7 PRMT5
7.1 Background
Initially identified as a Janus kinase-binding protein [142], PRMT5 is the first
identified and most abundant member of the type II PRMTs. The type II PRMTs
preferentially produce symmetrically dimethylated arginine (sDMA) [13]. PRMT5
recognizes both the GAR and PGM motifs in a wide variety of substrates, including
histone arginines H2AR3, H4R3, H3R2 and H3R8 (in vivo) [13, 143, 144]. Nonhistone substrates include ribosomal proteins (RPS10) [145], nuclear factor NF-κB,
tumour suppressor protein p53, transcription factor E2F-1 [143, 146, 147], tumour
suppressor PDCD4 (programmed cell death protein 4) [148] and the MAPK/ERK
pathway (mitogen-activated protein kinase/extracellular signal-regulated kinase)
[149], among many others [56]. PRMT5 has multiple associations with binding
partners of which MEP50 (methylosome protein 50) is known to be necessary for
regulating its specificity in methylating H2A and H4 [150, 151]. Other binding
partners regulate the activity and substrate specificity of PRMT5 [152].
PRMT5 is upregulated in wide variety of human cancers, including breast [148],
colorectal [153], lung [154, 155] and epithelial ovarian cancer [156], lymphomas
[157–159] and melanoma [160]. In addition, recent studies have shown PRMT5 to
be a unique anticancer target [47–49]. These recent findings suggest that
methylthioadenosine (MTA) plays a role in regulating the activity of PRMT5.
This finding originated with the discovery that 5-methylthioadenosine phosphorylase (MTAP) is often co-deleted with a commonly deleted tumour suppressor gene,
CDKN2A (cyclin-dependent kinase inhibitor 2A), through close chromosomal
proximity [161]. This leads to the accumulation of MTA, which in turn inhibits
PRMT5 in a surprisingly specific manner. Through this pathway the MTAP/
CDKN2A-deleted tumours have a hypomorphic PRMT5 state, making them sensitive towards further inhibition of PRMT5.
7.2 Inhibitors: In Vitro and Cell-Based Activities
The majority of inhibitors developed against PRMT5 have only been described in
recent years. A hit-to-lead optimization study was recently published by Epizyme Inc.
describing in detail the optimization of compounds identified through HTS
[162]. Extensive SAR studies were performed with compound EPZ007345 (61,
IC 50 of 326 nM), yielding compound EPZ015666 (62, Fig. 15a) [163]. EPZ015666
was found to inhibit PRMT5:MEP50 with an IC 50 of 22 nM with no activity detected
against a panel of 20 other methyltransferases [162]. Furthermore, the compound was
found to be substrate-competitive and AdoMet-uncompetitive. The co-crystal structure confirms EPZ015666 binds in the substrate-binding pocket of PRMT5 (see
Fig. 15c). Interestingly, binding to the PRMT5:MEP50 complex was observed only
178
M. J. van Haren and N. I. Martin
7.1 Background
Initially identified as a Janus kinase-binding protein [142], PRMT5 is the first
identified and most abundant member of the type II PRMTs. The type II PRMTs
preferentially produce symmetrically dimethylated arginine (sDMA) [13]. PRMT5
recognizes both the GAR and PGM motifs in a wide variety of substrates, including
histone arginines H2AR3, H4R3, H3R2 and H3R8 (in vivo) [13, 143, 144]. Nonhistone substrates include ribosomal proteins (RPS10) [145], nuclear factor NF-κB,
tumour suppressor protein p53, transcription factor E2F-1 [143, 146, 147], tumour
suppressor PDCD4 (programmed cell death protein 4) [148] and the MAPK/ERK
pathway (mitogen-activated protein kinase/extracellular signal-regulated kinase)
[149], among many others [56]. PRMT5 has multiple associations with binding
partners of which MEP50 (methylosome protein 50) is known to be necessary for
regulating its specificity in methylating H2A and H4 [150, 151]. Other binding
partners regulate the activity and substrate specificity of PRMT5 [152].
PRMT5 is upregulated in wide variety of human cancers, including breast [148],
colorectal [153], lung [154, 155] and epithelial ovarian cancer [156], lymphomas
[157–159] and melanoma [160]. In addition, recent studies have shown PRMT5 to
be a unique anticancer target [47–49]. These recent findings suggest that
methylthioadenosine (MTA) plays a role in regulating the activity of PRMT5.
This finding originated with the discovery that 5-methylthioadenosine phosphorylase (MTAP) is often co-deleted with a commonly deleted tumour suppressor gene,
CDKN2A (cyclin-dependent kinase inhibitor 2A), through close chromosomal
proximity [161]. This leads to the accumulation of MTA, which in turn inhibits
PRMT5 in a surprisingly specific manner. Through this pathway the MTAP/
CDKN2A-deleted tumours have a hypomorphic PRMT5 state, making them sensitive towards further inhibition of PRMT5.
7.2 Inhibitors: In Vitro and Cell-Based Activities
The majority of inhibitors developed against PRMT5 have only been described in
recent years. A hit-to-lead optimization study was recently published by Epizyme Inc.
describing in detail the optimization of compounds identified through HTS
[162]. Extensive SAR studies were performed with compound EPZ007345 (61,
IC 50 of 326 nM), yielding compound EPZ015666 (62, Fig. 15a) [163]. EPZ015666
was found to inhibit PRMT5:MEP50 with an IC 50 of 22 nM with no activity detected
against a panel of 20 other methyltransferases [162]. Furthermore, the compound was
found to be substrate-competitive and AdoMet-uncompetitive. The co-crystal structure confirms EPZ015666 binds in the substrate-binding pocket of PRMT5 (see
Fig. 15c). Interestingly, binding to the PRMT5:MEP50 complex was observed only
178
M. J. van Haren and N. I. Martin