Redox lipidomics to better understand brain aging and function

Human prefrontal cortex (PFC) is a recently evolutionary emerged brain region involved in cognitive functions. Human cognitive abilities decline during aging. Yet the molecular mechanisms that sustain the preservation or deterioration of neurons and PFC functions are unknown. In this review, we focus on the role of lipids in human PFC aging. As the evolution of brain lipid concentrations is particularly accelerated in the human PFC, conferring a specific lipid profile, a brief approach to the lipidome of PFC was consider along with the relationship between lipids and lipoxidative damage, and the role of lipids in human PFC aging. In addition, the specific targets of lipoxidative damage in human PFC, the affected biological processes, and their potential role in the cognitive decline associated with aging are discussed. Finally, interventions designed to modify this process are considered. We propose that the dysfunction of key biological processes due to selective protein lipoxidation damage may have a role the cognitive decline of PFC during aging.

. The meaning of lipid oxidation and lipoxidation-derived protein adducts 4. Redox lipidomics and adductomics in human prefrontal cortex aging 5. Lipoxidation-derived protein damage targets in human prefrontal cortex 6. Advanced lipoxidation end-products and soluble oligomers 7. Interventions against prefrontal cortex aging 8. Healthy aging brain as prevention for neurodegenerative diseases 9. Conclusions

Introduction
One critical trait of human evolution is the rapid expansion of brain size and the emergence of unique and complex cognitive capabilities. These changes have demanded metabolic adaptations, in particular with respect to increased energy supplies [1]. Thus, humans allocate about 20% of their total energy to the brain, compared with 12% for apes and 5% for other mammalian species [2]. This higher metabolic activity, supported by increased expression of genes related to neuronal functions and energy metabolism [3,4], as well as the increased density of glia cells relative to neurons [5,6], shows its greatest demand in human prefrontal cortex (PFC) [5,7]. Furthermore, PFC has been shown to undergo greater expansion at the gross anatomy level in human evolutionary lineage than other brain regions [8]; it is the brain region that has emerged and evolved most recently during primate evolution. PFC is involved in complex cognitive functions including reasoning, planning, social behavior, and general intelligence [9][10][11].
Cognitive abilities decline during aging. Yet brain aging does not affect all neurons identically. Non-invasive techniques such as structural brain imaging and functional magnetic resonance imaging (MRI) are being widely used and combined with postmortem analyses of aging human brain to identify the most age-sensitive regions of the brain. Volumetric brain measurements as well as positron emission tomography (PET) measures of whole brain and regional cerebral metabolic rates of glucose (CMRg) reinforce the idea that aging affects some brain regions more than others. Thus, in addition to the volume changes in, particularly, the association cortex during aging, it has been shown that the PFC shows the greatest and most consistent decrements in CMRg as compared with all other regions of the aging brain [12][13][14][15]. In this line, cognitive functions that rely on PFC, such as learning, memory, and executive functions, show considerable age-related decline. Consequently, it is plausible to postulate that the decline in both energy metabolism and cognition are closely linked.
These changes in metabolic activity in PFC during aging have been associated with different, and non-exclusive, molecular mechanisms and biological processes comprising altered neuronal expression of proteins involved in the metabolic state of neurons and glial cells, altered activation of synapses, and the disruptive effects of abnormal neuronal excitability. But, in any case, the evidence clearly suggests that neuronal loss during aging is very modest, whereas a decrease in the number of synapses appears to be a more prominent trait of brain aging. Indeed, most of the functional decline associated with physiological aging is caused by relatively subtle changes, such as loss of dendrites, reductions in spine density, altered spine morphology, changes in the molecular profile of synapses, and a general loss of synaptic plasticity, all of which is paralleled by declining cognitive function and elevated risk of neurodegenerative disease [16][17][18][19][20][21][22]. Nevertheless, the molecular mechanisms behind aging-related phenotypic changes are only scarcely understood.
Lipids have allowed and lent support to the brain's evolution toward complexity [23]. In addition to their quantitative relevance (50% of the dry matter [24]), brain lipids show great structural and functional diversity, with most lipid categories represented in cells as expressions of the different needs and functions ascribed to them which are related to the generation of membranes, cell signaling, and energy storage [23,25]. By regulating the chemical and physical properties of membranes, lipids direct vesicle fusion and fission processes, ion flux, and lateral diffusion of membrane proteins, they also create the microdomains necessary for optimal cellular communication, among other functions [26] [ 27,28]. However, despite growing evidence of the critical role of lipids in brain function, little is known about the participation of lipids in PFC aging. In this review, we used data from targeted and untargeted mass spectrometry (MS)-based lipidomics and redox lipidomics to characterize the lipidome composition of human PFC and its association with the cognitive decline linked to the aging process.

