Two-year effort identifies several compounds suitable for proof-of-concept studies.

Recently I had the opportunity to present a scientific poster about a Charles River Labs’ collaboration with CHDI that addressed the “hot topic” of targeting kinases in CNS, in our case, for the treatment of Huntington’s disease, and I was pleasantly surprised when it was awarded the “best poster prize.”

For those unaware, kinases bind substrate proteins and a cofactor, ATP and transfer a phosphate group from ATP to amino acids with free hydroxyl (-OH) groups (serine, threonine or tyrosine), playing a key role in most cellular activities. Drugs that inhibit kinases are being used to treat cancer and more recently rheumatoid arthritis. Growing evidence suggests kinases play a key role in CNS diseases.

We started working on ATM (ataxia telangiectasia mutated) kinase at the beginning of 2013. By the end of 2015 we had produced several candidate molecules suitable for proof-of-concept studies in HD mouse models and demonstrated in vivo ATM inhibition of our best molecule in the mouse brain. CRL medicinal chemistry (computational and synthetic chemistry being an important part of it!), DMPK and biology expertise, worked together with the experienced CHDI leadership to prove how quickly a project can progress if we work as one fully-integrated team.

ATM is a phosphatidylinositol-3-kinase-like kinase (PIKK) involved in the DNA damage response. It plays a central role in maintaining genome integrity by regulating the detection and repair of DNA double-strand breaks. Recent studies have shown that genetic and pharmacological reduction of ATM signalling can ameliorate mutant huntingtin (mHTT) toxicity in cellular and animal models of HD. Additionally, ATM kinase signalling was shown to be altered in the post-mortem brains of HD patients. Selective inhibition of ATM could therefore provide a novel clinical intervention strategy for the treatment of HD.

We chose a known potent and selective ATM inhibitor (KU60019), as starting point for an optimization program, with the ultimate objective of creating brain penetrant compounds suitable for proof-of-concept studies in HD models. We quickly established biochemical and In-Cell Western™ ATM assays to assess compound inhibitory effects. Given the role of PGP in keeping small molecules out of the brain, we chose the MDCK/MDR1 assay to measure permeability and P-gp mediated efflux.

KU60019 exhibited poor brain exposure in mouse, due to poor physicochemical properties and P-gp efflux. Using a homology model of the ATM kinase domain and optimizing CNS-compliant physical chemical properties, we have been able to design potent and selective ATM inhibitors with improved brain exposure.

Furthermore, compounds have been tested to evaluate the neuroprotective role of ATM inhibition in a primary neuronal model of HD. We assessed ATM inhibitors in a mHTT phenotypic assay, using rat primary cortico-striatal co-cultures. Our compounds have shown protection of cortical and striatal cells with EC50 in accord with their ATM cell EC50. Our lead compound, CHDI-‘194, displayed excellent oral bioavailability and pharmacokinetics. Oral administration to mice showed distribution into the brain and linear pharmacokinetics in a dose escalation study.

We then evaluated CHDI-‘194 in an acute PK/PD biomarker study. ATM exists as an inactive dimer in unstressed cells. In response to ionizing radiation, X-ray irradiation or other treatments that introduce double-strand breaks, ATM kinase is rapidly activated by intermolecular autophosphorylation of serine 1981, leading to dimer dissociation. ATM activation is then required for downstream substrate phosphorylation. With this in mind, we studied the effect of X-ray irradiation on phosphorylation of the ATM substrate KAP1 and, consistent with the literature, we observed a dose-dependent and highly transient effect of irradiation. Significant dose-dependent inhibition of KAP1 phosphorylation by CHDI-‘194 was detected at 30 and 60 minutes post-irradiation.

So, what does this experiment mean over the long run, as we continue to search for drugs that can successfully treat HD? A logistical problem in HD drug development is finding compounds that can penetrate the CNS? So the more we learn from projects like this about the physiological role of kinases the more likely we are to develop new medicines that stick. Toward that end, our next step will be will be to evaluate the compound in disease mouse models and we await these results with great expectation. Stay tuned!

How to cite:

Breccia, Perla. Optimizing Kinases in HD. Eureka blog. June 7, 2016. Available: