Low-Dose Radiation Therapy Reduces Amyloid Load in Young 3xTg-AD Mice

Abstract

Background:

Low-dose radiation therapy (LD-RT) has been shown to decrease amyloidosis or inflammation in systemic diseases and has recently been proposed as possible treatment of Alzheimer’s disease (AD). A positive effect of LD-RT on tauopathy, the other marker of AD, has also been suggested. These effects have been shown in preclinical studies, but their mechanisms are still not well understood.

Objective:

This study aimed to evaluate if anti-amyloid and anti-inflammatory effects of LD-RT can be observed at an early stage of the disease. Its impact on tauopathy and behavioral alterations was also investigated.

Methods:

The whole brain of 12-month-old 3xTg-AD mice was irradiated with 10 Gy in 5 daily fractions of 2 Gy. Mice underwent behavioral tests before and 8 weeks post treatment. Amyloid load, tauopathy, and neuroinflammation were measured using histology and/or ELISA.

Results:

Compared with wild-type animals, 3xTg-AD mice showed a moderate amyloid and tau pathology restricted to the hippocampus, a glial reactivity restricted to the proximity of amyloid plaques. LD-RT significantly reduced Aβ₄₂ aggregated forms (–71%) in the hippocampus and tended to reduce other forms in the hippocampus and frontal cortex but did not affect tauopathy or cognitive performance. A trend for neuroinflammation markers reduction was also observed.

Conclusion:

When applied at an early stage, LD-RT reduced amyloid load and possibly neuroinflammation markers, with no impact on tauopathy. The long-term persistence of these beneficial effects of LD-RT should be evaluated in future studies.

Keywords

Alzheimer’s disease amyloid neuroinflammation radiation therapy tau

INTRODUCTION

Alzheimer’s disease (AD) is a neurodegenerative disease characterized by memory and cognitive deficits, ultimately leading to loss of autonomy. The amyloid pathology has been the first hallmark described. Amyloid-β (Aβ) peptides are produced by the consecutive cleavage of β-secretase and γ-secretase leading to the production of soluble Aβ peptides, Aβ₄₀ and Aβ₄₂ being the pathological forms. Aβ peptides aggregate into oligomers, fibrils, and finally amyloid plaques [1]. The second hallmark consists in hyperphosphorylation of the tau protein that finally forms intracellular aggregates. Multiple sites of hyperphosphorylation have been identified and seem to be specific for AD [2, 3]. The third hallmark of AD, increasingly recognized, is the neuroinflammatory response of glial cells, namely astrocytes and microglia. Both cell types display numerous morphological, transcriptomic, and functional changes in response to the pathological environment [4 –7]. Classical markers are used to identify this response such as GFAP and Vimentin overexpression by reactive astrocytes and IBA1 and CD68 upregulation by activated microglia. Glial cells produce both pro- and anti-inflammatory cytokines and chemokines. These molecules are involved in the fine communications between astrocytes and microglia in the brain, but they also induce stress signals on neighboring cells, making difficult to interpret their beneficial or deleterious role in the pathogenesis [5, 8].

Numerous treatments have been developed to target one of these biological alterations, so far with limited success [1], highlighting the necessity to develop new therapeutic strategies. Radiation therapy (RT), one of the main treatments for cancer, has been recently studied in the context of AD. Indeed, it has been postulated that this treatment, applied at low doses, could achieve two important effects in AD: a decrease of amyloid load and a reduction of inflammation [9 –11]. These combined effects have been observed separately in some organ-specific amyloidosis or in the context of chronic inflammatory diseases (results summarized in [9]). Its specific effect in the brain has been explored only recently and the first results are highly encouraging, with a decrease of amyloid plaques and/or anti-inflammatory effects demonstrated in AD mouse models or in in vitro studies [12 –16]. Marples et al. showed that the regimen of 5 daily fractions of 2 Gy was the most efficient to reduce amyloid plaques in the brain of 30-week-old APP/PS1 mice (mild stage of the disease; 2, 4, or 8 weeks post treatment) [12]. This regimen was also studied in 3xTg-AD mice at 14 months of age (mild stage; 8 weeks post treatment) [14], in 5xFAD mice at 6 months (mild stage; 8 weeks post treatment) [13] or at 8–9 months (advanced stage; 5 weeks post treatment) [16]. The impact of low-dose RT (LD-RT) on amyloid plaques was evaluated using immunohistology (6E10 or Thioflavin S staining) in all studies. The anti-inflammatory effects was demonstrated using immunohistology (glial cell staining), multiplex ELISA (cytokine/chemokine quantification), or at the mRNA levels (cytokine/chemokine quantification) [12 , 16]. In a fifth study, authors showed an early effect of LD-RT on neuroinflammation with a reduction of glial reactivity by histology 4 days post LD-RT (1.8 Gy×5 fractions delivered daily) in 4-month-old 5xFAD mouse model [15]. Interestingly, a preliminary study even suggested that LD-RT could reduce the tauopathy in the 3xTg-AD model [14], but the mechanisms involved still have to be deciphered. Moreover, the stage of disease at the time of treatment is probably an important factor, as we recently showed that a treatment in advanced stages does not reduce amyloid pathology or neuroinflammation, despite a positive effect on cognition [17].

In this study, we evaluated the potential of LD-RT to modulate amyloid, tau, and neuroinflammation in the 3xTg-AD mouse model treated at an earlier stage than in the previous study using this model [14]. We evaluated the effect of LD-RT on amyloid, tau, and neuroinflammation by assessing markers of glial cell reactivity and cytokine/chemokine levels. For the first time, we also evaluated the impact of LD-RT on different forms of Aβ peptides in the brain.

MATERIAL AND METHODS

Animals

Triple transgenic AD mice (3xTg-AD) expressing human APP_swe and human Tau^P301L mutations under a Thy-1 promoter and a point mutation on the mouse Psen1 gene (PS1^M146V), on a mixed C57BL/6J×129Sv background were used. Wild-type (WT) littermates were used as controls. Only females were used as they develop a more aggressive pathology than males [18]. Mice were housed with food and water ad libitum, in a 12-h light-dark cycle. All experimental procedures were approved by the Ethics Committee for Animal Experimentation of the Canton of Geneva, Switzerland. Data are reported in accordance with Animal Research: Reporting In vivo Experiments (ARRIVE) guidelines.

