Design and synthesis of selective CDK8/19 dual inhibitors: Discovery of 4,5-dihydrothieno[3′,4′:3,4]benzo[1,2-d]isothiazole derivatives

Article abstract of DOI:10.1016/j.bmc.2017.02.038

To develop a novel series of CDK8/19 dual inhibitors, we employed structure-based drug design using docking models based on a library compound, 4,5-dihydroimidazolo[3′,4′:3,4]benzo[1,2-d]isothiazole 16 bound to CDK8. We designed various [5,6,5]-fused tric

Full text of DOI:10.1016/j.bmc.2017.02.038

Accepted Manuscript
Design and synthesis of selective CDK8/19 dual inhibitors: Discovery of 4,5-
dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole derivatives
Koji Ono, Hiroshi Banno, Masanori Okaniwa, Takaharu Hirayama, Naoki
Iwamura, Yukiko Hikichi, Saomi Murai, Maki Hasegawa, Yuka Hasegawa,
Kazuko Yonemori, Akito Hata, Kazunobu Aoyama, Douglas R. Cary
PII:
S0968-0896(16)31440-7
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.02.038
BMC 13568
Reference:
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
21 December 2016
17 February 2017
17 February 2017
Please cite this article as: Ono, K., Banno, H., Okaniwa, M., Hirayama, T., Iwamura, N., Hikichi, Y., Murai, S.,
Hasegawa, M., Hasegawa, Y., Yonemori, K., Hata, A., Aoyama, K., Cary, D.R., Design and synthesis of selective
CDK8/19 dual inhibitors: Discovery of 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole derivatives, Bioorganic
& Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/j.bmc.2017.02.038
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Design and synthesis of selective CDK8/19 dual inhibitors: Discovery of
4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole derivatives
Koji Ono,a, * Hiroshi Banno,a Masanori Okaniwa,a Takaharu Hirayama,a Naoki Iwamura,b Yukiko Hikichi,c
Saomi Murai,a Maki Hasegawa,a Yuka Hasegawa,a Kazuko Yonemori,a Akito Hata,a Kazunobu Aoyama,a
and Douglas R. Carya, †
a Pharmaceutical Research Division, Takeda Pharmaceutical Co., Ltd., Shonan Research Center, 26-1,
Muraoka-Higashi 2-Chome, Fujisawa, Kanagawa 251-8555, Japan
b Takeda Union, 26-1, Muraoka-Higashi 2-Chome, Fujisawa, Kanagawa 251-8555, Japan
c Product Information Group Japan Oncology Business Unit, Takeda Pharmaceutical Co., Ltd., 12-10,
Nihonbashi 2-Chome, Chuo-ku, Tokyo 103-8686, Japan
* Corresponding author. Tel.: +81 466 32 1168; fax: +81 466 29 4448.
E-mail address: kouji.ono@takeda.com (K. Ono)
† Present address: Peptidream Inc.
Abstract
To develop a novel series of CDK8/19 dual inhibitors, we employed structure-based drug design
using docking models based on a library compound, 4,5-dihydroimidazolo[3',4':3,4]benzo[1,2-d]isothiazole
16 bound to CDK8. We designed various [5,6,5]-fused tricyclic scaffolds bearing a carboxamide group to
maintain predicted interactions with the backbone C=O and NH of Ala100 in the CDK8 kinase hinge
region. We found that 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole derivative 29a showed
particularly potent enzymatic inhibitory activity in both CDK8/19 (CDK8 IC50: 0.76 nM, CDK19 IC50: 1.7
nM). To improve the physicochemical properties and kinase selectivity of this compound, we introduced a
substituted 3-pyridyloxy group into the scaffold 8-position. The resulting optimized compound 52h showed
excellent in vitro potency (CDK8 IC50: 0.46 nM, CDK19 IC50: 0.99 nM), physicochemical properties, and
kinase selectivity (only 5 kinases showed <35% unbound fraction at 300 nM. CDK19: 4.6%, CDK8: 8.3%,
HASPIN: 23%, DYRK1B: 27%, HIP1: 32%). Based on a docking model of 52h bound to CDK8, we could
explain the highly specific kinase activity profile found for this compound, based on the interaction of the
pyridyl group of 52h interacting with Met174 of the CDK8 DMG activation loop.
In vitro pharmacological evaluation of 52h revealed potent suppression of phosphorylated STAT1
in various cancer cells. The high oral bioavailability found for this compound enabled in vivo studies, in
which we demonstrated a mechanism-based in vivo PD effect as well as tumor growth suppression in an
RPMI8226 human hematopoietic and lymphoid xenograft model in mouse [T/C: −1% (2.5 mg/kg, qd)].
Graphical abstract

Keywords
CDK8; CDK19; Cyclin-dependent kinases (CDKs); transcriptional regulation; STAT1; RPMI8226;
SW480; DMG; 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole
1. Introduction
Cyclin-dependent kinases (CDKs)1 are serine/threonine kinases that are regulated through their
interaction with a cyclin subunit. CDKs are categorized based on their action through one of two different
biological mechanisms: CDKs that play an important role in cell cycle regulation by binding to multiple
cyclins, and CDKs that control transcriptional regulation by binding to a single cyclin. In mammals, the
CDK family is grouped into the cell cycle-related (CDK1, CDK2, CDK4/6, etc.) and transcriptional (CDK7,
CDK8, CDK9, CDK12, CDK19, etc.) subfamilies. Although various pan-CDK inhibitors known to
function as cell-cycle regulators have demonstrated antitumor efficacy in clinical trials,2 the antitumor
activity of some of these have been reported as also arising from their effects on transcriptional regulation
through the inhibition of CDK7 and CDK9, which are members of the transcriptional CDK subfamily.3
CDK8 has been reported to act as a transcriptional regulator,4 and has been found to be
overexpressed in colorectal cancer (CRC) and adenocarcinoma relative to adjacent normal tissue.5
Furthermore, CDK8 was identified as a major oncogenic driver for amplification at 13q12.13–12q12.2 in
CRC in two RNA interference-based loss-of-function screens.6 Immunohistochemical analysis of the
CDK8 expression correlated with the tumor grade, stage and the local recurrence or metastasis in
extrauterine leiomyosarcoma,7 and is associated with shorter relapse-free survival in breast cancer.8
CDK8 plays various roles in different signaling pathways to modulate gene expression levels as a
part of the Mediator complex (CDK8, MED12, MED13, and cyclin-C),9 such as in the regulation of RNA
polymerase (RNAP) II,10 activation at the nucleosome level by phosphorylation at Ser10 of histone H3,11
enhancement of β-catenin dependent gene expression by phosphorylation of E2F1,6a, 10c, 10f, 12 activation of
INFγ-dependent gene expression by phosphorylation of STAT1,13 inhibition of Notch1-dependent gene
expression by phosphorylation of ICN1,14 modulation of TGF-β signaling,10c, 15 and regulation of p27
protein expression by Skp2-mediated p27 ubiquitination and degradation.16
These biological effects of CDK8 on context-related gene expression may be beneficial for
patients suffering from various types of tumors. 4, 17 Current pan-CDK inhibitors in clinical trials have
potential risks, such as narrow therapeutic windows due to their indiscriminate toxicity toward normal cells,

and complications in attempting to understand their mechanisms of action.1, 2 Therefore, a highly selective
inhibitor of CDK8 could be a promising candidate for cancer therapy. Although CDK1910c, 12b, 18 is closely
related to CDK8 in terms of structure and its association with C-type cyclins, its biological role is unclear.
Several classes of small molecules that modulate CDK8 activity have been reported in the patent
literature (Figure 1). A quinazoline derivative, Senexin B (SNX2-1–165, 1),19 showed CDK8 enzyme
inhibitory activity with an IC50 value of 24–50 nM in different assay19b, and displayed potent, selective
enzymatic inhibitory activity against CDK8 and CDK19 compared with other CDK family kinases.
Furthermore, Senexin B inhibited β-catenin activity in human colon HCT-116 cancer cells, and
demonstrated suppression of tumor growth in various types of in vivo model. Macrocyclic compound 220
showed CDK8 binding activity with an IC50 value of 31 nM in LanthaScreenTM assay, and also regulated
transcriptional activity of β-catenin with an EC50 value of 337 nM in reporter assay with human colon
HCT-116 cancer cells. Benzimidazole 321 inhibited CDK8 enzyme activity by 96% at 1 µM in an
ADP-GLoTM assay, and demonstrated good kinase selectivity against the CDK family (CDK1, 2, 5, 7, 9).
Furthermore, 3 showed cell growth inhibitory activity against various cancer cell lines, and demonstrated
tumor growth inhibition in NOD/SCID mice bearing a HCT116 tumor xenograft. Pyrazine 422 and pyridine
523, based on similar scaffolds, demonstrated enzyme inhibitory activity against CDK8 bound to cyclin C
with IC50 values of 0.4 nM and 0.2 nM, respectively, in a LANCE® TR-FRET assay. Pyrrole 624 showed
CDK8 enzyme inhibitory activity with an IC50 value of 7.0 nM in a LANCE® TR-FRET assay, and
displayed anti-proliferative activity against human gastric AGS and colorectal HCT116 cancer cells in
vitro. Tricyclic compound 725 showed enzyme inhibitory activity for CDK8 with an IC50 value of 22 nM
against human lung carcinoma A549 live cells in a mass spectrometry assay using the KiNativTM platform.
Figure 1. Chemical structures of reported CDK8 inhibitors from the patent literature.
Recently, new mechanistic insights into the potential of CDK8 and CDK19 inhibition in cancer
therapy were revealed by the discovery of high-quality small molecule probes (Type I kinase inhibitors
shown in Figure 2).26 The natural product Cortistatin A (8)27 selectively inhibited the kinase activity of the

