Protein lysine methyltransferase g9a inhibitors: Design, synthesis, and structure activity relationships of 2,4-diamino-7-aminoalkoxy-quinazolines.

Article abstract of DOI:10.1021/jm100478y

Protein lysine methyltransferase G9a, which catalyzes methylation of lysine 9 of histone H3 (H3K9) and lysine 373 (K373) of p53, is overexpressed in human cancers. Genetic knockdown of G9a inhibits cancer cell growth, and the dimethylation of p53 K373 results in the inactivation of p53. Initial SAR exploration of the 2,4-diamino-6,7-dimethoxyquinazoline template represented by 3a (BIX01294), a selective small molecule inhibitor of G9a and GLP, led to the discovery of 10 (UNC0224) as a potent G9a inhibitor with excellent selectivity. A high resolution X-ray crystal structure of the G9a?10 complex, the first cocrystal structure of G9a with a small molecule inhibitor, was obtained. On the basis of the structural insights revealed by this cocrystal structure, optimization of the 7-dimethylaminopropoxy side chain of 10 resulted in the discovery of 29 (UNC0321) (Morrison K_i = 63 pM), which is the first G9a inhibitor with picomolar potency and the most potent G9a inhibitor to date.

Full text of DOI:10.1021/jm100478y

5844 J. Med. Chem. 2010, 53, 5844–5857
DOI: 10.1021/jm100478y
Protein Lysine Methyltransferase G9a Inhibitors: Design, Synthesis, and Structure Activity
Relationships of 2,4-Diamino-7-aminoalkoxy-quinazolines.†
Feng Liu,‡ Xin Chen,‡ Abdellah Allali-Hassani,§ Amy M. Quinn, Tim J. Wigle,‡ Gregory A. Wasney,§ Aiping Dong,§
Guillermo Senisterra,§ Irene Chau,§ Alena Siarheyeva,§ Jacqueline L. Norris,‡ Dmitri B. Kireev,‡ Ajit Jadhav,
J. Martin Herold,‡ William P. Janzen,‡ Cheryl H. Arrowsmith,§ Stephen V. Frye,‡ Peter J. Brown,§ Anton Simeonov,
Masoud Vedadi,§ and Jian Jin*,‡
‡Center for Integrative Chemical Biology and Drug Discovery, Division of Medicinal Chemistry and Natural Products, Eshelman School of
Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, §Structural Genomics Consortium, University of
Toronto, Toronto, Ontario M5G 1L7, Canada, and NIH Chemical Genomics Center, National Human Genome Research Institute,
National Institutes of Health, Bethesda, Maryland 20892
Received April 19, 2010
Protein lysine methyltransferase G9a, which catalyzes methylation of lysine 9 of histone H3 (H3K9) and
lysine 373 (K373) of p53, is overexpressed in human cancers. Genetic knockdown of G9a inhibits cancer
cell growth, and the dimethylation of p53 K373 results in the inactivation of p53. Initial SAR
exploration of the 2,4-diamino-6,7-dimethoxyquinazoline template represented by 3a (BIX01294), a
selective small molecule inhibitor of G9a and GLP, led to the discovery of 10 (UNC0224) as a potent
G9a inhibitor with excellent selectivity. A high resolution X-ray crystal structure of the G9a-10
complex, the first cocrystal structure of G9a with a small molecule inhibitor, was obtained. On the basis
of the structural insights revealed by this cocrystal structure, optimization of the 7-dimethylaminopro-
poxy side chain of 10 resulted in the discovery of 29 (UNC0321) (Morrison Ki=63 pM), which is the first
G9a inhibitor with picomolar potency and the most potent G9a inhibitor to date.
Introduction
lysine acetylation, sumoylation, ADP-ribosylation, ubiquiti-
nation, glycosylation and phosphorylation, and DNA methyl-
ation.2 Given the widespread importance of chromatin reg-
ulation to cell biology, the enzymes that produce these
modifications(the “writers”), the proteins thatrecognizethem
(the “readers”), and the enzymes that remove them (the
“erasers”) are critical targets for manipulation to further
understand the histone code3,4 and its role in human disease.
Indeed, small molecule histone deacetylase inhibitors5 and
DNA methyltransferase inhibitors6 have already proven use-
ful in the treatment of cancer.
Among the “writers” of the histone code, protein lysine
methyltransferases (PKMTs) that catalyze mono-, di-, and/
or trimethylation of lysine residues of various proteins includ-
ing histones have received great attention because of the
essential function of histone lysine methylation in many
biological processes including gene expression and trans-
criptional regulation, heterochromatin formation, and X-
chromosome inactivation.7 Since the first PKMT was characteri-
zed a decade ago,8 more than 50 human PKMTs have been
identified.9,10 Growing evidence suggests that PKMTs play
important roles in the development of various human diseases
including cancer,10-13 inflammation,14 drug addiction,15
mental retardation,16 and HIV-1 latency maintenance.17 In
particular, G9a (also known as EHMT2), which was initially
identified as a H3K9 methyltransferase,18 is overexpressed
in human cancers. It has been shown that knockdown of
G9a inhibits cancer cell growth.19,20 In addition to catalyz-
ing mono- and dimethylation of H3K9, it has been shown
that G9a methylates lysine 373 (K373) of p53, a tumor
suppressor.21 The dimethylation of p53 K373 results in the
Epigenetics is defined as changes in gene function that
are mitotically and/or meiotically heritable and that do not
entail a change in DNA sequence. Over the past decade, the
cellular machinery that creates these heritable changes has
been intensely investigated by the biomedical research com-
munityasthere is noarea of biology, orfor thatmatternoarea
of human health, where epigenetics may not play a funda-
mental role.1 The template upon which the epigenome is
written is chromatin that is primarily made up of DNA
wrapped around octamers of histone proteins. The state of
chromatin, and therefore access to the genetic code, is mainly
regulated by covalent and reversible PTMs to histones and
DNA, and the recognition of these marks by other proteins
and protein complexes. The PTMsa of histones and DNA
include: histone lysine methylation, arginine methylation,
†The coordinates and structure factors of the cocrystal structure of
the G9a-10 complex have been deposited in the Protein Data Bank
(www.pdb.org, PDB code 3K5K).
*To whom correspondence should be addressed. Phone: 919-843-
8459. Fax: 919-843-8465. E-mail: jianjin@unc.edu.
aAbbreviations: PTMs, post-translational modifications; PKMT, pro-
tein lysine methyltransferase; EHMT2, euchromatic histonelysine methyl-
transferase 2; H3K9, histone H3 lysine 9; p53 K373, lysine 373 of p53;
SAM, S-adenosyl-L-methionine; SAR, structure-activity relationships;
iPS cells, induced pluripotent stem cells; GLP, G9a like protein; EHMT1,
euchromatic histone lysine methyltransferase 1; SET, suppressor of var-
iegation 3-9, enhancer of zeste, and trithorax; SAH, S-adenosyl-L-homo-
cysteine; CLOT, chemiluminescence-based oxygen tunneling; ECSD,
enzyme-coupled SAH detection; ITC, isothermal titration calorimetry;
FP, fluorescence polarization; DSF, differential scanning fluorimetry;
MCE, microfluidic capillary electrophoresis; PRMT, protein arginine
methyltransferase.
r
pubs.acs.org/jmc
Published on Web 07/09/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5845
Figure 1. Structure of 3a (BIX01294).22,29
inactivation of p53, which is implicated in over 50% of
cancers.21 These observations suggest that inhibition of G9a
is a potential approach for cancer treatment. Additionally, 3a
(BIX01294), a small molecule G9a inhibitor,22 is efficacious as
a replacement for oct3/4, one of the four original genetic
factors used for reprogramming of mammalian somatic cells
into induced pluripotent stem (iPS) cells.23,24 Thus, small
molecule PKMT inhibitors could play an important role in
stem cell biology and regenerative medicine.
Figure 2. The lipophilic benzyl group (top) is exposed to solvent
and does not makes any interactions with the protein in the X-ray
crystal structure of the GLP-3a complex (PDB code: 3FPD).29
Despite the tremendous progress made in identifying new
PKMTs and elucidating their biological function, generating
small molecule PKMT inhibitors is lagging significantly
behind. In fact, only two selective small molecule PKMT
inhibitors (SU(VAR)3-9 inhibitor Chaetocin and G9a in-
hibitor 3a)22,25-27 have been reported since the first PKMT
was characterized in 2000.8 This gap in creating small mole-
cule PKMT inhibitors presents a major challenge and oppor-
tunity for the biomedical research community. Thecreation of
a “tool-kit” of potent, selective, and well-characterized che-
mical probes28 of PKMTs will permit biological and thera-
peutic hypotheses to be tested with high confidence in cell-
based and animal models of human biology and disease.
To provide potent and selective G9a inhibitors as research
tools and make them available to the biomedical research
community without restrictions on their use, we have explored
the 2,4-diamino-6,7-dimethoxyquinazoline template rep-
resented by 3a (Figure 1). When we began our work, this
compound was the only known selective small molecule
inhibitor of G9a and GLP (also known as EHMT1), which
is another H3K9 and p53 K373 methyltransferase that shares
80% sequence identity with G9a in their respective SET do-
mains.22,29 Although 3a was initially reported to be selective
for G9a (IC50=1.7 μM) over GLP (IC50=38 μM),22 a second
study using linear assay conditions for both enzymes revealed
that 3a inhibited GLP (IC50=0.7 μM) as well as or better than
G9a (IC50=1.9 μM).29 This observed selectivity difference is
likely due to that the original study did not use identical assay
conditions for both enzymes: the GLP assay was conducted
under conditions where the reaction was oversaturated while
the G9a assay was carried out under linear assay conditions.
Morerecently, 3a has been shown tobe about 10-fold selective
for GLP (IC50=27 nM) over G9a (IC50=250 nM) in chemi-
luminescence-based oxygen tunneling (CLOT) assays.30 Be-
cause no SAR was reported for this quinazoline scaffold, we
decided to investigate multiple regions of this template to
elucidate SAR and improve potency and selectivity. In our
previous letter,31 we reported preliminary SAR findings, the
discovery of 10 (UNC0224) as a potent and selective G9a
using the structural insights revealed by the G9a-10 cocrystal
structure, we optimized the 7-dimethylaminopropoxy side
chain, which was not extensively explored in our previous
letter.31 This optimization resulted in the discovery of 29
(UNC0321) (Morrison Ki = 63 pM), the most potent G9a
inhibitor to date.
Results and Discussion
Our initial inhibitor design was based on the structural
insightsrevealedbythe X-raycrystal structureoftheGLP-3a
complex.29 Because the benzyl group of 3a is outside the
peptide binding groove and does not make any interactions
(Figure 2),29 it was proposed that the 1-benzyl piperidin-4-yl-
amino group (marked in red in Figure 1) of 3a could be
replaced with a smaller group such as 1-methyl piperidin-4-yl-
amino and other smaller amino groups as shown in Table 1.
This modification would lower the molecular weight, which
is an important consideration for cellular activity. Addition-
ally, exploration of the 2-amino region (marked in green in
Figure 1) with various cyclic and acyclic amines shown in
Table 1 was carried out.
An efficient two-step synthetic sequence was developed to
prepare the compounds in Table 1 (Scheme 1). Displacing the
4-chloro of commercially available 2,4-dichloro-6,7-dimeth-
oxyquinazoline (1) with the first set of amines at room tem-
peratureaffordedthe intermediates 2. The displacement of the
2-chloro of the intermediates 2 with the second set of amines
under microwave heating conditions yielded the desired 2,4-
diamino-6,7-dimethoxyquinazolines 3. Using this efficient
synthesis, the compounds listed in Table 1 were rapidly pre-
pared in high yields.
Newly synthesized compounds were evaluated in two
orthogonal and complementary G9a biochemical assays:
enzyme-coupled SAH detection (ECSD) and chemilumin-
escence-based oxygen tunneling (CLOT). The ECSD assay
monitors the conversion of cofactor SAM to SAH. This assay
technology was previously employed to study Schizosaccharo-
myces pombe CLR4, a H3K9 specific PKMT.32 The CLOT
assay detects monomethylation of histone H3 peptide 1-11.30
The use of these two orthogonal and complementary bio-
chemical assays in parallel enables us to rule out false positives
and other artifacts of one particular assay format and in-
creases our confidence in SAR findings.
˚
inhibitor, and a 1.7 A X-ray crystal structure of the G9a-10
complex, the first cocrystal structure of G9a with a small
molecule inhibitor. In this article, we describe full accounts
of the design, synthesis, biological evaluation, and SAR of
compounds that explore three regions (i.e., 2-amino, 4-amino,
and 7-methoxy) of this quinazoline scaffold. In particular,

5846 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Liu et al.
Table 1. SAR of 4- and 2-Amino Regions of the Quinazoline Templatea
a IC50 values are the average of at least duplicate assay runs with standard deviation (SD) values that are 3-fold less than the average.
Scheme 1. Synthesis of 2,4-Diamino-6,7-dimethoxy Quinazolines 3a
a (a) R0 amines, DMF, DIEA, RT; (b) R00 amines, i-PrOH, 4 M HCl/dioxane, microwave, 160 °C.