Radiation treatment

Under anesthesia, the whole brain of 3xTg-AD mice (n = 8) was treated with 5 daily fractions of 2 Gy delivered using a 100 kV direct field from a TOPEX SRT-100™ superficial system (Sensus Healthcare, US). Sham-treated 3xTg-AD mice (n = 7) and WT mice (n = 8) were only anesthetized during 5 consecutive days. A customized lead shield was used to protect the eyes of animals from irradiation. Animals were treated at 11–12 months and postmortem analysis was performed 8 weeks after the end of treatment (Fig. 1A). The same groups of mice were used for all postmortem experiments.

Fig. 1

LD-RT reduces amyloid load in 3xTg-AD mice treated early. A) Experimental design of low-dose radiation treatment in the 3xTg-AD mice. Both hemispheres were treated with the following low-dose radiation treatment: 2 Gy×5 fractions delivered daily. B) Representative image of amyloid deposition (Methoxy-XO4, MXO4⁺, Blue) in the subiculum of a sham-treated and a treated animal, 8 weeks after RT treatment performed at 12 months (Left: scale bar = 250μm; Right: scale bar = 25μm). C) Percentage of the hippocampus positively stained with MXO4, representing the percentage of coverage by plaques. D) MXO4⁺ plaque density per μm² in the entire hippocampus. E) Mean size of MXO4⁺ amyloid plaques in the hippocampus. F) Concentration in Aβ oligomers in the Triton X100 (Tx)-soluble fraction from the hippocampus measured by ELISA. Concentrations in Aβ₄₀ in the Tx-soluble fraction (G) or guanidine (Gua)-soluble fraction (H) from the hippocampus or the frontal cortex. Concentrations in Aβ₄₂ in the Tx-soluble fraction (I) or guanidine (Gua)-soluble fraction (J) from the hippocampus or the frontal cortex. When two Y axes a represented, hippocampus values are reported on the left Y axis and the frontal cortex values referred to the right Y axis. ^**p < 0.01. RT, low-dose radiation therapy.

Behavior

3xTg-AD mice performed the following behavioral tests before and 8 weeks post treatment.

Open field

This test was used to assess the general locomotion of animals. Mice were placed at the center of the square (45×45 cm) and the general exploration was automatically recorded during 1 h and analyzed using Ethovision™ Software (Noldus, Wageningen, Netherlands). The total distance travelled during the 1 h was analyzed.

Elevated plus maze

This test, based on the natural curiosity of the animals, allows to evaluate anxiety-like behavior. Animals were placed at the center of the device (5×5cm) and free to explore the closed arms (20×5×30 cm) or the open arms (20×5 cm; anxiogenic areas) for 5 min. Movements of animals were automatically recorded and analyzed using Ethovision™ software. Total duration in closed and open arms or in the center was measured. The number of head dipping, another index to evaluate anxiety, were automatically calculated. Animals performing less than 3 entries in different areas were excluded.

Alternative Y maze

The device is composed by 3 arms (20×10×30 cm) freely accessible to animals for 5 min. The subsequent exploration of the three different arms constitutes a good alternation. The number of good alternations was automatically recorded and analyzed using Ethovision™ software and used as an index of working memory. Mice visiting less than 4 arms were excluded from the analysis.

Immunohistology

Animals were intracardiacally perfused with a saline solution under anesthesia (2% isoflurane). Half brains were then post-fixed in a 4% paraformaldehyde solution for 24 h and cryoprotect with a sucrose gradient (5 to 20%). Serial brain sections (30μm) were cut and stored in an anti-freeze solution at –20°C until use for immunostaining. The other hemisphere was used for biochemistry.

Immunofluorescence

Slices were rinsed 3×10 min in PBS0.1M and mounted on gelatin sliced and dried O/N at room temperature. After being rehydrated in PBS0.1M for 10 min, a methoxy-XO4 (MXO4; Tocris, Bristol, UK) plaque labelling was realized. For this, slices were incubated with 20μg/ml of MXO4 in PBS0.1M for 30 min at room temperature. After 3 washes, slices were incubated O/N at 4°C with the following antibodies: anti-GFAP-Cy3 (1/1,000, Sigma, St. Louis, MO), anti-IBA1 (1/600, Rabbit; Wako, Osaka, Japan), anti-Vimentin (1/1,000, Chicken; Abcam, Cambridge, UK), anti-CD68 (1/1,000, Rabbit; Invitrogen, Carlsbad, CA) diluted in 1% BSA/PBS0.1M/0.3% Triton X-100. After 3×5 min of washes, slices were incubated for 1 h at room temperature in 1% BSA/PBS0.1M/0.3% Triton X-100 with appropriate secondary Alexa-fluor conjugated antibodies (1/200, Invitrogen). Slices not stained with MXO4 were stained with DAPI for 10 min at room temperature. Slices were rinsed before being coverslipped with FluorosaveTM (Calbiochem, Darmstadt, Germany).

Immunohistochemistry

Slices were rinsed 3×10 min in PBS0.1M and mounted on gelatin slides and dried O/N at room temperature. After being rehydrated in PBS0.1M for 10 min, slices were incubated O/N at 4°C with an anti-HT7 antibody (1/1000, ThermoScientific, Waltham, MA) diluted in 1% BSA/PBS0.1M/0.3% Triton X-100. After 3 washes in PBS0.1M, slices were incubated for 1 h at room temperature in the secondary antibody goat anti-mouse-HRP (1/200, Dako, Troy, MI) in 1% BSA/PBS0.1M/0.3% Triton X-100. Slices were rinsed and revealed with a solution of 0.2 mg/ml DAB (Sigma) in PBS0.1M, containing 100μl/l H₂O₂ for 5 min at room temperature. Slices were mounted and let dry O/N. After dehydration, the coverslip was added using a mounting medium.

Image analysis

All images were acquired using an Axioscan.Z1® (Zeiss, Oberkochen, Germany) at 10x and analyzed using ImageJ software. For the MXO4 labelling quantification, a manual segmentation of the hippocampus was realized, and the percentage of the hippocampus occupied by plaques was measured using an intensity threshold. The plaque density (total number of amyloid plaques/μm²) and their individual size were measured using a semi-automatic detection of amyloid plaques with size and intensity threshold. The percentage positively stained with HT7⁺ tau was quantified in the entire hippocampus (3–5 slices per animal). For GFAP, Vimentin, IBA1 and CD68 staining, the mean grey value and the percentage of the area which was positively stained were quantified in the hippocampus, or in the subiculum, using two separated regions of interest (ROI) manually delineated (2–5 slices per animal). The background was subtracted for the analysis of the entire region.