CDK8/CDK19 module in phosphorylation of the RNAP II C-terminal domain (CTD) with a CDK8 IC50
value of 12 nM in vitro, and also showed the complete selectivity against 387 kinases in cell lysate using
KiNativ.TM Furthermore, Cortistatin A afforded efficacy in a disseminated human AML model, and 71%
tumor volume reduction in an AML xenograft model using SET-2 cells by intraperitoneal injection. The
antiproliferative and cell lineage-specific activity of Cortistatin A was caused by the enhancement of
master cell fate-determining tumor-suppressor genes regulated by superenhancers.26a Identification of
Mediator kinase substrates targeted by CDK8 and CDK19 using Cortistatin A suggested effects on
pathways implicated in inflammation, growth, and metabolic regulation.28 In another report, an orally
bioavailable small molecule CCT251545 (9)29, which was discovered through cell-based WNT pathway
screening, was found to be a CDK8 and CDK19 selective inhibitor with an IC50 value of 7.2 nM against
CDK8 in Lanthascreen TR-FRET assay.30 This compound was also found to show >100-fold selectivity
against 291 kinases in vitro.26b Further optimization to improve metabolic stability lead to the identification
of CCT251921 (10)29a, 30 as a good compromise between in vitro biochemical and pharmacokinetic
properties. CCT251921 showed strong affinity for CDK8 with an IC50 value of 2.3 nM in Lanthascreen
TR-FRET assay, and demonstrated potent inhibition of reporter-based readouts measuring basal WNT
pathway activity in human cancer cell lines LS174T (β-catenin mutant), SW480, Colo205 (APC mutant),
and PA-1 (human teratocarcinoma). On the other hand, scaffold hopping from the 3,4,5-trisubstituted
pyridine series lead to the identification of 4,6-disubstituted-isoquinoline derivative 11,31 which showed
strong binding affinity for CDK8 with an IC50 value of 5.1 nM in Lanthascreen TR-FRET assay.
Furthermore, CCT251921 and compound 11 suppressed tumor growth in vivo in an APC-mutant SW620
human colorectal carcinoma xenograft model.
Figure 2. Chemical structures of reported CDK8 inhibitors in the journal literature.

Three additional Type I CDK8 and CDK19 kinase inhibitors have also been recently reported.
Orally available 3-benzylindazole 1232 was discovered by optimization of compounds with HSP90 affinity
to yield potent and selective CDK8 inhibitors. Compound 12 showed strong binding activity for CDK8
with an IC50 value of 10 nM in Lanthascreen TR-FRET assay. MSC2530818 (13),33 which was identified
by the modification of an imidazothiazole compound, is also an orally available potent and selective CDK8
inhibitor with an IC50 value of 2.6 nM in Lanthascreen TR-FRET assay. Both compounds 12 and 13
demonstrated reduction of tumor growth rates in human SW620 colorectal carcinoma xenograft model.
Thieno[2,3-c]pyridine compound 1434 was also discovered to be a Type I CDK8 and CDK19 inhibitor, with
strong affinity for CDK8 indicated by an IC50 value of 1.5 nM in Lanthascreen TR-FRET assay. However,
CDK8 dependent antiproliferative activity of compound 14 in HCT116 colon cancer cell lines was not
observed, despite the confirmation of STAT1 phosphorylation inhibition at Ser727. The modification of the
well-known kinase inhibitor Sorafenib also led to the identification of a highly potent and selective Type II
CDK8 inhibitor 15,35 which showed high binding affinity in Lanthascreen TR-FRET assay (IC50: 17.4 nM).
However, compound 15 did not significantly suppress phosphorylation of STAT1 at Ser727.
As stated above, CDK8 inhibition is expected to be a promising approach for treating cancer,
however none of the compounds mentioned appear to have entered into clinical studies at this time.
The 4,5-dihydroimidazolo[3',4':3,4]benzo[1,2-d]isothiazole derivative 16 (Figure 3), bearing a
unique [5,6,5]-fused tricyclic scaffold, was originally identified in our laboratories as a potential
prophylactic or therapeutic drug for treating bone or articular diseases, possessing the ability to induce cell
differentiation in osteoblasts and chondrocyte precursor cells; later reports focused on the use of these
compounds to regulate muscle and fat cell differentiation.36 However, the target protein of this compound
remained unknown for several years. Recently, this compound was screened by affinity selection-mass
spectrometry (AS/MS) against a library of approximately 13,000 proteins, and found to bind specifically to
CDK19.37 High throughput screening of an in-house compound library against CDK8 confirmed that a
variety of 4,5-dihydroimidazolo[3',4':3,4]benzo[1,2-d]isothiazole derivatives, including 16, inhibit both
CDK8 and CDK19 (IC50: 18 nM and 19 nM, respectively). These results prompted us to initiate a campaign
to discover novel, potent CDK8/19 dual inhibitors for use in anticancer therapy.
A docking model (shown in Figure 3) was constructed of 16 bound to CDK8 using the program
GOLD, version 5.2, and the reported co-crystal structure of CDK8 bound to a small molecule inhibitor
(PDB ID: 4F6W).38 An amide proton and the carbonyl group of the aminocarbonyl moiety at the
4,5-dihydroimidazolo[3',4':3,4]benzo[1,2-d]isothiazole 6-position were suggested to be involved in
interaction with the backbone C=O and NH of Ala100 in the kinase hinge region of CDK8. The nitrogen
atom at the N-2 position of this scaffold was implicated as being involved in a hydrogen bond with +NH3 of
Lys52. Additionally, the methyl sulfide group at the 8-position of this scaffold was directed toward the
front pocket region, which consists of a large hydrophobic space. On the basis of these modeling studies, a
series of novel CDK8/19 dual inhibitors based on [5,6,5]-fused tricyclic scaffolds were designed (Figure 4).
An aryl group introduced into the 8-position of this scaffold was expected to occupy the hydrophobic front

pocket region to improve CDK8 potency and kinase selectivity, while allowing for the modification of
ADME properties.
We report herein the synthesis and structure–active relationships (SARs) for a series of tricyclic
4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole compounds as novel CDK8/19 dual inhibitors.39 The
optimization of this series led to the identification of 52h, which showed the antitumor efficacy in a
RPMI8226 human hematopoietic and lymphoid xenograft model in mice.
Figure 3. Modeling prediction of compound 16 binding to CDK8 kinase.
Figure 4. Design of CDK8/19 inhibitors with alternative fused heterocyclic ring A and functionalized at
the scaffold 8-position.
2. Chemistry
The 8-methylsulfanyl-4,5-dihydro-1H-thieno[3,4-g]isothiadizaole scaffold was synthesized from
commercially available 1,3-dioxolane-2,1'-cyclohexane-4-one 17 (Scheme 1). The isothiadizaole ring was
formed with SOCl2 from N-carboxamide hydrazine imine 19, which was prepared from 17 and

