g, 5.27 mmol) was added and the reac-
tion mixture was heated at 100 ◦C in the microwave for three hours. The
reaction mixture was diluted in dichloromethane, washed with satu-
rated sodium bicarbonate, dried over magnesium sulfate, filtered, and
concentrated. The resulting residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to 100:0)
to give ethyl 7-((2,2-difluoroethyl)amino)-1,8-naphthyridine-3-carbox-
ylate 6 k (0.4796 g, 1.620 mmol, 92% yield). 1H NMR (400 MHz,
CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H), 3.84–3.98 (m, 2H), 4.36 (q, J = 7
Hz, 2H), 6.24 (t, J = 56 Hz, 1H), 7.00 (d, J = 9 Hz, 1H), 8.14 (d, J = 9 Hz,
1H), 8.18 (br s, 1H), 8.66 (s, 1H), 9.14 (s, 1H); LC-MS (LC-ES) M + H =
282.
7-((2,2,2-Trifluoroethyl)amino)-1,8-naphthyridine-3-carboxylic
acid lithium salt 7q
7-((2,2-Difluoroethyl)amino)-1,8-naphthyridine-3-carboxylic acid
lithium salt 7r
Lithium hydroxide (0.041 g, 1.723 mmol) was added to ethyl 7-
((2,2,2-trifluoroethyl)amino)-1,8-naphthyridine-3-carboxylate
6j
(0.1719 g, 0.574 mmol) in methanol (2.3 mL) and water (0.57 mL) at
room temperature and the reaction mixture was stirred sixteen hours at
60 ◦C. The reaction mixture was concentrated to give 7-((2,2,2-tri-
fluoroethyl)amino)-1,8-naphthyridine-3-carboxylic acid lithium salt 7q
(0.2390 g, 0.816 mmol, 142% yield). 1H NMR (400 MHz, CD3SOCD3) δ
4.35 (p, J = 7 Hz, 2H), 6.90 (d, J = 9 Hz, 1H), 7.83 (br s, 1H), 8.02 (d, J
= 9 Hz, 1H), 8.36 (s, 1H), 9.12 (s, 1H); LC-MS (LC-ES) M + H = 272.
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-((2,2,2-tri-
Lithium hydroxide (0.123 g, 5.12 mmol) was added to ethyl 7-((2,2-
difluoroethyl)amino)-1,8-naphthyridine-3-carboxylate 6k (0.4796 g,
1.705 mmol) in methanol (6.8 mL) and water (1.7 mL) at room tem-
perature and the reaction mixture was stirred three hours at 60 ◦C. The
reaction mixture was concentrated to give 7-((2,2-difluoroethyl)amino)-
1,8-naphthyridine-3-carboxylic acid lithium salt 7r (0.4567 g, 1.668
mmol, 98% yield). 1H NMR (400 MHz, CD3SOCD3) δ 3.71 (t, J = 16 Hz,
2H), 6.53 (t, J = 57 Hz, 1H), 6.57 (d, J = 9 Hz, 1H), 7.54 (d, J = 7 Hz,
1H), 8.10 (s, 1H), 8.91 (s, 1H); LC-MS (LC-ES) M + H = 254.
fluoroethyl)amino)-1,8-naphthyridine-3-carboxamide 1w
7-((2,2-Difluoroethyl)amino)-N-(trans-4-(2-hydroxypropan-2-yl)
cyclohexyl)-1,8-naphthyridine-3-carboxamide 1x
N,N-Diisopropylethylamine (0.216 mL, 1.234 mmol) was added to 7-
((2,2,2-trifluoroethyl)amino)-1,8-naphthyridine-3-carboxylic
acid
lithium salt 7q (0.0572 g, 0.206 mmol) in N,N-dimethylformamide
(0.69 mL) at room temperature. Then, 2-(trans-4-aminocyclohexyl)
propan-2-ol 8a (0.039 g, 0.247 mmol, PharmaBlock) was added and the
reaction mixture was stirred for five minutes. Then, n-propylphosphonic
acid anhydride (0.245 mL, 0.411 mmol) was added and the reaction
mixture was stirred for sixteen hours. The reaction mixture was
concentrated. The resulting residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to
100:0), then further purified by silica gel chromatography, eluting with
methanol:ethyl acetate (0:1 to 1:4) to give N-(trans-4-(2-hydroxypropan-
N,N-Diisopropylethylamine (0.269 mL, 1.543 mmol) was added to 7-
((2,2-difluoroethyl)amino)-1,8-naphthyridine-3-carboxylic acid lithium
salt 7r (0.0669 g, 0.257 mmol) in N,N-dimethylformamide (0.86 mL) at
room temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a
(0.049 g, 0.309 mmol, PharmaBlock) was added and the reaction
mixture was stirred for five minutes. Then, n-propylphosphonic acid
anhydride (0.306 mL, 0.514 mmol) was added and the reaction mixture
26

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
was stirred for sixteen hours. The reaction mixture was concentrated.
