Identification of potent, selective KDM5 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.07.026

This communication describes the identification and optimization of a series of pan-KDM5 inhibitors derived from compound 1, a hit initially identified against KDM4C. Compound 1 was optimized to afford compound 20, a 10?nM inhibitor of KDM5A. Compound 20

Full text of DOI:10.1016/j.bmcl.2016.07.026

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of potent, selective KDM5 inhibitors
a,
⇑
a
a
a
a
Victor S. Gehling , Steven F. Bellon , Jean-Christophe Harmange , Yves LeBlanc , Florence Poy ,
a
a
a
a
a
a
Shobu Odate , Shane Buker , Fei Lan , Shilpi Arora , Kaylyn E. Williamson , Peter Sandy ,
Richard T. Cummings , Christopher M. Bailey , Louise Bergeron , Weifeng Mao , Amy Gustafson ,
Yichin Liu , Erica VanderPorten , James E. Audia , Patrick Trojer , Brian K. Albrecht
a
a
a
c
b
b
b
a
a
a
a
Constellation Pharmaceuticals, Inc., 215 First Street, Suite 200, Cambridge, MA 02142, United States
Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080, United States
WuXi Apptec, 288 Fute Zhong Road, Waigaoqiao Free Trade Zone, Shanghai 200131, China
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
This communication describes the identification and optimization of a series of pan-KDM5 inhibitors
derived from compound 1, a hit initially identified against KDM4C. Compound 1 was optimized to afford
compound 20, a 10 nM inhibitor of KDM5A. Compound 20 is highly selective for the KDM5 enzymes ver-
sus other histone lysine demethylases and demonstrates activity in a cellular assay measuring the
increase in global histone 3 lysine 4 tri-methylation (H3K4me3). In addition compound 20 has good
ADME properties, excellent mouse PK, and is a suitable starting point for further optimization.
Ó 2016 Published by Elsevier Ltd.
Received 14 May 2016
Revised 8 July 2016
Accepted 9 July 2016
Available online xxxx
Keywords:
Histone lysine demethylase
KDM5
JARID1
Epigenetics
Histone methylation plays an important role in the mainte-
nance of chromatin structure and the regulation of transcription.
by sustaining a slow cycling subpopulation of cells that is required
for the proliferation of the main cancer cell population. In the
1
16
Aberrant changes in histone methylation profiles contribute to
cancer development and progression.2 Dynamic regulation of his-
tone lysine methylation states is maintained by the interplay of
two enzyme families, histone lysine methyltransferases (KMTs)
presence of various anti-cancer agents, including BRAF inhibitors
and cisplatin, this KDM5B-expressing melanoma subpopulation
,3
1
7
gives rise to drug-resistant melanoma cells. Collectively, multiple
KDM5 members are implicated in tumor initiation, maintenance,
drug tolerance, and suggest that KDM5 inhibitors may have utility
4
and demethylases (KDMs). KMTs and KDMs are highly selective
3
,7
for a given histone lysine residue and the number of methyl groups
that are added or removed by a given enzyme. The KDM5 (JARID1)
proteins (KDM5A, KDM5B, KDM5C, KDM5D) demethylate
tri-methylated lysine 4 on histone 3 (H3K4me3), a modification
associated specifically with promoters of actively transcribing
and transcriptionally poised genes.5 Tumor-promoting and
tumor-suppressive roles for the KDM5 family members have
as cancer therapeutics.
Several KDM inhibitors have been reported towards a variety of
1
8
targets and this area has recently been reviewed. Many of the
literature inhibitors are analogs of the 2-oxoglutarate co-factor
(2-OG), such as N-oxalyl glycine (NOG), 2,4-pyridine dicarboxylic
acid (2,4-PDA), or contain strong metal chelators, such as
2,2-bipyridines or hydroxyquinolines. While these compounds
can be potent biochemical inhibitors, in general the literature
compounds lack the selectivity profiles necessary to be useful tool
compounds and have chemical properties that limit cell permeabil-
ity. Two reports from GSK have described inhibitors with improved
properties and cell-permeability, but these compounds still have
,6
7
,8
been described. KDM5A (JARID1A, RBP2) and KDM5B (JARID1B,
PLU-1) show increased gene expression in a number of human
9
–14
cancers,
suggesting that these enzymes may be required for
cancer cell survival.
