Early Detection of Aβ Deposition in the 5xFAD Mouse by Amyloid PET.

Purpose.18F-FC119S is a positron emission tomography (PET) tracer for imaging β-amyloid (Aβ) plaques in Alzheimer's disease (AD). The aim of this study is to evaluate the efficacy of 18F-FC119S in quantitating Aβ deposition in a mouse model of early amyloid deposition (5xFAD) by PET. Method. Dynamic 18F-FC119S PET images were obtained in 5xFAD (n = 5) and wild-type (WT) mice (n = 7). The brain PET images were spatially normalized to the M. Mirrione T2-weighted mouse brain MR template, and the volumes of interest were then automatically drawn on the cortex, hippocampus, thalamus, and cerebellum. The specific binding of 18F-FC119S to Aβ was quantified as the distribution volume ratio using Logan graphical analysis with the cerebellum as a reference tissue. The Aβ levels in the brain were also confirmed by immunohistochemical analysis. Result. For the 5xFAD group, radioactivity levels in the cortex, the hippocampus, and the thalamus were higher than those for the WT group. In these regions, specific binding was approximately 1.2-fold higher in 5xFAD mice than in WT. Immunohistochemistry supported these findings; the 5xFAD showed severe Aβ deposition in the cortex and hippocampus in contrast to the WT group. Conclusion. These results demonstrated that 18F-FC119S PET can successfully distinguish Aβ depositions in 5xFAD mice from WT.

1. Introduction

Alzheimer's disease (AD), the most common neurodegenerative disorder, is characterized by cognitive and memory deterioration as a consequence of abnormal deposition of amyloid-beta (Aβ) and neurofibrillary tangles of hyperphosphorylated tau [1-3].

Early diagnosis of AD, before the advent of structural changes, is important to reduce the socioeconomic burden of the disease. Molecular imaging with positron emission tomography (PET) is a promising tool because it can functionally detect the burden of Aβ plaques and evaluate longitudinal changes in patients with AD [4, 5]. Radiochemists designed many other Aβ-targeted PET radiotracers in an effort to satisfy one with more drug-like properties. Klunk et al. developed 11C-Pittsburgh compound B (PIB), which has high specificity and affinity for Aβ [6]. However, this radiopharmaceutical has an intrinsic limitation for multicenter studies. 11C has a short half-life of only 20 min, and thus, 11C-based radiotracers can be used only at a center with an on-site synthetic facility [7]. Later, 18F-florbetapir, 18F-flutemetamol, and 18F-florbetaben, which detect the presence of Aβ plaques, were developed and approved by the US Food and Drug Administration [8-10]. Previous PET studies have shown that they have both succeeded and failed in detecting amyloidosis in AD animal models [11-15]. Recently, 18F-FC119S (2-[2-(N-monomethyl)aminopyridine-6-yl]-6-[(S)-3-fluoro-2-hydroxypropoxy]benzothazole) is developed Aβ imaging PET tracer. This tracer has high selectivity and metabolic stability against troublesome in vivo defluorination [16]. In addition, 18F-FC119S showed a greater binding value in the APP/PS1 than their wild-type (WT) counterparts [17]. In the first in human study including patients with mild cognitive impairment and AD, 18F-FC119S displays significant linear correlation with 11C-PIB [18].

Several transgenic mouse lines have been used for amyloid PET research, but only a limited range of AD mice can be used [14, 15]. The 5xFAD mouse is a transgenic model of AD carrying five mutations associated with early onset familial Alzheimer's disease (FAD): the K670N/M671L (Swedish), I716V (Florida), and V717I (London) mutations in human amyloid precursor protein (APP) and the M146L and L286V mutations in human presenilin-1 (PS1) [19-22]. All five FAD mutations synergistically promote the occurrence of Aβ42; therefore, this model shows very aggressive Aβ deposition in the brain starting at 1.5 months of age [23].

