**3. Mitochondrial dysfunction can help us to predict AD?**

activity is increased [113], suggesting an important role of the mPTP in mitochondrial injury

**Figure 2. Improving mitochondrial health as a valid therapy for AD**. (**A**) **Mitochondrial morphology therapy**. Ma‐ nipulating the processes of mitochondrial dynamics has a considerable potential for treating neurodegenerative diseas‐ es. Therapies that increase mitochondrial biogenesis and fission/fusion cycle may improve mitochondrial function and decreased oxidative stress. Mitophagy is a selective autophagy process that removes dysfunctional mitochondria and maintains adequate mitochondria quality control. Increasing PINK1‐mediated mitophagy improves mitochondrial in‐ tegrity and function. (**B**) **Mitochondrial bioenergetics therapy**. Several agents that boost bioenergetics could have effi‐ cacy in improving mitochondrial function. These compounds show neuroprotective effects, which may be a useful target for treating neurodegenerative diseases. Treatment with CoQ10, MitoQ and MitoE prevented oxidative stress; cyclosporine A, a substance that blocks the opening of mPTP, prevented mitochondrial depolarization, blocks cyto‐ chrome c release and increased superoxide dismutase activity. Drugs that mediate the activation of Nrf2 induce the

In that context, several groups have found that some compounds not only improve one aspect of mitochondrial damage but also improve several alterations at once by the activation of several pathways like nuclear factor E2‐related factor 2 (Nrf2) [10, 101, 102]. The Nfr2 and the Nrf2‐Are pathways have been studied in mitochondrial dysfunction and neurodegeneration [10, 115]. In response to oxidative stress, the Nrf2 translocate from the cytoplasm into the nucleus and activates the expression of several antioxidant genes [116]. Nrf2 is the principal regulator of the antioxidant cellular response and seems to be a promising target in the treatment of age‐related neurodegenerative diseases [10, 101, 102, 117]. Nrf2 activation induces changes in mitochondrial structure and function, which is of particular importance under conditions of oxidative stress [10, 102, 118]. In primary murine cortical cultures, neurons lacking Nrf2 are more susceptible to oxidative stress induced by H2O2 and glutamate [116, 119] and overexpression of Nrf2, totally prevented these changes [116]. Furthermore, overex‐ pression of Nrf2 can rescue neurons from mitochondrial complex II inhibition and ischemic

expression of antioxidant enzymes and improve mitochondrial function and biogenesis.

insult in animal models of Huntington disease and stroke [116, 120, 121].

in AD [113, 114].

358 Update on Dementia

In 2011, the National Institute of Aging (NIA) and the Alzheimer's Association proposed a revised criteria and new guidelines for diagnosing Alzheimer's disease [123]. They proposed three stages of progression of AD, preclinical AD, mild cognitive impairment (MCI) due to AD and dementia due to AD. Also, they incorporated the use of biomarker tests to corroborate the presence or absence of AD or the risk to develop it [124]. Biomarker tests will be essential to identify which individuals are in the early stages of the disease and if they should receive some disease‐modifying treatment. They are also critical for monitoring the effects of treatment against AD [123, 124].

AD mainly affects memory and cognitive functions and to this date, there is no early biomarker that shows the reliability and accuracy needed to diagnose the disease [125]. Currently, AD can be diagnosed with over 90% of confidence but with invasive and expensive tools based on cerebrospinal fluid (CSF) analysis and neuroimaging with positron emission tomography, with Pittsburgh compound‐B radiotracer (PET/PiB) [126]. For this reason, the diagnosis is based on neuropsychological surveys and in the exclusion of other age‐related dementias only when there is an advanced cognitive impairment [127]. The conclusive diagnosis of AD is only possible in autopsy with the presence of characteristic pathological brain lesions [125, 127].

Despite that AD early treatment can slow down the progression of the disease, the ability to diagnose AD at early stages is currently limited. In the search for potential biomarkers for early diagnosis of AD, several studies have shown that a significant number of peripheral tissues, both in animal models and patients, showed from early stages of the disease an abnormal presence of markers normally associated with nerve tissue [128].

For example, deposits of Aβ have been reported in skin, blood vessels, glandular structures and fibroblasts in human tissue [129–131], and the presence of total and phosphorylated tau protein were detected in plasma of AD and healthy patients [132, 133]. These facts suggest that the use of peripheral tissues as a source of inexpensive and minimally invasive samples is taking force in the diagnosis of AD. Interestingly, several studies have shown that there is an important relationship between the peripheral tissue in patients and animal models that develop AD and mitochondrial damage. Here, we show that AD peripheral tissues present different mitochondrial alterations that include mitochondrial defects in morphology, dynamics and bioenergetics.