Protein extraction

Half brains were dissected on ice and the hippocampus and frontal cortex were isolated, snap frozen in liquid nitrogen, and stored at –80°C before use. Both hippocampus and frontal cortex samples were homogenized by sonication in Triton X100 (Tx) lysis buffer [50 mM Tris-HCl pH = 7.4, 150 mM NaCl, 1% Triton X-100 with 1x protease and phosphatase inhibitors (Pierce, Waltham, MA); 300μl] centrifuged at 20,000 g for 20 min at 4°C. The supernatant contains Tx-soluble proteins. The pellet was resuspended in a Guanidine (Gua) lysis buffer [50 mM Tris-HCl pH = 8, 5 M Guanidine HCl with 1x protease and phosphatase inhibitors (Pierce); 180μl], incubated for 3 h on ice and centrifuged at 20,000 g for 20 min at 4°C. The supernatant contains Gua-soluble proteins. The total protein concentration was quantified using a BCA test (Pierce), as described by the manufacturer.

ELISA tests

Aβ measurements

Different amyloid peptide forms were quantified using ELISA kits: human Aβ40 ELISA kit (Life Technologies), human Aβ42 ELISA kit (Human Aβ42 Ultrasensitive ELISA Kit, Life Technologies), and the human Aβ oligomers ELISA kit (Amyloid-β Oligomers (82E1-specific) ELISA, IBL International, Hamburg, Germany). The following dilution of samples were used for Aβ₄₀ measurement: 1/10 for Tx-soluble fractions and 1/500 for Gua-soluble samples in diluent provided in the appropriate kit. For Aβ₄₂ measurement: 1/10 for Tx-soluble fractions and 1/10 Gua-soluble samples from the frontal cortex in diluent provided in the appropriate kit. An intermediate dilution of Gua-soluble hippocampal samples was realized (1/100) in the Gua extraction buffer before a second dilution at 1/500 in the diluent provide in the kit. For Aβ oligomers measurement: 1/2 in diluent provided in the kit for all samples. Manufacturer’s protocols were followed. Absorbances were measured using the EZ read 400 microplate reader (Biochrom, Cambourne, GB). Aβ concentrations were normalized to the total protein concentration. The absence of Aβ detection was validated in WT animals for each test.

Tau measurements

The Tau (Phospho) [pT231] Human ELISA kit (ThermoFisher) was used to quantify pT231-tau in Tx-soluble and Gua-soluble samples from both hippocampus and frontal cortex. Samples were diluted at 1/40 for Tx-soluble fraction and 1/10 for Gua-soluble fractions in diluent provided in the kit. Manufacturer’s protocols were followed. Absorbances were measured using the EZ read 400 microplate reader (Biochrom, Cambourne, GB). pT231-tau concentrations were normalized to the total protein concentration. The absence of pT231-tau detection was validated in WT animals.

Inflammation markers measurements

Tx-soluble samples were diluted at 1/2 in diluent provided in the Luminex multiplex assay kit (Cyto-kine & chemokine 22-plex mouse ProcartaPlex^TM Panel, ThermoFisher Scientific). Samples were incubated for 4h with beads, otherwise the protocol of the manufacturer was followed. Inflammation protein concentrations were measured using the MagPix instrument (Luminex, ThermoFisher Scientific) and analyzed using Procartaplex software (ThermoFisher Scientific). Concentrations were normalized to the total protein concentration.

Statistical analysis

A sample size analysis with the graphical Douglas Altman’s nomogram approach [19] was performed and significant data were reported if p≤0.05 and β< 0.2. Analysis was performed in blind conditions and normality of residues was assessed using the Shapiro-Wilks test. For two-group analyses, unpaired two-tailed student’s t-test or Mann-Whitney test were used for parametric and non-parametric data respectively. For three group comparisons, a one-way-ANOVA or Kruskal-Wallis tests were used for parametric and non-parametric data respectively. ELISA tests and behavioral results were analyzed using a two-way ANOVA (Brain region and treatment as between factors for ELISA; Group and Treatment as between factors for behavior) and uncorrected Fisher’s LSD post hoc test. Outliers were identified using the ROUT method (Maximal false discovery rate = 1%). All analyses were performed on Prism 8 (GraphPad, San Diego, CA). Results are presented as individual values and mean±SEM.

RESULTS

LD-RT reduces amyloid load in 3xTg-AD mice treated early

At the age of 14 months, 3xTg-AD mice showed only few amyloid plaques, and limited to the subiculum (Fig. 1B). The daily treatment with 2 Gy of X-ray radiation performed at 12 months did not influence the percentage of the hippocampus occupied by plaques (Fig. 1C), the plaque density (Fig. 1D), or the individual size of plaques (Fig. 1E). To go further, we measured in the hippocampus and in the frontal cortex the concentration in Aβ oligomers (Fig. 1F), Aβ₄₀ (Fig. 1G, H), and Aβ₄₂ (Fig. 1I, J) in the Tx-soluble fraction or in the Gua-soluble fraction, the last one corresponding to more aggregated forms of amyloid. Aβ oligomers were only detectable in the Tx-soluble fraction of the hippocampus and the concentration of Aβ₄₀ and Aβ₄₂ was significantly higher in the hippocampus than in the frontal cortex (Two-way ANOVA, Region effect: p < 0.01 for all forms). LD-RT significantly decreased Gua-soluble Aβ₄₂ (–71%; Two-way ANOVA: F_(1,9) = 5.424, p = 0.045 for treatment main effect; F_(1,9) = 17.72, p = 0.002 for region effect; F_(1,9) = 5.424, p = 0.045 for treatment×region interaction; LSD post-hoc test for treatment effect: p = 0.004) and tended to reduce Tx-soluble Aβ₄₀ in the frontal cortex without reaching significance (–67%; Two-way ANOVA: F_(1,9) = 4.071, p = 0.074 for treatment main effect). Globally, other Aβ peptide concentrations also tended to be reduced by LD-RT in both regions.

