Arches. Photo by Daniel Chia
Jun
30
2010

Caspase-6 inhibition

There is ample evidence that Huntington’s disease is associated with a specific genetic mutation that produces an expanded polyglutamine chain in the huntingtin protein. This mutation causes huntingtin to become a misfolded protein with an altered shape. One of the hallmarks of HD is the build-up of short, broken fragments of the altered huntingtin protein in the nucleus of the nerve cell. There are many theories regarding the actual role of these fragments of altered huntingtin protein in the nerve cell’s nucleus. However, many scientists believe that the accumulation of these fragments in the nucleus directly underlies the death of nerve cells in HD. Nerve cell death is responsible for the many cognitive, behavioral, and motor symptoms of HD (for more information about HD symptoms, click here.

The nucleus of a mammalian cell is enclosed by a nuclear envelope, a membrane that features many small openings or “pores.” (The nuclear membrane and its pores can be seen in Segment 4 of the “Basics of HD” video: click here to view that segment.) These pores allow different molecules to move back and forth between the nucleus and the cytoplasm. But these pores are very small and allow only smaller molecules to cross the nuclear envelope. Larger molecules require other, more complex mechanisms to be transported into the nucleus, and these mechanisms often take longer as well.

The altered huntingtin protein associated with HD normally resides in the cytoplasm of a nerve cell because it is too big to be able to easily cross the envelope into the nucleus. But when that intact protein is cut up into small fragments, those fragments can easily move into the nucleus and cause dangerous problems for the cell. Proteases are a family of proteins that break up other proteins into smaller pieces. Studies have shown that a specific group of proteases called caspases play a big role in cutting up altered huntingtin into small fragments that can move from the cytoplasm into the nucleus.

A recent study from the lab of Michael Hayden at the University of British Columbia has shown that a particular caspase protein, named caspase-6, may be responsible for the type of huntingtin fragments that lead to nerve cell death and symptoms in HD. In the study, scientists used a mouse model of HD and changed the altered huntingtin protein so that caspase-6 could no longer cut it into fragments. They found that these mice showed no evidence of nerve cell death and they never developed any symptoms of HD. This finding suggests that a drug that inhibits the activity of caspase-6 may be a treatment for HD.

Background on caspases and HD^

Caspase-6 is not the only protein to cut up altered huntingtin into fragments. Previous studies have shown that caspases can be divided into three rough categories (for more on caspases, click here). There are “ICE-like” caspases (named for their similarity to another kind of protease called the interleukin-1b converting enzyme), “initiator” caspases, and “effector” caspases. ICE-like caspases include caspase-1, 4, and 5. These three seem to play a role in fragmenting proteins involved in processes like inflammation, rather than fragmenting the huntingtin protein. Initiator caspases include caspase-3, 7, and 2, and convert the inactive form of an effector caspase to an active form by cutting off one or two small fragments from the inactive effector caspase. The three effector caspases include caspase-6, 8, and 9, which are the caspases that (when activated by the initiator caspases) actually break down most other proteins.

But there is much overlap between all the caspases, and some fit in more than one category. Furthermore, each of these nine caspases have different target sites where they interact with other proteins. Target sites are specific short sequences of amino acids within a protein where the caspase cuts the protein. Studies have shown that caspase-1, 3, and 6 all target altered huntingtin protein, but they do so at different target sites. The altered huntingtin protein has three locations that have the right target sequences for cleavage by caspase-1. However, for unknown reasons, caspase-1 does not fragment the altered huntingtin protein very much. There are four sites in the altered huntingtin protein that serve as targets for caspase-3. Two of them are active, and caspase-3 does indeed fragment the huntingtin protein at these points. The other two sites are considered “silent” because caspase-3 does not use those targets to fragment huntingtin. Finally, there is only one site that caspase-6 can target, and it’s an active site, so caspase-6 does fragment the huntingtin protein.

The Hayden lab study^

While it was known for some time that both caspase-3 and caspase-6 break down huntingtin protein into fragments, it was not known if all of the resulting fragments enter the nucleus and cause nerve cell death. It seemed possible that the fragments that were particularly toxic to the nerve cell were specifically generated by one of the two caspases. So in their study, Hayden and co-workers used a mouse model of HD, and mutated the altered huntingtin protein so that either the caspase-3 or caspase-6 protein would not find its usual target. This involved changing the specific amino acid sequence that caspase-3 (or -6) usually targets, and only changing that part so that the rest of the huntingtin protein acts the same. Most of the target sites for caspases are only 4 amino acids long, so it is not difficult to selectively change that part.

Hayden’s group generated one mouse that had all four of the caspase-3 target sites changed and inactivated, one mouse that had the single caspase-6 target sites changed and inactivated, and one mouse that had all of the caspase-3 sites and the caspase-6 site changed and inactivated. They tested all of the types of mice to make sure that they were expressing similar amounts of the huntingtin protein, and that the expanded polyglutamine chains were roughly the same length. In so doing, the researchers ensured that the main difference between these mice was the ability for caspases-3 and 6 to fragment the huntingtin protein.