The resulting residue was purified by RP HPLC, eluting with acetonitrile:
water with 0.1% ammonium hydroxide (5:95 to 100:0), then further
purified by silica gel chromatography, eluting with methanol:ethyl ac-
etate (0:1 to 1:4) to give 7-((2,2-difluoroethyl)amino)-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-1,8-naphthyridine-3-carboxamide 1x
(0.0611 g, 0.148 mmol, 57.5% yield). 1H NMR (400 MHz, CD3SOCD3) δ
1.04 (s, 6H), 1.04–1.24 (m, 3H), 1.31 (q, J = 11 Hz, 2H), 1.83 (br d, J =
12 Hz, 2H), 1.92 (br d, J = 12 Hz, 2H), 3.66–3.80 (m, 1H), 3.91 (br t, J =
15 Hz, 2H), 4.01 (s, 1H), 6.23 (t, J = 57 Hz, 1H), 6.98 (d, J = 9 Hz, 1H),
8.00 (br s, 1H), 8.04 (d, J = 9 Hz, 1H), 8.32 (d, J = 8 Hz, 1H), 8.52 (s,
1H), 9.10 (s, 1H); LC-MS (LC-ES) C20H26F2N4O2 M + H = 393; tR = 0.50
min, 99% purity.
N,N-Diisopropylethylamine (1.977 mL, 11.32 mmol) was added to 7-
cyclopropyl-1,8-naphthyridine-3-carboxylic acid 7s (0.6061 g, 2.83
mmol) in dichloromethane (14.15 mL) at room temperature. Then, 2-
(trans-4-aminocyclohexyl)propan-2-ol 8a (0.667 g, 4.24 mmol, Phar-
maBlock) was added and the reaction mixture was stirred for five mi-
nutes. Then, n-propylphosphonic acid anhydride (3.03 mL, 5.09 mmol)
was added and the reaction mixture was stirred for sixteen hours. The
reaction mixture was concentrated. The resulting residue was purified
by RP HPLC, eluting with acetonitrile:water with 0.1% ammonium hy-
droxide (5:95 to 100:0), then further purified by silica gel chromatog-
raphy, eluting with methanol:ethyl acetate (0:1 to 1:4) to give 7-
cyclopropyl-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-naph-
1
thyridine-3-carboxamide 1y (0.5820 g, 1.564 mmol, 55.3% yield). H
Ethyl 7-cyclopropyl-1,8-naphthyridine-3-carboxylate 6l
NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 7H), 1.32 (q,
J = 11 Hz, 2H), 1.83 (br d, J = 11 Hz, 2H), 1.93 (br d, J = 11 Hz, 2H),
2.32–2.40 (m, 1H), 3.75 (dtt, J = 8, 4, 4 Hz, 1H), 4.06 (s, 1H), 7.63 (d, J
= 9 Hz, 1H), 8.39 (d, J = 8 Hz, 1H), 8.57 (br d, J = 8 Hz, 1H), 8.78 (d, J
= 2 Hz, 1H), 9.32 (d, J = 2 Hz, 1H); 13C NMR (400 MHz, CD3SOCD3) δ
12.20 (2C), 18.41, 26.54 (2C), 27.52 (2C), 32.89 (2C), 48.41, 49.41,
70.97, 119.98, 122.73, 127.43, 137.20, 138.51, 152.13, 156.68, 164.03,
169.15; LC-MS (LC-ES) C21H27N3O2 M + H = 354, tR = 0.65 min, 100%
purity; HRMS C21H27N3O2 Calculated: M + H = 354.2171; Measured: M
+ H = 354.2182; Elemental Analysis for C21H27N3O2 C Theoretical
71.3%, Measured 71.0%; H Theoretical 7.7%, Measured 8.0%; N
Theoretical 11.9%, Measured 11.6%.
Cyclopropylboronic acid (3.16 g, 36.8 mmol) was added to a solution
of ethyl 7-chloro-1,8-naphthyridine-3-carboxylate 17c (3 g, 12.27
mmol), dicyclohexyl(2′,6′-dimethoxy-[1,1′-biphenyl]-2-yl)phosphine
(0.504 g, 1.227 mmol), tris(dibenzylideneacetone)dipalladium(0)
(2.247 g, 2.453 mmol), and 2 M sodium carbonate (15.33 mL, 30.7
mmol) in toluene (30 mL) at room temperature under nitrogen and the
reaction mixture was purged with argon for 10 min. The resulting re-
action mixture was heated to 110 ◦C and stirred for sixteen hours. On
completion, the reaction mixture was cooled, filtered through a Celite®
pad, and the filtrate was evaporated under reduced pressure. The res-
idue was purified via neutral alumina column chromatography, eluting
with ethyl acetate:petroleum ether (1:1) to afford ethyl 7-cyclopropyl-
1,8-naphthyridine-3-carboxylate 6l (700 mg, 2.58 mmol, 21% yield)
as an off white solid. 1H NMR (400 MHz, CDCl3) δ 1.12–1.23 (m, 2H),
1.43–1.53 (m, 5H), 2.25–2.36 (m, 1H), 4.48 (q, J = 7 Hz, 2H), 7.46 (d, J
= 8 Hz, 1H), 8.11 (d, J = 8 Hz, 1H), 8.78 (d, J = 2 Hz, 1H), 9.56 (d, J = 2
Hz, 1H); LC-MS (LC-ES) M + H = 243.