Lipidomics of human prefrontal cortex
Recent progress in mass spectrometry (MS)-based lipidomics has provided experimental opportunities for the large-scale quantitative profiling of thousands of individual lipid molecular species in human brain, providing novel insights into the mechanisms and evolution of brain functionality. Thus, current data demonstrate the existence of lipidomic signatures which are tissue-specific [29,30] and brain region-dependent [29,31,32], with the lipid profile of the brain seen to be markedly different from that of the other tissues (verified in mammals, non-human primates, and humans) [29]. This supports the idea that lipids play specific roles in brain function [25]. Interestingly, the largest lipidome divergence between the brain and other tissues is found in humans, who show the greatest cognitive complexity. Furthermore, lipidome divergence follows the genetic divergence among animal species, supporting the idea that concentration levels of lipids enriched in brain evolved at a much more rapid pace than the concentration levels of lipids characteristic of non-neural tissues [29]. As stated by Khaitovich and coworkers [29], these features suggest a link between brain lipidome evolution and the evolution of brain functionality.
This specific lipid pattern present in human PFC seems to also be projected and reflected at the cellular, subcellular, and molecular levels. Thus, for example, at the cellular level, by using imaging mass spectrometry (IMS) [33,34] it is possible to generate density maps for several lipid species, mainly glycerophosphocholines (ChoGpl), glycerophosphoethanolamines (EtnGpl), and sphingomyelins (SM), offering important information about the lipidome of the human brain and, in particular, the human PFC.
The IMS images show that the diverse lipid molecular species found in human PFC exhibit a different and specific distribution, and even allow us to distinguish among the six different layers of the PFC based on the slight differences in lipid density throughout the cortex [33] (see Figure 1).
A recent study has used a lipidomic approach to analyze the lipid profile at the mitochondrial and microsomal levels in grey matter of human PFC [35]. Other minor phospholipids such us phosphatidylinositols (PIGpl) and cardiolipin were not analyzed.
At the membrane level, lipid rafts are microdomains which provide a highly-saturated liquid-ordered microenvironment that promotes protein-lipid and protein-protein interactions [36,37]. Lipid rafts comprise a highly dynamic clustering of proteins and lipids playing a central role in signal transduction and intercellular communication. Lipid rafts from human PFC are characterized by their high level of cholesterol and sphingolipids (SM, cerebrosides, and sulfatides), and lower content in GP (particularly EtnGpl and SerGpl, both aminophospholipids, and to a lesser degree ChoGpl and PIGpl) compared to the non-rafts membranes [38] (see Table 1).
Fatty acids are prime components of the structural diversity of lipids of neurons and glial cells. Indeed, with different permutations of head group and fatty acids, up to 10,000 theoretical lipid molecular species could be generated [39]. Fatty acids determine the functional properties of lipids in PFC which can be ascribed to their roles in the structural and functional integrity of membranes, the generation of lipid signaling mediators, and the chemical reactivity of the acyl chains [23]. In human PFC, the fatty acid profile is characterized by an average chain length of 18 carbon atoms, and a relative distribution  Table   2). The content of these PUFAs in PFC is the highest across human brain regions [32], is supported at transcriptional and protein expression levels, and can probably be ascribed to inherent neuronal activity [32]. The specific finding that DHA (22:6n-3) is the main PUFA present in brain regions is in line with previous findings pointing up the very high level of DHA in the whole brain [24]. Interestingly, lipid rafts from human PFC are characterized by their high level of SFA, as well as by reduced amounts of PUFA compared to the non-rafts membranes [38].
The high concentration of PUFAs in neuronal and glial cell membrane GP in PFC not only makes them prime targets for lipoperoxidative damage but also enables them to participate in long free radical chain reactions. Considering that the peroxidizability index (PI) combines the membrane fatty acid composition with the relative susceptibility of individual fatty acids to peroxidation [42], the available evidence indicates that PFC shows the highest value of PI in the human brain and, consequently, helps explain the greater susceptibility of the membrane bilayer to lipid peroxidation [32]. Considering that lipid rafts from PFC show enrichment in SFA, it is proposed that this vulnerability particularly affects non-raft territories of the cell membrane. However, a greater PI is not necessarily associated with a higher degree of molecular damage. In fact, quite the reverse is the case. Thus, the higher PI of the human PFC compared to other human brain regions is associated with a lower steady-state level of lipoxidation-derived protein damage [32,43]. This dissociation is probably due to the lower mitochondrial stress present in PFC (based on the levels of the mitochondrial stress marker succinyl-cysteine (SC)) and the relevant presence of neuronal adaptive response mediated by the antioxidant responsesignaling pathway Nrf2 (Nuclear factor erythroid 2-related factor), as well as by the stress resistance adaptive response mediated by the transcriptional factor REST (RE1-silencing transcription factor, also known as neuron-restrictive silencer factor (NRSF)) [32,44].
Globally, human PFC has a specific lipidome that significantly differs from those of other human brain regions, tissues, and animal species. Inside the PFC, this lipidome varies in the cell layers of the grey matter, the subcellular level (mitochondria and microsomes), and even the membrane levels (non-raft vs lipid raft domains). Furthermore, PFC presents a specific fatty acid profile characterized by the presence of a high content in PUFAs, conferring to PFC a high vulnerability to oxidative conditions, which in turn probably demands the presence of adaptive responses to ensure neuronal survival during adult lifespan.

