Novel SAR for Quinazoline Inhibitors of EHMT1 and EHMT2

Ruben Leenders, Remco Zijlmans, Bart van Bree, Marc van de Sande, Federica
Trivarelli, Eddy Damen, Anita Wegert, Daniel Müller, Jan Erik Ehlert, Daniel
Feger, Carolin Heidemann-Dinger, Michael Kubbutat, Christoph Schächtele,
Danny C. Lenstra, Jasmin Mecinović, Gerhard Müller

Novel SAR for Quinazoline Inhibitors of EHMT1 and EHMT2
and Gerhard Müllerd
aMercachem BV, Kerkenbos 1013, 6546 BB Nijmegen, The Netherlands
bProQinase GmbH, Breisacher strasse 117, 19106 Freiburg im Breisgau, Germany
cRadboud University, Institute for Molecules and Materials, Heyendaalseweg 135, 6525 AJ Nijmegen, The Netherlands
dGotham Therapeutics, 430 East 29th Street, New York, NY 10016, USA
Histone Lysine Methyl Transferases (HKMTs) comprise a group of epigenetic writers
responsible for the transfer of one to three methyl groups from cofactor S-adenosyl￾methionine (SAM) to the terminal amino-group of the target lysine residues in histone tails1
and several non-histone targets,2 including tumour suppressor p53.3 To date, more than 5O
human HKMTs have been identified4 and many of them have been implicated in various
diseases.5,6,7 Besides EZH2 and DOT1L, EHMT1 (Euchromatic histone methyltransferase 1,
also known as G9a-like protein, GLP) and EHMT2 (Euchromatic histone methyltransferase 2,
also known as G9a) are among the most widely studied HKMTs. EHMT1 and EHMT2 are both
capable of mono-, di- or tri-methylation of the lysine-9 residue of histone-3 (H3K9). In
addition, EHMT2 has also been shown to methylate H3K27.8,9
Dysregulation of the histone methylation state has been implicated in many cancers and
other diseases.10,11,12,13 Histone methylation is upregulated in hepatocellular carcinoma,14 B
cell acute lymphoblastic leukaemia15 and lung cancers.16 In addition, enhanced expression
of EHMT2 in aggressive lung cancer correlates with poor prognosis while its knockdown in
Corresponding author: e-mail: [email protected]
ARTICLE INFO ABSTRACT

Detailed structure activity relationship of two series of quinazoline EHMT1/EHMT2
inhibitors (UNC0224 and UNC0638) have been elaborated. New and active alternatives
are presented for the ubiquitous substitution patterns found in literature for the linker to
the lysine mimicking region and the lysine mimic itself. These findings could allow for
advancing EHMT1/EHMT2 inhibitors of that type beyond tool compounds by fine￾tuning physicochemical properties making these inhibitors more drug-like.

highly invasive lung cancer cells suppressed metastasis in an in vivo mouse model.17 In
prostate cancer cells EHMT2 knockdown caused significant morphological changes and
inhibition of cell growth.18
Because EHMT1 and EHMT2 are not functionally redundant and have distinct biological
functions, we also studied the EHMT1/2 selectivity of these two closely related HKMTs.
In conclusion, HKMTs are promising targets and some NCE’s have reached human clinical
trials.19,20
Whereas a handful of HKMT SAM-site inhibitors is described in literature,21 histone-site
inhibitors are prevailing and seem to be more promising with respect to their optimization
and development potential. BIX-01294 (Figure-1) was the first selective EHMT1/EHMT2
inhibitor that emerged from a high-throughput screen.22 Optimization of this scaffold based
on a complex structure with EHMT223 by adding a lysine mimicking side chain designed to
interact with the narrow lysine binding channel, led to the discovery of the potent and
selective EHMT1/EHMT2 inhibitor UNC0224.
24 In further studies, a number of different
lysine mimics were employed, identifying UNC032125 and E7226 as the most interesting
compounds. Subsequent generation of analogues focussed on improving cell membrane
permeability while maintaining high in vitro potency. This approach resulted in the discovery
of UNC0646 and UNC0631.
27 Further optimization of these quinazoline scaffolds resulted
in the development of UNC0638 which shows cellular activity alongside low cellular
toxicity.28 From this cyclohexyl substituted quinazoline series, UNC0642 has emerged and
was claimed29 to be the first in vivo chemical probe for EHMT1/EHMT2 possessing improved
pharmacokinetics. In a recent paper,30 quinazolines lacking a lysine mimicking tail were
Figure-1: Literature compounds (selection)
described and one of them, MS012, was claimed to possess a 140-fold selectivity for
EHMT1 over EHMT2.