For the 4-amino moiety of the quinazoline scaffold, repla-
cing the N-(1-benzyl piperidin-4-yl)-amino group (3a) with
the N-(1-methyl piperidin-4-yl)-amino group (3b) resulted
in little or no potency loss (Table 1). This result confirmed
our prediction that the benzyl group of 3a could be replaced
by a small alkyl group such as methyl in 3b based on struc-
tural insights revealed by the cocrystal structure of the
GLP-3a complex.29 This SAR finding allowed us to reduce

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5847
the molecular weight while maintaining potency for this
chemical series. On the other hand, replacing the N-(1-methyl
piperidin-4-yl)-amino group (3b) with the N-(piperidin-4-yl)-
amino (3c), N-(tetrahydropyran-4-yl)-amino (3d), or N-cyclo-
hexyl-amino (3e) group led to significant potency loss, indi-
cating that an alkylated nitrogen is important for inhibitory
activity. We next explored the preferred substitution pattern
(N,N-disubstitution versus N-monosubstitution) of the 4-
amino group and whether the nitrogen could be replaced by
an oxygen atom. Compound 3f that contains a N-(1-methyl
piperidin-4-yl),N-methyl-4-amino group was drastically less
potent compared with compound 3b that has a N-monosub-
stituted-4-amino group. Similarly, compound 3g, an oxygen
analogue of compound 3b, had no G9a inhibitory activity in
both assays. These results together indicate that the secondary
nitrogen is important for G9a inhibition, likely via a hydrogen
bond interaction. This finding was subsequently confirmed by
the cocrystal structure of G9a-10 (vide infra). Replacing the
piperidine ring in 3b with a smaller pyrrolidine ring (3h) led to
some potency loss, but compound 3h33 still had appreciable
G9a inhibitory activity. In addition, analogues containing a
smaller amino group such as N-cyclopropyl-4-amino (3i), N-
isopropyl-4-amino (3j), and N-methyl-4-amino (3k) were
significantly less potent compared with compounds 3a and 3b.
Modifications to the 2-amino region of the quinazoline
template werewelltolerated in general(Table 1). For example,
the methyl homopiperazine group (3b) could be replaced with
the methyl piperazine (3l) or piperidine (3m) group without
significant potency loss. Analogues containing a morpho-
line (3n), diethylamine (3o), or dimethylamine (3p) group
had moderate potency. The 2-choloro analogue 2a had poor
potency in both assays. In general, SAR trends from the
ECSD and CLOT assays are in good agreement.
Having established tractable SAR for the 2- and 4-amino
regions, we next explored the 7-methoxy moiety of the quina-
zoline scaffold, again using structure-based design and synth-
esis. The X-ray cocrystal structure of the GLP-3a complex
revealed that3a occupiedthe GLP histonepeptide binding site
but did not interact with the narrow lysine binding channel
(Figure 3).29 To take advantage of this nearby lysine binding
channel, it was hypothesized that adding a 7-aminoalkoxy
side chain to the quinazoline template would make new
interactions with the lysine binding channel while the rest of
molecule maintained interactions with the peptide binding
groove. Thus, compound 10 (structure shown in Scheme 2),
which possesses a 7-dimethylaminopropoxy chain and also
combines the best 2- and 4-amino moieties identified pre-
viously, was designed and modeled. Replacing GLP and 3a in
the X-ray cocrystal structure of the GLP-3a complex29 with
G9a and 10 yielded a model that suggests that 10 is capable of
interacting with the narrow lysine binding channel. This
docking model (not shown here) is virtually identical to the
G9a-10 X-ray cocrystal structure subsequently determined
(vide infra).
To test this hypothesis, compound 10 was synthesized via
the 10-step synthetic sequence outlined in Scheme 2. Benzyl
protection of commercially available 2-methoxy-4-cyanophe-
nol (4), followed by nitration and subsequent reduction of the
nitro group afforded the aniline 5, which was then converted
to the quinazolinedione 6 via formation of methyl carbamate
and subsequent saponification of the cyano group and ring
closure. POCl3 treatment of the quinazolinedione 6 afforded
the 2,4-dichloro quinazoline 7, which underwent two conse-
cutive chloro displacement reactions to yield the 2,4-diamino
quinazoline 8. Debenzylation of the intermediate 8, followed
by a Mitsunobu reaction with3-(dimethylamino)propan-1-ol,
produced the desired compound 10.
Figure 3. Compound 3a (gray, red, and blue) does not occupy the
lysine binding channel in the X-ray cocrystal structure of the
GLP-3a complex (PDB code: 3FPD).29 A fragment of the histone
peptide (magenta) was transposed into this crystal structure to
illustrate the lysine binding channel.
In G9a ECSD and CLOT assays, 10 was about 5-fold more
potent compared to 3a (Table 2). The high potency of 10 was
confirmed in several secondary assays. In ITC experiments
Scheme 2. Synthesis of Compound 10a
a (a) BnBr, K2CO3, DMF, RT; (b) HNO3, Ac2O, 0 °C to RT; (c) Fe dust, NH4Cl, i-PrOH-H2O, reflux, 67% over 3 steps; (d) methyl chloroformate,
DIEA, DMF-DCM, 0 °C to RT; (e) NaOH, H2O2, EtOH, reflux, 70% over 2 steps; (f) N,N-diethylaniline, POCl3, reflux, 59%; (g) 4-amino-1-
methylpiperidine, DIEA, THF, RT; (h) 1-methylhomopiperazine, HCl, i-PrOH, 160 °C, microwave, 82% over 2 steps; (i) Pd/C, 1,4-cyclohexadiene,
EtOH, reflux; (j) 3-(dimethylamino)propan-1-ol, PPh3, DIAD, THF, 0 °C to RT, 46% over 2 steps.

5848 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Liu et al.
that measure the binding affinity of a small molecule to the
G9a protein,34 10 had about 5-fold higher binding affinity
compared to 3a (Table 2). In addition, 10 displaced a fluor-
escein labeled 15-mer H3 peptide (1-15) with higher effi-
ciency than 3a in a G9a fluorescence polarization (FP) assay
and stabilized G9a better compared to 3a in differential
scanning fluorimetry (DSF) experiments.31 These results to-
gether suggest that 10 is a more potent G9a inhibitor com-
pared to 3a.
potency of this series. Compounds 11, 14-16, 19-23, 25-29
were synthesized from the phenol 9 via Mitsunobu reactions
with corresponding alcohols (Scheme 3), similar to the synth-
esis of 10. Compound 24 was made via Mitsunobu reaction of
the phenol 9 with the Boc-protected 5-amino-1-pentanol,
followed by the deprotection of the Boc group. Compounds
12, 13, 17, and 18 were synthesized from the chlorides
30, which were prepared from the phenol 9 via Mitsunobu
reactions with corresponding chloroalkyl alcohols.
These newly synthesized compounds bearing various 7-
alkoxy side chains were evaluated in G9a ECSD and CLOT
assays and SAR results are summarized in Tables 3-5. Com-
pounds 11, 10, 12, and 13 that contain, respectively, a 2-
5-carbon chain, had similar potency while the 6-carbon chain
containing analogue 14 was significantly less potent (Table 3),
indicating the lysine binding channel of G9a can accommodate
a longer aminoalkoxy side chain up to 5 carbons in length.
Compounds possessing various amino capping groups were
evaluated next (Table 4). Compounds 15, 16, and 17 contain-
ing, respectively, N-methylamino, N,N-diethylamino, and N-
methyl,N-propylamino groups, were as potent as 10, whichhas
˚
A high resolution (1.7 A) X-ray cocrystal structure of the
G9a-10 complex was obtained. This cocrystal structure, the
first X-ray crystal structure of G9a in complex with a small
molecule inhibitor, revealed that the 7-dimethylamino pro-
poxy side chain of 10 indeed occupied the lysine binding chan-
nel of G9a and the basic nitrogen interacted with Tyr1154 and
Leu1086 (Figure 4), thus validating our binding hypothesis.
The higher potency of 10 compared to 3a can be explained by
these additional interactions between the 7-dimethylamino
propoxy side chain and the lysine binding channel. These
interactions were absent in the GLP-3a complex as shown
in Figure 3. Other key structural insights revealed by the
G9a-10 cocrystal structure include: (1) the bulk of 10 occu-
pied the histone peptide binding site, similar to the GLP-3a
complex, (2) the hydrogen of the secondary amino group at
the 4-position formed a hydrogen bond with Asp1083, pro-
viding a potential explanation for the drastic potency loss of
compound 3fcompared tocompound 3b(Table1), and (3) the
lysine binding channel was not fully occupied by the 7-di-
methylamino propoxy group and there is space at the end of
the channel to accommodate a longer chain or larger amino
group (Figure 5).
On the basis of this observation, the compounds in
Tables 3-5 were designed and synthesized to extensively
explore the SAR of this 7-aminoalkoxy region and optimize
Table 2. Potency Comparison of 10 with 3aa
G9a
compd
ECSD IC50 (nM)
CLOT IC50 (nM)
ITC Kd (nM)
Figure 5. The 7-dimethylamino propoxy side chain (gray, blue, and
red) of 10 does not fully occupy the lysine binding channel in the
X-ray crystal structure of the G9a-10 complex (PDB code: 3K5K).
A fragment of the H3K9 peptide (magenta) was transposed into this
crystal structure to illustrate the lysine binding channel.
10
3a
43
180
57
250
23
130
a IC50 and Kd values are the average of at least duplicate assay runs
with SD values that are 3-fold less than the average.
Figure 4. X-ray crystal structure of the G9a-10 complex (PDB code: 3K5K). The inhibitor-protein interactions are annotated as follows:
“Es”, electrostatic; “Hb”, hydrogen bonding; “xπ”, cation-π interactions; “(s)”, interaction with a side chain; “(b)”, interaction with the
backbone.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5849
Table 4. SAR of the Capping Groupa
Table 3. SAR of the Chain Lengtha
G9a IC50 (nM)
compd
n
ECSD
CLOT
11
10
12
13
14
2
3
4
5
6
110
43
120
57
140
95
110
49
1500
3200
a IC50 values are the average of at least duplicate assay runs with SD
values that are 3-fold less than the average.
a N,N-dimethylamino group. Similarly, replacing the tertiary
amino group of compound 13 with the primary amino group
(compound 24) did not result in significant potency loss. Cyclic
amino groups were also explored. Although the morpholine 20
was >10-fold less potent, the pyrrolidine 18 and piperidine 19
were more potent compared to their acyclic amine analogues
such as 16. In particular, the pyrrolidine 18 with a respective
IC50 of 8 and 11 nM in ECSD and CLOT assays was about
5-fold more potent than 10. These results together suggest that
modifications to the substituents on the basic nitrogen are well
tolerated and a larger amino capping group can be accommo-
dated. These SAR findings are consistent with the structural
insights revealed by the cocrystal structure of the G9a-10
complex (vide supra).
The necessity for a basic nitrogen in the aminoalkoxy side
chain was investigated next (Table 4). Replacing the nitrogen
in the 7-dimethylamino propoxy group of 10 with a methine
(21) led to about a 100-fold potency loss. Switching the N,N-
dimethylamino group in compound 12 to the methoxy group
in compound 22 or the N-Boc protected amino group in
compound 23 completely abolished G9a inhibitory activity.
Similarly, compound 25 that contains an amide group was
about 100-fold less potent compared to its amine analogue
24. Taken together, these results suggest that the basic nitro-
gen in the aminoalkoxy side chain is critical for retaining high
potency for G9a.
In addition, it was found that replacing the 5-carbon chain
in compound 13 with an ethoxyethyl chain resulted in com-
pound 29 (IC50=6 nM (CLOT) and 9 nM (ECSD)), the most
potent G9a inhibitor to date (Table 5). Increasing the size
of the amino capping group from dimethylamino (29) to
pyrrolidine (26) led to >5-fold potency decrease. Compound
27 that possesses a 2-(N-ethyl,N-methylamino) ethyl chain
was also less potent compared to its carbon analogue 13
and oxygen analogue 29. Additionally, conformationally con-
strained analogues were explored. For example, compound
2833 containing a (piperidin-3-yl)methoxy group was signifi-
cantly less potent compared to its acyclic analogues 15 and 17.
In general, the extensive exploration of the 7-aminoalkoxy
region of this quinazoline scaffold yielded valuable SAR
findings, which are useful for future inhibitor design.
a IC50 values are the average of at least duplicate assay runs with SD
values that are 3-fold less than the average.
studies of G9a inhibitors was utilized to determine the Morri-
son Ki, which is an effective method developed by Morrison
and co-workers that uses a quadratic fit for steep dose res-
ponse data over a narrow compound concentration range for
assessing tight binding inhibitors.36,37 As shown in Figure 6,
29 had a Morrison Ki of 63 pM and was about 40-fold more
potent than 10 (Morrison Ki =2.6 nM) and 250-fold more
potent than 3a (Morrison Ki=16 nM) in this G9a MCE assay.
In addition, the Morrison Ki (2.6 nM) of 10 matched well with
its Ki (1.6 nM),35 which was previously determined using this
G9a MCE assay.
The selectivity of compounds 29, 10, and 3a were evaluated
next. As shown in Table 6, 3a was more selective for GLP over
G9a in ECSD assays. This result is consistent with the
previous finding: 3a was about 10-fold selective for GLP over
G9a in CLOT assays.30 In contrast to 3a, 29 inhibited G9a as
Because compound 29 likely reached the IC50 limits of the
G9a ECSD and CLOT assays, a newly developed endopro-
teinase-coupled microfluidic capillary electrophoresis (MCE)
assay35 that is suitable for mechanism of action and kinetics

5850 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Table 5. Additional SAR of the 7-Aminoalkoxy Side Chaina
Liu et al.
Figure 6. Morrison Kis of compounds 29, 10, and 3a in the G9a
MCE assay.
formaldehyde dehydrogenase-coupled biochemical assay of
JMJD2E,38 a Jumonji protein demethylase. Additionally, 10
was clean (less than 20% inhibition at 1 μM) against a broad
range of G-protein coupled receptors, ion channels, and tran-
sporters in a 30-target selectivity panel (tested by MDS
Pharma Services) with the exception of 82% inhibition of the
muscarinic M2 receptor at 1 μM and 31% inhibition of the
histamine H1 receptor at 1 μM.