LD-RT does not influence the tauopathy in 3xTg-AD mice treated early

As shown in the Fig. 2A, the tauopathy was already present in the hippocampus of the 3xTgAD mice as compared to WT. However, LD-RT did not impact the percentage of the hippocampus positively stained with HT7 (Total tau, Fig. 2B). To go further, we quantified pT231-tau levels in the hippocampus and frontal cortex of 3xTg-AD mice. Same levels of pT231-tau were observed in both regions in the Tx-soluble fractions (Fig. 2C), but pT231-tau levels were significantly higher in the hippocampus than in the frontal cortex in the Gua-soluble fractions (Two-way ANOVA: F_(1,9) = 81.06, p < 0.0001 for region main effect) (Fig. 2D), suggesting that the tauopathy is globally more advanced in the hippocampus. LD-RT did not influence pT231-tau levels (Fig. 2C, D).

Fig. 2

No impact of LD-RT on Tau in 3xTg-AD mice treated early. A) Representative image of HT7⁺ cells in the hippocampus of WT, sham-treated and RT-3xTg-AD mice 8 weeks after treatment performed at 12 months. No staining was observed in WT mice. Scale bar = 250μm. Right panel: High magnification showing HT7⁺ neurons in the hippocampus of a 3xTg-AD mouse. B) Percentage of the hippocampus positively stained with HT7 (total Tau). C) Concentration in pT231-tau in the triton (Tx)-soluble from the hippocampus or the frontal cortex. D) Concentration in pT231-tau in the guanidine (Gua)-soluble fraction from the hippocampus (left Y axis) or the frontal cortex (right Y axis). RT, low-dose radiation therapy.

LD-RT tends to modulate neuroinflammation markers in 3xTg-AD mice treated early

Astrocyte and microglial reactivities were assessed in the entire hippocampus or specifically in the subiculum (Fig. 3A) using classical markers such as GFAP and Vimentin or IBA1 and CD68, respectively, known to be overexpressed in AD conditions. An overexpression of GFAP and Vimentin and a hypertrophy of primary processes of astrocytes were observed only around amyloid plaques (Fig. 3B). IBA1⁺ microglial cells also display an ameboid morphology, typical of microglial reactivity, and overexpressed CD68, exclusively around plaques (Fig. 3C). However, as mentioned previously, the number of amyloid plaques was very limited and restricted to the subiculum, suggesting only a slight glial activation in the 3xTg-AD mice at this age.

Fig. 3

LD-RT tends to reduce inflammation markers in 3xTg-AD mice treated early. A) Illustration of the regions of interest used for quantifications (Sub = Subiculum, Hipp = all hippocampus [Subiculum included]). Scale bar = 250μm. B) Representative confocal images of reactive astrocytes stained with anti-GFAP (red) and anti-Vimentin (magenta) antibodies around amyloid plaques (blue) in the subiculum of 3xTg-AD mice (Top panel). Scale bar = 25μm. Bottom panel: Confocal images illustrating morphological differences of astrocytes (GFAP⁺, red) distant vs close to plaques. Scale bar = 10μm. C) Representative confocal images of reactive microglial cells stained with anti-IBA1 (green) and anti-CD68 (grey) antibodies around amyloid plaques (blue) in the subiculum (Top panel). Scale bar = 25μm. Bottom panel: Confocal images illustrating morphological differences of microglia (IBA1⁺, green) distant vs close to plaques. Scale bar = 10μm. Percentage of the hippocampus positively stained or mean intensity of GFAP (D-E) or Vimentin (F-G), two proteins overexpressed by reactive astrocytes. Percentage of the hippocampus positively stained or mean intensity of IBA1 (H-I) or CD68 (J-K), two proteins overexpressed by reactive microglia. Percentage of the subiculum positively stained or mean intensity of GFAP (L-M), Vimentin (N-O), IBA1 (P-Q), or CD68 (R-S). RT, low-dose radiation therapy.

GFAP (Fig. 3D, E) or Vimentin expression (Fig. 3F, G) were not affected by LD-RT in the entire hippocampus. Same results were obtained for IBA1 (Fig. 3H, I) and CD68 (Fig. 3J, K), suggesting that LD-RT did not impact either astrocyte or microglial cells. Then, we restricted our analysis to the subiculum, where few amyloid plaques were present. In this area, GFAP expression was not impacted (Fig. 3L, M), but the treatment tended to reduce Vimentin levels (Fig. 3N, O; t(12) = 2.031, p = 0.065), suggesting a slight reduction of astrocyte response. However, microglial markers were not impacted (Fig. 3P-S).

To go further, the Tx-soluble fraction of the hippocampus was analyzed using a Luminex multiplex assay to measure 22 cytokines and chemokines (Fig. 4). Nine pro-inflammatory cytokines (Fig. 4A; IL-1β, IL-2, IL-5, Il-6, IL-18, MIP1β, IP-10, MCP-3, GM-CSF), 1 anti-inflammatory cytokine (Fig. 4B; Il-4), and 2 chemokines (Fig. 4C; Eotaxin, MIP-2) were detectable. However, at this age, no upregulation was observed in 3xTg-AD mice compared to WT mice, suggesting that there is not a clear neuroinflammatory response yet. LD-RT significantly reduced MIP-2 levels compared to WT animals. It also tended to reduce pro-inflammatory cytokines such as IL-1β (One-way ANOVA: F_(2,14) = 2.954, p = 0.09), IL-2 (Kruskall-Wallis test, p = 0.0854), IL-5 (Kruskall-Wallis test, p = 0.0753), and IL-6 (One-way ANOVA: F_(2,15) = 2.837, p = 0.09) levels.

Fig. 4

LD-RT tends to reduce cytokines/chemokine levels in the hippocampus of 3xTg-AD mice treated early. Concentrations of pro-inflammatory (A), anti-inflammatory (B) cytokines, and chemokines (C) measured using a Luminex multiplex assay on Tx-soluble protein fraction from the hippocampus of WT, sham-treated or LD-RT treated 3xTg-AD mice. ^*p < 0.05. RT, low-dose radiation therapy.