A completely distinct class of compounds inhibiting EHMT1/EHMT2 was identified in a
peptide-based AlphaLISA high throughput screening quantifying the levels of H3K9
dimethylation. This resulted in the development of aminoindole A-366.
31
We set out to investigate the underexplored regions in the quinazoline inhibitors UNC0224
and UNC0638 (Figure-2, quinazoline-2- and 4-substituents a and b respectively). As the 2-
and 4-substitution patterns of these inhibitors are relatively well explored, we focussed on
the eastern part of the molecule where the lysine mimic and the 6-substituent reside.
1) In order to investigate wether additional interactions can be picked up, we prepared
amides, reversed amides and amines and to learn whether the ether linkage is
needed, C-coupled lysine mimics were prepared (orange in Figure-2).
2) Rigidification of a flexible part in a drug molecules is often a driver for potency and
hence we incorporated an alkyne and/or a ring-system.
3) In order to obtain more drug-like properties, the basicity of the lysine capping-group
was substituted by non-basic end-groups (magenta in Figure-2) .
4) To further explore the SAR, we replaced the 6-methoxy group by other small groups
to see if the methoxy is crucial for inhibitory activity.32 (green in Figure-2)
Figure-2: UNC0224 and UNC0638 structural modifications
Figure-3: Docking poses of 5n and 6n
Docking pose of compound 5n (A: green) and 6n (B: green) in co-crystal structure of EHMT2, UNC0638 (orange)
and SAH (cyan) (PDB-code 3RJW). The nitrogens of the lysing mimicking tail of UNC0638 and 5n and 6n exactly
overlap and are positioned in the lysine channel pointing towards the SAH sulphur atom (yellow).
A B
UNC0224 point mutations (Scheme-1)
We thought to use an advanced central building block, facilitating the syntheses of multiple
types of targets to scan the SAR as broadly as possible. In this respect, novel triflate 2 was
synthesized from the known phenol 1.
24 Downstream chemistry from 2 as follows:
1) In order to arrive at amides 4, methyl ester 3 was prepared by a carbonylation
reaction. From 3, amides 4a-f were prepared in neat amine at elevated
temperatures as amide coupling reactions with acid 3a to give 4 failed under all
conditions tested (Route-A). 3b was assessed from 3 with ammonia in MeOH.
2) Alkynes 5g-q prepared by Sonogashira reactions. From these, after reduction of the
alkyne moiety, targets 6g-q were accessed (Route-B). Reduction of the pyridyl
moiety gave 6r and 6t. Compounds 6v-x (see Table-1) were obtained as side￾products.
3) Products 7p, q and u were obtained by Suzuki cross-coupling and 7o by a Stille
reaction. Selective reduction of the pyridyl-group in 7o-q gave 8r-t (Route-C).
4) Aniline 13 (Route-D, Scheme-2) was prepared in a 6-step sequence from known 10
which was accessed from 9.
33 From 13 reversed amides 14a-d were prepared.
Unfortunately, Buchwald-Hartwig chemistry on triflate 2 was not feasible under all
conditions tested: aminolysis of the triflate was the exclusive reaction seen.
UNC0638 point mutations (Scheme-3)
Here again we employed a central advanced building block (21) giving access to a wide
range of different targets for broad SAR scanning. Analogousely to literature,28 compounds
20 were prepared in a microwave mediated reaction form methyl anthranilates 18 and
cyclohexanecarbonitrile. Then, the 4-position of quinazoline scaffold 20 was functionalized
with 1-isopropylpiperidin-4-amine either by substitution of the corresponding 4-chloride (for
R = F and Cl) or by a direct PyBOP coupling (for R = OMe, OEt and OCF3) giving
intermediates 21. From these bromides we were able to synthesize:
1) Alkynes 22-25 by Sonogashira reaction. From these, after reduction of the alkyne
moiety, targets 26-29 were accessed (Route-E).
2) Amine-linked products 30 and 31 by Buchwald-Hartwig chemistry (Route-F).
3) Cyclic C-coupled lysine mimic 32 by Suzuki-Miyaura cross-coupling. Reduction of the
pyridyl-group gave piperidine 33 (Route-G).
Scheme-1: Synthesis Routes-A to -C
i: PhN(Tf)2, K2CO3, THF, rt; ii: MeOH, CO, Pd(OAc)2, dppp, Et3N, DMF, 90 °C; iii: neat amine, 70 °C; iv: Pd(PPh3)2Cl2, CuI, Cs2CO3, MeCN,
80 °C; v: H2, Pd/C, EtOH; vi: 7o: 2-(tributylstannyl)pyridine, Pd(PPh3)4, CsF2, LiCl, CuI, THF, 65 °C and 7p, q and u: boronic acid,
Pd(PPh3)2Cl2, Cs2CO3, 1,2-DME, water, 80 °C; vii: H2, PtO2, TFA, i-PrOH.

