β- but not γ-secretase proteolysis of APP causes synaptic and memory deficits in a mouse model of dementia

Abstract

A mutation in the BRI2/ITM2b gene causes loss of BRI2 protein leading to familial Danish dementia (FDD). BRI2 deficiency of FDD provokes an increase in amyloid-β precursor protein (APP) processing since BRI2 is an inhibitor of APP proteolysis, and APP mediates the synaptic/memory deficits in FDD. APP processing is linked to Alzheimer disease (AD) pathogenesis, which is consistent with a common mechanism involving toxic APP metabolites in both dementias. We show that inhibition of APP cleavage by β-secretase rescues synaptic/memory deficits in a mouse model of FDD. β-cleavage of APP yields amino-terminal-soluble APPβ (sAPPβ) and β-carboxyl-terminal fragments (β-CTF). Processing of β-CTF by γ-secretase releases amyloid-β (Aβ), which is assumed to cause AD. However, inhibition of γ-secretase did not ameliorate synaptic/memory deficits of FDD mice. These results suggest that sAPPβ and/or β-CTF, rather than Aβ, are the toxic species causing dementia, and indicate that reducing β-cleavage of APP is an appropriate therapeutic approach to treating human dementias. Our data and the failures of anti-Aβ therapies in humans advise against targeting γ-secretase cleavage of APP and/or Aβ.

Figure 1. Mapping the BRI2 domain that binds APP and inhibits APP processing

1. A. APP is cleaved by β-secretase into sAPPβ and β-CTF. γ-cleavage of β-CTF yields Aβ and AID/AICD peptides. Alternatively, α-secretase clips APP into sAPPα and α-CTF. α-CTF is cut by γ-secretase into P3 and AID.

2. B-C. BRI2 binds APP and inhibits processing by α- and β-secretases. Binding of BRI2 to β-CTF inhibits cleavage by γ-secretase.

3. D. Constructs and domains [cytoplasmic (Cyt), transmembrane (TM), extracellular (Lumen), brichos (B) and convertases-cleavage site, myc-tag]. Lysates (L) and α-myc immunoprecipitates (myc-IP) from transfected cells were analysed by Western blot (WB) for α-Tubulin, BRI2, APP and APP-CTFs. Supernatants (SN) were analysed for sAPPα and sAPPβ. *Indicates an APP-CTF larger than β-CTF, which is routinely observed when BRI2 is over-expressed. This band, whose origin is unknown, also binds to BRI2.

4. E. APP-Gal4, AID-Gal4, Gal4-depended promoter, luciferase reporter, cytoplasm (Cyt) and nucleus (Nc) are schematically indicated. Luciferase activity is expressed as % of the activity in cells transfected with APP-Gal4, luciferase reporter and empty vector (vec). Overall, our analysis shows that BRI2 residues comprised between amino acids 102 and 134 retained APP-binding properties and inhibitory effects on APP processing.

Figure 2. A BRI2-derived peptide binds APP and inhibits β-cleavage of APP

1. A-B. HEK293-APP cells were incubated with the indicated peptides. β- and α-cleavage of APP were quantified by measuring sAPPβ and sAPPα in media by WB. WB of cell lysates detected APP and α-Tubulin.

2. C. WB analysis of cell lysates and conditioned media from HEK293-APP cells treated either with the indicated concentrations of either N3-2A or N1. In a duplicate experiment, cells were treated with compound-E (+) while incubated with the indicated peptides. Lysates were probed for APP, APP-CTFs and α-Tubulin, culture media was probed for sAPPα and sAPPβ. In the right panel, the results of a similar experiment in HeLa-APP cells are shown.

3. D-F. WB analysis of lysates (L) or α-Flag IP (IP) from HeLa/APP cells incubated for 2 h with Flag-tagged peptides.

4. D. Cells were incubated at either 37 or 4 °C with or without 40 µM N3-2A-F.

5. E. The indicated concentrations of N3-2A were added to the media containing 40 µM N3-2A-F.

6. F. Cells were incubated with 40 µM N3-2A-F, N4-F or N3-4A-F.

7. G. Brain cells were cultured as in (D).

8. H. Biotinylated cells were cultured as in (D). The reduced and not reduced samples are indicated (+red and −red, respectively). Lysates (L), α-Flag IP eluted with Flag-peptide (E), eluted sample precipitated with streptavidin-beads [both the fraction unbound (U) and bound (B) to streptavidin-beads], were probed for APP in WB.