The meaning of lipid oxidation and lipoxidation-derived protein adducts
Polyunsaturated fatty acids (PUFAs) are abundant constituents preferentially located at the sn-2 position in glycerophospholipids (to pointing up plasmalogen-PE) within cellular membranes. The fact that lipids constitute a major portion of the plasma, and mitochondrial and endoplasmic membranes, establishes the presence of substantial concentrations of PUFAs within membranous structures. It is known that PUFA residues of lipids (and particularly GP) are very sensitive to oxidation, and this sensitivity increases as a function of the number of double bonds per fatty acid molecule [45,46]. In this context, PFC is the human brain region with the highest content in PUFAs which also determines that it has the highest content in double bond and peroxidation indexes [32] (see Table 2).
PUFAs are vulnerable to diverse oxidation reactions and radical reactions, both of which result in the formation of so-called electrophilic lipid products. Certain oxidation reactions involving PUFAs are enzymatically mediated by families of enzymes including the lipoxygenases (LOX), cyclooxygenases (COX), and cytochrome P450 [47]. In contrast to the highly regulated PUFA oxidation mediated by enzymes, the electrophilic lipid products generated by free radical-mediated lipid oxidation are part of the lipid peroxidation process (for a review see [48,49]). Lipid peroxidation occurs in three major phases, comprised of an initiation event, chain propagation, and termination (for more details, see [50,51]). Of note, a single initiation reaction is postulated to result in 200 to 400 propagation cycles, rapidly amplifying free radical damage in highly oxidizing PUFA-rich environments. The resulting electrophilic lipid products are reactive aldehydes characterized by the common group R-CHO consisting of a carbonyl center bonded to hydrogen and an R group arising predominantly as a consequence of oxidative damage within the cellular microenvironment. These reactive carbonyl species (RCS) have signaling properties, but can also be cytotoxic [23,42]. It is estimated that over a hundred different reactive species can be generated, each with a wide range of reactivity, size, and phospholipids are other significant aldehydic products of lipid peroxidation of PUFAs.
As mentioned above, lipid rich membranes provide an optimal environment for producing a large abundance of these cytotoxic compounds. For instance, microsomal and mitochondrial intra-membrane concentrations of these lipid peroxidation-derived electrophilic compounds have been estimated to accumulate to as many as 1-10 mM in vivo [50]. The regulatory function and/or cytotoxicity of these aldehydes depend on abundance, reactivity, and half-life. Half-life varies significantly, with compounds such as ONE and HNE exhibiting half-lives of about 1 second and 2 minutes, respectively. So compared with reactive oxygen species (ROS), RCSs have a much longer half-life. This property along with the non-charged structure of aldehydes allows them to transiently modify distant proteins, lipid membranes (particularly aminophospholipids), and nucleic acids [42]. These properties may contribute to a localized bias for molecular adduction by RCSs, particularly those with short half-life, as cellular components located within the vicinity of lipid peroxidation targets remain more susceptible to generated RCS. By contrast, long-lived RCSs can be more destructive because they may have a more pronounced global impact by altering a broader array of molecular targets due to their transient nature [42,49].
RCS react with nucleophilic groups in macromolecules (lipoxidation reactions) such as proteins, nucleic acids, and aminophospholipids, resulting in their chemical and nonenzymatic modification (see Figure 2). The result is the formation of a diversity of adducts and intra-and inter-molecular cross-links collectively called Advanced Lipoxidation Endproducts (ALEs) [53]. Thus, by reacting with nucleophilic sites in proteins (basically to cysteine (Cys), lysine (Lys), arginine (Arg), and histidine (His) residues), RCSs generate ALE adducts such as MDA-Lys, HNE-Lys, NKT-lys, FDP-Lys, carboxymethyl-lysine (CML), and Scarboxymethyl-cysteine, as well as the cross-links glyoxal-lys dimer (GOLD) and lys-MDAlys, among others [53][54][55]. Several of these compounds have been detected, characterized, and located (by mass spectrometry, redox proteomics, and immunohistochemistry) in the human PFC. RCSs can also react with the exocyclic amino groups of nucleosides to form alkylated products. Guanine is the most vulnerable and commonly modified DNA base because of its high nucleophilicity, while MDA-deoxyguanosine (M1dG) is the most common adduct [56]. Finally, the amino group of aminophospholipids can also react with RCS, leading to the formation of adducts such as MDA-PtdEtn and carboxymethyl-PE [57]. To the best of our knowledge, no data are currently available about lipoxidation-derived damage to DNA and aminophospholipids in human brain.
The molecular consequences of ALE formation in proteins mostly include deleterious structural and functional changes [42]. Thus, ALE formation induces alterations in physico-chemical properties (e.g., conformation, charge, hydrophobicity, and solubility), formation of intra-and inter-molecular protein crosslinks and aggregates, and loss of enzymatic activity, among others; yet it also includes protein activation when the ALE formation is in proteins particularly designed to sense and respond to cellular RCS production. In other words, protein activation seems to be associated with specific protein targets designed to detect RCS which will act as signaling products (see next paragraph). In a deleterious way, ALE formation on nucleic acids induces DNA damage and mutagenesis, and alterations in physico-chemical and biological properties of the lipid bilayer [56,57]. In this context, a relevant recent study demonstrated that MDA causes neuronal mitochondrial dysfunction by directly promoting ROS generation and modifying mitochondrial proteins [58].
In contrast to their cytotoxic effects, RCSs can also work by sending regulatory signals activating specific protein targets which act as adaptive responses designed to decrease lipoxidative damage and improve antioxidant defenses. Among these mechanisms involved in the modulation of oxidative damage by RCS are i) the modification and activation of uncoupling proteins (UCPs) by the RCS hydroxynonenal and the subsequent decrease in mitochondrial ROS production [59], and ii) the activation of the antioxidant response signaling pathway Nrf2 that includes, among others, the expression of enzymes such as glutathione-S-transferase (GST), specifically designed to detoxify reactive carbonyl compounds, and GPx4 (phospholipid hydroperoxide glutathione peroxidase), designed to restore reduced states of membrane fatty acids from phospholipids to ensure membrane lipid homeostasis [60][61][62]. The existence of both mechanisms at the brain level and the relevant content of Nrf2 in PFC suggest a particular adaptive response in the face of the presence of a lipid microenvironment showing the higher unsaturation degree of human brain [32].

