Days to criterion as an indicator of toxicity associated with human Alzheimer amyloid-beta oligomers

Abstract

Objective: Recent evidence suggests that high molecular weight soluble oligomeric Abeta (oAbeta) assemblies (also known as Abeta-derived diffusible ligands, or ADDLs) may represent a primary neurotoxic basis for cognitive failure in Alzheimer disease (AD). To date, most in vivo studies of oAbeta/ADDLs have involved injection of assemblies purified from the cerebrospinal fluid of human subjects with AD or from the conditioned media of Abeta-secreting cells into experimental animals. We sought to study the bioactivities of endogenously formed oAbeta/ADDLs generated in situ from the physiological processing of human amyloid precursor protein (APP) and presenitin1 (PS1) transgenes.

Methods: We produced and histologically characterized single transgenic mice overexpressing APP(E693Q) or APP(E693Q) X PS1DeltaE9 bigenic mice. APP(E693Q) mice were studied in the Morris water maze (MWM) task at 6 and 12 months of age. Following the second MWM evaluation, mice were sacrificed, and brains were assayed for Abetatotal, Abeta40, Abeta42, and oAbeta/ADDLs by enzyme-linked immunosorbent assay (ELISA) and were also histologically examined. Based on results from the oAbeta/ADDL ELISA, we assigned individual APP(E693Q) mice to either an undetectable oAbeta/ADDLs group or a readily detectable oAbeta/ADDLs group. A days to criterion (DTC) analysis was used to determine delays in acquisition of the MWM task.

Results: Both single transgenic and bigenic mice developed intraneuronal accumulation of APP/Abeta, although only APP(E693Q) X PS1Delta9 bigenic mice developed amyloid plaques. The APP(E693Q) mice did not develop amyloid plaques at any age studied, up to 30 months. APP(E693Q) mice were tested for spatial learning and memory, and only 12-month-old APP(E693Q) mice with readily detectable oAbeta/ADDLs displayed a significant delay in acquisition of the MWM task when compared to nontransgenic littermates.

Interpretation: These data suggest that cerebral oAbeta/ADDL assemblies generated in brain in situ from human APP transgenes may be associated with cognitive impairment. We propose that a DTC analysis may be a sensitive method for assessing the cognitive impact in mice of endogenously generated oligomeric human Abeta assemblies. ANN NEUROL 2010.

Figure 1. Immunoblot characterization of APP^E693Q single transgenic mice

(A) Both blots are from identical samples. Lane 1 is from a non-transgenic mouse brain. Lanes 2–7 are from the brains of 6 F1 generation mice from 6 unique APP^E693Q single transgenic mice. Levels of holoAPP expression were measured from each mouse based on levels of mature and immature APP as measured by pan-species anti-APP cytoplasmic tail pAb369 or human APP/Aβ-specific mAb6E10. Lane 5 represents an F1 mouse from the line of the APP^E693Q breeder mice used in this experiment, indicating a 5–8-fold overexpression of holoAPP in comparison to non-transgenic mice (Lane 1). Immunoreactivity to mAb6E10 (bottom) represents expression of the human APP transgene. (B) Immunoblot characterization of APP-CTFs in the brains of transgenic and non-transgenic mice is shown. In comparison to mice harboring the Swedish (APP^{K670N, M671L}) mutation (our laboratory’s unpublished Swedish mouse) or TgCRND8 (APP^{K670N, M671L, V717F}), courtesy of Dr. David Westaway, University of Toronto), the APP^E693Q mutation does not appear to alter cleavage by either α- or β-secretases, noted by the lack of C99-CTF (β-CTF) in comparison to both TgCRND8 and Tg2576 mice (Swedish APP, courtesy of Dr. Karen Hsiao) shown here.

Figure 2. Pathological analysis of CAA and amyloid deposition in APP^E693Q and APP^E693Q X PS1ΔE9 mice

(A) Cerebral amyloid angiopathy is represented by vascular pathology in APP^E693Q mice. A representative image of staining of amyloid-laden cerebral vessels in an APP^E693Q mouse (right) compared to a NTg (left) is shown here. (B) Representative Perls’ Berlin Blue stain of the brain of a 20-month old APP^E693Q mouse shows hemosiderin indicative of the local extravasion of blood (arrows). This degree of Perls’ positivity is likely due to a combination of both aging and CAA. (C) Immunostaining of the hippocampus of APP^E693Q X PS1ΔE9 bigenic mouse using mAb6E10 antibody, revealing the development of plaque pathology, which is not present in APP^E693Q single transgenic animals. (D) Intraneuronal APP/Aβ accumulation in APP^E693Q mice compared to Tg2576 mice represented by immunostaining with mAb6E10 and mAb4G8. Vesicular staining in the two murine lines was found to differ qualitatively and quantitatively, with APP^E693Q showing more discrete and more intense immunoreactivity compared to Tg2576 mice.

