Modeling an anti-amyloid combination therapy for Alzheimer's disease

Abstract

As only symptomatic treatments are now available for Alzheimer's disease (AD), safe and effective mechanism-based therapies remain a great unmet need for patients with this neurodegenerative disease. Although gamma-secretase and BACE1 [beta-site beta-amyloid (Abeta) precursor protein (APP) cleaving enzyme 1] are well-recognized therapeutic targets for AD, untoward side effects associated with strong inhibition or reductions in amounts of these aspartyl proteases have raised concerns regarding their therapeutic potential. Although moderate decreases of either gamma-secretase or BACE1 are not associated with mechanism-based toxicities, they provide only modest benefits in reducing Abeta in the brains of APPswe/PS1DeltaE9 mice. Because the processing of APP to generate Abeta requires both gamma-secretase and BACE1, it is possible that moderate reductions of both enzymes would provide additive and significant protection against Abeta amyloidosis. Here, we test this hypothesis and assess the value of this novel anti-amyloid combination therapy in mutant mice. We demonstrate that genetic reductions of both BACE1 and gamma-secretase additively attenuate the amyloid burden and ameliorate cognitive deficits occurring in aged APPswe/PS1DeltaE9 animals. No evidence of mechanism-based toxicities was associated with such decreases in amounts of both enzymes. Thus, we propose that targeting both gamma-secretase and BACE1 may be an effective and safe treatment strategy for AD.

Fig. 1

Reduction of γ-secretase and BACE1 additively attenuates Aβ1–40 and Ab1–42 peptides and Aβ deposition. (A and B) ELISA assay of insoluble (formic acid extraction, pg/μl) Aβ1–40 and Aβ1–42 extracted from brain hemisphere of APPswe/PS1ΔE9 (AP), APPswe/PS1ΔE9;Aph-1a^+/− (APA), APPswe/PS1ΔE9;BACE1^+/− (APB), and APPswe/PS1ΔE9;Aph-1a^+/−;BACE1^+/− (APAB) mice at 6 months of age. (C and D) Average amyloid plaque number per section in the hippocampus and cerebral cortex of 6-month-old APPswe/PS1ΔE9, APPswe/PS1ΔE9;Aph-1a^+/−, APPswe/PS1ΔE9;BACE1^+/−, and APPswe/PS1ΔE9;Aph-1a^+/−;BACE1^+/− mice. Sagittal serial brain sections were selected at eight-section intervals for the analysis. *P < 0.05, **P < 0.01, ***P < 0.001. (E to H) Representative sagittal hippocampal sections immunostained with ubiquitin-specific antisera from 6-month-old APPswe/PS1ΔE9 (E), APPswe/PS1ΔE9;Aph-1a^+/− (F), APPswe/PS1ΔE9;BACE1^+/− (G), and APPswe/PS1ΔE9;Aph-1a^+/−;BACE1^+/− (H) mice. Scale bars, 200 μm.

Fig. 2

Reductions of both γ-secretase and BACE1 protect against Aβ amyloidosis in aged APPswe/PS1ΔE9 mice. (A to D) Representative sagittal hippocampal sections immunostained with ubiquitin-specific antisera from 23-month-old APPswe/PS1ΔE9 (A), APPswe/PS1ΔE9;Aph-1a^+/− (B), APPswe/PS1ΔE9;BACE1^+/− (C), and APPswe/PS1ΔE9;Aph-1a^+/−;BACE1^+/− (D) mice. (E and F) Unbiased stereology analysis of area fraction covered by amyloid plaques in the hippocampus (E) and cerebral cortex (F). (G) Average number of A11 immunoreactive clusters in hippocampus and cortex per brain section. *P < 0.05, **P < 0.01, ***P < 0.001. (H to K) Immunohistochemical staining with A11, an oligomer-specific antisera, of the hippocampus in APPswe/PS1ΔE9 (H and J) and APPswe/PS1ΔE9;Aph-1a^+/−;BACE1^+/− (I and K) mice. Magnifications, ×4 [(H) and (I)] and ×20 of dentate gyrus from hippocampal section from (H) and (I) defined by the rectangular box. Scale bars, 200 μm [(H) and (I)], 50 μm [(J) and (K)].

Fig. 3

Reductions of both γ-secretase and BACE1 rescue memory deficits in aged APPswe/PS1ΔE9 mice. (A to F) Testing of spatial learning [(A) and (B)] and reference memory [(C) to (F)] in the classic Morris water maze. (A) Distances to the platform are shown as averages for consecutive blocks of five training trials for wild-type control (WT; n = 10) and APPswe/PS1ΔE9 mice (n = 14). (B) Distance to the platform for groups of APPswe/PS1ΔE9 (n = 8), APPswe/PS1ΔE9;Aph-1a^+/− (n = 9), APPswe/PS1ΔE9;BACE1^+/− (n = 6), and APPswe/PS1ΔE9;Aph-1a^+/−; BACE1^+/− (n = 6) mice. These four groups of mice were tested at the same age and in the same strain background as mice shown in (A). (C) Representative examples of swimming patterns for each of the genotypes (n = 6 mice per genotype) are shown as bivariate histograms of coordinates of the animal location recorded every 0.1 s during the probe trial. The most frequently visited areas are indicated by orange and red. An insert shows a scheme of the water maze with a position of start location, platform location, and areas of 20 and 40 cm around the platform. [(D) to (F)] Accuracy of spatial reference memory assessed in the probe trial conducted 20 hours after the last training trial. Measures of spatial memory are presented in accordance with increasing demands for accuracy: Spatial preferences for the area of 40 cm (D) and 20 cm (E) in diameter around the platform are followed by the frequency of crossings (number of crossings per 10 s) through the exact platform location (10 cm in diameter) (F). The higher scores in the indices of probe trials reflect a better performance. *P < 0.05, **P < 0.01, significant between-group differences in the measures for the correct quadrant; #P < 0.05, significant differences between the measures taken for the correct and opposite quadrants within the same genotype.