Redox lipidomics and adductomics in human prefrontal cortex aging
Aging induces changes at all levels of the biological organization which are offset by allostatic adaptive response mechanisms geared to preserving the composition and function within homeostatic limits. The cell membrane is not an exception [63], and, consequently, the longer the optimal membrane lipid composition is maintained, the better the cell survival and function. These adaptive mechanisms seem to be particularly effective in human PFC since cognitive functions are relatively well-preserved during normal aging. Among these mechanisms, the preservation of the lipid membrane composition seems to be a key component. Lipidomic changes with aging have also been described in lipid raft from human PFC.
Thus, Diaz and coworkers [38], using lipid rafts isolated from PFC of healthy subjects aged 24 to 85, demonstrated that that human cortical lipid rafts are modified by aging in a gender-dependent way. The main changes were in levels of plasmalogens, PUFAs (especially AA and DHA), total polar lipids (mainly InoGpl, SM, sulfatides, and cerebrosides), and total neutral lipids (particularly cholesterol and sterol esters). These findings may provide molecular support for the modifications in protein-protein dynamics and protein clustering in lipid raft underlying alterations in synaptic plasticity, memory retention, and learning during aging [38]. Further studies are, however, needed to consolidate these new ideas. functioning during normal aging [67]. For this reason, it is proposed that during normal human brain aging, the lipid profile is particularly resistant to changes with age as a result of strict control designed to ensure neuronal survival and function.
Although the membrane lipid composition is largely maintained to guarantee cell survival and function, there is a continuous physiological lipoperoxidative attack on the membrane, and the membrane becomes a source of carbonyl compounds with the ability to damage other cellular components. In this line, a recent initial study analyzing selected specific protein damage markers with mass spectrometry in PFC from healthy humans ranging in age from 43 to 86 years old was published [68]. The results demonstrated that there is an increase in the steady-state level of oxidative and lipoxidative protein damage in human PFC over the adult lifespan, with a breakpoint at 60 years of age (the detected and quantified markers for protein oxidation were glutamic semialdehyde and aminoadipic semialdehyde, and for protein lipoxidation carboxymethyl-lysine (CML) and carboxyethyl-lysine (CEL)). In two additional recent studies, changes with age in the protein damage markers derived from the RCS neuroketals (NKT) and malondialdehyde (MDA) in human PFC in two groups of individuals, middle-aged and old-aged, were analyzed. Steady-state levels of these markers were also analyzed as continuous variables in function of age. The levels of NKT-and MDA-protein adducts were significantly increased in old-aged human PFC when comparing the two groups, and they remained significantly increased considering age as a continuous variable [43,69].
Consequently, it is suggested that progressive lipoxidation-derived molecular damage is a conserved, central mechanism of age-related functional decline in human PFC. Indeed, several alterations occur in the human PFC with age including increased levels of cytokines and of mediators of the innate inflammatory response [70], increased cytotoxic oligomeric species [69], regression of dendritic arbors and dendritic spines [71], neurotransmitter deficits [72], and reduction in the volume of the gray and white matter [73].

