Novel Hits in the Correction of δf508-Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) Protein: Synthesis, Pharmacological, and ADME Evaluation of Tetrahydropyrido[4,3-d]pyrimidines for the Potential Treatment of Cystic Fibrosis

Article abstract of DOI:10.1021/acs.jmedchem.5b00771

Cystic fibrosis (CF) is a lethal genetic disease caused by mutations of the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) with a prevalence of the δF508 mutation. Whereas the detailed mechanisms underlying disease have yet to be fully elucidated, recent breakthroughs in clinical trials have demonstrated that CFTR dysfunction can be corrected by drug-like molecules. On the basis of this success, a screening campaign was carried out, seeking new drug-like compounds able to rescue δF508-CFTR that led to the discovery of a novel series of correctors based on a tetrahydropyrido[4,3-d]pyrimidine core. These molecules proved to be soluble, cell-permeable, and active in a disease relevant functional-assay. The series was then further optimized with emphasis on biological data from multiple cell systems while keeping physicochemical properties under strict control. The pharmacological and ADME profile of this corrector series hold promise for the development of more efficacious compounds to be explored for therapeutic use in CF.

Full text of DOI:10.1021/acs.jmedchem.5b00771

Subscriber access provided by TUFTS UNIV
Article
Novel Hits in the Correction of #F508-Cystic Fibrosis
Transmembrane Conductance Regulator (CFTR) Protein: Synthesis,
Pharmacological, and ADME Evaluation of Tetrahydropyrido[4,3-
d]pyrimidines for the Potential Treatment of Cystic Fibrosis
Elisabetta Pesci, Laura Bettinetti, Paola Fanti, Luis J.V. Galietta, Salvatore La Rosa,
Letizia Magnoni, Nicoletta Pedemonte, Gian Luca Sardone, and Laura Maccari
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00771 • Publication Date (Web): 11 Nov 2015
Downloaded from http://pubs.acs.org on December 6, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Novel Hits in the Correction of ꢀF508ꢁCystic
Fibrosis Transmembrane Conductance Regulator
(CFTR) Protein: Synthesis, Pharmacological, and
ADME Evaluation of Tetrahydropyrido[4,3ꢁ
d]pyrimidines for the Potential Treatment of Cystic
Fibrosis
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Elisabetta Pesci,§ Laura Bettinetti,§ Paola Fanti,§ Luis J.V. Galietta,‡ Salvatore La Rosa,§
Letizia Magnoni,§ Nicoletta Pedemonte,‡ Gian Luca Sardone,§ Laura Maccari,§*.
§
Siena Biotech S.p.A., Strada del Petriccio e Belriguardo 35, Siena, Italy.
‡
Istituto Giovannina Gaslini, Genova, Italy
KEYWORDS. Cystic Fibrosis, Correctors, CFTR, ꢀF508ꢁCFTR, Ussing Chamber, ADME,
drugꢁlike
ABSTRACT. Cystic Fibrosis (CF) is a lethal genetic disease caused by mutations of the gene
encoding the cystic fibrosis transmembrane conductance regulator (CFTR) with a prevalence of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the ꢀF508 mutation. Whereas the detailed mechanisms underlying disease have yet to be fully
elucidated, recent breakthroughs in clinical trials have demonstrated that CFTR dysfunction can
be corrected by drugꢁlike molecules. Based on this success, a screening campaign was carried
out, seeking new drugꢁlike compounds able to rescue ꢀF508ꢁCFTR that led to the discovery of a
novel series of correctors based on a tetrahydropyrido[4,3ꢁd]pyrimidine core. These molecules
proved to be soluble, cellꢁpermeable and active in a disease relevant functionalꢁassay. The series
was then further optimized with emphasis on biological data from multiple cell systems while
keeping physicoꢁchemical properties under strict control. The pharmacological and ADME
profile of this corrector series hold promise for the development of more efficacious compounds
to be explored for therapeutic use in CF.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
Cystic Fibrosis (CF) is an autosomal recessive genetic disease with a lethal outcome, mostly
affecting the lungs together with a variety of organs throughout the body such as pancreas and
1
meconium ileus. Around 1700 different mutations in the cystic fibrosis transmembrane
2
conductance regulator (CFTR) have been identified as causing CF. CFTR is a cAMPꢁregulated
chloride channel expressed in epithelial membranes of airways, intestinal tracts, pancreas duct,
testes and sweat glands. Due to anion flow blockade, water movement slows and abnormally
1
thick mucus clogs the ducts leading to severe bacterial infections.
3
CF affects 1 in 2500 live births in caucasians, and about twoꢁthirds of all CF cases can be
4
attributed to the deletion of phenylalanine 508 (ꢀF508) in the CFTR protein. The ꢀF508 mutant
is largely unable to exit the endoplasmic reticulum, presumably due to a folding defect, and only
a small fraction of the synthesized protein reaches the plasma membrane, giving trafficking
5
,6
defect. However, when localized to the plasma membrane by pharmacological treatment or
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
lowꢁtemperature incubation, the ꢀF508 protein shows an accelerated rate of internalization and
4
an intrinsic low channel activity (gating defect) compared to wild type CFTR.
Until recently, therapies available to CF patients were not designed to correct the basic defect,
but focused on preventing and treating its consequences: the airway bacterial colonization, the
mucociliary clearance deficit, the intestinal malabsorption, and included antiꢁinflammatory,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
antibiotics, vitamins and enzymes as nutrition supplements. These agents have improved the
8
quality and duration of patients’ lives leading to a median survival rate of about 40 years.
Recently, however, the impressive growth of research around CFTR malfunctioning has started
to deliver new small molecules therapies addressing the root cause of CF by acting on the faulty
gene or faulty protein.
In recent years, a number of small molecules have progressed into clinical trials targeting
defective CFTR protein (Figure 1). Usually, the term ‘potentiators’ refers to compounds able to
interfere with the gating defect, whereas ‘correctors’ are those drugs acting on the mistrafficking
of mutated CFTR. Ivacaftor (1), initially called VXꢁ770 (Vertex Pharmaceuticals) is a potentiator
for the G551D mutation that has shown a significant improvement in patient lung function in
9
Phase III study. This drug is the first FDAꢁapproved medicine targeting the basic defect in CF,
though the patients bearing the G551D mutation, who will benefit from this therapy, represent
only 4% of the CF population. More recently (December 2014), the FDA approved Ivacaftor for
use in patients with CF with one of the following ten mutations: R117H, G551D, G178R,
S549N, S549R, G551S, G1244E, S1251N, S1255P or G1349D, expanding the number of
patients who can be treated by it.