Figure 3. Immunoelectron microscopy reveals APP accumulation in intraneuronal organelles

Representative immunoelectron microscopy images of pAb369-immunopositive staining of small-, medium-, and especially large-sized cytoplasmic vesicular structures, including the internal and external lamellae of multivesicular bodies (MVBs) and external lamellae of dense lysosomes and lipofuscin. APP accumulation in MVBs and dense lysosomes has been previously associated with increased pathogenic processing of APP^,. These results suggest that APP^E693Q transgenic mice exhibit a large amount of intraneuronal accumulation of APP.

Figure 4. Behavioral characterization of APP^E693Q on the MWM task

Non-transgenic (n=8) and APP^E693Q single transgenic (n=17) mice were trained for 12 days and then tested on probe trials at 12 and 21 days-post training at 6 months of age, extinguished, then trained and tested again at 12 months of age. No significant differences were observed between NTg and APP^E693Q mice during training or probe test (either 12 or 21 days) at 6 months of age. (A) No significant differences were observed for percent of time in the target quadrant between 12-month old NTg and APP^E693Q mice at either 12 (gray; p=0.06) or 21 (purple; p=0.754) day probe trials. (B) During the 12-day training/acquisition period, NTg mice reached the escape platform with shorter escape latencies from day-to-day and with a low amount of variability among NTg mice, indicating acquisition of the task. (C) No difference was observed between NTg and APP^E693Q during the 12-day training period. In comparison to NTg littermates, APP^E693Q single transgenic mice displayed a large amount of intragenotype variability, especially in the later days of training. Levene’s test for equality of variances for NTg versus APP^E693Q mice during training revealed significantly non-homogeneous variances on day 8 [F(2,22)=5.208, p=0.014], day 10 [F(2,22]=3.634, p=0.043], day 11 [F(2,22)=3.428, p=0.05), and day 12 [F(2,22)=4.108, p=0.03]. Error bars: +/− S.E.M.

Figure 5. Aged APP^E693Q mice exhibited an oAβ/ADDL-dependent delay in acquisition of the MWM task

We tested the a priori hypothesis that cerebral oAβ/ADDL level accounts for cognitive deficits on the MWM task in non-plaque forming, pre-pathological APP^E693Q single transgenic mice. Post-mortem biochemical analysis of brain oAβ/ADDL concentration was used to group experimental mice as NTg (unable to make oAβ/ADDLs, n=8), or APP^E693Q with undetectable (below LLRQ; (ud)oAβ/ADDL, n=12) or detectable (above LLRQ; (d)oAβ/ADDL, n=5) ADDLs. (A) No significant difference was observed for percent of time in target quadrant between 12-month old NTg, (ud)oAβ/ADDL, or (d)oAβ/ADDL groups at the day 12 post-training probe trial, suggesting that APP^E693Q single transgenic mice do not have oAβ/ADDL-dependent long-term memory deficits. (B) Further analysis showed no significant difference between 12-month old NTg and (ud)oAβ/ADDL or (d)oAβ/ADDL mice on day 12 of training, indicating that all mice performed equally, on average, on the final day of training. (C) A repeated measures ANOVA revealed significant between-groups differences for escape latency [F(2,22)=6.005, p=0.008] and Dunnett’s T3 multiple comparisons analysis (homogeneity of variance not assumed) revealed a significant between-groups difference only between NTg and (d)oAβ/ADDL mice (p=0.027). A MANOVA was further utilized to determine the days on which between-subjects differences occurred, indicating a significant between-subjects effects on day 5 [F(2,22)=6.551, p=0.006], day 6 [F(2,22)=4.641, p=0.021], and day 9 [F(2,22)=11.730, p<0.001]. Bonferroni’s multiple comparisons analysis (homogeneity of variance assumed) indicated a significantly higher escape latency only for (d)oAβ/ADDL mice compared to NTg mice on day 6 (p=0.021), and also on day 5 and day 9 of training between (d)oAβ/ADDL mice and both NTg (p=0.010, p<0.001, respectively) and (ud)oAβ/ADDL (p=0.010, p=0.002, respectively) mice. (D) A days-to-criterion (DTC) analysis was utilized to more specifically assess the relationship between oAβ/ADDL level and acquisition of the MWM task. A criterion score for reliable acquisition of the MWM task was set to two consecutive trials with escape latencies of 25 seconds or less and each mouse received a score reflecting the day on which criterion was met. A oneway ANOVA revealed significant between-groups differences for DTC [F(2,22)=5.526, p=0.011] and Bonferroni’s multiple comparisons analysis (homogeneity of variance assumed) revealed a significant increase in DTC only for (d)oAβ/ADDL in comparison to NTg mice (p=0.01). Taken together, these results suggest that APP^E693Q mice do eventually acquire and retain the MWM task, however these mice exhibit a clear oAβ/ADDL-dependent delay in acquisition of the task in comparison to NTg mice. Error bars: +/− S.E.M; *p<0.05; **p<0.01 with a two-tailed α=0.05.