hydrazinecarboxamide 18. Aldol reaction of 20 with carbon disulfide and the subsequent methylation with
MeI afforded the precursor 21 of the desired tricyclic scaffold. Thiophene ring formation was successfully
achieved by the treatment of 21 with ethyl 2-mercaptacetate under basic conditions. Transformation of
ethyl ester 22 to carboxamide 24 was conducted by hydrolysis with aqueous NaOH, followed by
condensation with NH4Cl and HATU.
Scheme 1. Synthesis of 4,5-dihydro-1H-thieno[3,4-g]isothiadiazole scaffold.
aReagents and conditions: (a) NaOAc, MeOH, rt, 83%; (b) SOCl2, CH2Cl2, 0 °C, 10%; (c) CS2, K2CO3,
DMF, 0–5 °C, then MeI, 0–5 °C, 23% in 2 steps; (d) HSCH2CO2Et, K2CO3, EtOH, reflux, 50%; (e) NaOH,
water/THF/EtOH, rt, 99%; (f) NH4Cl, HATU, DMF, rt, 86%.
The
synthetic
procedure
for
the
preparation
of
the
8-methylsulfanyl-4,5-dihydro-1H-thieno[3,4-g] isothiazole scaffold is shown in Scheme 2.
Vilsmeier–Haack reaction of commercially available 25 with PBr3 and DMF gave formyl derivative 26 in
39% yields. Isothiazole ring formation of 26 with sodium thiocyanate afforded the desired tricyclic
derivative 27. Transformation of ethyl ester 27 to carboxamide 29a was conducted by hydrolysis with
aqueous NaOH, followed by condensation with aqueous ammonia via acid chloride formation with oxalyl
chloride. Dehydration of the carboxyl group of 29a with POCl3 provided nitrile derivative 30.
N-Substituted carboxamide derivatives 29b–h were synthesized by two methods (Scheme 3). The
direct condensation of carboxylic acid 28 using HATU in the presence of NMM provided 29b–d, f in
36–76% yields. Stepwise amidation via acid chloride 31, obtained from 28 by treatment with SOCl2,
afforded 29e, g, h in 13–28% overall yield in two steps.
Scheme 2. Synthesis of 4,5-dihydro-1H-thieno[3,4-g]isothiazole scaffold.

aReagents and conditions: (a) PBr3, DMF, DME, 80 °C, 39%; (b) NaSCN, DMF, 140 °C, 59%; (c) NaOH,
water/THF/EtOH, rt, quant.; (d) (COCl)2, cat. DMF, THF, then aq. NH3, 98% in 2 steps; (e) POCl3,
imidazole, pyridine, 67%.
Scheme 3. Synthesis of various 8-methylsulfanyl-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives
functionalized at the N-carboxamide moiety.
aReagents and conditions: (a) HNR1R2, NMM, HATU, DMF, rt, 36–76%; (b) SOCl2, toluene, rt, 67%: (c)
HNR1R2, LiHMDS, THF, −20 °C, 19–42%.
The synthesis of various 8-substituted-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives is
shown in Scheme 4. Sulfoxide 32 and sulfone 33 were synthesized separately by oxidation of sulfide 29a
with the appropriate number of equivalents of mCPBA. For preparation of the final
8-aryloxy-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives, SNAr reaction of sulfone 33 with phenol in
the presence of Cs2CO3 afforded desired product 34b in only 10% yield. On the other hand, it was
previously
reported
that
the
SNAr
reaction
of
bicyclic
4,5,6,7-tetrahydro-3-methylsulfonyl-4-oxobenzo[c]thiophene derivative 35 successfully provided aryl ether
derivatives.36a b In fact, nucleophilic addition was successfully employed to prepare 34b by treatment of 35
–
with phenol in the presence of K2CO3 in 88% yield. The reason for this substantial improvement in yield is
likely
due
to
the
improved
electrophilicity
of
the
bicyclic
4,5,6,7-tetrahydro-3-methylsulfonyl-4-oxobenzo[c]thiophene intermediate relative to the tricyclic
4,5-dihydro-1H-thieno[3,4-g]isothiazole intermediate. These findings led us to the facile preparation of a
wide variety of 8-aryloxy-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives 34a–i. Introduction of the
isothiazole ring into 36a–h, followed by transformation of the ethyl ester group to a carboxamide group

was accomplished using the same method shown in Scheme 2 to afford derivatives 34a–h. Oxidation of
34e with mCPBA gave N-oxypyridine analog 34i in good yield.
Scheme 4. Synthesis of various 8-substituted-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives.
aReagents and conditions: (a) mCPBA, DMF, rt, 55% for 32, 49% for 33; (b) phenol, Cs2CO3, NMP,
140 °C, 10%; (c) aryl alcohol, K2CO3, toluene/EtOAc, 80 °C, 16–95%; (d) PBr3, DMF, CH2Cl2 or CH3CN
or DMF, 80 °C; (e) ammonium thioacetate, acetone or CH3CN, 55 °C, 17–79% in 2 steps; (f) KOTMS,
THF, rt; or aq. NaOH, THF/EtOH (or MeOH), rt, 70–97%; (g) NH4OH, HOBt, WSC·HCl, DMF, rt; or
(COCl)2, DMF, THF, rt, then NH4OH, 0 °C, 17–79%; (h) mCPBA, DMF, rt, 72%.
The synthesis of various 8-(pyridin-2-yloxy)-4,5-dihydro-1H-thieno[3,4-g]isothiazole derivatives
was accomplished by Pd-catalyzed coupling as the key reaction (Scheme 5). Bromo analogs 44, 45 and 47
as key intermediates were synthesized in a similar manner to that shown in Scheme 4. A carbon
monoxide-free, Pd-catalyzed Heck aminocarbonylation40 reaction of 47 employing the Herrmann-Beller
palladacycle catalyst in conjunction with Mo(CO)6 as a solid CO source in the presence of DBU41
successfully provided a variety of N-substituted-6-carboxamide-3-oxy-pyridine derivatives 52e–f. The
Pd-catalyzed Heck carbonylation42 of 47 with Pd(OAc)2 and dppf catalyst in the presence of MeOH under
CO atmosphere gave methyl ester intermediate 52l in 64% yield. The transformation of methyl ester 52l to
primary amide 52d was conducted by hydrolysis with aqueous NaOH, followed by the condensation with
HOBt·NH3 and WSC.
Scheme 5. The synthesis of substituted 8-(pyridin-2-yloxy)-4,5-dihydro-1H-thieno[3,4-g]isothiazole
derivatives.

aReagents and conditions: (a) aryl alcohol, K2CO3, toluene/EtOAc, 80 °C, 68%; (b) PBr3, DMF, 80 °C,
83%; (c) ammonium thioacetate, acetone, 55–60 °C, 85%; (d) aqueous NaOH, THF/EtOH, rt, 55–100%;
(e) NH4OH and HOBt or HOBt·NH3, WSC·HCl, DMF, rt, 47-100%; (f) amine, Herrmann-Beller
palladacycle catalyst, Mo(CO)6, DBU, THF, 125 °C, microwave, 36-75%; (g) MeOH, Pd(OAc)2, dppf,
NEt3, DMF, 70 °C, CO2 atmosphere, 3 atm, 64%; (h) amine, (PPh3)2PdCl2, NEt3, DMA, 80 °C, CO2
atmosphere, 1 atm, 68–75%; (i) diphenylmethanimine, Pd2(dba)3, t-BuXPhos, NaOtBu, toluene, 60 °C,
47%; (j) N-methylcyclopropylamide, Pd2(dba)3, XPhos, toluene, 80 °C, 20%; (k) AcCl, DMA, rt, 33%.
Other N-substituted-6-carboxamide-3-oxy-pyridine derivatives 52a, 52h, 52i, and 53h were
synthesized using an alternative route. The Pd-catalyzed Heck aminocarbonylation of 44 or 45 with

