Amyloid double trouble

John Hardy

John Hardy is at the Laboratory of Neurogenetics, National Institute on Aging, Porter Neuroscience Building, 35 Convent Drive, Bethesda, Maryland 20892, USA. hardyj@mail.nih.gov

A new study shows that some cases of early-onset Alzheimer disease result from duplications of the APP locus, which encodes the amyloid precursor protein. This finding fulfills a 20-year-old prediction that genetic variability in APP expression could lead to disease and provides further, perhaps definitive, evidence for the amyloid hypothesis of the disorder.

When Glenner and Wong first reported the isolation and identification of the amyloid (A) peptide in the meningeal vessels of individuals with Alzheimer disease¹, and later in meningeal vessels of adults with Down syndrome, they wrote², "This is the first chemical evidence of a relationship between Down syndrome and Alzheimer's disease...Assuming [A] is a human gene product, it also suggests that the genetic defect in Alzheimer's disease is localized on human chromosome 21." Not only have the direct predictions in these remarkable papers been shown to be essentially accurate, but the implicit prediction that genetic variability in the expression of the normal APP gene product could cause Alzheimer disease has now been shown to be correct. On page 24 of this issue, Rovelet-Lecrux et al.³ report several independent duplications of the APP locus in French families with a variable, autosomal dominant phenotype intermediate between the pure Alzheimer phenotype seen in most families with APP mutations⁴ and the cerebral hemorrhage phenotype of Dutch angiopathy associated with the APP E693Q (Dutch) mutation⁵. These findings highlight the importance of APP gene dosage and provide strong support for the amyloid hypothesis⁶, which postulates that accumulation of A in the brain drives Alzheimer disease pathogenesis (Fig. 1).

Figure 1. Genetic factors influencing A42 production and Alzheimer disease onset. The amyloid precursor protein (APP) can be cleaved to yield either of two smaller peptides, A40 and A42, the latter of which contributes to disease pathogenesis. Trisomy 21 (Down syndrome), duplication of the APP locus or normal variation in APP expression can result in elevated levels of APP, resulting in increased production of A peptides. Mutations in APP or in the presenilin genes, which encode the proteases responsible for APP cleavage, can also lead to increased production of A42. Normal variation in ApoE4 influences A42 accumulation, contributing to Alzheimer disease pathogenesis. Full Figure and legend (62K)

The report by Rovelet-Lecrux et al. is of interest for several reasons. First, when considered with the report of an individual with Down syndrome who was trisomic distal to the APP locus and did not develop Alzheimer pathology⁹, it demonstrates that modest (50%) increases in APP expression are sufficient to cause disease, with a clinical onset in the fifties. This suggestion is concordant with the notion that genetic variability in the normal disomic expression of APP can predispose individuals to late-onset disease¹⁰. It is difficult to interpret these results in any way other than as strong evidence for the amyloid hypothesis⁶. A clear implication is that individuals homozygous for an APP allele with an expression level 25% higher than normal would also be expected to develop Alzheimer disease in their fifties, with smaller increases in APP expression presumably leading to disease onset at a later age. Second, in the cases reported by Rovelet-Lecrux et al., as in earlier cases involving copy number increases at the -synuclein (4q21) locus causing Parkinson disease¹¹ and the classic cases of PMP22 duplication (17q11.2) causing Charcot-Marie-Tooth disease Type 1A (ref. 12), several other genes are duplicated without obvious phenotypic consequences. This suggests that the expression of most genes is plastic enough that aberrant dosage does not have severe effects. Third, this study confirms that genes in the vicinity of 21q21 do not contribute to the Down syndrome phenotype, except with respect to the occurrence of Alzheimer disease in older individuals with Down syndrome⁹. Fourth, it reinforces the notion that there is a close pathogenic link between vascular and parenchymal amyloid deposition¹³. And fifth, it adds to the increasing evidence that the genome is less tidy than first thought, with many more polymorphic insertions, deletions and inversions than suspected—some of which, like the ones reported here, have dire phenotypic consequences but many others of which probably have no discernable effects¹⁴.

Several questions remain. Why is the phenotype in these families subtly different from that of Alzheimer disease in individuals with Down syndrome, in whom hemorrhages seem rare? Are there individuals who are monosomic for APP, and, if so, what is their phenotype? And finally, are there other examples of neurodegenerative diseases in which copy number increases involving established mendelian loci lead to disease phenotypes? Prime candidates for this would include the MAPT (17q21.1) and PRNP (20pter-p12) loci. With these loci, as with APP and -synuclein, there is clear evidence that variability in expression contributes to risk of the sporadic forms of the associated diseases¹⁵. With regard to Alzheimer disease, all the genetic evidence points to A42 and the process of amyloid deposition, if not the amyloid deposits themselves, as the initiator of pathogenesis (Fig. 1)⁶.

	Top

1. Glenner, G.G. & Wong, C.W. Biochem. Biophys. Res. Commun. 120, 885–890 (1984). | Article | PubMed | ISI | ChemPort |
2. Glenner, G.G. & Wong, C.W. Biochem. Biophys. Res. Commun. 122, 1131–1135 (1984). | Article | PubMed | ISI | ChemPort |
3. Rovelet-Lecrux, A. et al. Nat. Genet. 38, 24–26 (2006). | Article |
4. Goate, A. et al. Nature 349, 704–706 (1991). | Article | PubMed | ISI | ChemPort |
5. Levy, E. et al. Science 248, 1124–1126 (1990). | PubMed | ISI | ChemPort |
6. Hardy, J. & Selkoe, D.J. Science 297, 353–356 (2002). | Article | PubMed | ISI | ChemPort |
7. Hendriks, L. et al. Nat. Genet. 1, 218–221 (1992). | Article | PubMed | ISI | ChemPort |
8. Remes, A.M. et al. Neurology 63, 234–240 (2004). | PubMed | ISI | ChemPort |
9. Prasher, V.P. et al. Ann. Neurol. 43, 380–383 (1998). | Article | PubMed | ISI | ChemPort |
10. Wavrant-De Vrièze, F. et al. Neurosci. Lett. 269, 67–70 (1999). | PubMed |
11. Singleton, A.B. et al. Science 302, 841 (2003). | Article | PubMed | ISI | ChemPort |
12. Lupski, J.R. et al. Cell 66, 219–232 (1991). | Article | PubMed | ISI | ChemPort |
13. Kumar-Singh, S. et al. Am. J. Pathol. 167, 527–543 (2005). | PubMed | ISI | ChemPort |
14. Tuzun, E. et al. Nat. Genet. 37, 727–732 (2005). | Article | PubMed | ISI | ChemPort |
15. Singleton, A. , Myers, A. & Hardy, J. Hum. Mol. Genet. 13, R123–R126 (2004). | Article | PubMed | ISI | ChemPort |

	Top