*4.3. Advanced Effects in the Early Stage Resulting from Mitochondrial Dysfunction and Oxidative-Derived Damage*

The selective oxidative damage of ATP synthase leads to a progressive mitochondrial bioenergetic failure with the cell's deleterious effects postulated as the earliest molecular

event from which AD pathology will progress in EC and spread to the rest of the brain. These deleterious effects, representing the first steps of disease progression, may be grouped as follows: (a) expansion of the modified proteins and functional consequences at the mitochondrial and cellular levels, and (b) mitochondrial functional alterations, cellular dysfunction, and neuronal death.

Thus, once the oxidative lesion of ATP synthase and its functional defect are established as an early marker, and given the persistence of free radical production and lipid peroxidation, there will be an increase in the diversity of modified proteins, particularly at the mitochondrial level, amplifying the functional defects associated with them. Thus, there is a significant number of selectively modified proteins described in different cortical regions of the brain at different stages of AD [18,141,144,146,149,152]. Interestingly, this apparent diversity of modified proteins can be grouped into highly restricted functional categories such as bioenergetics, proteostasis, neurotransmission, antioxidant, and ion channel [18,144,149]. Bioenergetics is the most affected functional category, deepening the bioenergetic worsening already initiated by the ATP synthase injury. Table 1 shows a list of modified proteins at the mitochondrial level by lipid peroxidation-derived compounds (lipoxidation reactions) identified with redox proteomics in different regions of the human cerebral cortex during different stages of AD. In this context, it may be hypothesized that this mitochondrial bioenergetic defect will end up affecting an additional number of mitochondrial functions that, in turn, will affect cellular functions beyond the mitochondria.

Thus, protein damage can be extended beyond mitochondria, increasing the pool of modified proteins. Effectively, several observations confirm a greater number of modified proteins, and these can again be restricted to very specific functional categories such as neurotransmission, cytoskeleton, and oxygen metabolism [144]. Importantly, synaptic proteins constitute another group of major deregulated AD targets [144,152–155]. Although mostly described in other brain regions such as hippocampus and frontal cortex, these targets probably are also modified in EC. However, this point needs to be confirmed.


**Table 1.** Mitochondrial lipoxidized proteins identified with redox proteomics in different cerebral cortex regions affected by aging and AD pathology in different stages.

Brain regions from which modified proteins have been identified are entorhinal cortex, cingulate gyrus, hippocampus, parietal cortex, temporal cortex, and frontal cortex, see also: [18,141,144,146,149,152].

> As a consequence of the bioenergetic defects and loss-of-function of many mitochondrial proteins, it is feasible to postulate that additional mitochondrial activities, e.g., the machinery for the import of proteins/subunits of nuclear origin, and the communication between mitochondrion and nucleus, among others, can be affected, inducing a dysfunc-

tion in the gene expression of mitochondrial structural and regulatory components. In light of this bioenergetics compromise, we might also explain early EC changes in AD as the altered expression of several subunits of mitochondrial complexes and enzymes involved in energy metabolism [138,161–165], altered mitochondrial DNA methylation pattern [161,166], and alteration of the phosphorylation state of the mitochondrial channel VDAC (Voltage-dependent anion channel) [167]. Remarkably, bioinformatics processing has identified a large cluster of altered protein expression in the EC at relatively early stages of the disease, as shown in Figure 5 [144]. A dominant cluster is composed of mitochondrial proteins. This cumulative evidence points to an aggravation of mitochondrial dysfunction.

Other changes at the neuronal level are also described in the EC of AD-related pathology at initial stages includes deregulation of purine metabolism [168], alteration of pro-NGF [169], minor changes in microRNA expression [170], abnormal expression and distribution of metalloproteinase MMP2 [171], and alterations in the phosphorylation of the translation initiation factor 2 alpha (TIF2) [172].

An additional point that needs to be explored is related to lipid metabolism. The accrual of lipid granules was noted in the central studies of Alois Alzheimer. Recent studies have confirmed and provided further details of alteration in brain lipid metabolism in AD in general [18], and in the EC in particular [173–175]. These alterations include changes in lipidomic profiles, which must be added to the role as a prime target of oxidative damage for unsaturated acyl chains in the context of lipid peroxidation and lipoxidation-derived molecular damage previously treated in this review. This increased damage to PUFAs, along with potential alterations in biosynthesis pathways (currently unknown), could explain the reduced content of these fatty acids described in the EC and in the lipid rafts from EC at early stages I-II of AD [173,174]. This change in fatty acid profile is relevant for neuronal membrane properties (fluidity, thickness, curvature, packing, and activities of membrane-bound proteins) because the biophysical traits of polyunsaturated phospholipids do not favor the formation of highly ordered lamellar microdomains, whereas a relative increase in phospholipids containing short-chain saturated or monounsaturated fatty acids interacts favorably with cholesterol and sphingolipids in lipid rafts [176]. Thus, these features point to an increased propensity of neuronal membranes of EC in the earliest stages of AD (AD I/II) to form lipid rafts [173,174].