(PPh3)2PdCl2 catalyst in the presence of Et3N under CO atmosphere successfully provided various
N-substituted-6-carboxamide-3-oxy-pyridine derivatives 48h–i, 49h. Transformation of the ethyl ester
moiety of 48h–i, 49h to a carboxamide group by an analogous method to that shown in Scheme 3 afforded
the desired compounds 52h–i and 53h. N-Substituted-6-amino-3-oxy-pyridine derivatives 52a–c were
synthesized using ethyl ester intermediate 44. The Pd-Catalyzed Buchwald–Hartwig amination42,43 of 44
and diphenylmethanimine with Pd2(dba)3 and t-BuXPhos afforded 48j in 47% yield. Hydrolysis of 48j with
aqueous NaOH achieved double hydrolysis of both the ethyl ester and the imine group to afford
6-amino-3-oxy-pyridine compound 50a. Transformation of the carboxylic acid group of 50a to a
carboxamide group in a manner similar to Scheme 3 provided 52a, which was in turn acetylated to provide
6-acetylamino-3-oxy-pyridine derivative 52c. On the other hand, conducting the analogous Pd-Catalyzed
Buchwald–Hartwig amidation43,44 of 44 with N-methylcyclopropylamide with Pd2(dba)3 and t-BuXPhos
afforded 48k in low yield (20%). Subsequent transformation of the ethyl ester moiety of 48k to a
carboxamide in a similar manner to that described above afforded 50k. The hydrolysis of
N-methy-N-cyclopropylamide group (50k) afforded 6-methylamino-3-oxy-pyridine derivative 52b.
3. Results and discussion
3.1. In vitro biological evaluation
To discover novel, potent CDK8/19 dual inhibitors, novel [5,6,5]-fused tricyclic derivatives
were evaluated in a ligand displacement assay against CDK8 and CDK19 kinases. In-cell western assay
was used to assess the phosphorylation level of a direct downstream substrate of CDK8, STAT1 (pSTAT1),
in SW480 cells.
Biological activity data for several [5,6,5]-fused tricyclic derivatives with different
heterocycles in the A-ring position are shown in Table 1. Initial lead compound 16 showed good potency
against CDK8 and CDK19, with IC50 values of 18 and 19 nM, respectively. The docking model of
compound 16 bound to CDK8 suggested that a nitrogen atom at the N-2 position of the tricyclic scaffold
forms a hydrogen bond with +NH3 of Lys52. Therefore, we examined the SAR for the introduction of other
5-membered heterocyclic rings in place of the A-ring with library compounds 16, 54, 55, and 56.
Changing placement of the methyl group from the N-1 position (16) to the N-2 position (54), and removal
of the methyl group (55) dramatically decreased binding potency. On the other hand, isoxazole compound
56 showed slightly decreased binding potency compared with imidazole 16. These results are consistent
with those of our modeling studies, suggesting that the lone pair of electrons at the N-2 nitrogen, along with
hydrophobic interactions, are important for binding activity. As examples of more hydrophobic
5-membered heterocyclic rings with an electron lone pair at the N-2 position, isoxazole isothiazole
compound 29a was designed and evaluated. Isothiazole is a more hydrophobic heterocyclic ring than either
imidazole or isoxazole. In fact, isothiazole compound 29a showed dramatically increased binding activity
against CDK8/19 with IC50 values of 0.65 and 1.3 nM, respectively. The introduction of an additional
nitrogen atom at the 4-position, resulting in thiadiazole derivative 24, resulted in somewhat diminished

potency along with increased hydrophilicities. Generally good correlation was found for this series of
compounds between the CDK8 cell free binding activity and cellular activity, measured in terms of STAT1
phosphorylation. Thus, compound 29a demonstrated high inhibitory activity in cell with an IC50 of 8.9
nM. Thus, we chose to focus on the 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole scaffold for further
efforts in the discovery of CDK8/19 dual inhibitors.
Table 1. Structure–activity relationships at the A-ring.
b
an = 2. Values in parentheses indicate 95% confidence interval. Concentration producing 50% inhibition
(EC50) values against CDK8 substrate STAT1 phosphorylation in SW480 cultured human colon cancer cell
lines.
Next, we investigated the effect of introducing R groups with the potential to interact with the
backbone C=O and NH of Ala100 in the kinase hinge region of CDK8; the results are summarized in Table

2. Initially, various alternative functional groups with hydrogen acceptor functionality were examined as
potential replacements for the carboxamide moiety (29a), such as a carboxyl group (28) or nitrile (30).
However, both modifications resulted in decreasing binding activity. Based on these results, we focused on
N-substituted carboxamide derivatives. Compounds with increasing substitution at this position, such as
N-methyl (29b), N,N-dimethyl (29c), N-phenyl (29d), showed diminished potency. Because the
introduction of R groups at this position would place them in proximity to the hinge region, according to
our CDK8 model, bulky groups might be expected to be unfavorable for binding affinity. On the other hand,
the introduction of a 2-pyridyl group (29e) as the N-substituent showed weak binding activity. According to
our modeling studies, the nitrogen atom of the 2-pyridyl ring might be able to interact with Alg365 in the
kinase hinge region. Replacement of the 2-pyridyl ring with smaller 5-membered heteroaryl rings, such
N-methylimidazole (29f) or isoxazole (29g), resulted in a dramatic recovery of enzymatic potency. On the
other hand, isothiazole derivative 29h showed decreased binding activity. Thus, our designs aimed at
increasing interaction with Alg365 in the kinase hinge region led us to discover compound 29g, possessing
an N-3-oxazolo-aminocarbonyl group as a novel hinge binding moiety, with IC50 values of 12 nM and 15
nM against CDK8 and CDK19, respectively. However, compound 29g still showed lower potency than lead
compound 29a bearing an unsubstituted carboxamide group.
Table 2. Structure–activity relationships around the carboxamide group

an = 2. Values in parentheses indicate the 95% confidence interval. bConcentration producing 50%
inhibition (EC50) values for STAT1 phosphorylation in SW480 cultured human colon cancer cells.
Because 29a showed sufficient in vitro potency to act as a CDK8/19 dual kinase inhibitor, the
ADME properties of this compound were evaluated. Compound 29a showed poor metabolic stability in

human liver microsomes, with a metabolic clearance value of 137 µL/min/mg (Table 3). To improve the
metabolic stability of this series, we focused on reducing overall molecular lipophilicity to reduce
compound recognition by metabolizing enzymes.45 Oxidized sulfide analogs such as methyl sulfoxide (32)
or methyl sulfone (33) derivatives, with lower cLogP values than methyl sulfide compound 29a, were
synthesized and tested. These compounds showed excellent metabolic stability, but also exhibited reduced
in vitro potency. Replacement of the methylthio moiety of 29a with a methoxy group (34a) resulted in an
acceptable lipohilicity profile, with a cLogP value of 1.89, and recovered in vitro potency. However, the
metabolic stability of 34a could not be measured due to analytical issues. On the other hand, the results of
our modeling studies suggested that there should be sufficient space in the hydrophobic front pocket region
around the 8-position of 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole scaffold to introduce a bulkier
functional groups than methoxy. Therefore, we designed phenoxy derivative 34b, which demonstrated
comparable enzymatic potency to compound 29a, albeit with decreased metabolic stability likely due to a
high ClogP value of 3.88. Furthermore, 34b demonstrated the most potent cellular activity among these
derivatives, with an EC50 value of 0.77 nM. These results indicated that the introduction of a lipophilic
front pocket moiety could be effectively employed to increase CDK8 inhibitory activity in cells. To further
explore this possibility, we introduced pyridyloxy groups such as 2-pyridyloxy (34c) or 3-pyridyloxy (34d)
(the corresponding 4-pyridyloxy derivative could not be obtained by the same SNAr reaction shown in
Scheme 5). Both compounds exhibited increased metabolic stability, along with a reduction in cLogP
compared with 34b. In particular, 3-pyridyloxy derivative (34d) exhibited potent enzymatic activity against
CDK8 and CDK19 with IC50 values of 1.5 and 2.7 nM, respectively, and cellular activity with an IC50 value
of 3.0 nM. Next, we explored the possibility of introducing functional groups into the pyridyl ring of
3-pyridyloxy derivative 34d; initially, we examined the introduction of a methyl group into each position.
Among the four methyl derivatives (34e–h) thus prepared, the 3-(2-methyl-pyridyloxy) (34e) and
3-(6-methyl-pyridyloxy) (34f) derivatives showed slightly increased enzymatic potency. Transformation to
N-oxy-pyridyloxy derivative 34i showed excellent metabolic stability with a significant reduction in cLogP
to 1.00, but exhibited reduced potency. Based on these results, we focused further optimization efforts for
3-pyridyloxy derivatives on substitutions at the 2- or 6-positions, with a goal of delicately tuning the
lipophilic properties of this series to obtain an appropriate balance of potency and metabolic stability.
Table 3. Structure–activity relationships at the 8-position of 4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]
isothiazole derivatives

an = 2. Values in parentheses indicate the 95% confidence interval. bConcentration producing 50%
inhibition (EC50) values for STAT1 phosphorylation in SW480 cultured human colon cancer cells.
d
cMetabolic clearance measured using human liver microsomes and NADPH. cLogP was calculated by
Daylight Software. eNot determined.