Another investigational drug from Vertex, the corrector lumacaftor (2, VXꢁ809), acts on the
10
most common mutation ꢀF508. The molecule did not give outstanding results as a single agent
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in a Phase II trial and that formed the basis for combination studies (Phase III trial) of ivacaftor
together with lumacaftor or together with VXꢁ661 (3, Phase II trial), a second corrector
11
developed by Vertex. The combination is intended to work synergistically by first correcting
the defect and allowing CFTR to be tracked up to the surface, and then potentiating the protein
activity by prolonging channel opening. Results from Phase III trials of the combination drug for
people with two copies of the ꢀF508 mutation ages 12 and older demonstrated significant
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
improvements in lung function and other key measures of the disease, and lumacaftor in now
part of an approved drug combination (ORKAMBI™) with ivacaftor. Another compound,
ataluren (4, PTC124; PTC Therapeutics) is aimed at treating patients with the CFTR ‘nonsense
1
3
mutation’ that affects 10% of CF patients. Ataluren is in Phase III after encouraging results
showing improvements in chloride channel activity and positive trends in lung function.
Figure 1. Small molecules currently in clinical trials for CF or used as benchmark in this study.
Given these data, the identification of small molecules able to halt disease progression in the
vast majority of the CF population (i.e. carrying ꢀF508 mutation) by restoring CFTR
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
malfunction at the folding or trafficking level still represents a challenging objective. The
potentiator ivacaftor showed the success of a small molecule approach to the gating defect of
mutated CFTR. However, when targeting ꢀF508ꢁCFTR with a potentiator, only a low level of
mutant protein could be modulated at the cell surface. A preferred option would be to use a
corrector compound targeting earlier stages of malfunctioning, i.e. allowing CFTR to properly
fold and traffic to the cell surface, considering that all other liabilities (degradation, instability,
low activity) are a direct consequence of the folding defect.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bearing the above factors in mind we embarked on a drug discovery program with the aim of
identifying new diseaseꢁmodifying molecules targeting the most common ꢀF508 CFTR
mutation. In particular, we focused on compounds acting as correctors. Herein we describe the
discovery, synthesis, pharmacological and ADME characterization of a novel corrector series
based on the tethrahydropyridopyrimidine scaffold. Compounds were optimized against multiple
cell models whilst aiming to maintain a good physicoꢁchemical profile. Although moderately
potent, the new series showed promising overall profile in terms of in vitro pharmacology and
ADME properties and represent a promising starting point for lead discovery.
CHEMISTRY
Several tetrahydropyridopyrimidines were identified as ꢀF508ꢁCFTR correctors during a
screening campaign (Figure 2). The most interesting compounds were synthesized as depicted in
Scheme 1.
Scheme 1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Reagents and conditions: (a) pꢀsubstitutedꢁmethylꢁ1ꢁbenzylꢁ4ꢁoxopiperidineꢁ3ꢁcarboxylate,
amidinopyridine hydrochloride, NaOCH , EtOH, 80 °C, 2.5 h; (b) triflic anhydride, pyridine, ꢁ20
3
°
C, then room temperature 2h; (c) amine, K CO , dioxane, 90 °C, overnight; (d) POCl , toluene,
2 3 3
1
30 °C, 2 h; (e) amine, Et N, DMF, 90 °C, overnight.
3
14
The 2ꢁpyridinyltetrahydropyrido[4,3ꢁd]pyrimidine core in compound 9-14 was obtained by
thermal condensation of commercially available piperidone 5-8 in the presence of sodium
methoxide and amidinopyridine in ethanol. Activation of the hydroxyl group as a triflate using
trifluoromethanesulfonic anhydride in pyridine followed by reaction with a primary or secondary
amine afforded the corresponding 4ꢁsubstituted tetrahydropyrido[4,3ꢁd]pyrimidines 21-23, 27-
2
8, 31-35 and 41 in moderate yields. An alternative activation of the hydroxyl group was
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
identified by reaction of compounds 9-13 in the presence of phosphoryl chloride in refluxing
toluene, yielding the 4ꢁchloro derivative in good yields (compounds 19-20 and 46-48). The 4ꢁ
chloro intermediate underwent nucleophilic substitution with the appropriate amines in DMF in
the presence of triethylamine to yield the final compounds 29, 30, 36-40, and 42-45.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to explore the activity of the substituent at the 6ꢁposition (Scheme 2) of the
tetrahydropyrido[4,3ꢁd]pyrimidine core, commercially available 49 was reacted in the same
reaction sequence as described in Scheme 1 including ring closure with 2ꢁamidinopyridine to
obtain 50, followed by activation with POCl (compound 51) and subsequent nucleophilic
3
substitution with dimethylaminoethylpiperazine to give compound 52.
Scheme 2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Reagents and conditions: (a) 2ꢁamidinopyridine hydrochloride, NaOCH , EtOH, 80 °C, 2.5
3
h; (b) POCl , CH CN, 60 °C, 2 h; (c) amine, K CO , CH CN, 90 °C, overnight ; (d) H , 5%
3
3
2
3
3
2
Pd/C, EtOH, rt, 24 h, then at 50 °C, 24 h; (e) aldehyde, titanium (IV) isopropoxide, HOAc , 1,2ꢁ
dichloroethane, rt, 1h, then sodium triacetoxyborohydride (STAB), rt, overnight; (f) benzoic acid,
Oꢁ(7ꢁazabenzotriazolꢁ1ꢁyl)ꢁN,N,N',N'ꢁtetramethyluroniumꢁhexafluorophosphate
(HATU),
DIPEA, CH CN, rt, 1 h, then amine (compound 53), rt, 48 h (g) compound 53, Et N, DCM, rt,
3
3
1
0 min, then benzenesulfonyl chloride, rt 2 h.
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Derivative 52 was deprotected by catalytic hydrogenation affording the free amine 53, which
was reacted with the corresponding aldehyde under reductive conditions to afford products 54-56
or employed in the coupling with the benzoic acid and the benzenesulfonyl chloride to yield
respectively the final products 57 and 58.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
In an effort to discover small drugꢁlike molecules able to target the root cause of the disease
and to correct ꢀF508ꢁCFTR, a screening campaign was carried out on 6000 compounds. The
screening library was defined through a diverse selection approach aimed at sampling the
available chemical space in a corporate compound collection consisting of about 70,000 unique
compounds. In silico molecular properties were carefully checked and filtered as described
below.