Furthermore, these lipid rafts show alterations in their profiles of lipid classes (increased content of phosphatidylcholine, sphingomyelins, and gangliosides) [173,174] which determine an increased membrane order and viscosity in these microdomains [174]. The physiopathological consequences of these changes in lipid profile in the onset and progression of AD in the EC are given credence by the specific accumulation of beta-secretase within AD subjects' lipid rafts even at the earliest stages. So, these findings provide a mechanistic connection between lipid alterations in these microdomains and amyloidogenic processing of amyloid precursor protein (APP) and subsequent cytotoxic effects [174]. Consequently, these changes in lipid metabolism seem to precede and play a causal role in forming SPs and probably NFTs, and thus represent an early event in the onset of AD at the EC. Similarly, the relevant role of amyloid fragments as cholesterol-binding proteins in cellular homeostasis of this vital lipid [177] and the upregulation of its subcellular transport towards mitochondrially associated membranes of endoplasmic reticulum in neurons [178] reinforces the relevance of lipid changes in the pathophysiology of AD.

We hypothesize the existence of a detrimental self-sustained loop between mitochondrial oxidative stress, lipid peroxidation-alteration of lipid metabolism, and bioenergetic defects in AD. We propose that this loop could be the basis of additional functional alteration at mitochondrial levels described in EC (and other brain regions) during AD. These secondary alterations would include altered mitochondrial genomic homeostasis, dysfunctional mitochondrial fusion and fission, deficits in mitochondrial axonal trafficking and distribution, impaired mitochondrial biogenesis, abnormal endoplasmic reticulummitochondrial interaction, and impaired mitophagy [140,179,180]. Overall, all these phe-

nomena would lead to cell failure and eventual neuronal death. Table 2 shows a summary of alterations described in human EC during aging and AD.

– **Figure 5.** Interactome map of deregulated proteins in the EC of AD at III–IV stage. Edge colors: (i) protein level correlations of proteomic data obtained by Orbitrap Velos, and (ii) interactions retrieved from the public databases BIND, CCSB, DIP, GRID, PubMed, Reactome, KEGG, HPRD, IntAct, MDC, and MINT. Node colors indicate cellular components provided by GO. A large cluster in the center corresponds to deregulated mitochondria-related proteins. Reproduced and modified with permission from [144].


**Table 2.** Summary of changes reported in human entorhinal cortex during aging and AD.

#### **5. Conclusions**

The origin of AD pathology seems to be associated with alterations located explicitly in the EC. Particular neurons from this brain region possess an inherent selective vulnerability prone to oxidative damage with early involvement of energy metabolism. This bioenergetic alteration is the onset for subsequent changes in a multitude of cell mechanisms leading to neuronal dysfunction and, eventually, cell death. Changes induced during physiological aging are the substrate for the emergence of dysfunctional mechanisms which will evolve toward neurodegeneration and, consequently, the development of the AD pathology. Current observations allow us to propose the existence of an altered allostatic mechanism in the EC whose central nucleus is made up of increased mitochondrial oxidative stress, lipid oxidation, and bioenergetic failure, and which in detrimental self-sustained feedback evolves to neurodegeneration, laying the basis for the onset and progression of the AD pathology. Since these alterations are already identified in middle-aged individuals, it seems reasonable to act upon the appropriate free radical-producing targets and lipid metabolism at the appropriate middle-age window. The present observations form a framework for further experiments and provide potential new targets for neuroprotective therapeutic interventions in AD pathology at the earliest stages. Curiously, the evolutionary traits that support human longevity are based on our vulnerability to age-related degenerative processes. Two key traits that define long-lived animal species (humans included) are the presence of cellular components resistant to oxidative stress and a low generation rate of molecular damage. These two characteristics are expressed at the biological level in the lipid profile and the mitochondrial production of free radicals. Curiously, both mechanisms, when altered, are the basis for a pathological condition, in this case AD, which limits human longevity.

**Author Contributions:** The manuscript was written by M.J., N.M.-M., P.T., V.A., M.P.-O., I.F. and R.P. and edited by R.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** Research by the authors was supported by the Institute of Health Carlos III (FIS grants PI14/00757, PI14/00328, PI20/0155), the Spanish Ministry of Science, Innovation, and Universities (Ministerio de Ciencia, Innovación y Universidades, grant RTI2018-099200-B-I00), and the Generalitat of Catalonia: Agency for Management of University and Research Grants (2017SGR696) to M.P-O., I.F., and R.P. This study was co-financed by FEDER funds from the European Union ('A way to build Europe').

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of IDIBELL and IRBLleida.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Acknowledgments:** M.J. is a 'Serra-Hunter' Fellow. N.M.-M. and P.T. received predoctoral fellowships from the Generalitat of Catalonia (AGAUR, ref 2018FI\_B2\_00104) and the Spanish Ministry of Science, Innovation, and Universities (Ministerio de Ciencia, Innovación y Universidades, ref. FPU16/01446), respectively. We thank T. Yohannan for editorial help.

**Conflicts of Interest:** The authors declare no competing financial interest.