Further substituted 3-pyridyloxy derivatives (52a–i, 53h) were prepared and evaluated, as shown
in Table 4. Initially, we examined 3-pyridyloxy derivatives including mono-substitution at the pyridine
6-position, which were more easily synthesized than modifications at the 2-position. To improve metabolic
stability, replacements of methyl with more hydrophilic functional groups were examined. The
3-(6-amino-pyridyloxy) derivative 52a showed significantly improved human metabolic stability (30
µL/min/mg), while maintaining in vitro potency. However, compound 52a showed poor solubility in
aqueous solutions under neutral conditions (pH 6.8), with a solubility value of 4.5 µg/mL. The introduction
of the further functional groups into the amino moiety of 52a, such as methyl group (52b) or acetyl group
(52c), decreased metabolic stability. On the other hand, preparation of aminocarbonyl compound 52d
significantly reduced the molecular lipophilicity (cLogP: 1.88) and improved metabolic stability, while
maintaining in vitro enzymatic potency and cellular activity. However, compound 52d also showed poor
aqueous solubility. Based on these results, we examined the possibility of extending the functional group
with more flexible substituents to improve aqueous solubility while maintaining good metabolic stability.
To confirm our hypothesis, we prepared N-methylaminocarbonyl derivative 52e and
N,N-dimethylaminocarbonyl derivative 52f. Monomethyl derivative 52e showed improved aqueous
solubility with high in vitro potency and good metabolic stability; dimethyl derivative 52f demonstrated
dramatically improved aqueous solubility with good metabolic stability, but showed slightly reduced in
vitro potency. These results led us to continue studies of N-methylaminocarbonyl derivatives. Among a
variety of compounds prepared by parallel synthesis, we found that N-(2-methoxyethyl)-aminocarbonyl
derivative 52h showed excellent metabolic stability and aqueous solubility. Furthermore, 52h maintained
enzyme inhibitory activity against CDK8 and CDK19 (IC50: 0.46 nM and 0.99 nM, respectively), and
demonstrated enhanced cellular potency relative to 34f with an EC50 value of 1.3 nM. The introduction of
an additional methyl group into the pyridyl 2-position (53h) or a methoxyethyl moiety into the amide
functionality (52i) dramatically reduced metabolic stability likely due to high cLogP values, as well as
aqueous solubility. Therefore, we selected 52h as a promising compound with sufficient in vitro potency
and physicochemical properties for further in-depth evaluation.
Table 4. Structure–activity relationships of 3-pyridyloxy derivatives.

an = 2. Values in parentheses indicate the 95% confidence interval. bConcentration producing 50%
inhibition (EC50) values for STAT1 phosphorylation in SW480 cultured human colon cancer cells.
d
cMetabolic clearance measured using human liver microsomes and NADPH. cLogP was calculated by
e
Daylight Software. 10 mmol/L solution of the compound in DMSO was evaluated using JP2 buffer (pH
6.8), as described in the Japanese Pharmacopoeia. eNot determined.
3.2. Kinase inhibitory profiles of 29a and 52h.
The in vitro kinase inhibitory activity of 29a and 52h was assessed against a panel of 27 kinases.
A listing of IC50 values for those kinases that were inhibited more than 50% at 1.0 µM compound
concentration by at least one of the compounds tested is shown in Table 5. Methyl sulfide compound 29a
showed multiple kinase inhibitory activities. On the other hand, pyridyl ether compound 52h exhibited
dramatically more selective kinase activity, and enzyme inhibitory activity against only CDK family

kinases (IC50 values of 0.99 nM for CDK19, 62% inhibition at 1 µM for CDK2, 270 nM for CDK9).
Furthermore, the inhibitory activity of 52h against CDK2 was significantly less potent than that against
CDK8. The origin of such high selectivity is suggested in a general sense by the data shown Table 6, which
outlines variations in amino acid residues found across the kinase family in the DXG loop and gatekeeper
regions. CDK8 and CDK19 are somewhat rare examples, among CDK family proteins as well as the
general kinase family, of kinases that possess a Met residue in the DXG activation loop and a Phe residue
as the gatekeeper. In our comparison of kinase selectivity for 29a and 52h, the kinases that show a dramatic
reduction of inhibitory activity for 52h (GSK3β, PLK1, ASK1, CK1δ, PKA, ROCK1, PKCθ, CDC7) are all
kinases that differ from CDK8/19 in the amino acid residues found in the DXG motif and gatekeeper.
CDK2 and CDK9 do each possess a Phe gatekeeper residue, but differ from CDK8/19 in the DXG loop (X
= Phe), which could explain some residual but diminished inhibitory activity for these kinases. A docking
study of 52h bound to CDK8 suggested that the tricyclic scaffold interacts with the Phe gatekeeper, with
the pyridyl ether moiety locates near the Met residue in the DMG motif of CDK8 kinase (Figure 5). Based
on these results, we reason that the scaffold itself, combined with the introduction of the pyridyl ether
sidechain result in the high kinase specificity found for 52h with respect to the CDK8 and CDK19 kinases.
Compound 52h was further tested against a panel of 456 kinases in a panel binding assay
(KINOMEscanTM from DiscoveRx Corp.)46 to determine its selectivity (Figure 6.(a)). At a concentration
of 300 nM, the percent of unbound compound 52h was 8.3% for CDK8 and 4.6% for CDK19, with
HASPIN (23%), DYRK1B (27%), and HIP1 (32%) being the kinases with the next highest binding affinity.
Only 10 kinases showed <50% unbound fraction at 300 nM (YSK4: 41%, EPHA3: 43%, DYRK1A: 47%,
HIPK2: 48%, IRK1: 49%), and full Kd curves were measured for seven of these (Figure 6.(b)).
Table 5. Kinase selectivity of 29a and 52h.a

aNo significant inhibition observed at concentrations <1 µM for the following kinases: AKT1 (v-akt murine
thymoma viral oncogene homolog 1), Aurora B (aurora kinase A), MEK1 (mitogen-activated protein kinase
kinase 1), MAPKAPK2 (mitogen-activated protein kinase-activated protein kinase 2), CHK1 (checkpoint
kinase 1), EGFR (epidermal growth factor receptor), VEGFR2 (vascular endothelial growth factor receptor
2), cMET (met proto-oncogene), EPHA5 (ephrin type-A receptor 5 precursor), JAK1 (janus kinases 1),
FAK (focal adhesion kinase), SRC (v-src sarcoma), IRK (insulin receptor kinase), PI3Kα (phosphoinositide
3-kinase α), ERK1 (extracellular signal-regulated kinase 1), p38α (mitogen-activated protein kinase 14),
TrkA (neurotrophic tyrosine kinase, receptor, type 1). bn = 2. cGSK3β (glycogen synthase kinase 3β), PLK1
(polo-like kinase 1), ASK1 (Apoptosis signal-regulating kinase 1), CK1δ (casein kinase 1δ), PKA (protein
kinases-A), ROCK1 (rho-associated protein kinase 1), PKCθ (protein kinase Cθ), CDC7 (cell division
cycle 7).
Table 6. Analysis of amino acid residues found in DXG motif and gatekeeper for kinase family proteins.

Figure 5. Docking model of 52h bound to CDK8.

(a)
(b)
HASPIN
CDK8
DYRK1B
HIPK1
CDK19
Figure 6. (a) Selectivity profile of 52h at 300 nM against 456 kinases (evaluated by KINOMEscan™
profiling). Percent relative to control (unbound fraction) for each kinase is depicted by the size of the red
circle. (b) Kd values for kinases with most potent binding affinity shown. DYRK1B (Dual specificity
tyrosine-phosphorylation-regulated
kinase
1A),
DYRK1B
(Dual
specificity
tyrosine-phosphorylation-regulated kinase 1B), EPHA3 (ephrin type-A receptor 3), HASPIN (GSG2: germ
cell assiciated 2), HIPK1 (Homeodomain-interacting protein kinase 1), IRK1 (Interleukin-1
receptor-associated kinase 1), YSK4 (Sps1/Ste20-related kinase homolog).
3.3. In vitro pharmacology of 52h.
To evaluate the anti-proliferative activity of 52h against cancer cells, we measured inhibitory
activity against phosphorylation of STAT1, a direct substrate of CDK8, as well as cell growth inhibitory
activity in vitro (Table 7). In several types of colon, multiple myeloma, acute myelogenous leukemia
(AML), lung cancer cells, 52h suppressed phospho-STAT1 levels. 52h also demonstrated anti-proliferative
activities among these cell lines with GI50 of 0.43–2.5 nM, with the exception of the SW480 cell line. GI50
was not determined in SW480 cells because 52h did not suppress SW480 cell growth to a level below 50%.
Based on these and other results, we have found that phospho-STAT1 is an appropriate target engagement
marker, which is suppressed by 52h in all of cell lines, but does not always reflect sensitivity of cell growth
in the cell lines. A sensitivity prediction marker for cell growth inhibition is under investigation and will be
published in the near future.
Table 7. STAT1 suppression effect by 1 day treatment of 52h (Western blotting: WB) and its
anti-proliferative activity (GI50) in various cancer cell lines.

