Technical Comments

Comment on “ApoE-Directed Therapeutics Rapidly Clear β-Amyloid and Reverse Deficits in AD Mouse Models”

See allHide authors and affiliations

Science  24 May 2013:
Vol. 340, Issue 6135, pp. 924
DOI: 10.1126/science.1235809


Cramer et al. (Reports, 23 March 2012, p. 1503; published online 9 February 2012) demonstrated in a mouse model for Alzheimer’s disease (AD) that treatment of APP/PS1ΔE9 mice with bexarotene decreased Aβ pathology and ameliorated memory deficits. We confirm the reversal of memory deficits in APP/PS1ΔE9 mice expressing human APOE3 or APOE4 to the levels of their nontransgenic controls and the significant decrease of interstitial fluid Aβ, but not the effects on amyloid deposition.

The inheritance of apolipoprotein (APOE) ε4 allele is the only established genetic risk factor for late-onset Alzheimer’s disease (AD) (1, 2).

Liver X receptors α and β (LXRs) are transcription factors that act as heterodimers with retinoid X receptors (RXR) to activate the transcription of their target genes. LXR and RXR agonists were shown to ameliorate memory deficits and decrease β-amyloid (Aβ) in AD mouse models presumably through up-regulation of Abca1 and Apoe (36).

Cramer et al. (3) demonstrated that treatment with U.S. Food and Drug Administration (FDA)–approved RXR agonist bexarotene in APP/PS1ΔE9 mice significantly decreased interstitial fluid (ISF) Aβ, amyloid plaques, and soluble and insoluble Aβ in the brain. Importantly, they demonstrated a rapid reversal of cognitive, social, and olfactory deficits. The authors postulated that RXR activation stimulates normal Aβ clearance processes and concluded that bexarotene facilitates APOE-dependent clearance of soluble Aβ from ISF, which correlates to the improvement of behavior.

At the time the Report was available online, we had been working with APP/PS1ΔE9 mice expressing human APOE3 and APOE4 isoforms (referred to as APP/E3 and APP/E4 mice) (7). Mice (7 months old) were treated by oral gavage with 100 mg per kg of weight (mg/kg) of Targretin and controls with vehicle (0.2 mg/kg glycerol) for 15 days. Comparable numbers of male and female mice were used in all experiments. Cognitive deficits were evaluated by a radial arm water maze (RWM) task (7) on the 10th day of treatment and by novel object recognition (8) on the 12th day of treatment. The mice were perfused (7) after completion of the behavioral testing (15th day of treatment). One hemibrain was fixed, sectioned, and stained with X-34 and antibody 6E10 to Aβ (7). The other hemibrain was homogenized for Western blotting to extract Aβ for enzyme-linked immunosorbent assay (ELISA) (4, 9) or soluble oligomers for dot blotting (10, 11). For microdialysis experiments (4, 7), 3.4-month-old APP/E3 and APP/E4 mice were treated with 100 mg/kg bexarotene or vehicle for 48 hours, and ISF Aβ40 and Aβ42 were determined by ELISA (4, 7).

Bexarotene treatment did not affect weight or general behavior of treated mice compared with controls. Target engagement was confirmed by the increased protein levels of known RXR targets: ABCA1 (2-fold), APOE (1.3-fold), and APOA-I (1.5-fold) in brain and high-density lipoproteins in plasma.

RWM testing showed that bexarotene restored spatial memory deficits in both APP/E3 (Fig. 1A) and APP/E4 (Fig. 1B) mice; that is, there was no statistical difference between bexarotene-treated APP transgenic mice and their respective nontransgenic controls. Novel object recognition testing of long-term memory demonstrated that bexarotene restored memory deficits and neither APP/E3- nor APP/E4-treated mice differed from their nontransgenic controls (Fig. 1C). Bexarotene treatment was equally effective in both genders of APP/E3 and APP/E4 mice. We conclude that bexarotene significantly improves memory functions in APP mice expressing both APOE isoforms and restores cognitive performance to that of nontransgenic controls, which is in agreement with Cramer et al. (3).