i: KOCN, AcOH, H2O; ii: NaOH, H2O; iii: POCl3, Et4NCl, MeCN, 82 °C; iv: 1-methylpiperidin-4-amine, THF,
DIPEA, rt; v: 1-methylhomopiperazine, 4N HCl in dioxane, 2-PrOH, 125 °C 30 min. MW; vi: H2, Pd/C, EtOH;
vii: for a and b: acryloyl chloride, Et3N, DCM, 0 °C; for c: 5-azidopentanoyl chloride, DMAP, Et3N, DCM, 0 °C;
for d: 3-isocyanatopropanenitrile, MeCN, 0 °C; viii: for a: NH3, MeOH, 70 °C; for b: Me2NH, THF, MeOH, 55
°C; for c: H2, Pd/C, MeOH; for d: H2, Ra-Ni, NH3, MeOH, H2O.
i: KNO3, H2SO4, 0 °C -> rt; ii: NBS, H2SO4, 80 °C; iii: Fe, NH4Cl, i-PrOH, water, 100 °C; iv: 1N HCl in MeOH, 65°C; v: MeOH,
KOtBu, 0 °C > rt; vi: EtOH, KOtBu, 0 °C -> rt; vii: Fe, NH4Cl, MeOH, THF, water, 65 °C; viii: cyclohexanecarbonitrile, 4N HCl
in dioxane, 130 °C microwave, 4-20 h; ix (for R = F, Cl): 1) SOCl2, pyridine, 70 °C; 2) 1-isopropylpiperidin-4-amine, THF, 70
°C; x (for R = OMe, OEt, OCF3): 1-isopropylpiperidin-4-amine, PyBOP, DBU, DMF, rt; xi: alkyne, Pd(PPh3)2Cl2, CuI, Cs2CO3, MeCN, 80 °C; xii: H2, Pd/C, EtOH; xiii: amine, Pd2dba3, BINAP, LiHMDS, toluene, 80 °C; xiv: boronic acid, Pd(PPh3)2Cl2, Cs2CO3, 1,2-DME, water, 80 °C; xv: H2, PtO2, TFA, i-PrOH.
ix or x
Biochemical assay
The in-vitro methyltransferase activity of recombinant EHMT1/EHMT2 was monitored by
means of an enzyme coupled detection of the HKMT-activity associated with the generation
of SAH (S-adenosyl-homocysteine, Scheme-4). EHMT1 and EHMT2 transfer a methyl group
from co-substrate SAM (S-adenosyl-methionine) to their substrate, a peptide derived from
the human histone H3 sequence (amino acids 1-25). SAH is generated as a consequence
and SAH levels were detected by enzymatic conversion of SAH to adenosine and 5-ribosyl￾homocysteine using recombinant 5′-methylthioadenosine/S-adenosylhomocysteine
nucleosidase (MTAN). Adenosine is further converted to AMP by adenine
phosphoribosyltransferase (APRT) catalysed condensation with phosphoribosyl￾pyrophosphate (PRPP). AMP levels were detected by means of the AMP-Glo® detection kit.
The assay approach was validated by determination of IC50 values of published reference
compounds UNC0224 and UNC0638, which gave comparable values as previously
published.22,23,28 Table-1 summarizes IC50 values against EHMT1 and EHMT2 of all
compounds derived from UNC0224 structural mutations; Table-2 those derived from
UNC0638.
Cellular assay
Cellular methyltransferase activity was monitored for selected compounds after 48 h
incubation in AsPC-1 pancreas tumour cell-line. Cell lysates were analysed for the presence
of histone H3K9Me2 in comparison to total histone H3 (panHis3) by means of Western blot
detection. Reduction of H3K9Me2 in cells represent a relevant readout for the cellular
inhibitor of EHMT1 and EHMT2.
Significant cellular potencies in reduction of H3K9Me2 levels has been shown for UNC0638
and A-366. A direct comparison of compound 5n and 6n with reference compounds
UNC0224, UNC0638 and A-336 in the pancreatic cancer cell-line AsPC-1 demonstrates
that both compounds have similar potencies in reducing cellular levels of H3K9Me2 (Table-3
and Figure-4). This also suggests good cell permeability for both compounds.28,34
MALDI-TOF Assay
A selected group of compounds was used for monitoring the inhibition of EHMT2/EHMT1
using a MALDI-TOF MS based assay. Added peak areas of each methylation state were
compared to a blank run without inhibitor and the % inhibition was calculated.
