Betts V, Leissring MA, Dolios G, Wang R, Selkoe DJ, Walsh DM.
Aggregation and catabolism of disease-associated intra-Abeta mutations: reduced proteolysis of AbetaA21G by neprilysin.
Neurobiol Dis. 2008 Sep;31(3):442-50.
Please login to recommend the paper.
To make a comment you must login or register.
Indigestion and Alzheimer Disease
In a number of familial forms of Alzheimer disease (FAD), mutations in the amyloid-β protein (Aβ) precursor, APP, result in single amino acid substitutions in Aβ. These substitutions, named after the ethnicities of the kindreds in which they are found, include the Flemish (Ala21Gly), Dutch (Glu22Gln), Italian (Glu22Lys), Arctic (Glu22Gly), and Iowa (Asp23Asn). The diseases caused by these substitutions present as classic AD, cerebral amyloid angiopathy (CAA), or a combination of the two, and may include unusual plaque morphology or especially severe vessel pathology. Cures for AD and CAA, and for neurodegenerative disorders involving other proteins, may depend upon the elucidation of key factors controlling protein metabolism and self-assembly. Unfortunately, both phenomena are complex and operate contemporaneously in the brain. This has made the identification of targets for disease therapy exceedingly difficult.
Now, in the June 17 Neurobiology of Disease online, Betts et al. report results of studies addressing one aspect of the protein metabolism puzzle, Aβ catabolism, as well as the kinetics of peptide assembly. To do so, wild-type Aβ(1-40) and its five mutant homologues were chemically synthesized and then their susceptibility to digestion by three different endoproteolytic enzymes was assessed. Evidence suggests that two of the enzymes, neprilysin (NEP) and insulin-degrading enzyme (IDE), are involved in the normal proteolysis of Aβ in vivo. A third enzyme, plasmin, may also degrade Aβ, but only in the diseased brain.
The results show, as has been long known in the amyloid field, that peptide aggregates are protease-resistant. However, what has been less well appreciated are potential differences in protease sensitivity among the different Aβ alloforms. Such differences could result from structural variation in a folded monomer state as well as intrinsic differences in enzyme activity that depend upon peptide primary structure at, or adjacent to, susceptible peptide bonds. However, to address these last questions, no aggregation can exist. Betts et al. ensured this was the case by carefully preparing aggregate-free samples. The ensuing experiments revealed that all the peptides could be digested, but that the Flemish peptide was more resistant to NEP action than were any of the others.
The data contrast with earlier results that showed that all the peptides except the Iowa (which was not studied) were either “totally resistant to proteolysis by NEP,” or were “degraded poorly.” The authors suggest that this contradiction could be due to differences in aggregation state among the experiments. This is highly likely, as a basic tenet of protein primary structure analysis is the necessity of denaturing the substrate, something that rarely has been done in studies of Aβ. In fact, our own studies have shown that every peptide bond in monomeric Aβ42 (and hence Aβ40 as well) is susceptible to proteolysis (Lazo et al., 2005).
If we stipulate to the fact that an intrinsic insensitivity to proteolysis does not exist among the Aβ peptides studied, and the data extant and now those from Betts et al. show this is true, then we are left with the conclusion that differences in the conformational dynamics of the monomer must explain the relatively low digestion rate of the Flemish peptide. The implication of this conclusion is central to understanding the Aβ system—namely that pathogenesis depends on the kinetics and not on the thermodynamics (absolute resistance or sensitivity) of the catabolism of toxic Aβ structures.
A kinetics argument is consistent with the results of studies of the time-dependence of β-sheet formation among the Aβ peptides studied. These results showed two general kinetic classes, one comprising peptides assembling more rapidly than wild-type Aβ and one comprising peptides assembling more slowly. The only member of the latter class was the Flemish peptide.
Betts et al. suggest that a mechanistic explanation for the relatively slow digestion of the Flemish peptide by NEP is glycine-mediated secondary structure destabilization. They likely “got it wrong” here, because glycine does not intrinsically destroy structure, as evidenced, e.g., by the stability of the collagen triple helix and silk, and also because peptide bond cleavage requires denaturation of “secondary structure” to eliminate hydrogen bonds involving the Nα and Cα atoms of the labile peptide bond. Finally, conspicuous by their absence were data from studies of Aβ42. This longer peptide, which many argue is the primary neurotoxin in AD, assembles distinctly from Aβ40. Do the differences in kinetics and protease sensitivity observed with Aβ40 extend to Aβ42? This latter question may be the only clinically relevant one.
Lazo ND, Grant MA, Condron MC, Rigby AC, Teplow DB.
On the nucleation of amyloid beta-protein monomer folding.
Protein Sci. 2005 Jun;14(6):1581-96.