Ethyl 7-isopropyl-1,8-naphthyridine-3-carboxylate 6m
[1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium(II)
dichloromethane complex (0.282 g, 0.345 mmol) was added to ethyl 7-
chloro-1,8-naphthyridine-3-carboxylate 17c (0.797 g, 3.37 mmol), fol-
lowed by copper(I) iodide (0.136 g, 0.714 mmol). Then, tetrahydrofuran
(35 mL) was added and the reaction mixture was purged with nitrogen.
Isopropylzinc(II) bromide (12.0 mL, 6.00 mmol) was added and the
reaction mixture was stirred for 30 min, then quenched with ethanol and
concentrated. 1 M Potassium carbonate (25 mL) and dichloromethane
(50 mL) were added and the reaction mixture was filtered through
Celite®, extracted with dichloromethane, dried over magnesium sulfate,
filtered, and concentrated. The residue was purified via silica gel chro-
matography, eluting with methanol:dichloromethane (0:1 to 1:12) with
10% ammonium hydroxide to give impure ethyl 7-isopropyl-1,8-naph-
7-Cyclopropyl-1,8-naphthyridine-3-carboxylate 7s
Sodium hydroxide (17.58 mL, 35.2 mmol) was added to a solution of
ethyl 7-cyclopropyl-1,8-naphthyridine-3-carboxylate 6l (7.1 g, 29.3
mmol) in tetrahydrofuran (30 mL) at room temperature and reaction
mixture was stirred for fifteen hours. Then, the reaction mixture was
concentrated, acidified with diluted hydrochloric acid and the precipi-
tated solid was filtered, washed with water, and dried to give a crude
solid. This material was washed with n-pentane (500 mL) and diethyl
ether (500 mL), then dried to give 7-cyclopropyl-1,8-naphthyridine-3-
carboxylic acid 7 s (5.33 g, 23.89 mmol, 82% yield). 1H NMR (400
MHz, CD3SOCD3) δ 1.04–1.16 (m, 4H), 2.22–2.35 (m, 1H), 7.50 (d, J =
8 Hz, 1H), 8.31 (d, J = 8 Hz, 1H), 8.59 (s, 1H), 9.36 (s, 1H); LC-MS (LC-
ES) Mꢀ H = 213.
thyridine-3-carboxylate 6m (0.885 g, 108% yield) as
red–orange gum. LC-MS (LC-ES) M + H = 245.
a dark
7-Isopropyl-1,8-naphthyridine-3-carboxylic acid 7t
1 M Sodium hydroxide (5.0 mL, 5.00 mmol) was added to impure
ethyl 7-isopropyl-1,8-naphthyridine-3-carboxylate 6m (0.823 g, 3.37
mmol) in methanol (25 mL) and the reaction mixture was stirred for four
hours, then quenched with 1 M hydrochloric acid (5 mL) and concen-
trated. The residue was purified by reverse phase HPLC, eluting with
acetonitrile:water (0:1 to 1:4) with 0.1% trifluoroacetic acid to give 7-
Isopropyl-1,8-naphthyridine-3-carboxylic acid 7t (0.137 g, 0.634
mmol, 19% yield) as a tan-yellow solid. 1H NMR (400 MHz, CD3SOCD3)
δ 1.35 (d, J = 7 Hz, 6H), 3.29 (h, J = 7 Hz, 1H), 7.72 (d, J = 9 Hz, 1H),
8.59 (d, J = 9 Hz, 1H), 9.02 (d, J = 2 Hz, 1H), 9.42 (d, J = 2 Hz, 1H),
13.56 (br s, 1H); LC-MS (LC-ES) Mꢀ H = 217.
7-Cyclopropyl-N-((trans)-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-
naphthyridine-3-carboxamide 1y
N-(trans-4-(2-Hydroxypropan-2-yl)cyclohexyl)-7-isopropyl-1,8-
27

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
naphthyridine-3-carboxamide 1z
hydrochloric acid (3.2 mL) and concentrated to ~ 6 mL volume. The
solids were filtered off and rinsed with water, air-dried, and then dried
under vacuum to give 7-cyclobutyl-1,8-naphthyridine-3-carboxylic acid
7u (0.294 g, 1.288 mmol, 78% yield) as an orange-yellow powder. 1H
NMR (400 MHz, CD3SOCD3) δ 1.84–1.96 (m, 1H), 2.00–2.14 (m, 1H),
2.32–2.48 (m, 4H), 3.91 (p, J = 8 Hz, 1H), 7.59 (d, J = 8 Hz, 1H), 8.53
(d, J = 8 Hz, 1H), 8.98 (br s, 1H), 9.41 (d, J = 2 Hz, 1H), 13.59 (br s, 1H);
LC-MS (LC-ES) M + H = 229.