Interestingly, cancer cell subpopulations require KDM5A to
acquire reversible tolerance to agents targeting EGF signaling and
to cytotoxic agents such as cisplatin.15 Similarly, KDM5B is impli-
cated in melanoma tumor maintenance and metastatic progression
1
9,20
significant activity against related KDM’s.
The combination of
poor selectivity profiles and limited cell permeability complicates
further target validation and drug discovery efforts with the
reported KDM5 inhibitors. At the onset of this program our goal
was to identify a potent, selective KDM5 inhibitor that could serve
as a useful tool to interrogate the contribution of the catalytic
activity of KDM5 enzymes in disease biology.
⇑
Corresponding author. Tel.: +1 617 714 0540; fax: +1 617 577 0472.
E-mail address: victor.gehling@constellationpharma.com (V.S. Gehling).
http://dx.doi.org/10.1016/j.bmcl.2016.07.026
960-894X/Ó 2016 Published by Elsevier Ltd.
0
Please cite this article in press as: Gehling, V. S.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.07.026

2
V. S. Gehling et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Our initial chemical matter originated from a high throughput
Table 1
Initial SAR around 1
2
1
screen against KDM4C. Hits from the KDM4C screen were tested
against a panel of KDMs representative of the various KDM sub-
families and several compounds were found to inhibit KDM5A
with sub-micromolar potency. This effort identified compound 1
as a potent inhibitor of KDM5A with approximately 10-fold selec-
tivity against KDM4C (Fig. 1). Compound 1 serves as an attractive
starting point for further optimization as it does not contain an
obvious metal chelator or the carboxylic acid functional group
common to literature KDM inhibitors.
Attempts to obtain a crystal structure of 1 bound to KDM5A
were initially unsuccessful. Due to the difficulty of obtaining co-
crystal structures in KDM5A, we used KDM4C as a surrogate for
KDM5A. A co-crystal structure of compound 1 bound to KDM4C
was solved and this structure was used to determine the mode of
inhibition, identify important key interactions, and guide further
compound design (Fig. 1).
O
R2
N
N
R1
N
H
N
Compound
R1
R2
KDM5A IC50a
(
l
M) Standard deviation LipE23
1
2
3
4
5
6
7
Me
NH
H
H
H
Et
Et
Et
0.237b
1.18b
0.020
0.649
0.024
0.065
0.351
0.009
0.405
0.033
0.013
0.111
6.0
6.3
7.5
7.0
7.0
6.6
4.5
2
H
iPr
nPr
iBu
b
H
H
0.055
2.24b
a
IC50 value reported as an average of P3 determinations with standard deviation
reported (SD).
The crystal structure of 1 bound to KDM4C shows the inhibitor
22
b
binding in the 2-OG binding site. Compound 1 chelates the cat-
alytic iron via the nitrile functional group while the pyrazolo-
pyrimidine core participates in a well-ordered hydrogen bonding
network from lysine 208 and asparagine 282 to both the pyrazole
ring and the carbonyl of the inhibitor. One interesting feature of
this structure is the movement of tyrosine 134, as this residue
swings out of the way to accommodate inhibitor binding.
IC50 values reported as an average of 2 determinations with standard deviation
reported.
2
the R group to an isobutyl afforded 7, but this modification led to a
significant potency loss. The SAR demonstrates that KDM5 has a
well-defined binding pocket formed by tryptophan 210 and tyro-
sine 134 and that these residues limit the size of substituents at
Using the crystal structure as a rough guide we began assem-
bling our structure activity relationships (SAR) by modifying the
2
R to small alkyl groups.
2
-methyl group on the pyrazole ring. We replaced the 2-methyl
After defining the size of the binding pocket, we next investi-
gated changes to the core of the inhibitor (Table 2). We began by
group with an amino-group to afford 2 in an attempt to pick up
a favorable interaction with tryptophan 210. This modification
was not productive and resulted in a 5-fold loss in potency sug-
gesting that this area of the binding pocket is sensitive to perturba-
tions in the inhibitor structure. Next, we removed the 2-methyl
group of 1 to afford 3. This change resulted in a 12-fold increase
in potency relative to 1. We attribute the increased potency of 3
to removal of unfavorable steric interactions with tryptophan
2
3
cyclizing the R and R groups as in 8. This modification was not
well tolerated, resulting in a compound with weak inhibitory activ-
ity against KDM5A. The weak activity of this cyclized compound
demonstrates the importance of interacting with tyrosine 134 for
potent KDM5A inhibition. We then replaced the pyrazole ring of
8 with an imidazole to afford 9. This change resulted in a nearly
inactive compound against KDM5A and emphasizes the impor-
tance of the hydrogen bonding network between inhibitor and pro-
tein in affording good KDM5A inhibition (Fig. 1).