Beside lumacaftor, which is the result of a drug discovery process that produced a clinical
1
5
16
candidate, other known correctors such as Corr4a and VRTꢁ325 (24 and 25, respectively,
Figure 1) are routinely used as reference compounds for pharmacological studies although they
hold a subꢁoptimal physicoꢁchemical profile, in particular in terms of solubility or permeability
(
see Table 1).
In order to expedite the drug discovery process and maximize the probability of finding high
quality hits, a series of filtering rules were applied to take out molecules with an undesired
profile. A ‘drugꢁlikeness’ filter was used to remove molecules with unsuitable calculated
physicochemical properties as defined by the widely applied Lipinski’s rule of five and Veber’s
1
7,18
rule.
Moreover, in order to avoid complex stereoisomers mixtures, the number of chiral
centers with defined and undefined stereochemistry was set not to exceed 3 and 1, respectively.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Molecules having experimentally determined aqueous solubility lower than 20 µM were
removed, while for those molecules without experimental solubility data, we applied an in house
generated insolubility model to filter out compounds predicted to have aqueous solubility lower
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
19
than 40 µM. Molecules containing highly reactive groups or expected to be interferent or
2
0
promiscuous were discarded. Finally, lipophilic molecules bearing more than 4 aromatic rings
and flat compounds, which are often associated with low aqueous solubility and poor
2
1
developability, were removed. Having applied these filters, a subset of about 6000 compounds
were tested according to the following screening cascade.
FRT cells coꢁexpressing ꢀF508ꢁCFTR and a fluorescent YFP halide sensor were used as the
2
2
primary screening system allowing for mode of action (corrector vs. potentiator) determination.
Our main aim was the identification of corrector compounds; however, since compounds with
23ꢁ26
both potentiator and corrector mode of action (dualꢁacting compounds) have been reported,
we opted for testing our compound set in both assay modes.
To identify correctors, compounds were tested at 2 and 20 µM in order not to lose slightly less
active compounds. Normalization was performed for each plate with respect to negative (DMSO,
0%) and positive control (24, 100%). Moreover, as corrector identification is usually challenging
we decided to use a relatively low 30% cutoff value for considering a compound worthy of
further characterization. Thus, 115 compounds acting as correctors with an efficacy higher than
30% at 2 µM or 20 µM with respect to 24 were selected. They were first clustered and 42
representative structures were characterized by concentrationꢁresponse curve determination in
ꢀF508ꢁCFTR FRT cells coꢁexpressing YFP. Thirtyꢁsix compounds were finally confirmed with
EC s ranging in [0.2, 17] µM and efficacy from 30% to 100%.
50
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In potentiator mode, a single concentration of 10 µM was used to test compounds.
Normalization was done according to negative (forskolin, 0%) and positive (genistein, 100%)
controls and a 40% cutoff value was applied to select potentiators; a total of 242 compounds
were selected for further investigation. Compounds were clustered according to their chemical
structure to select 123 representative molecules to be characterized by concentrationꢁresponse
curve determination in ꢀF508ꢁCFTR FRT cells coꢁexpressing YFP. Eightyꢁsix compounds were
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
confirmed as potentiators with K ranging from 0.1 to 18 µM and efficacy from 30% to 100%.
d
Two dualꢁacting compounds were identified and characterized by concentrationꢁresponse
curves in both corrector and potentiator assays. They were close structural analogues of VRTꢁ532
2
7
(
26, Figure 1), a tool compound known to exhibit both corrector and potentiator functions.
These observations, while confirming the validity of our screening cascade, led us to abandon the
series, which does not offer any novelty, and to focus our attention on the corrector series.
Corrector compounds with confirmed EC were evaluated in a series of early in vitro ADME
50
assays to assess aqueous solubility at pH 3 and pH 7.4, passive permeability in a Parallel
Artificial Membrane Permeability Assay (PAMPA), and stability upon incubation with
recombinant cytochrome h3A4. Compounds with an appropriate balance of activity and early
ADME properties underwent further in vitro profiling consisting of counter screening on FRT
null cells to detect unspecific activity signals. Electrophysiology was also done by Ussing
ꢁ
chamber experiments on active compounds measuring transepithelial Cl currents in FRT cells
expressing ∆F508ꢁCFTR to corroborate the data obtained using the fluorescent assay. At this
point, the tetrahydropyrido[4,3ꢁd]pyrimidine series was identified as ꢀF508ꢁCFTR correctors
with an overall promising profile for both pharmacological and physicoꢁchemical parameters
(Figure 2 and Table 1).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Hit series identified in screening campaign
Table 1. In vitro activity data in FRT cells and ADME profile for compounds 21ꢁ23 and known
corrector compounds 24, 25, and 2
d
e
Compound
ClogP MPSA ꢀF508ꢁ
WTꢁ
CFTR
Ussing
Sol
Perm
Met
Stab
%
a
b
f
CFTR
ꢀF508ꢁ
µM
2
c
Å
EC50 µM Emax % CFTR
(Emax %)
Emax %
21
22
23
24
2.8
1.7
0.9
6.2
65.4
88.2
91.5
132.6
2.8
1.4
2.8
1.1
(
(
(
(
62.9)
56.6)
57.9)
97.8
102.9
94.0
68.3
247
250
207
1
High
High
High
No
25
71
57
55
74.6
30.1
113.8) 115.3
100
result
2
5
5.9
5.7
93.2
97.8
2.2
2.6
(
(
92.9)
66.7)
Not
tested
41.4
75
1
No
13
95
result
2
Not
126
Mediu
m
tested
a
Values for EC and Emax are obtained from dose–response relationships of YFP fluorescence
5
0
b
experiments. Data are normalized according to DMSO (100%), all compounds are tested at 10
µM. Data reported for biological assays are expressed as mean values from 2 or 3 independent
experiments. Data are normalized according to DMSO (0%) and 24 at 10 µM (100%), 25 tested
c
d
at 10 µM and our compounds were tested at 15 µM. Solubilities were determined at pH 7.4 at
e
pseudothermodynamic equilibrium. Permeability is based on measuring the permeation rate of
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
f
the test article through an artificial membrane. Metabolic stability was determined as percentage
remaining after incubation for 1 h with recombinant hCYP3A4.