7-Cyclobutyl-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-
naphthyridine-3-carboxamide 1aa
2-(trans-4-Aminocyclohexyl)propan-2-ol 8a (0.110 g, 0.699 mmol,
PharmaBlock) was added to 7-isopropyl-1,8-naphthyridine-3-carboxylic
acid 7t (0.135 g, 0.624 mmol) in tetrahydrofuran (6 mL). Then, N,N-
diisopropylethylamine (0.13 mL, 0.746 mmol) was added, followed by
1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium
3-oxide hexafluorophosphate (0.286 g, 0.752 mmol) and the reaction
mixture was stirred for three hours, then quenched with 1 M aqueous
potassium carbonate:tetrahydrofuran (1:1, 10 mL). The layers were
separated, and the organic layer was washed with saturated sodium
chloride (5 mL), dried over magnesium sulfate, filtered, and concen-
trated. The residue was purified by silica gel chromatography, eluting
with ethanol:ethyl acetate (1:3):hexanes (0:1 to 1:0) to give N-(trans-4-
(2-hydroxypropan-2-yl)cyclohexyl)-7-isopropyl-1,8-naphthyridine-3-
carboxamide 1z (0.176 g, 0.495 mmol, 79% yield) as an off-white
powder. 1H NMR (400 MHz, CD3SOCD3) δ 1.05 (s, 6H), 1.08–1.24 (m,
3H), 1.34 (q, J = 12 Hz, 2H), 1.35 (d, J = 7 Hz, 6H), 1.84 (br d, J = 12
Hz, 2H), 1.94 (br d, J = 12 Hz, 2H), 3.28 (h, J = 7 Hz, 1H), 3.70–3.84 (m,
1H), 4.07 (s, 1H), 7.67 (d, J = 8 Hz, 1H), 8.48 (d, J = 8 Hz, 1H), 8.62 (d,
J = 8 Hz, 1H), 8.84 (d, J = 2 Hz, 1H), 9.38 (d, J = 2 Hz, 1H); LC-MS (LC-
ES) C21H29N3O2 M + H = 356; tR = 0.70 min, 100% purity.
2-(trans-4-Aminocyclohexyl)propan-2-ol 8a (0.040 g, 0.254 mmol,
PharmaBlock) was added to 7-cyclobutyl-1,8-naphthyridine-3-carbox-
ylic acid 7u (0.057 g, 0.250 mmol) in N,N-dimethylformamide (2.5
mL). Then, N,N-diisopropylethylamine (0.05 mL, 0.287 mmol) was
added, followed by 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo
[4,5-b]pyridinium 3-oxide hexafluorophosphate (0.117 g, 0.308 mmol)
and the reaction mixture was stirred for five hours and concentrated.
Dichloromethane and methanol were added to the residue and it was
purified via silica gel chromatography, eluting with (3:1 ethyl acetate:
ethanol):hexanes (0:1 to 7:3) to give a material which was triturated/
sonicated with ethyl acetate and the solids collected by filtration, air-
dried, then dried under vacuum overnight to give 7-cyclobutyl-N-
(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-1,8-naphthyridine-3-car-
boxamide 1aa (0.068 g, 0.185 mmol, 74.1% yield) as a cream colored
powder. 1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m,
3H), 1.32 (q, J = 12 Hz, 2H), 1.84 (br d, J = 12 Hz, 2H), 1.86–1.98 (m,
3H), 2.00–2.14 (m, 1H), 2.30–2.48 (m, 4H), 3.70–3.82 (m, 1H), 3.90 (p,
J = 9 Hz, 1H), 4.06 (s, 1H), 7.58 (d, J = 8 Hz, 1H), 8.45 (d, J = 8 Hz, 1H),
8.61 (d, J = 8 Hz, 1H), 8.82 (d, J = 2 Hz, 1H), 9.37 (d, J = 2 Hz, 1H); LC-
MS (LC-ES) C22H29N3O2 M + H = 368; tR = 0.72 min, 100% purity.
6-Bromo-3-chloro-1,8-naphthyridin-2-amine 16c
Ethyl 7-cyclobutyl-1,8-naphthyridine-3-carboxylate 6n and Cyclo-
butyl 7-cyclobutyl-1,8-naphthyridine-3-carboxylate
[1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium
(II)-
dichloromethane adduct (0.168 g, 0.206 mmol) was added to ethyl 7-
chloro-1,8-naphthyridine-3-carboxylate 17c (0.476 g, 2.011 mmol),
followed by tetrahydrofuran (20 mL) and the reaction mixture was
purged with nitrogen. Then, cyclobutylzinc(II) bromide (4.5 mL, 2.250
mmol, 0.5 M in tetrahydrofuran) was added and the reaction mixture
was heated at 60 ◦C for 90 min. Then, additional cyclobutylzinc(II)
bromide (0.8 mL, 0.400 mmol, 0.5 M in tetrahydrofuran) was added.
