Design, Synthesis, and Biological Evaluation of 4-Quinoline Carboxylic Acids as Inhibitors of Dihydroorotate Dehydrogenase

Article abstract of DOI:10.1021/acs.jmedchem.7b01862

We pursued a structure-guided approach toward the development of improved dihydroorotate dehydrogenase (DHODH) inhibitors with the goal of forming new interactions between DHODH and the brequinar class of inhibitors. Two potential residues, T63 and Y356, suitable for novel H-bonding interactions, were identified in the brequinar-binding pocket. Analogues were designed to maintain the essential pharmacophore and form new electrostatic interactions through strategically positioned H-bond accepting groups. This effort led to the discovery of potent quinoline-based analogues 41 (DHODH IC₅₀ = 9.71 ± 1.4 nM) and 43 (DHODH IC₅₀ = 26.2 ± 1.8 nM). A cocrystal structure between 43 and DHODH depicts a novel water mediated H-bond interaction with T63. Additional optimization led to the 1,7-naphthyridine 46 (DHODH IC₅₀ = 28.3 ± 3.3 nM) that forms a novel H-bond with Y356. Importantly, compound 41 possesses significant oral bioavailability (F = 56%) and an elimination t_1/2 = 2.78 h (PO dosing). In conclusion, the data supports further preclinical studies of our lead compounds toward selection of a candidate for early-stage clinical development.

Full text of DOI:10.1021/acs.jmedchem.7b01862

Subscriber access provided by Kaohsiung Medical University
Article
Design, Synthesis, and Biological Evaluation of 4-Quinoline
Carboxylic Acids as Inhibitors of Dihydroorotate Dehydrogenase
Joseph T. Madak, Christine R Cuthbertson, Yoshinari Miyata, Shuzo Tamura, Elyse M Petrunak,
Jeanne A. Stuckey, Yanyan Han, Miao He, Duxin Sun, Hollis Daniel Showalter, and Nouri Neamati
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01862 • Publication Date (Web): 04 May 2018
Downloaded from http://pubs.acs.org on May 5, 2018
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 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 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.
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 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis, and Biological Evaluation of 4ꢀQuinoline
Carboxylic Acids as Inhibitors of Dihydroorotate Dehydrogenase
†
†
†
†
Joseph T. Madak, Christine R. Cuthbertson, Yoshinari Miyata, Shuzo Tamura, Elyse M.
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
‡
‡
¶
¶
¶
†,*
Petrunak, Jeanne A. Stuckey, Yanyan Han, Miao He, Duxin Sun, Hollis D. Showalter,
†
,*
Nouri Neamati
†
¶
Departments of Medicinal Chemistry, Pharmaceutical Sciences, College of Pharmacy, North
‡
Campus Research Complex, 1600 Huron Parkway, and Life Sciences Institute and Department
of Biological Chemistry, University of Michigan, Ann Arbor, MI, 48109, United States
ABSTRACT:
We pursued a structureꢀguided approach toward the development of improved dihydroorotate
dehydrogenase (DHODH) inhibitors with the goal of forming new interactions between DHODH
and the brequinar class of inhibitors. Two potential residues, T63 and Y356 suitable for novel Hꢀ
bonding interactions, were identified in the brequinarꢀbinding pocket. Analogues were designed
to maintain the essential pharmacophore and form new electrostatic interactions through
strategically positioned Hꢀbond accepting groups. This effort led to the discovery of potent
quinoline based analogues 41 (DHODH IC = 9.71 ± 1.4 nM) and 43 (DHODH IC = 26.2 ±
5
0
50
1
.8 nM). A coꢀcrystal structure between 43 and DHODH depicts a novel water mediated Hꢀbond
interaction with T63. Additional optimization led to the 1,7ꢀnaphthyridine 46 (DHODH IC =
5
0
2
8.3 ± 3.3 nM) that forms a novel Hꢀbond with Y356. Importantly, compound 41 possesses
significant oral bioavailability (F = 56%) and an elimination t1/2 = 2.78 hours (PO dosing). In
conclusion, the data supports further preclinical studies of our lead compounds toward selection
of a candidate for early stage clinical development.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 108
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
INTRODUCTION
Dihydroorotate dehydrogenase (DHODH) catalyzes the oxidation of dihydroorotate to
1
, 2
orotate, which represents a committed step in the de novo pyrimidine biosynthesis pathway.
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
Inhibitors of DHODH induce pyrimidine depletion and halt cell cycle progression at Sꢀphase,
3
where a sufficient concentration of nucleotides is required for continued growth. Pyrimidine
depletion, through DHODH inhibition, has been exploited to develop therapies for many diseases
including bacterial and viral infections, parasitic diseases (i.e., malaria), autoimmune disorders,
4, 5
and cancer.
Beyond directly halting cell growth, DHODH inhibitors sensitize cells to
6
7, 8
9
doxorubicin , fludarabine, and TRAIL therapy. Additionally, PTENꢀdeficient cancer cells are
1
0
significantly more sensitive to DHODH inhibition. DHODH inhibitors have also been shown to
11
thwart selfꢀrenewal capacity of neural crest and melanoma cells. Excitingly, DHODH
inhibitors were found to induce differentiation in vivo, leading to the promise of DHODHꢀ
1
2
targeted therapy for acute myelogenous leukemia. Collectively, these studies bolster interest for
DHODH as an anticancer target.
Previous DHODH inhibitors have failed in clinical trials as single agents and in selected
drug combinations for the treatment of various cancers. These include the wellꢀknown DHODH
inhibitors, brequinar (1) and leflunomide (2, Figure 1), which have not demonstrated widespread
success in cancer clinical trials. Despite promising preclinical data, brequinar failed to generate
1
3
14
15
an objective response in multiple phase II clinical trials for breast, colon, head and neck,
1
6
17
18
gastrointestinal, lung, and skin cancers. Brequinar sodium has poor water solubility (< 0.10
mg/mL in room temperature water at pH 7.4) and molecular aggregation occurs at high
1
9
concentrations. In addition, common serum electrolytes, such as sodium chloride, have been
1
9
demonstrated to reduce aqueous solubility by over 200ꢀfold. With a cLogP of 5.94, there is
ACS Paragon Plus Environment

Page 3 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
much room for improvement to increase water solubility and reduce formation of molecular
aggregates. As such, a previous attempt to utilize brequinar in combination with cyclosporine A
changed brequinar’s pharmacokinetic profile in which its terminal halfꢀlife was extended and
2
0
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
oral clearance rate was lowered. Altered pharmacokinetic properties such as this present a
significant liability for safe dosing, thus limiting the selection of potential patient populations for
combination therapy. Leflunomide, an FDA approved drug for rheumatoid and psoriatic
2
1
arthritis, is currently being evaluated as a single agent in clinical trials for multiple myeloma
22
(
ClinicalTrials.gov, NCT02509052) but is no longer being evaluated in combination with
2
3
vemurafenib for melanoma (ClinicalTrials.gov, NCT01611675). Leflunomide is a reported
agonist for the arylꢀhydrocarbon receptor, which controls expression of drug metabolizing
24, 25
enzymes.
Despite the disappointing results of both these agents in clinical trials, we feel that
DHODH remains a promising anticancer target awaiting the discovery of small molecule
inhibitors with better drugꢀlike properties. As such, many drug discovery campaigns have been
1
, 2, 26
carried out for DHODH.
In this paper, we report the design and synthesis of a novel class of
brequinarꢀlike inhibitors toward the discovery of agents with improved physicochemical
properties, principally better aqueous solubility, which we hypothesized would translate into
improved enzyme and cellular potency.
RESULTS AND DISCUSSION
Analogue Design. Our approach was guided by the basic pharmacophore interactions
between brequinar and DHODH. Initially, we began with an unbiased cellꢀbased screening
approach to identify new small molecule hits possessing antiproliferative properties. This led to
lead compound (3) that showed potent DHODH activity (IC = 0.250 ± 0.11 ꢁM) (Figure 1). We
5
0
postulated that 3 occupies a similar binding site as brequinar based on its structural similarity.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 108
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 brequinar binding pocket of DHODH is primarily filled with nonpolar residues, therefore
requiring lipophilic moieties. A highꢀresolution coꢀcrystal structure of DHODH with a brequinar
2
7
analogue (PDB: 1D3G) provided insight into the essential pharmacophore. The carboxylate of
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
brequinar forms a salt bridge with R136 and a potential hydrogen bond interaction with Q47,
27, 28
exemplifying the importance of the carboxylic acid.
The remaining interactions between
brequinar and DHODH occur in a hydrophobic channel with the biphenyl group or the quinoline
2
8,29
core with residues such as M43, L58, A59, and P364.
To develop a DHODH inhibitor with a
lower cLogP, we identified sites where ring heteroatom replacements could be tolerated toward
enhancing binding without disrupting the essential pharmacophore.
Two regions were explored to potentially form electrostatic interactions. These regions
were previously identified and are of interest because of their polar nature in a primarily
3
0
lipophilic pocket. The first region, located within a hydrophobic channel, has two hydroxyls,
one on Y38 and another on T63. Each hydroxyl is in close proximity (Y38, 3.7 Å and T63, 3.9
Å) to form an interaction with the meta position of the distal phenyl of a brequinar analogue
27
(
Figure 2). The distance is suitable for an Hꢀbonding interaction and, if achieved, may provide
31
an inhibitor with improved enthalpyꢀdriven potency and a lower cLogP. Additionally, the
hydroxyl of Y356 is nearly 2.5 Å away from C7 on the quinoline ring (Figure 2) and presents
another opportunity to form a Hꢀbond or other electrostatic interaction. With these residues in
mind, we postulated that strategically positioned Hꢀbond acceptors on the brequinar
pharmacophore would improve potency while lowering the cLogP. The overall decrease in
cLogP should significantly improve the aqueous solubility leading to inhibitors with better drugꢀ
like properties.
ACS Paragon Plus Environment