The three compounds were found to have micromolar corrector activity on FRT cells carrying
the ꢀF508ꢁCFTR mutation. Despite being less efficacious with respect to the reference
compounds, they showed a promising physicoꢁchemical profile, being highly soluble and
permeable. Metabolic stability versus Cyp3A4 was generally good; only compound 21 showed
some liability that was viewed as acceptable as a hit of a screening campaign.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ussing chamber experiment on ꢀF508ꢁCFTR FRT epithelia further confirmed activity of
compounds 21-23, with compounds 21 and 22 being able to significantly decrease the CFTRꢁ
ꢁ
dependent Cl current.
Finally, compounds 21-23 did not exert any effect on cells expressing wildꢁtype CFTR thus
confirming their specificity of action on malfunctioning CFTR cell system.
All these data were rather encouraging and the series was deemed to be worthy of further
development. At that point we turned our attention on having compounds acting on multiple cell
models.
It has been reported that correctors may be strongly affected by the cell background and many
2
8
correctors have been found to work in one cell type only. Two different modes of action have
been proposed for correctors where they could act as pharmacological chaperones able to
directly bind to CFTR and allow its proper folding, or proteostasis regulators which modulate
other proteins involved in CFTR maturation. While the former hypothesis implies that correctors
should work in any cell type, the latter paves the way to the possibility of having cellꢁtype
dependent correctors. In order to identify compounds with a more general mechanism of action
and avoid cellꢁtype specific compounds, we introduced testing in ꢀF508ꢁCFTR transfected A549
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cells, which are of human lung epithelial origin, as a second in vitro pharmacologic tool for
further characterizing correctors.
As shown in Table 2, two out of three original hits showed some activity on A549 cells.
Although the efficacy was very partial, these data further prompted us to deepen the investigation
on the series. Medicinal chemistry efforts were thus focused on generating compounds with
significant biological activity in both FRT and A549 cell systems while maintaining the optimal
physicoꢁchemical profile. Compounds were deemed of interest when they possessed an efficacy
higher than 30% with respect to 24 in both cell lines.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In the initial phase of structureꢁactivity relationship (SAR) study on the series, we identified
three main regions to explore corresponding to the pyridine ring in the 2ꢁposition (A), the amine
in 4ꢁposition (B) and the Nꢁsubstitution on 6ꢁposition (C) of the tetrahydropyrido[4,3ꢁ
d]pyrimidine core (Figure 3).
Figure 3. The three regions explored in our SAR studies.
On the basis of results obtained from the original three hit compounds (21, 22, 23) the first set
of analogues synthesized explored the effect of the pyridine ring on position 2ꢁ of the core
structure. Table 2 contains examples of synthesized compounds bearing a benzyl group in
position 6ꢁ and various amines in 4ꢁposition of the tetrahydropyrido[4,3ꢁd]pyrimidine core.
Table 2. In vitro Activity data and ADME profile for compounds 21ꢁ23, 27-34.
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
R1
N
N
N
R2
a
b
c
d
Cpd
R1
R2
ClogP MPSA
ꢀF508ꢁCFTR EC
µM (Emax %)
Sol Perm Met Stab
µM
50
2
%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Å
FRT
A549
OH
21
2ꢁpyridine 2.8
65.4
2.8 (62.9) 0.4 (30.6) 247 High
1.4 (56.6) 7.1 (34.4) 250 High
25
N
N
NH2
2
2
3
2ꢁpyridine 1.7
2ꢁpyridine 0.9
88.2
91.5
71
57
O
2
2.8 (57.9) inactive
207 High
N
N
NH2
O
H
2ꢁpyridine 2.4
2ꢁpyridine 3.6
60.4
51.6
3.1 (53.4) inactive
1.5 (41.3) 11.7 (57)
180 High
218 High
100
100
2
7
8
N
N
N
2
N
N
N
NH2
2
9
0
3ꢁpyridine 1.4
3ꢁpyridine 0.7
88.2
91.5
inactive
inactive
inactive
inactive
224 High
156 High
22
17
N
N
O
3
N
NH2
O
H
3ꢁpyridine 2.2
3ꢁpyridine 3.4
60.4
51.6
inactive
inactive
inactive
inactive
234 High
247 High
15
41
3
1
2
N
N
N
3
N
N
N
H
4ꢁpyridine 2.2
4ꢁpyridine 3.4
60.4
51.6
inactive
inactive
inactive
inactive
250 High
250 High
13
21
33
N
N
N
34
N
N
N
a
Values for EC and Emax are obtained from dose–response relationships of YFP fluorescence
experiments. Data reported for biological assays are expressed as mean values from 2 or 3
5
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
b
independent experiments.. Solubilities were determined at pH 7.4 at pseudothermodynamic
c
equilibrium. Permeability is based on measuring the permeation rate of the test article through
an artificial membrane. Metabolic stability was determined as percentage remaining after
incubation for 1 h with recombinant hCYP3A4.
d
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Two of the newly synthesized compounds (27 and 28) proved active in the FTR cell assay with
efficacy ranging from 41 to 53%. They were all characterized by a 2ꢁpyridine substituent.
Among these, only compound 28 bearing the dimethylaminoethylpiperazine substituent in 4ꢁ
position was active on both cell lines, with higher efficacy on A549 (57%) with respect to FRT
cells.
In general, the molecules had excellent solubility and passive permeability while metabolic
stability towards cytochrome 3A4 was more variable. 3ꢁPyridyl and 4ꢁpyridyl derivatives proved
inferior to the 2ꢁpyridyl analogues not only for biological activity but also for metabolic stability
as clearly showed by going from 28 to 32 and 34 or from 27 to 31 and 33. Taken together, these
data indicate the 2ꢁpyridyl substitution is essential for both activity and ADME properties.
With the aim of a multiparameter optimization of the molecules, we then decided to: i.
investigate the metabolic stability by inserting simple substitutions on 4ꢁposition of the benzyl
group, which is a known labile site, while keeping the 2ꢁpyridyl substituent fixed, and ii. expand
the initial selection of amines on 4ꢁposition of the pyridopirimidine core (Table 3).
Table 3. In vitro activity data and ADME profile for compounds 35ꢁ40.