In general, the results were found to be in good agreement with the IC50 values obtained
by the in vitro biochemical assay. Compounds 4b, 4d-f, 7p and 8s inhibited EHMT2 and
EHMT1 activity less than 50% even at 100 µM concentration (Table-4 and Figure-5).
Excellent inhibition, i.e. mainly unmethylated H3K9 peptide was observed for compounds 5h
and 6h even at 10 µM concentration. In line with biochemical assay, the test compounds
inhibited EHMT1 typically 20-30% better than EHMT2.
Scheme-4: Biochemical assay setup
POTENT: significant inhibition at 10 µM
GOOD: significant inhibition at 100 µM, some inhibition at 10 µM
POOR: no inhibition
Figure-5: MALDI-TOF functional assay results
From the above data SAR is seen as follows:
General
HKMT-inhibitors described in literature so far addressed the lysine channel exclusively via
an ether-linked lysine mimic. We showed that a lysine mimic can be connected to the
core of the inhibitor by other functionalities i.e. by an amide-, reversed amide-, amine￾and carbon connector. In addition, we found that a basic lysine end-group is not required
for inhibitory activity.
Linker
In the UNC0224 series, quinazoline-NH-C(O)-R amides 14a-d are far more potent than
the reversed amides -C(O)-NH-R 4a-f, these latter linkers are hardly tolerated.
A good alternative to the lysine mimic–O linkage is the C-linkage, which is of similar
activity but easier synthetically tractable in both the UNC0224 and -0638 series.
EHMT1/EHMT2 Selectivity
Most compounds show a similar activity for EHMT1 and EHMT2 (factor 1-5 difference).
Higher selectivity for EHMT1 (ratio>10) is seen in the C-linked series for compounds
23a, 24a and 27a. The most pronounced selectivity was observed for the Buchwald
products 30a and 30c with selectivity factors of ~30 and ~50 respectively. In the
UNC0224 series this selectivity is less pronounced.
Quinazoline-6-position
With respect to the quinazoline-6-position in the UNC0638 series: replacing the OMe for
an OEt reduced the activity with a factor of >100 (compare, e.g. 23a/b and 30a/b). In
addition, F-, Cl- and F3CO- are not tolerated. A special case is the cluster of compounds
30. In these piperazine Buchwald products a fluorine in 6-position is very well tolerated.
Lysine mimic
In addition to all literature inhibitors with flexible lysine tails, we found that rigidifying the
mimic with an alkyne in both series gave highly active inhibitors as well. Furthermore, a
basic cap-group is not required as lactams 5l, 6l and 24a gave good inhibitors.
These findings give the opportunity to fine-tune physicochemical properties of the
EHMT1/EHMT2 inhibitors by choosing less flexible linkers with non-basic cap groups and
advance EHMT1/EHMT2 inhibitors beyond tool compounds.
Acknowledgments
This work was financed by an Eurostars fund, project application E!7300 iSelect.
The authors declare no conflicting interests.
References and Notes
All reagents and solvents were purchased from commercial sources and were used without
further purification. Dry solvents were used as purchased. 1H-NMR spectra were recorded on
a Bruker Avance 400 MHz spectrophotometer with TMS as internal standard in CDCl3 or d6-
DMSO as indicated on the NMR spectrum. LCMS was recorded on an Agilent 1100 Bin.
Machine; the actual method mentioned on the chromatogram.
Preparative silica TLC’s were used as purchased from Merck KgaA type: PLC silica gel P254, 2
mm with concentration zone. Compounds 5n and 6n were purified on a Reveleris® machine
(Grace) employing a Xselect CSH C18 column run in the basic eluent as indicated in the
preparative procedure.
Synthesis of triflate 2
To a suspension of phenol 1124 (4.04 g, 10.09 mmol) in dry THF (100 mL) was added
K2CO3 (2.79 g, 20.17 mmol, 2.0 eq.) and N-phenyl-bis(trifluoromethanesulfonimide) (3.96
g, 11.10 mmol, 1.1 eq.) and the reaction mixture was stirred at room temperature under a
nitrogen atmosphere for 1h. The solids dissolved almost completely. After 6h, LCMS
revealed almost complete and clean conversion to a product with the desired target mass.
