The discovery of small-molecule inhibitors of ERK5 for the treatment of cancer

Abstract

Mitogen-activated protein kinases (MAPKs) play an essential role in the transduction of extracellular signals to cytoplasmic and nuclear effectors. MAP kinase kinases (MEKs/MAPKKs) represent protein kinases upstream of MAPKs, critically controlling cellular proliferation and apoptosis. Mitogen/extracellular signal regulated kinase kinase-5 (MEK5) specifically activates extracellular signal-rlated kinase-5 (ERK5), also known as Big MAP kinase 1 (BMK1). Studies conducted in Newcastle have shown that abnormal MEK5/ERK5 signalling is present in certain types of prostate cancer and correlates with shorter disease-specific survival. ERK5 is over-expressed in 20% of breast cancers and is also thought to promote cell growth in hepatocellular carcinoma (HCC) by regulating mitotic entry. In addition, ERK5 has been shown to interact with, and inactivate promyleocytic leukaemia (PML) protein. This prevents activation of its downstream effector, tumour suppressor p21. An ERK5 inhibitor based on a series of isoform selective polo kinase inhibitors was reported in the literature. XMD8-92 displayed selectivity for ERK5 when tested against a panel of 402 kinases in an ATPsite competition binding assay (ERK5; IC50 = 300 nM) and the compound showed reasonable activity in HeLa cells (GI50 = 1.5 μM). XMD8-92 was also shown to inhibit growth in two human tumour xenograft models (HeLa cells and Lewis lung cells), highlighting that inhibition of ERK5 is a valid target for anti-cancer therapy. Encouragingly, both in vitro and in vivo activities of XMD8-92 have been tested and confirmed in our assays. A high throughput screening (HTS) campaign was conducted using an immobilised metal affinity polarisation (IMAPTM) format assay at Cancer Research Technology- Discovery Laboratories (CRT-DL), identifying three distinct chemical series for further investigation. A pyrrole carboxamide series (1) has undergone hit validation by resynthesis and re-testing of the original HTS hits along with synthesis of closely related series. Initial hit-to-lead studies focussed on modification of R1 in structure 1, and the best activity against ERK5 was observed when R1 is an aromatic ring disubstituted with a 5 halogen at the 2,3- or 2,6-positions (e.g. 2, ERK5; IC50 = 2.3 M), with the R2 substituent being a pyridyl moiety linked to the amide via a methylene group. Further structure-activity relationship studies (SARs) revealed that removal of the methylene group led to an increase in potency (3, ERK5; IC50 = 1.1 M) and more importantly, in more than 100-fold selectivity for ERK5 over the closely related MAPK, p38α. Introduction of a different halogen atom into the 2-position of the aromatic ring conferred good potency and retained selectivity for ERK5 over p38α (e.g. 4, ERK5; IC50 = 0.6 M and 5, ERK5; IC50 = 0.8 M). Inhibitor 5 was selected for in vitro and in vivo pharmacokinetic (PK) studies with promising results [in vitro PK; solubility >100 μM, PPB = 94% and in vivo PK; oral bioavailability = 68% and t1/2 = 65 min (p.o. mice)]. More recently, introduction of a 2,3,6-tri-substituted aromatic ring led to an increase in potency against ERK5, identifying the first compound with an ERK5 IC50 <100 nM, 6 (ERK5; IC50 = 0.07 μM). Modification of R2 has also been investigated in order to improve CYP450 inhibition profiles. Synthesis of a series of compounds with various substituted pyridyl, pyrimidyl or pyridazyl rings was conducted. The SARs revealed compounds with a pyrimidyl moiety to be potent inhibitors of ERK5 (IC50 <5 M) with excellent CYP450 inhibition profiles when tested against a panel of five CYP450 isoforms (IC50>25 μM). Further lead optimisation studies are currently being undertaken in order to find an inhibitor suitable for clinical evaluation.