Cell line
SW480
Organ
WB: Phospho-STAT1
30 (nM)
Cell Proliferationa
0
3
GI50 (nM)
pS727-STAT1
STAT1
Colon
–
-actine
pS727-STAT1
STAT1
RPMI8226
Multiple Myeloma
2.5 (1.9–4.5)
-actine
pS727-STAT1
-actine
1.6 (1.3–1.8)
MV-4-11
Calu-1
AML
pS727-STAT1
-actine
0.43 (0.067–2.8)
Lung
aCell proliferation determined by ATP content (SW480, RPMI8226 and MV-4-11) or nuclear count
(Calu-1).
3.4. In vivo pharmacology of 52h.
Given its promising in vitro biological activity, a preliminary pharmacokinetic study of 52h was
conducted using cassette dosing in ICR mice. As shown in Table 8, compound 52h displayed good
pharmacokinetic properties, with an oral bioavailability of 98.5% and an in vivo half-life of 2.1 h. With
sufficient oral bioavailability confirmed, we then proceeded to examine the in vivo antitumor efficacy of
52h in a mouse xenograft model bearing RPMI8226 cells, which was one of the most sensitive cell lines in
vitro (Table 8).
The antitumor efficacy and body weight loss of 52h observed after oral administration of 1.25
mg/kg twice daily or 2.5 mg/kg once daily in an RPMI8226 human hematopoietic and lymphoid xenograft
model in mice is shown Figure 7 (a). Compound 52h showed tumor growth suppression with T/C of 27%,
without severe body weight loss, two weeks after administration at a dose of 1.25 mg/kg. Furthermore, we
observed tumor stasis with T/C of 1% two weeks after administration at a dose of 2.5 mg/kg for 52h,
accompanied by moderate body weight loss. We measured STAT1 phosphorylation levels in mouse blood
as a PD maker, sampling 2 h and 8 h after the last administration (day 14) (Figure 7 (b)). Compound 52h
demonstrated decreased STAT1 phosphorylation levels at each time point for both doses. Based on these

results, we concluded that the tumor suppression observed for 52h was caused by the inhibition of
CDK8/19 activity.
Most tolerable dose (MTD) of compound 52h is 2.5 mg/kg in SCID mouse model. We conducted
the efficacy study at MTD and a half of MTD to evaluate the margin between efficacy and tolerability. The
result suggests that 52h effectively suppresses tumor growth with more than two-fold safety margin in
mice.
Table 8. Pharmacokinetic parameters for 52h in ICR mice.a
aData are expressed as the mean SD (n = 3)
(a)

(b)
Figure 7. (a) Mean (n = 6) tumor volumes and body weights in nude mice bearing RPMI8226 human
hematopoietic and lymphoid xenograft tumors dosed with 52h or vehicle. P ≤ 0.05 vs control at day 14
(Dunnett's t-test). (b) Mean (n = 6) phosphorylated STAT1 levels in nude mice bearing RPMI8226 human
hematopoietic and lymphoid xenograft tumors 2 h and 8 h after the last administration (day 14) with a dose
of 1.25 mg/kg, b.i.d., po or 2.5 mg/kg, q.d., po of 52h.
4. Conclusion
We designed and developed CDK8/19 dual inhibitors bearing
4,5-dihydrothieno[3',4':3,4]benzo[1,2-d]isothiazole scaffold utilizing structure-based drug design
techniques based on docking models of our lead compound,
a
novel
4,5-dihydroimidazolo[3',4':3,4]benzo[1,2-d]isothiazole 16 bound to CDK8. To improve physicochemical
properties and kinase selectivity, we introduced a substituted 3-pyridyloxy group into the 8-position of the
scaffold in the front pocket region. The resulting optimized compound 52h showed excellent in vitro
potency, physicochemical properties, and kinase selectivity. Based on our docking model of 52h bound to
CDK8, we could explain the highly specific kinase activity profile found for this compound, based on the
interaction of the pyridyl ether sidechain of 52h interacting with Met174 of the DMG activation loop of
CDK8.
In vitro pharmacological evaluation of 52h revealed potent suppression of phosphorylated STAT1
in various cancer cells. The high oral bioavailability found for this compound enabled in vivo studies, in
which we demonstrated a mechanism-based in vivo PD effect as well as tumor growth suppression in an

RPMI8226 human hematopoietic and lymphoid xenograft model in mice. These results suggested that
inhibition of CDK8/19 may provide strong antitumor efficacy and that 52h is a promising candidate for the
treatment of various human cancers. Further pharmacological evaluation of this compound and of
CDK8/19 inhibition is in progress and will be reported in due course.
Acknowledgements
The authors thank Takashi Motoyaji and Akiyoshi Tani for target identification studies of
compound 16. We also would like to express our appreciation to Tsuneo Yasuma, Akira Mori, Hitoaki
Nishikawa, Yoshihiro Banno, Jun Fujimoto, Abhijit Nayek, and Goutam Saha for their helpful advice and
support in organic synthesis. We also thank Yukio Toyoda for his assistance in cell Western blotting assay
development, and Tsuyoshi Ishii for his useful discussion about in vitro assay development. We are also
grateful to Hiroshi Miyake, Hikichi Yuichi, and Tomoyasu Ishikawa for helpful discussion and guidance in
the overall execution of this research.
Supplementary material
Supplementary data for experimental section associated with this article can be found, in the
online version, at XXXXXX.
References and notes
1. For a representative review, see: Malumbres, M. Genome Biol. 2014, 15, 122.
2. For selected reviews, see: (a) Galons, H.; Oumata, N.; Gloulou, O.; Meijer, L. Expert Opin. Ther.
Patents 2013, 23, 945–963; (b) Blachly, J. S.; Byrd, J. C. Leuk. Lymphoma 2013, 54, 2133–2143; (c)
Mariaule, G.; Belmont, P. Molecules 2014, 19, 14366–14382; (d) Asghar, U.; Witkiewicz, A. K.; Turner, N.
C.; Knudsen, E. S. Nat. Rev. Drug Discov. 2015, 14, 130–146; (e) Sánchez-Martínez, C.; Gelbert, L. M.;
Lallena, M. J.; de Dios, A. Bioorg. Med. Chem. Lett. 2015, 25, 3420–3435; (f) Vymětalová, L.; Kryštof, V.
Med. Res. Rev. 2015, 35, 1156–1174. (g) Malínková, V.; Vylíčil, J.; Kryštof, V. Expert Opin. Ther. Patents
2015, 25, 953–970; (h) Di Giovanni, C.; Novellino, E.; Chilin, A.; Lavecchia, A.; Marzaro, G. Expert Opin.
Investig. Drugs 2016, 25, 1215–1230.
3. For selected reviews, see: (a) Romano, G.; Giordano, A. Cell Cycle 2008, 7, 3664–3668; (b) Kryštof,,
V.; Uldrijan, S. Curr. Drug. Targets 2010, 11, 291–302; (c) Węsierska-Gądek, J.; Kramer, M. P. Expert
Opin. Investig. Drugs 2011, 20, 1611–1628; (d) Fisher, R. P. Genes Cancer 2012, 3, 731–738; (e) Kryštof,
V.; Baumli, S.; Fürst, R. Curr. Pharm. Des. 2012, 18, 2883–2890; (f) Bose, P.; Simmons, G. L.; Grant, S.
Expert Opin. Investig. Drugs 2013, 22, 723–738; (g) Romano, G. ISRN Oncol. 2013, 2013, 305371; (h)
Sonawane, Y. A.; Taylor, M. A.; Napoleon, J. V.; Rana, S.; Contreras, J. I.; Natarajan, A. J. Med. Chem.
2016, 59, 8667–8684; (i) Rahaman, M. H.; Kumarasiri, M.; Mekonnen, L.; Yu, M.; Diab, S.; Albrecht, H.;
Milne, R. W.; Wang, S. Endocr. Relat. Cancer 2016, ASAP.