The purpose of the present study is to examine the efficacy of 18F-FC119S PET for distinguishing Aβ deposition in an AD mouse model of early amyloid deposition from WT.

2. Materials and Methods

2.1. Animals

Two groups of mice were used for these studies: 5xFAD mice (male, n = 5) and WT counterparts (male, n = 7), all at the age of 5.5 months. The care, maintenance, and treatment of animals in these studies followed protocols approved by the Institutional Animal Care and Use Committee of Korea Institute of Radiological & Medical Sciences (KIRAMS), and the experiments involving animals were performed according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health. The animal housing room was automatically controlled with a temperature of 22 ± 3°C, a 12 h/12 h light/dark cycle and 55 ± 20% humidity. Sterilized rodent diet and purified tap water were supplied ad libitum.

2.2. Preparation of 18F-FC119S

18F-FC119S was synthesized by nucleophilic substitution of F-18 on the precursor as previously described in Byun et al.'s work [18]. The specific activity of 18F-FC119S was greater than 44 GBq/μmol, and the mean radiochemical purity was 99% (n = 4).

2.3. PET/CT Scan

PET images of the mice were obtained using a small-animal PET scanner (nanoScan®, Mediso, Budapest, Hungary), consisting of two rings with 12 PET detector modules. The spatial resolution at 1 mm off the center of the scanner in the axial, radial, tangential directions is 0.85, 0.8, and 0.8 mm, respectively. Mice were anesthetized with 2.5% isoflurane, and 18F-FC119S (8.9 ± 1.3 MBq/200 uL) was injected through the tail vein with a syringe pump (KDS 210, KD Scientific, Holliston, MA) over the course of 1 min. Simultaneously, dynamic PET scanning was performed for 60 min. Images were reconstructed with user defined time frames (14 × 30 s, 3 × 60 s, 4 × 300 s, 3 × 600 s, 24 frames in total) using a 3-dimensional ordered subset expectation maximization algorithm. For attenuation correction and anatomical reference, a computed tomography (CT) scan was acquired immediately after PET (50 kVp of X-ray voltage at 0.16 mAs).

2.4. Image Analysis

According to Oakley et al., Aβ deposition begins to appear in the cortex, the hippocampus and the thalamus of 5xFAD mice at 1.5 months of age [23]. However, no significant increase in Aβ was found in the olfactory bulbs, striatum, hypothalamus, or thalamus at 5 months of age [24]. Therefore, we selected the cortex, hippocampus, and the thalamus as volumes of interest (VOIs). To create a study-specific PET brain template, dynamic PET images were motion-corrected and spatially normalized to the T2-weighted mouse brain MR template (M. Mirrione, embedded PMOD software). Four VOIs were defined on the MRI template; cortex (149.2 mm3), hippocampus (24.9 mm3), thalamus (28.2 mm3), and cerebellum (57.2 mm3, Figure 1). Finally, regional time-activity curves (TACs) were generated. The obtained uptake value was determined for each brain region and presented as a standardized uptake value (SUV). The SUVs were obtained by normalizing tissue radioactivity concentration to injected dose and body weight [25].

Figure 1

Definition of VOIs from a 5xFAD mouse in the horizontal (a), coronal (b), and sagittal (c) planes. The PET was spatially normalized to the T2-weighted mouse brain MR template.

To compare the specific binding of 18F-FC119S to Aβ deposits, we used Logan graphical analysis (t = 10 min) to derive a distribution volume ratio (DVR) based on the cerebellum [26]. All image analysis was conducted with PMOD (version 3.4, PMOD Group, Graubunden, Switzerland).