Lipoxidation-derived protein damage targets in human prefrontal cortex
Although the steady-state levels of ALEs in proteins increase in human PFC during aging, increased lipoxidation is not due to a higher content of the corresponding protein but rather to increased flux of protein damage [69,74]. Importantly, this small but significant pool of lipoxidized protein will probably increase over time because the ALEs detected are limited to those derived from MDA and NKTs. Consequently, further studies are needed to develop a more detailed view.
The identification of specific lipoxidation-modified proteins in the PFC of adult and aged subjects provides an overview of the selective cellular functions that are affected during the aging process. The proteins listed in Table 3 represent the biological processes affected by RCS-related protein damage. If we consider these molecular pathways, we can verify that energy metabolism (with 34.2% of modified proteins, n=12), cytoskeleton under normal conditions, glucose is the exclusive energy substrate for the brain [76]. In addition, CKB, being a key player in the 'phosphocreatine circuit' for cellular energy homeostasis, is in charge of rapid ATP production from phosphocreatine reservoirs in response to acute increased energy demands in neurons [77]. Thus, CKB provides neurons with a reservoir and also an alternative source of ATP to glycolysis, TCA cycle, and respiration.
Cytoskeletal proteins are also lipoxidized in the second major biological process with a great number of modified proteins.  [78,79]. Consequently, general neuronal and glial processes linked to cytoskeleton can become dysfunctional as a consequence of ALE formation in their protein components, but additional functions more specifically linked to neurons which require cytoskeleton integrity can also be affected secondarily to protein lipoxidation [18,19]. Some of these functions are vesicle trafficking, maintenance of the neuronal caliber, axon guidance, synaptogenesis, and synaptic plasticity.
The third major biological process that is affected during aging based on the number of Whereas HSPD1 participates in mitochondrial protein import and macromolecular assembly, folding of proteins, and apoptotic process, among other potential functions, the rest of the proteins (UCHL1, CRYAB, and PARK7) play a relevant role in oxidative stress homeostasis. The case of HSPD1 may be particularly important since decreased content of mitochondrial ETC complexes during aging has been described, without any explanation [68]. So, it is proposed that defects in HSPD1 activity could induce a decreased import of nuclear subunits to the mitochondria with the subsequent deficiencies in complex formation which, in turn, affect energy production. As for the other proteins, UCHL1 participates in processing of ubiquitin precursors and ubiquitinated proteins for proteasomal degradation [80], and CRYAB shows chaperone-like activity and prevents aggregation of proteins under stress conditions [81]. PARK7 is a protein and nucleotide deglycase that catalyzes the degradation of adducts formed between amino groups of proteins or nucleotides and RCS acting as a protein repair system [82][83][84][85]. where it probably plays a role in oxygen transport or in the regulation of cytosolic neuronal oxygen [87]. Lipoxidative damage to these three proteins tags cell oxygen and carbon dioxide regulation as a target of putative cellular respiratory dysfunction in the aged PFC.