Conclusions
Inhibitor design and synthesis based on the GLP-3a X-ray
cocrystal structure in combination with early SAR explora-
tion led to the discovery of 10, a 2,4-diamino-7-aminoalkoxy-
quinazoline, as a potent and selective inhibitor of PKMT
G9a. A high resolution X-ray crystal structure of the G9a-10
complex, the first cocrystal structure of G9a with a small
molecule inhibitor, was obtained. This high resolution cocrys-
tal structure of the G9a-10 complex validated our binding
and inhibitor design hypothesis. Using the structural insights
revealed by this cocrystal structure, the 7-dimethylaminopro-
poxy side chain of 10 was optimized, resulting in the discovery
of 29 (Morrison Ki=63 pM), the most potent G9a inhibitor to
date. Exploration of the 2-amino, 4-amino, and 7-aminoalk-
oxy regions of this quinazoline scaffold yielded valuable SAR
findings, which can be exploited for design of novel inhibitors
in the future. Compounds 29 and 10 are potentially useful
small molecule tools for the biomedical research community
to investigate the biology of G9a and its role in chromatin
remodeling as well as PTMs of other proteins.
a IC50 values are the average of at least duplicate assay runs with SD
values that are 3-fold less than the average.
Scheme 3. Synthesis of Compounds 11-29a
a (a) various alcohols, PPh3, DIAD, THF, 0 °C to RT; (b) TFA,
DCM, RT; (c) various amines, i-PrOH, 160 °C, microwave.
well as or better than GLP while 10 was equipotent versus
G9a and GLP in both ECSD and CLOT assays. This dif-
ferent selectivity profile makes 29 and 10 potentially useful
tool compounds that are complementary to 3a. In addition
to G9a and GLP, compounds 29 and 10 were evaluated in
SET7/9 (a H3K4 PKMT), SET8/PreSET7 (a H4K20 PKMT),
and PRMT3 (a protein arginine methyltransferase (PRMT))
ECSD assays and found to be inactive (IC50>40 μM) versus
these protein lysine and arginine methyltransferases, providing
a >1000-fold selectivity window for G9a over these PKMTs
and PRMT. 29 and 10 were also inactive (IC50>57 μM) in the
Experimental Section
Chemistry General Procedures. HPLC spectra of all com-
pounds except 3k, 10-14, 17-20, 22, 23, 25, and 27-29 were
acquired from an Agilent 6110 series system with UV detector
set to 254 nm. Samples were injected (5 μL) onto an Agilent
Eclipse Plus 4.6 mm ꢀ 50 mm, 1.8 μM, C18 column at room
temperature. A linear gradient from 10% to 100% B (MeOH þ
0.1% acetic acid) in 5.0 min was followed by pumping 100% B
for another 2 min with A being H2O þ 0.1% acetic acid. The
flow rate was 1.0 mL/min. The HPLC spectra of 3k, 10-14,
17-20, 22, 23, 25, and 27-29 were recorded from an Agilent
Table 6. Selectivity of Compounds 29, 10, and 3aa
G9a IC50 (nM)
GLP IC50 (nM)
SET7/9
IC50 (nM)
SET8/PreSET7
IC50 (nM)
PRMT3
IC50 (nM)
JMJD2E
IC50 (nM)
compd
ECSD
9.0
43
180
CLOT
6.0
57
250
ECSD
CLOT
29
10
3a
15
50
34
23
58
27
>40000
>40000
ND
>40000
>40000
ND
>40000
>40000
ND
>57000
>57000
ND
a IC50 values are the average of at least duplicate assay runs with SD values that are 3-fold less than the average. ND: not determined.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5851
6110 series system with UV detector set to 254 nm. Samples were
injected (10 μL) onto a Phenomenex Kinetex 2.6 μM, 100 mm ꢀ
2.1 mm, C18 column at room temperature. Then they were
eluted out with 72% B (MeCN) and A being H2O þ 0.1% formic
acid. The flow rate was 0.3 mL/min. Mass spectra (MS)
data were acquired in positive ion mode using an Agilent 6110
single quadrupole mass spectrometer with an electrospray ioni-
zation (ESI) source. High-resolution (negative ion) mass spec-
tra (HRMS) were acquired using a Shimadzu LCMS-IT-Tof
time-of-flight mass spectrometer. Nuclear magnetic resonance
(NMR) spectra were recorded at Varian Mercury spectrometer
with 400 MHz for proton (1H NMR) and 100 MHz for carbon
(13C NMR) or Varian 300 MHz for proton (1H NMR); chemical
shifts are reported in ppm (δ). Preparative HPLC was perfor-
med on Agilent Prep 1200 series with UV detector set to 220 nm.
Samples were injected onto a Phenomenex Luna 75 mm ꢀ
30 mm, 5 μΜ, C18 column at room temperature. The flow rate
was 30 mL/min. Various linear gradients were used with A being
H2O þ 0.5% TFA and B being MeOH. HPLC was used to
establish the purity of targeted compounds. All compounds that
were evaluated in biological assays had >95% purity using the
HPLC methods described above.
(m, 1H), 4.00 (s, 3H), 4.02 (s, 3H), 2.96 (d, J=11.5 Hz, 2H), 2.32
(s, 3H), 2.19-1.95 (m, 4H), 1.84-1.71 (m, 2H). HPLC: 100%,
RT 4.148 min. MS (ESI) 333 [M þ H]þ. 3b (239.5 mg, yield 78%)
was prepared as a yellow solid from 2-chloro-6,7-dimethoxy-N-
(1-methylpiperidin-4-yl)quinazolin-4-amine (2a, 247.6 mg, 0.74
mmol), 1-methylhomopiperazine (0.18 mL, 1.47 mmol), HCl
in dioxane (4.0 M, 0.37 mL, 1.47 mmol), and isopropyl alcohol
1
(3 mL) using the same procedure as 3a. H NMR (400 MHz,
MeOH-d4) δ ppm 7.38 (s, 1H), 6.89 (s, 1H), 4.15-4.07 (m, 1H),
3.93-3.89 (m, 2H), 3.88 (s, 3H), 3.87 (s, 3H), 3.82 (t, J=6.4 Hz,
2H), 2.94 (d, J=12.1 Hz, 2H), 2.76-2.70 (m, 2H), 2.60-2.57
(m, 2H), 2.34 (s, 3H), 2.30 (s, 3H), 2.17 (td, J=12.1, 2.2 Hz, 2H),
2.12-2.04 (m, 2H), 2.04-1.96 (m, 2H), 1.76-1.66 (m, 2H). 13C
NMR (100 MHz, MeOH-d4) δ ppm 160.24, 159.36, 156.07,
147.11, 105.26, 104.84, 104.28, 59.62, 58.19, 56.98, 56.39, 56.12,
46.97, 46.71, 46.63, 46.31, 32.38, 28.43. HPLC: 100%, RT 0.571
min. MS (ESI) 415 [M þ H]þ. HRMS calcd for C22H34N6O2:
414.2743. Found: 413.2685 [M - H].
6,7-Dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(piperidin-4-
yl)quinazolin-4-amine (3c). 3c was prepared as a yellow solid by
the same procedure as3b, yield 81%. 1H NMR (400 MHz, MeOH-
d4) δ ppm 7.05 (s, 1H), 6.97 (s, 1H), 4.15 (d, J=13.3 Hz, 2H),
4.00-3.95 (m, 2H), 3.94 (s, 3H), 3.91-3.85 (m, 5H), 3.12-3.06 (m,
2H), 3.01-2.92 (m, 1H), 2.80-2.73 (m, 2H), 2.64-2.57 (m, 2H),
2.38 (s, 3H), 2.10-1.94 (m, 4H), 1.67-1.57 (m, 2H). 13C NMR
(100 MHz, MeOH-d4) δ ppm 166.40, 159.50, 156.35, 152.68,
146.43, 106.55, 106.24, 106.22, 59.70, 58.28, 56.71, 56.41, 50.25,
50.03, 47.02, 46.81, 46.68, 35.56, 28.49. HPLC: 100%, RT 0.670
min. MS (ESI) 401 [M þ H]þ.
N-(1-Benzylpiperidin-4-yl)-6,7-dimethoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)quinazolin-4-amine (3a). 4-Amino-1-benzylpiperidine
(6.8 mL, 36.02 mmol) was added to a solutionof 2,4-dichloro-6,7-
dimethoxyquinazoline (1, 4.22 g, 16.28 mmol) in DMF (40 mL),
followed by the addition of N,N-diisopropylethylamine (3 mL,
17.24 mmol), and the resulting mixture was stirred at room
temperature for 2 h until TLC showed that the starting material
had disappeared. Water was added to the reaction mixture, and
the resulting solution was extracted with ethyl acetate. The orga-
nic layer was washed with 0.5% acetic acid aqueous solution and
brine, dried, and concentrated to give the crude product, which
was purified on an ISCO chromatography system using 40 g silica
gel column eluting with hexane-ethyl acetate to give 6.2 g of
6,7-Dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(tetrahydro-
2H-pyran-4-yl)quinazolin-4-amine (3d). 3d was prepared as an
off-white solid by the same procedure as 3b, yield 84%. 1H NMR
(400 MHz, MeOH-d4) δ ppm 7.44 (s, 1H), 6.95 (s, 1H), 4.44-4.36
(m, 1H), 4.08-4.05 (m, 2H), 4.00-3.95 (m, 2H), 3.94 (s, 3H), 3.93
(s, 3H), 3.89 (t, J=6.3 Hz, 2H), 3.60 (t, J=11.9 Hz, 2H), 2.83-
2.77 (m, 2H), 2.69-2.63 (m, 2H), 2.41 (s, 3H), 2.10-2.04 (m, 4H),
1.83-1.72 (m, 2H). 13C NMR (100 MHz, MeOH-d4) δ ppm
160.15, 160.12, 155.92, 150.15, 146.84, 105.89, 104.89, 104.17,
68.53, 59.83, 58.27, 56.96, 56.32, 46.91, 46.78, 46.70, 34.04, 28.60.
HPLC: 100%, RT 1.316 min. MS (ESI) 402 [M þ H]þ.
1
the desired compound as a yellow solid, yield 92%. H NMR
(300 MHz, MeOH-d4) δ ppm 7.70 (s, 1H), 7.59-7.42 (m, 5H),
7.11 (s, 1H), 4.47-4.36 (m, 1H), 4.09 (s, 3H), 4.08 (s, 3H), 3.87
(s, 2H), 3.26 (d, J=12.0 Hz, 2H), 2.61-2.52 (m, 2H), 2.24 (d, J=
12.0 Hz, 2H), 2.02-1.82 (m, 2H). HPLC: 99%, RT 4.112 min.
MS (ESI) 413 [M þ H]þ. N-(1-Benzylpiperidin-4-yl)-2-chloro-
6,7-dimethoxyquinazolin-4-amine (232 mg, 0.56 mmol) was dis-
solved in 3 mL of isopropyl alcohol. To this solution was added
1-methylhomopiperazine (0.14 mL, 1.12 mmol) and HCl in dio-
xane (4.0 M, 0.28 mL, 1.12 mmol). The resulting solution was
heated using microwave irradiation to 160 °C for 10 min. After
cooling, TLC indicated the completion of the reaction. After
removal of the solvent by rotary evaporation, the residue was
redissolved in CH2Cl2 and washed with saturated NaHCO3 solu-
tion. The organic layer was dried, concentrated, and purified by
ISCO to give 221.2 mg of the desired compound 3a as a yellow
N-Cyclohexyl-6,7-dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-
quinazolin-4-amine (3e). 3e was prepared as a yellow solid by the
same procedure as 3b, yield 82%. 1H NMR (400 MHz, MeOH-
d4) δ ppm 7.38 (s, 1H), 6.89 (s, 1H), 4.12-4.04(m, 1H), 3.93-3.89
(m, 2H), 3.88 (s, 3H), 3.87 (s, 3H), 3.82 (t, J=6.4 Hz, 2H), 2.76-
2.70 (m, 2H), 2.60-2.54 (m, 2H), 2.33 (s, 3H), 2.10-2.04 (m, 2H),
2.03-1.95 (m, 2H), 1.85-1.81 (m, 2H), 1.75-1.66 (m, 1H), 1.44-
1.34 (m, 4H), 1.28-1.21 (m, 1H). 13C NMR (100 MHz, MeOH-
d4) δ ppm 159.97, 159.92, 155.76, 149.54, 146.76, 105.64, 104.94,
104.27, 59.75, 58.16, 56.96, 56.29, 51.77, 46.90, 46.74, 46.71,
33.93, 28.53, 27.17, 26.92. HPLC: 100%, RT 2.841 min. MS
(ESI) 400 [M þ H]þ.
1
solid, yield 80%. H NMR (400 MHz, MeOH-d4) δ ppm 7.36
(s, 1H), 7.32-7.24 (m, 5H), 6.87 (s, 1H), 4.09 (m, 1H), 3.89 (m,
2H), 3.86 (s, 3H), 3.85 (s, 3H), 3.81 (t, J=4 Hz, 2H), 3.53 (s, 2H),
2.96 (d, J=12 Hz, 2H), 2.71 (m, 2H), 2.56 (m, 2H), 2.32 (s, 3H),
2.14 (t, J=12 Hz, 2H), 2.06 (d, J=12 Hz, 2H), 1.98 (m, 2H), 1.69
(m, 2H). HPLC: 100%, RT 2.044 min. MS (ESI) 491 [M þ H]þ.
6,7-Dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methyl-
piperidin-4-yl)quinazolin-4-amine (3b). 4-Amino-1-methylpiper-
idine (1.12 mL, 8.92 mmol) was added to a solution of 2,4-
dichloro-6,7-dimethoxyquinazoline (1, 1.05 g, 4.05 mmol) in
THF (10 mL), followed by the addition of N,N-diisopropylethyl-
amine (1.41 mL, 8.11 mmol), andthe resulting mixturewas stirred
at room temperature for 2 h until the TLC indicated that the
reaction completed. The solvent was removed on the rotary
evaporation, and the residue was purified by an ISCO chromato-
graphy system using 40 g silica gel column eluting with hexane-
ethyl acetate to give 1.26 g of the targeted compound 2a as a
yellow solid, yield 92%. 1H NMR (300 MHz, DMSO-d6) δ ppm
8.13 (d, J=7.4 Hz, 1H), 7.78 (s, 1H), 7.18 (s, 1H), 4.28-4.10
6,7-Dimethoxy-N-methyl-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-
methylpiperidin-4-yl)quinazolin-4-amine (3f). 3f was prepared as
an off-white solid by the same procedure as 3b, yield 80%. H
1
NMR (400 MHz, MeOH-d4) δ 7.14 (s, 1H), 6.96 (s, 1H), 4.29-
4.19 (m, 1H), 3.94 (m, 5H), 3.90-3.83 (m, 5H), 3.15 (s, 3H), 3.02
(d, J=11.5 Hz, 2H), 2.79-2.71 (m, 2H), 2.64-2.56 (m, 2H), 2.37
(s, 3H), 2.32 (s, 3H), 2.15 (t, J=11.5 Hz, 2H), 2.10-1.97 (m, 4H),
1.92 (d, J=10.5 Hz, 2H). 13C NMR (100 MHz, MeOH-d4) δ
165.57, 159.30, 155.99, 153.00, 145.83, 107.08, 106.21, 106.19,
59.79, 58.70, 58.26, 56.81, 56.55, 56.35, 46.94, 46.80, 46.74, 46.31,
34.55, 29.61, 28.59. HPLC: 100%, RT 0.868 min. MS (ESI)
429 [M þ H]þ.
