Identification of novel quinazolinedione derivatives as RORγt inverse agonist

Article abstract of DOI:10.1016/j.bmc.2017.12.039

Novel small molecules were synthesized and evaluated as retinoic acid receptor-related orphan receptor-gamma t (RORγt) inverse agonists for the treatment of inflammatory and autoimmune diseases. A hit compound, 1, was discovered by high-throughput screeni

Full text of DOI:10.1016/j.bmc.2017.12.039

Accepted Manuscript
Identification of novel quinazolinedione derivatives as RORγt inverse agonist
Yoshiyuki Fukase, Ayumu Sato, Yoshihide Tomata, Atsuko Ochida, Mitsunori
Kono, Kazuko Yonemori, Keiko Koga, Toshitake Okui, Masashi Yamasaki,
Yasushi Fujitani, Hideyuki Nakagawa, Ryoukichi Koyama, Masaharu
Nakayama, Robert Skene, Bi-Ching Sang, Isaac Hoffman, Junya Shirai, Satoshi
Yamamoto
PII:
S0968-0896(17)31967-3
https://doi.org/10.1016/j.bmc.2017.12.039
BMC 14144
DOI:
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
9 October 2017
22 December 2017
24 December 2017
Please cite this article as: Fukase, Y., Sato, A., Tomata, Y., Ochida, A., Kono, M., Yonemori, K., Koga, K., Okui,
T., Yamasaki, M., Fujitani, Y., Nakagawa, H., Koyama, R., Nakayama, M., Skene, R., Sang, B-C., Hoffman, I.,
Shirai, J., Yamamoto, S., Identification of novel quinazolinedione derivatives as RORγt inverse agonist, Bioorganic
& Medicinal Chemistry (2017), doi: https://doi.org/10.1016/j.bmc.2017.12.039
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Identification of novel quinazolinedione derivatives as RORγt inverse agonist
Yoshiyuki Fukase a,1, Ayumu Satoa,2,*, Yoshihide Tomata a, Atsuko Ochida a, Mitsunori Kono a,
Kazuko Yonemori a, Keiko Koga a, Toshitake Okui a, Masashi Yamasaki a,2, Yasushi Fujitani a,
Hideyuki Nakagawa a, Ryoukichi Koyama a, Masaharu Nakayama a, Robert Skeneb, Bi-Ching Sang b,
Isaac Hoffman b, Junya Shirai a, and Satoshi Yamamoto a.
aPharmaceutical Research Division, Takeda Pharmaceutical Company Ltd, 26-1, Muraokahigashi
b
2-chome, Fujisawa, Kanagawa 251-8555, Japan. Takeda California, Inc., 10410 Science Center
Drive, San Diego, California 92121, United States.
1Current address: Tri-Institutional Therapeutics Discovery Institute. 2Current address: Axcelead
Drug Discovery Partners, Inc.

Abstract
Novel small molecules were synthesized and evaluated as retinoic acid receptor-related orphan
receptor-gamma t (RORγt) inverse agonists for the treatment of inflammatory and autoimmune
diseases. A hit compound, 1, was discovered by high-throughput screening of our compound library.
The structure–activity relationship (SAR) study of compound 1 showed that the introduction of a
chlorine group at the 3-position of 4-cyanophenyl moiety increased the potency and a
3-methylpentane-1,5-diamide linker is favorable for the activity. The carbazole moiety of 1 was also
optimized; a quinazolinedione derivative 18i suppressed the increase of IL-17A mRNA level in the
lymph node of a rat model of experimental autoimmune encephalomyelitis (EAE) upon oral
administration. These results indicate that the novel quinazolinedione derivatives have great
potential as orally available small-molecule RORγt inverse agonists for the treatment of Th17-driven
autoimmune diseases. A U-shaped bioactive conformation of this chemotype with RORγt protein
was also observed.
Key words: RORγt (retinoic acid receptor-related orphan receptor gamma t), Inverse agonist, IL-17,
Th17, Quinazolinedione, EAE model.

Abbreviations: AUC, area under the plasma concentration-time curve; BA, bioavailability; EAE,
experimental autoimmune encephalomyelitis; LBD, ligand-binding domain; RORγt, retinoic acid
receptor-related orphan receptor-gamma t; SAR, structure–activity relationship; T3P,
propylphosphonic anhydride; TR-FRET, time-resolved fluorescence resonance energy transfer.
1.
Introduction
Th17 cells and inflammatory cytokines (IL-17A, IL-17F, and IL-23) have been associated with
various autoimmune diseases such as inflammatory bowel disease (IBD), rheumatoid arthritis (RA),
multiple sclerosis (MS), and psoriasis.[1-10] Autoimmune diseases decrease quality of life (QOL) of
patients as a result of the enhancement of an abnormal systemic immune response. Because existing
therapeutic drugs show only limited effects, other novel therapeutic drugs should be developed.
Therefore, the involvement of Th17 cells and inflammatory cytokines in the pathology of these
autoimmune diseases has drawn attention in recent years.[4, 11] Moreover, a retinoic acid
receptor-related orphan receptor-gamma t (ROR γt), one of the member of nuclear receptors, plays
an important role in the differentiation of Th17 cells and production of IL-17A/IL-17F and
IL-23.[12] RORγt is mainly expressed in Th17 cells and functions as a transcription factor of IL-17A
and IL-17F, as well as a master regulator of Th17 cells differentiation.[13] Deletion of RORγt results
in impaired Th17 differentiation in murine and human cells. RORγt-knockout (KO) mice are
resistant to various autoimmune diseases such as colitis and the experimental autoimmune
encephalomyelitis (EAE). Therefore, a drug that inhibits the action of RORγt can be used in the
treatment of various autoimmune diseases by suppressing the differentiation and activation of Th17
cells.[14-18] We started the investigation of novel orally available RORγt inverse agonists. A
number of RORγt inhibitors have been reported. Some of them showed efficacy for disease models,
and a number of compounds such as VTP-43742, JNJ-3534, ARN-6039, AZD-0284, and JTE-451
are under clinical trials. [19-45] Examples of the structures of known RORγt inverse agonists are
shown in Figure 1.
During the investigation of novel RORγt inverse agonists, a screening of our compound library
showed that carbazole compound 1 has potent RORγt inverse agonist activity (Figure 1). Compound
1 has a unique structure distinct from the known RORγt inverse agonists. This compound displayed
>1000 fold selectivity over other isoforms of human ROR (ROR-alpha and ROR-beta) and more

than >258 fold selectivity over twelve nuclear receptors (AR, ER-alpha, ER-beta, FXR, GR, MR,
PPAR-gamma, RAR-alpha, RAR-beta, RAR-gamma, RXA-alpha and RXR-beta) in cell assay panel
in agonist or antagonist mode. We expected that a series of compounds derived from compound 1
may provide a novel class of RORγt inverse agonists, which could be effective against Th17-driven
autoimmune diseases. Compound 1 was selected for the design of novel RORγt inverse agonists.
Because compound 1 has already shown potent activity in the binding and reporter gene assays, we
investigated the structure–activity relationship (SAR) of each region (scaffold A) divided into three
regions, left-hand-side fused ring (a), central linker (b), and right-hand-side ring (c), to generate the
optimized compounds that were effective in an animal model. In this paper, we describe the
synthesis, SAR, and scaffold hopping of carbazole ring, leading to potent inverse agonists of RORγt
that inhibited IL-17 production in the EAE-model rats. The crystal structure of this chemical series
with RORγt protein (C-terminal truncated) that showed a unique mode of binding to the RORγt
ligand-binding domain (LBD) is also described.
Figure 1. Published RORγt inhibitors and hit compound 1 from Takeda compound library

2. Chemistry
Scheme 1 shows the synthesis of carbazole derivatives with various linkers 5a–e and 6.
Dehydration of dicarboxylic acid 2 afforded anhydride 3e. Carboxylic acid derivatives 4a–e were
prepared from 3e or commercially available acid anhydrides 3a–d by ring-opening amidation with
4-cyanoaniline. Amidation of carboxylic acids 4a–e with 9-ethyl-9H-carbazol-3-amine afforded the
corresponding carbazole derivatives 5a–e. N-methylated compound 6 was synthesized by the
methylation of compound 1.
Scheme 1. Synthesis of compounds 5a–e and 6
Anilines 7a–g were treated with 4-methyldihydro-2H-pyran-2,6(3H)-dione, affording the
corresponding carboxylic acids 8a–g (Scheme 2). Amidation of carboxylic acids 8a–f with
9-ethyl-9H-carbazol-3-amine gave carbazole derivatives 9a–f. The chiral resolution of 9f by chiral
preparative HPLC furnished 10a and 10b.