Page 5 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To optimize our screening lead 3, designed analogues were evaluated in the DHODH enzyme
assay and MTT assay. We sought to develop analogues with a lower cLogP that may interact
with T63 and Y356. Obviously, reducing the cLogP with analogues containing an ionizable acid
could significantly reduce cellular permeability. However, we rationalized that, if necessary, an
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
32
improved DHODH inhibitor could utilize a proꢀdrug strategy to address this issue. Therefore,
our optimization process was primarily informed by the DHODH enzyme assay. Cellular activity
was evaluated in HCTꢀ116, a colon cancer cell line, and MIA PaCaꢀ2, a pancreatic cancer cell
line. Colon cancer has a high expression of DHODH and HCTꢀ116 is sensitive to DHODH
inhibition (Supplemental Figure 1). DHODH is not overexpressed in pancreatic cancer and
therefore should be less sensitive to DHODH inhibitors. Collectively, these two cell lines may
help distinguish between DHODH induced growth inhibition and potential offꢀtarget effects.
3
3
Additionally, cLogP and LipE calculators were utilized to approximate aqueous solubility.
Using the cLogP, calculations of LipE should provide a statement of the lipophilicity associated
with potency and approximate a compound’s drugꢀlike properties.
Synthesis. The synthesis of analogues containing the quinoline core utilized the classic
3
4
Pfitzinger condensation reaction. 2,4ꢀSubstituted quinoline analogues were generated using
synthetic routes AꢀC (Scheme 1). Route A generates analogues in one step. However, this
approach was limited primarily to commercially available 4’ꢀsubstituted acetophenones. To
expand our SAR beyond these limitations, we developed routes B and C, which introduce
diversity through carbonꢀcarbon bond coupling via the Suzuki reaction on suitable precursors.
These routes differ principally when such couplings take place with a phenylꢀR moiety
1
generated at a late and early stage, respectively. The advantage of Route B (Scheme 1) is that R1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 108
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
moieties sensitive to the harsh Pfitizinger reaction conditions can be successfully installed,
whereas Route C (Scheme 1) is less tolerant to such moieties.
Analogues containing a C3 methyl were incorporated through the Pfitzinger reaction with
the appropriate 4'ꢀbromopropiophenone (54ꢀ57, Scheme 2). Surprisingly, FischerꢀSpeier
esterification conditions were low yielding on carboxylic acid analogues containing the C3
methyl. Alternatively, methyl ester analogues (59ꢀ62) were obtained through generation of the
cesium salt of the carboxylic acid followed by exposure to iodomethane. Subsequent Suzuki
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
coupling was used for installation of the terminal aromatic R substituents (79ꢀ90). However, the
1
C3 methyl thwarted mild ester hydrolysis, therefore harsher conditions with strong base (> 60 °C
with NaOH) were required for the reaction to occur. Unfortunately, this favored side product
formation with little to no generation of acid, thus alternative deprotection methods were
evaluated. Gratifyingly, the carboxylic acid could be unveiled upon exposure to BBr in
3
dichloromethane at room temperature with minimal side product formation (34ꢀ44, Scheme 2).
Additional analogues were generated to evaluate a scaffold hop from the quinoline series to
synthetically challenging naphthyridine congeners. As previously noted, we postulated that Y356
is close enough to potentially form an Hꢀbond interaction with the N7 nitrogen of a 1,7ꢀ
naphthyridine core (Figure 2 and Scheme 3). Initial attempts to generate this focused on
oxidation of azaꢀindoles to yield the corresponding azaꢀisatin for subsequent Pfitzinger
35
36
condensation. However, reactions involving PCC oxidation or a reusable polyꢀaniline catalyst
on 1,6ꢀazaꢀindoles did not provide the desired 1,6ꢀazaꢀisatin. Inspired by the work of Stockmann
37
38
et al. and Zong et al., we sought to generate our naphthyridine series by installing an αꢀketo
3
8
ester on an aminopyridine amide precursor, which would provide an intermediate for a
3
9
40
subsequent Pfitzinger condensation. Drawing on the work of Turner et al., Meanwell et al.,
ACS Paragon Plus Environment

Page 7 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
38
and Zong et al., we generated our desired intermediate through directed metalation to control
regioselectivity as shown in Scheme 3. A pivaloyl amide (92ꢀ95) was utilized to direct orthoꢀ
lithiation. Exposure of this intermediate to nꢀBuLi/TMEDA under strict temperature control
provided the anion, which was then quenched with diethyl oxalate to give the key αꢀketoꢀester
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
(
97ꢀ100). This was then transformed to the naphthyridine (45ꢀ50) through baseꢀcatalyzed
cyclization under Pfitzinger reaction conditions.
The synthesis of C6ꢀsubstituted 1,7ꢀnaphthyridines presented an obstacle for SAR evaluation.
Attempts to generate the C6ꢀfluoro substituted 1,7ꢀnaphthyridine were foiled by the harsh
reaction conditions necessary for the Pfitzinger cyclization (Scheme 3, Box i). The high
temperatures and basic conditions favor fluorine displacement through an SN mechanism,
Ar
which likely proceed at a faster rate than cyclization. LCMS studies indicated that when the
reaction temperature was above 50 °C, the fluorine was displaced under basic conditions within a
few hours. Attempts to generate the C6ꢀsubstituted methyl analogue were foiled using the orthoꢀ
lithiation protocol as the excess of nꢀBuLi favored kinetic deprotonation of the methyl group,
leading to a variety of side products (Scheme 3, Box ii). To overcome this, we opted to make the
C6ꢀtrifluoromethyl analogue and were gratified to secure it despite low yields.
The overall low yield of our route to 1,7ꢀnaphthyridines presented an obstacle to expanding
the SAR around this core. Efforts to improve the orthoꢀlithiation/acylation step initially focused
on improving the electrophile’s reactivity. Diethyl oxalate was replaced with asymmetric
electrophiles such as ethyl chlorooxoacetate or ethyl 2ꢀ(methoxy(methyl)amino)ꢀ2ꢀoxoacetate
(
Weinreb amide derivative), but without any yield enhancement. We then shifted our efforts to
improving the Pfitzinger cyclization reaction (Scheme 3, reaction c). LCMS traces indicated that
ester hydrolysis was followed by decarboxylation, which occurred at a faster rate than hydrolysis
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 108
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
of the pivaloyl amide, resulting in low yields of product. To avoid the unwanted decarboxylation,
we decided to swap the orthoꢀlithiating pivaloyl amide group with a Boc carbamate. Formation
of the oxalate ester followed by mild Boc acid hydrolysis could then unmask the amine needed
for intramolecular cyclization onto an αꢀketo acid. The reduction of this to practice is shown in
Scheme 4. With αꢀketo ester intermediate 101 in hand, acid hydrolysis unveiled the amine
followed by its condensation with 4’ꢀbromopropionone and cyclization to provide the key 1,7ꢀ
naphthyridine intermediate. Methyl esterification via cesium salt/methyl iodide failed, but was
successful using TMSꢀdiazomethane to give 102. Suzuki coupling resulting in installation of the
terminal 2ꢀfluorophenyl substituent was followed by methyl ester hydrolysis to yield the final
product (52). Alternatively, initial ester hydrolysis followed by Suzuki coupling on the resultant
carboxylic acid successfully yielded final product 51. With the improved reaction conditions, we
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
were able to incorporate the C6 CF substituent into the 1,7ꢀnaphthyridine scaffold. This
3
methodology should be readily amenable to the synthesis of additional naphthyridine analogues.
Structure activity relationships. Our initial goal was to validate that our synthesized
analogues followed the same SAR trends (Table 1) as observed previously for the brequinar
2
8
series. With screen lead 3 and subsequent analogues, we noted an immediate decrease in
potency for analogues containing a methyl ester substituent (e.g., compare 3 to 4, Table 1).
Attempts to lower cLogP by incorporating piperidine (5) or morpholine (7) substituents were not
well tolerated and resulted in a marked potency decrease in comparison to 3. Replacement of the
cyclohexyl group with branched aliphatic groups (8, 9) maintained modest potency in the
DHODH assay, but were far less potent than 3. As expected, replacement of the cyclohexyl ring
with a phenyl ring (11) resulted in a potency enhancement. This analogue has a lower cLogP
than 3, but substitution of the conformationally flexible cyclohexyl with the planar phenyl allows
ACS Paragon Plus Environment

