Missing LDL receptor increases amyloid buildup and lowers immune cell activity in an Alzheimer's mouse model

To elucidate the effect of LDLR deficiency on the AD-like phenotype we used 4 months old female 5XFAD transgenic mice as male 5XFAD mice of the same age displayed a significant delay in the development of the amyloid pathology. A similar sex-specific effect on amyloid deposition has been reported on other AD mice [27]. Our analysis showed that 5XFAD/LDLR-/- mice showed a significant increase in Thioflavine-S positive amyloid deposits compared to the 5XFAD mice both in the hippocampus (Fig. 1B upper photos) and the cortex (Fig. 1A upper photos) (n = 7). Deletion of ApoE greatly reduced Thioflavine-S positive amyloid plaque formation in 5XFAD/ApoE-/- transgenic mice, with amyloid plaques restricted mainly at the subiculum (Fig. 1A and B lower photos). 5XFAD/ApoE-/-LDLR-/- double knock-out mice displayed a significant increase in amyloid plaque deposition, compared to the 5XFAD/ApoE-/- mice in the hippocampus (Fig. 1C) (n = 5−7). Quantification of the Thioflavine-S positive amyloid plaque load in the hippocampus and the cortex in these mice confirmed a significant increase in the absence of LDLR both in the presence and in the absence of ApoE (Fig. 1D). These data confirm that the absence of LDLR in the 5XFAD mouse model results in the increase of the amyloid plaque formation independently of ApoE.

To confirm the effect of LDLR deficiency on ApoE levels in the brains of the 5XFAD/LDLR-/- and 5XFAD control mice we extracted and immunoblotted the plaque-associated fraction from brain homogenates of the 5XFAD and 5XFAD/LDL-/- mice (Fig. 2A). We found a significant increase in the brain levels of ApoE in the 5XFAD/LDLR -/- mice compared to the 5XFAD mice (Fig. 2B), as it has been previously reported [20], [25].

To evaluate the effect of LDLR deletion on APP processing, we analyzed protein extracts from mouse brains. Western- blot analysis of total brain protein extracts for the full-length APP and its proteolytic fragments (α- and β-CTFs) did not detect any differences among the different genotypes (Fig. 2C). These data support that the LDLR deletion has no effect on the levels of APP or CTFs in the brains of the 5XFAD transgenic mice.

Consistent with the increase in the Thioflavine-S positive amyloid deposits, 5XFAD/LDLR-/- and 5XFAD/ApoE-/-LDLR-/- transgenic mice displayed increased Aβ deposition detected by 6E10 antibody. Mouse brains of 4 months old 5XFAD/LDLR-/- mice had increased total Aβ immunoreactivity in the hippocampus and the cortex compared to 5XFAD control mice (Fig. 3A upper photos) (n = 7). This increase was statistically significant only in the hippocampus and not in the cortex (Fig. 3B). The pattern of Aβ-deposition in the 5XFAD and the 5XFAD/LDLR-/- mice was similar with more intense deposition in the subiculum and in the molecular layer of the dentate gyrus of the hippocampus. A significant increase in Aβ immunoreactivity was also noticed in the hippocampus of the 5XFAD/ApoE-/-LDLR-/- mice compared to the 5XFAD/ApoE-/- mice (n = 5−7) (Fig. 3A lower photos and 3B left graph). Also the pattern of Aβ-deposition was similar with a characteristic lack of deposition in the molecular layer of the dentate gyrus of the hippocampus both in the 5XFAD/ApoE-/- and the 5XFAD/ApoE-/-LDLR-/- mice. Comparison of Aβ-deposition in the cortices of the 5XFAD/ApoE-/- and the 5XFAD/ApoE-/-LDLR-/- mice showed a non significant increase in the 5XFAD/ApoE-/- brains (Fig. 3B right graph). This could be explained by the different pattern of Aβ deposits between the two genotypes in the cortex. 5XFAD/ApoE-/- mice developed a more ‘diffuse’ pattern covering a larger area, compared to the 5XFAD/ApoE-/-LDLR-/- mice that showed a more ‘compact’ pattern covering a smaller area (Fig. 3A lower photos).