Scheme 2. Synthesis of carbazole derivatives 9a–f, and chiral resolution of 9f
The synthesis of anilines 13a–e is shown in Scheme 3. The treatment of 2-amino-5-nitrophenol
(11) with methyl 2-bromobutyrate resulted in a ring formation, affording benzoxazine 12. Ethylation
of the amide group of 12 followed by the reduction of the nitro group resulted in aniline 13a.
Quinazolinedione 15a was prepared from 2-amino-5-nitrobenzoic acid (14) and urea. Compound
15a was alkylated using ethyl iodide or (bromomethyl)cyclopropane, and subsequent reduction of
the nitro group produced anilines 13b and 13c, respectively. Amidation of acid 14 and ring
formation afforded 15b. Compound 15b was converted to 13d in a similar fashion. Compound 16
was subjected to SNAr reaction with ethylamine in the presence of potassium carbonate; subsequent
cyclization with urea afforded quinazolinedione 17. Alkylation of 17 with
(bromomethyl)cyclopropane and subsequent reduction of the nitro group afforded aniline 13e.
Condensation of acid 8g or 8f with various amines produced diamide derivatives 18a–k (Scheme 4).

Scheme 3. Synthesis of fused ring anilines 13a-e

Scheme 4. Synthesis of fused ring derivatives 18a–k
3. Results and discussion
The synthesized compounds were primarily evaluated by binding and reporter gene assays in vitro.
A fluorescent-labeled synthetic ligand discovered by our group was used to measure the affinity of
the test compounds to human RORγt by time-resolved fluorescence resonance energy transfer
(TR-FRET, binding assay, cell-free system).[46] Human Jurkat cells were used to evaluate the
functional ability of the test compounds to affect the transcriptional activity of RORγt by measuring
the luciferase activity (reporter gene assay, cell-based). Cassette dosing pharmacokinetic (PK)
studies in mice were performed to determine the oral availability of selected compounds by
measuring the areas under the curve (AUC) 0–8h values at 1 mg/kg.

Table 1. SAR of linker part
Bindinga IC50 (nM)
[95%CI]b
Reporterc IC50 (nM)
Compound
Linker
[95%CI]b
7.5
92
1d
5a
5b
[6.7–8.4]
[68–120]
4.8
200
[4.4–5.3]
[140–280]
410
>30000
[340–500]
140
>30000
5c
5d
5e
5f
[70–280]
290
7800
[3300–18000]
1600
[230–360]
57
[49–65]
81
[1000–2500]
4200
[74–88]
6700
[2400–7200]
15000
6d
[5100–8800]
[7600–28000]
aDisplacement assay of the fluorescent-labeled synthetic ligand with apo RORγt.
b95% confidence intervals (CI) in duplicate (n = 2).
cInhibition of RORγt transcriptional activity in human Jurkat cells. dRacemic.
First, SAR trends for the linker parts were investigated (Table 1). Initially, the effect of linker
length was investigated. Compounds 5a, 5b, and 5c bearing unsubstituted alkylene tethers were
synthesized and evaluated. Regarding the length of the linker, seven-atom linkage (5a) was found to
be superior to six-atom linkage (5b) and eight-atom linkage (5c) in the binding assay. For the
seven-atom length compounds, the presence of one methyl group at the center of the linker slightly
increased the reporter activity (1 vs. 5a). On the other hand, the introduction of one more methyl
group largely decreased the binding and reporter activities (5d). The introduction of a nitrogen atom
(5e) or an oxygen atom (5f) to the center of the linker did not affect the binding affinity, but the
reporter activity decreased. N-methylation of both the amides on the linker significantly decreased

the binding affinity (6). It is noteworthy that there were discrepancies in SAR trends between
binding and reporter assay. The function of the ligand was affected by linker moiety. This SAR
showed that N-unsubstituted 3-methylpentanediamide (1) is the most favorable linker in terms of the
inverse agonistic activities.
Then, the right-hand-side ring was investigated (region c, Table 2). First, the effect of the cyano
group on the phenyl ring was evaluated. Transposition of the 4-cyano group to 3- (9a) or 2-position
(9b) decreased the binding affinity. Removal of the cyano group was not tolerated (9c), and
replacement of the cyano group to a fluorine group also decreased the potency (9d). These results
indicate that the cyano group located at the 4-position has an important role in the biological activity.

Table 2. SAR of right hand sidea
Compound
R1
R2
R3
Bindingb
IC50 (nM)
[95%CI]c
7.5
Reporterd
IC50 (nM)
[95%CI]c
92
CN
H
H
H
H
1
[6.7–8.4]
190
[68–120]
6300
CN
9a
[160–210]
>10000
6300
[3800–10000]
>30000
H
H
H
H
CN
H
9b
9c
>30000
[4900–8100]
840
F
H
CF3
Cl
H
H
H
H
H
>30000
9d
9e
[750–940]
15
CN
CN
CN
CN
190
[140–270]
11
[13–17]
2.1
9f
[1.6–2.8]
1.3
[7.8–16]
13
10ae
10bf
Cl
[1.2–1.5]
1.6
[9.3–18]
12
Cl
[1.5–1.7]
[9.2–17]
aAll the compounds are racemic except for 10a and 10b. bDisplacement assay of the
fluorescent-labeled synthetic ligand with apo RORγt. c95% confidence intervals (CI) in duplicate (n
= 2). dInhibition of RORγt transcriptional activity in human Jurkat cells. e(+)-isomer. f(-)-isomer
Then, the effect of the substituent adjacent to the cyano group was evaluated.
4-Cyano-3-trifluorophenyl compound 9e did not show a higher activity. In contrast, the introduction
of a chloro group (9f) increased both the binding and reporter activities. Based on these results,
3-chloro-4-cyanophenyl group was selected as the structure of region c. Compound 9f had a chiral
center on the central methyl group; thus, chiral resolution of the compound was performed to
determine the eutomer. Interestingly, the resulting single enantiomers 10a and 10b were equally

potent. Thus, further investigation was continued with the racemic compounds.
Compound 9f showed very low aqueous solubility (0.15 μg/mL, Table 3). Because the poor
physicochemical properties such as solubility for oral availability could be attributed to the left hand
side of the carbazole moiety (region a), the chemical hopping of this region was investigated. We
supposed that the highly aromatized and planer carbazole ring and a relatively high lipophilicity
(logD at pH 7.4 of 9f: 3.76) of the molecule might cause the low solubility, and the exchange of the
tricyclic carbazole moiety to a smaller aromatic ring with reduced hydrophobicity can provide more
soluble molecules. Initially, the structural requirement for the tricyclic ring was evaluated. Removal
of the ethyl group significantly decreased the potency (18a vs. 18b), indicating that the presence of a
substituent on the nitrogen atom highlighted by a red circle is necessary (Figure 2). Conversion of
the terminal benzene ring to the cyclohexyl ring was tolerated (18a). On the other hand, removal of
the terminal benzene ring remarkably decreased the activity (18c). These results indicate that the
presence of the moiety highlighted by a blue circle is important for the biological activities (Figure
2). Based on these SARs, we expected that the replacement of the carbazole moiety with a bicyclic
system possessing substituents X and Y will afford a new terminal scaffold with reduced
hydrophobicity. Conversion of the indole ring to carbonyl containing ring was tolerated (18c, 18d,
18e). The addition of an alkyl group to 18e increased the activities (18f, 18g). These two compounds
were less lipophilic and soluble than 9f (logD at pH 7.4 of 18f and 18h: 2.66 and 2.99).
Figure 2. Hopping design from carbazole compounds

Modification of 18g afforded quinazolinedione 18h, further improving the solubility as well as
activity. These results encouraged us to evaluate the effects of substituents on the quinazolinedione
ring.
Table 3. Conversion of the left hand sidea
Compound
R1
R2
Bindingb
IC50 (nM)
[95%CI]c
2.1
Reporterd
IC50 (nM)
[95%CI]c
11
Solubility
(μg/mL)e
Cl
H
0.15
0.06
9f
[1.6–2.8]
[7.8–16]
5.9
57
18a
[4.1–8.5]
[43–76]
5500
[4400–6800]
170
7700
[4200–14000]
4000
H
0.19
0.1
0.9
12
18b
18c
18d
18e
18f
Cl
Cl
Cl
Cl
[150–200]
260
[1900–8500]
5200
[240–280]
110
[2800–10000]
1400
[73–160]
13
[930–2000]
200
23
[11–15]
[80–490]
7.5
100
Cl
Cl
5.6
42
18g
18h
[6.8–8.3]
[41–260]
3.6
47
[2.7–4.6]
[32–68]
aRacemic. bDisplacement assay of the fluorescent-labeled synthetic ligand with apo RORγt. c95%
confidence intervals (CI) in duplicate (n = 2). dInhibition of RORγt transcriptional activity in human
Jurkat cells. eKinetic solubility in the 2nd fluid of disintegration test of the Japanese Pharmacopoeia
(pH 6.8).

