Survival of cultured neurons from amyloid precursor protein knock-out mice against Alzheimer's amyloid-beta toxicity and oxidative stress

Abstract

Studies on the amyloid precursor protein (APP) have suggested that it may be neuroprotective against amyloid-beta (Abeta) toxicity and oxidative stress. However, these findings have been obtained from either transfection of cell lines and mice that overexpress human APP isoforms or pretreatment of APP-expressing primary neurons with exogenous soluble APP. The neuroprotective role of endogenously expressed APP in neurons exposed to Abeta or oxidative stress has not been determined. This was investigated using primary cortical and cerebellar neuronal cultures established from APP knock-out (APP-/-) and wild-type (APP+/+) mice. Differences in susceptibility to Abeta toxicity or oxidative stress were not found between APP-/- and APP+/+ neurons. This observation may reflect the expression of the amyloid precursor-like proteins 1 and 2 (APLP1 and APLP2) molecules and supports the theory that APP and the APLPs may have similar functional activities. Increased expression of cell-associated APLP2, but not APLP1, was detected in Abeta-treated APP-/- and APP+/+ cultures but not in H2O2-treated cultures. This suggests that the Abeta toxicity pathway differs from other general forms of oxidative stress. These findings show that Abeta toxicity does not require an interaction of the Abeta peptide with the parental molecule (APP) and is therefore distinct from prion protein neurotoxicity that is dependent on the expression of the parental cellular prion protein.

Fig. 1.

APP^−/− and APP^+/+ cortical neurons do not have differences in susceptibility to Aβ25–35 or Aβ1–42 inhibition of MTT reduction. Primary cortical neurons were grown at (A) high density (450,000 cells/cm²) or (B) low density (250,000 cells/cm²) for 3 d and exposed to Aβ25–35 for an additional 4 d. No differences between MTT reduction were observed between APP^−/− and APP^+/+ cortical neurons exposed to Aβ25–35 at either density. Treatment of cultures with 10 ng/ml bFGF (applied concomitantly with Aβ) resulted in a significant increase in cell viability as compared with non-bFGF-treated cultures when measured 4 d after exposure to Aβ25–35. *p < 0.05, **p < 0.01: differences in MTT reduction between bFGF and non-bFGF-treated cultures were determined using ANOVA and Newman–Keuls tests. C, APP^−/− and APP^+/+ cortical neurons reveal no differences in MTT reduction when treated with Aβ1–42. D, APP^−/− and APP^+/+ cortical neurons have no differences in susceptibility to Aβ25–35-induced cell death as determined using the LDH assay. E, APP^−/− and APP^+/+ cortical neurons reveal no differences in survival when exposed to Aβ1–42-induced cell death as determined using the LDH assay.F, APP^−/− and APP^+/+ cerebellar granule neurons reveal no differences in susceptibility to Aβ25–35 inhibition of MTT reduction. Primary cerebellar neurons were grown for 1 d and exposed to Aβ25–35 for an additional 6 d.

Fig. 2.

APP^−/− and APP^+/+ neurons do not have differences in susceptibility to intracellular- or extracellular-generated oxidative stress. A, Primary cortical neurons were grown for 2 d and exposed to glutamate for 24 hr. B, Primary cortical cultures were grown for 4 d and then exposed to H₂O₂ for 24 hr. C, Primary cerebellar granule neurons were grown for 7 d and then exposed to glutamate for 30 min. D, Primary cortical cultures were grown for 14 d and then exposed to glutamate for 30 min. Cell viability was determined 24 hr later. E, Primary cortical neurons were exposed to increasing concentrations of xanthine oxidase and 50 μm xanthine for 24 hr at either 4 or 6 d in vitro. F, Primary cortical neurons were grown for 14 d before incubation in glucose-free Locke’s media. Cell viability was determined after the given incubation period.

Fig. 3.

Characterization of the specificity of the APLP2 and APLP1 antibodies. Western blots of recombinant sAPP751 (lane 1), sAPLP2 (lane 2), and sAPLP1 (lane 3) probed with 22C11 (anti-APP/APLP2, 1:2000), 95/11 (anti-APLP2, 1:1000), or 25104 (anti-APLP1, 1:1000). The lower bands correspond to breakdown products as described previously (Henry et al., 1997). The position of the molecular weight markers is indicated on the left.

Fig. 4.

Quantitative immunoblotting of cell-associated APP, APLP1, and APLP2 in neurons exposed to Aβ25–35 and H₂O₂. Primary cortical neurons were grown for 2 d and then exposed to either 10 μmAβ25–35(Aβ) or 25 μmH₂O₂, or were untreated [control (C)] for 24 hr. The antibodies are anti-APP/APLP2 (22C11, 1:2000), anti-APLP2 (95/11, 1:1000), or anti-APLP1 (25104, 1:1000). The brackets correspond to the proteins described in Results and their molecular weights are as follows: anti-APP (95–105), anti-APLP1 (87 and 126), and anti-APLP2 (110 kDa). The position of the molecular weight markers is indicated on theright-hand side. A, Analysis of APLP2, APLP1, and APP expression under basal conditions. B, Analysis of APP expression detected in Aβ25–35-treated APP^+/+ cultures. C, Analysis of APLP2 expression in Aβ25–35-treated APP^−/− and APP^+/+ neurons shows a significant increase in APLP2 expression in both APP^−/− and APP^+/+ neurons exposed to Aβ. D, Analysis of APLP1 expression in Aβ25–35-treated APP^−/− or APP^+/+ neurons.E, Analysis of APP expression in APP^+/+ neurons in response to H₂O₂. F, Analysis of APLP2 expression in H₂O₂-treated APP^−/− and APP^+/+ neurons.G, Analysis of APLP1 expression in H₂O₂-treated APP^−/− and APP^+/+ neurons.