#### **3.1. Evidence for mitochondrial dynamics defects in AD peripheral tissues.**

Mitochondrial dynamics is a complex cellular process that controls the shape, localization, turnover and function of mitochondria. As we previously discussed, several findings in patients and animal models of AD suggest that the deregulation of mitochondrial dynamics is a common feature in the disease, but may vary from case to case [134]. In the case of peripheral tissue of patients with AD, different studies indicated an altered mitochondrial morphology that could be related with changes in mitochondrial dynamics [21, 140, 141].

Several studies had proposed that the platelets could be a promising peripheral surrogate to detect AD [135], which is because these cell fragments express high levels of APP [136], tau protein [137, 138] and they have an increased GSK3β activity, a kinase responsible for tau hyperphosphorylation [139]. More important, in studies with cytoplasmic hybrid (cybrid) cells created from human neuroblastoma cells repopulated with mitochondria from platelets obtained from sporadic AD and control donors, it was shown that cybrid cells from AD patients contained a significantly increased percentage of enlarged or swollen mitochondria, and they also present a reduced mitochondrial membrane potential [140].

Using another blood cell component, the analysis of peripheral blood lymphocytes from AD patients showed an increase in SNO‐Drp1 and Fis1 and reduced Drp1 levels compared with healthy controls, PD patients and vascular dementia patients [141]. The protein expression pattern observed here suggests the presence of morphological alterations of mitochondria [141].

On the other hand, in a study with fibroblasts of sporadic AD patients, an abnormal mito‐ chondrial distribution characterized by elongated mitochondria that are accumulated in perinuclear areas with a significant decreased in Drp1 levels was found [21]. These findings are very relevant because several publications suggest that the basic pathogenic mechanism of amyloidogenesis is similar in brain and skin fibroblasts, with an increase in the production and depositions of Aβ [128, 142]. Therefore, a mitochondrial deregulation in the fibroblasts of AD patients could be indicative of the neurological progression of the disease [143, 144].

#### **3.2. Mitochondrial bioenergetics is altered in AD peripheral tissues.**

Evidence of a primary role for mitochondrial damage in AD development has also been provided through post‐mortem examination of AD brains, revealing oxidative stress, mito‐ chondrial DNA damage and bioenergetic deficiencies in MCI and AD patients [145–147]. In contrast, studies on peripheral tissues of AD patients have generated inconsistent findings [135, 140, 148–174].

Different studies reviewed by Cervellati's group have reported changes in the hydroperoxide levels, a biomarker of oxidative stress, in plasma and serum of AD patients [148]. In addition, these studies revealed that the levels of the oxidant damage markers, MDA and 4‐HNE, were increased in plasma and serum of AD and MCI patients compared to controls [148]. Comple‐ mentary, in blood samples of individuals with mild cognitive impairment and AD, there are evidence of mitochondrial dysfunction with decreased expression of respiratory complex genes, TOMM40, and subunits of the core mitochondrial ribosome complex [149, 150]. In addition, in human peripheral blood mononuclear cells was found an increase in oxidative stress and phosphorylated levels of Nrf2 [151].

important relationship between the peripheral tissue in patients and animal models that develop AD and mitochondrial damage. Here, we show that AD peripheral tissues present different mitochondrial alterations that include mitochondrial defects in morphology,

Mitochondrial dynamics is a complex cellular process that controls the shape, localization, turnover and function of mitochondria. As we previously discussed, several findings in patients and animal models of AD suggest that the deregulation of mitochondrial dynamics is a common feature in the disease, but may vary from case to case [134]. In the case of peripheral tissue of patients with AD, different studies indicated an altered mitochondrial morphology that could be related with changes in mitochondrial dynamics [21, 140, 141]. Several studies had proposed that the platelets could be a promising peripheral surrogate to detect AD [135], which is because these cell fragments express high levels of APP [136], tau protein [137, 138] and they have an increased GSK3β activity, a kinase responsible for tau hyperphosphorylation [139]. More important, in studies with cytoplasmic hybrid (cybrid) cells created from human neuroblastoma cells repopulated with mitochondria from platelets obtained from sporadic AD and control donors, it was shown that cybrid cells from AD patients contained a significantly increased percentage of enlarged or swollen mitochondria, and they

Using another blood cell component, the analysis of peripheral blood lymphocytes from AD patients showed an increase in SNO‐Drp1 and Fis1 and reduced Drp1 levels compared with healthy controls, PD patients and vascular dementia patients [141]. The protein expression pattern observed here suggests the presence of morphological alterations of mitochondria

On the other hand, in a study with fibroblasts of sporadic AD patients, an abnormal mito‐ chondrial distribution characterized by elongated mitochondria that are accumulated in perinuclear areas with a significant decreased in Drp1 levels was found [21]. These findings are very relevant because several publications suggest that the basic pathogenic mechanism of amyloidogenesis is similar in brain and skin fibroblasts, with an increase in the production and depositions of Aβ [128, 142]. Therefore, a mitochondrial deregulation in the fibroblasts of AD patients could be indicative of the neurological progression of the disease [143, 144].