Fig. 4

Aph-1a^+/−;BACE1^+/− mice show no emotional or cognitive deficits. (A) Spontaneous arm alternation (%) in the working memory Y-maze task is shown for wildtype control, Aph-1a^+/− (A), BACE1^+/− (B), and Aph-1a^+/−;BACE1^+/− (AB) mice. (B and C) Novelty-induced exploration tested in the open-field task. (B) Dynamic of motor activity in the open field during consecutive 3-min blocks of testing. (C) Total activity accumulated throughout a whole period of testing. (D and E) Testing the anxiety levels in the plus maze. (D) Percentage of visits to the open arms of the plus maze. (E) Total number of arms visited during the test was used as a measure of motor activity in this task. *P < 0.05, significant difference relative to the wild-type control mice. (F) Psychostimulant-induced hyperactivity was tested after the intraperitoneal injection of MK-801 (0.3 mg/kg). Mice were first habituated for 45 min to a novel cage. After the MK-801 injection, activity was monitored for 60 min more. (G to I) PPI of ASR. (G) Amplitude of startle reaction to startle stimuli of different intensities. (H) Latency of startle reaction averaged for all trials with 110 dB. (I) PPI of ASR in four types of trials: with two levels of prepulses (8 or 16 dB above background) and two levels of startle stimuli (110 or 120 dB). The data shown are for mice tested at 5 to 6 months of age and are expressed as the mean ± SEM. An additional cohort of 12- to 15-month-old mice was tested in the Y maze, open field, and plus maze with no significant effect of genotype (data not shown).

Fig. 5

BACE1^+/− and Aph-1a^+/−;BACE1^+/− mice exhibit mild hypomyelination. (A) Representative examples of electron micrographs of sections of optic and sciatic nerves from 3-month-old wild-type control mice, Aph-1a^+/−, BACE1^+/−, and Aph-1a^+/−;BACE1^+/− mice. Scale bars, 1 mm (optic nerves), 4 μm (sciatic nerves). (B and C) Scatterplots of axonal diameters are shown against the g ratios of myelinated axons from the optic (B) and sciatic (C) nerves. (D and E) Quantitative analyses of genotype-related differences in the g ratio for the optic (D) and sciatic (E) nerves (n = 6 mice per genotype, at least 100 axons per nerve per mouse). For the optic nerve, significantly higher g ratios were observed in BACE1^+/− (P < 0.001) and Aph-1a^+/−;BACE1^+/− (P < 0.001) mice relative to that of control mice (D). (E) For the sciatic nerve, g ratios of Aph-1a^+/− (P < 0.01), BACE1^+/− (P < 0.001), and Aph-1a^+/−;BACE1^+/− (P < 0.001) were higher than that of control mice.

Fig. 6

Aph-1a^+/−;BACE1^+/− mice exhibit normal life span and organ histology. (A) Kaplan-Meier survival plots for Aph-1a^+/− [n = 52, median life expectancy (MLE) = 95 weeks], BACE1^+/− (n = 61, MLE = 105 weeks), Aph-1a^+/−;BACE1^+/− mice (n = 66, MLE = 104 weeks), and wild-type control mice (n = 67, MLE = 103 weeks). No significant differences in life span were found between any of these genotypes. (B) Average body weight in Aph-1a^+/− (n = 12), BACE1^+/− (n = 27), and Aph-1a^+/−;BACE1^+/− (n = 23) mice is shown as percentage of average body weight in wild-type control (n = 23) mice. No significant differences in body weight were observed. (C) Average spleen weight in grams. Splenomegaly (>3 SDs from the average spleen size) was rare and observed only in one animal in each of the wild-type control, Aph-1a^+/−, and Aph-1a^+/−;BACE1^+/− groups. (D) Average liver weight in grams. The incidents of hepatomegaly (>3 SDs from the average liver size) increased as aging progressed; however, no significant differences in liver size were found between different genotypes. An individual case of 91-week-old wild-type control male [denoted in (C) and (D) with asterisks] represented with splenomegaly and hepatomegaly accompanied by metastases in multiple organs (fig. S4). (E to L) H&E staining of spleen [(E), (G), (I), and (K)] and liver [(F), (H), (J), and (L)]. No histopathology was found in wild-type control or Aph-1a^+/−;BACE1^+/− mice. Scale bars, 100 μm [(E) to (H)], 50 μm [(I) to (L)].