Preliminary results from GSS mice treated with anle138b
Lab Note #1
May 22, 2014
As promised in our last update, we are now writing with news on the experiments James Mastrianni has been conducting in Chicago, treating his mouse model of GSS with the candidate therapeutic compound anle138b. The mice for this experiment were born at different times, so they are all currently in various stages of the experiment. To review, our experimental plan called for three groups of GSS mice: 15 mice treated with anle138b from their 1st month of life, 15 mice treated later in the disease course, and 15 placebo-treated mice as a control. As of this writing, 12 of the 15 early-treated mice and 4 of the 15 placebo mice have died, which is enough that Dr. Mastrianni feels comfortable beginning to draw at least some very preliminary conclusions.
It appears fairly clear that anle138b is not extending the survival of the GSS mice. The below figure shows how long the mice of each group have survived - as you can see, the curves are almost exactly overlapping, with both groups succumbing to disease at an age of around 157 days. These are, of course, preliminary data – but there isn’t the slightest whiff of a difference here, so it seems unlikely that the rest of the data will change this picture much.
Surprisingly, though, the neuropathology of the treated mice seems dramatically altered. Normally, these GSS mice get a large number of prion protein plaques in their brains as they age [Yang 2009], similar to what is seen in human GSS patients. These plaques are seen in the placebo-treated mice just as usual. The anle138b-treated mice, on the other hand, have only about ~40% as many plaques as the placebo mice. Data from the cerebellum, one particularly affected brain region, are shown below:
The difference is highly statistically significant - a p value of .0001 means it's vanishingly unlikely we would see such a difference by chance alone. Of course, just because it’s not by chance doesn’t mean for sure it’s due to anle138b. There could be some other bias or confounder - these things happen all the time in science. Still, because these mice have the same food, same environmental conditions and same genetic background, anle138b does seem to be the most likely explanation.
The below image shows an example of what these differences in plaques look like. On the left, the placebo mouse's cerebellum is peppered with plaques stained an orangish color. On the right, the anle138b-treated mouse has only a few of these plaques.
It is surprising that the drug seems to be changing of the neuropathology of the mice without extending their lives. Probably the most likely explanation is that anle138b just doesn’t work against GSS in the same way it works against the RML prion strain. RML is the most-studied laboratory strain of prions, against which anle138b was originally tested and shown to double survival [Wagner 2013]. It may be that anle138b works less well against the GSS prions than RML prions, or it might affect them in a different way that doesn’t change their neurotoxicity.
An additional possibility, which we’d like to be able to rule out, is that the effective dose of anle138b that the mice received in this experiment, where the drug was given in food pellets, isn’t as much as the original experiment, where the drug was given in peanut butter.
So on the basis of the preliminary results, we’ve discussed with Dr. Mastrianni making a slight change to the experimental plan. Originally, we were slated to have 15 mice treated with anle138b only after the onset of symptoms, but because mouse breeding is a slow process, none of the mice for that group were born yet by the time we got these results. Now that we know anle138b is not extending survival even when mice are treated right away, it seems unlikely we’ll see an effect in the mice treated later on. Instead, we’d like to devote the funds to running a parallel experiment in which mice infected with RML prions are treated with anle138b or placebo in the same food pellets. If anle138b increases survival in those mice we’ll know that the food pellets are working and that the difference is really between GSS and RML. If it doesn’t, then we’ll know it was a problem with the food pellets.
Assuming we find out that the difference really is between GSS and RML, what does all this mean?
On one hand, it is disappointing that anle138b has not extended the lives of these mice. This compound has been reported to work against multiple prion strains, and even multiple different prion-like proteins, in vitro [Wagner 2013]. This data had provided some basis to hope that it would show similarly broad effects in vivo, extending survival in mice affected with a wide range of prion strains. So far, that’s not what we see. This makes us somewhat less optimistic that anle138b will work in the FFI or E200K mouse models, but we won’t know until we do those experiments.
On the other hand, anle138b does appear to be doing something in these mice, and this may be able to teach us something about this mouse model, this compound, or both. The finding that the plaques are so reduced without extending survival certainly suggests that the plaques are not what’s killing the mice. Indeed, Dr. Mastrianni has previously treated his GSS mice with a drug called rapamycin, and found that the drug completely abolished the plaques, yet there was only a slight survival benefit [Cortes 2012].
On a broader level, hopefully these observations can contribute to a larger effort to understand the ways in which small molecule anti-prion activity is, and is not, strain-specific. The more data we can gather across molecules and strains, the more we can learn about the principles underlying strain specificity and about the prospects for small molecule therapeutics as a strategy against prions.
We will keep you in the loop as these experiments continue to unfold. Thank you again for your support, and for your continued interest! We are excited to learn all that we can from this work. No given therapeutic trial is likely to reveal a magic bullet, and more often than not in science, the data paint a picture far more complex than the starting hypothesis. But working to understand that complexity is how we make sure that even experiments that don't yield a drug move us closer to ones that will.
Sonia & Eric