Fig. 1

Bexarotene restores cognitive function and decreases ISF Aβ in APP/E3 and APP/E4 mice. (A to C) APP/E3 and APP/E4 mice (7 months old) and matched nontransgenic littermates (E3 and E4) were treated with bexarotene or vehicle. N = 8 to 13 mice per group. RWM (7) was used to assess bexarotene effect on spatial learning deficits in APP/E3 (A) and APP/E4 mice (B). Analysis by two-way repeated-measures analysis of variance (ANOVA) shows a significant effect on treatment (P < 0.0001) and training (P < 0.001) in both APOE isoforms. Tukey’s post hoc test shows that bexarotene-treated APP mice were significantly different from vehicle-treated APP mice (P < 0.05 for both APOE isoforms) but not from their nontransgenic controls. (C) Novel object recognition test was performed on the same mice after RWM. Analysis by one-way ANOVA and Tukey’s post hoc test shows that bexarotene-treated APP mice were significantly different from vehicle-treated APP mice but not from their nontransgenic controls. (D) Bexarotene significantly decreases Aβ level in ISF. APP/E3 and APP/E4 mice (3.4 months old) were treated with bexarotene for 48 hours, and in vivo microdialysis was performed in the hippocampus as in (7). ISF Aβ40 and Aβ42 were determined by ELISA. Analysis by t test; N = 5 mice per group.

Using microdialysis (Fig. 1D), we found that bexarotene treatment decreased ISF Aβ in APP/E3 and APP/E4 mice approximately by the same factor as Cramer et al. reported. Bexarotene decreased Aβ40 in APP/E3 mice by 23% and Aβ42 by 26% (compared with APP/E3-vehicle). The treatment also caused a statistically significant decrease of Aβ in APP/E4 mice (compared with APP/E4-vehicle): 17% for Aβ40 and 12% for Aβ42.

We visualized the compact fibrillar amyloid plaques with X-34 and did not find bexarotene effect in APOE3- or APOE4-expressing mice (Fig. 2A and representative picture below). Staining with 6E10 antibody did not show any significant difference in Aβ plaques between treated and control mice of both genotypes (Fig. 2B). Soluble Aβ was extracted from cortices and hippocampi by tris buffer followed by extraction of insoluble Aβ from the remaining pellets with formic acid, and Aβ levels were determined by ELISA (7, 9). The results (Fig. 2, C and D) demonstrate no effect of bexarotene on insoluble Aβ40 and Aβ42 in cortex and hippocampus. Soluble Aβ40 and Aβ42 were also unchanged in cortex (Fig. 2E) and hippocampus (not shown). Finally, we examined the level of soluble oligomers on dot blots using A11 antibody (11) and found that bexarotene decreased A11-positive oligomers in both APP/E3 and APP/E4 mice (Fig. 2F).

Fig. 2

Bexarotene treatment does not affect amyloid deposition. After the behavior tests, the level of amyloid pathology was compared between treated and control mice within each genotype. (A) X-34 staining of compact fibrillar amyloid plaques in cortex and hippocampus (7). Representative pictures for X-34 (20X) are shown on the right. (B) Results of staining of brain sections with antibody to Aβ. For (A) and (B), N = 8 to 12 mice per group. Insoluble Aβ40 and Aβ42 in cortex (C) and hippocampus (D), and soluble Aβ in cortex (E) were measured by ELISA (N = 9 to 17 mice per group). (F) Bexarotene significantly decreased A11-positive oligomers (4). Intensity of the dots was quantified and normalized on the total protein measured on dot blots stained with coomassie blue. N = 7 to 9 mice. For all panels, analysis is by t test.

In summary, our study shows that bexarotene significantly improved cognitive deficits in APP/PS1ΔE9 mice expressing human APOE3 and APOE4. We also found a significant decrease of ISF Aβ in both APOE isoforms. However, we could not confirm bexarotene’s effect on Aβ or amyloid plaques in cortex and hippocampus. It is possible that bexarotene exerts its effect on memory not by modifying amyloid plaques but by reducing soluble oligomers in the brain or by a non-Aβ related mechanism. Finally, it is important to note that the effect of bexarotene on memory was equally beneficial in mice expressing human APOE4 and APOE3 isoform. Regardless of the discrepancy, we consider the data relevant to a particular therapeutic approach in AD, targeting APOE at the transcriptional level with the possibility to modulate the LXR/RXR-ABCA1-APOE regulatory axis and functional aspects of APOE in the brain that have been revealed during the last decade (2, 1214).

References and Notes

  1. Acknowledgments: This work was supported by NIH grants R01AG037481, R01AG037919, R21ES021243, and F32AG034031 and by NIRG-12-242432 from the Alzheimer’s Association.. We are grateful for the excellent technical support of S. P. Kancherla and A. Carter. A11 antibody was provided by C. Glabe, University of California, Irvine, CA.

Stay Connected to Science

Navigate This Article