Elsevier

Neurobiology of Aging

Volume 34, Issue 7, July 2013, Pages 1790-1798
Neurobiology of Aging

Regular article
In vivo evaluation of amyloid deposition and brain glucose metabolism of 5XFAD mice using positron emission tomography

https://doi.org/10.1016/j.neurobiolaging.2012.12.027Get rights and content

Abstract

Positron emission tomography (PET) has been used extensively to evaluate the neuropathology of Alzheimer's disease (AD) in vivo. Radiotracers directed toward the amyloid deposition such as [18F]-FDDNP (2-(1-{6-[(2-[F]Fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malononitrile) and [11C]-PIB (Pittsburg compound B) have shown exceptional value in animal models and AD patients. Previously, the glucose analogue [18F]-FDG (2-[(18)F]fluorodeoxyglucose) allowed researchers and clinicians to evaluate the brain glucose consumption and proved its utility for the early diagnosis and the monitoring of the progression of AD. Animal models of AD are based on the transgenic expression of different human mutant genes linked to familial AD. The novel transgenic 5XFAD mouse containing 5 mutated genes in its genome has been proposed as an AD model with rapid and massive cerebral amyloid deposition. PET studies performed with animal-dedicated scanners indicate that PET with amyloid-targeted radiotracers can detect the pathological amyloid deposition in transgenic mice and rats. However, in other studies no differences were found between transgenic mice and their wild type littermates. We sought to investigate in 5XFAD mice if the radiotracers [11C]-PIB, and [18F]-Florbetapir could quantify the amyloid deposition in vivo and if [18F]-FDG could do so with regard to glucose consumption. We found that 5XFAD animals presented higher cerebral binding of [18F]-Florbetapir, [11C]-PIB, and [18F]-FDG. These results support the use of amyloid PET radiotracers for the evaluation of AD animal models. Probably, the increased uptake observed with [18F]-FDG is a consequence of glial activation that occurs in 5XFAD mice.

Introduction

In vivo evaluation of amyloid deposition in Alzheimer's disease (AD) has attracted great interest because it permits investigation of the pathophysiology of the disease and facilitates the evaluation of new therapeutic strategies (Rinne et al., 2010). In this way, the first successful radiotracer for amyloid imaging of AD patients was [18F]-FDDNP (2-(1-{6-[(2-[F]Fluoroethyl)(methyl)amino]-2-naphthyl}ethylidene)malononitrile) which binds to hyperphosphorylated tau aggregates as well (Shoghi-Jadid et al., 2002). AD patients showed an increase of 10%–20% in the tracer binding in the affected brain regions compared with control subjects (Burggren et al., 2006). The next radiotracer developed was [11C]-PIB (Pittsburg compound B) which does not bind to hyperphosphorylated tau aggregates but presents a higher specific binding to amyloid aggregates in comparison with [18F]-FDDNP (Klunk et al., 2004). To date, [11C]-PIB has been the most widely used radiotracer to evaluate amyloid load in AD patients. However, clinical use of this tracer is restricted to positron emission tomography (PET) centers equipped with a cyclotron because it is labeled with C-11 (Herholz and Ebmeier, 2011). Recently, new fluorinated radiotracers like [18F]-Florbetapir (Frisoni, 2011), [18F]-Florbetapen (Barthel et al., 2011), and [18F]-Flutemetamol (Vandenberghe et al., 2010) have been developed to overcome this limitation. These new compounds present properties similar to [11C]-PIB but can be used in imaging centers far away from a cyclotron, thus enabling their widespread distribution and use in routine clinical practice. One of them, [18F]-Florbetapir, has been recently approved by the US Food and Drug Administration and the European Medicines Agency for the clinical management of patients with dementia. Besides, [18F]-FDG (2-[(18)F]fluorodeoxyglucose) is the most frequently used PET radiotracer, particularly, for the clinical evaluation of oncology patients. As a glucose analogue, it has been used extensively to evaluate cerebral glucose consumption in AD and other dementias (Mosconi et al., 2010). Decreased [18F]-FDG uptake has been consistently reported in parietal, temporal, medial, and frontal cortices of patients with AD and mild cognitive impairment years before the onset of clinical symptoms (Herholz et al., 2011).

A number of animal models that recapitulate essential features of AD have been proposed (Woodruff-Pak, 2008). The most relevant ones are transgenic animals which contain in their genome 1 or more sequences of human mutated proteins responsible for familial forms of AD, namely amyloid beta precursor protein (APP), presenilin 1 (PS1), and presenilin 2 (Ertekin-Taner, 2007). In this way, transgenic mice expressing a mutated APP (Higgins et al., 1993), double transgenic for mutated APP and PS1 (Citron et al., 1997) or triple transgenic for mutated APP, PS1, and tau (Oddo et al., 2003) are widely used as animal AD models because they develop cerebral amyloid deposition with aging. Recently, a new strain containing 5 human mutations related to familial AD has been proposed as an animal model for AD investigations (Oakley et al., 2006). These animals develop amyloid deposits faster than any other strain to date. Therefore, they can reduce the amount of time required to obtain suitable animals for investigation, hence simplifying the realization of longitudinal studies.