LD-RT does not impact behavior in 3xTg-AD mice treated early

3xTg-AD mice typically show early behavioral alterations, including locomotor alteration, higher anxiety, and deficits in learning and memory [20 –27]. For example, based on literature, the percentage of good alternation of WT mice in the Y maze test is around 75% at 12 months [26, 27], whereas our sham-treated 3xTg-AD mice only performed ∼40.1±2.9% of good alternation, showing their memory alteration. We evaluated the potential impact of LD-RT on cognitive and behavioral performances of animals, including working memory (Fig. 5A), general locomotion (Fig. 5B), and anxiety-like behaviors (Fig. 5C, D). No difference was observed between RT and sham-treated animals or between the spatial memory performances of animals before and after treatment (Fig. 5A). However, we noticed a decrease of the total distance travelled in the open field in both groups after treatment or anesthesia only, suggesting an increase of anxiety (Fig. 5B). This effect being also observed in sham-treated animals, it suggests that it was not due to the treatment but probably to repeated anesthesia/manipulations of animals (Two-way ANOVA: Group effect: F(_1,13) = 0.130, p = 0.724, Treatment effect: F(_1,13) = 30.48, p < 0.001, Group×treatment interaction: F(_1,13) = 0.008, p = 0.928; LSD post hoc test for treatment effect: p = 0.002 in RT group, p = 0.002 in sham-treated group). Indeed, we observed a tendency to a higher anxiety after treatment in the elevated plus maze test, when analyzing the total distance in the open arms (anxiogenic areas; Fig. 5C), and the number of head dipping (Fig. 5D), another index inversely correlated to anxiety.

Fig. 5

LD-RT does not impact behavioral and cognitive performances of 3xTg-AD mice treated early. A) Short term memory evaluation in the Y maze of 3xTg-AD mice before and 8 weeks after treatment. B) General locomotion in the open field. ^**p < 0.01. C) Total distance travelled in the open arms of the elevated plus maze (anxiogenic areas). D) Total number of head dipping in the open arms of the elevated plus maze. RT, low-dose radiation therapy.

DISCUSSION

Pathophysiological hallmarks of AD include amyloid plaques, abnormal hyperphosphorylation of tau, and neuroinflammation. Using a LD-RT schedule of 10 Gy in 5 daily fractions of 2 Gy applied at 12 months of age, we observed a significant decrease of aggregated Aβ₄₂ forms in the hippocampus of 3xTg-AD mice. Neuroinflammation markers also tended to be reduced by LD-RT. On the other hand, the tauopathy and cognitive performances were not modulated by the treatment.

The 3xTg-AD model is a slow progression model that displays only few amyloid plaques at 12 months, restricted to the subiculum [28]. The amyloid pathology was more pronounced in the hippocampus than in the frontal cortex, as shown by the higher concentration in the hippocampus of all Aβ forms (Aβ oligomers, Aβ₄₀, Aβ₄₂ in Tx-soluble and Gua-soluble fractions). The MXO4 staining, showing dense-core Aβ plaques [29], is not modulated by LD-RT. For the first time, we evaluated the impact of LD-RT on different forms of Aβ peptides in the brain. Interestingly, LD-RT significantly reduced Gua-soluble Aβ₄₂ peptides in the hippocampus (–71.2%). Other forms of Aβ also tended to be reduced after treatment (up to –67% depending on the peptide and the region studied) but did not reach significance, probably due to inter-individual variability. Globally, these results suggest that LD-RT is not able to reduce highly aggregated forms of amyloid (dense-core MXO4⁺ plaques) but may nevertheless decrease other forms of Aβ (soluble peptides, oligomers, fibrils). The reduction of amyloid plaques has been demonstrated in other studies with the same regimen on 3xTg-AD mice using 6E10 immunostaining [14] or in other mouse models of AD (Thioflavin S or 6E10 staining) [12, 13] but the effects are not well understood. For example, it has been proposed that LD-RT disintegrates plaques, through a direct physical mechanism and the depolymerization of glycosaminoglycans which are highly sensitive to radiation and present in amyloid fibrils [11]. However, the impact of LD-RT seems to be more complex: a recent study observed only a reduction of plaque size by not of the total plaque number in the 5xFAD mouse model [16], suggesting a slow mechanism of degradation, maybe by neighboring glial cells [30]. LD-RT applied on cell cultures could bring important insights to dissect the underlying mechanisms and should be considered for future studies.

In our study, we did not observe a reduction of the tauopathy (total Tau HT7⁺ and pT231-tau) after LD-RT. These results are in opposition to those reported in the only other study which studied the impact of LD-RT on tauopathy. Indeed, authors described a reduction of around 20% of total tau (5A6 tau staining) in the treated hippocampus of 3xTg-AD mice [14]. However, it is important to note that the treatment was not applied at the same stage of the pathology (16-month-old versus 12-month-old in our study), as tauopathy and amyloid increase with age in this model [31 –33]. It is possible that LD-RT only could have an impact on aggregated forms of pTau proteins, appearing at later stage, as shown for amyloid. Further studies are necessary to clarify the impact of LD-RT on various forms of aggregated tau and the optimal disease stage to deliver the treatment.

The anti-inflammatory effect of LD-RT in AD has been observed in three studies with different doses or delay post RT (1.8 Gy×5 fractions daily with a delay of 4 days [15]; 2 Gy×5 fractions daily with a delay of 8 weeks [13]; 0.6 Gy×5 fractions daily with a delay of 4 weeks [16]). Consequently, either the anti-inflammatory effect was observed using another regimen or delay post RT [15, 16], or the anti-inflammatory indexes measured were limited at the mRNA level [13]. In our previous study, in the TgF344-AD rat model, developing amyloid, tau, and cognitive deficits, we did not observe a modification of amyloid load or neuroinflammation in old females [17]. We hypothesize that the animals were treated at a too advanced stage (important amyloid load and neuroinflammation) to impact the neurochemical mechanisms of AD. In our study, animals were treated at an early stage of the disease, with slight amyloid deposition and tauopathy. However, 3xTg-AD mice did not display a clear neuroinflammation yet, excepted around amyloid plaques, as shown by the absence of an upregulation of neuroinflammation markers (GFAP, Vimentin, IBA1, CD68) or cytokines/chemokines compared to WT mice. Vimentin levels tended to be reduced after LD-RT, mainly in the subiculum, the only sub-region affected by amyloid plaques at this age. Vimentin being mainly detectable in response to pathological insults in the hippocampus, it may be more sensitive to LD-RT than GFAP. However, the decrease did not reach significance probably because the number of Vimentin⁺ astrocytes was very limited. MIP-2 levels were significantly reduced by LD-RT compared to WT mice and other pro-inflammatory cytokines (Il-1β, IL-2, IL-5, IL-6) tended to be reduced, suggesting also a slight reduction of neuroinflammation after treatment. Moreover, brain irradiation, using different regimens from ours (higher total doses, non-fractionated protocols or in juvenile mice) but also defined as LD-RT, has been shown to modulate inflammation markers in WT [32 –36]. The absence of a clear effect of LD-RT on cytokines in 3xTg-AD mice may suggest a dysregulation of the cellular response to LD-RT and may be explored in future studies.