Structural characterization of the modified proteins in the healthy adult human PFC with
PredictProtein software analysis (https://www.predictprotein.org/) reveals some shared specific traits which may explain this specificity, rendering proteins more susceptible to oxidative damage [69]. Thus, i) the predominant structures of lipoxidized proteins are alpha helix and loops, ii) most proteins (excepting cytoskeletal proteins) are globular and form soluble coiled-shaped molecules with hydrophobic groups at the core and exposed hydrophilic groups to the medium, and iii) among the most recurrent amino acids encountered in the exposed regions is lysine, in addition to glutamic acid and aspartic acid, a frequent target of lipoxidation on proteins. Another factor that may be important Overall, these findings show that i) the PFC in healthy adults contains lipoxidized proteins, ii) the ALE-modified proteins increase in aged subjects, and iii) the selectivity of molecular damage is associated with specific structural traits and spatial location.

Advanced lipoxidation end-products (ALEs) and soluble oligomers
Aberrant protein structures are able to generate soluble oligomers with a high cytotoxic potential [88]. For instance, NKT-modified proteins generate aberrant structures with cross-linking and aggregation [52]. For this reason, global levels of NKT-protein adducts and the presence of oligomers were analyzed in a recent work [69]. A positive significant relation between NKT-modified proteins and soluble oligomer levels in human PFC was found, suggesting that modifications of proteins parallels soluble oligomer formation.
Furthermore, levels of oligomers and NKT increase with age in the PFC. These findings are complementary to previous observations showing increased expression levels of soluble oligomers in the PFC when comparing middle-aged individuals lacking sAD (Alzheimer disease)-related pathology with older cases with sporadic AD-related pathology stages I-II.
Thus, increased levels of soluble oligomers are indeed a characteristic feature of the aging human brain coincidental with the first stages of AD-related pathology [70]. Importantly, increased levels of oligomeric species parallel regulation of brain cytokines and mediators of the immune response in old age [70]. Whether these coincidences have a cause-effect relationship needs further study using appropriate in vitro and in vivo models. Studies in post-mortem brains serve to demonstrate that such changes do really exist in the human PFC and that they deserve to be investigated deeper.

