AGED BEAGLE DOGS DEMONSTRATE REDUCED BRAIN METABOLISM MEASURED BY FDG-­‐PET-­‐MR Joseph A Araujo1,2, Kate Sokolnicki3, Jacob Hesterman3, Jack Hoppin3, Dale E. Araujo4, Howard Dobson2 1Dept. of Pharmacology, University of Toronto, Toronto; 2InterVivo Solu@ons Inc., Toronto; 3InviCRO, LLC., Boston; 4VivoCore Inc., Toronto; 5CanCog Technologies Inc., Toronto Novel models with greater translaUonal value have been recognized as being necessary for evaluaUng Alzheimer’s disease (AD) therapeuUcs. Here we describe several features of the aged dog that supports its value as a) a model of AD progression, and b) as a tool for either rapid or longitudinal screening of novel AD therapeuUcs. The aging dogs models aspects of both the pathophysiology and cogniUve decline observed in Alzheimer’s disease progression. The current study assessed the effects of age and task on brain-­‐region specific uptake of 18F-­‐FDG in young and aged dogs. Fig. 4. Immunohistochemical stain with mAb tau AT-­‐8 in aged dog brain in the cerebral cortex (adapted from Papaioannou et al., 2001). Rest 1 Rest 2 0.4 0.2 0 Young Fig. 7. Example of generated brain atlas used for region of interest analyses. Rest 2 0.2 0 Young Aged Parietal Lobe Temporal Lobe Parietal Lobe % (ID/g) Normalized to Cerebellum 1.2 1.1 1 0.9 0.8 Old 0.7 Young 0.6 0.5 1 2 3 1.3 1.2 1.1 1 0.9 Old 0.8 Young 0.7 0.6 0.5 4 1 2 3 Test Test 1.2 1.1 1 0.9 0.8 Old 0.7 Young 0.6 0.5 1 2 4 Frontal Lobe Occipital Lobe 18F-­‐FDG PET-­‐MR data were acquired on six young (mean age 2.3 years) and six aged (mean age 12.6 years) dogs. Repeatability was examined using two scans separated by 48 hours, and effect of cogni?ve load was assessed following simple object discrimina?on learning and more complex variable object discrimina?on (Fig. 7). An isotropic anatomical atlas of the canine brain was developed using six T1-­‐weighted anatomical MR scans (Mean age=6.66 years) registered to form an average MR brain template for region of interest analysis (Fig. 8). Effect of age and task on region specific uptake (% ID/g normalized to cerebellum) was evaluated using repeated-­‐
measures ANOVA. Rest 1 0.4 3 Frontal Lobe % (ID/g) Normalized to Cerebellum Methods Aged 0.8 0.6 Effect of age on FDG uptake Fig. 6. A variable object discrimina?on task reveals both age related deficits in performance (A) and processing speed (B) in Beagles. EFFECT OF AGE ON BRAIN METABOLISM Occipitla Lobe %(ID/g) Normalized to Cerebellum 0.6 1 Fig. 9. FDG uptake in young and old dogs. Repeatability across 48 hours was excellent [p<0.002], especially in young dogs. Error bars represent SEM. Aged dogs show domain specific and progressive cogniUve decline Aged dogs show cogni?ve impairment that is correlated with pathology and is domain specific, such as on measures of selec?ve aJen?on (Fig. 6). Repeatability Occipital Lobe 1 0.8 1.2 1.1 1 0.9 0.8 Old Young 0.7 0.6 0.5 4 1 2 Test 3 4 Test Fig. 10. FDG uptake in young and old dogs across brain regions. Uptake was significantly lower in aged compared to young dogs across several brain regions. Error bars represent SEM. Effect of task on FDG uptake 0.6 0.58 0.56 0.54 0.52 Old 0.5 Young 0.48 0.46 0.44 1 2 3 0.91 0.89 0.87 0.85 0.83 0.75 1 1.1 1 0.95 0.9 Old 0.85 Young 0.8 0.75 Test 2 3 Test 1.05 2 Young 0.79 0.77 Test 1 Old 0.81 3 Parietal Lobe % (ID/g) Normalized to Cerebellum Intraneuronal hyperphosphoylated tau is also detected in aged dogs that is similar in some respects to the neurofibrillary tangle pathology seen in AD (Fig. 4). COGNITION 1.2 1.2 1.15 1.1 1.05 1 0.95 0.9 0.85 0.8 0.75 Old Young 1 2 3 Test 1.1 1.05 1 0.95 0.9 Old 0.85 Young 0.8 0.75 1 2 Test 3 Occipital % (ID/g) Normalized to Cerebellum Fig. 3 Signal to noise (S/N) ra?o for detec?ng oligomeric forms of Aβ in brains of young and aged Beagles. Oligomeric Aβ is exponen?ally higher in aged dogs, as is variability, compared to young dogs using the A4 assay developed by Amorfix (Toronto, Canada). Repeatability Frontal Lobe 1.2 Forebrain % (ID/g) Normalized to Cerebellum Oligomeric Aβ is also increased in the aged brain (Fig. 3). Repeatability at 48 hours