Astrocyte reactivity has been associated with higher amyloid load, suggesting a deleterious role of reactive astrocytes in AD [39 –47]. We could consequently suppose that LD-RT decreased Aβ₄₂ levels through a slight modulation of astrocytes. Functional measure of amyloid uptake by astrocytes or production and degradation pathways should be studied. This hypothesis also assumes that astrocytes are involved earlier in AD pathogenesis than microglial cells, which is in agreement with the literature [48 –51].

3xTg-AD mice show early behavioral alterations, including locomotor alteration, higher anxiety, and deficits in learning and memory [20 –27]. In the TgF344-AD rat model, we described a reduction of the hyperlocomotion displayed by AD rats and an improvement of spatial memory deficits after LD-RT (2 Gy×5 fractions daily) compared to sham-treated animals, even in the absence of molecular changes [17]. Here we did not measure any improvement of behavioral and memory alteration after LD-RT in 3xTg-AD mice, in contrast to previous preliminary studies [12, 13]. This absence of therapeutic effect could be explained by the absence of a clear modulation of pathological hallmarks as discussed before.

Overall, this study confirmed that LD-RT is able to reduce amyloid load, and particularly aggregated Aβ₄₂ peptides. The effect of the treatment on inflammation and tauopathy should be tested in older mice presenting a clear neuroinflammation and more extended tauopathy or with a longer post RT delay to verify a potential preventive effect.

Footnotes

ACKNOWLEDGMENTS

We are grateful to Maria Surini and Mickael Das Neves for the technical assistance. We thank all the team of the radiation-oncology division, Geneva University Hospitals, for its help with the radiation treatments of animals. This work was supported by the Velux foundation (grant number 1123). PM was funded by the Swiss National Science Foundation (project 320030_184713). GF was funded by Association Suisse pour la Recherche sur l’Alzhei-mer, Genève; Fondation Segré, Genève; Ivan Pictet, Genève; Fondazione Agusta, Lugano; Fondation Chmielewski, Genève; Velux Stiftung; Swiss National Science Foundation (projects n. 320030_182772 and n. 320030_169876); Horizon 2020 (projects n. 667375); Human Brain Project; Innovative Medicines Initiatives (IMI contract n. 115736 and 115952). VG was funded by the Swiss National Science Foundation (projects 320030_169876 and 320030_185028) and the Schmidheiny foundation. TZ was funded by the Swiss National Science Foundation (project 320030_182366).

Authors’ disclosures available online ().

References

Ceyzériat

, Zilli

, Millet

, Frisoni

, Garibotto

, Tournier

(2020) Learning from the past: A review of clinical trials targeting amyloid, tau and neuroinflammation in Alzheimer’s disease. Curr Alzheimer Res 17, 112–125.

Luna-Muñoz

, Chávez-Macías

, García-Sierra

, Mena

(2007) Earliest stages of tau conformational changes are related to the appearance of a sequence of specific phospho-dependent tau epitopes in Alzheimer’s disease. J Alzheimers Dis 12, 365–375.

Luna-Muñoz

, Harrington

, Wischik

, Flores-Rodríguez

, Avila

, Zamudio

, De la Cruz

, Mena

, Meraz-Ríos

, Floran-Garduño

(2013) Phosphorylation of tau protein associated as a protective mechanism in the presence of toxic, C-terminally truncated tau in Alzheimer’s disease. In Understanding Alzheimer’s Disease, Zerr I, ed. IntechOpen, doi: 10.5772/54228.

Escartin

, Galea

, Lakatos

, O’Callaghan

, Petzold

, Serrano-Pozo

, Steinhäuser

, Volterra

, Carmignoto

, Agarwal

, Allen

, Araque

, Barbeito

, Barzilai

, Bergles

, Bonvento

, Butt

, Chen

W-T

, Cohen-Salmon

, Cunningham

, Deneen

, De Strooper

, Díaz-Castro

, Farina

, Freeman

, Gallo

, Goldman

, Götz

, Gutiérrez

, Haydon

, Heiland

, Hol

, Holt

, Iino

, Kastanenka

, Kettenmann

, Khakh

, Koizumi

, Lee

, Liddelow

, MacVicar

, Magistretti

, Messing

, Mishra

, Molofsky

, Murai

, Norris

, Okada

, Oliet

SHR

, Oliveira

, Panatier

, Parpura

, Pekna

, Pekny

, Pellerin

, Perea

, Pérez-Nievas

, Pfrieger

, Poskanzer

, Quintana

, Ransohoff

, Riquelme-Perez

, Robel

, Rose

, Rothstein

, Rouach

, Rowitch

, Semyanov

, Sirko

, Sontheimer

, Swanson

, Vitorica

, Wanner

I-B

, Wood

, Wu

, Zheng

, Zimmer

, Zorec

, Sofroniew

, Verkhratsky

(2021) Reactive astrocyte nomenclature, definitions, and future directions. Nat Neurosci 24, 312–325.

Ben Haim

, Carrillo-de Sauvage

, Ceyzériat

, Escartin

(2015) Elusive roles for reactive astrocytes in neurodegenerative diseases. Front Cell Neurosci 9, 278.

Hansen

, Hanson

, Sheng

(2018) Microglia in Alzheimer’s disease. J Cell Biol 217, 459–472.

Pascoal

, Benedet

, Ashton

, Kang

, Therriault

, Chamoun

, Savard

, Lussier

, Tissot

, Karikari

, Ottoy

, Mathotaarachchi

, Stevenson

, Massarweh

, Schöll

, de Leon

, Soucy

, Edison

, Blennow

, Zetterberg

, Gauthier

, Rosa-Neto

(2021) Microglial activation and tau propagate jointly across Braak stages. Nat Med 27, 1592–1599.

Domingues

, Cruz e Silva

OAB

, Henriques

(2017) Impact of cytokines and chemokines on Alzheimer’s disease neuro-pathological hallmarks. Curr Alzheimer Res 14, 870–882.

Ceyzériat

, Tournier

, Millet

, Frisoni

, Garibotto

, Zilli

(2020) Low-dose radiation therapy: A new treatment strategy for Alzheimer’s disease? J Alzheimers Dis 74, 411–419.

10.

Wilson

, Marples

(2016) A new use for an old treatment: Radiation therapy and Alzheimer’s disease. Radiat Res 185, 443–448.