4. For selected reviews, see: (a) Nemet, J.; Jelicic, B.; Rubelj, I.; Sopta, M. Biochimine 2014, 97, 22–27; (b)
Rzymski, T.; Mikula, M.; Wiklik, K.; Brzózka, K. Biochim. Biophys. Acta 2015, 1854, 1617–1629; (c)
Clark, A. D.; Oldenbroek, M.; Boyer, T. G. Crit. Rev. Biochem. Mol. Biol. 2015, 50, 393–426.
5. Firestein, R.; Shima, K.; Nosho, K.; Irahara, N.; Baba, Y.; Bojarski, E.; Giovannucci, E. L.; Hahn, W. C.;
Fuchs, C. S.; Ogino, S. Int. J. Cancer 2010, 126, 2863–2873.
6. (a) Firestein, R.; Bass, A. J.; Kim, S. Y.; Dunn, I. F.; Silver, S. J.; Guney, I.; Freed, E.; Ligon, A. H.;
Vena, N.; Ogino, S.; Chheda, M. G.; Tamayo, P.; Finn, S.; Shrestha, Y.; Boehm, J. S.; Jain, S.; Bojarski,
E.; Mermel, C.; Barretina, J.; Chan, J. A.; Baselga, J.; Tabernero, J.; Root, D. E.; Fuchs, C. S.; Loda, M.;
Shivdasani, R. A.; Meyerson, M.; Hahn, W. C. Nature 2008, 455, 547–551; (b) Salari, K.; Spulak, M. E.;
Cuff, J.; Forster, A. D.; Giacomini, C. P.; Huang, S.; Ko, M. E.; Lin, A. Y.; van de Rijn, M. Pollack, J. R.
Proc. Natl. Acad. Sci. USA 2012, 109, E3196–E3205.
7. Yergiyev, O.; Garib, G.; Schoedel, K.; Palekar, A.; Bartlett, D.; Rao, U. N. Appl. Immunohistochem. Mol.
Morphol. 2016, ASAP.
8. Broude, E. V.; Győrffy, B.; Chumanevich, A. A.; Chen, M.; McDermott, M. S.; Shtutman, M.; Catroppo,
J. F.; Roninson, I. B. Curr. Cancer Drug Targets 2015, 15, 739–749.
9. For selected reviews, see: (a) Taatjes, D. J. Trens Biochem. Sci. 2010, 35, 315–322; (b) Conaway, R. C;
Conaway, J. W. Curr. Opin. Genet. Dev. 2011, 21, 225–230; (c) Conaway, R. C.; Conaway, J. W. Semin.
Cell Dev. Biol. 2011, 22, 729–734; (d) Borggrefe, T.; Yue, X. Semin. Cell Dev. Biol. 2011, 22, 759–768;
(e) Hentges, K. E. Semin. Cell Dev. Biol. 2011, 22, 769–775; (f) Napoli, C.; Sessa, M.; Infante, T.;
Casamassimi, A. Biochimie 2012, 94, 579–587; (g) Conaway, R. C.; Conaway, J. W. Biochim. Biophys.
Acta 2013, 1829, 69–75; (h) Ansari, S. A.; Morse, R. H. Cell Mol. Life Sci. 2013, 70, 2743–2756; (i)
Carlsten, J. O. P.; Zhu, X.; Gustafsson, C. M. Trends Biochem. Sci. 2013, 38, 531–537; (j) Poss, Z. C.;
Ebmeier, C. C.; Taatjes, D. J. Crit. Rev. Biochem. Mol. Biol. 2013, 48, 575–608; (k) Schiano, C.;
Casamassimi, A.; Rienzo, M.; de Nigris, F.; Sommese, L.; Napoli, C. Biochem. Biophys. Acta 2014, 1845,
66–83; (l) Yin, J.-W.; Wang, G. Development 2014, 141, 977–987; (m) Allen, B. L.; Taatjes, D. J. Nat. Rev.
Mol. Cell Biol. 2015, 16, 155–166.
10. (a) Pavri, R.; Lewis, B.; Kim, T.-K.; Dilworth, F. J.; Erdjument-Bromage, H.; Tempst, P.; de Murcia,
G.; Evans, R.; Chambon, P.; Reinberg, D. Mol. Cell 2005, 18, 83–96; (b) Knuesel, M. T.; Meyer, K. D.;
Bernecky, C.; Taatjes, D. J. Genes Dev. 2009, 23, 439–451; (c) Galbraith, M. D.; Donner, A. J.; Espinosa, J.
M. Transcription 2010, 1, 4–12; (d) Donner, A. J.; Ebmeier, C. C.; Taatjes, D. J.; Espinosa, J. M. Nat.
Struct. Mol. Biol. 2010, 17, 194–201; (e) Takahashi, H.; Parmely, T.; Sato, S.; Tomomori-Sato, C.; Banks,
C. A. S.; Kong, S. E.; Szutorisz, H.; Swanson, S. K.; Martin-Brown, S.; Washburn, M. P.; Florens, L.;
Seidel, C. W.; Lin, C.; Smith, E. R.; Shilatifard, A.; Conaway, R. C.; Conaway, J. W. Cell 2011, 146,
92–104; (f) Galbraith, M. D.; Allen, M. A.; Bensard, C. L.; Wang, X.; Schwinn, M. K.; Qin, B.; Long, H.
W.; Daniels, D. L.; Hahn, W. C.; Dowell, R. D.; Espinosa, J. M. Cell 2013, 153, 1327–1339.
11. (a) Meyer, K. D.; Donner, A. J.; Knuesel, M. T.; York, A. G.; Espinosa, J. M.; Taatjes, D. J. EMBO J.
2008, 27, 1447–1457; (b) Knuesel, M. T.; Meyer, K. D.; Donner, A. J.; Espinosa, J. M.; Taatjes, D. J. Mol.
Cell. Biol. 2009, 29, 650–661; (c) Lai, F.; Orom, U. A.; Cesaroni, M.; Beringer, M.; Taatjes, D. J.; Blobel,