Optimization of the substituents on the quinazolinedione ring was performed (Table 4). An
increase in the reporter activity was observed when a cyclopropylmethyl group was present at the
3-position (18k, reporter IC50 = 5 nM). In contrast, the presence of a cyclopropylmethyl group at the
1-position (R2) did not increase the activity (18j). Similar to compound 18k, the introduction of a
cyclopropylmethyl group at the 3-position of the mono-cyclopropylmethyl derivative 18j increased
the reporter activity (18i). The reporter activity of compound 18i was comparable to that of 18k.
These results indicate that the terminal cyclopropyl group located at the 3-position (R1) is more
sensitive to the reporter activity of the compound than that at the 1-position (R2). Interestingly in the
case of 3-cyclopropylmethyl derivatives, 10-fold increase of the AUC values in rat PK study was
observed by introducing a cyclopropylmethyl group at the 1-position (18i vs. 18k). Combination of
the two cyclopropylmethyl groups at R1 and R2 is important for the AUC of the compound. These
compounds showed relatively low BA at the dose of 1 mg/kg because these compounds were
P-glycoprotein substrates. At a higher dose, P-glycoprotein would be saturated and concentration of
the drug would be more dependent on in vivo clearance. Because compound 18i had lowest in vivo
clearance among them and potent in the reporter assay, 18i was selected for further in vitro and in
vivo studies.

Table 4. Optimization of the quinazolinedione derivatives and their pharmacokinetic
parametersa
ID
R1
Et
R2
Bindingb
Reporterd
Rat PKe
CL
IC50 (nM) IC50 (nM)
AUC, po
BA
(%)
[95%CI]c
3.6
[95%CI]c
47
(ng•h/mL)
(mL/h/kg)
Et
Et
39
1609
6.2
18h
[2.7–4.6]
[32–68]
2.5
5.0
10
33
2639
1702
672
2.7
5.5
7.1
18k
18j
18i
[1.9–3.3]
[2.2–11]
3.9
32
Et
[3.2–4.9]
4.9
[25–41]
7.2
106.2
[4.4–5.4]
[2.7–19]
aRacemic. bDisplacement assay of the fluorescent-labeled synthetic ligand with apo RORγt. c95%
confidence intervals (CI) in duplicate (n = 2). dInhibition of RORγt transcriptional activity in human
Jurkat cells. eThe values are shown as means of three determinations (0.1 mg/kg, iv, and 1.0 mg/kg,
po).
The inhibitory activity of compound 18i on IL-17 production in rat whole blood was evaluated
using phorbol 12-myristate 13-acetate (PMA) and ionomycin stimulation. Compound 18i suppressed
the production of IL-17A with an IC50 of 57 nM. The in vivo activity of compound 18i was evaluated
by the rat EAE model, which is generally used as a model of Th17-cell-mediated autoimmune
diseases (Figure 4). The compound was orally administrated in 30, 100, or 300 mg/kg dosages to
immunized rats from day 0 to day 5. On day 5, gene expression analysis of the homogenized extracts
of lymph nodes indicated the dose-dependent suppression of IL-17A, and 100 mg/kg was the
minimal effective dose (Figure 4, A). Importantly, no notable effect was observed on the INFγ levels
in any dose level (Figure 4, B), indicating that 18i did not affect the cytokines outside of the RORγt

pathway. Taken together, these findings indicate that the quinazolinedione derivatives have a great
potential to become a novel class of small-molecule RORγt inverse agonists for the treatment of
Th17-driven autoimmune diseases.
A
B
Figure 4. Assessment of 18i in rat EAE model. (A) Dose dependent suppression of IL-17A mRNA.
(B) Effect of 18i on IFNγ level.
An X-ray structure analysis of the complex structure of 18h bound to the RORγt protein was also
carried out. Compound 18h was located on the cholesterol-binding pocket. Interestingly, the
compound adopts a U-shaped conformation in the binding pocket (Figure 5A). Because the inverse
agonist digoxin shows hydrogen-bonding interactions with arginines (i.e., Arg364 and Arg367) and
His479 in the LBD (Figure 5B),[47] we expected that 18h can also bind both to His479 and Arg
residues or either of them. In fact, compound 18h interacted with only His479 among these residues,
whereas the Arg residues were away from 18h due to the U-shaped conformation. Interactions
between His479 and the cyano group of 18h are considered as one of the key factors affecting the
RORγt inhibitory activity.[47] The hydrogen-bond formation between small-molecule ROR inverse
agonist and His479 was also observed in several co-crystal structures recently reported by
Astrazeneca, Biogen, and Boehringer Ingelheim,[41, 42, 48, 49] and the possible mechanism for the
inverse agonism was discussed in detail. This binding mode was consistent with the biological data:
The removal of the cyano group (compound 9c) resulted in the loss of inverse agonist activity.
Furthermore, in the complex structure, the ethyl group located on the 1-position of the
quinazolinedione ring was oriented to a space consisting of lipophilic residues such as Phe388. The
U-shaped form would be the consequence of the-stacking interactions of the benzene rings and

packing of the linker.[50-54] The bis-amide linker packed on the top of each other, making direct
and water-mediated hydrogen bonds with the backbone of Phe377 and Glu379. N-methylation of the
amides of the linker (compound 6) destroyed the hydrogen interactions with Phe377 and also might
destabilize the U-shaped conformation, thus decreasing the potency.[55] Elucidation of this
U-shaped conformation led us to further modify the linker of the quinazolinedione analog 18h using
various conformationally constrained linkers that enforce the U-shaped conformation. The results of
this effort will be the subject of another report from our group.
A
B
Arg367
3.1
Arg367
Phe377
Val361
His479
Phe377
His479
2.7
2.9
3.4
3.0
Glu379
Phe388
Glu379
Leu391
Figure 5. A: Structure of inverse agonist 18h complex with the C-terminal truncated RORγt protein.
Hydrogen bonds are marked as dotted lines. B: Reported crystal structure of the RORγt protein LBD
in complex with digoxin (PDB code:3B0W).[47]
4. Conclusion
Starting from a hit compound bearing a carbazole ring discovered by an HTS campaign, novel
orally available RORγt inverse agonists were investigated using SAR studies on the three
components of the hit compound. Replacement of the carbazole ring was effective in improving the
physicochemical properties, resulting in the identification of quinazolinedione ring (18h) that
retained the activity with a high solubility. By tuning the PK profiles of 18h, 18i with a decent
plasma exposure upon oral administration in rats was discovered; it was effective in a rat EAE model.

These results indicate that the quinazolinedione derivatives have a great potential to become a novel
class of small-molecule RORγt inverse agonists for the treatment of Th17-driven autoimmune
diseases. In addition, 18h exists in a U-shaped conformation as a result of the -stacking interactions
of the benzene rings and packing of the linker. Elucidation of the conformation led us to synthesize
quinazolinedione derivatives bearing a conformationally constrained linker to enforce the U-shaped
conformation. The results of this effort will be the subject of another report from our group.
5. Experimental section
5.1 Chemistry
5.1.1. General
Reagents and solvents were obtained from commercial sources and used without further
purification. Reaction progress was determined by thin layer chromatography (TLC) analysis on
Merck Kieselgel 60 F254 plates or Fuji Silysia NH plates. Silica gel column chromatography was
performed on Purif-Pack (SI or NH, SHOKO SCIENTIFIC). Proton nuclear magnetic resonance (1H
NMR) spectra were recorded on Bruker Ultra Shield-300 (300 MHz) instruments. Chemical shifts
are given in parts per million (ppm) with tetramethylsilane as an internal standard. Abbreviations are
used as follows: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, dd = doublets of
doublet, br = broad. Coupling constants (J values) are given in hertz (Hz). LC−MS analysis was
performed on a Shimadzu LC-20AD separations module, Agilent 1200 series, or Applied
Biosystems API 2000, operating in ESI (+ or −) ionization mode. Analytes were eluted using a linear
gradient of 0.05% TFA containing H2O/CH3CN or 0.01% TFA containing H2O/CH3CN or 5 mM
ammonium acetate containing H2O/CH3CN or 10 mM NH4HCO3 containing H2O/CH3CN or 0.01%
heptafluorobutyric acid/1.0% isopropyl alcohol containing H2O/CH3CN mobile phase. Preparative
HPLC was performed on a Waters 2525 separations module (L-column2 ODS (20 × 150 mm ID),
CERI, Japan) or Waters auto purification instrument (X Terra RP18 OBD Prep Column (19 × 250
mm ID); MS spectra were recorded using a Waters ZQ2000 with electrospray ionization or on a
GILSON system, equipped with a L-column2 ODS (20 × 150 mm ID, CERI, Japan) or on a Waters
Deltaprep 300 system or on a Shimadzu 10A VP system. Samples were eluted using a linear gradient
of 0.1% TFA in H2O/CH3CN or 10 mM NH4HCO3 in H2O/CH3CN, or H2O/CH3CN. Purity data
were collected by a HPLC with Corona CAD (Charged Aerosol Detector), Nano quantity analyte