Analysis of Aβ₄₀ and Aβ₄₂ levels by ELISA between 5XFAD and 5XFAD/LDLR-/- mice showed an increase both in the guanidine and lysis fractions in the 5XFAD/LDLR-/- brains, reflecting the increase in Thioflavine-S deposition observed in the 5XFAD/LDLR-/- mice. The same analysis between the 5XFAD/ApoE-/- and the 5XFAD/ApoE-/-LDLR-/- mice showed an increase in both Aβ₄₀ and Aβ₄₂ levels in the guanidine extract.

This increase of Aβ₄₀ and Aβ₄₂ levels observed mainly in the guanidine fraction suggests that LDLR deficiency promotes fibrillar Aβ formation (Fig. 3C.)

5XFAD mice have been shown to have extended neuroinflammatory responses displaying a significant increase in astrocytic and microglial response that is associated with age and amyloid deposition [28]. To evaluate the effect of LDLR deficiency on the glial response, astrocytic and microglial, we analysed female 4 month old 5XFAD and 5XFAD/LDLR-/- mice as well as 5XFAD/ApoE-/- and 5XFAD/ ApoE-/-LDLR-/- mice. Despite the increase in the amyloid deposition that was expected to further increase the glial response, in the 5XFAD/LDLR-/- mice GFAP (astrocytic) and Iba-1 (microglial) staining was reduced compared to 5XFAD littermates of the same gender and age. A similar effect was exerted in the 5XFAD/ApoE-/-LDLR-/- mice where astrocytic and microglial response was reduced compared to the 5XFAD/ApoE-/- (Figs. 4A,B,C and 5A,B,C). This decrease of GFAP and Iba1 staining was localized in the subiculum of the hippocampus (Figs. 4A and 5A) as well as in the cortex where it was more profound (Figs. 4B and 5B). In the thalamus of the 5XFAD/LDLR-/- mice we also noticed the formation of Iba-1 positive round cell formations surrounding Thioflavine-S positive amyloid deposits not observed in any other genotype (Fig. 5C).

These results indicate that the LDLR deficiency is affecting the neuroinflammatory response in the 5XFAD mouse model. Our results also suggest that this effect is not mediated by ApoE.

LDLR-/-, ApoE-/- and 5XFAD transgenic mice were obtained from The Jackson Laboratory, Bar Harbor, ME. ApoE-/- and 5XFAD mice were on the C57Bl6/J background. LDLR-/- were on C57Bl6/J;129S2 mixed genetic background and have been backcrossed for at least six generations to C57Bl6/J background. All test and control mouse groups analyzed for Aβ related pathology were 4 months old. Only female mice were analyzed as female 5XFAD transgenic mice develop the amyloid related phenotype earlier compared to males of the same age. To generate 5XFAD/LDLR-/- mice, 5XFAD mice were mated with LDLR-/- mice and the resulting 5XFAD/LDLR+/- mice were mated with the LDLR-/- mice. 5XFAD/LDLR-/- were identified by PCR. The same strategy was used to generate 5XFAD/ApoE-/- mice. 5XFAD mice were mated with ApoE-/- mice and the resulting 5XFAD/ApoE+/- mice were mated with the ApoE-/- mice. 5XFAD/ApoE-/- mice were identified by PCR. To generate ApoE-/-LDLR-/- mice, ApoE-/- mice were mated to LDLR-/- mice. The ApoE+/-LDLR+/- that resulted from this cross were mated with each other and ApoE-/-LDLR-/- mice were identified by PCR. To generate 5XFAD/ApoE-/-LDLR-/- mice 5XFAD mice were mated with ApoE-/-LDLR-/- mice and the resulting 5XFAD/ApoE+/-LDLR+/- mice were mated with the ApoE-/-LDLR-/- mice. 5XFAD/ApoE-/-LDLR-/- mice were identified by PCR. All mice were maintained on a standard chow diet containing 5% fat. All animal procedures were approved by the Bioethical committee of the Biomedical Research Institute of the Academy of Athens and by the Prefecture of Athens, Greece. All animal experimentations were carried out in agreement with ethical recommendation of the European Communities Council Directive of 24 November 1986 (86/609/EEC). Standard PCR techniques were used to genotype the 5XFAD, ApoE and LDLR genotypes.

