hits for the lysine methyltransferase SET7 (also named KMT7, SETD7, SET9), has
been recently described by Meng et al. [119]. Using the crystal structure of SET7 in
complex with SAM, a pharmacophore model was generated and used to screen a
filtered SPECS library (182,014 compounds). The pharmacophore hits were subsequently docked into the SAM cofactor binding site using Lys294 as a hydrogen bond
constraint. The top-ranked 649 molecules were clustered, and at least one molecule
for each cluster was picked to cover a large chemical space. Program Glide was used
as docking program due to its better performance in preliminary studies. Indeed, the
authors investigated the performance of diverse docking programs with respect to
enrichment factors. Moreover, the ability of the selected procedure to reproduce the
cocrystallized binding mode of SAM was evaluated by re-docking studies. Using
such a protocol, seven compounds, among the 127 tested, were discovered to inhibit
SET7 activity. Compound DC-S100 (IC 50 ¼ 30.04 μM) was chosen as a query for
the similarity search and structure-guided optimization studies. This approach led to
the identification of compound DC-S239 (29, Fig. 6) with an IC 50 of 4.59 μM and
selectivity over DNMT1, DOT1L, EZH2, NSD1, SETD8, and G9a.
5 Histone Demethylases
In 2004, lysine-specific demethylase-1 (LSD1, also named KDM1A) was discovered
as the first demethylase which can specifically remove methyl groups from histone
H3K4 (lysine 4 of histone 3) utilizing flavin adenine dinucleotide (FAD) as a cofactor
[120]. Later on, JumonjiC (JmjC) domain-containing proteins were identified as
another superfamily of histone demethylases which use Fe
2+ and α-ketoglutarate
(α-KG) as cofactors in an oxygenation reaction to remove the methyl groups from
lysine residues [121]. Numerous histone lysine demethylases were found to be
overexpressed in primary tumors, which render them as highly promising biological
targets [122–124]. Over the past years, huge efforts have been made to explore the
structural requirements for modulating these targets. This culminated in the discovery
of numerous small molecule inhibitors of diverse histone demethylases; one of them,
namely, the irreversible LSD1 inhibitor GSK2879552, is currently undergoing clinical trials for the treatment of acute myeloid leukemia (AML), myelodysplastic
syndromes (MDS), and small cell lung carcinoma.
Plenty of crystal structures of LSD1 and various JmjC domains have been solved
in the last decade which helped to understand the binding of substrates and some
available inhibitors.
5.1 LSD1 (KDM1A)
LSD1 is composed of three domains: the tower domain (aa 419-520), the SWIRM
domain (aa 166-260), and the amine oxidase domain (AOD, aa 520-852) [125, 126];
Structure-Based Design of Epigenetic Inhibitors
469
been recently described by Meng et al. [119]. Using the crystal structure of SET7 in
complex with SAM, a pharmacophore model was generated and used to screen a
filtered SPECS library (182,014 compounds). The pharmacophore hits were subsequently docked into the SAM cofactor binding site using Lys294 as a hydrogen bond
constraint. The top-ranked 649 molecules were clustered, and at least one molecule
for each cluster was picked to cover a large chemical space. Program Glide was used
as docking program due to its better performance in preliminary studies. Indeed, the
authors investigated the performance of diverse docking programs with respect to
enrichment factors. Moreover, the ability of the selected procedure to reproduce the
cocrystallized binding mode of SAM was evaluated by re-docking studies. Using
such a protocol, seven compounds, among the 127 tested, were discovered to inhibit
SET7 activity. Compound DC-S100 (IC 50 ¼ 30.04 μM) was chosen as a query for
the similarity search and structure-guided optimization studies. This approach led to
the identification of compound DC-S239 (29, Fig. 6) with an IC 50 of 4.59 μM and
selectivity over DNMT1, DOT1L, EZH2, NSD1, SETD8, and G9a.
5 Histone Demethylases
In 2004, lysine-specific demethylase-1 (LSD1, also named KDM1A) was discovered
as the first demethylase which can specifically remove methyl groups from histone
H3K4 (lysine 4 of histone 3) utilizing flavin adenine dinucleotide (FAD) as a cofactor
[120]. Later on, JumonjiC (JmjC) domain-containing proteins were identified as
another superfamily of histone demethylases which use Fe
2+ and α-ketoglutarate
(α-KG) as cofactors in an oxygenation reaction to remove the methyl groups from
lysine residues [121]. Numerous histone lysine demethylases were found to be
overexpressed in primary tumors, which render them as highly promising biological
targets [122–124]. Over the past years, huge efforts have been made to explore the
structural requirements for modulating these targets. This culminated in the discovery
of numerous small molecule inhibitors of diverse histone demethylases; one of them,
namely, the irreversible LSD1 inhibitor GSK2879552, is currently undergoing clinical trials for the treatment of acute myeloid leukemia (AML), myelodysplastic
syndromes (MDS), and small cell lung carcinoma.
Plenty of crystal structures of LSD1 and various JmjC domains have been solved
in the last decade which helped to understand the binding of substrates and some
available inhibitors.
5.1 LSD1 (KDM1A)
LSD1 is composed of three domains: the tower domain (aa 419-520), the SWIRM
domain (aa 166-260), and the amine oxidase domain (AOD, aa 520-852) [125, 126];
Structure-Based Design of Epigenetic Inhibitors
469