First author Yuequan Shen and colleagues obtained high-resolution, x-ray crystallographic analysis of IDE bound to several of its substrates, including insulin and Aβ. Their analysis shows the enzyme is composed of four domains which come together to make two halves of a cage that traps substrates. Domains 1 and 2 in the N-terminal make up one-half of the clam shell and are opposed by domains 4 and 3, respectively, which make up the C-terminal half of the molecule. A large polar cavity, flanked by hydrophobic and charged residues, lies between domains 1 and 4 and forms the catalytic site. Despite the structural diversity of IDE substrates, they form remarkably similar crystal structures with the protease. In all four substrates studied, insulin, Aβ, amylin, and glucagon, the N-terminal 3-5 amino acids and the substrate cleavage site, comprising 7-13 residues, form β-sheets with β strands on the N-terminal half of the clam shell. The rest of the substrates show no particular structural order, though they do lie inside the IDE chamber.

Insulin and IDE—Happy as a Clam
X-ray crystallographic structure of IDE and insulin shows that the substrate is buried within a closed shell formed between the two halves of the protease. The N-terminal half, comprising domains 1 (green) and 2 (blue), binds the N-terminal and cleavage site of the substrate (brown). The IDE N- and C-terminal—domains 3 (yellow) and 4 (red)—fit neatly together to enclose the substrate and form a catalytic pore that houses the Zn2+ cofactor. [Image courtesy of Wei-Jen Tang, University of Chicago, Illinois.]

Together, enzyme and substrate form a closed shell, which begs the question, How do substrates enter and products exit the catalytic chamber? Shen and colleagues compared their structures with that of the similar bacterial protease, pitrilysin. That analysis suggests that the C-terminal of IDE must rotate away from the N-terminal by about 54 degrees to form an open conformation that would allow access to substrate. However, the closed conformation may be the default because the crystal structure shows that the N- and C-terminal halves of IDE scallop together very well and are held by numerous hydrogen bonds.

All this suggests that IDE may be less than the optimal catalyst. To test this idea, Shen and colleagues mutated the protein to weaken the bonding between the two halves. They found that simply replacing two amino acids, aspartic acid 426 and lysine 899, with cysteine boosted degradation of Aβ1-42 by 30- to 40-fold. When they similarly replaced asparagine 184 and glutamine 828, or serine 132 and glutamic acid 817, they got similar increases in activity. In contrast, adding an oxidizing agent to drive disulphide bond formation caused two of the mutants to clam up—activity was completely abolished.

IDE Clams Up
a) A latch mechanism keeps IDE predominantly in the closed formation, slowing down catalysis by delaying entry of substrate and release of products. b) Mutations that abolish hydrogen bonds and weaken the latch speed up catalysis. Small molecules that weaken the hydrogen bonds might achieve similar results. [Model courtesy of Malcolm Leissring, The Scripps Research Institute, Jupiter, Florida, and Dennis Selkoe, Brigham and Women’s Hospital, Boston.]

As Malcolm Leissring, The Scripps Research Institute, Jupiter, Florida, and Dennis Selkoe, Brigham and Women’s Hospital, Boston, write in an accompanying Nature News & Views, the hydrogen bonding between the two halves of IDE form a latch that keeps the protease closed and delays entry of substrate and exit of product (see model above). “Shen and colleagues’ results excitingly suggest that drug molecules might also be able to activate IDE by mimicking the latch disrupting mechanism,” they write. This could be a valuable therapeutic strategy for tackling the build-up of Aβ in Alzheimer disease. In fact, overexpression of IDE or another Aβ-degrading enzyme, neprilysin, attenuates pathology in animal models of AD (see Leissring et al., 2003).

While the crystallographic data from Shen and colleagues suggest a rare opportunity to design small molecules that boost enzymatic activity (small molecules are overwhelmingly inhibitors), traditional small molecule inhibition of IDE may also be useful for treating diabetes. And as Leissring and Selkoe point out, careful design of these inhibitors to prevent them from crossing the blood-brain barrier would ensure that IDE-catalyzed degradation of Aβ in the brain goes on unimpeded.—Tom Fagan.

References:
Shen Y, Joachimiak A, Rich Rosner M, Tang W-J. Structures of human insulin-degrading enzyme reveal a new substrate recognition mechanism. Nature. Advanced online publication. October 11, 2006. Abstract

Leissring MA, Selkoe DJ. Enzyme target to latch on to. Nature. Advance online publication. October 11, 2006.