The reaction mixture was then diluted with DCM (200 mL) and washed with saturated
aqueous NH4Cl (150 mL). The aqueous layer was back-extracted twice with DCM (150 mL)
and the combined organic extract was dried over anhydrous Na2SO4, filtered and evaporated
at 40 °C under reduced pressure to afford a light brown foamy residue (8.27 g). The crude
material was purified by flash column chromatography on a Reveleris® (300 g silica column,
100 mL/min, gradient: 0-10% 7 N NH3 in MeOH/DCM) collecting on UV response. All
collected fractions were pooled, evaporated and co-evaporated twice with diisopropyl ether
and once with Et2O at 40 °C under reduced pressure to obtain a yellow foam. The foam was
dried at 40 °C in a vacuum stove overnight yielding 4.39 g (8.24 mmol, 80.0% yield) of 2
as a yellow foam;
LCMS: m/z(+)= 533 [M+H]+, 98.57% (DAD).
water DMSO-d6 TMS
Synthesis of 6n
To a solution of 5n (20 mg, 0.043 mmol) in absolute EtOH (4 mL) was added 10%
palladium on activated carbon (23 mg, 0.022 mmol, 0.5 eq.). The suspension was flushed
with nitrogen and then brought under a hydrogen atmosphere and vigorously stirred for 4h.
LCMS showed formation of a main product with desired target mass. The reaction mixture
was flushed with nitrogen and the catalyst was filtered off over a short pad of celite, washed
with EtOH (2x 4 mL) and DCM (2x 4 mL). The filtrate was concentrated at 40 °C under
reduced pressure and the residue was purified on a preparative silica TLC by elution with
(DCM/7N NH3 in MeOH 95:5) followed by (DCM/7N NH3 in MeOH 9:1). The highest running
product was scraped off and the silica was eluted with DCM/MeOH 9:1 (50 ml). The extract
was concentrated at 40 °C under reduced pressure and the residue was re-dissolved in
CH3CN/H2O 10:1 (4 mL) and lyophilized to obtain 8 mg (0.017 mmol, 37.7% yield) of 6n as

Biology
Production of recombinant EHMT1 and EHMT2
cDNAs coding for the open reading frames of EHMT1 (amino acids 794-1294, GenBank
accession number NM_024757.4) and EHMT2 (amino acids 785-1210, GenBank accession
number NM_006709.3) were isolated by PCR from a human cDNA library. cDNAs coding for
APRT (full length, GenBank accession number NC_000913.3) and MTAN (full length,
GenBank accession number NC_010473.1) were isolated by PCR from E. coli genomic DNA.
EHMT1 and EHMT2 were expressed as N-terminally GST-tagged proteins in insect cells using
the Baculo virus expression system and purified by GST-affinity chromatography. APRT and
MTAN were expressed as N-terminally GST-tagged proteins in E. coli and purified by GST￾affinity chromatography.
Biochemical Assay
In-vitro methyltransferase activity of recombinant EHMT1 and EHMT2 was monitored by
means of an enzyme coupled detection of the HKMT-activity associated with the generation
of SAH (S-adenosyl-homocysteine, Scheme-4). EHMT1 and EHMT2 transfer a methyl group
from co-substrate SAM (S-adenosyl-methionine) to their substrate, a peptide derived from
the human Histone H3 sequence (amino acids 1-25), SAH is generated as a consequence
and SAH levels were detected by enzymatic conversion of SAH to adenosine and 5-ribosyl￾homocysteine using recombinant MTAN. Adenosine was further converted to AMP by APRT
catalysed condensation with phosphoribosyl-pyrophosphate (PRPP). AMP levels were
detected by means of the AMP-Glo® detection kit as specified by the supplier (Promega
GmbH, Mannheim, Germany). The assay signal was SAH concentration dependent, while no
assay signal was generated by increasing SAM concentrations (data not shown). The SAM
assay concentration was 20 µM corresponding to the apparent SAM-KM of EHMT1 and
EHMT2. The PRPP concentration was 40 µM. EHMT1 and EHMT2 were used in the assay at
concentrations of 9.4 and 10.3 nM respectively. The assay was performed in 50 mM HEPES
pH 7.5 with 100 mM NaCl, 5 mM MgCl2, 3 µM Na-ortho-vanadate and 0.01% Triton X-100.
compound was added to the HKMT/substrate mix and the reaction was started by the
addition of SAM. The reaction was incubated for 30 min at 30 °C. APRT and MTAN were
added to obtain concentrations of 40 and 100 nM respectively together with the PRPP and
AMP-Glo® reagents. The conversion of SAH to AMP was allowed to proceed for 60 min at 24
°C. The AMP-Glo® detection reagent was added, and the reaction was incubated for 30 min
at 24 °C. Luminescence signals were read out using a Victor plate reader (PerkinElmer,
USA). IC50 curves were obtained using 10 concentrations in semi log dilution starting with a
highest assay concentration of 10 µM or 100 µM.