After 30 min, the reaction was allowed to cool to room temperature,
combined with material from another reaction and purified via silica gel
chromatography, eluting with (3:1 ethyl acetate:ethanol):hexanes (0:1
to 1:1 to 1:0) to give material that was further purified via silica gel
chromatography, eluting with (9:1 methanol:ammonium hydroxide):
dichloromethane (0:1 to 1:15) to give a mixture of ethyl 7-cyclobutyl-
1,8-naphthyridine-3-carboxylate 6n and cyclobutyl 7-cyclobutyl-1,8-
naphthyridine-3-carboxylate (0.421 g, ~1.643 mmol, ~65% combined
yield) as a tan-orange powder. LC-MS (LC-ES) M + H = 257; LC-MS (LC-
ES) M + H = 283.
Hydrogen peroxide (51.6 mL, 505 mmol, 30 wt% in water) was
added to a solution of 6-bromo-1,8-naphthyridin-2-amine 16b (15 g,
◦
63.1 mmol) in concentrated hydrochloric acid (60 mL) at 27 C in a
sealed tube. The resultant reaction mixture was stirred for 30 h at 27 ◦C.
On completion, the reaction mixture was neutralized with 50% sodium
hydroxide solution (100 mL) to pH = ~8. The precipitated solid com-
pound was filtered and dried under vacuum to give impure material,
which was purified via neutral alumina column chromatography,
eluting with methanol:dichloromethane (1:99) to afford 6-bromo-3-
chloro-1,8-naphthyridin-2-amine 16c (6.5 g, 15.7 mmol, 24.9% yield)
as an off-white solid. 1H NMR (400 MHz, CD3SOCD3) δ 7.31 (br s, 2H),
8.22–8.24 (m, 1H), 8.37 (d, J = 3 Hz, 1H), 8.76 (d, J = 3 Hz, 1H); LC-MS
(LC-ES) M + H = 258.
7-Cyclobutyl-1,8-naphthyridine-3-carboxylic acid 7u
Ethyl 7-amino-6-chloro-1,8-naphthyridine-3-carboxylate 17f
1 N Sodium hydroxide (3.2 mL, 3.20 mmol) was added to the mixed
ester ethyl/cyclobutyl 7-cyclobutyl-1,8-naphthyridine-3-carboxylate 6n
(0.421 g, 1.643 mmol) in methanol (10 mL) and the reaction mixture
was stirred for 2.5 h. Then, the reaction was quenched with 1 N
[1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium(II)
dichloromethane adduct (1.283 g, 1.571 mmol) and triethylamine (4.38
mL, 31.4 mmol) were added to a solution of 6-bromo-3-chloro-1,8-
28

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
naphthyridin-2-amine 16c (6.5 g, 15.71 mmol) in ethanol (100 mL) in a
steel bomb at 27 ◦C. The reaction mixture was stirred at 100 ◦C under a
carbon monoxide atmosphere (80 psi) for two hours. On completion, the
reaction mixture was filtered through Celite® and the filtrate was
concentrated under reduced pressure to give impure material. This
material was purified via neutral alumina column chromatography,
eluting with methanol:dichloromethane (1:19) to afford ethyl 7-amino-
6-chloro-1,8-naphthyridine-3-carboxylate 17f (2 g, 7.22 mmol, 45.9%
yield) as an off-white solid. 1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J
= 7 Hz, 3H), 4.37 (q, J = 7 Hz, 2H), 7.60 (br s, 2H), 8.45 (s, 1H), 8.68 (d,
J = 2 Hz, 1H), 9.17 (d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H = 252.
Ethyl 6-chloro-7-hydroxy-1,8-naphthyridine-3-carboxylate 17g
carboxylate 6o (0.125 g, 0.495 mmol, 82% combined yield) as a tan
1
powder. H NMR (400 MHz, CD3SOCD3) δ 3.95 (s, 3H), 4.15 (s, 3H),
8.80 (s, 1H), 8.98 (d, J = 2 Hz, 1H), 9.35 (d, J = 2 Hz, 1H); LC-MS (LC-
ES) M + H = 253.
6-Chloro-7-methoxy-1,8-naphthyridine-3-carboxylic acid 7v
1 N Sodium hydroxide (0.50 mL, 0.500 mmol) was added to the
methyl 6-chloro-7-methoxy-1,8-naphthyridine-3-carboxylate 6o (0.125
g, 0.495 mmol) in tetrahydrofuran (5 mL) at room temperature and the
reaction mixture was stirred for five hours. Then, more 1 N sodium
hydroxide (0.50 mL, 0.500 mmol) was added and the reaction mixture
was stirred for nineteen hours. The reaction mixture was acidified with
1 N hydrochloric acid (1.0 mL) and a precipitate formed. The solids were
filtered off and rinsed with water, and air-dried, then dried under vac-
uum to give 6-chloro-7-methoxy-1,8-naphthyridine-3-carboxylic acid
Aqueous sulfuric acid (~12 mL in 200 mL water) (200 mL, 400
mmol) was added dropwise to ethyl 7-amino-6-chloro-1,8-naphthyri-
dine-3-carboxylate 17f (4 g, 15.89 mmol) at 0 ◦C, followed by the
addition of 2 M sodium nitrite in water (15.9 mL, 31.8 mmol) dropwise
at 0 ◦C. The resultant reaction mixture was allowed to warm to 27 ◦C and
stirred for sixteen hours. On completion, the reaction mixture was
filtered and dried under vacuum to give an impure material. This ma-
terial was washed with diethyl ether (50 mL) and dried under vacuum to
afford ethyl 6-chloro-7-hydroxy-1,8-naphthyridine-3-carboxylate 17g
(3.8 g, 10.17 mmol, 64% yield) as an off white solid. 1H NMR (400 MHz,
CD3SOCD3) δ 1.35 (t, J = 7 Hz, 3H), 4.37 (q, J = 7 Hz, 2H), 8.49 (s, 1H),
8.68 (d, J = 2 Hz, 1H), 9.02 (d, J = 2 Hz, 1H), 13.07 (br s, 1H); LC-MS
(LC-ES) M + H = 253.