detector (NQAD), or photo diode array detector. The column was a Capcell Pak C18AQ (50 mm ×
3.0 mm ID, Shiseido, Japan) or L-column 2 ODS (30 mm × 2.0 mm ID, CERI, Japan) with a
temperature of 50 °C and a flow rate of 0.5 mL/min. Mobile phases A and B under a neutral
condition were a mixture of 50 mM ammonium acetate, H2O, and CH3CN (1:8:1, v/v/v) and a
mixture of 50 mM ammonium acetate and CH3CN (1:9, v/v), respectively. The ratio of mobile phase
B was increased linearly from 5% to 95% over 3 min, 95% over the next 1 min. Mobile phases A
and B under an acidic condition were a mixture of 0.2% formic acid in 10 mM ammonium formate
and 0.2% formic acid in CH3CN, respectively. The ratio of mobile phase B was increased linearly
from 14% to 86% over 3 min, 86% over the next 1 min. All final compounds were purified to >95%
purity unless otherwise noted as determined by analytical HPLC. Elemental analyses (Anal.) was
carried out at Takeda Analytical Laboratories, Ltd. Yields were not optimized.
5.1.2. 5-[(4-cyanophenyl)amino]-5-oxopentanoic acid (4a). A mixture of dihydro-2H-pyran-
2,6(3H)-dione 3a (483 mg, 4.23 mmol) and 4-aminobenzonitrile (500 mg, 4.23 mmol) in THF (15
mL) was refluxed overnight. The resulting precipitations were filtered and rinsed with EtOAc to give
4a (782 mg, 3.37 mmol, 80 %) as pale yellow powder. MS (ESI) m/z: 233.2 (M+H)+. 1H NMR (300
MHz, DMSO-d6): δ 1.81 (2H, m), 2.17–2.35 (2H, m), 2.41 (2H, t, J = 7.4 Hz), 7.56–7.96 (4H, m),
10.33 (1H, s), 12.08 (1H, br s).
5.1.3. 4-[(4-cyanophenyl)amino]-4-oxobutanoic acid (4b). Compound 4b was prepared from 3b in
1
52% yield by a procedure similar to that described for 4a as an off-white powder. H NMR (300
MHz, DMSO-d6): δ 2.53–2.69 (4H, m), 7.54–7.92 (4H, m), 10.40 (1H, s), 12.15 (1H, br s).
5.1.4. 6-[(4-cyanophenyl)amino]-6-oxohexanoic acid (4c). Compound 4c was prepared from 3c in
90% yield by a procedure similar to that described for 4a as a colorless powder. MS (ESI) m/z:
247.1 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.42–1.70 (4H, m), 2.13–2.30 (2H, m), 2.36 (2H,
t, J = 7.0 Hz), 7.58–7.95 (4H, m), 10.32 (1H, d, J = 3.0 Hz), 12.00 (1H, br s).
5.1.5. 5-[(4-cyanophenyl)amino]-3,3-dimethyl-5-oxopentanoic acid (4d). Compound 4d was
prepared from 3d in quantitative yield by a procedure similar to that described for 4a as a pale
yellow solid. 1H NMR (300 MHz, DMSO-d6): δ 1.00–1.14 (6H, m), 2.25–2.47 (4H, m), 6.10 (1H, br
s), 6.51–6.69 (1H, m), 7.29–7.46 (1H, m), 7.76 (2H, d, J = 3.0 Hz), 11.99 (1H, br s).

5.1.6. ({2-[(4-cyanophenyl)amino]-2-oxoethyl}(methyl)amino)acetic acid (4e).
Step 1: 4-Methylmorpholine-2,6-dione (3e). A mixture of 2e (500 mg, 3.40 mmol) and acetic
anhydride (5 mL) was stirred at 165 °C for 30 min. The reaction mixture was cooled to rt and
evaporated in vacuo to give 3e as brown oil. This compound was used for next reaction without
further purification.
Step 2: ({2-[(4-Cyanophenyl)amino]-2-oxoethyl}(methyl)amino)acetic acid (4e). A mixture of
3e (439 mg, 3.40 mmol), p-aminobenzonitrile (401 mg, 3.40 mmol) in THF (10 mL) was refluxed for
22 h. The reaction mixture was evaporated in vacuo, and the precipitates were washed with MeOH and
Et2O to afford 4e (467 mg, 56%, from 2e) as a light brown solid. MS (ESI) m/z: 248.1 (M+H)+. 1H
NMR (300 MHz, DMSO-d6): δ 2.42 (3H, s), 3.41 (4H, s), 7.76–7.85 (4H, m), 10.30 (1H, s).
5.1.7. N-(4-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)pentanediamide (5a). To a solution of 4a
(200 mg, 0.86 mmol), 9-ethyl-9H-carbazol-3-amine (217 mg, 1.03 mmol) and DIEA (0.451 mL,
2.58 mmol) in DMF (6 mL) was added HATU (491 mg, 1.29 mmol), and the mixture was stirred at
rt overnight. The mixture was quenched with water at rt and extracted with EtOAc. The organic
layer was separated, washed with water and brine, dried over MgSO4 and concentrated in vacuo. The
residue was purified by column chromatography (silica gel, eluted with 5%−75% EtOAc in hexane),
followed by recrystallization from EtOAc to give 5a (229 mg, 0.540 mmol, 63%) as colorless
powder. MS (ESI) m/z: 425.2 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.30 (3H, t, J = 7.0 Hz),
1.85–2.11 (2H, m), 2.31–2.44 (1H, m), 2.49–2.63 (3H, m), 4.41 (2H, q, J = 6.8 Hz), 7.17 (1H, t, J =
7.4 Hz), 7.36–7.65 (4H, m), 7.69–7.90 (4H, m), 8.04 (1H, d, J = 7.6 Hz), 8.43 (1H, s), 9.92 (1H, s),
10.37 (1H, s).
5.1.8. N-(4-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)succinamide (5b). Compound 5b was
prepared from 4b in 71% yield by a procedure similar to that described for 5a as a colorless powder.
MS (ESI) m/z: 411.2 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.30 (3H, t, J = 7.0 Hz), 2.60-2.82
(4H, m), 4.41 (2H, q, J = 6.8 Hz), 7.16 (1H, t, J = 7.4 Hz), 7.43 (1H, t, J = 7.4 Hz), 7.50–7.64 (3H,
m), 7.69–7.89 (4H, m), 8.04 (1H, d, J = 8.0 Hz), 8.42 (1H, s), 10.02 (1H, s), 10.48 (1H, s).
5.1.9. N-(4-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)hexanediamide (5c). Compound 5c was
prepared from 4c in 14% yield by a procedure similar to that described for 5a as a colorless powder.
MS (ESI) m/z: 439.1 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.30 (3H, t, J = 7.0 Hz), 1.68 (4H,
br s), 2.27–2.46 (4H, m), 4.41 (2H, q, J = 7.1 Hz), 7.07–7.25 (1H, m), 7.36–7.63 (4H, m), 7.68–7.85
(4H, m), 8.04 (1H, d, J = 7.6 Hz), 8.40 (1H, s), 9.89 (1H, s), 10.35 (1H, s).