6,7-Dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-4-(1-methyl-
piperidin-4-yloxy)quinazoline (3g). 3g was prepared as a yellow
solid by the same procedure as 3b, yield 77%. H NMR (400
1
MHz, MeOH-d4) δ 7.25 (s, 1H), 6.99 (s, 1H), 5.38-5.28 (m, 1H),
4.00-3.93 (m, 5H), 3.93-3.86 (m, 5H), 2.88-2.75 (m, 4H),
2.67-2.60 (m, 2H), 2.52-2.42 (m, 2H), 2.40 (s, 3H), 2.37 (s, 3H),

5852 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Liu et al.
2.23-2.13 (m, 2H), 2.10-1.93 (m, 4H). HPLC: 99%, RT 1.724
min. MS (ESI) 416 [M þ H]þ.
6,7-Dimethoxy-N2,N2-dimethyl-N4-(1-methylpiperidin-4-yl)-
quinazoline-2,4-diamine (3p). 3p was prepared as a white solid by
the same procedure as3b, yield 76%. 1H NMR (400 MHz, MeOH-
d4) δ ppm 7.44 (s, 1H), 6.94 (s, 1H), 4.25-4.17 (m, 1H), 3.94
(s, 3H), 3.92 (s, 3H), 3.21 (s, 6H), 3.00 (d, J=12.5 Hz, 2H), 2.35 (s,
3H), 2.27-2.20 (m, 2H), 2.19-2.11 (m, 2H), 1.81-1.71 (m, 2H).
13C NMR (100 MHz, MeOH-d4) δ ppm 161.04, 160.21, 155.91,
149.84, 146.82, 105.65, 104.67, 104.28, 56.99, 56.32, 56.24, 46.41,
37.73, 32.46. HPLC: 99%, RT 2.623 min. MS (ESI) 346 [M þ H]þ.
2-Amino-4-benzyloxy-5-methoxybenzonitrile (5). To an ice-
cooled solution of 4-hydroxy-3-methoxybenzonitrile (3.40 g,
22.8 mmol) in DMF (45 mL) was slowly added potassium
carbonate (4.73 g, 34.2 mmol) and then benzyl bromide (3.0
mL, 25.1 mmol). The reaction mixture was stirred overnight at
room temperature, and saturated sodium chloride aqueous solu-
tion (80 mL) was added. The resulting precipitate was collected,
washed with water and dried to provide 4-benzyloxy-3-methoxy-
benzonitrile as a yellow solid (5.40 g, 93% yield). 1H NMR (400
MHz, CDCl3) δ ppm 7.36-7.27 (m, 5H), 7.14 (dd, J=8.4, 2.0 Hz,
1H), 7.02 (d, J=2.0 Hz, 1H), 6.83 (d, J=8.4 Hz, 1H), 5.13 (s, 2H),
3.83 (s, 3H). A solution of 4-benzyloxy-3-methoxybenzonitrile
(3.50 g, 14.62 mmol) in acetic anhydride (40 mL) was cooled to
0 °C, and69wt%nitricacid(3.6 mL, 55mmol)wasaddedslowly.
The reaction mixture was stirred overnight at room temperature
and then poured into ice-water. The resulting precipitate was
collected, washedwithwater, anddried toprovide 4-benzyloxy-5-
methoxy-2-nitrobenzonitrile. A mixture of crude 4-benzyloxy-5-
methoxy-2-nitrobenzonitrile, iron dust (3.0 g, 54.8 mmol), and
ammonium chloride (4.4 g, 82.2 mmol) in isopropyl alcohol-
water (50:30 mL) was heated to reflux for 4 h. Then, the resulting
mixture was filtered and washed with 5% methanol in dichloro-
methane (150 mL). The organic phase of the filtrate was dried,
concentrated, and purified by silica gel chromatography (DCM
to 10% MeOH in DCM) to afford 2-amino-4-benzyloxy-5-
methoxybenzonitrile as a yellow solid (5, 2.68 g, 72% yield over
2 steps). 1H NMR (400 MHz, MeOH-d4) δ ppm 7.43-7.28 (m,
5H), 7.12 (s, 1H), 6.41 (s, 1H), 5.07 (s, 2H), 3.76 (s, 3H).
7-(Benzyloxy)-6-methoxyquinazoline-2,4(1H,3H)-dione (6).
Methyl chloroformate (0.94 mL, 12.1 mmol) was added into the
solution of 2-amino-4-benzyloxy-5-methoxybenzonitrile (2.05 g,
8.06 mmol) and DIEA (2.6 mL, 16.1 mmol) in cosolvent DMF:
DCM (20:10 mL) at 0 °C. The reaction solution was stirred over-
night at room temperature, and then 60 mL of water was added.
The resulting precipitate was collected, washed with water, dried,
and used in the next step without further purification. A mixture
of the product, 30 wt % H2O2 (20 mL), and NaOH (322 mg, 8.06
mmol) in ethanol (100 mL) was stirred for 2 h under reflux. After
adding 100 mL of water, the reaction mixture was cooled to RT,
and the resulting precipitate was collected, washed with water,
and dried toafford compound6 as a white solid (1.68 g, 70% yield
over 2 steps). 1H NMR (400 MHz, DMSO-d6) δ ppm 11.03 (br, s,
2H), 7.48-7.36 (m, 5H), 7.28 (s, 1H), 6.78 (s, 1H), 5.14 (s, 2H),
3.79 (s, 3H). 13C NMR(100 MHz, DMSO-d6) δppm 162.8, 154.2,
150.8, 145.6, 136.8, 136.3, 128.9, 128.6, 128.5, 107.8, 106.8, 99.5,
70.5, 56.2. MS (ESI): 299 [M þ H]þ.
6,7-Dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpyr-
rolidin-3-yl)quinazolin-4-amine (3h). 3h was prepared as a yellow
solid by the same procedure as 3b, yield 79%. H NMR (400
1
MHz, MeOH-d4) δ 7.39 (s, 1H), 6.94 (s, 1H), 4.84-4.74 (m, 1H),
4.00-3.95 (m, 2H), 3.93 (s, 3H), 3.92 (s, 3H), 3.88 (t, J=6.3 Hz,
2H), 3.00 (dd, J=9.8, 7.1 Hz, 1H), 2.87-2.75 (m, 3H), 2.72-2.60
(m, 4H), 2.50-2.41 (m, 4H), 2.39 (s, 3H), 2.09-2.01 (m, 2H),
2.01-1.92 (m, 1H). 13C NMR (100 MHz, MeOH-d4) δ 160.43,
159.98, 156.01, 149.97, 146.91, 105.84, 104.83, 104.08, 63.66,
59.83, 58.26, 56.95, 56.34, 56.24, 52.25, 46.93, 46.78, 46.65,
42.48, 33.18, 28.57. HPLC: 100%, RT 0.528 min. MS (ESI) 401
[M þ H]þ.
N-Cyclopropyl-6,7-dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-
quinazolin-4-amine (3i). 3i was prepared as a yellow solid by the
same procedure as 3b, yield 79%. 1H NMR (400 MHz, MeOH-
d4) δ ppm 7.17 (s, 1H), 6.80 (s, 1H), 5.38 (s, 0.34 H, NH), 3.91-
3.84 (m, 2H), 3.83-3.71 (m, 8H), 2.85-2.79 (m, 1H), 2.68-2.64
(m, 2H), 2.52-2.48 (m, 2H), 2.25 (s, 3H), 1.97-1.89 (m, 2H),
0.73-0.67 (m, 2H), 0.55-0.50 (m, 2H). HPLC: 100%, RT 2.631
min. MS (ESI) 358 [M þ H]þ.
N-Isopropyl-6,7-dimethoxy-2-(4-methyl-1,4-diazepan-1-yl)-
quinazolin-4-amine (3j). 3j was prepared as a yellow solid by the
same procedure as 3b, yield 76%. 1H NMR (400 MHz, MeOH-
d4) δ 7.38 (s, 1H), 6.90 (s, 1H), 5.48 (s, 1H, NH), 4.47 (dt, J=13.1,
6.6 Hz, 1H), 3.97-3.91 (m, 2H), 3.89 (s, 3H), 3.88 (s, 3H), 3.84
(t, J=6.4 Hz, 2H), 2.76-2.71 (m, 2H), 2.61-2.55 (m, 2H), 2.34 (s,
3H), 2.05-1.97 (m, 2H), 1.31 (d, J=6.6 Hz, 6H). HPLC: 99%,
RT 2.532 min. MS (ESI) 360 [M þ H]þ.
6,7-Dimethoxy-N-methyl-2-(4-methyl-1,4-diazepan-1-yl)quina-
zolin-4-amine (3k). 3k was prepared as a yellow solid by the same
procedure as 3b, yield 78%. 1H NMR (400 MHz, CDCl3) δ 6.89
(s, 1H), 6.76 (s, 1H), 5.34 (s, 1H), 4.06-3.97 (m, 2H), 3.95-3.83
(m, 8H), 3.10 (d, J=4.7 Hz, 3H), 2.73-2.67 (m, 2H), 2.60-2.52
(m, 2H), 2.36 (s, 3H), 2.08-1.95 (m, 2H). HPLC: 100%, RT0.673
min. MS (ESI) 332 [M þ H]þ.
6,7-Dimethoxy-2-(4-methylpiperazin-1-yl)-N-(1-methylpiperi-
din-4-yl)quinazolin-4-amine (3l). 3l was prepared as a yellow
solid by the same procedure as 3b, yield 78%. H NMR (400
1
MHz, MeOH-d4) δ ppm 7.39 (s, 1H), 6.85 (s, 1H), 4.16-4.08 (m,
1H), 3.88 (s, 3H), 3.87 (s, 3H), 3.80 (m, 4H), 2.94 (d, J=12.1
Hz, 2H), 2.55-2.45 (m, 4H), 2.31 (s, 3H), 2.30 (s, 3H), 2.22-2.16
(m, 2H), 2.11-2.02 (m, 2H), 1.76-1.66 (m, 2H). HPLC: 100%,
RT 1.068 min. MS (ESI) 401 [M þ H]þ.
6,7-Dimethoxy-N-(1-methylpiperidin-4-yl)-2-(piperidin-1-yl)-
quinazolin-4-amine (3m). 3m was prepared as a yellow solid by
the same procedure as 3b, yield 80%. 1H NMR (400 MHz,
MeOH-d4) δ ppm 7.64 (s, 1H), 7.11 (s, 1H), 4.58-4.46 (m, 1H),
3.94 (s, 3H), 3.90 (s, 3H), 3.87-3.81 (m, 4H), 3.63 (d, J=12.8 Hz,
2H), 3.26-3.19 (m, 1H), 2.89 (s, 3H), 2.34 (d, J=12.9 Hz, 2H),
2.10-1.96 (m, 2H), 1.82-1.66 (m, 7H). HPLC: 100%, RT 3.896
min. MS (ESI) 386 [M þ H]þ.
6,7-Dimethoxy-N-(1-methylpiperidin-4-yl)-2-morpholinoquina-
zolin-4-amine (3n). 3n was prepared as a yellow solid by the same
procedure as 3b, yield 81%. 1H NMR (400 MHz, MeOH-d4) δ
ppm 7.40 (s, 1H), 6.86 (s, 1H), 4.17-4.09 (m, 1H), 3.88 (s, 3H),
3.87 (s, 3H), 3.73 (s, 8H), 2.94 (d, J=11.8 Hz, 2H), 2.31 (s, 3H),
2.19 (t, J=12.2 Hz, 2H), 2.10-2.04 (m, 2H), 1.77-1.67 (m, 2H).
HPLC: 100%, RT 0.554 min. MS (ESI) 388 [M þ H]þ.
7-(Benzyloxy)-2,4-dichloro-6-methoxyquinazoline (7). A mix-
ture of compound 5 (1.95 g, 6.54 mmol) and N,N-diethylaniline
(1.15 mL, 7.19 mmol) in POCl3 (10 mL) was stirred for 7 h under
reflux. The reaction mixture was concentrated in vacuo, and
saturated sodium bicarbonate aqueous solution (20 mL) was
added. The resulting mixture was extracted with chloroform
(30 mL ꢀ 2). The combined organic phases were dried, filtered,
concentrated, and purified by silica gel chromatography (hex-
anes to 20% ethyl acetate in hexanes) to afford compound 7 as a
yellow solid (1.30 g, 59% yield). 1H NMR (400 MHz, CDCl3) δ
ppm 7.39-7.27 (m, 5H), 7.25 (s, 1H), 7.19 (s, 1H), 5.21 (s, 2H),
3.97 (s, 3H).