One way to test for nerve cell death is simply to measure the weight of the brain at a certain age in HD mice and compare it with the weight of the brain in other strains of mice. The less the brain weighs, the more you can assume there is nerve cell death. Previous studies have shown that mice with altered huntingtin protein (that can be targeted by both caspase-3 and 6) lose about 10% of their brain mass as compared to healthy, wild-type mice without the altered huntingtin protein. This loss of brain mass can be attributed to nerve cell death due to the HD associated protein. The first thing that the Hayden group observed was that the mice with altered huntingtin protein resistant to both caspase-3 and caspase-6 did not have that 10% loss of brain mass. Instead, the mice were much more similar to the healthy, wild-type mice.

Then, to determine which of the caspases—3, 6, or both—were necessary for brain mass loss, Hayden and coworkers tested each of the other two mouse lines they had generated. They found that the mice with huntingtin protein resistant to caspase-3 cleavage had similar brain mass loss as mice with the HD associated huntingtin protein. In other words, fragments generated by caspase-3 are not the fragments that cause nerve cell death. But these mice still generated fragments due to caspase-6.

Next, they tested the mice with huntingtin protein resistant to caspase-6 cleavage, and they found that these mice had no significant brain mass loss. They were similar to healthy, wild-type mice and to the mice that had huntingtin protein resistant to both caspase-3 and 6 cleavage. Notably, these mice were still generating fragments due to caspase-3. But since these mice had no evidence of brain mass loss, it is evident that fragments selectively generated by the action of caspase-6, but not caspase-3, are toxic and cause nerve cell death.

Additionally, Hayden and his group tested the motor coordination of each type of mouse. What they found was that both the mice resistant to caspase-6 action and the mice resistant to caspase-3 and -6 action, were able to perform normally, just like healthy, wild-type mice. The mice resistant to only caspase-3 action performed poorly, just like mice with the regular HD-associated huntingtin protein. This result shows that selective inhibition of caspase-6 not only prevents brain mass loss, it also prevents motor symptoms of HD.

Finally, they looked specifically at the location of fragments of huntingtin protein within the nerve cell. Mice with the HD-associated huntingtin protein and mice that have caspase-3 resistance (but generate fragments cut by caspase-6) both have fragments that enter the nucleus early in the mouse’s lifetime. Both healthy, wild-type mice and mice that are resistant to caspace-6 (but generate caspase-3 fragments), show little to no signs of huntingtin fragments entering the nucleus. In the caspase-6 resistant mice, researchers saw some fragments enter the cell very late in life, but they still did not cause nerve cell death or symptoms. This points to the idea that it is the action of fragments (created selectively by caspase-6) inside the nucleus that causes toxicity and nerve cell death. If fragments created selectively by caspase-6 are the ones to enter the nucleus, then caspase-6 inhibition might prevent that toxicity and might prevent HD symptoms.

Directions for the future^

A few uncertainties remain to be considered in the Hayden Lab study. Most significantly, it is unknown whether altering the specific amino acid sites that caspase-3 and -6 target has any effect on the rest of the huntingtin protein itself. Perhaps in addition to being targets for caspases, those sites determine huntingtin structure, stability, or clearance. If so, we cannot know whether the lab’s findings on the role of caspase-6 would hold true in human patients. Furthermore, caspase-6 might have other important functions in the cell that an inhibitory drug would impede. Finally, we do not know if there are other caspases or caspase sites that play a significant role in creating the specific huntingtin fragments that lead to nerve cell death in humans. More work will have to be done to answer all of these questions and ensure that any caspase-6 inhibitors developed as drugs are safe and effective.

Nevertheless, this research has excited the HD community. Finding a drug to inhibit the action of caspase-6 seems a promising direction for treatments, and the work that must be done is very practical and logical. Researchers will look for chemical compounds that might inhibit caspase-6, test them in mice with the human version of HD, and if successful, then move on to clinical trials. Hayden says he hopes to start human clinical trials within five years. In fact, at the Cure HD Initiative (CHDI), a nonprofit drug development research organization for Huntington’s disease, efforts are already underway to develop a safe, effective caspase-6 inhibitor. This area of research will be important to watch for the next few years.

For further reading^

  1. Graham RK, et al. (2006). Cleavage at the caspase-6 site is required for neuronal dysfunction and degeneration due to mutant huntingtin. Cell. Jun 16;125(6):1179-91
    This is the main paper discussed in this article: a fairly technical research paper.
  2. Slow EJ, et al. (2003). Selective striatal neuronal loss in a YAC128 mouse model of Huntington disease. Hum Mol Genet. Jul 1;12(13):1555-67.
    This paper discusses the creation of the mouse model used in the study described by Graham et al (2006).
  3. Thornberry NA, et al. (1997). A combinatorial approach defines specificities of members of the caspase family and granzyme: Functional relationships established for key mediators of apoptosis. J Biol Chem. Jul 18;272(29):17907-11.
    A more general review of caspases and their three different functions. Still technical, but more comprehensible.
  4. Wellington CL, et al. (2002). Caspase cleavage of mutant huntingtin precedes neurodegeneration in Huntington’s disease. J Neurosci. Sep 15;22(18):7862-72.
    A preliminary study of the role of caspases in a different model system of HD
  5. Gutekunst CA, et al. (1999). Nuclear and neuropil aggregates in Huntington’s disease: relationship to neuropathology. J Neurosci. Apr 1;19(7):2522-34.
    This paper is a technical but readable research article about where it is in the nerve cell that huntingtin protein and huntingtin fragments tend to localize.

-J. Seidenfeld