Page 9 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
for the formation of stacking interactions at high concentrations in aqueous solutions. Overall,
these initial SAR trends are consistent with those observed for established DHODH inhibitors
2
8
and suggest that our inhibitors bind in a similar pocket as brequinar. To further confirm that
DHODH is the primary target for cell growth inhibition, we evaluated screen lead 3 in the
presence and absence of excess uridine, a nucleoside which, when entering the pyrimidine
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
4
1, 42
salvage pathway, ablates DHODH inhibition.
In the absence of uridine, cells are susceptible
to growth inhibition by 3 and brequinar (Supplemental Figure 2). Uridine supplementation (5
µM) significantly decreased cell growth inhibition (below 50%), suggesting 3 inhibits an enzyme
upstream of uridine nucleotide production. This data strengthened our initial postulate that 3 has
a similar mechanism of action as brequinar, which shifted our attention towards incorporation of
Hꢀbond acceptors into our next set of analogues.
R pyridine substituents were incorporated to probe for potential interactions with Y38 and
1
T63. Our first goal was to evaluate positional pyridine isomers (Table 2). In the DHODH assay,
the pyridine analogue 14 is 25ꢀ and 19ꢀfold more potent than isomeric pyridine analogues 13 and
1
6, respectively. This potency difference between these suggests a potential interaction between
the pyridine nitrogen of 14 and DHODH. When docked into DHODH, 13, 14, and 16 all showed
similar molecular docking energies (Supplemental Table 1), and all placements of the nitrogen
were too far away to form an Hꢀbond with either T63 or Y38 (≥ 4.0 Å). The pyridine nitrogen of
1
4 points toward the hydroxyl group of T63 at a distance of 4.6 Å and is therefore too far to form
a direct Hꢀbond. While we were not able to rationalize the difference in potency experimentally
between the pyridine isomers, we observed that the metaꢀisomer (14) retained nanomolar activity
against DHODH and had the highest LipE of any analogue at that juncture of our SAR
campaign. In cells, 14 has a modest effect on cell growth in the HCTꢀ116 line. In contrast, the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 108
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
methyl ester analogue 15 is inactive in the DHODH assay but possesses improved cytotoxicity in
cells, which may be a result of improved permeability followed by intracellular hydrolysis to
generate 14. However, this has not been confirmed experimentally. The addition of the methoxy
substituent at the 2’ position of the pyridine ring to increase its electron density (17) improves
potency in both the DHODH assay and cellꢀbased viability assays. However, this enhancement is
greater for phenyl analogue 18, suggesting that enhancements from 14 to 17 are not due to
electronic effects alone. In both cell lines 18 is more potent than 17, which may be a result of
decreased permeability for 17 considering the relative cLogP values and modest DHODH
potency differences.
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
Replacement of the pyridine ring with pyrimidine (19) resulted in a marked decrease in
potency. Incorporation of the additional nitrogen increases tPSA, which may result in a repulsive
interaction in a lipophilic pocket as this potency decrease is also observed with pyridazine (20)
and reverse pyrimidine (21) analogues. The electron withdrawing CF substituent was
3
incorporated at the 2’ꢀ (22) and 4’ꢀ (23) positions (pyridine ring numbering; Table 3). Analogue
2
2 is more potent than 23 in the DHODH assay but shows no significant difference in cells. The
same is observed for methyl substituents (25, 26) installed at various positions. Installation of a
strongly electronꢀwithdrawing fluoro substituent at the 2’ꢀ (28) and 6’ꢀ (27) positions of the
pyridine ring resulted in a marked loss of potency. In fact, compound 28 is 64ꢀfold less potent
than 26 and 9ꢀfold less potent than 22 suggesting that steric effects also play an important role.
Docking of these three analogues did not offer any insights as to why 28 is much less potent as it
had the same predicted binding energy as 26. Both 26 and 28 were predicted to bind about 2ꢀfold
weaker than 22 (Supplemental Table 1). Based on the steric effect hypothesis, we incorporated a
chlorine at the 2’ position to provide 29. Chlorine has a similar van der Waals radius as a methyl
ACS Paragon Plus Environment

Page 11 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
group but different inductive effects. Compound 29 is amongst the most potent analogues
synthesized with predicted metabolic stability better than 26. In cells, 29 is more potent in the
HCTꢀ116 and MIA PaCaꢀ2 lines than 26. Further installation of a methyl group at the 6’ position
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
(30) of the pyridine ring was completed to increase binding in a lipophilic pocket. However, this
does not improve potency. Neither does replacement of the pyridine with an imidazole (31).
Altogether, 2’ꢀsubstutited pyridine analogues (e.g., 26, 29) are amongst the most potent of
the core quinoline series. Potency differences between 2’ꢀposition substituents may be a
reflection of lipophilicity, inductive, or entropic effects that lower the total number of
conformations the pyridine ring can adopt. A comparison between 17 and 22 suggests that
inductive effects may not have a significant effect on potency. However, a comparison of
analogues 17, 22, 26, 28, and 29 suggests that substituent size may be a contributing factor. The
size of the substituent may factor into undesirable clashes or limit the degrees of conformational
freedom. Nonetheless, since analogues 26 and 29 were the most potent analogues synthesized at
this stage of our campaign, the SAR was extended around 29.
On the basis of SAR developed around the pendant R position (Tables 1ꢀ3), we focused on
1
further incorporating a C3 methyl substituent as is found in brequinar (Figure 1). We postulated
that this added substituent would further limit conformational freedom around the C2 biaryl bond
and hence minimize entropic penalties required for interaction with the nitrogen on the pendant
pyridine ring for analogues shown in Tables 2 and 3. Additionally, the C3 methyl substituent
should minimize potential stacking interactions between molecules in solution, leading to better
43
aqueous solubility. Synthesized analogues with the C3 methyl were generally more potent in
the DHODH assay than the corresponding C3 desmethyl congeners with the notable exceptions
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 108
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
of 32 and 35 (Table 4). Analogue 32 is less potent than 11 in the DHODH enzyme and cell based
assays whereas 35 is moderately less potent than 29 in both assays.
The next iteration of our optimization campaign was to evaluate core modifications of the
C6 position of the quinoline ring. In general, the crystal structure (PDB: 1D3G) depicts a small
pocket that could be occupied by a suitable C6 substituent (Figure 3). Additionally, SMARTCyp,
a software for predicting cytochrome P450ꢀmediated metabolism, identified the C6 position of a
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
4
4
nonꢀsubstituted analogue as a site of metabolic liability. Thus, incorporation of a C6 substituent
might improve potency and decrease metabolic liabilities. The SAR around this position was
limited to substituents that could be easily incorporated by our synthetic methodology. The data
in Table 5 show that C6 substitutions improve the potency of the analogue series in the DHODH
assay. In fact, analogues 39 (IC = 0.0542 ± 0.012 ꢁM), 41 (IC = 0.00971 ± 0.0014 ꢁM), and
5
0
50
4
2 (IC = 0.0360 ± 0.0058 ꢁM) were all more potent than their corresponding C6 proton
50
congener 35 (IC = 0.0543 ± 0.037 ꢁM). Furthermore, the data in Table 5 show that C6 fluorine
5
0
analogues are more potent than corresponding chloro or methyl congeners. This trend suggests
that the size of the substituent is important with smaller lipophilic functional groups better
tolerated at this position. For these analogues, 41 (IC = 0.00971 ± 0.0014 ꢁM) possesses an
50
IC on par with brequinar (IC = 0.00730 ± 0.0031 ꢁM) but with a lower cLogP (4.76 vs 5.94,
5
0
50
respectively). The lowered cLogP may explain the potency differences in the HCTꢀ116 cell line.
Despite similar potency in the DHODH assay, 41 is less potent than brequinar in HCTꢀ116 (IC50
= 3.02 ± 0.35 ꢁM vs. IC = 0.679 ± 0.19 ꢁM, respectively). This may be due to analogue 41
50
being less cell permeable than brequinar. Nonetheless, compound 41 is our most potent analogue
in the DHODH assay within the quinoline series.
ACS Paragon Plus Environment

