What about proteins with fewer CAG repeats?

   Bates: This year Erich and Hans looked in more detail using this system, and they have made proteins that contain a whole range of CAG repeats around the threshold for this disease in humans. They find that as you go from 32 to 37 repeats, there’s a dramatic increase in the rate of this aggregation. That’s just a beautiful piece of work. It really suggests that it’s the ability of these glutamines to aggregate that is the molecular trigger of these diseases.
   Obviously, then, there’s an aggregation process, and we don’t know much about that yet. We don’t know if a conformation change is the first step of the aggregation process, and presumable the end stage is the huge lumps we see in the cells.

So if it’s not cell death that’s causing the disease symptoms, what is it?

   Bates: Because we didn’t see cell death in mice by 12 weeks, it suggests that the symptoms in the mice are caused by cells going wrong rather than cells dying. We have collaborated with many labs to use whatever special techniques they might have to look at the brains, and it is quite clear now that you can see this aggregation occurring very early on in the mouse brains. You can see clear clumps by three weeks in some regions, and the load of these polyglutamine aggregates dramatically increases up to 12 weeks, at which point these mouse brains are loaded with the stuff. And we see evidence of neuronal dysfunction and the down-regulation of specific neurotransmitter receptors in these cells by four weeks. This is work done primarily by Anne Young, who is chief of neurology at Massachusetts General Hospital, and Jang Ho Cha, who is in her department. And Steve Davies finally did see cell death from 14 weeks onward.

Are the affected areas of the brain consistent with what you see in HD in humans?

   Bates: The remarkable thing about that is when we did see cell death, we observed that it is in the same regions of the brain as one would see in HD. We never expected this. Remember, we have this short fragment of the protein, not the whole protein, being expressed in mice. So in terms of what this polyglutamine load is doing in brains, we can say that it doesn’t easily kill cells, because we don’t see any cell death until the very, very end stage. And it’s very selective. It’s making cells dysfunction, and that’s what’s causing the symptoms. This may well be what’s happening in patients.

It seems like this might this give you a potential therapeutic strategy.

   Bates: Yes. We think that a target for therapeutic intervention is to try and prevent the aggregation, and in the process try to correct some of the dysfunction. In Berlin, Hans has established a collaboration with Merck to screen their pharmaceutical library of 200,000 compounds for small compounds that might be turned into a drug, get across the blood-brain barrier, and have some effect blocking this aggregation process. Anything that comes out of the screen we will try in mice. That is the major focus of work going on in our lab at the moment.

Are you optimistic?

   Bates: Obviously, trying to block something like this is quite a tall order. But all we have to do for most HD patients is slow down the rate at which the disease progresses. One method has already been found to slow down symptoms in our mice. This was done recently by Rob Friedlander at Harvard. He crossed our mice with some other mice in which he had expressed something called a dominant negative caspase-1 transgene, and he has found that everything is delayed–the aggregate formation, motor dysfunction, and all the other symptoms. I don’t understand how it happens yet. It’s not an experiment I would have ever thought of doing.
   So it's not like we must prevent HD from ever happening. Even if we could slow it down by a factor of two in many cases, we would shift the age of onset to beyond the normal life expectancy. If it’s possible to do this, then I think we’ll do it in the relatively near future.