R2
N
N
N
N
R1
a
b
c
Cpd R1
R2
ClogP MPSA ꢀF508ꢁCFTR EC50 µM Sol Perm
Met
Stab
d
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
Å
(Emax %)
FRT
µM
%
A549
3
5
6
ꢁOCH
3
3.5
3.8
60.9 inactive
51.6 inactive
77.9 0.4
inactive
229 High
231 High
124 High
70
82
97
N
N
N
N
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ꢁF
inactive
inactive
N
3
7
8
ꢁOCH
ꢁOCH
ꢁOCH
ꢁOCH
3
3
3
3
2.5
1.6
3.1
0.7
N
N
N
OH
(
106.9)
NH2
3
97.5 3.5
(
(
(
86.6) inactive
40.6) inactive
35.9) 10.9
224 Medium 78
O
39
81.0 0.2
100.7 2.3
217 High
28
H
N
N
N
40
217 Medium 79
N
N
NH2
(36.8)
O
a
Values for EC and Emax are obtained from dose–response relationships of YFP fluorescence
experiments. Data reported for biological assays are expressed as mean values from 2 or 3
independent experiments.. Solubilities were determined at pH 7.4 at pseudothermodynamic
equilibrium. Permeability is based on measuring the permeation rate of the test article through
an artificial membrane. Metabolic stability was determined as percentage remaining after
incubation for 1 h with recombinant hCYP3A4.
5
0
b
c
d
The pꢁOMe substituent gave the highest activity values observed on the FRT cell line with
compounds 37 and 38 showing efficacy values >85%, but were completely inactive on the A549
cell line. One example (compound 40) showed weak activity on both cell lines with values
around 35%. The introduction of the pꢁOMe substituent also exerted some effect on the physicoꢁ
chemical profile of the molecules: compounds 38 and 40, direct analogues of primary hits 22 and
2
3, benefited from the paraꢁsubstitution in terms of stability versus cytochrome 3A4. On the
other hand, passive permeability was not optimal for compounds 38 and 40: this trend was in line
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with ClogP and MPSA values in this array of molecules, i.e. the compounds with lower
permeability were characterized by a lower ClogP and higher MPSA values.
Maintaining activity across the series on the A549 cell line proved to be more difficult than
attaining consistent response of this class of compounds on the FRT cell line. In fact, in all cases
compounds active on the A549 cell line showed activity also on the FRT cells. This trend
encouraged us to select the dimethylaminoethylpiperazine substituent, which showed the highest
level of activity on the A549 cell line (compound 28), and explore the left handside of the
molecule, switching from benzyl to different type of substituents on 6ꢁposition (Table 4).
Unexpectedly, none of the compounds synthesized in this mini array showed activity on the
A549 cell line, whereas some of these maintained activity on FRT.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 4. In vitro activity data and ADME profile for compounds 54ꢁ58.
N
N
N
R1
N
N
N
N
a
b
c
Cpd R1
ClogP MPS
ꢀF508ꢁCFTR EC50 µM Sol
Emax %)
Perm
Met
Stab
%
2
d
(
A Å
µM
FRT
A549
3
.3
79.9
51.6
0.2
(
79.8)
57.2)
inactive
235 High
186 High
66
58
5
4
S
55
5.0
0.3
(
inactive
ACS Paragon Plus Environment

Page 19 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
5
5
6
7
8
4.5
2.7
3.0
51.6
68.7
94.2
0.9
(
41)
inactive
inactive
inactive
191 High
65
inactive
inactive
244 Mediu
m
100
84
O
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
S
195 High
a
Values for EC and Emax are obtained from dose–response relationships of YFP fluorescence
experiments. Data reported for biological assays are expressed as mean values from 2 or 3
independent experiments.. Solubilities were determined at pH 7.4 at pseudothermodynamic
5
0
b
c
equilibrium. Permeability is based on measuring the permeation rate of the test article through
d
an artificial membrane. Metabolic stability was determined as percentage remaining after
incubation for 1 h with recombinant hCYP3A4.
This set of molecules completed the preliminary SAR around the scaffold: 2ꢁpyridyl
substitution on position 2ꢁ of the tetrahydropyrido[4,3ꢁd]pyrimidine core and a substituted benzyl
group in position 6ꢁ proved almost essential to find activity on both cell lines. The amine on
position 4ꢁ of the core structure could accommodate various groups which were active mostly on
FRT cells, but with the best result in both cell lines obtained only by cyclic or linear aliphatic
amines with hydrogen bond acceptor/donor groups located at the far end of the moiety.
With these observations in mind, we proceeded to the synthesis of a focused set of compounds,
selecting pꢁsubstituted benzyl groups at position 6 and a new set of amines in position 4. Indeed
in this table (Table 5) we found a more robust set of results with three compounds (43-45)
showing activity on both cell lines with EC s in the lowꢁtoꢁsubmicromolar range. In particular
5
0
compound 43 showed the best combination of activity observed among this series both in terms
of EC (0.7 ꢂM and 1.9 ꢂM on FTR and A549 respectively) and Emax values (>60% on both
50
assays). Compounds 44 and 45 showed a lower activity but an improved metabolic stability with
respect to 43.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 5. In vitro activity data and ADME profile for compounds 41ꢁ45.
R2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N
N
N
N
R1
a
b
Cpd R1
R2
Clog MPSA ꢀF508ꢁCFTR
EC50 Sol cPerm Met
2
d
µM (Emax %)
P
Å
µM
Stab
%
FRT
0.3
A549
36.9) inactive
41
OCH
3
3.0
3.8
2.4
74.7
65.5
74.3
(
218 High 61
N
N
N
O
4
2
3
Cl
inactive
inactive
223 High
3
N
N
O
4
Cl
H
N
0.7
(
65.6) 1.9
73.8) 0.4
(
61.7) 198 High 25
O
O
44
F
F
H
N
2.7
4.4
83.0
57.2
0.2
0.2
(
(
(
(
44.3) 226 High 80
40.6) 196 High 71
N
H
H
N
45
40)
0.5
N
a
Values for EC and Emax are obtained from dose–response relationships of YFP fluorescence
5
0
experiments. Data reported for biological assays are expressed as mean values from 2 or 3
b
independent experiments.. Solubilities were determined at pH 7.4 at pseudothermodynamic
c
equilibrium. Permeability is based on measuring the permeation rate of the test article through
d
an artificial membrane. Metabolic stability was determined as percentage remaining after
incubation for 1 h with recombinant hCYP3A4.