Evidence of a primary role for mitochondrial damage in AD development has also been provided through post‐mortem examination of AD brains, revealing oxidative stress, mito‐ chondrial DNA damage and bioenergetic deficiencies in MCI and AD patients [145–147]. In contrast, studies on peripheral tissues of AD patients have generated inconsistent findings

Different studies reviewed by Cervellati's group have reported changes in the hydroperoxide levels, a biomarker of oxidative stress, in plasma and serum of AD patients [148]. In addition,

**3.1. Evidence for mitochondrial dynamics defects in AD peripheral tissues.**

also present a reduced mitochondrial membrane potential [140].

**3.2. Mitochondrial bioenergetics is altered in AD peripheral tissues.**

dynamics and bioenergetics.

360 Update on Dementia

[141].

[135, 140, 148–174].

On the other hand, several studies had shown that blood platelets from AD patients also present an increase in markers related to mitochondrial bioenergetics damage [135]. Platelets presented intracellular calcium deregulation [152, 153], an increase in oxidative damage [152, 153], a decrease in CoxIV and ATP synthase activities [154–158], and as we previously mentioned, a reduced mitochondrial potential in the cybrid condition [140]. Interestingly, in a study with cognitively normal individuals with maternal history of late onset of AD was found a reduced activity of platelet CoxIV compared to those with paternal or negative family history [159]. These findings suggest not only a possible mitochondrial peripheral biomarker but also an exclusively maternally inherited marker in humans [159].

Mitochondria isolated from AD lymphocytes showed an increase in several markers of oxidative stress [160, 161], increased susceptibility to oxidative death [162, 163], and the extent of this oxidative damage inversely correlated with dementia severity [161, 162, 164]. Also, this cell type presented alterations in proteins levels of mitochondrial‐related factors categorized as energetic, structural and antioxidants such as glyceraldehyde 3‐phosphate dehydrogenase (GAPDH), lactate dehydrogenase B chain and ATP synthase [164]. Furthermore, analysis of mitochondrial function in lymphocytes of AD patients showed a reduction in basal respiration and a lower ATP turnover that could finally lead to accumulate mutations in mitochondrial DNA [161].

Furthermore, a recent study determined that mitochondrial population, ATP production and respiratory function are altered in fibroblasts of patients with genetic type of AD [165]. While genetic forms of the disease do not account for the majority of cases, these observations marks an important precedent that directly links mitochondrial dysfunction in peripheral tissue of AD patients [165]. Also, in this cell type, mitochondrial dysfunction is associated with high levels of ROS and oxidative damage [166–168]. This alterations could be explained because of the lower levels of antioxidant defences observed in AD patients [169], and more interesting is the fact that these fibroblasts exhibit an alteration of the calcium buffering capacity compared to control cells [170, 171].

Based on that, recent studies have shown that fibroblasts of sporadic and familial AD present an enhanced link between the endoplasmic reticulum (ER) and mitochondria, through the mitochondria‐associated ER membranes structures (MAMs) [172]. This alteration in the communication between these organelles could affect the mitochondrial dynamics and function, calcium homoeostasis and production of ROS [172]. This is an interesting observa‐ tion, since a recent study showed that nanomolar concentrations of oligomeric Aβ regulated MAM and mitochondrial calcium in neuronal cells of human AD cortical tissue, as well as in AD mouse models [173]. These findings suggest that these subcellular structures are affected in AD and this would not be considered an isolated effect of fibroblasts culture.

**Figure 3. Mitochondrial impairment as a potential biomarker for early diagnosis of AD**. Diagram shows that possi‐ ble markers of mitochondrial damage could be present in blood plasma, blood cells and skin fibroblasts from AD pa‐ tients. The compressive evaluation of mitochondrial health in these tissues could early detect neurodegenerative changes reported in AD.

Interestingly, a recent study with fibroblasts from an AD patient demonstrated that is possible to induce the differentiation of dermal fibroblasts into neuronal cells [174]. This study demonstrated that those neurons derived from fibroblasts expressed significant levels of phosphorylated tau and presented significant changes in the expression of genes associated with AD [174]. These studies indicate that the fibroblasts of patients could be a reliable tool for obtaining physiological information that reflects the neurological state of the patients.

Peripheral biomarkers with effective action in the early detection of Alzheimer's pathology are currently unknown, but the evidence of possible markers of mitochondrial damage in blood plasma, blood cells and skin fibroblasts represents an important step in the search for an AD biomarker (**Figure 3**). Although the fact that these tissues may provide less invasive and inexpensive sources to investigate AD progression, the finding of a new biomarker would not only be important for early diagnosis but also be an opportunity to prove direct and person‐ alized therapies in patients with AD. Future research should focus not only in search for therapies of the disease but also in the search for a good and safe model to test the effectiveness of these pathways proposed.