Page 13 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To form potential interactions with Y356, a scaffold hop was made from the quinoline to
naphthyridine core (Table 6). The SAR highlights that 1,7ꢀnaphthyridines are the most potent
within the broad series synthesized and supports our hypothesis of the importance of Hꢀbonding
interactions with Y356. Both the 1,6ꢀ and 1,8ꢀ naphthyridines (47 and 45) are less potent than the
quinoline (32) and suggest that these positional nitrogen placements generate unfavorable
interactions. In contrast, the 1,7ꢀnaphthyridine (46) displays greater potency than the quinoline
congener (IC = 0.0283 ± 0.0033 ꢁM vs. IC = 0.0754 ± 0.017 ꢁM, respectively). Molecular
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
5
0
50
docking of these four compounds also support these findings. Looking at the predicted binding
energies alone was not informative with respect to potency differences, but when the optimal
poses for 45, 46, 47 were overlaid in the binding site, it was clear that the 1,7ꢀnapthyridine (46)
is in a favorable position to form a hydrogen bond with Y356 (Supplemental Table 1 and
Supplemental Figure 3). While 46 did not possess potent cell activity in both HCTꢀ1116 and
MIA PaCaꢀ2 compared to brequinar, we were confident that incorporation of a substitution at the
C6 position, as found in brequinar, would improve activity.
Our focus subsequently shifted to C6 substituted 1,7ꢀnaphthyridines. For the quinoline series,
analogues possessing C6 substitution are far more potent than their nonꢀsubstituted analogues
(35 vs 41, 36 vs 43). We postulated that C6 substituted 1,7ꢀnaphthyridines would be more potent
than corresponding C6 substituted quinolines and possess a lower cLogP. While there was not
much difference in DHODH potency between 46 and 48, there was a significant increase in cell
activity with the C6ꢀsubstituted 1,7ꢀnapthyridine (48). The next obvious step was to incorporate
the terminal 2ꢀchloroꢀ3ꢀpyridyl moiety onto the core 1,7ꢀnaphthyridine scaffold.
Unfortunately, 1,7ꢀnaphthyridine analogues with a pyridine terminal attachment (49ꢀ51)
(Table 7) are not as potent as naphthyridines with a terminal phenyl group (46, 48) or
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 108
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
corresponding analogues within the quinoline series (41, 43). This result was unexpected and
may be the result of the terminal aromatic ring adopting different conformations when binding to
Hꢀbond donors. If binding with the new residues modifies the binding conformation, it may
increase the distance between small molecule Hꢀbond acceptor and donor residues leading to
more unfavorable interactions (e.g., clashing with hydrophobic residues). However, this has not
been confirmed experimentally and is being evaluated as part of another study. Furthermore,
while terminal pyridyl analogues of Table 7 possess modest activity in the DHODH assay, they
are not active in HCTꢀ116 cells. In fact, analogues with a cLogP < 4 (49, 50) do not adversely
affect cell growth below 100 ꢁM. In contrast, analogue 52, containing the terminal 2ꢀ
fluorophenyl substituent found in brequinar, is potent in the DHODH assay and is the most
potent analogue in the MTT assay against both cell lines, by at least 2ꢀfold over brequinar.
Collectively, the potency of the 1,7ꢀnaphthyridines in the DHODH assay suggests potential
interactions are being formed with DHODH, possibly via Y356. As DHODH inhibitors were
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
recently discovered to have activity in acute myeloid leukemia, we evaluated our optimized
analogs for antiproliferative activity against HLꢀ60 (Figure 5). Our inhibitors exhibited similar
antiproliferative effects as were seen in HCTꢀ116 and MIA PaCaꢀ2. Inhibitors with a lower
DHODH IC were generally more potent in HCTꢀ116 than in MIA PaCaꢀ2 cells. Additionally,
5
0
the antiꢀproliferative effects of our potent inhibitors of DHODH were rescued by uridine
supplementation, validating their mechanism of action (Figure 5 and Supplemental Figure 4). In
2
6, 45
agreement with previously published DHODH inhibitors,
exerted their antiꢀproliferative effects through a cytostatic rather than cytotoxic mechanism
Supplemental Figure 5).
we observed that our inhibitors
(
ACS Paragon Plus Environment

Page 15 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Crystallography. The structures of DHODH coꢀcrystallized with 43 and 46 were solved to
.63 Å and 2.85 Å resolution, respectively. DHODH:43 (PDB ID: 6CJF) was solved with two
1
molecules in the asymmetric unit, while DHODH:46 (PDB ID: 6CJG) contained only one. Their
overall structures are similar in that the single chain of DHODH:46 aligns with RMSD values of
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
0
.741 Å and 0.739 Å to the A and B chains of DHODH:43, respectively. The structures also
2
7,
adopt the same fold as other brequinar analogues bound to DHODH reported in the literature.
4
6, 47
27, 46, 47
Consistent with previous structures of DHODH bound to brequinar analogs,
inhibitors
4
3 and 46 occupy the proposed ubiquinone binding site, a hydrophobic channel formed by α1
and α2 helices directed toward the proximal redox site with the 4ꢀcarboxylic acid quinoline ring
system oriented toward the active site. Both compounds are stabilized by a substantial number
of hydrophobic interactions with the side chains of DHODH residues lining the pocket (L42,
M43, L46, P52, A55, A59, F62, and T63) (Figure 4, A and B). The more deeply buried
quinoline ring of 43 and naphthyridine ring of 46 form van der Waals contacts with a series of
hydrophobic residues near the redox site including F96, V134, V143, T360, Y356, L359, and
P364 (Figure 4, A and B). The C7ꢀfluorine substituent of 43 is directed toward the FMN
27, 47
cofactor as in previous structures with other brequinar analogs,
and the C4 carboxylate of
both inhibitors form salt bridges with R136 and participate in Hꢀbonding with Q47 and T360
through a water molecule (Figure 4, C and D).
In contrast to previous structures, the additional nitrogen incorporated into the 1,7ꢀ
naphthyridine ring of 46 is within 2.9 Å of the Y356 hydroxyl group, allows this inhibitor to
participate in an additional Hꢀbonding interaction within the binding site (Figure 4C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 108
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 place of the unsubstituted phenyl ring of 46 and the 2ꢀfluorophenyl ring of brequinar,
3 features a 2ꢀchloroꢀ6ꢀmethyl pyridine ring that can adopt two different orientations. Based on
4
the ligand omit map density (Supplemental Figure 6), the chlorine substituent can be directed
toward the α1 helix, but this orientation has low occupancy, likely in part due to steric clashes
with the hydrophobic residues lining the binding pocket. The higher occupancy orientation of
the pyridine ring directs the chlorine toward F98. In this orientation, the pyridine nitrogen is
directed toward space usually occupied by the disordered 68ꢀ72 loop. In one molecule of the
asymmetric unit, density corresponding to a water lies within 3.4 Å of the pyridine nitrogen and
mediates a Hꢀbond with the side chain of T63 (Figure 4D). The 6ꢀmethyl substituent is also
directed toward space usually occupied by the loop connecting the two domains; steric clashes
with this highly flexible loop might contribute to the intrinsic disorder of this loop and explain
why density corresponding to residues 68ꢀ70 is absent in the structure with 43 in contrast to the
DHODH:46 structure in which these residues were able to be modeled.
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
There is additional density near the solventꢀexposed entrance to the channel occupied by the
inhibitor. The extra electron density in this site has been reported in previous structures of
2
7, 47
DHODH with brequinar analogs.
Liu and colleagues modeled a detergent, DDAO, in this
2
7
site.
In our structures of DHODH:43 and DHODH:46, we modeled the aliphatic chain in
common to the two detergents, Anzergent 3ꢀ10 and HEGAꢀ8, used for crystallization
(Supplemental Figure 7). The aliphatic chain forms van der Waals interactions with the side
chains of F62 and P69 as well as the backbone of residues 67ꢀ69. Additionally, the detergent
makes contact with the phenyl group of 46. In the DHODH:43 structure, the aliphatic chain does
not interact with the loop, but forms van der Waals interactions with the side chains of F62, L46,
L42 and F37 along with the terminal substituted pyridine of 43. Although the detergent is not
ACS Paragon Plus Environment

