With the increase in the average life expectancy of the population worldwide, cognitive frailty is emerging as one of the most important issues of human health, with a rising percentage of the population affected by cognitive decline and dementia. However, it remains impossible to delineate the precise moment at which the brain begins to age (Fjell et al., 2014).

Alzheimer’s Disease (AD) is the most prevalent dementia and its progression is influenced by both genetic and environmental factors (Grupe et al., 2007; Avramopoulos, 2009; Harold et al., 2009). These gene–environment interactions may influence and trigger pathogenic pathways that determine the severity and progression of several degenerative diseases (Mastroeni et al., 2009; Bradley-Whitman and Lovell, 2013; Millan, 2014; Griñán-Ferré et al., 2015). In addition, a growing body of evidence demonstrated that the effects of environmental factors are mainly orchestrated by epigenetic mechanisms. In AD, only a small percentage of the cases (<5%) are consistent with inheritance. Even in the familial forms of AD, penetrance is not complete and is also highly variable between individuals with the same genetic background (Mastroeni et al., 2009, 2011). Consequently, the effect of environmental factors is of remarkable importance. In this respect, epigenetic changes, or other forms of gene-environment interaction, could be involved in determining the onset of AD and also aging, which is the most important risk factor for development of the pathology (López-Otín et al., 2013; Spiegel et al., 2014).

In particular, lines of evidence indicate that the methylation and hydroxymethylation state of the DNA plays an important role in controlling gene expression, and also learning and memory, including memory formation and maintenance of learning and memory decline during aging (Levenson et al., 2004; Miller and Sweatt, 2007; Liu et al., 2009; Day and Sweatt, 2010; Mikaelsson and Miller, 2011; Morris et al., 2014).

Similarly, epigenetic mechanisms, i.e., alterations of histone proteins and DNA, are essential for hippocampal synaptic plasticity and cognitive function disease (López-Otín et al., 2013). Additionally, recent works (Holliday, 2006; Fischer et al., 2007; Gräff and Tsai, 2013) demonstrate that chromatin regulation, and in particular histone acetylation, could be important for modulating the cognitive outcome of several neurodegenerative disorders. Moreover, exposure to an EE has been shown to have a protective effect in mouse models by slowing disease progression and reducing AD-like cognitive impairment (Barak et al., 2013).

Although the mechanisms by which EE preserves cognition remain unknown, our work team recently demonstrated that EE is an experimental strategy to alleviate cognitive decline and neuroprotection in Senescence Accelerated Mice P8 (SAMP8) (Griñán-Ferré et al., 2015). Moreover, changes in modulators of epigenetic machinery, such as LSD1 or CoREST, were found changed after EE in this strain and were accompanied by cognitive feature improvements (Griñán-Ferré et al., 2015). These results correlated with the activation of endogenous neuroprotective pathways, thus implicating these in the beneficial effects of EE on cognition.

Many studies have demonstrated that EE changes promoter DNA methylation states while changing the expression of DNA MethylTransferases (DNMTs) (Christensen et al., 2009; Marsit et al., 2009; Madrigano et al., 2011; Barrès et al., 2012). Therefore, gene-environment interactions, mediated at the epigenetic level, may be an intermediary step in providing an appropriate organism response to changes in the environment (Irier et al., 2014).

There is evidence that EE attenuated Oxidative Stress (OS) through elevated Nrf2 levels in hippocampi (Snow et al., 2015). In addition, EE elicits anti-inflammatory effects through interaction with several immune signaling pathways, including Interleukin 6 (IL-6). Diminution in the proinflammatory process after EE might be due by changes in NF-κB translocation to the nucleus, reducing chemokines (Costa et al., 2007; Jurgens and Johnson, 2012; Williamson et al., 2012).

The goal of the present work was to delve deep into the epigenetic mechanisms influenced by EE that rendered the previously described beneficial effects by preventing cognitive impairment and neuronal dysfunction. We focus on antioxidant and inflammaging processes modulated by EE exposure in SAMP8.