11.

Bistolfi

(2008) Localized amyloidosis and Alzheimer’s disease: The rationale for weekly long-term low dose amyloid-based fractionated radiotherapy. Neuroradiol J 21, 683–692.

12.

Marples

, McGee

, Callan

, Bowen

, Thibodeau

, Michael

, Wilson

, Maddens

, Fontanesi

, Martinez

(2016) Cranial irradiation significantly reduces beta amyloid plaques in the brain and improves cognition in a murine model of Alzheimer’s disease (AD). Radiother Oncol 118, 579–580.

13.

Kim

, Chung

, Ngoc Mai

, Nam

, Shin

, Park

, Chung

, Lee

, Rhee

, Jahng

, Kim

, Lim

, Kong

, Moon

, Chung

(2020) Low-dose ionizing radiation modulates microglia phenotypes in the models of Alzheimer’s disease. Int J Mol Sci 21, 4532.

14.

Wilson

, Wilson

, Hanna

, Fontanesi

, Kulchycki

, Buelow

, Pruetz

, Michael

, Chinnaiyan

, Maddens

, Martinez

, Fontanesi

(2020) Low dose brain irradiation reduces Amyloid-β and Tau in 3xTg-AD mice. J Alzheimers Dis 75, 15–21.

15.

Kim

, Nam

, Kim

, Lee

, Hong

, Kim

, Shin

, Park

, Mai

, Oh

, Kim

, Yoo

, Chung

, Moon

(2020) Neuroprotective and anti-inflammatory effects of low-moderate dose ionizing radiation in models of Alzheimer’s disease. Int J Mol Sci 21, 3678.

16.

Yang

, Kim

, Choi

, Kim

, Yoon

, Seo

, Kim

(2021) Modulation of neuroinflammation by low-dose radiation therapy in an animal model of Alzheimer’s disease. Int J Radiat Oncol Biol Phys 111, 658–670.

17.

Ceyzériat

, Zilli

, Fall

, Millet

, Koutsouvelis

, Dipasquale

, Frisoni

, Tournier

, Garibotto

(2021) Treatment by low-dose brain radiation therapy improves memory performances without changes of the amyloid load in the TgF344-AD rat model. Neurobiol Aging 103, 117–127.

18.

Carroll

, Rosario

, Kreimer

, Villamagna

, Gentzschein

, Stanczyk

, Pike

(2010) Sex differences in β-amyloid accumulation in 3xTg-AD mice: Role of neonatal sex steroid hormone exposure. Brain Res 1366, 233–245.

19.

O’Hara

(2008) How I do it: Sample size calculations. Clin Otolaryngol 33, 145–149.

20.

Guillemaud

, Ceyzériat

, Saint-Georges

, Cambon

, Petit

, Ben Haim

, Carrillo-de Sauvage

, Guillermier

, Bernier

, Hérard

, Joséphine

, Bémelmans

, Brouillet

, Hantraye

, Bonvento

, Escartin

(2020) Complex roles for reactive astrocytes in the triple transgenic mouse model of Alzheimer disease. Neurobiol Aging 90, 135–146.

21.

Clinton

, Billings

, Green

, Caccamo

, Ngo

, Oddo

, McGaugh

, LaFerla

(2007) Age-dependent sexual dimorphism in cognition and stress response in the 3xTg-AD mice. Neurobiol Dis 28, 76–82.

22.

Filali

, Lalonde

, Theriault

, Julien

, Calon

, Planel

(2012) Cognitive and non-cognitive behaviors in the triple transgenic mouse model of Alzheimer’s disease expressing mutated APP, PS1, and Mapt (3xTg-AD). Behav Brain Res 234, 334–342.

23.

Gloria

, Ceyzériat

, Tsartsalis

, Millet

, Tournier

(2021) Dopaminergic dysfunction in the 3xTg-AD mice model of Alzheimer’s disease. Sci Rep 11, 19412.

24.

Stimmell

, Baglietto-Vargas

, Moseley

, Lapointe

, Thompson

, LaFerla

, McNaughton

, Wilber

(2019) Impaired spatial reorientation in the 3xTg-AD mouse model of Alzheimer’s disease. Sci Rep 9, 1311.

25.

Webster

, Bachstetter

, Nelson

, Schmitt

, Van Eldik

(2014) Using mice to model Alzheimer’s dementia: An overview of the clinical disease and the preclinical behavioral changes in 10 mouse models. Front Genet 5, 88.

26.

Carroll

, Rosario

, Chang

, Stanczyk

, Oddo

, LaFerla

, Pike

(2007) Progesterone and estrogen regulate Alzheimer-like neuropathology in female 3xTg-AD mice. J Neurosci 27, 13357–13365.

27.

Knight

, Martins

IVA

, Gümüsgöz

, Allan

, Lawrence

(2014) High-fat diet-induced memory impairment in triple-transgenic Alzheimer’s disease (3xTgAD) mice is independent of changes in amyloid and tau pathology. Neurobiol Aging 35, 1821–1832.

28.

Tournier

, Tsartsalis

, Rigaud

, Fossey

, Cailly

, Fabis

, Pham

, Grégoire

M-C

, Kövari

, Moulin-Sallanon

, Savioz

, Millet

(2019) TSPO and amyloid deposits in sub-regions of the hippocampus in the 3xTgAD mouse model of Alzheimer’s disease. Neurobiol Dis 121, 95–105.

29.

Serrano-Pozo

, Frosch

, Masliah

, Hyman

(2011) Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med 1, a006189.

30.

Ries

, Sastre

(2016) Mechanisms of abeta clearance and degradation by glial cells. Front Aging Neurosci 8, 160.

31.

Chiquita

, Ribeiro

, Castelhano

, Oliveira

, Sereno

, Batista

, Abrunhosa

, Rodrigues-Neves

, Carecho

, Baptista

, Gomes

, Moreira

, Ambrósio

, Castelo-Branco

(2019) A longitudinal multimodal in vivo molecular imaging study of the 3xTg-AD mouse model shows progressive early hippocampal and taurine loss. Hum Mol Genet 28, 2174–2188.

32.

Oddo

, Caccamo

, Shepherd

, Murphy

, Golde

, Kayed

, Metherate

, Mattson

, Akbari

, LaFerla

(2003) Triple-transgenic model of Alzheimer’s disease with plaques and tangles: Intracellular Abeta and synaptic dysfunction. Neuron 39, 409–421.