In order to further investigate the extent of metabolic stability of the compounds, intrinsic
clearance (CL ) was determined for the most promising correctors (22, 27, 43).
int
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vitro measurement of CL in human hepatic microsomes was used to estimate the stability
int
towards CYP450ꢁdependent metabolism. Results are reported in Table 6 together with additional
data for a summary of the ADME in vitro profile including permeability in the MDCK cell line
and plasma stability.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 6. ADME in vitro profile of most interesting compounds.
a
b
d
Cpd
CLint
MDCK AB
MDCK
BA/AB
Plasma Stability 3 h; rat
c
µL/min/m
g
ꢁ6
1
0 cm/s
22
36.5
30.1
97.5
0.4
1.8
3.1
52.3
18.1
12.1
85
2
7
3
101
4
93.2
a
b
In vitro intrinsic clearance from human liver microsomal stability assay. MSꢁbased
quantification of apical → basolateral transfer rate of a test compound at 10 ꢂM across
contiguous monolayers of MDCK (Madin−Darby canine kidney) cells. Ratio of (basolateral →
c
apical) to (apical → basolateral) transfer rate of a test compound at 10 ꢂM across contiguous
d
monolayers of MDCK cells. Plasma stability was determined as percentage remaining after
incubation for 3 h with fresh rat plasma.
Intrinsic clearance in the presence of human hepatic microsomes for compounds 22 and 27,
although not optimal, was considered acceptable (ca. 30 µL/min/mg). Compound 43, confirmed
the low metabolic data observed with CYP3A4 also in the intrinsic clearance assay. Interestingly,
while no particular liabilities were highlighted in terms of passive permeability, data from an
MDCK cell line pointed out a poor ability of compounds to enter cellular barriers (MDCK AB)
and a clear indication of a role for efflux transporters (MDCK BA/AB > 3). Improving active
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
efflux for the series will be one of the major goals for the next round of optimization. Finally,
stability towards the hydrolysis by plasma circulating enzymes appears to be optimal.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chemistry
General Methods. Nuclear magnetic resonance spectra were recorded using a Varian Mercury
Plus 400 MHz spectrometer equipped with a PFG ATB broadband probe. UPLCꢁMS analyses
were run using a Acquity Waters UPLC equipped with a Waters SQD (ES ionization) and Waters
Acquity PDA detector, using a column BEH C18 1.7 ꢂm, 2.1 mm × 50 mm. Basic method:
gradients were run using 0.1% NH HCO water/acetonitrile 95/5 and acetonitrile with a gradient
4
3
95/5 to 15/85, flow: 0.8 mL/min over 3 min. Acid method: gradients were run using 0.05%
formic acid water/acetonitrile 95/5 and acetonitrile with a gradient 95/5 to 100, flow: 0.8 mL/min
over 3 min as stated in the examples. Retention times were expressed in minutes. Temperature:
4
0 °C. UV detection at 215 nm and 254. ESI+ detection in the 80−1000 m/z range. HPLCꢁMS
analysis were run using a Waters 2795 separation module equipped with a Waters Micromass ZQ
ES ionization) and Waters PDA 2996, using a XꢁBridge C18 3.5 ꢂm 2.10 x 50 mm column.
Basic method: gradients were run using 0.1% ammonia/water and acetonitrile with gradient
5/15 to 5/95, flow 0.8 ml/min over 5/10 minutes. Acid method: gradients were run using 0.1%
(
8
formic acid/water and 0.1% formic acid/acetonitrile with gradient 95/15 to 5/95, flow 0.8 ml/min
over 5/10 minutes. Temperature: 40 °C. UV Detection at 215 nm and 254. ESI+ detection in the
80ꢁ1000 m/z range. Preparative HPLC was run using a Waters 2767 system with a binary
gradient module Waters 2525 pump and coupled to a Waters Micromass ZQ (ES) or Waters 2487
DAD, using a XꢁBridge C18 5 ꢂm, 19 mm × 150 mm, using a 0.1% formic acid/ water and 0.1%
formic acid/methanol flow: 17 mL/min. Alternatively in basic condition it was used 0.1%
ACS Paragon Plus Environment

Page 23 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ammonia/water and methanol flow: 17 mL/min. The purity of compounds submitted for
screening was > 95% as determined by integrating at 215 nm the peak area of the LC
chromatograms. To further support the purity statement, all compounds were also analyzed at a
different wavelengths (254 nm), and total ion current (TIC) chromatogram and NMR spectra
were used to further substantiate results. HRMS data were obtained through an LTQꢁOrbitrap
mass spectrometer via direct infusion of a 1 ꢂM sample in acetonitrile 30%/water 69.9%/formic
acid 0.1% (flow: 7 ꢂL/min). The LTQꢁOrbitrap instrument was calibrated prior to analysis with
the manufacturer’s positive ion Calibration Solution in the mass range 100−2000 m/z. Each
determination was the combination of five successive scans acquired at 100.000 fwhm
resolution. Elemental composition calculations and isotopic pattern simulations were executed
using the specific tool included in the QualBrowser module of Xcalibur (Thermofisher
Scientific, release 2.0.7) software, using a tolerance of 5 ppm. All TLC analyses were performed
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
on silica gel (Merck 60 F254) and spots revealed by UV visualization at 254 nm and KMnO or
4
ninhydrin stain.
General procedure A for the synthesis of 6-(p-substituted-benzyl)-2-pyridin-5,6,7,8-
tetrahydropyrido[4,3-d]pyrimidin-4-ols (9-14 and 50). To a solution of pꢁsubstitutedꢁ1ꢁ
benzylꢁ4ꢁoxopiperidineꢁ3ꢁcarboxylic acid methyl ester (3.18 g, 12 mmol, 1 equiv) in EtOH (50
mL) was added NaOMe (24 mmol, 2 equiv) followed by amidinopyridine hydrochloride (12
mmol, 1 equiv). The reaction mixture was heated to 80 °C for 2.5 h, then cooled to rt. The
solvent was removed under reduced pressure, and a saturated solution of ammonium chloride
was added to the residue. The insoluble material was filtered off, washed with H O (60 mL) and
2
then ether (20 mL) and concentrated in vacuo to give a solid, which was not further purified.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
-Benzyl-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (9). 5 was prepared
according to general procedure A, using 5 (1.8 g, 7.37 mmol) and 2ꢁamidinopyridine
hydrochloride (1.16 g, 7.37 mmol) to afford 6ꢁbenzylꢁ2ꢁpyridinꢁ2ꢁylꢁ5,6,7,8ꢁ
tetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol (9) as an orange solid (1.54 g, yield: 66%) which was
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
used in the next step without further purification.