In the investigation of transgenic animal models of AD, PET imaging enables the longitudinal evaluation of the same animal, thus facilitating the investigation of different neuropathological alterations. In addition to this, using the same animal as its own control eases the preclinical evaluation of new therapeutic strategies. In this way, amyloid deposition has been investigated in different transgenic mice breeds using [11C]-PIB (Klunk et al., 2005; Toyama et al., 2005) and [18F]-FDDNP (Kuntner et al., 2009) with contradictory results. Some authors reported that [11C]-PIB enables the in vivo quantification of amyloid load in mice (Maeda et al., 2007), but others failed to find any differences between transgenic and wild type (WT) animals (Klunk et al., 2005). These discrepancies have been attributed to differences in the amount of amyloid deposition between strains and to the specific activity of the radiotracer at the time of injection. These studies indicate that a considerably higher specific activity than that usually used to image humans is required to successfully detect amyloid deposition in mice. Achieving such high specific activities is challenging and represents a major limitation to the technique and for the molecular imaging community. Higher specific activities are sought because [11C]-PIB presents a lower affinity for amyloid plaques in transgenic mice than amyloid deposits found in humans. Moreover, because of differences in size and other physiological factors between humans and rodents, there is a dramatic divergence in the mass effect which might lead to significant blocking of the specific binding sites with relatively low radiotracer doses (Kung and Kung, 2005). Indeed, the radiotracer binds specifically to a specific binding site of the amyloid protein, so the specific activity could be a key factor to prevent the blocking of this site with the unlabelled compound (Higuchi, 2009).

The long half-life of fluorine-labeled amyloid radiotracers, such as [18F]-Florbetapir could facilitate studies in animals with high specific activities and provide better count statistics for image reconstruction. Recently, the usefulness of this radiotracer to evaluate the amyloid deposition in double transgenic mice has been demonstrated and could represent and advantageous alternative to [11C]-PIB for the evaluation of new therapies in animal models of AD (Poisnel et al., 2012a).

Another important limitation of the in vivo PET studies in rodents is the relatively low resolution of the images. This hampers the quantification of the brain because of the partial volume effect and might preclude the quantification of certain structures like the cortex or hippocampus. Recently, transgenic rats expressing mutations related to familial AD have been proposed as a model of the disease (Flood et al., 2009). Because rats are 10 times bigger than mice, the resolution limitations of PET can be partially overcome. In this way, a recent study demonstrated that transgenic rats present an age-dependent increase in the uptake of [18F]-FDDNP compared with WT animals (Teng et al., 2011).

Besides amyloid imaging, autoradiographic studies performed with FDG labeled with 14C showed reductions in glucose consumption in some brain areas of transgenic mice (Reiman et al., 2000). These animals presented reductions of FDG uptake mainly at the level of the retrosplenial cingulate cortex that has reciprocal connectivity with the entorhinal cortex (Dubois et al., 2010). In addition, other transgenic mice strains developed reductions at the level of the thalamus (Valla et al., 2008). Because the observed alterations were correlated with the age of the animal, the authors paralleled them to the severe decline of [18F]-FDG uptake described in AD patients. However, transgenic animals did not develop the severe neuronal loss and brain atrophy that contribute to the observed hypometabolism in the [18F]-FDG PET scans of AD patients. PET studies performed with [18F]-FDG in transgenic mice models of AD reported contradictory results compared with previous ex vivo studies. In fact, 1 study did not find any differences between transgenic and WT animals. The authors argued that the hypometabolic areas described in autoradiographic studies, were too small to be individually quantified in vivo, because of the relatively low resolution of microPET systems (Valla et al., 2002). However 2 posterior studies performed with PET in transgenic AD models reported dependent increments of brain glucose consumption compared with age-matched control animals (Luo et al., 2012; Poisnel et al., 2012b). The different transgenic AD mice strains that were evaluated presented important variations in cerebral amyloid load and its distribution that could explain the differences in cerebral glucose metabolism described in the literature.

Here we sought to investigate in vivo if [18F]-Florbetapir, [11C]-PIB, and [18F]-FDG could detect the amyloid deposition and the changes in brain glucose consumption of 5XFAD mice.

Section snippets

Animals

Heterozygous 5XFAD transgenic mice (n = 10) and their WT (n = 12) littermates were used. Breeding progenitors were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Genotypes were determined by polymerase chain reaction analysis of DNA obtained from ear biopsies. Mice were housed in controlled laboratory conditions with the temperature maintained at 21 ± 2 °C and humidity at 55 ± 10%. Food and water were available ad libitum. Animal procedures were conducted in strict accordance with

[18F]-Florbetapir

The image quality provided by [18F]-Florbetapir was reasonably high but because of the variability of injected doses and animal weight, the differences between transgenic and WT animals were too small to be visually detected (Fig. 2A). However, image quantification revealed that transgenic animals presented a significantly higher uptake of [18F]-Florbetapir (p < 0.001). However, the magnitude of the difference was relatively low because 5XFAD animals only presented a 14.5% mean increase in

Discussion

In agreement with a recent study published with another AD murine model (Manook et al., 2012), we found a significant increase in [11C]-PIB binding in 5XFAD animals compared with WT animals using specific activities in the conventional range used for human studies. Previous works that failed to find any significant increases in the [11C]-PIB binding in murine models of AD argued that the affinity of the [11C]-PIB binding site was lower in mouse models than in patients. Consequently, it has been

Disclosure statement

The authors disclose no conflicts of interest.

Animal procedures were conducted in strict accordance with the guidelines of the European Community Directive 86/609/EEC regulating animal research and were approved by the local ethical committee for the use of animals in research and scientific purposes of Barcelona Biomedical Research park (CEEA-PRBB).

Acknowledgements

This study was supported by grant SAF2009-13093-C02-02 and CSD2010-00045 from the Spanish Ministerio de Ciencia e Innovación (MCINN), MOIF-CT-2006-039637 from the European Union, 2009/SGR/214 from the Generalitat, Catalonia, and CDTI under the CENIT Programme (AMIT Project). The authors thank Ms Carme Xaus for her skillful technical assistance in the histology procedures and Dr Roser Cortés for comments and suggestions on the microscope procedures.

References (34)

Cited by (0)

View full text