2.5. Immunohistochemistry

The animals were euthanized, and brain samples were fixed in 4% paraformaldehyde for 48 hours, embedded in paraffin, and sectioned at 5 μm intervals. Immunohistochemistry was conducted using the Vectastain Elite ABC kit (Vector Laboratories Inc., Burlingame, CA, USA) following the manufacturer's protocol. For antigen retrieval, the sections were placed in a citrate buffer (pH 6.0) and heated in boiling water for 30 minutes. The sections were then placed in 0.3% H2O2 in absolute methanol for 15 minutes at room temperature to block the endogenous peroxidase. Sections were then incubated overnight at 4°C with mouse anti-6E10 antibody (1 : 1000, SIG-39320, Covance, Emeryville, CA), washed, and incubated with the corresponding secondary antibody. ImageJ was used to quantify the amount of Aβ in each brain section. As a control, the primary antibody was omitted from several test sections in each experiment. The sections were counterstained with Harris' hematoxylin prior to mounting.

For analysis of the amyloid plaques, the ImageJ software (https://rsbweb.nih.gov/ij/) was used. After adjusting for threshold, ImageJ was used to measure total area of the plaques and the percentage of total brain area occupied by plaques. The brain area (cortex or hippocampus) was outlined in the right hemisphere using the edit plane function. Data were pooled from 8~12 sections at ×200 magnification of each mouse and 5 mice were used for the statistical analysis.

2.6. Statistical Analysis

The quantitative results are expressed as the mean ± SD. All statistical results were analyzed with Prism (GraphPad Software, Inc., CA). Student's t-test was used to determine statistical significance at the 95% confidence level, with p < 0.05 indicating a significant difference.

3. Results

3.1. 18F-FC119S PET Images

The ability of 18F-FC119S to quantify Aβ burden was assessed by PET. PET images showing the brain SUV ratio (30–60 min) are shown in Figure 2. By visual inspection, the 5xFAD showed higher cortical and hippocampal 18F-FC119S uptake than the WT group. Radioactivity was also detected in the olfactory bulb in both groups, but this uptake was due to spill-over from the Harderian gland [27].

Figure 2

18F-FC119S summed PET images (30–60 min) of the WT and the 5xFAD in the one representative animal. The color scale indicates the SUV ratio.

Figure 3 represents the regional TACs. After approximately 10 minutes, the cortex and hippocampus and the thalamus of the 5xFAD showed greater uptake than those of the WT (Figures 3(a)–3(c)). In contrast, radioactivities in the cerebellum, which we used as the reference region, did not differ between groups (Figure 3(d)). Area under the curve (AUC) values was obtained from 30 to 60 minutes by the trapezoid rule (Figures 3(e)–3(h)). In 5xFAD mice, the AUC values of the cortex, hippocampus, and thalamus showed 33.3, 41.7, and 25.9% increments, respectively, compared with the corresponding values in WT mice. Among VOIs, only hippocampus showed statistically significant difference (p = 0.0479). The difference of cerebellar uptakes between 5xFAD and WT was not statistically significant (p = 0.9299).

Figure 3

Time-activity curves of the cortex (a), hippocampus (b), thalamus (c), and cerebellum (d). Area under the curve (AUC) for the cortex (e), hippocampus (f), thalamus (g), and cerebellum (h). Data are presented as the mean ± SD. (p < 0.05, n.s. = statistically nonsignificant difference).

3.2. Distribution Volume Ratios

To elucidate the specific binding level of 18F-FC119S, we calculated the DVR values (Figure 4). The mean DVR values for the 5xFAD were 10–21% higher than those for the WT group (DVR values in cortex: 5xFAD = 0.99 ± 0.05 versus WT = 0.86 ± 0.04, p = 0.0007; in hippocampus: 5xFAD = 1.03 ± 0.09 versus WT = 0.85 ± 0.03, p = 0.0006; in thalamus: 5xFAD = 1.01 ± 0.12 versus WT = 0.92 ± 0.04, p = 0.0748).

Figure 4

The DVR values according to Logan graphical analysis. Statistical significance was defined as a p value less than 0.05 (p < 0.001, n.s. = statistically nonsignificant difference).