5.1.10. N-(4-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3,3-dimethylpentanediamide (5d).
Compound 5d was prepared from 4d in 55% yield by a procedure similar to that described for 5a as
1
a pale yellow solid. MS (ESI) m/z: 453.4 (M+H)+. H NMR (300 MHz, CDCl3): δ 1.06–1.33 (6H,
m), 1.42 (3H, t, J = 7.2 Hz), 2.45 (2H, s), 2.56 (2H, s), 4.36 (2H, q, J = 7.2 Hz), 7.12-7.32 (1H, m),
7.32–7.66 (6H, m), 7.76 (2H, d, J = 8.7 Hz), 7.94 (1H, s), 8.08 (1H, d, J = 7.9 Hz), 8.25 (1H, d, J =
1.5 Hz), 10.56 (1H, s).
5.1.11. 2-({2-[(4-Cyanophenyl)amino]-2-oxoethyl}(methyl)amino)-N-(9-ethyl-9H-carbazol-3-
yl)acetamide (5e). Compound 5e was prepared from 4e in 57% yield by a procedure similar to that
1
described for 5a as a colorless powder. MS (ESI) m/z: 440.2 (M+H)+. H NMR (300 MHz,
DMSO-d6): δ 1.31 (3H, t, J = 7.2 Hz), 2.49 (3H, s), 3.43 (2H, s), 3.48 (2H, s), 4.42 (2H, q, J = 7.2 Hz),
7.19 (1H, t, J = 7.5 Hz), 7.42–7.48 (1H, m), 7.56–7.67 (3H, m), 7.81 (2H, d, J = 8.7 Hz), 7.91 (2H, d,
J = 8.7 Hz), 8.07 (1H, d, J = 7.8 Hz), 8.44 (1H, d, J = 2.1 Hz), 10.07 (1H, s), 10.55 (1H, s).
5.1.12. N-(4-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-N,N',3-trimethylpentanediamide (6).
To an ice-cooled solution of 1 (200 mg, 0.46 mmol) and DMF (3 mL) were added NaH (219 mg,
4.56 mmol) and iodomethane (0.142 mL, 2.28 mmol). After being stirred at rt overnight, the mixture
was quenched with water at rt and extracted with EtOAc. The organic layer was separated, washed
with water and brine, dried over MgSO4 and concentrated in vacuo. The residue was purified by
column chromatography (silica gel, eluted with 50−100% EtOAc in hexane) to give 6 (21.00 mg,
1
0.045 mmol, 10.0 %) as a colorless amorphous solid. MS (ESI) m/z: 467.3 (M+H)+. H NMR (600
MHz, DMSO-d6): δ 0.72 (3H, d, J = 6.2 Hz), 1.34 (3H, t, J = 7.2 Hz), 1.83 (1H, dd, J = 15.0, 7.7 Hz),
1.89 (1H, brs), 2.02 (1H, dd, J = 15.2, 5.7 Hz), 2.12 (1H, d, J = 10.6 Hz), 2.32 (1H, m), 3.13 (3H, s),
3.21 (3H, s), 4.47 (2H, q, J = 7.0 Hz), 7.22 (1H, t, J = 7.3 Hz), 7.29 (1H, dd, J = 8.4, 1.5 Hz), 7.39
(2H, d, J = 8.4 Hz), 7.49 (1H, t, J = 7.5 Hz), 7.64 (2H, d, J = 8.4 Hz), 7.73 (2H, d, J = 8.1 Hz), 8.06
(1H, d, J = 1.8 Hz), 8.16 (1H, d, J = 7.7 Hz). 13C NMR (151 MHz, DMSO-d6): δ 13.7, 19.8, 27.5,
36.6, 37.0, 37.4, 39.8, 40.0, 109.2, 109.3, 109.8, 118.3, 118.9, 119.2, 120.6, 121.8, 122.6, 124.9,
126.2, 127.7, 133.2, 135.3, 138.2, 140.0, 147.9, 170.7, 171.0.
5.1.13. 5-[(3-Cyanophenyl)amino]-3-methyl-5-oxopentanoic acid (8a).
A
mixture of
4-methyldihydro-2H-pyran-2,6(3H)-dione (1.28 g, 10.00 mmol) and 7a (1.181 g, 10 mmol) in THF
(20 mL) was stirred at 90 °C for 20 h. After being cooled to rt, the reaction mixture was concentrated
in vacuo. The residue was crystallized from EtOAc−hexane to give compound to give compound 4a

(2.210 g, 8.97 mmol, 90 %) as white powder. 1H NMR (300 MHz, DMSO-d6): δ 0.96 (3 H, d, J = 6.0
Hz), 2.05–2.45 (5 H, m), 7.43–7.62 (2 H, m), 7.78 (1 H, dt, J = 6.7, 2.3 Hz), 8.10 (1 H, s), 10.25 (1 H,
s), 12.10 (1 H, br s). Anal. Calcd for C13H14N2O3: C, 63.40; H, 5.73; N, 11.38. Found: C, 63.35; H,
5.77; N, 11.28.
5.1.14. 5-[(2-Cyanophenyl)amino]-3-methyl-5-oxopentanoic acid (8b). Compound 8b was
prepared from 7b in 91% yield by a procedure similar to that described for 8a as a white powder. 1H
NMR (300 MHz, DMSO-d6): δ 0.99 (3 H, d, J = 5.7 Hz), 2.08–2.22 (1 H, m), 2.22–2.46 (4 H, m),
7.30–7.41 (1 H, m), 7.52 (1 H, d, J = 7.9 Hz), 7.63–7.73 (1 H, m), 7.80 (1 H, d, J = 7.9 Hz), 10.16 (1
H, s), 12.10 (1 H, br s). Anal. Calcd for C13H14N2O3: C, 63.40; H, 5.73; N, 11.38. Found: C, 63.38;
H, 5.77; N, 11.33.
5.1.15. 5-Anilino-3-methyl-5-oxopentanoic acid (8c). Compound 8c was prepared from 7c in 89%
1
yield by a procedure similar to that described for 8a as a white powder. H NMR (300 MHz,
DMSO-d6): δ 0.95 (3 H, d, J = 6.0 Hz), 2.02–2.43 (5 H, m), 6.96–7.08 (1 H, m), 7.28 (2 H, t, J = 7.9
Hz), 7.58 (2 H, d, J = 7.6 Hz), 9.88 (1 H, s), 12.10 (1 H, br s). Anal. Calcd for C12H15NO3: C, 65.14;
H, 6.83; N, 6.33. Found: C, 65.04; H, 6.91; N, 6.28.
5.1.16. 5-[(4-Fluorophenyl)amino]-3-methyl-5-oxopentanoic acid (8d). Compound 8d was
prepared from 7d in 76% yield by a procedure similar to that described for 8a as a white powder. 1H
NMR (300 MHz, DMSO-d6): δ 0.95 (3 H, d, J = 6.1 Hz), 2.00–2.43 (5 H, m), 7.06–7.20 (2 H, m),
7.53–7.65 (2 H, m), 9.94 (1 H, s), 12.08 (1 H, br s). Anal. Calcd for C12H14FNO3: C, 60.24; H, 5.90;
N, 5.85. Found: C, 60.32; H, 5.93; N, 5.83.
5.1.17. 5-{[4-Cyano-3-(trifluoromethyl)phenyl]amino}-3-methyl-5-oxopentanoic acid (8e).
Compound 8e was prepared from 7e in quantitative yield by a procedure similar to that described for
1
8a as pale brown viscous oil. H NMR (300 MHz, DMSO-d6): δ 0.80–1.08 (3H, m), 2.08–2.47 (5H,
m), 7.88–8.16 (2H, m), 8.28 (1H, d, J = 1.9 Hz), 10.68 (1H, s), 12.09 (1H, s).
5.1.18. 5-[(3-Chloro-4-cyanophenyl)amino]-3-methyl-5-oxopentanoic acid (8f). Compound 8f
was prepared from 7f in quantitative yield by a procedure similar to that described for 8a as pale
brown oil. 1H NMR (300 MHz, DMSO-d6): δ 0.82–1.07 (3H, m), 2.05–2.42 (5H, m), 7.60 (1H, dd, J
= 8.7, 1.9 Hz), 7.88 (1H, d, J = 8.7 Hz), 8.06 (1H, d, J = 1.5 Hz), 10.50 (1H, s), 12.08 (1H, s).