G. A.; Shiekhattar, R. Nature 2013, 494, 497–501.
12. (a) Morris, E. J.; Ji, J.-Y.; Yang, F.; Stefano, L. D.; Herr, A.; Moon, N.-S.; Kwon, E.-J.; Haigis, K. M.
Näär, A. M.; Dyson, N. J. Nature 2008, 455, 552–556; (b) Tsutsui, T.; Fukasawa, R.; Tanaka, A.; Hirose,
Y.; Ohkuma, Y. Genes Cells 2011, 16, 1208–1218; (c) Porter, D. C.; Farmaki, E.; Altilia, S.; Schools, G.
P.; West, D. K.; Chen, M.; Chang, B.-D.; Puzyrev, A. T.; Lim, C.-U., Rokow-Kittell, R.; Friedhoff, L. T.;
Papavassiliou, A. G.; Kalurupalle, S.; Hurteau, G.; Shi, J.; Baran, P. S.; Gyorffy, B.; Wentland, M. P.;
Broude, E. V.; Kiraris, H.; Roninson, I. B. Proc. Natl. Acad. USA 2012, 109, 13799–13804; (d) Zhao, J.;
Ramos, R.; Demma, M. Oncogene 2013, 32, 3520–3530; (e) Cai, W. S.; Shen, F.; Feng, Z.; Chen, J.-W.;
Liu, Q.-C.; Li, E.-M.; Xu, B.; Cao, J. Biomed. Pharmacother. 2015, 74, 153–157; (f) Xu, W.; Wang, Z.;
Zhang, W.; Qian, K.; Li, H.; Kong, D.; Li, Y.; Tang, Y. Cancer Lett. 2015, 356, 613–627.
13. (a) Bancerek, J.; Poss, Z. C.; Steinparzer, I.; Sedlyarov, V.; Pfaffenwimmer, T.; Mikulic, I.; Dölken, L.;
Strobl, B.; Müller, M.; Taatjes, D. J.; Kovarik, P. Immunity 2013, 38, 250–262; (b) Putz, E. M.; Gotthardt,
D.; Hoermann, G.; Csiszar, A.; Wirth, S.; Berger, A.; Straka, E.; Rigler, D.; Wallner, B.; Jamieson, A. M.;
Pickl, W. F.; Zebedin-Brandl, E. M.; Müller, M.; Decker, T.; Sexl, V. Cell Rep. 2013, 4, 437–444; (c) Staab,
J.; Herrmann-Lingen, C.; Meyer, T. JAKSTAT 2013, 2, e24275; (d) Putz, E. M.; Gotthaedt, D.; Sexl, V.
Oncoimmunology 2014, 3, e955441.
14. (a) Fryer, C. J.; White, J. B.; Jones, K. A. Mol. Cell 2014, 16, 509–520; (b) Trakala, M.; Malumbres, M.
Nat. Cell Biol. 2014, 16, 1031–1033; (c) Li, N.; Fassl, A.; Chick, J.; Inuzuka, H.; Li, X.; Mansour, M. R.;
Liu, L.; Wang, H.; King, B.; Shaik, S.; Gutierrez, A.; Ordureau, A.; Otto, T.; Kreslavsky, T.; Baitsch, L.;
Bury, L.; Meyer, C. A.; Ke, N,; Mulry, K. A.; Kluk, M. J.; Roy, M.; Kim, S.; Zhang, X.; Geng, Y.;
Zagozdzon, A.; Jenkinson, S.; Gale, R. E.; Linch, D. C.; Zhao, J. J.; Mullighan, C. G.; Harper, J. W.; Aster,
J. C.; Aifantis, I.; von Boehmer, H.; Gygi, S. P.; Wei, W.; Look, A. T.; Sicinski, P. Nat. Cell Biol. 2014, 16,
1080–1091.
15. Alacón, C.; Zaromytidou, A. I.; Xi, Q.; Gao, S.; Yu, J.; Fujisawa, S.; Barlas, A.; Miller, A. N.;
Manova-Todorova, K.; Macias, M. J.; Sapkota, G.; Pan, D.; Massaqué, J. Cell 2009, 139, 757–769.
16. Xu, D.; Li, C.-F.; Zhang, X.; Gong, Z.; Chan, C.-H.; Lee, S.-W.; Jin, G.; Rezaeian, A.-H.; Han, F.;
Wang, J.; Yang, W.-L.; Feng, Z.-Z.; Chen, W.; Wu, C.-Y.; Wang, Y.-J.; Chow, L.-P.; Zhu, X.-F.; Zeng,
Y.-X.; Lin, H.-K. Nat. Commun. 2015, 6, 6641.
17. (a) Adler, A. S.; McCleland, M. L.; Truong, T.; Lau, S.; Modrusan, Z.; Soukup, T. M.; Roose-Girma,
M.; Blackwood, E. M.; Firestein, R. Cancer Res. 2012, 72, 2129–2139; (b) McCleland, M. L.; Soukup, T.
M.; Liu, S. D.; Esensten, J. H.; de Sousa e Melo, F.; Yaylaoglu, M.; Warming, S.; Roose-Girma, M.;
Firestein, R. J. Pathol. 2015, 237, 508–519.
18. Brägelmann, J.; Klümper, N.; Offermann, A.; von Mässenhausen, A.; Böhm, D.; Deng, M.; Queisser,
A.; Sanders, C.; Syring, I.; Merseburger, A. S.; Vogel, W.; Sievers, E.; Vlasic, I.; Carlsson, J.; Andrén, O.;
Brossart, P.; Duensing, S.; Svensson, M. A.; Shaikhibrahim, Z.; Kirfel, J.; Perner, S. Clin. Cancer. Res.
2016, ASAP.
19. (a) Roninson, I. World Patent 2013040153, March, 21, 2013; (b) Robinson, I. B.; Porter, D. C.;
Wentland, M. P. World Patent 2013116786, August, 8, 2013. (c) Roninson, I.; Chen, M. World Patent

2014071143, May, 8, 2014; (d) Broude, E.; Roninson, I. B. World Patent 2014134169, September, 4, 2014;
(e) Roninson, I. B.; Cui, T. World Patent 2016100782, June, 23, 2016.
20. Pastor Fernández, J.; Álvarez Escobar, R. M.; Varela Busto, C.; García García, A. B.; Dimitrievich
Zarich, N.; Blanco Aparicio, C.; Rodríguez López, A.; Oyarzabal Santamarina, J.; Bischoff, J. R. World
Patent 2013001310, January, 3, 2013.
21. Rzymski, T.; Zarebski, A.; Dreas, A.; Osowska, K.; Kucwaj, K.; Fogt, J.; Cholody, M.; Galezowski,
M.; Czardybon, W.; Horvath, R.; Wiklik, K.; Milik, M.; Brzózka, K. World Patent 20141072435, May, 15,
2014.
22. Cheng, Z.; Han, X.; Jiang, M.; Wang, J.; Wang, M.; Yang, S.; Zhou, C. World Patent 20141090692,
June, 19, 2014.
23. (a) Hu, H.; Jiang, M.; Jin, T.; Niu, R.; Wang, J.; Yang, S.; Yuan, T.; Zhou, C.; Wang, M.; Zhou, Z.
World Patent 20141029726, February, 27, 2014; (b) Wang, J.; Wang, M.; Yang, S.; Zhou, C. World Patent
2014106606, July, 10, 2014.
24. Han, X.; Jiang, M.; Mayweg, A. V.; Wang, L.; Yang, S. World Patent 2014154723, October, 2, 2014.
25. Romero, D. L.; Chaudhary, D.; Robinson, S.; Masse, C. E.; Morin, M. J. World Patent 2014194201,
December, 4, 2014.
26. (a) Pelish, H. E.; Liau, B. B.; Nitulescu, I. I.; Tangpeerachaikul, A.; Poss, Z. C.; Da Silva, D. H.;
Caruso, B. T.; Arefolov, A.; Fadeyi, O.; Christie, A. L.; Du, K.; Banka, D.; Schneider, E. V.; Jestel, A.;
Zou, G.; Si, C.; Ebmeier, C. C.; Bronson R. T.; Krivtsov, A. V.; Myers, A. G.; Kohl, N. E.; Kung, A. L.;
Armstrong, S. A.; Lemieux, M. E.; Taatjes, D. J.; Shair, M. D. Nature 2015, 526, 273–276; (b) Dale, T.;
Clarke, P. A.; Esdar, C.; Waalboer, D.; Adeniji-Popoola, O.; Ortiz-Ruiz, M.-J.; Mallinger, A.; Samant, R.
S.; Czodrowski, P.; Musil, D.; Schwarz, D.; Schneider, K.; Stubbs, M.; Ewan, K.; Fraser, E.; TePoele, R.;
Court, W.; Box, G.; Valenti, M.; de Haven Brandon, A.; Gowan, S.; Rohdich, F.; Raynaud, F.; Schneider,
R.; Poeschke, O.; Blaukat, A.; Workman, P.; Schiemann, K.; Eccles, S. A.; Wienke, D.; Blagg, J. Nat.
Chem. Biol. 2015, 11, 973–980; (c) Boyer, T. G. Nat. Chem. Biol. 2016, 12, 4–5.
27. (a) Aoki, S.; Watanabe, Y.; Sanagawa, M.; Setiawan, A.; Kotoku, N.; Kobayashi, M. J. Am. Chem. Soc.
2006, 128, 3148–3149; (b) Cee, V. J.; Chen, D. Y.-K.; Lee, M. R.; Nicolaou, K. C. Angew. Chem. Int. Ed.
2009, 48, 8952–8957.
28. Poss, Z. C.; Ebmeier, C. C.; Odell, A. T.; Tangpeerachaikul, A.; Lee, T.; Pelish, H. E.; Shair, M. D.;
Dowell, R. D.; Old, W. M.; Taatjes, D. J. Cell Rep. 2016, 15, 436–450.
29. (a) Schiemann, K.; Stieber, F.; Blagg, J.; Mallinger, A.; Waalboer, D.; Rink, C.; Crumpler, S. R. World
Patent 2014063778, May, 1, 2014; (b) Mallinger, A.; Crumpler, S.; Pichowicz, M.; Waalboer, D.; Stubbs,
M.; Adeniji-Popoola, O.; Wood, B.; Smith, E.; Thai, C.; Henley, A. T.; Georgi, K.; Court, W.; Hobbs, S.;
Box, G.; Ortiz-Ruiz, M.-J.; Valenti, M.; de Haven Brandon, A.; TePoele, R.; Leuthner, B.; Workman, P.;
Aherne, W.; Poeschke, O.; Dale, T.; Wienke, D.; Esdar, C.; Rohdich, F.; Raynaud, F.; Clarke, P. A.; Eccles,
S. A.; Stieber, F.; Schiemann, K.; Blagg, J. J. Med. Chem. 2015, 58, 1717–1735.
30. Mallinger, A.; Schiemann, K.; Rink, C.; Stieber, F.; Calderini, M.; Crumpler, S.; Stubbs, M.;
Adeniji-Popoola, O.; Poeschke, O.; Busch, M.; Czodrowski, P.; Musil, D.; Schwarz, D.; Ortiz-Ruiz, M.-J.;