33.

Belfiore

, Rodin

, Ferreira

, Velazquez

, Branca

, Caccamo

, Oddo

(2019) Temporal and regional progression of Alzheimer’s disease-like pathology in 3xTg-AD mice. Aging Cell 18, e12873.

34.

Monje

, Toda

, Palmer

(2003) Inflammatory blockade restores adult hippocampal neurogenesis. Science 302, 1760–1765.

35.

Boyd

, Byrne

, Middleton

, Banati

, Liu

(2021) Control of neuroinflammation through radiation-induced microglial changes. Cells 10, 2381.

36.

Zhou

, Boström

, Ek

, Li

, Xie

, Xu

, Sun

, Blomgren

, Zhu

(2017) Radiation induces progenitor cell death, microglia activation, and blood-brain barrier damage in the juvenile rat cerebellum. Sci Rep 7, 46181.

37.

Ung

, Garrett

, Dalke

, Leitner

, Dragosa

, Hladik

, Neff

, Wagner

, Zitzelsberger

, Miller

, de Angelis

, Rößler

, Vogt Weisenhorn

, Wurst

, Graw

, Hölter

(2021) Dose-dependent long-term effects of a single radiation event on behaviour and glial cells. Int J Radiat Biol 97, 156–169.

38.

Osman

, Sun

, Burns

, He

, Kee

, Oliva-Vilarnau

, Alevyzaki

, Zhou

, Louhivuori

, Uhlén

, Hedlund

, Betsholtz

, Lauschke

, Kele

, Blomgren

(2020) Radiation triggers a dynamic sequence of transient microglial alterations in juvenile brain. Cell Rep 31, 107699.

39.

Kiyota

, Okuyama

, Swan

, Jacobsen

, Gendelman

, Ikezu

(2010) CNS expression of anti-inflammatory cytokine interleukin-4 attenuates Alzheimer’s disease-like pathogenesis in APP+PS1 bigenic mice. FASEB J 24, 3093–3102.

40.

Krauthausen

, Kummer

, Zimmermann

, Reyes-Irisarri

, Terwel

, Bulic

, Heneka

, Müller

(2015) CXCR3 promotes plaque formation and behavioral deficits in an Alzheimer’s disease model. J Clin Invest 125, 365–378.

41.

Tweedie

, Ferguson

, Fishman

, Frankola

, Van Praag

, Holloway

, Luo

, Li

, Caracciolo

, Russo

, Barlati

, Ray

, Lahiri

, Bosetti

, Greig

, Rosi

(2012) Tumor necrosis factor-α synthesis inhibitor 3,6’-dithiothalidomide attenuates markers of inflammation, Alzheimer pathology and behavioral deficits in animal models of neuroinflammation and Alzheimer’s disease. J Neuroinflammation 9, 106.

42.

Vom Berg

, Prokop

, Miller

, Obst

, Kälin

, Lopategui-Cabezas

, Wegner

, Mair

, Schipke

, Peters

, Winter

, Becher

, Heppner

(2012) Inhibition of IL-12/IL-23 signaling reduces Alzheimer’s disease-like pathology and cognitive decline. Nat Med 18, 1812–1819.

43.

Shi

, Chowdhury

, Ma

, Le

, Hong

, Caldarone

, Stevens

, Lemere

(2017) Complement C3 deficiency protects against neurodegeneration in aged plaque-rich APP/PS1 mice. Sci Transl Med 9, eaaf6295.

44.

Fonseca

, Zhou

, Botto

, Tenner

(2004) Absence of C1q leads to less neuropathology in transgenic mouse models of Alzheimer’s disease. J Neurosci 24, 6457–6465.

45.

Fernandez

, Jimenez

, Mecha

, Davila

, Guaza

, Vitorica

, Torres-Aleman

(2012) Regulation of the phosphatase calcineurin by insulin-like growth factor I unveils a key role of astrocytes in Alzheimer’s pathology. Mol Psychiatry 17, 705–718.

46.

Furman

, Sama

, Gant

, Beckett

, Murphy

, Bachstetter

, Van Eldik

, Norris

(2012) Targeting astrocytes ameliorates neurologic changes in a mouse model of Alzheimer’s disease. J Neurosci 32, 16129–16140.

47.

Ceyzériat

, Ben Haim

, Denizot

, Pommier

, Matos

, Guillemaud

, Palomares

, Abjean

, Petit

, Gipchtein

, Gaillard

, Guillermier

, Bernier

, Gaudin

, Aurégan

, Joséphine

, Déchamps

, Veran

, Langlais

, Cambon

, Bemelmans

, Baijer

, Bonvento

, Dhenain

, Deleuze

, Oliet

SHR

, Brouillet

, Hantraye

, Carrillo-de Sauvage

, Olaso

, Panatier

, Escartin

(2018) Modulation of astrocyte reactivity improves functional deficits in mouse models of Alzheimer’s disease. Acta Neuropathol Commun 6, 104.

48.

Tournier

, Tsartsalis

, Ceyzériat

, Fraser

, Grégoire

, Kövari

, Millet

(2020) Astrocytic TSPO upregulation appears before microglial TSPO in Alzheimer’s disease. J Alzheimers Dis 77, 1043–1056.

49.

Heneka

, Carson

, El Khoury

, Landreth

, Brosseron

, Feinstein

, Jacobs

, Wyss-Coray

, Vitorica

, Ransohoff

, Herrup

, Frautschy

, Finsen

, Brown

, Verkhratsky

, Yamanaka

, Koistinaho

, Latz

, Halle

, Petzold

, Town

, Morgan

, Shinohara

, Perry

, Holmes

, Bazan

, Brooks

, Hunot

, Joseph

, Deigendesch

, Garaschuk

, Boddeke

, Dinarello

, Breitner

, Cole

, Golenbock

, Kummer

(2015) Neuroinflammation in Alzheimer’s disease. Lancet Neurol 14, 388–405.

50.

Chun

, Marriott

, Lee

, Cho

(2018) Elucidating the interactive roles of glia in Alzheimer’s disease using established and newly developed experimental models. Front Neurol 9, 797.

51.

Gordon

, Holcomb

, Jantzen

, DiCarlo

, Wilcock

, Boyett

, Connor

, Melachrino

, O’Callaghan

, Morgan

(2002) Time course of the development of Alzheimer-like pathology in the doubly transgenic PS1+APP mouse. Exp Neurol 173, 183–195.