9. I. Purified β-secretase was incubated with fluorescent β-secretase substrate for 30 min, resulting in β-cleavage that could be detected by fluorescence increase. In separate samples, the indicated concentrations of N3-2A or β-secretase-inhibitor IV were added to the reaction. The data are shown as % of inhibition of β-secretase activity in samples without inhibitors.

10. J. Model depicting the mechanism of action of N3-2A/MoBA. The peptide interferes with processing of APP by β-secretase but, unlike full-length BRI2, does not modulate γ-cleavage of β-CTF.

Figure 3. MoBA and a β-secretase inhibitor rescue the LTP deficit of FDD_KI mice—a GSI does not

1. Sixty-minutes perfusion with MoBA reverses LTP impairment in FDD_KI mice [WT to FDD_KI: F(1,12) = 12.372, p = 0.004; WT to FDD_KI + MoBA 1 µM: F(1,12) = 0.012, p = 0.914; WT to FDD_KI + MoBA 10 nM: F(1,11) = 0.202, p = 0.662; FDD_KI to FDD_KI + MoBA 1 µM: F(1,12) = (10.078), p = 0.006; FDD_KI to FDD_KI + MoBA 10 nM: F(1,11) = 15.049, p = 0.008]. N6 does not rescue the LTP deficit [FDD_KI to FDD_KI + N6 1 µM: F(1,10) = 0.053, p = 0.821]. MoBA does not alter LTP of WT mice [WT to WT + MoBA 1 µM: F(1,12) = 0.361, p = 0.560].

2. β-secretase-inhibitor IV (50 nM; IC₅₀ = 15 nM) rescues LTP impairment in FDD_KI mice [FDD_KI to FDD_KI + β-secretase-inhibitor IV: F(1,14) = 12.258, p = 0.004; WT to FDD_KI: F(1,13) = 12.272, p = 0.004; WT to FDD_KI + β-secretase-inhibitor IV: F(1,13) = 0.604, p = 0.451]. There was a trend towards increased LTP in inhibitor IV-treated WT and FDD_KI samples versus vehicle-treated WT controls, but this difference was not statistically significant. Compound-E (1nM; IC₅₀ = 300/240pM) does not rescue the LTP defect in FDD_KI samples [FDD_KI to FDD_KI + compound-E: F(1,11) = 0.838, p = 0.380]. The β- and GSIs do not alter LTP of WT mice [WT to WT + β-secretase-inhibitor IV: F(1,10) = 0.413, p = 0.535; WT to WT + compound-E: F(1,11) = 0.041, p = 0.844].

3. Lysates from hippocampal slices treated with (+) or without (−) compound-E for 3 h, were analysed by WB for APP and CTFs. The bottom graph represents quantization of triplicate samples. The CTFs levels are expressed as a % of APP.

Figure 4. Inhibiting β-cleavage of APP rescue the memory deficit of FDD_KI mice

Mice were injected in the lateral ventricle with either 1 µl of PBS/100 µM β-secretase-inhibitor IV, 1 µl of PBS/300 nM compound-E, 1 µl of PBS/100 µM-MoBA or 1 µl of PBS/3 µM compound-E. Injections were performed 1 h prior to the training section and, the following day, 1 h before testing.

1. WT and FDD_KI mice spent the same amount of time exploring the two identical objects on day 1. As the mice develop habituation to the test, they tend to explore the objects more.

2. WT mice spent more time exploring the novel object 24 h later, showing normal object recognition (discriminatory ratio = 0.63), while FDD_KI mice present amnesia and do not distinguish the new object from the old one (discriminatory ratio = 0.5). β-secretase-inhibitor IV and MoBA transiently rescue this memory deficit, while GSI does not. The number of days between the day 2 of a test and day 1 of the following test are indicated (× d.).

3. Model depicting early pathogenic events preceding amyloidosis and leading to memory loss. It is unlikely that full-length APP is pathogenic since either decreasing (Tamayev et al, 2011) or increasing (MoBA and inhibitor IV) its levels prevents/rescues the deficits of FDD_KI mice. Two inhibitors of β-cleavage of APP (Inhibitor IV and MoBA), but not a GSI, rescue the LTP/memory deficits, suggesting that newly synthesized sAPPβ and/or β-CTF, but not Aβ/P3/AID cause these deficits in FDD_KI mice (+ and in black). Whether sAPPα and/or α-CTF are pathogenic remains to be determined (?).