In the present study, we showed that EE affected epigenetic markers and produced changes in the transcriptional control of oxidative and inflammaging markers in SAMP8. We hypothesize that changing the environment protectively would modulates a large constellation or at least a large subset of cellular phenotypes of aging. SAMP8 comprises a well-enough, but still not fully characterized, model for studying brain aging and neurodegeneration (Takeda, 2009; Pallàs, 2012). Experimental studies on SAMP8 postulated it as a non-transgenic murine model for late-onset AD (Pallàs et al., 2008; Morley et al., 2012b; Pallàs, 2012; Cheng et al., 2014), in addition to being a spontaneous senescence-accelerated mouse strain. These mice exhibited cognitive and emotional disturbances from young ages, probably due to early development of brain pathological hallmarks, such as OS, inflammation, and activation of neuronal death pathways, which mainly affects cerebral cortex and hippocampus (Canudas et al., 2005; Sureda et al., 2006; Cristòfol et al., 2012; Morley et al., 2012a). We have revealed the first, to our knowledge, clue regarding epigenetic modulation in inflammation and oxidative processes in SAMP8 (Alvarez-López et al., 2013) and in a model of familial AD (Griñán-Ferré et al., 2016). Therefore, an adequate epigenetic intervention during early life would be able to have influence on the dysfunctional or harmful molecular mechanisms affected in such a way that this would be able to exert a positive influence (Cosín-Tomás et al., 2014).

Molecular changes in young-age SAMP8 are based on cerebral and cognitive dysfunction (Yuan et al., 2012; Alvarez-López et al., 2013), and these harmful changes were partially reverted in SAMP8 after EE early exposure, including cognitive, psychoemotional, and neurodegenerative hallmarks of AD (Griñán-Ferré et al., 2015). In addition to neuroprotection, EE was able to induce some type of epigenetic modulation in this strain, due to changes in CoREST and LSD1, in both hippocampus and cortex. Therefore, to open the focus on and to probe this epigenetic control, we demonstrated here that the DNA transcriptional control of specific genes indicated its being the initial and key step in the molecular stream developed in SAMP8 by EE, which improves cognition and ameliorates harmful factors gated to neurodegeneration, such as neuronal loss and apoptosis prevention, downregulation of kinase activity, and reduction in OS.

DNA methylation influences hippocampal memory formation, and growing evidence suggests that the regulation of epigenetic processes by EE, stress, and/or hormones can participate in memory function. In the present work and in accordance with the decrease in LSD1 transcription described in Griñán-Ferré et al. (2015), a higher degree of DNA hypermethylation was found in SAMP8 EE hippocampus. Conversely, only DNMT3b was reduced, whereas other DNMT were not modified. Usually, 5-mC decreases transcriptional access to DNA, although the functional effects of this gene silencing depend on the genes that are altered (Alvarez-López et al., 2013). However, recent data suggest a more complex role for 5-mC, which also includes stimulation of gene expression (Weber et al., 2007; Wu and Zhang, 2010). Both, increases and decreases of 5-mC have been described during aging (Richardson, 2003). Furthermore, pathological DNMT activity and aberrant 5-mC formation have been linked with neurodegeneration and apoptotic neuronal death (Chestnut et al., 2011; Hernandez et al., 2011). In these latter studies, neuroprotection was provided by DNMT inhibitors.

5-hmC is another epigenetic marker, catalyzed by the TET family, that appears to be particularly susceptible to developmental and aging-associated modifications (Flax and Soloway, 2011; Ito et al., 2011). According with other studies, diminution in 5-hmC, which paralleled a decrease in the Tet1 expression level, was found after EE in SAMP8. For instance, Irier et al. (2014) found that aged mice exposed to EE improved learning and memory, reducing 5-hmC-abundant hippocampus. However, in this work, no changes in Tet1 expression were found, and differences can be explained by different timing in EE and also because of differences in strain and age. In sum, aging-associated 5-hmC alterations appear to be likely participants in the neuroplasticity of the aging brain.

In addition to DNA methylation changes, histone alterations are one of epigenetic mechanisms that are essential for hippocampal synaptic plasticity and cognitive function diseases (Zhao et al., 2012). In total, histone acetylation increases gene transcriptional activity by relaxing chromatin (de Ruijter et al., 2003). The role of histone acetylation in memory has been intensively explored in recent years. Increasing histone H3 and H4 acetylation by means of HDAC inhibitors in the dorsal hippocampus enhances several types of hippocampal-dependent memories (Stefanko et al., 2009; Haettig et al., 2011), and also rescued hippocampal memory deficits in mouse models of AD (Ricobaraza et al., 2009; Kilgore et al., 2010) and on aged rats (Neidl et al., 2016). Different classes of HDAC can be distinctly involved in memory formation or in neurodegeneration. For instance, Hdac1 expression is upregulated in a mouse model for Huntington’s disease (Quinti et al., 2010). However, mice lacking or overexpressing neuronal HDAC1 from early developmental stages are viable and display normal memory function (Guan et al., 2009). Here, a significant increase in H3 and H4 acetylation levels were found after EE in SAMP8. Although changes in Hdac1. Sirt2, and Sirt6 revealed an increase in gene expression, Hdac2 gene expression was significantly reduced, pointing to a major role of this deacetylase in the beneficial action of EE on the SAMP8 strain (Griñán-Ferré et al., 2015). Interestingly, HDAC2 is one of the major regulators involved in the regulation of long-term memory. Specifically, mice overexpressing HDAC2 exhibited memory impairment, while HDAC2-knockout mice demonstrated enhanced synaptic plasticity (Guan et al., 2009).