The assay signal-to-background window was tested for each assay by measuring the signal
in presence of MTAN, APRT, SAM and AMP-Glo detection reagent in absence of HKMT￾enzyme (low control) and in presence of HKMT-enzyme (high control). Typically the
resulting luminscence signal was about 30000 cps in absence and 200000 cps in presence of
HKMT enzyme, Z’ factors calculated from 8 low and 8 high controls per 96 well assay plate
were in the range of >0.7. Assay signal dependency on the concentration of SAM and SAH
was evaluated in absence of HKMT-enzyme during the set-up of the assay. In presence of
the complete assay detection system (including MTAN) no signal dependency on SAM
concentration could be detected, while a clear concentration-dependency was seen for SAH
in the tested range of 0-100 µM.

Cellular Assay
3×105 cells were seeded overnight in 6 well plates in DMEM/10% FCS complete medium.
The cells were treated with 10 µM of the respective compound at a maximum DMSO
concentration of 0.1%. After 48 h of incubation, attached as well as detached cells were
harvested by trypsination, washed once with PBS and lysed in SDS-PAGE buffer. Upon SDS￾PAGE and Western blotting, histone H3K9Me2 as well as panHistone H3 were detected using
anti-panHistone H3 (CST Cat#9715) and anti-histone H3K9Me2 (CST Cat# 4658) antibodies
respectively, according to the providers recommendations. Chemiluminescent blots were
developed on X-ray film and scanned. The respective histone bands were quantified and
HisH3K9Me2/panHisH3 ratios were calculated. Setting the ratio of untreated cells to 100%,
effects of compounds were calculated as percental values of that ratio.
MALDI-TOF Assay
A MALDI-TOF MS based assay monitoring the mono-, di-, and trimethylation of a 15-mer
histone H3 peptide mimic (sequence: ARTKQTARKSTGGKA) was performed. Briefly,
recombinant enzyme (EHMT2 or EHMT1, at 2 µM final concentration) and inhibitor (1 µL of a
100% DMSO stock solution at appropriate concentration) were combined in 50 mM TRIS pH
8.0, to this was added histone peptide (final concentration 100 µM) and SAM (final
concentration 400 µM) to obtain a total reaction volume of 20 µL, containing 5% DMSO
(v/v). The mixture was incubated for 1 hour at 37 °C, after which it was quenched by the
addition of 20 µL of MeOH. For MALDI-TOF MS analysis, a 5 µL aliquot of the quenched
reaction mixture was mixed with a saturated solution of α-cyano-4-hydroxycinnamic acid
matrix (CHCA, in MeCN/H2O (1/1)) (4/1 (v/v)) and from this 1 µL was deposited onto the
MALDI plate for crystallization. MS data were recorded on a Bruker Microflex LRF MALDI￾TOF system. Total peak area of each methylation state, including all isotopes and adducts
([M+H]+ and [M+Na]+) were used for determination of enzymatic activity in the presence
of inhibitors relative to the activity in the absence of inhibitors (i.e. only with 5% DMSO).

Rea S, Eisenhaber F, O’Carroll D, Strahl BD, Sun ZW, Schmid M, Opravil S, Mechtler K, Ponting CP, Allis CD,
Jenuwein T. Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature
2000;406:593-599.
Huang J, Berger SL. The emerging field of dynamic lysine methylation of non-histone proteins. Curr Opin Genet
Dev 2008;18:152-158.
Huang J, Dorsey J, Chuikov S, Perez-Burgos L, Zhang X, Jenuwein T, Reinberg D, Berger SL. G9a and GLP
methylate lysine 373 in the tumor suppressor p53. J Biol Chem, 2010;285:9636-9641.
A) Arrowsmith CH, Bountra C, Fish PV, Lee K, Schapira M. Epigenetic protein families: A new frontier for drug
discovery. Nat Rev Drug Discovery. 2012;11:384−400. B) H. Kaniskan Ü, Martini ML, Jin J. Inhibitors of Protein
Methyltransferases and Demethylases. Chem Rev 2018;118:989-1068.
5Copeland RA, Solomon ME, Richon VM. Protein methyltransferases as a target class for drug discovery. Nat Rev
Drug Discovery 2009;8:724−732.
Campagna-Slater V, Mok MW, Nguyen KT, Feher M, Najmanovich R, Schapira M. Structural chemistry of the
histone methyltransferases cofactor binding site. J Chem Inf Model. 2011;51:612−623.
Wu H, Min J, Lunin VV, Antoshenko T, Dombrovski L, Zeng H, Allali-Hassani A, Campagna-Slater V, Vedadi M,
Arrowsmith CH, Plotnikov AN, Schapira M. Structural biology of human H3K9 methyltransferases. PLoS One
2010;5; e8570.
Tachibana M, Sugimoto K, Fukushima T, Shinkai Y. SET domain-containing protein, G9a, is a novel lysine￾preferring mammalian histone methyltransferase with hyperactivity and specific selectivity to lysines 9 and
27 of histone H3. J Biol Chem. 2001;276:25309–17.
