Calorie restriction attenuates astrogliosis but not amyloid plaque load in aged rhesus macaques: A preliminary quantitative imaging study

Abstract

While moderate calorie restriction (CR) in the absence of malnutrition has been consistently shown to have a systemic, beneficial effect against aging in several animals models, its effect on the brain microstructure in a non-human primate model remains to be studied using post-mortem histopathologic techniques. In the present study, we investigated differences in expression levels of glial fibrillary acid protein (GFAP) and β-amyloid plaque load in the hippocampus and the adjacent cortical areas of 7 Control (ad libitum)-fed and 6 CR male rhesus macaques using immunostaining methods. CR monkeys expressed significantly lower levels (∼30% on average) of GFAP than Controls in the CA region of the hippocampus and entorhinal cortex, suggesting a protective effect of CR in limiting astrogliosis. These results recapitulate the neuroprotective effects of CR seen in shorter-lived animal models. There was a significant positive association between age and average amyloid plaque pathology in these animals, but there was no significant difference in amyloid plaque distribution between the two groups. Two of the seven Control animals (28.6%) and one of the six CR animal (16.7%) did not express any amyloid plaques, five of seven Controls (71.4%) and four of six CR animals (66.7%) expressed minimal to moderate amyloid pathology, and one of six CR animals (16.7%) expressed severe amyloid pathology. That CR affects levels of GFAP expression but not amyloid plaque load provides some insight into the means by which CR is beneficial at the microstructural level, potentially by offsetting the increased load of oxidatively damaged proteins, in this non-human primate model of aging. The present study is a preliminary post-mortem histological analysis of the effects of CR on brain health, and further studies using molecular and biochemical techniques are warranted to elucidate underlying mechanisms.

Introduction

Calorie restriction (CR) without malnutrition slows the aging process and prolongs median and maximum lifespan in yeast, worms, flies, fish, and rodents (Kennedy et al., 2007, Anderson et al., 2009, Anderson and Weindruch, 2010). Three independent studies of CR in non-human primates (Macaca mulatta) have demonstrated that CR confers a protective effect against aging-related diseases (Bodkin et al., 2003, Colman et al., 2009, Mattison et al., 2012), although consensus has not been reached regarding the beneficial effect of CR on survival. The beneficial effects of CR in nonhuman primates include prevention of age-related loss of muscle mass, fewer incidences of cardiovascular disease, increased insulin sensitivity and glucose tolerance, lower cancer incidence, preservation of critical gray and white matter regions, and protective changes in gene expression (Colman and Anderson, 2011, Kemnitz, 2011). The specific mechanisms by which CR delays aging and the onset of age-associated disease have yet to be elucidated. It has been well established that CR is associated with reduced levels of oxidative damage and reduced inflammatory tone (Heilbronn and Ravussin, 2003, Weindruch, 2003), outcomes that may be particularly important with respect to brain aging.

Recently, our group has demonstrated the wide-ranging effects of CR in the rhesus macaque brain using neuroimaging techniques. Brain atrophy, a characteristic aging change in both humans and non-human primates, is attenuated by CR in rhesus macaques, particularly in the midcingulate cortex, bilateral lateral temporal cortex, and right dorsolateral frontal cortex (Colman et al., 2009). CR also preserves white matter integrity in the fronto-occipital fasciculus, superior longitudinal fasciculus, external capsule, and brainstem (Bendlin et al., 2011). In addition, age-related iron accumulation in the basal nuclei, red nucleus, and parietal, temporal, and perirhinal cortices is attenuated with CR, and this in turn is associated with improved performance on motor function tests (Kastman et al., 2010). Consistent with improved inflammatory tone, CR moderates the effect of important plasma-based inflammatory (e.g. IL-6) and vascular (e.g. homocysteine) markers on gray and white matter changes in several brain regions that are sensitive to aging (Willette et al., 2010, Willette et al., 2012a). Furthermore, CR improves glucoregulatory profiles in these animals and positively influences gray matter volume in the hippocampus and motor task performance (Willette et al., 2012b). While these findings point to an overall salubrious effect of CR on the brain, it is important to confirm and extend the effect of CR by examining neuropathological indices using post-mortem histologic techniques.

Several neuropathological alterations are associated with aging in the brain. The expression level of glial fibrillary acidic protein (GFAP), a marker of astrocytic activation, increases with age in rodents, rabbits, monkeys, and humans (Finch, 2003). This increased GFAP expression during aging has been suggested to be a consequence of the increased load of oxidatively damaged proteins, which occur in tissues throughout the body (Finch, 2003, Middeldorp and Hol, 2011). In rats, CR attenuates the age-associated increase in glial activation (Morgan et al., 1997, Kaur et al., 2008). Additionally, reactive astrogliosis is also predominant in several neurodegenerative diseases, including Down syndrome (Trisomy 21), Parkinson disease, Huntington disease, and Alzheimer disease (AD), and is suggested to be secondary to marked neurodegeneration and neuronal death characteristic of these diseases (Middeldorp and Hol, 2011). Higher GFAP expression levels are associated with AD in humans (Beach et al., 1989). Moreover, subjects with the ApoE ε4 allele (a risk factor for developing AD) express higher levels of GFAP than non-APOE ε4 carriers (Overmyer et al., 1999).

Another common histological change that occurs with aging in both monkeys and humans is a progressive increase in the formation of amyloid plaques (Heilbroner and Kemper, 1990, Sloane et al., 1997, Anderton, 2002). These are formed by the abnormal aggregation of amyloid-β peptide (Aβ), a small peptide that is involved in the pathogenesis of AD in humans (Zhang et al., 2012). Although Aβ deposition and plaque formation are signature pathologies of AD in humans, they are also commonly present in the brains of cognitively normal older adults (Rodrigue et al., 2009). Rhesus monkeys do not develop AD, but amyloid plaques are detected in the cortex in aged animals (Heilbroner and Kemper, 1990, Sloane et al., 1997, Uno, 1997). In this way, AD is associated with amyloid plaques but plaques are not always associated with cognitive impairment. Interestingly, both in rodent models of AD and in human AD, there is increased GFAP expression adjacent to Aβ plaques (Hanzel et al., 1999, Gallagher et al., 2012). Human AD amyloid plaques are surrounded by astrocytes expressing high levels of antioxidant enzymes, including superoxide dismutase, suggesting a role for oxidative stress in mediating age-associated astrocytic activation (Furuta et al., 1995).

The current report is an interim post-mortem histological analysis of the effects of age and CR on brain health. We examined GFAP expression and Aβ plaque distribution in the male rhesus macaque hippocampus and the adjacent entorhinal cortex (EC). The hippocampus exhibits extensive astrogliosis both during normal aging and in neurodegenerative states such as AD (Nichols et al., 1993). Since amyloid plaques in the aged rhesus brains preferentially accumulate in association cortical areas (Sani et al., 2003), we also examined the overlying neocortex (mainly temporal) for Aβ plaque distribution. We hypothesized that CR would lead to a decrease in the expression of GFAP and Aβ plaque frequency.