N2,N2-diethyl-6,7-dimethoxy-N4-(1-methylpiperidin-4-yl)quina-
zoline-2,4-diamine (3o). 3o was prepared as a yellow solid by the
same procedure as 3b, yield77%. 1H NMR (400 MHz, MeOH-d4)
δ ppm 7.39 (s, 1H), 6.92 (s, 1H), 4.20-4.13 (m, 1H), 3.91 (s, 3H),
3.90 (s, 3H), 3.66 (q, J=7.0 Hz, 4H), 2.97 (d, J=12.0 Hz, 2H), 2.33
(s, 3H), 2.20 (dd, J=12.1, 2.3 Hz, 2H), 2.16-2.07 (m, 2H), 1.73
(qd, J=12.3, 3.7 Hz, 2H), 1.22 (t, J=7.0 Hz, 6H). 13C NMR (100
MHz, MeOH-d4) δ ppm 160.28, 159.73, 155.77, 150.28, 146.53,
105.74, 104.73, 104.23, 56.95, 56.27, 46.41, 42.93, 32.63, 14.24.
HPLC: 100%, RT 2.711 min. MS (ESI) 374 [M þ H]þ.
7-(Benzyloxy)-6-methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-
(1-methylpiperidin-4-yl)quinazolin-4-amine (8). A mixture of
compound 7 (1.30 g, 3.88 mmol), 4-amino-1-methylpiperidine

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5853
(1.33 g, 11.64 mmol), and DIEA (0.96 mL, 5.82 mmol) in THF
(10 mL) was stirred overnight at room temperature. After con-
centration in vacuo, the crude product was purified by silica gel
chromatography (DCM to 10% MeOH in DCM) to afford
7-(benzyloxy)-2-chloro-6-methoxy-N-(1-methylpiperidin-4-yl)-
(d, J=11.2 Hz, 2H), 2.72 (t, J=6.0 Hz, 2H), 2.62 (t, J=4.3 Hz,
2H), 2.53-2.45 (m, 2H), 2.29 (s, 3H), 2.24 (s, 9H), 2.08 (t, J=9.3
Hz, 4H), 1.94 (dd, J=10.5, 5.5 Hz, 2H), 1.53 (dd, J=21.4, 10.6
Hz, 2H). 13C NMR (100 MHz, CDCl3) δ 158.58, 157.99, 153.65,
149.54, 145.07, 106.79, 102.82, 101.34, 66.57, 58.94, 57.83, 57.34,
56.38, 54.77, 47.96, 46.71, 46.26, 45.86, 45.84, 32.23, 27.89.
HPLC: 99%, RT: 0.713 min. MS (ESI): 472 [M þ H]þ.
1
quinazolin-4-amine as a yellow solid (1.60 g, 100% yield). H
NMR (400 MHz, CDCl3) δ ppm 7.46-7.26 (m, 5H), 7.13 (s, 1H),
6.86 (s, 1H), 5.51 (d, J=7.9 Hz, 1H), 5.22 (s, 2H), 4.42-4.22 (m,
1H), 3.99 (s, 3H), 2.97 (d, J=11.9 Hz, 2H), 2.38 (s, 3H), 2.32 (td,
J=11.9, 2.2 Hz, 2H), 2.15 (d, J=9.6 Hz, 2H), 1.81-1.71 (m, 2H).
13C NMR (100 MHz, CDCl3) δ 159.16, 156.07, 153.95, 149.30,
147.75, 135.48, 128.65, 128.18, 127.18, 108.66, 106.90, 100.25,
70.75, 56.36, 54.57, 47.98, 46.17, 32.11. MS (ESI): 413 [M þ H]þ.
Then, a mixture of 7-(benzyloxy)-2-chloro-6-methoxy-N-(1-me-
thylpiperidin-4-yl)quinazolin-4-amine (1.31 g, 3.17 mmol), 1-
methylhomopiperazine (0.80 mL, 6.35 mmol), and 4.0 M HCl
in dioxane (1.6 mL, 6.4 mmol) in isopropyl alcohol was heated by
microwave irradiation for 15 min to 160 °C in a sealed tube. The
reaction mixture was diluted with 30 mL of DCM and washed
with saturated sodium bicarbonate aqueous solution (10 mL).
The organic phase was dried, filtered, concentrated, and purified
by silica gel chromatography (DCM to 20% MeOH in DCM) to
afford compound8 as a yellow solid (1.30 g, 83% yield). 1H NMR
(400 MHz, MeOH-d4) δ ppm 7.42 (s, 1H), 7.40-7.20 (m, 5H),
6.96 (s, 1H), 5.08 (s, 2H), 4.11-4.03 (m, 1H), 3.90-3.84 (m, 5H),
3.78 (t, J=6.3 Hz, 2H), 2.93 (d, J=12.1 Hz, 2H), 2.80-2.68
(m, 2H), 2.58 (dd, J=6.4, 4.4 Hz, 2H), 2.34 (s, 3H), 2.29 (s, 3H),
2.17 (dd, J=12.0, 10.1 Hz, 2H), 2.05 (d, J=10.5 Hz, 2H), 2.00-
1.94 (m, 2H), 1.75-1.65 (m, 2H). 13C NMR (100 MHz, MeOH-
d4) δ 160.08, 158.83, 155.09, 147.93, 147.45, 137.98, 129.66,
129.15, 128.68, 106.64, 105.00, 104.60, 71.56, 59.50, 58.15, 57.06,
56.00, 49.47, 46.98, 46.67, 46.53, 46.23, 32.24, 28.30. MS (ESI):
491 [M þ H]þ.
7-(4-(Dimethylamino)butoxy)-6-methoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (12).
Intermediate 30 was prepared from phenol 9 (117 mg, 0.292
mmol) and 4-chlorobutan-1-ol (127 mg, 1.17 mmol) and purified
by basic alumina chromatography (DCM to 5% MeOH in
DCM) to afford the crude product (95 mg). Then a mixture of
30, dimethylamine (0.8 mL) and i-PrOH (2.5 mL) was stirred for
15 min at 160 °C in a microwave reactor. After purification (same
as 10), 17 mg of compound 12 was obtained as a yellow solid
(12% yield over three steps). 1H NMR (400 MHz, CDCl3) δ ppm
6.80 (s, 1H), 6.65 (s, 1H), 4.87 (d, J=6.9 Hz, 1H), 3.97-4.06 (m,
3H), 3.93-3.76 (m, 7H), 2.80 (d, J=11.0 Hz, 2H), 2.68-2.58 (m,
2H), 2.55-2.45 (m, 2H), 2.31 (s, 3H), 2.28-2.22 (m, 5H), 2.16 (s,
6H), 2.10 (t, J=11.1 Hz, 4H), 1.95 (dd, J=14.9, 9.8 Hz, 2H),
1.89-1.78 (m, 2H), 1.65-1.48 (m, 4H). 13C NMR (100 MHz,
CDCl3) δ 158.59, 158.00, 154.01, 149.65, 145.23, 106.86, 102.64,
101.47, 68.47, 59.37, 58.97, 57.35, 56.70, 54.78, 47.88, 46.71,
46.30, 45.87, 45.83, 45.47, 32.32, 27.88, 26.83, 24.23. HPLC:
100%; RT: 0.762 min. MS (ESI): 500 [M þ H]þ.
7-(5-(Dimethylamino)pentyloxy)-6-methoxy-2-(4-methyl-1,4-
diazepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (13).
Compound 13 was prepared as an off-white solid by the proce-
dure as the preparation of compound 12 (28% yield over three
steps). 1H NMR (400 MHz, CDCl3) δ 6.82 (s, 1H), 6.70 (s, 1H),
4.96 (d, J = 6.9 Hz, 1H), 4.04 (m, 3H), 3.97-3.90 (m, 2H),
3.89-3.78 (m, 5H), 2.82 (d, J = 11.3 Hz, 2H), 2.70-2.61 (m,
2H), 2.57-2.48 (m, 2H), 2.33 (s, 3H), 2.28 (s, 3H), 2.25-2.20 (m,
2H), 2.17 (s, 6H), 2.12 (t, J=10.2 Hz, 4H), 1.97 (dd, J=10.9, 5.5
Hz, 2H), 1.91-1.81 (m, 2H), 1.65-1.37 (m, 6H). 13C NMR (100
MHz, CDCl3) δ 158.59, 157.99, 153.95, 149.64, 145.16, 106.79,
102.63, 101.36, 68.60, 59.66, 58.96, 57.35, 56.60, 54.78, 47.90,
46.72, 46.29, 45.87, 45.84, 45.49, 32.29, 28.85, 27.89, 27.44, 23.88.
HPLC: 100%; RT: 0.762 min. MS (ESI): 514 [M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-4-(1-methylpiperi-
din-4-ylamino)quinazolin-7-ol (9). A mixture of compound 8
(1.29 g, 2.63 mmol), 1,4-cyclohexadiene (2.6 mL, 26 mmol),
and 10 wt % Pd/C (400 mg) in ethanol (30 mL) was heated to
reflux 2 h. The reaction mixture was filtered and concentrated to
provide the debenzylated product 9 as a yellow solid, which was
used without purification. MS (ESI): 401 [M þ H]þ.
7-(3-(Dimethylamino)propoxy)-6-methoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (10).
Under nitrogen gas, DIAD (250 μL, 1.25 mmol) was added into
the ice-cooled mixture of compound 9 (100 mg, 0.250 mmol),
3-(dimethylamino)propan-1-ol (120 μL, 1.00 mmol), and PPh3
(362 mg, 1.38 mmol) in THF (3.0 mL). The reaction mixture was
stirred overnight at RT. After concentration in vacuo, the crude
product was purified by preparative HPLC with a gradient from
10% of MeOH (A) in 0.1% TFA in H2O (B) to 100% of MeOH
(A). The resulting product was basified with a saturated solution
of NaHCO3 and extracted with CH2Cl2 to afford 56 mg of the
desired product 10 as a white solid (46% yield over 2 steps). 1H
NMR (400 MHz, CDCl3) δ ppm 6.87 (s, 1H), 6.71 (s, 1H), 4.96
(d, J=7.0 Hz, 1H), 4.11 (t, J=6.8 Hz, 2H), 4.08-3.98 (m, 1H),
3.94 (dd, J=5.6, 4.0 Hz, 2H), 3.87 (s, 3H), 3.84 (t, J=6.4 Hz, 2H),
2.83 (d, J=11.8 Hz, 2H), 2.72-2.63 (m, 2H), 2.59-2.51 (m, 2H),
2.47-2.37 (m, 2H), 2.34 (s, 3H), 2.29 (s, 3H), 2.20 (s, 6H), 2.10-
2.17 (m, 4H), 2.08-1.94 (m, 4H), 1.68-1.49 (m, 2H). 13C NMR
(100 MHz, CDCl3) δ 158.52, 157.99, 153.91, 149.58, 145.15,
106.95, 102.67, 101.44, 67.14, 58.90, 57.30, 56.65, 56.25, 54.75,
47.87, 46.66, 46.25, 45.83, 45.43, 32.25, 27.81, 27.20. MS (ESI):
486 [M þ H]þ. HPLC: 100%; RT: 0.765 min. HRMS calcd for
C26H43N7O2: 485.3478. Found: 484.3410 [M - H].
7-(6-(Dimethylamino)ethoxy)-6-methoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (14).
Compound 14 was prepared as a white solid from phenol 9 and 6-
(dimethylamino)hexan-1-ol via Mitsunobu reaction (39% yield over
2 steps). 1H NMR (400 MHz, CDCl3) δ 6.81 (s, 1H), 6.71 (s, 1H),
4.99(d, J=6.9 Hz, 1H), 4.02 (m, 3H), 3.96-3.88 (m, 2H), 3.90-3.76
(m, 5H), 2.81 (d, J=11.3 Hz, 2H), 2.69-2.60 (m, 2H), 2.57-2.45
(m, 2H), 2.32 (s, 3H), 2.26 (s, 3H), 2.23-2.03 (m, 12H), 1.96 (dd, J=
10.7, 5.5 Hz, 2H), 1.83 (dt, J=14.2, 6.9 Hz, 2H), 1.56 (dd, J=21.3,
10.7 Hz, 2H), 1.46-1.39 (m, 4H), 1.36-1.25 (m, 2H). 13C NMR
(100 MHz, CDCl3) δ 158.57, 157.98, 153.94, 149.63, 145.13, 106.79,
102.61, 101.40, 68.63, 59.75, 58.94, 57.33, 56.58, 54.76, 47.90, 46.70,
46.27, 45.84, 45.47, 32.26, 28.83, 27.87, 27.61, 27.17, 25.91. HPLC:
100%, RT: 0.758 min. MS (ESI): 528 [M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-7-(3-(methylamino)-
propoxy)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (15). Com-
pound 15 was prepared as an off-white solid from phenol 9 and
3-(methylamino)propan-1-ol via Mitsunobu reaction (33% yield
over 2 steps). 1H NMR (400 MHz, CD3OD) δ 7.43 (s, 1H), 6.92 (s,
1H), 4.22-4.10 (m, 3H), 3.98-3.94 (m, 2H), 3.93 (s, 3H), 3.87 (t,
J=6.3 Hz, 2H), 2.99 (d, J=11.8 Hz, 2H), 2.86 (t, J=6.8 Hz, 2H),
2.80-2.74 (m, 2H), 2.65-2.60 (m, 2H), 2.47 (d, J=0.9 Hz, 3H),
2.39 (s, 3H), 2.35 (s, 3H), 2.22 (t, J=11.8 Hz, 2H), 2.12-2.02 (m,
6H), 1.76 (dd, J = 21.1, 12.0 Hz, 2H). 13C NMR (100 MHz,
CD3OD) δ 160.27, 160.16, 155.10, 150.02, 147.01, 106.80, 105.04,
104.28, 68.62, 59.84, 58.25, 57.02, 56.22, 50.20, 46.90, 46.79, 46.72,
46.41, 35.99, 32.52, 29.43, 28.61. HPLC: 99%, RT 0.513 min. MS
(ESI): 472 [M þ H]þ.
7-(2-(Dimethylamino)ethoxy)-6-methoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (11).