Alkylation of the 4-nitrogen was then investigated as these
modifications extend into the large substrate binding pocket and
would allow us to probe for additional interactions. Alkylation
with a methyl group to form 10 resulted in a 10-fold loss in
potency. Further potency losses occurred when the alkyl group
was extended, as in compound 11. From the N-alkyl analogs, we
moved to triazine 12 and methyl ether 13. Surprisingly, compound
12 was completely inactive against KDM5A, despite the triazine
compound containing all of the key recognition elements for
binding.
2
10 which defines one edge of the binding pocket.
Since the co-crystal structure of 1 bound to KDM4C showed
2
movement of tyrosine 134 we investigated modifications of R in
an effort to optimize this interaction. Removal of the ethyl group
afforded 4, a compound that was 15-fold less potent than 1
(Table 1). The reduced potency of compound 4 suggests that the
induced fit of compound 1 with tyr134 contributes significantly
to the potency of these compounds. In an effort to keep the positive
interaction with tyr134 we then examined conservative modifica-
tions of the ethyl group, replacing it with an iPr group, as in 5, and
an nPr group, as in 6. While these modifications were tolerated,
they did not afford any improvements in potency over 1. Extending
One potential explanation for the absence of activity observed
with triazene 12 is that binding to KDM5A requires an acidic
hydrogen atom. Whether this acidic hydrogen atom is directly
involved with binding to KDM5A or affords access to an active tau-
tomer of the inhibitor is not obvious. What is consistent, from tri-
azene 12 and N-alkyl analogs 10 and 11, is that the absence of an
acidic hydrogen atoms is detrimental to KDM5 inhibition.
Exploring our core SAR further, we synthesized methyl ether 13.
Consistent with earlier discussion of the importance of an acidic
hydrogen atom, this modification resulted in a ꢀ16 fold loss in
activity compared to 3. As none of our core modifications afforded
an advantage relative to the starting point, we focused our atten-
tion on the optimization of the right hand side of our inhibitors.
O
N
N
N
H
N
1
KDM4C: 2.88 µM
KDM5A: 0.237 µM
3
A variety of functional groups were investigated at R , including
aryl groups, heterocycles, and amides (Table 3). In general aromatic
3
and heteroaromatic groups were well tolerated at R leading to
potent KDM5A inhibitors. The only exceptions are the 4-pyridine
Figure 1. Initial KDM5A hit (1) bound to KDM4C.
(18) and the m-tolyl (16) compounds, both of which demonstrate
Please cite this article in press as: Gehling, V. S.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.07.026

V. S. Gehling et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
Table 2
Table 3
Pyrazolo-pyrimidine core modifications
Optimization of the pyrazolopyrimidine core
Compound
Structure
KDM5A IC50a
M)
Standard
deviation
LipE
5.8
O
(l
R2
N
N
O
N
R3
N
N
H
8
0.878
22.5
0.397
1.78
N
H
N
N
R2
R3
Compound
KDM5A IC50
a
Standard
deviation
LipE
O
N
(lM)
3
Et
Et
Me
0.020b
0.009
7.5
7.0
N
N
9
3.9
6.5
H
N
N
H
0.712b
14
0.359
O
O
N
15
16
Et
Et
0.015
0.206
0.013b
0.011
0.144
6.3
5.0
N
N
1
0
1
0.214
0.098
N
N
1
7
8
Et
Et
0.006
2.27
7.9
5.0
O
N
N
8.94b
1
N
1
3.09
0.447
5.2
N
1
9
0
Et
0.011b
0.010
0.002
0.001
9.1
6.0
N
N
H
O
N
O
2
iPr
N
N
N
N
1
2
3
>25
—
—
a
IC50 values reported as an average of P3 determinations with standard devi-
ation reported (SD).
IC50 values reported as an average of 2 determinations with standard deviated
reported.
N
O
N
b
N
N
N
Interestingly, there was a dose-dependent increase in the KDM5B
and KDM5C protein levels upon treatment with active compound
0 that was not observed with inactive control 16. The increased
protein levels are not explained by increases in the transcription
levels of these proteins and suggests that inhibitor binding stabi-
lizes KDM5B and KDM5C (Fig. 2b).