5.1.19. 5-[(4-Cyanophenyl)amino]-3-methyl-5-oxopentanoic acid (8g). Compound 8g was
prepared from 7g in 87% yield by a procedure similar to that described for 8a as a pale yellow
1
powder. H NMR (300 MHz, DMSO-d6): δ 0.80–1.08 (3H, m), 2.05–2.45 (5H, m), 7.57–7.97 (4H,
m), 10.34 (1H, s), 12.11 (1H, br s).
5.1.20. N-(3-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3-methylpentanediamide (9a).
A
mixture of 8a (246 mg, 1 mmol), 9-ethyl-9H-carbazol-3-amine (210 mg, 1.00 mmol),
1H-benzo[d][1,2,3]triazol-1-ol hydrate (306 mg, 2.00 mmol) and WSC (383 mg, 2.00 mmol) in DMF
(5 mL) was stirred at rt for 20 h. The reaction was poured into aqueous NaHCO3 and EtOAc. The
separated organic layer was washed with water and brine, dried over MgSO4 and concentrated. The
residue was purified by column chromatography (silica gel, eluted with 66−80% (EtOAc in hexane) to
give compound 9a (283 mg, 0.645 mmol, 65%) as a white powder. MS (ESI) m/z: 439.1 (M+H)+. mp
1
164–166 °C. H NMR (300 MHz, DMSO-d6): δ 1.03 (3 H, d, J = 6.0 Hz), 1.23–1.34 (3 H, m),
2.23–2.50 (5 H, m), 2.72–2.83 (1 H, m), 4.41 (1 H, d, J = 7.2 Hz), 7.13–7.23 (1 H, m), 7.52 (7 H, d,
J = 12.1 Hz), 8.04 (1 H, d, J = 7.5 Hz), 8.13 (1 H, s), 8.41 (1 H, s), 9.93 (1 H, s), 10.25–10.34 (1 H,
m). Anal. Calcd for C27H26N4O2· H2O: C, 72.46; H, 6.08; N, 12.52. Found: C, 72.58; H, 6.02; N,
12.35.
5.1.21.
N-(2-Cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3-methylpentanediamide
(9b).
Compound 9b was prepared from 8b in 40% yield by a procedure similar to that described for 9a as
a white powder. MS (ESI) m/z: 439.2 (M+H)+. mp 212–214 °C. 1H NMR (300 MHz, DMSO-d6): δ
1.08 (3 H, d, J = 6.4 Hz), 1.30 (3 H, d, J = 7.2 Hz), 2.23–2.47 (5 H, m), 4.41 (2 H, d, J = 7.2 Hz),
7.13–7.22 (1 H, m), 7.32–7.40 (1 H, m), 7.40–7.49 (1 H, m), 7.51–7.62 (4 H, m), 7.64–7.73 (1 H,
m), 7.77–7.87 (1 H, m), 8.05 (1 H, d, J = 7.5 Hz), 8.43 (1 H, s), 9.85–10.35 (2 H, m). Anal. Calcd for
C27H26N4O2· 0.1H2O: C, 73.65; H, 6.00; N, 12.72. Found: C, 73.51; H, 6.04; N, 12.58.
5.1.22. N-(9-Ethyl-9H-carbazol-3-yl)-3-methyl-N'-phenylpentanediamide (9c). Compound 9c
was prepared from 8c in 39% yield by a procedure similar to that described for 9a as a white powder.
MS (ESI) m/z: 414.1 (M+H)+. mp 236–238 °C. 1H NMR (300 MHz, DMSO-d6): δ 1.03 (3 H, d, J =
6.4 Hz), 1.30 (3 H, t, J = 7.2 Hz), 2.19–2.47 (5 H, m), 4.41 (2 H, q, J = 7.0 Hz), 6.98–7.07 (1 H, m),
7.17 (1 H, t, J = 7.3 Hz), 7.29 (2 H, t, J = 7.9 Hz), 7.44 (1 H, t, J = 7.2 Hz), 7.50–7.67 (5 H, m), 8.05
(1 H, d, J = 7.9 Hz), 8.42 (1 H, s), 9.93 (2 H, s). Anal. Calcd for C26H27N3O2· 0.1H2O: C, 75.19; H,
6.60; N, 10.12. Found: C, 75.17; H, 6.65; N, 10.16.

5.1.23.
N-(9-Ethyl-9H-carbazol-3-yl)-N'-(4-fluorophenyl)-3-methylpentanediamide
(9d).
Compound 9d was prepared from 8d in 27% yield by a procedure similar to that described for 9a as
a white powder. MS (ESI) m/z: 432.1 (M+H)+.m .p. 250-252 °C. 1H NMR (300 MHz, DMSO-d6): δ
1.02 (3 H, d, J = 6.4 Hz), 1.30 (3 H, t, J = 7.2 Hz), 2.21–2.47 (5 H, m), 4.41 (2 H, d, J = 7.2 Hz),
7.06–7.23 (3 H, m), 7.44 (1 H, t, J = 7.7 Hz), 7.49–7.71 (5 H, m), 8.05 (1 H, d, J = 7.5 Hz), 8.41 (1
H, s), 9.93 (1 H, s), 10.00 (1 H, s). Anal. Calcd for C26H26FN3O2· H2O: C, 70.89; H, 6.18; N, 9.54.
Found: C, 71.06; H, 6.08; N, 9.46.
5.1.24. N-[4-Cyano-3-(trifluoromethyl)phenyl]-N'-(9-ethyl-9H-carbazol-3-yl)-
3-methylpentanediamide (9e). To
a
solution of 8e (300 mg, 0.95 mmol),
9-ethyl-9H-carbazol-3-amine (241 mg, 1.15 mmol), and DIEA (0.500 mL, 2.86 mmol) in DMF (6
mL) was added HATU (544 mg, 1.43 mmol), and the mixture was stirred at rt overnight. The
mixture was quenched with water at rt and extracted with EtOAc. The organic layer was separated,
washed with water and brine, dried over MgSO4 and concentrated in vacuo. The residue was
purified by column chromatography (silica gel, eluted with 5%−75% EtOAc in hexane, then
NH-silica gel, eluted with 5%−100% EtOAc in hexane) to give 9e (189 mg, 0.372 mmol, 39.0 %) as
1
colorless amorphous solid. MS (ESI) m/z: 507.1 (M+H)+. H NMR (300 MHz, DMSO-d6): δ 1.05
(3H, d, J = 6.0 Hz), 1.30 (3H, t, J = 7.0 Hz), 2.24–2.48 (3H, m), 2.56 (2H, d, J = 10.6 Hz), 4.41 (2H,
q, J = 6.8 Hz), 7.17 (1H, t, J = 7.4 Hz), 7.33–7.67 (4H, m), 7.89–8.16 (3H, m), 8.30 (1H, s), 8.41
(1H, s), 9.93 (1H, s), 10.72 (1H, s).
5.1.25.
N-(3-Chloro-4-cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3-methylpentanediamide
(9f). 9f was synthesized in a manner similar to 9e. 45% yield, colorless powder. MS (ESI) m/z:
473.1 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.03 (3H, d, J = 6.0 Hz), 1.20–1.40 (4H, m),
2.19–2.48 (3H, m), 2.54 (1H, br s), 4.41 (2H, q, J = 7.1 Hz), 7.07–7.27 (1H, m), 7.36–7.69 (5H, m),
7.87 (1H, d, J = 8.7 Hz), 7.97–8.14 (2H, m), 8.39 (1H, s), 9.91 (1H, s), 10.54 (1H, s).
5.1.26. Chiral resolution of 9f. 500 mg of 9f was separated by chiral resolution using AD column.
Column and conditions: Column: CHIRALPAK AD (Lot No. AF001) 50 mm ID X 500 mm. Mobile
phase: n-Hex/ EtOH = 500/500 (v/v), temperature: 40 °C. Flow rate: 80ml/min. Detection: UV 220
nm. Injection concentration: 1.5 mg/mL (in n-hexane/EtOH = 500/500). Injection volume: 100 mL
load.

(+)-N-(3-Chloro-4-cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3-methylpentanediamide (10a).
39% yield, colorless powder. MS (ESI) m/z: 473.2 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.03
(3H, d, J = 6.0 Hz), 1.30 (3H, t, J = 7.0 Hz), 2.22–2.48 (3H, m), 2.52–2.64 (2H, m), 4.41 (2H, q, J =
6.9 Hz), 7.17 (1H, t, J = 7.2 Hz), 7.36–7.66 (5H, m), 7.87 (1H, d, J = 8.7 Hz), 7.97–8.14 (2H, m),
8.39 (1H, s), 9.92 (1H, s), 10.55 (1H, s). [α]25D +4.3 (c 2.54, DMSO).
(-)-N-(3-Chloro-4-cyanophenyl)-N'-(9-ethyl-9H-carbazol-3-yl)-3-methylpentanediamide (10b).
40% yield, colorless powder. MS (ESI) m/z: 473.1 (M+H)+. 1H NMR (300 MHz, DMSO-d6): δ 1.03
(3H, d, J = 6.0 Hz), 1.30 (3H, t, J = 7.0 Hz), 2.18-2.48 (3H, m), 2.52–2.64 (2H, m), 4.41 (2H, q, J =
7.2 Hz), 7.08–7.24 (1H, m), 7.37–7.69 (5H, m), 7.87 (1H, d, J = 8.7 Hz), 7.96-8.16 (2H, m), 8.39
(1H, s), 9.92 (1H, s), 10.55 (1H, s). [α]25D −4.0 (c 2.50, DMSO).
5.1.27. 2-Ethyl-7-nitro-2H-1,4-benzoxazin-3(4H)-one (12). To a stirred suspension of potassium
fluoride (2.36 g, 40.55 mmol) in anhydrous DMF (15.0 mL) was added methyl 2-bromobutylate
(2.06 mL, 17.84 mmol) and reaction mixture was stirred at rt for 15 min. 2-amino-5-nitrophenol (11,
2.5 g, 16.22 mmol) was added to the mixture and heated at 60 °C for 6 h followed by stirring at rt for
additional 14 h. The reaction mixture was combined with another same quantity batch [potassium
fluoride (2.36 g, 40.55 mmol), DMF (15.0 mL), methyl 2-bromobutylate (2.06 mL, 17.84 mmol) and
2-amino-5-nitrophenol (2.5 g, 16.22 mmol)] and crushed ice (175 g) was added to the combined
mixture. White solid formed from the reaction mixture was filtered, washed with water (25 mL × 3)
followed by hexane (25 mL × 3) and dried under reduced pressure to afford 12 (6.0 g, 83%) as white
1
solid. MS (ESI) m/z: 221.0 (M-H)-. H NMR (400 MHz, DMSO-d6): δ 1.00 (t, 3H, J = 7.40 Hz),
1.75–1.89 (m, 2H), 4.70 (dd, 1H, J = 4.52 Hz, 7.64 Hz), 7.05 (d, 1H, J = 8.72 Hz), 7.79 (d, 1H, J =
2.44 Hz), 7.91 (dd, 1H, J = 2.44 Hz, 8.76 Hz ), 11.28 (br s, 1H).
5.1.28. 7-Amino-2,4-diethyl-2H-1,4-benzoxazin-3(4H)-one (13a).
Step A: 2,4-diethyl-7-nitro-2H-1,4-benzoxazin-3(4H)-one. To a stirred solution of 12 (3.0 g,
13.62 mmol) in anhydrous DMF (45 mL) was added K2CO3 (4.14 g, 29.97 mmol) at 0 °C under
nitrogen atmosphere. After stirring at rt for 15 minutes, ethyl iodide (1.63 mL, 20.44 mmol) was
added. The resulting reaction mixture was stirred at rt for 14 h. Reaction mixture was quenched with
water (30 mL) and extracted with ethyl acetate (50 mL × 3). Combined organic layers was washed
with water (30 mL) and brine (30 mL), dried over anhydrous Na2SO4 and concentrated under
reduced pressure to afford 2,4-diethyl-7-nitro-2H-1,4-benzoxazin-3(4H)-one (3.0 g, 88%) as white