1
7v (0.097 g, 0.406 mmol, 82% yield) as a cream colored powder. H
NMR (400 MHz, CD3SOCD3) δ 4.14 (s, 3H), 8.78 (s, 1H), 8.93 (d, J = 2
Hz, 1H), 9.33 (d, J = 2 Hz, 1H), 13.58 (br s, 1H); LC-MS (LC-ES) M + H
= 239.
6-Chloro-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-7-methoxy-
1,8-naphthyridine-3-carboxamide 1ab
Ethyl 6,7-dichloro-1,8-naphthyridine-3-carboxylate 17h
2-(trans-4-Aminocyclohexyl)propan-2-ol 8a (0.061 g, 0.388 mmol,
PharmaBlock) was added to 6-chloro-7-methoxy-1,8-naphthyridine-3-
carboxylic acid 7v (0.096 g, 0.402 mmol) in N,N-dimethylformamide
(4 mL) at room temperature. Then, N,N-diisopropylethylamine (0.09
mL, 0.517 mmol) was added to the suspension, followed by 1-[bis
Phosphorous oxychloride (6.0 mL, 64.3 mmol) was added in drop-
wise to a solution of ethyl 6-chloro-7-hydroxy-1,8-naphthyridine-3-
carboxylate 17g (5.6 g, 16.07 mmol) and N,N-diisopropylethylamine
(5.61 mL, 32.1 mmol) in 1,4-dioxane (50 mL) at 0 ◦C. The resulting
reaction mixture was heated to 80 ◦C and stirred for five hours. On
completion, the reaction mixture was quenched with ice water (250 mL)
and extracted with ethyl acetate (300 mL, 2X). The combined organic
layers were washed with brine (250 mL) and evaporated under reduced
pressure to give an impure material, which was purified by neutral
alumina column chromatography, eluting with ethyl acetate:petroleum
ether (3:2) to afford ethyl 6,7-dichloro-1,8-naphthyridine-3-carboxylate
(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium
3-
oxide hexafluorophosphate (0.189 g, 0.497 mmol) and the reaction
mixture was stirred for fourteen hours. The reaction mixture was
concentrated, and the residue was purified via silica gel chromatog-
raphy, eluting with ethyl acetate:ethanol (3:1):hexanes (0:1 to 3:1).
Ethyl acetate (2 mL) was added to the solid residue and it was triturated/
sonicated, then filtered and dried to give 6-chloro-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-7-methoxy-1,8-naphthyridine-3-car-
boxamide 1ab (0.119 g, 0.315 mmol, 81% yield) as an off white powder.
1H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m, 3H), 1.31
(q, J = 12 Hz, 2H), 1.83 (br d, J = 11 Hz, 2H), 1.93 (br d, J = 10 Hz, 2H),
3.68–3.80 (m, 1H), 4.06 (s, 1H), 4.12 (s, 3H), 8.60 (d, J = 8 Hz, 1H), 8.69
(s, 1H), 8.75 (d, J = 2 Hz, 1H), 9.30 (d, J = 2 Hz, 1H); LC-MS (LC-ES)
1
17h (2.39 g, 8.37 mmol, 52.1% yield) as an off-white solid. H NMR
(400 MHz, CDCl3) δ 1.47 (t, J = 7 Hz, 3H), 4.51 (q, J = 7 Hz, 2H), 8.38 (s,
1H), 8.82 (d, J = 2 Hz, 1H), 9.64 (d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H
= 271.
C
19H24N3O3 M + H = 378; tR = 0.79 min, 100% purity.