On the other hand, and in the same line of all previously mentioned herein, we found an increase in two transcriptional regulatory pathways; Nrf2 and NF-κB, despite the hypermethylation induced by EE. Thus, we can deduce that EE-induced hypermethylation of low-density CpG islands were the promoter for Nrf2, the Antioxidant Response Element (ARE), and for NF-κB.

It is well known that OS possesses a pre-eminent role in the pathogenesis of AD, and in addition, decreased expression of Nrf2 has been reported in AD (Cao et al., 2015). In response to OS, Nrf2 is translocated from the cytoplasm to the nucleus and subsequently binds with ARE, in turn promoting the expression of a variety of OS-related genes (Alam et al., 1999).

We found that in SAMP8 EE, there was significant Nrf2 nuclear translocation, and this in turn could induce the transcription of Heme oxygenase (decycling) 1 (Hmox1), Aldehyde oxidase 1 (Aox1), and Cyclooxygenase 2 (Cox2) genes. Hmox1 can exert cytoprotective effects against OS, whereas Aox1 and Cox2 mediated Reactive Oxygen Species (ROS). These latter factors can act as hormetic regulators in an oxidative environment, favoring the cellular defenses confronting OS (Cai et al., 2012; Maeda et al., 2012). Other target genes, such as Aldehyde dehydrogenase 2 (Aldh2), were not affected (data not shown). It is also relevant that memory improvement in Nrf2-level modulation has been reported in aged APP/PS1 mice (Kanninen et al., 2009) and in memory-impaired rats (Dwivedi et al., 2013).

On the other hand, the high levels of inflammatory mediators can increase a physiological response to certain pathological conditions in the nervous system, as AD or stroke, and It has been suggested that modulation of the signaling pathways of these mediators could be a potential therapeutic approach (Buga et al., 2013; Carriba and Comella, 2015). Here, we have studied NF-κB, that is a transcription factor in positive feedback with and exercising a regulatory role in relationship to classic pro-inflammatory signals that also impacts cell survival and apoptosis, allowing cells to adapt and respond to environmental changes (Oeckinghaus and Ghosh, 2009). In a previous work, we described that EE induced a significant decrease of IL-6 gene expression in the SAMP8 EE group compared with SAMP8 Ct (Griñán-Ferré et al., 2015). IL-6 is a pleiotropic inflammatory cytokine secreted by activated glia in the Central Nervous System (CNS) and is involved in mood disorders such as depression, also in aging process and the pathogenesis of neurodegenerative diseases such as AD (Quintanilla et al., 2004; Popa-Wagner et al., 2014). Our results demonstrate significant lowest genetic expression for IL-6, and also for Cxcl10, but not for Tumor necrosis factor alpha (Tnf-α) in the SAMP8 EE group. C-X-C motif chemokine 10 (Cxcl10) CXCL10 is secreted by several cell types in response to InterFeroN gamma (IFN-γ) and was found increased in AD mouse models (Zaheer et al., 2013). These results coincide with a decrease in the nuclear translocation of proinflammatory transcription factor NF-κB in SAMP8 mouse EE.

Taken together, our results showed that crucial epigenetic modifications were demonstrated in the SAMP8 EE paradigm, reducing proinflammatory and OS pathways. Therefore, epigenetics orchestrated the homeostasis among DNA methylation, histone acetylation, and transcription factors, acting as a crosstalk mechanism and leading cognitive improvement and neuroprotection after EE in SAMP8 (Figure 6).

CG-F, DP-I, VP-A, DP-C, and JC-A performed the experiments. CG-F, MTR, and DP-C carried out the experimental intervention. CG-F and VP-A performed behavior experiments. DP-I and JC-A performed ELISA, and western blot analysis. CG-F and DP-I performed the RT-PCR experiments. CG-F, AC, and DO-S analyzed the data and drafted the manuscript. MP designed the experiments and supervised the study. All authors read and approved the final manuscript. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.