Page 17 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
physiologically relevant, it is likely that this face of the protein and the bound inhibitor make
2
7
contacts with aliphatic chains of phospholipids comprising the membrane.
Pharmacokinetic evaluation and thermodynamic solubility. We deemed compound 41 to
possess an overall profile that merited pharmacokinetic evaluation in mice (Table 8). When 41
was administered either orally or intravenously, a similar elimination halfꢀlife of 2.73ꢀ2.78 hours
was observed. Oral bioavailability for 41 is 56%, which is suitable for DHODH inhibition
considering its potency. For PO dosing, the Cmax was 5313 ng/mL and was reached in 0.25 h. The
clearance rate and volume of distribution are favorable for continuous inhibition, suggesting that
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
4
1 is well suited for further investigation.
Analogue 46 is more soluble in aqueous solutions than compounds of the quinoline series. In
comparison to brequinar, compound 46 is 2.5 – 3x more soluble in PBS at pH 7.4 (Table 9). The
nitrogen of the naphthyridine core significantly increases the aqueous solubility in comparison to
the quinoline core. However, a comparison between 41 and brequinar highlights that despite a
lower cLogP, 41’s aqueous solubility is not significantly different from that of brequinar. The
large chlorine group may limit the total solvent exposure of 41’s pyridine moiety, which
minimizes its effect on aqueous solubility. Nonetheless, compound 46 is endowed with better
drugꢀlike properties compared to brequinar, with potent DHODH inhibition and improved
aqueous solubility.
CONCLUSION
This study reports on our drug discovery campaign to develop a novel DHODH inhibitor
with improved potency and aqueous solubility. Our efforts toward designing molecular
interactions with T63 led to the potent inhibitors 41 (DHODH IC = 0.00971 ± 0.0014 ꢁM) and
5
0
43 (DHODH IC = 0.0262 ± 0.018 ꢁM), each possessing a lower cLogP than brequinar (4.76,
5
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 108
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
5
.27 vs. 5.94 respectively). A coꢀcrystal structure of 43 and DHODH depicts a waterꢀmediated
Hꢀbonding interaction with the terminal pyridine and T63 (Figure 4D). Efforts to form an
interaction with Y356 led to the synthesis of 1,7ꢀnaphthyridine inhibitors; 46 (DHODH IC =
5
0
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
0
.0283 ± 0.0033 ꢁM), 48 (DHODH IC = 0.0212 ± 0.0039 ꢁM), and 52 (DHODH IC = 0.0118
50 50
±
0.00090 ꢁM). In a coꢀcrystal structure with DHODH, the 7’ꢀnitrogen of 46’s core forms an Hꢀ
bond with the hydroxyl of Y356 (Figure 4C). Furthermore, analogue 46 is significantly more
soluble in aqueous solution than brequinar. Additionally, 52 exhibited improved cellular activity
compared to brequinar. Moreover, optimized analogs displayed considerable antiproliferative
activity in an acute myeloid leukemia cell line. Pharmacokinetic evaluation of 41 highlights
suitable oral bioavailability (F = 56%) and halfꢀlife (t1/2 = 2.78 h), suitable for further
investigation. Collectively, these inhibitors demonstrate the potential to enhance binding with
DHODH through novel Hꢀbonding interactions. The results of our study highlight the possibility
to simultaneously improve potency against DHODH while lowering the cLogP for inhibitors that
occupy a lipophilic binding pocket. The disclosed inhibitors offer suitable leads for further
inhibitor design and early stage development. Future studies will seek to improve cell activity
through prodrug design strategies. Additionally, we plan to evaluate the capability of these
inhibitors to induce differentiation in acute myeloid leukemia in cell lines and in in vivo efficacy
studies.
EXPERIMENTAL SECTION
General methods:
Reagents and anhydrous solvents were used without further purification and purchased from
commercial sources. Cinchophen and brequinar were purchased from commercial sources and
ACS Paragon Plus Environment

Page 19 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
utilized without further purification. A Biotage® Initiator+ was used to perform microwave
catalyzed reactions in sealed vials. Reaction progress was monitored by UV absorbance using
thinꢀlayer chromatography (TLC) on aluminumꢀbacked precoated silica plates from Silicycle
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
(
SiliaPlate™, 200 ꢁm thickness, F ). Purifications using flash chromatography were performed
254
using Silicycle silica gel (SiliaFlash® F60, 40ꢀ63 ꢁm, 230ꢀ400 mesh, PN: R10030B) and a small
percentage of compounds were purified using a Biotage® Isolera Chromatography system
equipped with 10 g and 25 g UltraꢀSNAP Cartridge columns (25 ꢁM spherical silica). Glassware
for reactions were oven dried in preparation and reactions were performed using nitrogen or
1
argon atmosphere using standard inert conditions. H NMR spectra were obtained using a Bruker
(300 or 400 MHz) or a Varian (400 or 500 MHz) instrument. Spectral data are reported using
the following abbreviations: (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, dd =
1
3
doublet of doublets), coupling constants are reported in Hz, followed by integration. C NMR
spectra were obtained at 126 MHz on a Varian 500 MHz instrument with a proton decoupled
probe. A Shimadzu LCMS 20ꢀ20 system was utilized for generating HPLC traces, obtaining
mass spectrometry data, and evaluating purity. The system is equipped with a PDA UV detector
and Kinetex® 2.6 ꢁm, XBꢀC18 100 Å, 75 x 4.6 mm column, which was used at room
temperature. HPLC gradient method utilized a 1% to 90% MeCN in H O with 0.01% formic acid
2
over 20 minutes with a 0.50 mL/min flow rate. Purity of final compounds (≥ 95%) was assessed
at 254 nm using the described column and method. Reverseꢀphase preparatory purifications were
performed on a Shimadzu LCꢀ20 modular HPLC system. This system utilized a PDA detector
and a Kinetex® 5 ꢁm XBꢀC18 100 Å, 150 x 21.2 mm column. Purification methods used a 27ꢀ
minute gradient from 10% to 90% MeCN in H O with 0.02% trifluoroacetic acid. The chemicals
2
nꢀBuLi and TMSꢀdiazomethane present hazards. The chemical nꢀBuLi is pyrophoric and must be
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 108
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
utilized under an inert atmosphere. The chemical TMSꢀdiazomethane is toxic and potentially
explosive. Caution should be utilized when working with these chemicals.
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
General protocol A, Pfitzinger synthesis: Isatin (1.0 ꢀ1.2 eq) was added to a room temperature
solution of KOH (4.0 eq) in an EtOH/H O solution and mixed with various paraꢀsubstituted
2
derivatives of acetophenone (1.0 eq). The solution was heated to reflux for 24ꢀ48 hours,
concentrated in vacuo, and reꢀdissolved in 1 M NaOH/EtOAc. The aqueous layer was washed
with EtOAc (3x) and acidified using HCl or glacial acetic acid until precipitant was observed
(pH 2ꢀ3). The precipitant was filtered over a fritted funnel, washed with 1 M HCl, and washed
again with 2ꢀpropanol, diethyl ether, or ethanol to yield final compounds (2ꢀ81%).
General protocol B, acid catalyzed esterification: The corresponding carboxylic acid (1.0 eq)
was dissolved in solution containing anhydrous MeOH and a catalytic amount of H SO The
2
4.
reaction mixture was refluxed overnight before concentrating MeOH. Residue was reꢀdissolved
in H O and neutralized before extracting product with EtOAc (3x). The organic layer was dried
2
with MgSO filtered, concentrated, and purified via flash chromatography using a gradient
4
,
method from 5 to 60% EtOAc in hexane (63ꢀ88%).
General protocol C, base catalyzed esterification: The corresponding carboxylic acid (1.0 eq),
Cs CO (1.2 eq), and MeI (2.0 eq) were dissolved in anhydrous DMF. The mixture was stirred at
2
3
room temperature overnight. Upon completion, the mixture was diluted EtOAc and washed with
brine (8x). The organic layer was dried with MgSO filtered, concentrated, and purified via flash
4
,
chromatography using a gradient method from 5 to 60% EtOAc in hexane (43ꢀ93%).
General procedure D, Suzuki coupling Nonꢀmicrowave protocol: To a degassed round bottom
flask containing brominated starting material (1 eq), boronic acid (1.2ꢀ1.5 eq), K CO (3ꢀ4 eq),
2
3
ACS Paragon Plus Environment

Page 21 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Pd(PPh ) (5ꢀ10% mol) was dissolved in dioxane or toluene. The reaction was heated at reflux
3
4
until loss of starting material was observed via TLC (12ꢀ48 hrs). The reaction mixture was
concentrated, partitioned between EtOAc/H O, and washed with EtOAc (3x). The organic layer
2
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
was dried with MgSO and concentrated to a residue, which was then purified by flash silica
4
chromatography using a gradient of 10 to 60% EtOAc in hexane.
Microwave protocol: Brominated starting material (1 eq), corresponding boronic acid (1.5 eq),
base (Na CO or K HPO 3ꢀ4 eq), and Pd(PPh ) (≈ 5% eq) were dissolved in 2:1 dioxane/H O
2
3
2
4,
3 4
2
or toluene/H O in a microwave vial. The vial was capped, purged with argon, and used in a
2
microwave synthesizer. The reaction was heated at 125ꢀ130 °C for 1.5 – 2 hrs and followed the
same purification as the nonꢀmicrowave protocol. Yields for both protocols ranged from 7 –
9
7%.
General procedure E, basic hydrolysis: Ester derivatives were combined with 1ꢀ2 pellets of
NaOH (large excess), LiOH, or KOH and dissolved in a 1:1 mixture of THF/H O or
2
dioxane/H O The solution was heated to 40 °C until starting material was no longer observed (2ꢀ
2
.
6 hours). Upon completion, solvent was concentrated; residue was reꢀdissolved in 1 M KOH and
washed with EtOAc (3x). The aqueous layer was acidified with HCl until pH 2ꢀ3 was reached,
chilled overnight at 2ꢀ8 °C, then poured over a fritted funnel to collect precipitated product.
Product cake was washed with chilled deꢀionized H O and product was dried under vacuum (15ꢀ
2
1
00%).
General procedure F, acidic hydrolysis: Ester derivatives were dissolved in anhydrous DCM
and 1 M BBr in DCM was added to the solution. The solution was stirred at room temperature
3
overnight. Upon completion, the solvent was concentrated; residue was reꢀdissolved in 1 M
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 108
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
KOH and washed with EtOAc (3x). The aqueous layer was acidified with HCl until pH 2ꢀ3 was
reached, chilled overnight at 2ꢀ8 °C, then poured over a fritted funnel to collect precipitated
product. Product cake was washed with chilled deꢀionized H O and product was dried under
2
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
vacuum.
General procedure G, pivaloyl protecting group: The corresponding aminopyridine (1 eq)
was added to a round bottom flask containing anhydrous DCM and TEA/DIPEA (1.25 eq). The
solution was stirred on an ice bath for 20 minutes before trimethylacetyl chloride (1.1 eq) was
added dropꢀwise. The solution was allowed to warm to room temperature and quenched with
H O/NaHCO . The product was extracted in DCM (3x), dried with MgSO , and concentrated in
2
3
4
vacuo. The solid was recrystallized in hexanes to afford pivaloylꢀaminopyridines (53ꢀ98%).
48
General protocol H, orthoꢀlithiation: This procedure was adapted from Estel et al. and Zong
3
8
et al. The corresponding pivaloylamino pyridine (1.0 eq) was added to a vacuum purged round
bottom flask under argon atmosphere. The solid was dissolved in Et O and TMEDA (2.5 eq)
2
then chilled to ꢀ78 °C. After 15 minutes, nꢀBuLi (1.6 M in hexanes, 2.5 eq) was added dropwise
and the reaction mixture was stirred at ꢀ78 °C for 15 minutes and then stirred at ꢀ10 °C for 2.5
hours. Afterwards, the temperature was lowered to ꢀ78 °C for the addition of diethyl oxalate (3.0
eq) and warmed to room temperature over 1.5 hours. The reaction mixture was quenched with
1.0 M solution of HCl in ice water and extracted with Et O (3x). Organic layer was dried,
2
concentrated, and purified via silica chromatography in a gradient from 1% to 10% MeOH in
DCM gradient (1–51%).
General protocol I, generation of naphthyridine core: Corresponding ketone (1.0 eq), αꢀketoꢀ
ester (1.0 eq), and base (KOH or KOtBu, 3ꢀ6 eq) were combined in anhydrous EtOH in a sealed
vial. The mixture was heated to 100 °C overnight, concentrated to a residue, extracted with 1 M
ACS Paragon Plus Environment

Page 23 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
KOH, and washed with EtOAc. The basic layer was acidified by addition of HCl (pH 2ꢀ3) and
the product was collected via filtration. Product was further purified by reverseꢀphase
preparatory chromatography (3ꢀ21%).
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
2
ꢀ(4ꢀCyclohexylphenyl)quinolineꢀ4ꢀcarboxylic acid (3): Isatin (987 mg, 6.71 mmol), 1ꢀ(4ꢀ
cyclohexylphenyl)ethanꢀ1ꢀone (1.00 g, 4.95 mmol), and KOH (1.55 g, 27.6 mmol) were
dissolved in 30 mL EtOH and 10 mL H O. Following general protocol A, 2ꢀ(4ꢀ
2
cyclohexylphenyl)quinolineꢀ4ꢀcarboxylic acid was recovered as a white powder (374 mg, 1.13
1
mmol, 23%). H NMR (500 MHz, DMSOꢀd ) δ 8.64 (d, J = 8.6 Hz, 1H), 8.41 (d, J = 1.8 Hz,
6
1H), 8.21 – 8.15 (m, 2H), 8.14 – 8.08 (m, 1H), 7.85 – 7.79 (m, 1H), 7.71 – 7.61 (m, 1H), 7.40 –
7
4
.34 (m, 2H), 2.59 – 2.50 (m, 1H), 1.83 – 1.73 (m, 4H), 1.72 – 1.65 (m, 1H), 1.49 – 1.27 (m,
13
H), 1.27 – 1.10 (m, 1H); C NMR (126 MHz, DMSOꢀd )) δ 168.1, 156.2, 150.1, 148.9, 138.0,
6
136.0, 130.6, 130.1, 128.0, 127.7 (2H), 127.6 (2H), 125.9, 123.8, 119.4, 44.0, 34.2 (2H), 26.7
+
ꢀ
(
2H), 26.0; LCMS (ESI) 332.20 [M+H] , 330.20 [MꢀH] ; HPLC Purity at 254 nm, 98.6%.
Methyl 2ꢀ(4ꢀcyclohexylphenyl)quinolineꢀ4ꢀcarboxylate (4): Compound 3 (25 mg, 0.08 mmol)
was dissolved in 2 mL MeOH. H SO (16 drops, catalyst) was added to the reaction mixture and
2
4
the mixture was heated to reflux overnight. Following general protocol B, methyl 2ꢀ(4ꢀ
cyclohexylphenyl)quinolineꢀ4ꢀcarboxylate was recovered as an offꢀwhite solid (25 mg, 0.07
1
mmol, 88%) H NMR (300 MHz, CDCl ꢀd) δ 8.76 (d, J = 8.5 Hz, 1H), 8.41 (s, 1H), 8.24 (d, J =
3
8
.4 Hz, 1H), 8.15 (d, J = 8.0 Hz, 2H), 7.86 – 7.71 (m, 1H), 7.69 – 7.59 (m, 1H), 7.41 (d, J = 8.0
Hz, 2H), 4.09 (s, 3H), 2.70 – 2.57 (m, 1H), 2.03 – 1.85 (m, 4H), 1.84 – 1.67 (m, 1H), 1.61 – 1.21
+
(m, 5H); LCMS (ESI) 346.2 [M+H] ; HPLC Purity at 254 nm, 98.7%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 108
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
2
ꢀ(4ꢀ(Piperidinꢀ1ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (5): Isatin (247 mg, 1.68 mmol), 1ꢀ(4ꢀ
(piperidinꢀ1ꢀyl)phenyl)ethanꢀ1ꢀone (200 mg, 0.99 mmol), and KOH (336 mg, 6.00 mmol) were
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
dissolved in 7 mL EtOH and 2 mL H O. Following general protocol A, 2ꢀ(4ꢀ(piperidinꢀ1ꢀ
2
1
yl)phenyl)quinolineꢀ4ꢀcarboxylic acid was recovered as a red solid (88 mg, 0.27 mmol, 27%). H
NMR (400 MHz, DMSOꢀd ) δ 8.59 (d, J = 8.4 Hz, 1H), 8.37 (s, 1H), 8.18 (d, J = 8.6 Hz, 2H),
6
8
.09 (d, J = 8.4 Hz, 1H), 7.80 (t, J = 8.4, 6.9 Hz, 1H), 7.68 – 7.58 (m, 1H), 7.16 – 7.04 (m, 2H),
+
ꢀ
3
.34 – 3.27 (m, 4H), 1.71 – 1.54 (m, 6H); LCMS (ESI) 333.25 [M+H] , 331.15 [MꢀH] ; HPLC
Purity at 254 nm, 99.7%.
Methyl 2ꢀ(4ꢀ(piperidinꢀ1ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylate (6): Compound 5 (30 mg, 0.09
mmol) was dissolved in 2 mL MeOH. H SO (16 drops, catalyst) was added to the reaction
2
4
mixture and the mixture was heated to reflux for 12 hours. Following general protocol B, methyl
2
0
8
1
ꢀ(4ꢀ(piperidinꢀ1ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylate was recovered as a yellow solid (29 mg,
1
.08 mmol, 88%). H NMR (400 MHz, CDCl ꢀd) δ 8.71 (dd, J = 8.5, 1.4 Hz, 1H), 8.38 (s, 1H),
3
.21 – 8.12 (m, 3H), 7.80 – 7.70 (m, 1H), 7.63 – 7.53 (m, 1H), 7.13 – 7.00 (m, 2H), 4.08 (d, J =
.0 Hz, 3H), 3.39 – 3.28 (m, 4H), 1.80 – 1.71 (m, 4H), 1.70 – 1.60 (m, 2H); LCMS (ESI) 347.2
+
[
M+H] ; HPLC Purity at 254 nm, 97.3%.
2ꢀ(4ꢀMorpholinophenyl)quinolineꢀ4ꢀcarboxylic acid (7): Isatin (247 mg, 1.68 mmol), 1ꢀ(4ꢀ
morpholinophenyl)ethanꢀ1ꢀone (200 mg, 0.98 mmol), and KOH (336 mg, 6.00 mmol) were
dissolved in 7 mL EtOH and 1 mL H O. Following general protocol A, 2ꢀ(4ꢀ
2
morpholinophenyl)quinolineꢀ4ꢀcarboxylic acid was recovered as a red solid (132 mg, 0.40 mmol,
ACS Paragon Plus Environment