Ethyl 7-(azetidin-1-yl)-6-chloro-1,8-naphthyridine-3-carboxylate 6p
Methyl 6-chloro-7-methoxy-1,8-naphthyridine-3-carboxylate 6o
Sodium methoxide (25% in methanol, 0.65 mL, 2.84 mmol) was
added to ethyl 6,7-dichloro-1,8-naphthyridine-3-carboxylate 17h
(0.154 g, 0.568 mmol) in methanol (10 mL) at room temperature and
stirred for twenty-four hours. It was combined with a smaller scale re-
action and concentrated to ~ 4 mL volume. The solids were filtered off
and rinsed with methanol (2 mL) then air-dried, followed by drying
under vacuum to give methyl 6-chloro-7-methoxy-1,8-naphthyridine-3-
N,N-Diisopropylethylamine (0.264 mL, 1.515 mmol) was added to
ethyl 6,7-dichloro-1,8-naphthyridine-3-carboxylate 17h (0.1027 g,
0.379 mmol) in N-methyl-2-pyrrolidone (1.263 mL) at room tempera-
ture. Then, azetidine hydrochloride (0.071 g, 0.758 mmol) was added
and the reaction mixture was heated at 100 ◦C in the microwave for one
hour. The reaction mixture was diluted in dichloromethane, washed
with saturated sodium bicarbonate, dried over magnesium sulfate,
29

R. Cadilla et al.
Bioorganic & Medicinal Chemistry 28 (2020) 115791
filtered, and concentrated. The residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to
100:0), then further purified by silica gel chromatography, eluting with
ethyl acetate:hexanes (1:4 to 4:1) to give ethyl 7-(azetidin-1-yl)-6-
chloro-1,8-naphthyridine-3-carboxylate 6p (0.0783 g, 0.255 mmol,
67.3% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.34 (t, J = 7 Hz, 3H),
2.33 (p, J = 8 Hz, 2H), 4.35 (q, J = 7 Hz, 2H), 4.38–4.54 (m, 4H), 8.40 (s,
1H), 8.68 (d, J = 2 Hz, 1H), 9.15 (d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H
= 292.
Then (S)-2-methylazetidin-1-ium ((1R,4S)-7,7-dimethyl-2-oxobicyclo
[2.2.1]heptan-1-yl)methanesulfonate63 (1.261 g, 4.16 mmol) was added
and the reaction mixture was heated at 100 ◦C in the microwave for one
hour. The reaction mixture was diluted in dichloromethane, washed
with saturated sodium bicarbonate, dried over magnesium sulfate,
filtered, and concentrated. The residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95:100:0),
then further purified by silica gel chromatography, eluting with ethyl
acetate:hexanes (1:4 to 4:1) to give ethyl (S)-6-chloro-7-(2-methyl-
azetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6q (0.7191 g, 2.234
mmol, 81% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.35 (t, J = 7 Hz,
3H), 1.50 (d, J = 6 Hz, 3H), 1.92–2.02 (m, 1H), 2.44–2.56 (m, 1H), 4.31
(dt, J = 9, 7 Hz, 1H), 4.36 (q, J = 7 Hz, 2H), 4.56 (dt, J = 9, 6 Hz, 1H),
4.80 (h, J = 6 Hz, 1H), 8.43 (s, 1H), 8.70 (d, J = 2 Hz, 1H), 9.16 (d, J = 2
Hz, 1H); LC-MS (LC-ES) M + H = 306.
Lithium 7-(azetidin-1-yl)-6-chloro-1,8-naphthyridine-3-carboxylate
7w
Lithium (S)-6-chloro-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-
carboxylate 7x
Lithium hydroxide (0.019 g, 0.805 mmol) was added to ethyl 7-
(azetidin-1-yl)-6-chloro-1,8-naphthyridine-3-carboxylate 6p (0.0783 g,
0.268 mmol) in methanol (1.1 mL) and water (0.27 mL) at room tem-
perature and the reaction mixture was stirred three hours at 60 ◦C. The
reaction mixture was concentrated to give lithium 7-(azetidin-1-yl)-6-
chloro-1,8-naphthyridine-3-carboxylate 7w (0.0653 g, 0.230 mmol,
86% yield). 1H NMR (400 MHz, CD3SOCD3) δ 2.30 (p, J = 8 Hz, 2H),
4.34 (t, J = 8 Hz, 4H), 8.27 (s, 1H), 8.40 (d, J = 2 Hz, 1H), 9.17 (d, J = 2
Hz, 1H); LC-MS (LC-ES) M + H = 264.
Lithium hydroxide (0.068 g, 2.82 mmol) was added to ethyl (S)-6-
chloro-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 6q
(0.7191 g, 2.352 mmol) in methanol (9.4 mL) and water (2.4 mL) at
room temperature and the reaction mixture was stirred sixteen hours at
45 ◦C. The reaction mixture was concentrated to give lithium (S)-6-
chloro-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-carboxylate 7x
(0.6877 g, 2.303 mmol, 98% yield). 1H NMR (400 MHz, CD3SOCD3) δ
1.45 (d, J = 6 Hz, 3H), 1.90–2.00 (m, 1H), 2.38–2.50 (m, 1H), 4.11 (dt, J
= 9, 6 Hz, 1H), 4.47 (dt, J = 9, 6 Hz, 1H), 4.74 (h, J = 8 Hz, 1H), 8.30 (s,
1H), 8.41 (d, J = 2 Hz, 1H), 9.17 (d, J = 2 Hz, 1H); LC-MS (LC-ES) M + H
= 278.
