Biological Evaluation of Compounds

What are the best strategies for minimizing metabolic clearance in OSM Series 4? This has been, and remains, a key scientific issue for us, along with maintaining potency and improving solubility.

We're devising the next set of compounds to make (also GHI301) and we're meeting next week (GHI317), so this is a good time to re-engage with metabolic clearance.

Separate Issues will be created for each relevant To Do item on Github, but here are all the background data:

1) All the current clearance data are in the wiki. Some data inherited, others obtained more recently. For the moment we are deprioritising the amide series (top row of data in the main wiki scheme) because of a probable hERG liability. So let's focus mostly on the non-amides.

2) Chris Swain predicted metabolic hotspots (in red, below) for three Series 4 compounds, MMV669844, MMV669848 and MMV670936.

When Chris ran the same compounds through a Cyp predictor the same sites came up minus those on the ring, suggesting the ring could be a substrate for another enzyme such as aldehyde oxidase. However, measured AO clearance values suggest MMV669844, for example, is not an AO substrate (see below).

3) Met and MetID work on these same three compounds above from Sue Charman's lab suggested that oxidation on the triazolopyrazine-aromatic system was occurring in most cases, with other liabilities shown above. The analog of MMV669844 with a CHF₂ group in place of the CH₃ (MMV670652, 17 nM) was tested in microsomal studies, giving CLint HLM<8, RLM 30.

4) Aldehyde Oxidase Screen. Performed, with summary data. Amides appeared to be more susceptible to AO oxidation, but in general several potent members of the ether class were found not to be substrates for AO. We should both not worry about AO for this series and yet periodically re-assess Series 4 compounds in the assay.

5) The nature of the core heterocyclic ring has a major effect on clearance - compare the parent triazolopyrazine MMV639565 with the imidazopyrazines MMV669846 and MMV670250.

The potency and reasonable clearance of MMV669846 make it interesting to look at again. The very high clearance of MMV670250 suggests...what?

6) We know that blocking the pyrazine ring in the 8-position reduces potency and that the N-oxide in the pyrazine ring is inactive (below), and we also now know that replacement of the pyrazine southwest N with CH lowers potency. Is there any reason to investigate these further, e.g. for metabolic clearance rate on these compounds?

6) To Do/Questions

i) Benzylic Blocking: Are there blocking groups in the northwest benzylic position we ought to include that might improve metabolic stability? Is it possible to run some of these through predictive software to see if such structures might help?

ii) Acquiring New Metabolic Data on Existing Compounds: MMV670652 (above) has not yet been studied in hepatocytes, and has not had MetID performed. Is there justification for this, to confirm that we see improvements over MMV669844, or can this be assumed?

iii) Acquiring New Metabolic Data on New Compounds: Of the newer compounds synthesised (1, 2, 3), are there any that ought to be evaluated at the Charman lab for microsomal stability (GHI213) (and solubility: GHI260)? Should we re-run Chris Swain's analysis on any new target compounds?

iv) Trends in the Data: In general the newer Monash data (in green in the wiki) shows that a number of the analogs with blocking groups in the northwest benzylic position and OCHF₂ in the northeast have higher than expected clearance rates vs. what might be expected given the data for the analogs with similar blocking groups in the benzylic position and a cyano group in the northeast. Is it possible that the northeast Ar substituent is playing a role in metabolic liability of the aromatic system?

v) Triazole Mods: We should resynthesise the imidazopyrazine MMV669846 shown above to verify potency and microsomal stability. Underway. Are there other groups that could be entertained on the triazole, or in place of it? What is known of good mimics of such a ring system?

vi) Pinning Down Site of Oxidation: Is it worth establishing where on the triazolopyrazine-Ar structure the oxidation is occurring, given the prediction that Cyp will not oxidise on the TP ring and given the low AO clearance experimentally determined? (Related to GHI124).

Post originally authored by Mat Todd

Posted on behalf of Joie Garfunkle:

"Here is a list of compounds submitted for both MetID (3 compounds) and microsomal stability/LogD/solubility studies (8 compounds). When we receive the data we will post the results to the OSDD community. In the meantime, if there is another institution willing to assist in DMPK work for free, we are more than willing to help supply solid samples if available."