Page 25 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
4
1%). H NMR (400 MHz, DMSOꢀd ) δ 8.60 (d, J = 8.5 Hz, 1H), 8.39 (s, 1H), 8.21 (d, J = 8.7
6
Hz, 2H), 8.10 (d, J = 8.4 Hz, 1H), 7.85 – 7.76 (m, 1H), 7.68 – 7.59 (m, 1H), 7.11 (d, J = 8.7 Hz,
+
2H), 3.77 (t, J = 4.8 Hz, 4H), 3.26 (t, J = 4.9 Hz, 4H); LCMS (ESI) 335.90 [M+H] , 333.15 [Mꢀ
ꢀ
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
H] ; HPLC Purity at 254 nm, 98.8%.
2
ꢀ(4ꢀIsopropylphenyl)quinolineꢀ4ꢀcarboxylic acid (8): Isatin (264 mg, 1.79 mmol), 1ꢀ(4ꢀ
isopropylphenyl)ethanꢀ1ꢀone (0.27 mL, 1.62 mmol), and KOH (275 mg, 4.91 mmol) were
dissolved in 7 mL EtOH and 1 mL H O. Following general protocol A, 2ꢀ(4ꢀ
2
isopropylphenyl)quinolineꢀ4ꢀcarboxylic acid was recovered as a yellow solid (130 mg, 0.45
1
mmol, 28%). H NMR (400 MHz, DMSOꢀd ) δ 8.65 (d, J = 8.3 Hz, 1H), 8.44 (s, 1H), 8.23 (d, J
6
=
8.3 Hz, 2H), 8.16 (d, J = 8.5 Hz, 1H), 7.89 – 7.81 (m, 1H), 7.75 – 7.67 (m, 1H), 7.46 (d, J =
+
8
.3 Hz, 2H), 3.09 – 2.94 (m, 1H), 1.28 (d, J = 6.9 Hz, 6H); LCMS (ESI) 292.20 [M+H] , 290.15
ꢀ
[MꢀH] ; HPLC Purity at 254 nm, 95.8%.
2
ꢀ(4ꢀIsobutylphenyl)quinolineꢀ4ꢀcarboxylic acid (9): Isatin (500 mg, 3.40 mmol), 1ꢀ(4ꢀ
isobutylphenyl)ethanꢀ1ꢀone (0.57 mL, 3.08 mmol), and KOH (1.14 g, 20.4 mmol) were
dissolved in 12 mL EtOH + 2 mL H O. Following general protocol A, 2ꢀ(4ꢀ
2
isobutylphenyl)quinolineꢀ4ꢀcarboxylic acid was recovered as a pink solid (101 mg, 0.33 mmol,
1
1
0% yield). H NMR (300 MHz, DMSOꢀd ) δ 8.68 (d, J = 8.4 Hz, 1H), 8.14 (d, J = 7.9 Hz, 2H),
6
8.03 – 7.94 (m, 2H), 7.66 (t, J = 7.6 Hz, 1H), 7.46 (t, J = 7.6 Hz, 1H), 7.33 (d, J = 7.9 Hz, 2H),
+
2
3
.59 – 2.52 (m, 2H), 2.00 – 1.85 (m, 1H), 0.91 (d, J = 6.6 Hz, 6H); LCMS (ESI) 306.10 [M+H] ,
ꢀ
04.20 [MꢀH] ; HPLC Purity at 254 nm, 99.9%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 108
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
2
ꢀ([1,1'ꢀBiphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀcarboxylic acid (11): Isatin (168 mg, 1.14 mmol), 1ꢀ([1,1'ꢀ
biphenyl]ꢀ4ꢀyl)ethanꢀ1ꢀone (200 mg, 1.02 mmol), KOH (230 mg, 4.11 mmol) were dissolved in
3 mL EtOH with 1 mL H O. Following general protocol A, 2ꢀ([1,1'ꢀbiphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀ
2
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
carboxylic acid was recovered as a beige solid (74 mg, 0.23 mmol, 20% yield) after
1
recrystallization from ethanol. H NMR (300 MHz, CD ODꢀd ) δ 8.46 – 8.36 (m, 1H), 8.19 (d, J
3
4
=
8.4 Hz, 2H), 8.14 – 8.02 (m, 2H), 7.83 – 7.63 (m, 5H), 7.56 (ddd, J = 8.2, 6.8, 1.2 Hz, 1H),
+
ꢀ
7
.44 (t, J = 7.4 Hz, 2H), 7.40 – 7.28 (m, 1H); LCMS (ESI) 326.15 [M+H] , 324.10 [MꢀH] ;
HPLC Purity at 254 nm, 98.9%.
Methyl 2ꢀ([1,1'ꢀbiphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀcarboxylate (12): Intermediate 58 (100 mg, 0.29
mmol), phenylboronic acid (53 mg, 0.43 mmol), K HPO (151 mg, 0.87 mmol), and Pd(PPh )
3 4
2
4
(17 mg, 0.01 mmol) were dissolved in 2 mL dioxane and 1 mL H O. The mixture was heated to
2
130 °C for 1.5 hours in a microwave reactor. Following general protocol D, Methyl 2ꢀ([1,1'ꢀ
biphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀcarboxylate was recovered as a white crystal (28 mg, 0.08 mmol,
1
2
8%). H NMR (300 MHz, CDCl ꢀd) δ 8.76 (dd, J = 8.6, 1.3 Hz, 1H), 8.46 (s, 1H), 8.35 – 8.20
3
(m, 3H), 7.85 – 7.73 (m, 3H), 7.74 – 7.58 (m, 3H), 7.49 (t, J = 7.4 Hz, 2H), 4.09 (s, 3H); LCMS
+
(ESI) 340.10 [M+H] ; HPLC Purity at 254 nm, 99.5%.
2ꢀ(4ꢀ(Pyridinꢀ2ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (13): Isatin (82 mg, 0.56 mmol),
intermediate 63 (100 mg, 0.51 mmol), and KOH (112 mg, 2.00 mmol) were dissolved 4 mL in
EtOH with 1 mL H O. Following general protocol A, 2ꢀ(4ꢀ(pyridinꢀ2ꢀyl)phenyl)quinolineꢀ4ꢀ
2
1
carboxylic acid was recovered as a tan solid (62 mg, 0.19 mmol, 37%). H NMR (300 MHz,
CD ODꢀd ) δ 8.71 – 8.64 (m, 1H), 8.46 (d, J = 8.4 Hz, 1H), 8.31 (d, J = 8.1 Hz, 2H), 8.22 – 8.10
3
4
ACS Paragon Plus Environment