3.3. Immunohistochemical Staining

To assess the actual Aβ burden in the brain, we performed immunohistochemistry (Figure 5). While Aβ deposition was not detected in the WT mice, increased Aβ burden was identified in 5xFAD (Figure 5(a)). The quantification revealed that the Aβ deposition was greater in the cortex than in the hippocampus (Figure S1, 6.13 ± 0.76% and 3.08 ± 0.73%, resp.). These results were consistent with previous studies showing that Aβ filled most of the hippocampus and cortex of 5xFAD mice within 6 months [28]. However, Aβ plaques were not expressed in both 5xFAD and WT (Figure S2). The association between Aβ deposition and DVRs showed that the cortex has low correlation (Figure 5(b), r = 0.45) whereas the hippocampus has strong correlation (Figure 5(c), r = 0.85).

Figure 5

Immunohistochemical staining of Aβ in the brains of WT and 5xFAD mice (a). Insets represent high-magnification images of the cortex (A) and hippocampus (B). Relationship between IHC data and DVRs of cortex (b) and hippocampus (c).

4. Discussion

To the best of our knowledge, the present study is the first to demonstrate that 18F-FC119S PET can successfully differentiate the Aβ burden of 5xFAD mice from that of WT mice. These PET results were validated by immunohistochemistry. From this perspective, 18F-FC119S may be a useful PET radiotracer for Aβ imaging.

Pathologically, Aβ deposition in the brain occurs first in the cortical area (6–9 months), followed by the hippocampus (12–15 months) [29-32]. This implies that Aβ plaques have a relatively lower concentration in the limbic system than in the cortex at a young age [30]. Previous rodent amyloid PET studies using 11C-PIB or 18F-florbetaben failed to discriminate between young AD mice (PS2APP, G384A, APP/PS1; all mice age less than 12 months) and WT mice in terms of specific binding in the hippocampus [15], whereas the differences of binding values in 20-month-aged APP/PS1 mice were successfully detected by 18F-FC119S [17]. Rojas et al. also demonstrated that increased cerebral uptakes in 10–16-month-aged 5xFAD by 18F-florbetapir [12]. In addition to genetic modification, age may be a critical factor. Pathological changes in the hippocampus are observable in advanced AD mice. Longitudinal PET studies must be conducted for 20 months or longer to detect these changes [33]. However, this long timeframe makes it difficult to choose an appropriate age for analysis and to maintain the animals in good condition for the entire study. 5xFAD line displays high concentration of Aβ as well as cognitive dysfunction around 4 months of age. Moreover, these mice develop neuron loss, unlike most other hAPP and hAPP/PS1 models [34]. Regarding this issue, 5xFAD mice may offer an advantage over other transgenic mice. Due to the aggressive Aβ deposition in this model, we confirmed the presence of detectable changes even at the age of 5.5 months. The benefit of 5xFAD model is the early amyloid deposition, which could offer to longitudinal experiments over APP/PS1.

The current PET results are an underestimate of the in vitro values. In our immunohistochemistry analysis, the cortical and hippocampal Aβ plaque concentration was evidently increased in the 5xFAD, while no Aβ was detected in WT mice. Despite this in vitro result, the difference between binding values was only 16–22%. These patterns are consistent with previous studies using APP/PS1 transgenic mice [17, 19]. This observation may be related to experimental techniques. Dynamic PET images provide information on the pharmacokinetics and pharmacodynamics of the tracer in the body [35]. On the other hand, histological examination reflects antigen-antibody binding ability and is observed in a brain slice under static incubation conditions. Partial volume effect by the small brain structure and limited PET resolution also attribute the underestimation of PET signal.

5. Conclusion

In summary, 18F-FC119S PET showed selective binding to pathological Aβ deposits in 5xFAD mice. The cortical and hippocampal uptake of the tracer was higher in the 5xFAD than in WT mice. These results were reproduced by immunohistochemistry. Therefore, 18F-FC119S would be a suitable candidate preclinical PET radiotracer.