1
0.324
0.342
4.7
2
a
IC50 values reported as an average of P3 determinations with standard devi-
ation reported (SD).
The poor translation of compound 20 into the cellular assay
may be attributed to a variety of causes, including poor permeabil-
ity, high-intracellular 2-OG concentrations, and/or poor physical
properties. In an effort to understand the poor translation of com-
pound 20 into our cellular assay we measured it’s permeability in
MDCK cells and compound 20 demonstrates good permeability
poor activity against KDM5A. The ethyl amide compound (14) was
also not well tolerated resulting in significant potency losses. In
contrast to the amide and 4-pyridine examples, the 4-pyridone
compound (19) was among the most potent compounds synthe-
sized with a KDM5A IC50 of 11 nM. Finally, combining an isopropyl
substituent at R with a phenyl ring at R afforded 20, a highly
potent KDM5A inhibitor with a biochemical IC50 of 10 nM.
Compound 20 was profiled in our selectivity panel of KDMs and
while it demonstrated nearly equivalent inhibition of the different
members of KDM5 family, it displays a remarkable selectivity pro-
file against the other KDMs tested. In particular, 20 is greater than
À6
À6
(Papp (A to B) = 17.2 Â 10 cm/s, Papp (B to A) = 12.1 Â 10 cm/s)
2
3
25
with little efflux observed. Next we generated KDM5A IC50 data
on compound 20 at two 2-OG concentrations; one at approxi-
2
6
mately the Km of 2-OG and another with 2-OG concentrations
10-fold above Km (Fig. 3). In these experiments compound 20 dis-
plays a modestly shifted IC50 of 20 nM in the high 2-OG conditions.
Intracellular 2-OG concentrations are reported to be in the high lM
2
4
2
00 fold selective over KDM4C and has >500 fold selectivity against
range but do not fully explain the >500 fold shift between
biochemical and cellular activity observed with compound 20.
Further efforts directed at improving physical properties of the
series may afford inhibitors with improved biochemical to cell
translation. As compound 20 is a potent, selective, and cell-
permeable KDM5 inhibitor it was advanced into our ADME panel
and profiled in a mouse PK experiment (Fig. 3).
the other KDMs (Fig. 3). As compound 20 is a potent and selective
inhibitor of the KDM5 enzymes, it was profiled in our cellular
assays alongside compound 16 which served as an inactive control.
Our KDM5A cellular assay measures the increase of global
H3K4me3 levels in PC9 cells upon compound treatment. In this
assay compound 20 demonstrated a significant effect on the global
21
level of H3K4me3 with a cellular EC50 of 5.2
l
M. In the same
The synthesis of compound 20 entails a straightforward two
step sequence (Fig. 4). The first step is a Claisen condensation of
esters 21 and 22 to afford beta-keto ester 23. Condensation of 23
assay inactive control 16 had no effect on the H3K4me3 mark sug-
gesting that the mark change is due to selective inhibition of the
KDM5 family. In an alternative cellular assay with U2OS cells, com-
pound 20 demonstrates a modest increase in H3K4me3 (Fig. 2a, 1.6
2
7
with the amino-pyrazole 24 affords compound 20 in two steps.
Compound 20 demonstrates low microsomal metabolism and is
highly protein bound across species. This profile translates very
fold increase at 30 lM), while inactive control 16 shows no effect.
Please cite this article in press as: Gehling, V. S.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.07.026

4
V. S. Gehling et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Figure 2. (a) Western blot demonstrating the effect of compound on H3K4me3 levels and KDM5protein levels in U2OS cells. (b) Transcript levels of the KDM5 family proteins
upon compound treatment in U2OS cells.
Figure 3. In vitro and in vivo profile of 20.21
H
N
O
O
O
O
N
N
O
a
NH2
b
N
O
O
O
N
H
N
24
N
2
1
22
23
20
Figure 4. Synthesis of 20. Reagents: (a) NaH, DME (51%); (b) microwave, 160 °C (2%).
well into mouse PK experiments where 20 demonstrates low clear-
ance, good half-life, low volume, excellent exposure, and high
bioavailability.
3. Højfeldt, J. W.; Agger, K.; Helin, K. Nat. Rev. Drug Disc. 2013, 12, 917.
4
5
.
.
McGrath, J.; Trojer, P. Pharmacol. Ther. 2015, 1.
Christensen, J.; Agger, K.; Cloos, P. A.; Pasini, D.; Rose, S.; Sennels, L.; Rappsilber,
J.; Hansen, K. H.; Salcini, A. E.; Helin, K. Cell 2007, 128, 1063.
