In recent posts on depleting PrP and gene therapy, I’ve discussed the notion of a ‘magic threshold’ of PrP knockdown required to reverse symptomatic prion disease. This notion is based on the following evidence. Mallucci successfully reversed prion disease in mice by excising a floxed PrP transgene with Cre recombinase [Mallucci 2003, Mallucci 2007]. Although she never actually measured the level of PrP post-floxing, this Cre conditional knockout system probably gives something pretty darn close to a 100% knockout of PrP. In contrast, Tremblay & Safar have reduced (but not eliminated) PrP expression using doxycycline to turn off a PrP transgene expressed under a Tet-off switch, and this did not reverse prion disease – survival was prolonged almost threefold (from 149 dpi to 432 dpi) but the mice did eventually succumb to prion disease [Safar 2005 (ft)].*
* For those new to Tet-off switches, I introduce the concept in this post. It’s a genetic manipulation used to create genes that can be turned on and off using doxycycline. Doxycycline does not have this effect in normal mice or people, and it is inaccurate to cite this study as evidence for the therapeutic efficacy of doxycycline [e.g. Mitrova 2012].
Safar’s paper is pretty information-rich and I still discover new things every time I open it up. After poring over it a couple of times I gleaned that the Tet-off mice, when maintained on doxycycline, expressed ~0.1x wild-type PrP levels. Since this failed to outright reverse/prevent the disease, I had concluded that the magic threshold, if one exists, must be between 90% and 100% knockdown.
However, I tracked down Jiri Safar himself at Prion2013 to ask him more about this. He said his best estimate is that his mice when maintained on doxycycline actually express 0.2x wild-type PrP levels. That’s good – so the range for a hypothetical threshold is now somewhere between 80% and 100% knockdown. But he also pointed out to me that even if there is a magic threshold, you don’t necessarily need to reach it in order to have a drug be therapeutically valuable, delaying and at least temporarily reversing the disease.**
**As an aside, he also told me that his experiment was unfathomably long and labor-intensive and that personally he wouldn’t do it again. Well, thanks for doing it once, dude. It was great work.
So I took another look at his paper today. In one experiment, in which controls died at 149 dpi, a group of mice was treated with doxycycline beginning at 98 dpi, and this group lived to 432 dpi. Those mice were treated at timepoint 98/149 = 0.66 of the way through the control incubation time. Though Safar didn’t do any elaborate behavioral testing or histopathology to identify early changes in these mice, it’s likely that he would have found them if he’d looked hard enough. Mallucci 2007 did extensive behavioral testing on Tg37 mice – the paper is not super clear on expression level but based on Moreno 2012 I believe these were Tg37 heteterozygotes, so expressing 3x wild-type levels, compared to the 2x expressed by Safar’s mice when not maintained on doxycycline. Mallucci’s mice showed behavioral changes and synapse loss at 7 wpi, where controls died at 12 wpi, so symptoms were detectable at timepoint 7/12 = 0.58. All this is to say, Safar’s mice probably also had some early behavioral changes and synapse loss by timepoint 0.66 when he started them on doxycycline. They certainly had accumulated a great deal of proteinase K-resistant PrP (PrP-res), and much of this PrP-res was quickly degraded after PrPC expression levels were reduced. Within a week, PrP-res in the brain dropped ~20-fold [Fig 3] to “the minimum level to be detectable”, and it stayed that low for about 3 weeks before rising again. In at least two human prion diseases, it’s been reported that PrP-res accumulation tracks pretty well with the extent of pathology across brain regions – CJD [Parchi 1996] and FFI [Cortelli 1997] – so it seems reasonable to expect that a 20-fold reduction in PrP-res would have reversed any early pathological changes in Safar’s mice, temporarily, until PrP-res began to accumulate again.
That’s good news. If and when we discover a drug that can reduce PrP expression, even if it doesn’t reduce PrP enough to outright eliminate the disease, it still might be able to at least temporarily reverse symptoms. That would mean that it could bring not only survival benefits but quality of life benefits as well.