1
solid. MS (ESI) m/z: 251.2 (M+H)+. H NMR (400 MHz, DMSO-d6): δ 0.99 (t, 3H, J = 7.36 Hz),
1.15 (t, 3H, J = 7.04 Hz), 1.73–1.88 (m, 2H), 3.99 (q, 2H, J = 7.08 Hz), 4.74 (dd, 1H, J = 4.64 Hz,
7.92 Hz), 7.43 (d, 1H, J = 9.00 Hz), 7.82 (d, 1H, J = 2.60 Hz), 7.96 (dd, 1H, J = 2.52 Hz, 8.92 Hz).
Step B: 7-Amino-2,4-diethyl-2H-1,4-benzoxazin-3(4H)-one. To a stirred deoxygenated solution
of 2,4-diethyl-7-nitro-2H-1,4-benzoxazin-3(4H)-one (1.50 g, 6.00 mmol) in dry methanol (30 mL)
was added 10% Pd-C (0.225 g) under nitrogen atmosphere and stirred at rt for 14 h under hydrogen
atmosphere (balloon pressure). The reaction mixture was filtered through celite bed and washed with
methanol (10 mL × 2). Combined organic layers was collected and concentrated under reduced
pressure. The residue was triturated with pentane to afford 13a (1.0 g, 76%) as brown solid. MS
(ESI) m/z: 221.0 (M+H)+. 1H NMR (400 MHz, DMSO-d6): δ 0.95 (t, 3H, J = 7.36 Hz), 1.07–1.10 (m,
3H), 1.62–1.76 (m, 2H), 3.81 (q, 2H, J = 7.04 Hz), 4.35 (dd, 1H, J = 4.64 Hz, 8.20 Hz), 4.98 (s, 2H),
6.22–6.26 (m, 2H), 6.83 (d, 1H, J = 8.40 Hz).
5.1.29. 6-Nitroquinazoline-2,4(1H,3H)-dione (15a). A mixture of 14 (10.0 g, 54.91 mmol) and
urea (98.93 g, 1647.2 mmol) was heated to 160 °C for 48 h. After complete consumption of starting
material, the reaction mixture was cooled to 100 °C and water (10 mL) was added to the reaction
mixture. The reaction mixture was filtered and washed with water; solid residue was dissolved in
0.5N NaOH solution. Then the reaction mixture was heated at 100 °C for 40 minutes. The reaction
mixture was cooled to 0 °C and 1N HCl (aq.) was added to adjust pH = 5. Light yellow precipitate
was filtered, washed with water and dried under vacuum to give compound 15a (6.0 g, 53%) as a
light yellow solid. MS (ESI) m/z: 206.1 (M-H)-. 1H NMR (400 MHz, DMSO-d6): δ 7.31 (d, 1H, J =
9.04 Hz), 8.44 (dd, 1H, J = 2.60 Hz, 8.96 Hz), 8.57 (d, 1H, J = 2.64 Hz), 11.66 (br s, 2H).
5.1.30. 3-Ethyl-6-nitroquinazoline-2,4(1H,3H)-dione (15b).
Step A: 2-Amino-N-ethyl-5-nitrobenzamide. To a stirred solution of 14 (5.0 g, 27.45 mmol) in
dry DMF (100 mL) at 25 °C were added EtNH2 (2 M sol in THF, 41.18 mL, 82.35 mmol), DIPA
(47.81 mL, 274.51 mmol) and HATU (13.57 g, 35.69 mmol). Reaction mixture was stirred at rt for
24 h. After complete consumption of 14, DMF was distilled off under reduced pressure to get the
crude. Crude mass was quenched with ice-water and organic part was extracted with ethyl acetate (5
× 100 mL), washed with water (3 × 100 mL), dried over anhydrous Na2SO4 and concentrated under
reduced pressure to give crude which was purified by normal silica gel column chromatography
using 10-20% ethyl acetate in hexane as an eluent to afford 2-amino-N-ethyl-5-nitrobenzamide (5.0

g, 93.32%) as yellow solid. MS (ESI) m/z: 208.0 (M-H)-. 1H NMR (400 MHz, DMSO-d6): δ 1.12 (t,
3H, J = 7.20 Hz), 3.22-3.32 (m, 2H), 6.78 (d, 1H, J = 9.24 Hz), 7.77 (br s, 2H), 8.01 (dd, 1H, J =
9.16 Hz, 2.36 Hz), 8.48 (d, 1H, J = 2.36 Hz), 8.69 (br s, 1H).
Step B: 3-Ethyl-6-nitroquinazoline-2,4(1H,3H)-dione. To a solution of 2-amino-N-ethyl-
5-nitrobenzamide (6 g, 28.7 mmol) in THF (60 mL) was added triethyl amine (8.7 mL) at rt.
Phosgene (20% in toluene) (18.6 mL, 37.3 mmol) was added at rt and refluxed for 16 h. Reaction
was monitored by TLC. Starting material was present and again 20% phosgene (7.1 mL, 14.35
mmol) was added. The reaction mixture was heated to reflux for 2 h. Starting material was
consumed. The reaction mixture was concentrated, extracted with ethyl acetate (2 x 100 mL), dried
(Na2SO4) and purified by 15% ethyl acetate in hexane as an eluent to afford 15b (5g, 74.13%) as
yellow solid. MS (ESI) m/z: 234.0 (M-H)-. 1H NMR (400 MHz, DMSO-d6): δ 1.17 (t, 3H, J = 7.08
Hz), 3.90–3.95 (m, 2H), 7.32 (d, 1H, J = 9.00 Hz), 8.45 (dd, 1H, J = 9.04 Hz, 2.68 Hz), 8.62 (d, 1H,
J = 2.60 Hz), 12.03 (br s, 1H).
5.1.31. 6-Amino-1,3-diethylquinazoline-2,4(1H,3H)-dione (13b).
Step A: 1,3-Diethyl-6-nitroquinazoline-2,4(1H,3H)-dione. To a solution of 15a (750 mg, 3.62
mmol) and iodoethane (1.448 mL, 18.10 mmol) in DMF (15 mL) was added K2CO3 (2.502 g, 18.10
mmol), and the mixture was stirred at 70 °C for 16 h. After being cooled to rt, the mixture was
quenched with water at ambient temperature and extracted with EtOAc. The organic layer was
separated, washed with water and brine, dried over MgSO4 and concentrated in vacuo. The residual
powder was triturated with DMF−EtOAc−hexane, and the insoluble materials were filtered to give
1,3-diethyl-6-nitroquinazoline-2,4(1H,3H)-dione (664 mg, 2.52 mmol, 70%) as a pale yellow
1
powder. H NMR (300 MHz, DMSO-d6): δ 1.24 (3H, t, J = 7.2 Hz), 1.18 (3H, t, J = 7.2 Hz), 3.99
(2H, q, J = 6.9 Hz), 4.20 (2H, q, J = 6.9 Hz), 7.72 (1H, d, J = 9.4 Hz), 8.51 (1H, dd, J = 9.3, 2.8 Hz),
8.72 (1H, d, J = 3.0 Hz).
Step B: 6-Amino-1,3-diethylquinazoline-2,4(1H,3H)-dione. A mixture of 1,3-diethyl-
6-nitroquinazoline-2,4(1H,3H)-dione (655 mg, 2.49 mmol) and Pd-C (50 mg, 0.47 mmol) in MeOH
(15 mL) and EtOAc (15 mL) was stirred at rt under H2 (balloon, 3L) for 16 h. Pd catalyst was
removed by celite filtration. The filtrate was concentrated in vacuo to give compound 13b (572 mg,
1
2.452 mmol, 99%) as a gray powder. H NMR (300 MHz, DMSO-d6): δ 1.01–1.32 (6H, m),
3.84–4.16 (4H, m), 5.28 (2H, s), 7.04 (1H, dd, J = 8.7, 2.6 Hz), 7.15–7.32 (2H, m).