Page 27 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
m, 4H), 8.04 – 7.94 (m, 2H), 7.78 (t, J = 7.7 Hz, 1H), 7.61 (t, J = 7.7 Hz, 1H), 7.42 (t, J = 6.0
+
ꢀ
Hz, 1H); LCMS (ESI) 326.85 [M+H] , 324.85 [MꢀH] ; HPLC Purity at 254 nm, 99.4%.
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
2
ꢀ(4ꢀ(Pyridinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (14): Compound 15 (8 mg, 0.02 mmol)
and NaOH (45 mg, 1.13 mmol) was dissolved in a 1 mL THF and 1 mL H O. Following general
2
protocol E, a white solid was collected from the vacuum oven to yield 2ꢀ(4ꢀ(pyridinꢀ3ꢀ
ꢀ3
1
yl)phenyl)quinolineꢀ4ꢀcarboxylic acid (2 mg, 6.13 x 10 mmol, 30%). H NMR (400 MHz,
DMSOꢀd ) δ 9.24 – 9.20 (m, 1H), 8.82 – 8.78 (m, 1H), 8.67 (d, J = 8.5, 1.5, 0.6 Hz, 1H), 8.64 –
6
8
.61 (m, 1H), 8.56 (s, 1H), 8.50 (d, J = 8.5 Hz, 2H), 8.21 (d, 1H), 8.05 (d, J = 8.5 Hz, 2H), 7.93 –
+
ꢀ
7.85 (m, 2H), 7.78 – 7.71 (m, 1H); LCMS (ESI) 326.90 [M+H] , 324.85 [MꢀH] ; HPLC Purity
at 254 nm, 95.2%.
Methyl 2ꢀ(4ꢀ(pyridinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylate (15): Intermediate 58 (60 mg, 0.18
mmol), pyridinꢀ3ꢀylboronic acid (23 mg, 0.19 mmol), Pd(PPh ) (10 mg), and K CO (72 mg,
3
4
2
3
0
.53 mmol) were combined in 3 mL toluene and 1 mL H O. The mixture was heated at 110 °C
2
overnight. Following general protocol D, methyl 2ꢀ(4ꢀ(pyridinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀ
1
carboxylate was recovered as a white solid (20 mg, 0.06 mmol, 33%). H NMR (300 MHz,
CDCl ꢀd) δ 8.92 (s, 1H), 8.74 (d, J = 8.5 Hz, 1H), 8.62 (d, J = 4.8 Hz, 1H), 8.40 (s, 1H), 8.29 (d,
3
J = 8.2 Hz, 2H), 8.21 (d, J = 8.4 Hz, 1H), 7.91 (dd, J = 8.0, 2.0 Hz, 1H), 7.82 – 7.68 (m, 3H),
7
.66 – 7.57 (m, 1H), 7.37 (dd, J = 7.9, 4.8 Hz, 1H), 4.23 – 3.94 (m, 3H); LCMS (ESI) 341.1
+
[
M+H] ; HPLC Purity at 254 nm, 99.6%.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 108
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
2
ꢀ(4ꢀ(Pyridinꢀ4ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (16): Intermediate 72 (14 mg, 0.04
mmol) and NaOH (62 mg, 1.59 mmol) was dissolved in 1 mL THF and 1 mL H O. Following
2
general protocol E, a white solid was collected from the vacuum oven to yield 2ꢀ(4ꢀ(pyridinꢀ4ꢀ
ꢀ3
1
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
yl)phenyl)quinolineꢀ4ꢀcarboxylic acid (2 mg, 6.13 x 10 mmol, 15%). H NMR (400 MHz,
DMSOꢀd ) δ 8.72 – 8.66 (m, 3H), 8.39 (d, J = 8.3 Hz, 2H), 8.08 (s, 1H), 8.04 – 7.99 (m, 3H),
6
+
7
.85 – 7.80 (m, 2H), 7.73 – 7.66 (m, 1H), 7.53 – 7.47 (m, 1H); LCMS (ESI) 326.85 [M+H] ,
ꢀ
3
24.80 [MꢀH] ; HPLC Purity at 254 nm, 95.1%.
2
ꢀ(4ꢀ(2ꢀMethoxypyridinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (17): Intermediate 73 (15
mg, 0.04 mmol) and NaOH (63 mg, 1.62 mmol) were dissolved in 1 mL THF and 1 mL H O.
2
Following general protocol E, 2ꢀ(4ꢀ(2ꢀmethoxypyridinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid
1
was recovered as a beige solid (12 mg, 0.03 mmol, 75%). H NMR (400 MHz, DMSOꢀd ) δ 8.65
6
(d, J = 8.5 Hz, 1H), 8.50 (s, 1H), 8.36 (d, J = 8.0 Hz, 2H), 8.25 – 8.14 (m, 2H), 7.89 – 7.82 (m,
2
H), 7.76 (d, J = 7.9 Hz, 2H), 7.70 (t, J = 7.7 Hz, 1H), 7.13 (t, J = 6.2 Hz, 1H), 3.91 (s, 3H);
+
ꢀ
LCMS (ESI) 357.1 [M+H] , 355.1 [MꢀH] ; HPLC purity at 254 nm, 99.8%.
2
ꢀ(2'ꢀMethoxyꢀ[1,1'ꢀbiphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀcarboxylic acid (18): Intermediate 74 (15 mg,
0
.04 mmol) and NaOH (68 mg, 1.74 mmol) were dissolved in 1 mL THF and 1 mL H O.
2
Following general protocol E, 2ꢀ(2'ꢀmethoxyꢀ[1,1'ꢀbiphenyl]ꢀ4ꢀyl)quinolineꢀ4ꢀcarboxylic acid
1
was recovered as a beige solid (13 mg, 0.04 mmol, 100%). H NMR (400 MHz, DMSOꢀd ) δ
6
8
1
.69 – 8.61 (m, 1H), 8.49 (s, 1H), 8.37 – 8.27 (m, 2H), 8.17 (d, J = 8.4 Hz, 1H), 7.89 – 7.82 (m,
H), 7.73 – 7.65 (m, 3H), 7.41 – 7.34 (m, 2H), 7.17 – 7.12 (m, 1H), 7.10 – 7.04 (m, 1H), 3.79 (s,
+
ꢀ
3H); LCMS (ESI) 356.2 [M+H] , 354.1 [MꢀH] ; HPLC Purity at 254 nm, 98.6%.
ACS Paragon Plus Environment

Page 29 of 108
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
ꢀ(4ꢀ(Pyrimidinꢀ5ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (19): Isatin (56 mg, 0.38 mmol),
intermediate 64 (75 mg, 0.38 mmol), and KOH (64 mg, 1.14 mmol) were dissolved in 3 mL
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
EtOH and 1 mL H O. Following general protocol A, 2ꢀ(4ꢀ(pyrimidinꢀ5ꢀyl)phenyl)quinolineꢀ4ꢀ
2
1
carboxylic acid was recovered as a yellow solid (24 mg, 0.07 mmol, 18% yield). H NMR (300
MHz, DMSOꢀd ) δ 9.31 – 9.21 (m, 3H), 8.66 (d, J = 8.5 Hz, 1H), 8.56 (s, 1H), 8.48 (d, J = 8.3
6
Hz, 2H), 8.21 (d, J = 8.4 Hz, 1H), 8.05 (d, J = 8.4 Hz, 2H), 7.94 – 7.85 (m, 1H), 7.79 – 7.68 (m,
+
ꢀ
1
H); LCMS (ESI) 328.15 [M+H] , 326.85 [MꢀH] ; HPLC Purity at 254 nm, 99.5%.
2ꢀ(4ꢀ(Pyridazinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (20): (Isatin (203 mg, 1.38 mmol),
intermediate 65 (273 mg, 1.38 mmol), and KOH (309 mg, 5.52 mmol) were dissolved in 10 mL
EtOH. Following general protocol A, 2ꢀ(4ꢀ(pyridazinꢀ3ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid
1
was recovered as a tan solid (160 mg, 0.49 mmol, 36%). H NMR (400 MHz, DMSOꢀd ) δ 9.25
6
(dd, J = 4.8, 1.5 Hz, 1H), 8.66 (d, J = 8.5 Hz, 1H), 8.56 (s, 1H), 8.50 (d, J = 8.3 Hz, 2H), 8.42 –
8
3
.29 (m, 3H), 8.20 (d, J = 8.4 Hz, 1H), 7.91 – 7.79 (m, 2H), 7.74ꢀ7.70 (m, 1H); LCMS (ESI)
+
ꢀ
27.85 [M+H] , 325.85 [MꢀH] ; HPLC Purity at 254 nm, 96.7%.
2
ꢀ(4ꢀ(Pyrimidinꢀ2ꢀyl)phenyl)quinolineꢀ4ꢀcarboxylic acid (21): Isatin (67 mg, 0.46 mmol),
intermediate 66 (109 mg, 0.55 mmol), KOH (103 mg, 1.84 mmol) were dissolved in 3 mL EtOH
and 1 mL H O. Following general protocol A, 2ꢀ(4ꢀ(pyrimidinꢀ2ꢀyl)phenyl)quinolineꢀ4ꢀ
2
1
carboxylic acid was recovered as a tan solid (80 mg, 0.24 mmol, 52%). H NMR (400 MHz,
DMSOꢀd6) δ 8.96 (d, J = 4.8 Hz, 2H), 8.67 (d, J = 8.2 Hz, 1H), 8.61 – 8.54 (m, 3H), 8.50 – 8.46
ACS Paragon Plus Environment