Wu H, Chen X, Xiong J, Li Y, Li H, Ding X, Liu S, Chen S, Gao S, Zhu B. Histone methyltransferase
G9a contributes to H3K27 methylation in vivo. Cell Res. 2011;21:365–367.
10 Chi P, Allis CD, Wang GG. Covalent histone modifications: miswritten, misinterpreted and mis-erased in human
cancers. Nat Rev Cancer 2010;10:457-469.
11 Copeland RA, Moyer MP, Richon VM. Targeting genetic alterations in protein methyltransferases for personalized
cancer
therapeutics. Oncogene 2013;32;939−946.
12 Schneider R, Bannister AJ, Kouzarides T. Unsafe SETs: histone lysine methyltransferases and cancer. Trends
13 Li J, Zhu S, Ke X-X, Cui H. Role of several histone lysine methyltransferases in tumor development (Review),
Biomedical Reports 2016;4: 293 – 299.
14 Kondo Y, Shen L, Suzuki S, Kurokawa T, Masuko K, Tanaka Y, Kato H, Mizuno Y, Yokoe, M, Sugauchi F,
Hirashima
N, Orito E, Osada H, Ueda R, Guo Y, Chen X, Issa JP, Sekido Y. Alterations of DNA methylation and histone
modifications
contribute to gene silencing in hepatocellular carcinomas. Hepatol Res. 2007;37:974−983.
15 Huang J, Dorsey J, Chuikov S, Zhang X, Jenuwein T, Reinberg D, Berger SL. G9a and Glp methylate lysine 373 in
thetumor suppressor p53. J Biol Chem. 2010;285:9636−9641.
16 Watanabe H, Soejima K, Yasuda H, Kawada I, Nakachi I, Yoda S, Naoki K, Ishizaka A. Deregulation of histone
lysinemethyltransferases contributes to oncogenic transformation of human bronchoepithelial cells. Cancer Cell Int.
2008;8:15.
17 Chen M-W, Hua K-T, Kao H-J, Chi C-C, Wei L-H, Johansson G, Shiah S-G, Chen P-S, Jeng Y-M, Cheng T-Y, Lai
T-C, Chang J-S, Jan Y-H, Chien M-H, Yang C-J, Huang M-S, Hsiao M, Kuo M-L. H3K9 histone methyltransferase G9a
promotes lung cancer invasion and metastasis by silencing the cell adhesion molecule Ep-CAM. Cancer Res.
2010;70:7830-7840.
18 Kondo Y, Shen L, Ahmed S, Boumber Y, Sekido Y, Haddad BR, Issa J-P. Downregulation of histone H3 lysine 9
methyltransferase G9a induces centrosome disruption and chromosome instability in cancer cells. PLoS One
2008;3:e2037.
19 Sweis RF, Michaelides MR. Recent advances in small molecule modulation of epigenetic targets: discovery and
development of histone methyltransferase and bromodomain inhibitors. Annu. Rep. Med. Chem.
2013;48:185−203.
20 Kuntz KW, Campbell JE, Keilhack H, Pollock RM, Knutson SK, Porter-Scott M, Richon VM, Sneeringer CJ, Wigle TJ,
Allain CJ, Majer CR, Moyer MP, Copeland RA, Chesworth R. The Importance of Being Me: Magic Methyls,
Methyltransferase Inhibitors, and the Discovery of Tazemetostat. J Med Chem. 2016;59:1556-1564.
21 Yuan Y, Wang Q, Paulk J, Kubicek S, Kemp MM, Adams DJ, Shamji LF, Wagner BK, Schreiber SL. A Small￾Molecule Probe of the Histone Methyltransferase G9a. ACS Chem Biol. 2012;7:1152-1157. B) Devkota K, Lohse B,
Liu Q, Wang M-W, Staerk D, Berthelsen J, Clausen RP. Analogues of the Natural Product Sinefungin as Inhibitors of
EHMT1 and EHMT2 , ACS Med Chem Lett. 2014;5:293-297.
22 Kubicek S, O’Sullivan RJ, August EM, Hickey ER, Zhang Q, Teodoro ML, Rea S, Mechtler K, Kowalski JA, Homon
CA, Kelly TA, Jenuwein T. Reversal of H3K9me2 by a small-molecule inhibitor for the G9a histone
methyltransferase. Mol. Cell 2007;25: 473-481.
23 Chang Y, Zhang X, Horton JR, Upadhyay AK, Spannhoff A, Liu J, Snyder JP, Bedford MT, Cheng X. Structural
basis for G9a-like protein lysine methyltransferase inhibition by BIX-01294. Nat Struct Mol Biol. 2009;16: 312-317.