In this Letter we describe the identification of a KDM4C hit and
optimization against KDM5A. This resulted in the identification of
6. Benevolenskaya, E. V. Biochem. Cell Biol. 2007, 85, 435.
7.
8.
9.
Rasmussen, P. B.; Staller, P. Epigenomics 2014, 6, 277.
Blair, L. P.; Cao, J.; Zou, M. R.; Sayegh, J.; Yan, Q. Cancers 2011, 3, 1383.
Li, L.; Wang, L.; Song, P.; Geng, X.; Liang, X.; Zhou, M.; Wang, Y.; Chen, C.; Jia, J.;
Zeng, J. Mol. Cancer 2014, 13, 81.
2
0, a compound with low nanomolar biochemical potency and
micromolar cellular potency in an assay monitoring changes of glo-
1
0. Teng, Y.-C.; Lee, C.-F.; Li, Y.-S.; Chen, Y.-R.; Hsiao, P.-W.; Chan, M.-Y.; Lin, F.-M.;
Huang, H.-D.; Chen, Y.-T.; Jeng, Y.-M.; Hsu, C.-H.; Yan, Q.; Tsai, M.-D.; Juan, L.-J.
Cancer Res. 2013, 73, 4711.
bal H3K4me3 levels. Compound 20 has good ADME properties,
2
8
excellent mouse PK, and is highly selective for the KDM5 family.
Given the unprecedented degree of selectivity compound 20 serves
as a useful tool to study the role of KDM5 enzymes in various cel-
lular contexts. The promising ADME and PK properties suggest that
compound 20 provides a useful starting point for further optimiza-
tion and studies directed towards the optimization of this scaffold
will be reported in the future.
11. Vieira, F. Q.; Costa-Pinheiro, P.; Ramalho-Carvalho, J.; Pereira, A.; Menezes, F.
D.; Antunes, L.; Carneiro, I.; Oliveira, J.; Henrique, R.; Jerónimo, C. Endocr. Relat.
Cancer 2014, 21, 51.
1
1
1
2. Hou, J.; Wu, J.; Dombkowski, A.; Zhang, K.; Holowatyj, A.; Boerner, J. L.; Yang,
Z.-Q. Am. J. Transl. Res. 2012, 4, 247.
3. Xiang, Y.; Zhu, Z.; Han, G.; Ye, X.; Xu, B.; Peng, Z.; Ma, Y.; Yu, Y.; Lin, H.; Chen, A.
P.; Chen, C. D. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 19226.
4. Hayami, S.; Yoshimatsu, M.; Veerakumarasivam, A.; Unoki, M.; Iwai, Y.;
Tsunoda, T.; Field, H. I.; Kelly, J. D.; Neal, D. E.; Yamaue, H.; Ponder, B. A. J.;
Nakamura, Y.; Hamamoto, R. Mol. Cancer 2010, 9, 59.
References and notes
1
5. Sharma, S. V.; Lee, D. Y.; Li, B.; Quinlan, M. P.; Takahashi, F.; Maheswaran, S.;
McDermott, U.; Azizian, N.; Zou, L.; Fischbach, M. A.; Wong, K.-K.; Brandstetter,
K.; Wittner, B.; Ramaswamy, S.; Classon, M.; Settleman, J. A. Cell 2010,
1
2
.
.
Dawson, M. A.; Kouzarides, T. Cell 2012, 150, 12.
Wee, S.; Dhanak, D.; Li, H.; Armstrong, S. A.; Copeland, R. A.; Sims, R.; Baylin, S.
B.; Liu, X. S.; Schweizer, L. Ann. N.Y. Acad. Sci. 2014, 30.
1
41, 69.
Please cite this article in press as: Gehling, V. S.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.07.026

V. S. Gehling et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
5
1
1
6. Roesch, A.; Fukunaga-Kalabis, M.; Schmidt, E. C.; Zabierowski, S. E.; Brafford, P.
A.; Vultur, A.; Basu, D.; Gimotty, P.; Vogt, T.; Herlyn, M. Cell 2010, 141, 583.
7. Roesch, A.; Vultur, A.; Bogeski, I.; Wang, H.; Zimmerman, K. M.; Speicher, D.;
Korbel, C.; Laschke, M. W.; Gimotty, P. A.; Philipp, S. E.; Krause, E.; Patzold, S.;
Villanueva, J.; Krepler, C.; Fukunaga-Kalabis, M.; Hoth, M.; Bastian, B. C.; Vogt,
T.; Herlyn, M. Cancer Cell 2013, 23, 811.
