Liver stage results

[david1597]

At the start of this year we received back some results from the Winzeler lab regarding liver stage testing of some series 4 compounds, idea and shipping of compounds discussed in #4. We've had the results a while - sorry - but are now in the process of incorporating them into the paper.

The compounds sent were the resolved enantiomers from one of the frontrunners, plus the racemate. Liver Jan Results.zip

The results are below, I'm awaiting further info and details on the protocols used. Posting this issue here in advance as a reminder to myself, as much as anything. Data will be incorporated in the wiki and series 4 paper once we have clarified the protocols.

[david1597]

The data obtained at the turn of this year showed that none of the compounds showed activity in the liver stage assay; in each of the three samples sent, the potency was > 50 uM in each case.

This contrasts with an earlier result obtained in February 2014, when MMV669844 was tested and showed a potency of 1.4 uM against the liver. MMV669844 was an inherited compound with the same structure - it was enantioenriched, but we do not know the composition. This is problem number one - no matter the degree or direction of enentioenrichment, we saw nothing in the recent samples, for a racemic mix and both enantiopure (at 88% and 99%). Note the very large confidence interval for the Feb 2014 value. Note also that the original briefing document stated "the series has...no activity against Winzeler’s Pb liver stage", albeit without any further information or data.

There was no toxicology towards the HepG2 cells in any case.

The recent liver data also came back with asexual blood stage potencies. These results indicated that the S enantiomer was around 5,000x more potent than the R enantiomer. But, that the racemic compound was an order of magnitude more potent than either of the enantiopure samples - this is problem number two.

Additionally, whilst the relative order of these potencies matches those observed under the Dundee assay the absolute values are somewhat different, eg. for MMV1557951 the Dundee potency is 0.89 uM while the UCSD value is 0.003 uM. These assays use different strains (3D7 at Dundee, DD2 strain at UCSD).

[drc007]

@david1597 Have we double checked all the structures including the batch of MMV669844 that was originally tested in the liver stage assay?

With respect to the blood stage data, we would only expect a 2-fold difference between racemate and active enantiomer so I suspect the differences you are seeing are within experimental error. Particularly if the data is from an extreme end of the dose range used from serial dilutions.

[david1597]

Thanks for the response @drc007 (and also across in the hERG issue - interesting stuff).

OK, agreed that the blood stage data may just be within experimental error. We don't have the actual data points (we can probably ask) but it does look like we might have been at the very end with that 0.3 nM point. My concern with this data was also partly that this value of 0.3 nM is very low.

I will check - hopefully tomorrow - as far back as possible for all the info we have each of the samples that was actually sent.

[mattodd]

Hi @david1597 - yes, I think this issue must centre on the twin considerations of a) are we sure of the resolution data we received back from UQ - i.e. an unambiguous sense of which sample contained what approximate enantiomeric ratio, and b) are there differences in the liver stage assays here that may account for the data? If we have a sample of the material still in the lab (rac or not) we could re-confirm the purity/composition by 1H NMR spec and then consider re-sending if we need to.

[mattodd]

Tying up loose ends for the paper. This issue is adequately linked to from the wiki, so I'm closing it, despite our not getting the clarity we wanted.

Also attaching the original sheet from UCSD that I don't see posted anywhere, dated Jan 2018, also posted here.

Liver Stage Data UCSD Jan 2018.xlsx

In correspondence between Jenya Antonova and David Smith in April 2018, some of these issues of ambiguity were discussed, to paraphrase:

Jenya: The compounds tested in February 2014 were tested as described in the Swann et al 2016 paper (10.1021/acsinfecdis.5b00143). The other 2 documents attached are the assay descriptions for PbLuc liver stage and ABS which we are following now (posted here). The difference between work done in 2014 and now for liver stage are minor.

David: We had the racemic mixture looked at previously. Do you know if, or how, the assay that was used in this Feb 2014 experiment differs from the one run recently? Jenya: The liver stage protocols we use today is the same as before with sight modifications such are: a) we filter mosquito homogenate through 20uM filter, not 40uM as it was done to reduce amount of mosquito debris in the PbLuc mix used for HepG2 cells infections. b) We are using acoustic transfer system for compounds transfer not a pin-tool as it was used in February of 2014. Neither of these modifications should affect IC50 significantly.

David: The blood stage study indicates an IC50 for MMV897709 of 0.3 nM. This is a lot lower than we were expecting based on our ABS results from Dundee. Any explanation for difference? Jenya: Our ABS assay is run with Dd2 strain non-synchronized. Does Dundee use different strain or synchronized the parasites?