Mice were anesthetized and transcardially perfused with PBS 0.1 M pH 7.4. Tissues were harvested immediately after perfusion. Left hemibrains were snap-frozen in liquid nitrogen for protein analysis. Right hemibrains were immersion-fixed in 4% PFA in PBS for 48 hours and cryoprotected in 20% sucrose in PBS for immunohistochemistry.

Protein extraction from left hemibrains was performed in three consecutive steps in order to evaluate protein levels in PBS, Lysis buffer (containing 10% glycerol, 1% Triton X-100 and complete protease inhibitor (Roche Applied Science) in PBS) and Guanidine fractions (5 M Guanidine-HCl) respectively as described before [21]. Brain tissue from all animals was extracted in an identical manner, and all fractions were immediately frozen at -80°C until analysis. Protein concentrations were determined by BCA Protein Assay (Pierce).

Left hemibrains tissues were homogenised as designated above and 30 µg of total protein were separated in a 10% SDS-PAGE Tris-Glycine for ApoE and Tubulin. In brief, after electrophoresis, proteins were transferred to a PVDF membrane (Immobilon-P, Millipore), blocked for 1 hour at room temperature with 5% milk in TBS-Tween 0.05% (TBST) and incubated for 1 hour at room temperature with goat anti-ApoE (M20, 1∶2000, Santa Cruz) or mouse anti-Tubulin (3D10, 1∶500, Sigma) in blocking solution. After washes with TBST blots were incubated with appropriate secondary antibodies (Santa Cruz) for 1 hour. For APP western blot 10 µg of total protein were separated in a 12% Tris-Tricine gel and transferred to nitrocellulose membrane (Protran, Whatman). Membranes were blocked for 30 minutes at 37°C in 5% FCS and incubated overnight at 4°C with rabbit polyclonal anti-human APP 1∶2500 [R1(57), a kind gift by Dr. P. Mehta, Institute for basic research in developmental disabilities, Staten Island, NY] [40]. The following day the membrane was washed 3 times with TBST and incubated for 1 hour at room temperature with goat-anti-rabbit HRP-conjugated (Santa Cruz). Blots were developed using ECL (Amersham) according to the manufacturer's recommendations. Bands were quantitated by densitometry using NIH Image J software.

Brains were fixed in 4% PFA in PBS for 48 hrs and then were cryoprotected in 20% sucrose in PBS (0.1 M PBS, pH 7.4). Fixed hemibrains were cut in 40-µm sagittal floating sections from the genu of the corpus callosum to the most caudal hippocampus using a vibratome (Leica VT1000S). For total Aβ immunostaining 6–7 sagittal sections 240 µm apart from each other spanning all the hippocampal formation were chosen. Sections were washed with TBS-0.1% Triton X-100 3 times for 10 minutes each, incubated for 30 min with 0.3% H₂O₂ in TBS and washed with TBS 3 times. Antigen retrieval was performed with 98% formic acid for 5 min, followed by 3 washes with TBS. Sections were transferred into blocking solution containing 15% normal donkey serum (Vector Laboratories) in TBS-Triton X-100 0.1% for 1 hour and incubated overnight at 4°C in 5% normal donkey serum (Vector Laboratories) in TBS-Triton X-100 0.1% containing the monoclonal biotinylated 6E10 primary antibody (1∶1000, Signet). Sections were then washed five times with TBS-Triton X-100 0.1%, and incubation in avidin-biotinylated horseradish peroxidase complex (ABC Elite, Vector Laboratories) followed for 90 min at room temperature. Peroxidase labelling was visualized with DAB/Ni (Vector). After a 1-min incubation period, sections were washed, mounted on glass slides, dehydrated in increasing ethanol concentrations from 50 to 100% followed by xylene and coverslipped with DPX (BDH).