1
H NMR (400 MHz, d ꢁDMSO) δ 8.74 – 8.67 (m, 1H), 8.29 – 8.24 (m, 1H), 8.07 – 7.96 (m,
6
1H), 7.66 – 7.57 (m, 1H), 7.39 – 7.32 (m, 4H), 7.32 – 7.23 (m, 1H), 3.70 (s, 2H), 3.27 (s, 2H),
2
.74 (s, 4H).
Mass (ES) m/z 319 (M+1). HPLC (acid method, 3 min); R = 1.44 min.
t
6-(4-Methoxybenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (10).
1
0 was prepared according to general procedure A, using 6 (4.3 g, 15.5 mmol) and 2ꢁ
amidinopyridine hydrochloride (12.44 g, 15.5 mmol) to afford 6ꢁ(4ꢁmethoxybenzyl)ꢁ2ꢁpyridinꢁ2ꢁ
ylꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol (10) as an orange solid (5.3 g, yield: 98%)
which was used in the next step without further purification.
1
H NMR (400 MHz, d ꢁDMSO) δ 11.95 (s, 1H), 8.75 – 8.65 (m, 1H), 8.32 – 8.22 (m, 1H), 8.08
6
–
7.95 (m, 1H), 7.67 – 7.57 (m, 1H), 7.27 (d, J = 8.6 Hz, 2H), 6.91 (d, J = 8.6 Hz, 2H), 3.75 (s,
3H), 3.62 (s, 2H), 3.24 (s, 2H), 2.72 (s, 4H).
Mass (ES) m/z 349 (M+1).
6-(4-Fluorobenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (11). 11
was prepared according to general procedure A, using 7 (3.18 g, 12 mmol) and 2ꢁ
ACS Paragon Plus Environment

Page 25 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amidinopyridine hydrochloride (1.89 g, 12 mmol) to afford 6ꢁ(4ꢁfluorobenzyl)ꢁ2ꢁpyridinꢁ2ꢁylꢁ
5,6,7,8ꢁtetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol (11) as a brown solid (2.23 g, yield: 56%) which
was used in the next step without further purification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
H NMR (400 MHz, d ꢁDMSO) δ 11.32 – 10.57 (m, 1H), 8.74 – 8.69 (m, 1H), 8.32 – 8.25 (m,
6
1H), 8.06 – 7.97 (m, 1H), 7.64 – 7.57 (m, 1H), 7.45 – 7.34 (m, 2H), 7.24 – 7.10 (m, 2H), 3.68 (s,
2
H), 3.25 (s, 2H), 2.73 (s, 4H).
Mass (ES) m/z 337 (M+1).
6
-(4-Chlorobenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (12). 12
was prepared according to general procedure A, using 8 (1.4 g, 4.98 mmol) and 2ꢁ
amidinopyridine hydrochloride (785 mg, 4.98 mmol) to afford 6ꢁ(4ꢁchlorobenzyl)ꢁ2ꢁpyridinꢁ2ꢁ
ylꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol (12) as solid (1.4 g, yield: 80%) which was
used in the next step without further purification.
1
H NMR (400 MHz, d ꢁDMSO) δ 11.69 (s, 1H), 8.77 – 8.66 (m, 1H), 8.36 – 8.21 (m, 1H), 8.13
6
–
7.94 (m, 1H), 7.72 – 7.57 (m, 1H), 7.49 – 7.30 (m, 4H), 3.69 (s, 2H), 3.27 (s, 2H), 2.73 (s, 4H).
Mass (ES) m/z 353 (M+1).
6
-Benzyl-2-pyridin-3-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (13). 13 was
prepared according to general procedure A, using 5 (1.5 g, 6 mmol) and 3ꢁamidinopyridine
hydrochloride (950 mg, 6 mmol) to afford 6ꢁbenzylꢁ2ꢁpyridinꢁ3ꢁylꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁ
d]pyrimidinꢁ4ꢁol (13) as orange solid (1.46 g, yield: 76%) which was used in the next step
without any further purification.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
H NMR (400 MHz, d ꢁDMSO) δ 11.69 (s, 1H), 9.62 – 9.55 (m, 1H), 8.72 – 8.68 (m, 1H), 8.68
6
–
8.61 (m, 2H), 7.47 – 7.28 (m, 5H), 3.69 (s, 2H), 3.27 (s, 2H), 2.73 (s, 4H).
Mass (ES) m/z 319 (M+1). UPLC (basic method); R = 0.19 min.
t
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6-Benzyl-2-pyridin-4-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-ol (14). 14 was
prepared according to general procedure A, using 5 (1.5 g, 6 mmol) and 4ꢁamidinopyridine
hydrochloride (950 mg, 6 mmol) to afford 6ꢁbenzylꢁ2ꢁpyridinꢁ4ꢁylꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁ
d]pyrimidinꢁ4ꢁol (14) as orange solid (1.9 g, yield: 100%) which was used in the next step
without any further purification.
1
H NMR (400 MHz, d ꢁDMSO) δ 8.65 – 8.57 (m, 2H), 8.12 – 8.05 (m, 2H), 7.41 – 7.14 (m,
6
5H), 3.73 – 3.61 (m, 2H), 3.60 – 3.33 (m, 2H), 2.77 – 2.57 (m, 4H).
Mass (ES) m/z 319 (M+1). UPLC (basic method); R = 0.30 min.
t
General procedure B for the synthesis of trifluoro-methanesulfonic acid 6-(p-
substitutedbenzyl)-2-pyridin-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidin-4-yl esters (15-18).
To a solution of 6ꢁ(pꢁsubstitutedbenzyl)ꢁ2ꢁpyridinꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol
(
1 equiv, 2.38 mmol) in pyridine (20ml) was added trifluoromethanesulfonic anhydride (1 equiv,
0
2
.85 mmol) dropwise at ꢁ20 C. The reaction mixture was slowly warmed to room temperature
and then stirred for 2 h. The reaction was poured into water and the solution was filtered. A
brown oil (lowꢁmelting wax) was recovered which was dissolved in CH OH and evaporated
3
under vacuum to give the desired product.