Compound 11 was prepared as a white solid from phenol 9 and
2-(dimethylamino)ethanol via Mitsunobu reaction (43% yield
over 2 steps). 1H NMR (400 MHz, CDCl3) δ ppm 6.81 (s, 1H),
6.72 (s, 1H), 5.05 (d, J=6.8 Hz, 1H), 4.09 (t, J=6.0 Hz, 2H),
3.95-4.05 (m, 1H), 3.94-3.86 (m, 2H), 3.82-3.80 (m, 5H), 2.78
7-(3-(Diethylamino)propoxy)-6-methoxy-2-(4-methyl-1,4-dia-
zepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (16).

5854 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Liu et al.
Compound 16 was prepared as a yellow solid from phenol 9 and
3-(diethylamino)propan-1-ol via Mitsunobu reaction (47% yield
over 2 steps). 1H NMR (400 MHz, CD3OD) δ 7.42 (s, 1H), 6.92
(s, 1H), 4.19-4.11 (m, 3H), 3.99-3.93 (m, 2H), 3.91 (s, 3H), 3.87
(t, J=6.3 Hz, 2H), 2.99 (d, J=12.0 Hz, 2H), 2.80-2.70 (m, 4H),
2.67-2.59 (m, 6H), 2.39 (s, 3H), 2.35 (s, 3H), 2.21 (t, J=12.0 Hz,
2H), 2.13 (d, J=11.9 Hz, 2H), 2.08-1.98 (m, 4H), 1.74 (dt, J=
12.6, 8.2 Hz, 2H), 1.11 (t, J=7.0 Hz, 6H). 13C NMR (100 MHz,
CD3OD) δ 160.26, 160.16, 155.30, 150.07, 147.09, 106.84, 104.96,
104.46, 68.11, 59.85, 58.25, 57.06, 56.22, 50.41, 48.02, 46.91,
46.79, 46.72, 46.41, 32.53, 28.61, 26.93, 11.56. HPLC: 99%, RT
0.525 min. MS (ESI): 514 [M þ H]þ.
66.01, 57.97, 56.38, 55.66, 54.35, 53.79, 52.70, 46.92, 45.74, 45.31,
44.89, 31.34, 28.68, 26.91, 25.10. HPLC: 99%; RT: 0.762 min. MS
(ESI): 528 [M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-7-(4-methylpentyl-
oxy)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (21). Com-
pound 21 was prepared as a white solid from phenol 9 and
4-methylpentan-1-ol via Mitsunobu reaction (45% yield over
2 steps). 1H NMR (400 MHz, CD3OD) δ 7.41 (s, 1H), 6.89 (s,
1H), 4.12 (ddd, J=11.4, 9.8, 3.8 Hz, 1H), 4.04 (t, J=6.5 Hz, 2H),
3.96-3.88 (m, 5H), 3.84 (t, J=6.3 Hz, 2H), 2.95 (d, J=11.6 Hz,
2H), 2.78-2.70 (m, 2H), 2.63-2.55 (m, 2H), 2.36 (s, 3H), 2.32 (s,
3H), 2.18 (t, J=11.9 Hz, 2H), 2.10 (d, J=12.0 Hz, 2H), 2.05-
1.97 (m, 2H), 1.88-1.79 (m, 2H), 1.78-1.66 (m, 2H), 1.61 (td,
J=13.2, 6.6 Hz, 1H), 1.41-1.32 (m, 2H), 0.93 (d, J=6.6 Hz, 6H).
13C NMR (100 MHz, CD3OD) δ 160.21, 160.03, 155.44, 149.97,
147.07, 106.63, 104.82, 104.43, 70.14, 59.85, 58.24, 57.05, 56.20,
46.92, 46.78, 46.69, 46.41, 36.46, 32.51, 29.21, 28.59, 28.19,
23.12. HPLC: 99%; RT 2.570 min. MS (ESI): 485 [M þ H]þ.
6-Methoxy-7-(4-methoxybutoxy)-2-(4-methyl-1,4-diazepan-
1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (22). Com-
pound 22 was prepared as an off-white solid from phenol 9 and
4-methoxybutan-1-ol via Mitsunobu reaction (38% yield over
2 steps). 1H NMR (400 MHz, CDCl3) δ 6.84 (s, 1H), 6.71 (s, 1H),
4.96 (d, J=7.0 Hz, 1H), 4.12-3.99 (m, 3H), 3.93 (dd, J=5.7, 4.0
Hz, 2H), 3.86 (s, 3H), 3.84 (t, J=6.4 Hz, 2H), 3.40 (t, J=6.4 Hz,
2H), 3.30 (s, 3H), 2.83 (d, J=11.8 Hz, 2H), 2.69-2.63 (m, 2H),
2.53 (dd, J=6.5, 4.4 Hz, 2H), 2.33 (s, 3H), 2.28 (s, 3H), 2.20-
2.07 (m, 4H), 2.04-1.86 (m, 4H), 1.79-1.67 (m, 2H), 1.66-1.50
(m, 2H). 13C NMR (100 MHz, CDCl3) δ 158.58, 157.99, 153.92,
149.62, 145.18, 106.82, 102.65, 101.46, 72.24, 68.34, 58.94, 58.52,
57.34, 56.63, 54.77, 47.88, 46.70, 46.27, 45.85, 45.82, 32.27, 27.87,
26.15, 25.67. HPLC: 99%; RT 0.700 min. MS (ESI): 487 [M þ H]þ.
tert-Butyl 4-(6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-4-(1-me-
thylpiperidin-4-ylamino)quinazolin-7-yloxy)butylcarbamate (23).
Compound 23 was prepared as a white solid from phenol 9 and
tert-butyl 4-hydroxybutylcarbamate via Mitsunobu reaction
6-Methoxy-7-(3-(methyl(propyl)amino)propoxy)-2-(4-methyl-
1,4-diazepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine
(17). Compound 17 was prepared as a yellow solid from phenol 9
by the same general procedure as described for the preparation
1
of compound 12 (50% yield over three steps). H NMR (400
MHz, CDCl3) δ 6.86 (s, 1H), 6.70 (s, 1H), 4.93 (d, J=6.9 Hz,
1H), 4.11 (t, J=6.7 Hz, 2H), 4.08-3.99 (m, 1H), 3.99-3.90 (m,
2H), 3.90-3.78 (m, 5H), 2.84 (d, J=11.5 Hz, 2H), 2.70-2.63 (m,
2H), 2.58-2.51 (m, 2H), 2.48 (t, J=7.1 Hz, 2H), 2.34 (s, 3H),
2.31-2.23 (m, 5H), 2.19 (s, 3H), 2.13 (t, J = 10.1 Hz, 4H),
2.06-1.93 (m, 4H), 1.59 (dd, J=21.1, 11.0 Hz, 2H), 1.45 (dq,
J=14.8, 7.4 Hz, 2H), 0.85 (t, J=7.4 Hz, 3H). 13C NMR (100
MHz, CDCl3) δ 158.57, 158.00, 154.01, 149.66, 145.19, 106.97,
102.66, 101.47, 67.24, 59.72, 58.96, 57.35, 56.69, 54.78, 54.15,
47.90, 46.71, 46.30, 45.86, 42.22, 32.31, 27.87, 26.84, 20.46,
11.90. HPLC: 100%; RT: 0.759 min. MS (ESI): 514 [M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpiper-
idin-4-yl)-7-(3-(pyrrolidin-1-yl)propoxy)quinazolin-4-amine (18).
Compound 18 was prepared as a white solid from phenol 9 by
the same general procedure as described for the preparation of
compound 12 (48% yield over three steps). 1H NMR (400 MHz,
CDCl3) δ 6.87 (s, 1H), 6.71 (s, 1H), 4.95 (d, J=6.6 Hz, 1H),
4.14 (t, J=6.7 Hz, 2H), 4.10-3.99 (m, 1H), 3.98-3.90 (m, 2H),
3.90-3.79 (m, 5H), 2.84 (d, J=10.5 Hz, 2H), 2.66 (d, J=3.9 Hz,
2H), 2.64-2.42 (m, 8H), 2.34 (s, 3H), 2.29 (s, 3H), 2.21-2.02 (m,
6H), 2.03-1.91 (m, 2H), 1.83-1.69 (m, 4H), 1.59 (dd, J=21.6,
11.1 Hz, 2H). 13C NMR (100 MHz, CDCl3) δ 158.53, 157.99,
153.96, 149.61, 145.15, 106.95, 102.65, 101.47, 67.28, 58.93,
57.32, 56.66, 54.76, 54.14, 52.91, 47.88, 46.69, 46.27, 45.84,
32.28, 28.50, 27.84, 23.44. HPLC: 100%; RT: 0.761 min. MS
(ESI): 512 [M þ H]þ.
1
(45% yield over 2 steps). H NMR (400 MHz, CDCl3) δ 6.82
(s, 1H), 6.69 (s, 1H), 4.93-4.82 (m, 2H), 4.15-4.00 (m, 3H),
3.96-3.93 (m, 2H), 3.91 (s, 3H), 3.85 (t, J=6.1 Hz, 2H), 3.23-
3.10 (m, 2H), 2.85 (d, J=11.4 Hz, 2H), 2.72-2.62 (m, 2H), 2.61-
2.51 (m, 2H), 2.35 (s, 3H), 2.31 (s, 3H), 2.15 (t, J=9.5 Hz, 4H),
2.04-1.84 (m, 4H), 1.72-1.53 (m, 4H), 1.43 (s, 9H). HPLC: 99%,
RT 0.703 min. MS (ESI): 572 [M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpiper-
idin-4-yl)-7-(3-(piperidin-1-yl)propoxy)quinazolin-4-amine (19).
Compound 19 was prepared as a white solid from phenol 9
and 1-piperidinepropanol via Mitsunobu reaction (36% yield
over 2 steps). 1H NMR (400 MHz, CDCl3) δ 6.87 (s, 1H), 6.69
(s, 1H), 4.92 (d, J=6.9 Hz, 1H), 4.12 (t, J=6.6 Hz, 2H), 4.05 (dd,
J=14.2, 8.5 Hz, 1H), 3.99-3.91 (m, 2H), 3.91-3.79 (m, 5H),
2.84 (d, J=11.4 Hz, 2H), 2.72-2.63 (m, 2H), 2.60-2.50 (m, 2H),
2.44 (t, J=7.1 Hz, 2H), 2.35 (m, 7H), 2.30 (s, 3H), 2.14 (t, J=10.0
Hz, 4H), 2.08-1.92 (m, 4H), 1.66-1.50 (m, 6H), 1.47-1.35 (m,
2H). 13C NMR (100 MHz, CDCl3) δ 158.57, 157.98, 154.00,
149.65, 145.18, 106.96, 102.63, 101.46, 67.41, 58.95, 57.34, 56.69,
55.76, 54.76, 54.56, 47.86, 46.70, 46.28, 45.85, 45.83, 32.30,
27.86, 26.48, 25.98, 24.44. HPLC: 99%; RT: 0.713 min. MS
(ESI): 526 [M þ H]þ.
7-(5-Aminopentyloxy)-6-methoxy-2-(4-methyl-1,4-diazepan-1-
yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-amine (24). Compound
24 was prepared as a white solid from phenol 9 and tert-butyl
5-hydroxypentylcarbamate via Mitsunobu reaction, followed by
the deprotection of the Boc group by TFA (34% yield over three
steps). 1H NMR (400 MHz, CDCl3) δ 6.82 (s, 1H), 6.71 (s, 1H),
4.09-3.97 (m, 3H), 3.92 (d, J=4.1 Hz, 2H), 3.89-3.77 (m, 5H),
2.82 (d, J=10.9 Hz, 2H), 2.64 (m, 4H), 2.51 (d, J=5.0 Hz, 2H),
2.32 (s, 3H), 2.27 (s, 3H), 2.16-2.09 (m, 4H), 2.03-1.90 (m, 2H),
1.90-1.79 (m, 2H), 1.57 (dd, J=21.3, 10.3 Hz, 2H), 1.45 (m, 4H).
13C NMR (100 MHz, CDCl3) δ 158.60, 157.94, 153.88, 149.63,
145.12, 106.79, 102.61, 101.39, 68.54, 58.94, 57.35, 56.57, 54.77,
47.79, 46.71, 46.27, 45.85, 41.82, 33.43, 32.23, 28.72, 27.88, 23.28.
HPLC: 99%; RT 0.539 min. MS (ESI): 486 [M þ H]þ.