7-(Azetidin-1-yl)-6-chloro-N-(trans-4-(2-hydroxypropan-2-yl)cyclo-
hexyl)-1,8-naphthyridine-3-carboxamide 1ac
6-Chloro-N-(trans-4-(2-hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-
methylazetidin-1-yl)-1,8-naphthyridine-3-carboxamide 1ad
N,N-Diisopropylethylamine (0.169 mL, 0.969 mmol) was added to
lithium 7-(azetidin-1-yl)-6-chloro-1,8-naphthyridine-3-carboxylate 7w
(0.0653 g, 0.242 mmol) in N,N-dimethylformamide (0.81 mL) at room
temperature. Then, 2-(trans-4-aminocyclohexyl)propan-2-ol 8a (0.050
g, 0.315 mmol, PharmaBlock) was added and the reaction mixture was
stirred for five minutes. Then, n-propylphosphonic acid anhydride
(0.288 mL, 0.484 mmol) was added and the reaction mixture was stirred
for sixteen hours. The reaction mixture was concentrated. The resulting
residue was purified by RP HPLC, eluting with acetonitrile:water with
0.1% ammonium hydroxide (5:95 to 100:0), then further purified by
silica gel chromatography, eluting with methanol:ethyl acetate (0:1 to
1:4) to give 7-(azetidin-1-yl)-6-chloro-N-(trans-4-(2-hydroxypropan-2-
yl)cyclohexyl)-1,8-naphthyridine-3-carboxamide 1ac (0.0312 g, 0.074
mmol, 30.4% yield). 1H NMR (400 MHz, CD3SOCD3) δ 1.03 (s, 6H),
1.04–1.24 (m, 3H), 1.30 (q, J = 12 Hz, 2H), 1.82 (br d, J = 12 Hz, 2H),
1.91 (br d, J = 10 Hz, 2H), 2.32 (p, J = 8 Hz, 2H), 3.66–3.78 (m, 1H),
4.05 (s, 1H), 4.41 (t, J = 7 Hz, 4H), 8.30 (s, 1H), 8.43 (d, J = 8 Hz, 1H),
8.52 (d, J = 2 Hz, 1H), 9.13 (d, J = 2 Hz, 1H); LC-MS (LC-ES)
N,N-Diisopropylethylamine (0.162 mL, 0.927 mmol) was added to
lithium
(S)-6-chloro-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-
carboxylate 7x (0.0657 g, 0.232 mmol) in N,N-dimethylformamide
(0.77 mL) at room temperature. Then, 2-(trans-4-aminocyclohexyl)
propan-2-ol 8a (0.055 g, 0.347 mmol, PharmaBlock) was added and the
reaction mixture was stirred for five minutes. Then, n-propylphosphonic
acid anhydride (0.276 mL, 0.463 mmol) was added and the reaction
mixture was stirred for sixteen hours. The reaction mixture was
concentrated. The resulting residue was purified by RP HPLC, eluting
with acetonitrile:water with 0.1% ammonium hydroxide (5:95 to
100:0), then further purified by silica gel chromatography, eluting with
methanol:ethyl acetate (0:1 to 1:4) to give 6-chloro-N-(trans-4-(2-
hydroxypropan-2-yl)cyclohexyl)-7-((S)-2-methylazetidin-1-yl)-1,8-
naphthyridine-3-carboxamide 1ad (0.0560 g, 0.128 mmol, 55.1%
C
21H27ClN4O2 M + H = 403; tR = 0.59 min, 100% purity.
Ethyl (S)-6-chloro-7-(2-methylazetidin-1-yl)-1,8-naphthyridine-3-
carboxylate 6q
1
yield). H NMR (400 MHz, CD3SOCD3) δ 1.04 (s, 6H), 1.06–1.24 (m,
3H), 1.30 (q, J = 12 Hz, 2H), 1.48 (d, J = 6 Hz, 3H), 1.83 (br d, J = 11
Hz, 2H), 1.91 (br d, J = 10 Hz, 2H), 1.90–2.02 (m, 1H), 2.42–2.54 (m,
1H), 3.66–3.78 (m, 1H), 4.05 (s, 1H), 4.24 (dt, J = 9, 7 Hz, 1H), 4.53 (dt,
J = 9, 6 Hz, 1H), 4.79 (h, J = 8 Hz, 1H), 8.33 (s, 1H), 8.44 (d, J = 8 Hz,
1H), 8.54 (d, J = 2 Hz, 1H), 9.14 (d, J = 2 Hz, 1H); LC-MS (LC-ES)
N,N-Diisopropylethylamine (1.930 mL, 11.08 mmol) was added to
ethyl 6,7-dichloro-1,8-naphthyridine-3-carboxylate 17h (0.7511 g,
2.77 mmol) in N-methyl-2-pyrrolidone (5.54 mL) at room temperature.
C
22H29ClN4O2 M + H = 417; tR = 0.68 min, 100% purity.
Ethyl 6-chloro-7-cyclopropyl-1,8-naphthyridine-3-carboxylate 6r
30