Somatosensory Cortex % (ID/g) Normalized to Cerebellum Fig. 2. Cor?cal Aβ deposi?on visualized using immunohistochemical staining in (A) cogni?vely normal aged human (B) cogni?vely normal aged Beagle, (C) cogni?vely impaired aged Beagle, and (D) Alzheimer’s pa?ent (Adapted from Head et al., 2000). Results Frontal Lobe % (ID/g) Normalized to Cerebellum Endogenous Aβ is naturally deposited into plaques of the diffuse subtype in aged dogs, which are vulnerable to post transla?onal modifica?on and are fibrillar at the ultrastructural level -­‐ cerebral amyloid angiopathy (CAA) is also found in aged dogs. The paJern of Aβ deposi?on parallels that seen in humans and occurs over a 3-­‐ to 4-­‐ year window permiMng the examina?on of interven?ons that may slow or halt deposi?on (Fig.2). Fig. 5. Magne?c resonance imaging (MRI) of an aged (le_ panel) and young (right panel) Beagle. The aged (cogni?vely impaired) Beagle shows prefrontal and hippocampal atrophy as well as increased ventricular volume compared to the young, cogni?vely intact, Beagle (see Tapp et al., 2004). Fig. 8. Dogs were trained to respond to one (e.g, banana) of two objects in simple object discrimina?on learning. In the variable paradigm, the incorrect object distractor was also presented (0-­‐3 ?mes per trial). Temporal Lobe % (ID/g) Normalized to Cerebellum Fig. 1. Representa?ve MOLDI-­‐TOF mass spectra from (a) human and (b) canine CSF samples showing a similar paJern of Aβ isoforms (adapted from Portelius et al., 2010). CONCLUSIONS Occipital Lobe % (ID/g) Normalized to Cerebellum Canine amyloid-­‐β (Aβ) protein precursor shows approximately 98% homology to the human sequence and the protein is processed into Aβ isoforms that are analogous in concentra?on and iden?cal in sequence to that seen in humans (Fig. 1). AddiUonal correlates of AD in aged dogs include: 1) Increases in oxida?ve stress; 2) Neuronal loss and reduced neurogenesis; 3) Reduced markers (e.g. N-­‐acetyl aspartate) of neuronal health; 4)  Cholinergic deficits; and 5) Cor?cal atrophy (Fig. 5). Entorhinal % (ID/g) Normalized to Cerebellum Aged dogs exhibit both early stage amyloid-­‐β and tau pathologies Motor Cortex % (ID/g) Normalized to Cerebellum PATHOLOGY 1.15 1.1 1.05 1 0.95 0.9 0.85 0.8 0.75 Old Young 1 2 3 1)  The aged dog is a unique natural model of AD progression. 2)  Aβ pathology in aged dogs shows similar deposiUon, idenUcal protein isoforms and oligomeric forms as those seen in humans. 3)  AddiUonal pathology includes the development of tau-­‐like pathology, increases in oxidaUve stress, neuronal loss , decreased neurogenesis, reduced neuronal health, cholinergic deficits and brain atrophy. 4)  The pa_ern and progression of cogniUve decline in translaUonal neuropsychological tests are consistent with that seen in AD conversion. 5)  Here we examined the effects of age and task on uptake of FDG in young and aged dogs. FDG uptake measures were highly repeatable in scans separated by 48 hours. 6)  FDG uptake was significantly lower in aged dogs compared to young dogs across several brain regions, indicaUng that brain metabolism declines with age. 7)  There was evidence that FDG uptake varied by task, parUcularly in young dogs, which may reflect efficient recruitment of task-­‐
dependent brain regions. However, addiUonal studies are warranted to confirm this. 8)  CollecUvely, this supports the use of aging dogs for examining disease modifying AD therapeuUcs targeUng early stages of the AD and suggests that FDG-­‐PET may provide a useful biomarker in these endeavors. Test Fig. 11. FDG uptake varied by task condi?on primarily in young dogs. In entorhinal and forebrain, uptake increased with task difficulty. Reduced uptake was also evident in other brain regions. Error bars represent SEM. Email: [email protected] www.intervivo.com