5-(6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-4-(1-methyl-
piperidin-4-ylamino)quinazolin-7-yloxy)pentanamide (25). Com-
pound 25 was prepared as a yellow solid from phenol 9 and
5-hydroxypentanamide via Mitsunobu reaction (17% yield over
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpiperi-
din-4-yl)-7-(3-morpholinopropoxy)quinazolin-4-amine (20). Com-
pound 20 was prepared as a white solid from phenol 9 and
3-morpholinopropan-1-ol via Mitsunobu reaction (38% yield
over 2 steps). 1H NMR (400 MHz, CDCl3) δ 6.88 (s, 1H), 6.69
(s, 1H), 4.91 (d, J=6.7 Hz, 1H), 4.15 (t, J=6.6 Hz, 2H), 4.09-4.02
(m, 1H), 3.94 (d, J=4.1 Hz, 2H), 3.89 (s, 3H), 3.85 (t, J=6.3 Hz,
2H), 3.75-3.64 (m, 4H), 2.84 (d, J=10.8 Hz, 2H), 2.67 (d, J=3.9
Hz, 2H), 2.58-2.37 (m, 8H), 2.33 (d, J=13.6 Hz, 3H), 2.30 (s,
3H), 2.13 (d, J=10.4 Hz, 4H), 2.03 (td, J=13.9, 6.1 Hz, 4H), 1.59
(dd, J = 21.4, 10.4 Hz, 2H). 13C NMR (100 MHz, CDCl3) δ
157.63, 157.00, 152.91, 148.69, 144.15, 106.00, 101.70, 100.38,
1
2 steps). H NMR (400 MHz, CD3OD) δ 7.40 (s, 1H), 6.89 (s,
1H), 4.13 (m, 3H), 3.95-3.90 (m, 2H), 3.89 (s, 3H), 3.84 (t, J=6.4
Hz, 2H), 2.96 (d, J=12.0 Hz, 2H), 2.79-2.70 (m, 2H), 2.60 (dd,
J=6.5, 4.4 Hz, 2H), 2.36 (s, 3H), 2.34-2.29 (m, 5H), 2.25-2.14
(m, 2H), 2.10 (d, J=12.2 Hz, 2H), 2.06-1.97 (m, 2H), 1.95-1.66
(m, 6H). 13C NMR (100 MHz, CD3OD) δ 177.43, 158.64, 158.50,
153.69, 148.40, 145.39, 105.05, 103.29, 102.75, 68.03, 58.25, 56.65,
55.40, 54.61, 45.32, 45.19, 45.11, 44.81, 34.68, 30.92, 27.91, 27.00,
22.41. HPLC: 99%; RT 0.692 min. MS (ESI): 500 [M þ H]þ.

Article
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpiperi-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5855
product SAH to homocysteine and inosine. The abundance of
homocysteine can be quantified using ThioGlo (Calbiochem),
which reacts with thiols and fluoresces strongly. The substrate
peptide used in this assay was the first 25 residues of histone
3 [H3 (1-25)]. The SAHH clone was provided by Dr. Raymond
Trievel (University of Michigan). For IC50 determination, assay
mixtures were prepared with 5 μM SAHH, about 0.3 U/mL of
adenosine deaminase from Sigma, 16 μM SAM, 25 nM G9a,
and 15 μM Thioglo. The inhibitors were added at concentra-
tions ranging from 6 nM to 25 μM. After 5 min incubation,
reactions were initiated by the addition of 5 μM H3 (1-25)
peptide. The methylation reaction was followed by monitoring
the increase in fluorescence using BioTek Synergy2 plate reader
with 360/40 nm excitation filter and 528/20 nm emission filter
for 20 min in 384 well-plate format. Homocysteine generated in
the assay was quantitated using standard curves. Activity values
were corrected by subtracting background caused by the peptide
and the protein. IC50 values were calculated using four para-
meter logistic equation by Sigmaplot software. Standard devia-
tions were calculated from two independent experiments.
CLOT Assay Experimental Procedures. Substrate peptide
b-H3(1-11) (biotin-ARTKQTARKST) was synthesized, HPLC-
purified, and mass-analyzed by the Tufts Medical School Depart-
ment of Physiology Core Facility (Boston, MA). To each well of a
1536-well white solid-bottom plate (Greiner Bio-One, Monroe,
NC), 3 μL of G9a enzyme (final 7.5 nM) in PBS buffer, pH 7.4,
containing 0.01% Tween-20 were added using a BioRAPTR
(Beckman Coulter, Fullerton, CA) flying reagent dispenser
(FRD). Compounds in DMSO (23 nL) were transferred to the
assay plate with a Kalypsys pintool and assayed in duplicate in a
12-point dilution series, yielding a final concentration range of
32 pM to 57 μM and 0.5% DMSO. Following a 15 min incubation
of compound with enzyme, b-H3(1-11) peptide and SAM sub-
strates (final 500 nM and 20 μM, respectively) were added in a
1 μL FRD dispense. Reactions were incubated for 2 h at room
temperature. To detect methylated b-H3(1-11) peptide, 0.5 μg/
mL rabbitanti-methyllysine histone H3K9 antibody (Abcam Inc.,
Cambridge, MA) was added with 15 μg/mL streptavidin-coated
donor and anti-rabbit IgG acceptor AlphaScreen (PerkinElmer,
Waltham, MA)beadsina 1 μLFRD dispense. Following a20 min
incubation in the dark, plates were read on an EnVision multilabel
plate reader (PerkinElmer) using the 1536 Plate HTS AlphaScreen
aperture (80 ms excitation time, 240 ms measurement time). Data
were normalized to no-enzyme and no-compound controls and
dose-response curves were constructed. IC50 values were deter-
mined with in-house curve fitting software using a four-parameter
Hill equation as previously described (http://www.ncgc.nih.gov/
pub/openhts/curvefit/).
DSF Experimental Procedures. A real-time PCR device
(RTPCR 480 II) from Roche was used to monitor protein
unfolding by monitoring the increase in the fluorescence of the
fluorophor SYPRO Orange (Invitrogen, Carlsbad, CA) as de-
scribed before.39,40 Protein samples at 0.1 mg/mL in 100 mM
Hepes buffer (pH 7.5) containing 150 mM NaCl, and 0, 6.25,
12.5, 25, 50, 100, and 200 μM of compounds were screened in the
presence and absence of 500 μM SAH. Compound dilutions were
made from stock solutions of 100% DMSO. The final concentra-
tion of DMSO was kept at 0.2% throughout the dilutions. All
these solutions contained 5ꢀ Sypro Orange. Twenty μL aliquots
were transferred to a 384-well PCR plate and scanned at a heating
rate of 1 °C/min from 20 to 95 °C. Fluorescence intensities were
plotted as a function of temperature by using an internally
developed software package.40
Peptide Displacement Experimental Procedures. Fluorescence
polarization measurements were performed in 384-well plates,
using Synergy 2 microplate reader from BioTek.41 The H3
(1-15) peptide (ARTKQTARKSTGGKA) was synthesized,
N-terminally labeled with fluorescein [F-H3 (1-15)] and puri-
fied by Tufts University Core Services (Boston, MA). Displace-
ment of the F-H3 (1-15) peptide was carried out using the
din-4-yl)-7-(2-(2-(pyrrolidin-1-yl)ethoxy)ethoxy)quinazolin-4-amine
(26). Compound 26 was prepared as a yellow solid from phenol 9
and 2-(2-(pyrrolidin-1-yl)ethoxy)ethanol via Mitsunobu reaction
1
(49% yield over 2 steps). H NMR (400 MHz, CDCl3) δ 6.83
(s, 1H), 6.73 (s, 1H), 5.06 (d, J=6.9 Hz, 1H), 4.19 (t, J=5.0 Hz,
2H), 4.09-3.96 (m, 1H), 3.95-3.88 (m, 2H), 3.88-3.76 (m, 7H),
3.62 (t, J=6.0 Hz, 2H), 2.81 (d, J=11.5 Hz, 2H), 2.63 (t, J=5.9
Hz, 4H), 2.53-2.47 (m, 6H), 2.31 (s, 3H), 2.26 (s, 3H), 2.10 (t,
J=10.1 Hz, 4H), 2.01-1.92 (m, 2H), 1.78-1.65 (m, 4H), 1.64-
1.50 (m, 2H). 13C NMR (100 MHz, CDCl3) δ 158.51, 158.01,
153.64, 149.40, 145.12, 106.87, 102.93, 101.58, 70.41, 69.00, 67.85,
58.90, 57.31, 56.47, 55.50, 54.77, 54.53, 47.94, 46.67, 46.23, 45.84,
45.78, 32.19, 27.82, 23.38. HPLC: 99%; RT 0.549 min. MS (ESI):
542 [M þ H]þ.
N1-(2-(6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-4-(1-methyl-
piperidin-4-ylamino)quinazolin-7-yloxy)ethyl)-N1,N2,N2-trimethyl-
ethane-1,2-diamine (27). Compound 27 was prepared as a yellow
solid from phenol 9 and 2-((2-(dimethylamino)ethyl)(methyl)-
amino)ethanol via Mitsunobu reaction (29% yield over 2 steps).
1H NMR (400 MHz, CDCl3) δ 6.86 (s, 1H), 6.68 (s, 1H), 4.91 (d,
J=6.7 Hz, 1H), 4.18 (t, J=6.1 Hz, 2H), 4.12-3.99 (m, 1H), 3.95
(d, J=4.3 Hz, 2H), 3.88-3.84 (m, 5H), 2.92 (t, J=6.1 Hz, 2H),
2.85 (d, J=11.1 Hz, 2H), 2.72-2.63 (m, 2H), 2.62-2.53 (m, 4H),
2.42 (t, J=6.9 Hz, 2H), 2.35 (s, 6H), 2.30 (s, 3H), 2.23 (s, 6H), 2.15
(t, J=10.2 Hz, 4H), 2.04-1.90 (m, 2H), 1.60 (dd, J=21.1, 10.6 Hz,
2H). 13C NMR (100 MHz, CDCl3) δ 158.50, 157.97, 153.78,
149.48, 145.20, 106.83, 102.74, 101.23, 66.59, 58.91, 57.35, 57.31,
56.54, 56.20, 55.93, 54.75, 47.88, 46.67, 46.27, 45.82, 45.79, 45.76,
43.09, 32.29, 27.79. HPLC: 99%; RT 0.727 min. MS (ESI): 529
[M þ H]þ.
6-Methoxy-2-(4-methyl-1,4-diazepan-1-yl)-N-(1-methylpiperi-
din-4-yl)-7-(piperidin-3-ylmethoxy)quinazolin-4-amine (28). Com-
pound 28 was prepared as a white solid from phenol 9 and
3-piperidinemethanol via Mitsunobu reaction (40% yield over
2 steps). 1H NMR (400 MHz, CDCl3) δ 6.81 (s, 1H), 6.71 (s, 1H),
4.94 (d, J=7.0 Hz, 1H), 4.08-3.99 (m, 1H), 3.97-3.90 (m, 2H),
3.90-3.78 (m, 7H), 3.23 (dd, J=12.0, 2.7 Hz, 1H), 2.98 (d, J=12.0
Hz, 1H), 2.82 (d, J=11.6 Hz, 2H), 2.72-2.62 (m, 2H), 2.61-2.48
(m, 3H), 2.48-2.36 (m, 1H), 2.33 (s, 3H), 2.28 (s, 3H), 2.21-2.01
(m, 5H), 2.01-1.93 (m, 2H), 1.89 (d, J=9.5 Hz, 1H), 1.81 (br,
1H), 1.70-1.38 (m, 4H), 1.20 (ddd, J=24.4, 12.1, 3.8 Hz, 1H). 13
C
NMR (100 MHz, CDCl3) δ 158.59, 157.99, 154.14, 149.70,
145.24, 106.98, 102.70, 101.96, 71.77, 58.96, 57.35, 56.89, 54.77,
50.12, 47.90, 46.98, 46.73, 46.31, 45.87, 36.77, 32.31, 27.99, 27.90,
26.03. HPLC: 99%; RT 0.763 min. MS (ESI): 498 [M þ H]þ.
7-(2-(2-(Dimethylamino)ethoxy)ethoxy)-6-methoxy-2-(4-me-
thyl-1,4-diazepan-1-yl)-N-(1-methylpiperidin-4-yl)quinazolin-4-
amine (29). Compound 29 was prepared as an off-white solid
from phenol 9 and 2-(2-(dimethylamino)ethoxy)ethanol via Mit-
sunobu reaction (44% yield over 2 steps). 1H NMR (400 MHz,
CDCl3) δ 6.80 (s, 1H), 6.72 (s, 1H), 5.02 (d, J=6.9 Hz, 1H), 4.17
(t, J=5.0 Hz, 2H), 4.01 (qd, J=13.7, 8.3 Hz, 1H), 3.93-3.86 (m,
2H), 3.85-3.74 (m, 7H), 3.56 (t, J=5.8 Hz, 2H), 2.79 (d, J=11.2
Hz, 2H), 2.67-2.57 (m, 2H), 2.53-2.45 (m, 2H), 2.43 (t, J=5.8
Hz, 2H), 2.30 (s, 3H), 2.24 (s, 3H), 2.18 (s, 6H), 2.08 (t, J=9.0 Hz,
4H), 1.96-1.88 (m, 2H), 1.53 (dd, J=21.4, 10.4 Hz, 2H). 13C
NMR (100 MHz, CDCl3) δ 158.57, 157.98, 153.62, 149.54,
145.06, 106.95, 102.92, 101.59, 69.50, 68.99, 67.87, 58.93, 58.74,
57.33, 56.47, 54.77, 47.94, 46.71, 46.27, 45.84, 45.81, 32.23, 27.88.
HPLC: 100%; RT 0.759 min. MS (ESI): 516 [M þ H]þ.
Biochemical Assays General Procedures. The G9a enzyme
used in the following assays was cloned and purified in Struc-
tural Genomics Consortium (SGC) [Mammalian Gene Collec-
tion; MGC AU80-H7 (BC018718)]. The construct amino acid
sequence is from 913 to 1193.
ECSD Assay Experimental Procedures. The methyltransfer-
ase activity of PKMTs were measured using a coupled assay.32
In this assay SAHH (S-adenosylhomocysteine hydrolase) and
adenosine deaminase convert the methyltransferase reaction

5856 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Liu et al.
fluorescence polarization signal obtained upon peptide binding
to G9a protein. Five μM of G9a was incubated with 36 nM
peptide, and different concentrations of 3a, 10, or unlabeled H3
(1-25) from 0.5 to 500 μM were added. Displacement of the
peptide was monitored by following the decrease in FP signal.
Data were plotted and fit to a hyperbolic function using Sigma
Plot software.
Swedish Agency for Innovation Systems, the Swedish Founda-
tion for Strategic Research, and the Wellcome Trust.
Supporting Information Available: 1H and 13C NMR spectra
of compounds 10 and 29. This material is available free of charge
via the Internet at http://pubs.acs.org.
ITC Experimental Procedures. ITC measurements were per-
formed in duplicate at 25 °C, using a VP-ITC microcalorimeter
(MicroCal Inc.). Experiments were performed by injecting 10 μL
of 300 μM solution of the compound 3a or 10 into a sample cell
containing 25 μM G9a that was extensively dialyzed in ITC
buffer (25 mM Tris-HCl, pH 7.5, 200 mM NaCl, and 2 mM
β-mercaptoethanol) and degassed. A total of 25 injections were
performed with a spacing of 180 s and a reference power of
13 μcal/s. The ITC experiments were carried out in the presence
of 500 μM SAH by adding it to both G9a and final solution of
compounds. Compounds were first dissolved in 100% DMSO at
100 mM and were diluted in ITC buffer (final concentration of
0.3%). The concentration of DMSO was adjusted to 0.3% for
the G9a solution. Heat of dilution generated by compounds was
subtracted, and binding isotherms were plotted and analyzed
using Origin Software (MicroCal Inc.). The ITC measurements
were fit to a one-site binding model.