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Trifluoromethanesulfonic acid 6-benzyl-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-
d]pyrimidin-4-yl ester (15). Prepared following general procedure B using 9 (2 g, 6.28 mmol)
and trifluoromethanesulfonic anhydride (1.26 mL, 7.53 mmol). The product 15 (1.9 g, yield:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
7%) was obtained as a brown solid by precipitation from the reaction mixture by adding H O
2
and was used in the next step without further purification.
Mass (ES) m/z 481 (M+1). UPLC (basic method); R = 1.61 min.
t
Trifluoromethanesulfonic
tetrahydropyrido[4,3-d]pyrimidin-4-yl ester (16).
6 was prepared following general procedureB using10 (4.8 g, 13.8 mmol) and
acid
6-(4-methoxybenzyl)-2-pyridin-2-yl-5,6,7,8-
1
trifluoromethanesulfonic anhydride (2.78 mL, 16.56 mmol). The product 16 (1.8 g, yield: 30%)
was obtained as a brown solid by precipitation from the reaction mixture by adding H O and was
2
used in the next step without further purification.
Mass (ES) m/z 481 (M+1).
Trifluoromethanesulfonic acid 6-benzyl-2-pyridin-3-yl-5,6,7,8-tetrahydropyrido[4,3-
d]pyrimidin-4-yl ester (17). 17 was prepared following general procedure B using 13 (1.9 g, 6
mmol) and trifluoromethanesulfonic anhydride (1.21 mL, 7.2 mmol). The product 17 (2.1 g,
yield: 78%) was obtained as a brown solid by precipitation from the reaction mixture by adding
H O and was used in the next step without further purification.
2
Mass (ES) m/z 451 (M+1). UPLC (basic method); R = 1.64 min.
t
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Trifluoromethanesulfonic acid 6-benzyl-2-pyridin-4-yl-5,6,7,8-tetrahydropyrido[4,3-
d]pyrimidin-4-yl ester (18). 18 was prepared following general procedureB using 14 (1.4 g, 4.4
mmol) and trifluoromethanesulfonic anhydride (0.9 mL, 5.28 mmol). The product 18 (1 g, yield:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
1%) was obtained as a brown solid by precipitation from the reaction mixture by adding H O
2
and was used in the next step without further purification.
Mass (ES) m/z 451 (M+1). UPLC (basic method); R = 2.09 min.
t
General procedure C for the synthesis of 6-(p-substitutedbenzyl)-4-chloro-2-pyridin-
5,6,7,8-tetrahydropyrido[4,3-d]pyrimidines (19-20, 46-48 and 51). To a solution of 6ꢁ(pꢀ
substitutedbenzyl)ꢁ2ꢁpyridinꢁ5,6,7,8ꢁtetrahydropyrido[4,3ꢁd]pyrimidinꢁ4ꢁol (1 equiv, 0.9 mmol,
300 mg) in toluene, phosphoryl chloride (2 equiv, 1.8 mmol, 0.16 µL) was added and the mixture
was heated at reflux for 2 h. The reaction was allowed to cool to room temperature, the mixture
was poured into water and the excess of POCl was quenched with 1 N NaOH. The product was
3
extracted into DCM (3 x 10 mL) and the combined organic layers were dried over Na SO .
2
4
Evaporation of volatiles provided the desired product, which was used in the next step without
any further purification.
4-Chloro-6-(4-fluorobenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine
(19). By use of general procedure C, reaction of 11 (1.18 g, 3.5 mmol) with phosphoryl chloride
(0.64 mL, 7 mmol) gave the chlorinated derivative (850 mg, yield: 69%) as a brown solid.
1
H NMR (400 MHz, d ꢁDMSO) δ 8.78 – 8.70 (m, 1H), 8.38 – 8.30 (m, 1H), 8.03 – 7.94 (m,
6
1
H), 7.59 – 7.52 (m, 1H), 7.49 – 7.38 (m, 2H), 7.26 – 7.14 (m, 2H), 3.78 (s, 2H), 3.64 (s, 2H),
.02 (t, J = 5.8 Hz, 2H), 2.84 (t, J = 5.8 Hz, 2H).
3
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Mass (ES) m/z 355 (M+1).
4-Chloro-6-(4-chlorobenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
20). Prepared according to general procedure C, by reaction of 12 (300 mg, 0.85 mmol) with
phosphoryl chloride (0.15 mL,1.7 mmol) to give the title compound 20 (300 mg, yield: 100%) as
a brown solid.
1
H NMR (400 MHz, CDCl ) δ 8.83 – 8.77 (m, 1H), 8.47 – 8.39 (m, 1H), 7.87 – 7.75 (m, 1H),
3
7.41 – 7.32 (m, 1H), 7.33 – 7.28 (m, 4H), 3.72 (s, 2H), 3.67 (s, 2H), 3.12 (t, J = 5.8 Hz, 2H), 2.82
(t, J = 5.8 Hz, 2H).
Mass (ES) m/z 372 (M+1).
4-Chloro-6-(4-methoxybenzyl)-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine
(46). By use of general procedure C, reaction of 10 (3 g, 8.62 mmol) with phosphoryl chloride
(1.57 mL,17.24 mmol) gave the chlorinated derivative (3 g, yield: 100%) as a brown solid.
1
H NMR (400 MHz, CDCl ) δ 8.87 – 8.80 (m, 1H), 8.51 – 8.44 (m, 1H), 7.89 – 7.79 (m, 1H),
3
7.27 (d, J = 8.6 Hz, 2H), 6.91 (d, J = 8.6 Hz, 2H), 3.79 (s, 2H), 3.73 (s, 2H), 3.15 (t, J = 5.8 Hz,
2
H), 2.86 (t, J = 5.8 Hz, 2H).
Mass (ES) m/z 368 (M+1).
6-Benzyl-4-chloro-2-pyridin-2-yl-5,6,7,8-tetrahydropyrido[4,3-d]pyrimidine (47). General
procedure C was applied. Compound 9 (2.1 g, 7 mmol) was reacted with phosphoryl chloride (6
mL, 62 mmol, 9 equiv) in CH CN (20 mL) at 60 °C overnight. After work up, the crude was
3
ACS Paragon Plus Environment