24 Liu F, Chen X, Allali-Hassani A, Quinn AM, Wasney GA, Dong A, Barsyte D, Kozieradzki I, Senisterra G, Chau I,
Siarheyeva A, Kireev DB, Jadhav A, Herold JM, Frye SV, Arrowsmith CH, Brown PJ, 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.
25 Liu F, Chen X, Allali-Hassani A, Quinn AM, Wigle TJ, Wasney GA, Dong A, Senisterra G, Chau I, Siarheyeva A,
Norris JL, Kireev DB, Jadhav A, Herold JM, Janzen WP, Arrowsmith CH, Frye SV, Brown PJ, Simeonov A, Vedadi M,
Jin J. Protein lysine methyltransferase G9a inhibitors: Design, synthesis, and structure activity relationships of 2,4-
diamino-7-aminoalkoxy-quinazolines. J Med Chem. 2010;53:5844-5857.
26 Chang Y, Ganesh T, Horton JR, Spannhoff A, Liu J, Sun A, Zhang X, Bedford MT, Shinkai Y, Snyder J P, Cheng X.
Adding a lysine mimic in the design of potent inhibitors of histone lysine methyltransferases. J Mol Biol.
2010;400;1-7.
27 Liu F, Barsyte-Lovejoy D, Allali-Hassani A, He Y, Herold JM, Chen X, Yates CM, Frye SV, Brown PJ, Huang J,
Vedadi M, Arrowsmith CH, Jin J. Optimization of Cellular Activity of G9a Inhibitors 7-Aminoalkoxy-quinazolines, J
Med Chem. 2011;54: 6139-6150.
28 Vedadi M, Barsyte-Lovejoy D, Liu F, Rival-Gervier S, Allali-Hassani A, Labrie V, Wigle TJ, DiMaggio PA, Wasney
GA, Siarheyeva A, Dong A, Tempel W, Wang S-C, Chen X, Chau I, Mangano T, Huang X-P, Simpson CD, Pattenden
SG, Norris JL, Kireev DB, Tripathy A, Edwards A, Roth BL, Janzen WP, Garcia BA, Petronis A, Ellis J, Brown PJ, Frye
SV, Arrowsmith CH, Jin J. A chemical probe selectively inhibits G9a and GLP methyltransferase activity in cells. Nat
Chem Biol. 2011;7:566-574.
29 Liu F, Barsyte-Lovejoy D, Li F, Xiong Y, Korboukh V, Huang X-P, Allali-Hassani A, Janzen WP, Roth BL, Frye SV,
Arrowsmith CH, Brown PJ, Vedadi M, Jin J. Discovery of an in Vivo Chemical Probe of the Lysine Methyltransferases
G9a and GLP, J Med Chem. 2013;56:8931-8942.
30 Xiong Y, Li F, Babault N, Dong A, Zeng H, Wu H, Chen X, Arrowsmith CH, Brown PJ, Liu J, Vedadi M, Jin J.
Discovery of Potent and Selective Inhibitors for G9a-Like Protein (GLP) Lysine Methyltransferase, J Med Chem.
2017;60:1876-1891.
31 Sweis RF, Pliushchev M, Brown PJ, Guo J, Li F, Maag D, Petros AM, Soni NB, Tse C, Vedadi M, Michaelides MR,
Chiang GG, Pappano WN. Discovery and Development of Potent and Selective Inhibitors of Histone
Methyltransferase G9a, ACS Med Chem Lett. 2014;5:205-209.
32 Rotili D, Tarantino D, Marrocco B, Gros C, Masson V, Poughon V, Ausseil F, Chang Y, Labella D, Cosconati S, Di
Maro S, Novellino E, Schnekenburger M, Grandjenette C, Bouvy C, Diederich M, Cheng X, Arimondo PB, Mai A,
Properly Substituted Analogues of BIX-01294 Lose UNC0638 Inhibition of G9a Histone Methyltransferase and Gain Selective
Anti-DNA Methyltransferase 3A Activity. PLoS ONE 2014;9: e96941.
33 Compound 10 (Scheme-2): Thompson S, Vallinayagam R, Adler MJ, Scott, RTW, Hamilton AD. Tetrahedron
2012;68;4501-4505.
34 Pappano WN, Guo J, He Y, Ferguson D, Jagadeeswaran S, Osterling DJ, Gao W, Spence JK, Pliushchev M, Sweis
RF, Buchanan FG, Michaelides MR, Shoemaker AR, Tse C, Chiang GG. The Histone Methyltransferase Inhibitor A-
366 Uncovers a Role for G9a/GLP in the Epigenetics of Leukemia, PLoS One. 205 Jul 6;0(7):e0376. doi:
10.1371/journal.pone.0131716. eCollection 2015.