Morrison Ki Determination Using the Endoproteinase-coupled
MCE Assay. The test compounds were titrated in 10% DMSO
in 1X assay buffer (20 mM Tris-HCl pH=8, 25 mM NaCl, 2 mM
DTT and 0.025% Tween 20) using a 1.5-fold dilution scheme
over 16 points. The compounds (5 μL) were then spotted into the
assay wells of a 384-well microplate in triplicate. In the final
assay volume of 50 μL, the compounds spanned a concentration
range of 100 nM to 228 pM, and DMSO was 1%. A cocktail
made in 1X assay buffer containing G9a (12.5 nM) and an H3K9
(5 μM) substrate peptide, ARTKme1QTARKSTGGK-5/6-
FAM, where Kme1 indicates the presence of a monomethyl
lysine and 5/6-FAM is a carboxyfluorescein tracer attached to
the terminal lysine, was added and allowed to preincubate with
the compounds for 10 min at 25 °C. The methyltransferase
reaction was initiated by the addition of 10 μM SAM and
allowed to proceed for 20 min at 25 °C, and then 10 μL of
Endo-LysC (40 pg/μL) was added to proteolyze the remaining
unmethylated peptide. Maximum and minimum signal controls
were performed in the presence of 1% DMSO, and the minimum
signal control reactions were initiated with 1X assay buffer
rather than SAM. After 1 h, the plate was read on a Caliper
Life Sciences EZ Reader II using upstream voltage -500 V,
downstream voltage -1200 V, and pressure of -1.5 psi. The
fractional velocities of the reactions were calculated and fit to a
quadratic equation described by Morrison and co-workers36,37
using a fixed enzyme concentration of 12.5 nM, a fixed substrate
concentration of 5 μM, and a Km of 31 μM for the peptide.
References
(1) Bernstein, B. E.; Meissner, A.; Lander, E. S. The mammalian
epigenome. Cell 2007, 128, 669–681.
(2) Gelato, K. A.; Fischle, W. Role of histone modifications in defining
chromatin structure and function. Biol. Chem. 2008, 389, 353–363.
(3) Strahl, B. D.; Allis, C. D. The language of covalent histone
modifications. Nature 2000, 403, 41–45.
(4) Jenuwein, T.; Allis, C. D. Translating the histone code. Science
2001, 293, 1074–1080.
(5) Esteller, M. Epigenetics in cancer. N. Engl. J. Med. 2008, 358, 1148–
1159.
(6) Lyko, F.; Brown, R. DNA methyltransferase inhibitors and the
development of epigenetic cancer therapies. J. Natl. Cancer Inst.
2005, 97, 1498–1506.
(7) Martin, C.; Zhang, Y. The diverse functions of histone lysine
methylation. Nature Rev. Mol. Cell. Biol. 2005, 6, 838–849.
(8) Rea, S.; Eisenhaber, F.; O’Carroll, D.; Strahl, B. D.; Sun, Z. W.;
Schmid, M.; Opravil, S.; Mechtler, K.; Ponting, C. P.; Allis, C. D.;
Jenuwein, T. Regulation of chromatin structure by site-specific
histone H3 methyltransferases. Nature 2000, 406, 593–599.
(9) Kouzarides, T. Chromatin modifications and their function. Cell
2007, 128, 693–705.
(10) Copeland, R. A.; Solomon, M. E.; Richon, V. M. Protein methyl-
transferases as a target class for drug discovery. Nature Rev. Drug
Discovery 2009, 8, 724–732.
(11) Spannhoff, A.; Sippl, W.; Jung, M. Cancer treatment of the future:
inhibitors of histone methyltransferases. Int. J. Biochem. Cell Biol.
2009, 41, 4–11.
(12) Fog, C. K.;Jensen, K. T.; Lund, A. H. Chromatin-modifyingproteins
in cancer. Acta Pathol., Microbiol. Immunol. 2007, 115, 1060–1089.
(13) Spannhoff, A.; Hauser, A. T.; Heinke, R.; Sippl, W.; Jung, M. The
emerging therapeutic potential of histone methyltransferase and
demethylase inhibitors. ChemMedChem 2009, 4, 1568–1582.
(14) Li, Y.; Reddy, M. A.; Miao, F.; Shanmugam, N.; Yee, J.-K.;
Hawkins, D.; Ren, B.; Natarajan, R. Role of the Histone H3
Lysine 4 Methyltransferase, SET7/9, in the Regulation of NF-
{kappa}B-dependent Inflammatory Genes: Relevance to Diabetes
and Inflammation. J. Biol. Chem. 2008, 283, 26771–26781.
(15) Maze, I.; Covington, H. E., III; Dietz, D. M.; LaPlant, Q.; Renthal,
W.; Russo, S. J.; Mechanic, M.; Mouzon, E.; Neve, R. L.; Hag-
garty, S. J.; Ren, Y.; Sampath, S. C.; Hurd, Y. L.; Greengard, P.;
Tarakhovsky, A.; Schaefer, A.; Nestler, E. J. Essential role of the
histone methyltransferase G9a in cocaine-induced plasticity.
Science 2010, 327, 213–216.
(16) Schaefer, A.; Sampath, S. C.; Intrator, A.; Min, A.; Gertler, T. S.;
Surmeier, D. J.; Tarakhovsky, A.; Greengard, P. Control of
Cognition and Adaptive Behavior by the GLP/G9a Epigenetic
Suppressor Complex. Neuron 2009, 64, 678–691.
(17) Imai, K.; Togami, H.; Okamoto, T. Involvement of histone H3
Lysine 9 (H3K9) methyl transferase G9a in the maintenance of
HIV-1 latency and its reactivation by BIX01294. J. Biol. Chem.
2010, 285, 16538–16545.
(18) Tachibana, M.; Sugimoto, K.; Nozaki, M.; Ueda, J.; Ohta, T.;
Ohki, M.; Fukuda, M.; Takeda, N.; Niida, H.; Kato, H.; Shinkai,
Y. G9a histone methyltransferase plays a dominant role in euchro-
matic histone H3 lysine 9 methylation and is essential for early
embryogenesis. Genes Dev. 2002, 16, 1779–1791.
Acknowledgment. We thank Dr. Yizhou Dong and Profes-
sor K. H. Lee for HRMS and HPLC support and Dr. Stanley
Ng for supplying the JMJD2E enzyme. The research described
here was supported by the grant RC1GM090732 from the
National Institute Of General Medical Sciences of NIH, the
Carolina Partnership, the Molecular Libraries Initiative of
the NIH Roadmap for Medical Research, the Intramural
Research Program of NHGRI of NIH, the Ontario Research
Fund and the Structural Genomics Consortium, a registered
charity (number 1097737) that receives funds from the Cana-
dian Institutes for Health Research, the Canada Foundation
for Innovation, Genome Canada through the Ontario Geno-
mics Institute, GlaxoSmithKline, Karolinska Institutet, the
Knut and Alice Wallenberg Foundation, the Ontario Innova-
tion Trust, the Ontario Ministry for Research and Innovation,
Merck & Co., Inc., the Novartis Research Foundation, the
(19) McGarvey, K. M.; Fahrner, J. A.; Greene, E.; Martens, J.;
Jenuwein, T.; Baylin, S. B. Silenced tumor suppressor genes
reactivated by DNA demethylation do not return to a fully
euchromatic chromatin state. Cancer Res. 2006, 66, 3541–3549.
(20) Kondo, Y.; Shen, L.; Ahmed, S.; Boumber, Y.; Sekido, Y.;
Haddad, B. R.; Issa, J. P. Downregulation of histone H3 lysine 9
methyltransferase G9a induces centrosome disruption and chro-
mosome instability in cancer cells. PLoS ONE 2008, 3, e2037.
(21) Huang, J.; Dorsey, J.; Chuikov, S.; Zhang, X.; Jenuwein, T.;
Reinberg, D.; Berger, S. L. G9a and GLP methylate lysine 373 in
the tumor suppressor p53. J. Biol. Chem. 2010, 285, 9636–9641.
(22) Kubicek, S.; O’Sullivan, R. J.; August, E. M.; Hickey, E. R.;
Zhang, Q.; Teodoro, M. L.; Rea, S.; Mechtler, K.; Kowalski,
J. A.; Homon, C. A.; Kelly, T. A.; Jenuwein, T. Reversal of
H3K9me2 by a small-molecule inhibitor for the G9a histone
methyltransferase. Mol. Cell 2007, 25, 473–481.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5857
(23) Shi, Y.; Do, J. T.; Desponts, C.; Hahm, H. S.; Scholer, H. R.; Ding,
S. A combined chemical and genetic approach for the generation of
induced pluripotent stem cells. Cell Stem Cell 2008, 2, 525–528.
(24) Shi, Y.; Desponts, C.; Do, J. T.; Hahm, H. S.; Scholer, H. R.; Ding,
S. Induction of pluripotent stem cells from mouse embryonic
fibroblasts by Oct4 and Klf4 with small-molecule compounds. Cell
Stem Cell 2008, 3, 568–574.
(25) Greiner, D.; Bonaldi, T.; Eskeland, R.; Roemer, E.; Imhof, A.
Identification of a specific inhibitor of the histone methyltransfer-
ase SU(VAR)3-9. Nature Chem. Biol. 2005, 1, 143–145.
(26) Cole, P. A. Chemical probes for histone-modifying enzymes.
Nature Chem. Biol. 2008, 4, 590–597.
(33) Prepared as a racemic mixture.
(34) Min, J.; Allali-Hassani, A.; Nady, N.; Qi, C.; Ouyang, H.; Liu, Y.;
MacKenzie, F.; Vedadi, M.; Arrowsmith, C. H. L3MBTL1 recog-
nition of mono- and dimethylated histones. Nature Struct. Mol.
Biol. 2007, 14, 1229–1230.
(35) Wigle, T. J.; Provencher, L. M.; Norris, J. L.; Jin, J.; Brown, P. J.;
Frye, S. V.; Janzen, W. P. Accessing Protein Methyltransferase and
Demethylase Enzymology Using Microfluidic Capillary Electro-
phoresis. Chemistry & Biology 2010, in press.
(36) Morrison, J. F. Kinetics of the reversible inhibition of enzyme-
catalysed reactions by tight-binding inhibitors. Biochim. Biophys.
Acta 1969, 185, 269–286.
(27) Frye, S. V.; Heightman, T.; Jin, J. Targeting Methyl Lysine. Annu.
Rep. Med. Chem. 2010, in press, DOI: 10.1016/S0065-7743(10)45020-4.
(28) Frye, S. V. The art of the chemical probe. Nature Chem. Biol. 2010,
6, 159–161.
(29) Chang, Y.; Zhang, X.; Horton, J. R.; Upadhyay, A. K.; Spannhoff,
A.; Liu, J.; Snyder, J. P.; Bedford, M. T.; Cheng, X. Structural basis
for G9a-like protein lysine methyltransferase inhibition by BIX-
01294. Nature Struct. Mol. Biol. 2009, 16, 312–317.
(30) Quinn, A. M.; Allali-Hassani, A.; Vedadi, M.; Simeonov, A. A
chemiluminescence-based method for identification of histone
lysine methyltransferase inhibitors. Molecular BioSystems 2010,
6, 782–788.
(31) Liu, F.; Chen, X.; Allali-Hassani, A.; Quinn, A. M.; Wasney, G. A.;
Dong, A.; Barsyte, D.; Kozieradzki, I.; Senisterra, G.; Chau, I.;
Siarheyeva, A.; Kireev, D. B.; Jadhav, A.; Herold, J. M.; Frye,
S. V.; Arrowsmith, C. H.; Brown, P. J.; Simeonov, A.; Vedadi, M.;
Jin, J. Discovery of a 2,4-diamino-7-aminoalkoxyquinazoline as a
potent and selective inhibitor of histone lysine methyltransferase
G9a. J. Med. Chem. 2009, 52, 7950–7953.
(37) Williams, J. W.; Morrison, J. F. The kinetics of reversible tight-
binding inhibition. Methods Enzymol. 1979, 63, 437–467.
(38) Sakurai, M.; Rose, N. R.; Schultz, L.; Quinn, A. M.; Jadhav, A.;
Ng, S. S.; Oppermann, U.; Schofield, C. J.; Simeonov, A. A
miniaturized screen for inhibitors of Jumonji histone demethylases.
Mol. Biosyst. 2010, 6, 357–364.
(39) Niesen, F. H.; Berglund, H.; Vedadi, M. The use of differential
scanning fluorimetry to detect ligand interactions that promote
protein stability. Nature Protoc. 2007, 2, 2212–2221.
(40) Vedadi, M.; Niesen, F. H.; Allali-Hassani, A.; Fedorov, O. Y.;
Finerty, P. J., Jr.; Wasney, G. A.; Yeung, R.; Arrowsmith, C.; Ball,
L. J.; Berglund, H.; Hui, R.; Marsden, B. D.; Nordlund, P.;
Sundstrom, M.; Weigelt, J.; Edwards, A. M. Chemical screening
methods to identify ligands that promote protein stability, protein
crystallization, and structure determination. Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 15835–15840.
(41) Allali-Hassani, A.; Wasney, G. A.; Chau, I.; Hong, B. S.; Senis-
terra, G.; Loppnau, P.; Shi, Z.; Moult, J.; Edwards, A. M.; Arrow-
smith, C. H.; Park, H. W.; Schapira, M.; Vedadi, M. A survey of
proteins encoded by nonsynonymous single nucleotide polymorph-
isms reveals a significant fraction with altered stability and activity.
Biochem. J. 2009, 424, 15–26.
(32) Collazo, E.; Couture, J. F.; Bulfer, S.; Trievel, R. C. A coupled
fluorescent assay for histone methyltransferases. Anal. Biochem.
2005, 342, 86–92.