Interaction of two genes possibly involved in the regulation of the amyloid precursor protein (APP) processing

SIR – An intronic polymorphism of the gene encoding the transmembrane protein presenilin 1 (PS1) was found to be associated with increased risk for sporadic AD (ie over-representation of allele 1 in AD),1 however, this observation has been challenged by subsequent studies. These inconsistent findings might result from interactions of PS1 with other susceptibility genes and their products. Recently, Xia et al2 identified PS1 to be crucial for the catalytic activity of the γ-secretase, which is one of the two yet unidentified proteases (β- and γ-secretase) responsible for cleavage of the amyloid precursor protein (APP) into amyloidogenic components (βA4).

Cathepsin D (catD), an acid protease active in intracellular protein breakdown, has in vitro β- and γ-secretase-like properties and has also been implicated in the pathogenesis of AD. An exonic polymorphism of the catD gene (allele T) was recently found to be a major risk factor for the development of sporadic AD.3 Wolfe et al4 proposed that γ-secretase is an aspartyl protease with properties similar to those of cathepsin D. Thus, both genetic and biochemical data suggest an involvement of PS1 and catD in the cleavage and further processing of APP. Given this evidence, we hypothesised that genetic interactions between PS1 and catD would modify the risk for the development of sporadic AD and might be responsible for the conflicting reports on the association of the PS1 genotype with AD.

We investigated a sample of 217 patients with sporadic AD (mean age 72.6 ± 9.7 years, 62.2% female) and of 434 non-demented control subjects (mean age 68.2 ± 10.6 years, 56.0% female) recruited at two centres (Psychiatric Departments of the University of Bonn with 102 AD patients and 191 controls and the Technische Universitaet Muenchen with 115 AD patients and 243 controls). Gene interactions were tested by forward and backward logistic regression analysis.

The results (Table 1) show that, while allele 1 of the PS1 gene was not associated with increased risk for AD in the entire sample (odds ratio = 1.2; 95% CI: 0.8–1.8), there was a considerable difference in the distribution of the PS1 genotype when the population was stratified according to the catD genotype: Allele 1 of the PS1 gene doubled the AD risk in persons with at least one T allele of the catD gene (odds ratio = 2.54; 95% CI: 1.01–6.43), whereas it had no influence on AD risk in persons lacking the T allele (odds ratio = 1.00; 95% CI: 0.62–1.62). Logistic regression confirmed this observation. Besides age (P < 0.001) and the apolipoprotein E*4 allele (P < 0.001), the interaction of the PS1 and catD polymorphisms significantly modified the AD risk (odds ratio = 2.5; P = 0.004). Separate examination of the test hypothesis by recruitment site (Bonn/Munich) yielded consistent results between centres.

Table 1 Allele distribution of the intronic PS1 polymorphism after stratification for the catD genotype

The present study shows that interaction of the genes coding for PS1 and catD significantly modifies the risk for the development of sporadic AD. Allele 1 of the PS1 gene increased the AD risk only in subjects carrying at least one T allele of the catD gene, which is also a risk factor for AD. This genetic interaction might have pathophysiological consequences. The catD*T allele has been associated with altered intracellular maturation of catD,5 which is an enzyme with in vitro β- and γ-secretase-like properties.4, 6 Assuming PS1 is crucial for a proper γ-secretase activity and allele 1 of the PS1 genotype is associated with increased AD risk, the interaction of these two genes should influence APP processing and therefore modify the risk for AD. The present study provides an explanation for the conflicting reports on the association of the PS1 genotype with AD.


  1. 1

    Wragg M et al. Lancet 1996; 347: 509–512

  2. 2

    Xia W et al. Biochemistry 1998; 37: 16465–16471

  3. 3

    Papassotiropoulos A et al. Neurosci Lett 1999; 262: 171–174

  4. 4

    Wolfe MS et al. Biochemistry 1999; 38: 4720–4727

  5. 5

    Touitou I et al. Eur J Cancer 1994; 30A: 390–394

  6. 6

    Chevallier N et al. Brain Res 1997; 750: 11–19

Download references


The study was supported by the BONFOR grants program of the Medical Faculty of the University of Bonn (AP 111/34) and the Deutsche Forschungsgemeinschaft (He 2318).

Author information

Correspondence to A Papassotiropoulos.

Rights and permissions

Reprints and Permissions

About this article

Further reading