Interventions against PFC aging
The identification of interventions that delay PFC aging and can be easily implemented in humans is especially important because, to date, no pharmacological treatments able to maintain brain function and/or reduce cognitive decline, the risk of dementia, and neurodegenerative pathologies have been discovered. In recent decades there have been major advances in our understanding of the biology of aging, along with the development of nutritional interventions that delay aging including calorie restriction (CR), intermittent fasting (IF), protein restriction (PR), methionine restriction (MetR), and chemicals that influence pathways linking nutrition and aging processes [89,90].
Growing evidence supports the view that CR/IF/PR/MetR may be helpful in preventing age-related changes in the brain associated with cognitive dysfunction [89,[91][92][93]. CR improves cognitive function, including learning and memory, in old rodents and humans [94][95][96][97], as well as improving the plasticity and recovery of the brain [98,99], possibly mediated by its effects on cell-mitochondrial metabolism [91], nutrient-sensing pathways [89], and specific brain-related mechanisms [100].
One of the most relevant mechanisms proposed to explain this protective effect involves oxidative stress. In this line, a number of studies have shown that CR, as well as PR and MetR, may act at the mitochondrial level, reprogramming and improving mitochondrial metabolism and thereby decreasing free radical production [91,101,102]. In parallel to the improvement in mitochondrial free radical generation, there is a reduction in the steady-state levels of oxidative damage to proteins, lipids, and DNA in animals subjected to CR, IF, PR, and MetR, in whole brain as well as in brain cortex [32,[103][104][105][106][107][108][109][110].
Importantly, the set of brain proteins affected in normal aged rats is protected from oxidative damage by CR in association with improvement in cognitive functions [111].
With regards to lipids, several studies have demonstrated that these interventions (CR, PR, and MetR) have an impact on brain lipid profiles, inducing changes toward a lipid profile that is more resistant to lipid peroxidation which, in turn, leads to a significantly lower lipoxidation-derived protein damage level and, consequently, lower loss of function [32,105,106,108,111]. These interventions can also influence brain lipidome. Thus, it has been found that 80% MetR in mice induces marked changes in the brain, spinal cord, and liver lipidomes [108]. Interestingly, at least 50% of the lipids changed are common in the brain and spinal cord but not in the liver, suggesting a nervous system-specific lipidomic profile for MetR. The differentially expressed lipids basically include specific phospholipid species, which could reflect adaptive membrane responses, and sphingolipids, which could lead to changes in ceramide signaling pathways. In addition, specific oxidation products derived from cholesterol, ChoGpl, and EtnGpl are significantly decreased in brain, spinal cord, and liver from MetR mice. These results demonstrate the importance of adaptive responses of membrane lipids leading to increased stress resistance in brain.
Overall, these data provide a valuable insight into the mechanism of dietary restriction (CR, IF, PR, MetR) on oxidative stress reduction and functional improvements in the aged brain.

Healthy aging brain as prevention for neurodegenerative diseases
Various morphological, biochemical, metabolic and circulatory alterations, which are translated into functional changes, occur in brain aging. The degree of these functional changes will in turn determine whether or not neurodegenerative diseases take place.
When damage exceeds certain limits age-associated cognitive deterioration occurs.
Indeed, it has been shown that the physiopathological processes associated to the disease begin more than 20 years before the clinical onset of dementia [112]. This provides a window of opportunity for finding interventions which can slow or even prevent the progress of the disease. Genetics, along with the lifestyle, nutrition, environmental factors, etc. can prevent the appearance of these diseases that occur with cognitive impairment [113]. This is of the utmost importance, as in many occasions there is no treatment for these diseases once the symptoms have debuted. As stated by our group in [114] "Only prevention makes sense", i.e. maintaining a healthy aging brain is the best way to prevent neurodegenerative diseases. In this regard, there have been some attempts to intervene early in neurodegenerative diseases. The FINGER study is a randomised controlled trial with a multidomain intervention consisting of a 2 years intervention including diet, exercise, cognitive training and vascular risk monitoring, which demonstrated that this multidomain approach could improve or maintain cognitive functioning in at-risk elderly people [115]. Two years later, a 24-month multicentre trial in which prodromal Alzheimer's patients received a specific multinutrient showed that they had better results on the cognitive-functional measurements and less brain atrophy as assessed by magnetic resonance imaging (MRI) [116].
The Lancet Commission stated in 2017: that we should "be ambitious about prevention" [117]. "Active treatment of hypertension in middle aged (45-65 years) and older people (aged older than 65 years) without dementia to reduce dementia incidence. Interventions for other risk factors including more childhood education, exercise, maintaining social engagement, reducing smoking, and management of hearing loss, depression, diabetes, and obesity might have the potential to delay or prevent a third of dementia cases".
Therefore, maintaining a healthy brain throughout all life by healthy lifestyle habits seems to be one of the best ways to prevent neurodegenerative diseases.

Conclusions
Overall, the available evidence provides robust information about changes in lipidome and increased steady-state levels of protein damaged by lipoxidation reactions with PFC 27 aging. The set of damaged proteins compromise vital cell functions such as energy metabolism, cytoskeleton, neurotransmission, proteostasis, and oxygen/carbon dioxide/heme homeostasis. The affected biological processes may be considered putative collaborative factors contributing to the cognitive decline associated with PFC aging.
Different nutritional interventions can have an impact on these mechanisms, improving brain function and delaying changes associated with PFC aging. Since molecular damage is already identified in normal middle-aged individuals and increases physiologically in the elderly, it seems reasonable to act upon the identified free radical-producing targets and lipid metabolism in the appropriate middle-age window.

Acknowledgements
We thank T. Yohannan for editorial help.