The specificity of immunoreactivity was confirmed both by omitting the primary antibody and by the lack of signal when applying the same protocol on brain sections of wild-type littermates.

For the estimation of amyloid plaque load in transgenic mice, 6–7 sagittal sections 240 µm apart from each other spanning all the hippocampal formation were chosen for Thioflavine-S staining. Sections were incubated for 8 minutes in aqueous solution of Thioflavine-S (1% w/v) and then differentiated 2 times with 80% Ethanol for 3 minutes each, followed by another 3-minutes wash with 95% Ethanol. Sections were rinsed three times with ddH₂O and coverslipped with Vectashield fluorescence mounting medium (Vector). Imaging for Thioflavine-S was performed on a Leica DMRA 2 microscope (magnification 2.5x for cortices and 5x for hippocampi) and image captivation was performed using Leica Application Suite, version 2.8.1 software. For the analysis of the inflammatory response in the brains of mice, the GFAP and Iba1 markers for astrocytes and microglia respectively were used. For GFAP and Thioflavine-S, free-floating sections were washed 3 times with 1×TBS pH 7.6, blocked with 1×TBS-Triton X-100 0.4%, 5% Normal Goat Serum (Vector) for 1 hour at room temperature and incubated overnight at 4°C with rabbit polyclonal anti-GFAP (1∶1000, Sigma). The following day, sections were washed five times with 1×TBS and incubated in blocking solution with goat anti-rabbit Cy3-conjugated (1∶300, Jackson ImmunoResearch) for 1 hour at room temperature. Sections were washed with 1×TBS and immersed in Thioflavine-S (1% w/v aqueous solution) for 5 minutes. Sections were differentiated twice with 70% Ethanol and rinsed 3 times with PBS (0,1 M pH 7.4). Sections were mounted on glass slides and coversliped with Vectashield fluorescence mounting medium (Vector). For Iba1 and Thioflavine-S immunofluorescence sections were permeabilized for 30 minutes with PBS-Triton X-100 0.1% at room temperature, blocked for 3 hours with 10% fetal calf serum (FCS), 1% BSA (fraction V, Sigma) in PBS-Triton X-100 0.1% and incubated overnight at 4°C with rabbit anti-Iba1 (1∶300, Wako) in 1% FCS, 0.1% BSA, PBS-Triton X-100 0.01%. The following day, sections were washed 5 times with PBS and incubated with goat anti-rabbit Cy3-conjugated (1∶300, Jackson Immunoresearch) for 1 hour at room temperature. Sections were washed 5 times with 1×TBS and stained for Thioflavine-S as described above. Imaging was performed with a Leica TCS SP5 Confocal microscope and pictures were captured and analyzed with the Leica LAS AF Suite.

Images were analysed with the NIH Image J software. For the Aβ or Thioflavine-S load the images were thresholded within a linear range and the load was defined as the per cent area covered by Aβ or Thioflavin-S positive plaques (% Aβ or Thioflavin-S load). For Iba1 and GFAP the mean intensity of the immunofluorescent signal within a linear range was analysed. All data are expressed as mean±SEM. Statistical significance of differences was evaluated either with Student's t-test or with one way ANOVA followed by the two-way Student's t-test. In the t-test analyses, Welch's correction for unequal variances was applied when variances were significantly different between groups. Probability values P<0.05 were considered significant. All statistical analyses were performed using GraphPad Prism (version 4.0; Graphpad software for Science Inc., San Diego).