5.1.32. 6-Amino-1,3-bis(cyclopropylmethyl)quinazoline-2,4(1H,3H)-dione (13c).
Step A: 1,3-bis(cyclopropylmethyl)-6-nitroquinazoline-2,4(1H,3H)-dione. A mixture of 15a
(333 mg, 1.61 mmol), (bromomethyl)cyclopropane (0.624 mL, 6.43 mmol) and K2CO3 (489 mg,
3.54 mmol) in DMF (10 mL) was stirred at rt for 16 h. The mixture was quenched with brine at rt
and extracted with EtOAc. The organic layer was separated, washed with brine, dried over MgSO4
and concentrated in vacuo. The residue was purified by column chromatography (silica gel, eluted
with
5−30%
EtOAc
in
hexane)
to
give
1,3-bis(cyclopropylmethyl)-6-nitroquinazoline-2,4(1H,3H)-dione (402 mg, 1.28 mmol, 79%) as a
colorless solid. 1H NMR (300 MHz, CDCl3): δ 0.32–0.76 (8H, m), 1.11–1.44 (2H, m), 4.00 (2H, d, J
= 7.2 Hz), 4.12–4.16 (2H, m), 7.45 (1H, d, J = 9.4 Hz), 8.50 (1H, dd, J =9.3, 2.8 Hz), 9.11 (1H, d, J
= 2.6 Hz).
Step B: 6-Amino-1,3-bis(cyclopropylmethyl)quinazoline-2,4(1H,3H)-dione. A mixture of
1,3-bis(cyclopropylmethyl)-6-nitroquinazoline-2,4(1H,3H)-dione (400 mg, 1.27 mmol) and Pd-C
(135 mg, 0.13 mmol) in EtOH (10 mL) was stirred at rt for 2 h under H2 atmosphere. The mixture
was filtered through celite pad and the filtrate was concentrated in vacuo to give compound 13c (280
1
mg, 77%) as a colorless solid. H NMR (300 MHz, CDCl3): δ 0.30–0.68 (8H, m), 0.97–1.44 (2H,
m), 3.40–3.86 (2H, m), 4.01 (4H, dd, J = 14.0, 6.8 Hz), 6.96–7.08 (1H, m), 7.08–7.22 (1H, m), 7.51
(1H, d, J = 3.0 Hz).
5.1.33. 6-Amino-1-(cyclopropylmethyl)-3-ethylquinazoline-2,4(1H,3H)-dione (13d).
Step A: 1-(cyclopropylmethyl)-3-ethyl-6-nitroquinazoline-2,4(1H,3H)-dione. To a stirred
solution of 15b (250 mg, 1.06 mmol) in dry DMF (10 mL) under N2-atmosphere were added K2CO3
(294 mg, 2.13 mmol), and after 5 min, (bromomethyl)cyclopropane (0.43 mL, 3.19 mmol) was
added at rt. Reaction mixture was heated at 80 °C for 8 h. After complete consumption of 15b,
reaction was quenched with ice-water and organic part was extracted with EtOAc (4 × 100 mL),
washed with water (2 × 50 mL), dried over anhydrous Na2SO4 and solvent was distilled off under
reduced pressure. Crude mass was purified by washing with 2% EtOAc in hexane to give
1-(cyclopropylmethyl)-3-ethyl-6-nitroquinazoline-2,4(1H,3H)-dione (150 mg, 49%) as yellow sticky
1
solid. MS (ESI) m/z: 290.2 (M+H)+. H NMR (400 MHz, DMSO-d6): δ 0.45–0.53 (m, 4H),
1.16–1.25 (m, 4H), 3.97–4.02 (m, 2H), 4.11 (d, 2H, J = 6.88 Hz), 7.81 (d, 1H, J = 9.28 Hz), 8.50 (dd,
1H, J = 9.28 Hz, 2.80 Hz), 8.73 (d, 1H, J = 2.72 Hz).

Step B: 6-Amino-1-(cyclopropylmethyl)-3-ethylquinazoline-2,4(1H,3H)-dione. To a stirred
solution of 1-(cyclopropylmethyl)-3-ethyl-6-nitroquinazoline-2,4(1H,3H)-dione (150 mg, 0.52
mmol) in ethyl acetate (15 mL) was added 10% palladium-carbon (80 mg). Reaction mixture was
stirred at rt under hydrogen gas-pressure (30 Psi) for 3 h. After complete consumption of the starting
material, reaction mixture was filtered off through celite pad and washed with ethyl acetate (100 mL).
Filtrate was evaporated under reduced pressure to afford 13d (130 mg, 97%) as a yellow solid. MS
(ESI) m/z: 260.2 (M+H)+. 1H NMR (400 MHz, DMSO-d6): δ 0.42-0.47 (m, 4H), 0.75–0.85 (m, 1H),
1.12 (t, 3H, J = 7.04 Hz), 3.94–3.99 (m, 4H), 5.28 (s, 2H), 7.03 (dd, 1H, J = 8.88 Hz, 2.64 Hz), 7.24
(d, 1H, J = 2.64 Hz), 7.29 (d, 1H, J = 8.96 Hz).
5.1.34. 1-Ethyl-6-nitroquinazoline-2,4(1H,3H)-dione (17).
Step A: Methyl 2-(ethylamino)-5-nitrobenzoate. To a stirred solution of 2-fluoro-5-nitrobenzoic
acid methyl ester (16, 5.0 g, 25.11 mmol) in dry THF (50 mL) at 25 °C in a sealed tube were added
EtNH2 (2 M solution in THF, 2.83 g, 62.77 mmol, 31.5 mL), and K2CO3 (3.47 g, 25.11 mmol). The
reaction mixture was heated at 60 °C for 16 h. After complete consumption of 16, the reaction
mixture was evaporated under reduced pressure to give crude which was purified by washing with
2% EtOAc in hexane to afford methyl 2-(ethylamino)-5-nitrobenzoate (5.0 g, 89%) as a light yellow
solid. 1H NMR (400 MHz, DMSO-d6): δ 1.20 (t, 3H, J = 7.04 Hz), 4.10–4.16 (q, 2H), 7.66 (d, 1H, J =
9.28 Hz), 8.48 (dd, 1H, J = 9.22 Hz, 2.76 Hz), 8.66 (d, 1H, J = 2.64 Hz), 11.00 (br s, 1H).
Step
B:
1-Ethyl-6-nitroquinazoline-2,4(1H,3H)-dione
A
mixture
of
methyl
2-(ethylamino)-5-nitrobenzoate (5.0 g, 22.32 mmol) and urea (40.58 g, 669.64 mmol) was heated at
160 °C for 48 h. After complete consumption of the starting methyl ester, the reaction mixture was
cooled to rt and diluted with water. Precipitate was filtered off and crude was purified by washing
1
with dichloromethane to afford compound 17 (3.0 g, 57%) as a light yellow solid. H NMR (400
MHz, DMSO-d6): δ 1.20 (t, 3H, J = 7.04 Hz), 4.10–4.16 (q, 2H), 7.66 (d, 1H, J = 9.28 Hz), 8.48 (dd,
1H, J = 9.22 Hz, 2.76 Hz), 8.66 (d, 1H, J = 2.64 Hz), 11.00 (br s, 1H).
5.1.35. 6-Amino-3-(cyclopropylmethyl)-1-ethylquinazoline-2,4(1H,3H)-dione (13k).
Step A: 3-(cyclopropylmethyl)-1-ethyl-6-nitroquinazoline-2,4(1H,3H)-dione. To a stirred
solution of 17 (350 mg, 1.49 mmol) in dry DMF (10 mL) at 25 °C under N2-atmosphere were added
NaH (119.15 mg, 2.98 mmol) and after 5 min (bromomethyl)cyclopropane (0.239 mL, 2.98 mmol).
The reaction mixture was stirred at rt for 5 h. After complete consumption of 17, the reaction was