D.; Gangurde, P.; Manieri, W.; Busby, J.; Flynn, E. M.; Lan, F.; Kim, H.; Odate, S.;
Cochran, A. G.; Liu, Y.; Wongchenko, W.; Yang, Y.; Cheung, T. K.; Maile, T. M.;
Lau, T.; Costa, M.; Hegde, G. V.; Jackson, E.; Pitti, R.; Arnott, D.; Bailey, C.; Bellon,
S.; Cummings, R. T.; Albrecht, B. K.; Harmange, J. C.; Kiefer, J. R.; Trojer, P. T.;
Classon, M. Nat. Chem. Biol. 2016, 12, 531.
22. The X-ray coordinates for compound 1 bound to KDM4C are available from the
PDB, with ID 5KR7.
23. LipE = pIC50 – ClogP. ClogP values were calculated in Chemdraw 15.0.
24. Thirstrup, K.; Christensen, S.; Moller, H. A.; Ritzen, A.; Bergstrom, A. L.; Sager, T.
N.; Jensen, H. S. Pharmacol. Res. 2011, 64, 268.
1
8. McAllister, T. E.; England, K. S.; Hopkinson, R. J.; Brennan, P. E.; Kawamura, A.;
Schofield, C. J. J. Med. Chem. 2016, 59, 1308.
1
9. Westaway, S. M.; Preston, A. G. S.; Barker, M. D.; Brown, F.; Brown, J. A.;
Campbell, M.; Chung, C.; Diallo, H.; Douault, C.; Drewes, G.; Eagle, R.; Gordon,
L.; Haslam, C.; Hayhow, T. G.; Humphreys, P. G.; Joberty, G.; Katso, R.;
Kruidenier, L.; Leveridge, M.; Liddle, J.; Mosley, J.; Muelbaier, M.; Randle, R.;
Rioja, I.; Rueger, A.; Seal, G. A.; Sheppard, R. J.; Singh, O.; Taylor, J.; Thomas, P.;
Thomson, D.; Wilson, D. M.; Lee, K.; Prinjha, R. K. J. Med. Chem. 2016, 59, 1357.
0. Westaway, S. M.; Preston, A. G. S.; Barker, M. D.; Brown, F.; Brown, J. A.;
Campbell, M.; Chung, C.; Drewes, G.; Eagle, R.; Garton, N.; Gordon, L.; Haslam,
C.; Hayhow, T. G.; Humphreys, P. G.; Joberty, G.; Katso, R.; Kruidenier, L.;
Leveridge, M.; Pemberton, M.; Rioja, I.; Seal, G. A.; Shipley, T.; Singh, O.;
Suckling, C. J.; Taylor, J.; Thomas, P.; Wilson, D. M.; Lee, K.; Prinjha, R. K. J. Med.
Chem. 2016, 59, 1370.
25. Compound 20 also showed good permeability in MDCK cells expressing MDRI
À6
with some efflux observed. MDCK-MDRI Papp (A to B) 8.4 Â 10 cm/s and
À6
Papp (B to A) 27.6 Â 10 cm/s.
26. 2-OG Km for KDM5A was measured at ꢀ1–2
lM.
2
2
27. 3-Aminopyrazole-4-carbonitrile was purchased from Sigma–Aldrich and used
without further purification.
28. For further selectivity profiling against a variety of 2-OG utilizing enzymes,
see: Joberty, G.; Boesche, M.; Brown, J. A.; Eberhard, D.; Garton, N. S.;
Humphreys, P. G.; Mathieson, T.; Muelbauer, M.; Ramsden, N. G.; Reader, V.;
Rueger, A.; Sheppard, R. J.; Westaway, S. M.; Bantscheff, M.; Lee, K.; Wilson, D.
M.; Prinjha, R. K.; Drewes, G. ACS Chem. Biol. 2016. http://dx.doi.org/10.1021/
acschembio.6b00080.
1. For screening details, biochemical assay descriptions, cell-assay descriptions,
and further characterization of compound 20, see: Vinogradova, M.; Gehling, V.
S.; Gustafson, A.; Arora, S.; Tindell, C. A.; Wilson, C.; Williamson, K. E.; Guler, G.
Please cite this article in press as: Gehling, V. S.; et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.07.026