Alzheimer’s disease (AD) is the most common cause of dementia and the sixth cause of death in the world, constituting a major health problem for aging societies. This disease is a neurodegenerative continuum with well-established pathology hallmarks, namely the deposition of amyloid-β (Aβ) peptides in extracellular plaques and intracellular hyperphosphorylated forms of the microtubule associated protein tau forming neurofibrillary tangles (NFTs), accompanied by neuronal and synaptic loss. Interestingly, patients who will eventually develop AD manifest brain pathology decades before clinical symptoms appear. Among all the proposed pathogenic mechanisms to understand the etiology of Alzheimer’s disease (AD), increased oxidative stress seems to be a robust and early disease feature where many of those hypotheses converge.
Alzheimer’s disease (AD) is the most common cause of dementia and the sixth cause of death in the world, constituting a major health problem for aging societies . This disease is a neurodegenerative continuum with well-established pathology hallmarks, namely the deposition of amyloid-β (Aβ) peptides in extracellular plaques and intracellular hyperphosphorylated forms of the microtubule associated protein tau forming neurofibrillary tangles (NFTs), accompanied by neuronal and synaptic loss . Interestingly, patients who will eventually develop AD manifest brain pathology decades before clinical symptoms appear . Nevertheless, AD is still frequently diagnosed when symptoms are highly disabling and yet there is no satisfactory treatment.
Although the manifestations of AD are preponderantly cerebral, cumulative evidence shows that AD is a systemic disorder . Accordingly, molecular changes associated with AD are not exclusively manifested in the brain but include cells from different parts of the body, ranging from the blood and skin to peripheral olfactory cells. More recently, neurons derived from induced pluripotent stem cells (iPSCs) from AD patients have contributed to glean a more realistic insight of brain pathogenic mechanisms . Alternatively, the culture of olfactory neuronal precursors (ONPs) has emerged as a relatively simpler tool to study different brain disorders, taking advantage of their neuronal lineage and their readily non-invasive isolation . For instance, patient-derived ONPs manifest abnormal amyloid components together with tau hyperphosphorylation, which have recently led to the proposal of these cells as a novel diagnostic tool for AD .
Different hypotheses have attempted to explain AD pathogenesis. Some of them include Aβ cascade, tau hyperphosphorylation, mitochondrial damage, endoplasmic reticulum (ER) stress, and oxidative stress. Interestingly, although it has been difficult to establish a prevailing causative mechanism, increased levels of oxidative stress seem to be a common feature for many of these models. Furthermore, oxidative stress due to increased levels of reactive oxygen species (ROS) has been broadly recognized as a very early signature during the course of AD . Interestingly, AD-related oxidative stress is by no means restricted to neuronal cells but is also related to astrocytes’ oxidative damage and antioxidant capacity . Indeed, since the acknowledgment of the tripartite synapse, it has become increasingly clear that different antioxidant mechanisms of astrocytes can be harnessed by synaptically active neurons and surrounding cells . In the tripartite synapse, the astrocyte’s endfeet are close to synapses and can be activated by the spillover of synaptic glutamate to provide a timely antioxidant response . Moreover, it is not entirely understood how other glial cells such as pericytes may contribute to the damage induced by AD-related oxidative stress. For instance, oxidative damage may compromise the integrity of pericytes, which in turn could alter the blood-brain barrier’s integrity, favoring the infiltration of cytotoxic cells and the emergence of brain edema . In coherence with a broader systemic manifestation of this disease, the peripheral olfactory system shows AD-associated oxidative stress, which has been measured both in the olfactory neuroepithelium and in cultured ONPs . However, while the intriguing relationship between oxidative stress and AD has been long known, their translational impact has remained limited.
The olfactory neuroepithelium is a key structure for odor sensing. It consists of a pseudostratified columnar epithelium located on the outer domain of the olfactory mucosa settled on the basement membrane (BM) and the lamina propria (LP) . The cellular composition of these layers has been widely documented based on morphological analysis and the use of characteristic markers for each cell type . Figure 1 schematizes the location, cellular components, and molecular markers of the human olfactory mucosa.
Figure 1. Cytoarchitecture and cellular components of the human olfactory mucosa. Lamina propria components. Olfactory Ensheathing Cells, Bowman’s gland and Olfactory Ectomesenchymal Stem Cells (OE-MSCs). The image indicates the OE-MSCs markers: CD29, CD90, CD44, Nestin, and Vimentin. Olfactory epithelium components. Basal Cells, Olfactory sensory neurons (OSNs) or Olfactory receptor neurons (ORNs), Sustentacular cells, and Microvillar cells. The figure shows basal cell markers: K5 (Keratin 5), K17 (Keratin 17), p63, Sox-2 (SRY-Box Transcription Factor 2), Nestin, BrdU (Bromodeoxyuridine), and Ki-67; ORNs markers: GAP-43 (Growth Associated Protein 43), β-tubulin, OMP (Olfactory Marker Protein), GNG8 (Guanine Nucleotide-binding protein subunit Gamma), and GNG13 (Guanine Nucleotide-binding protein G(I)/G(S)/G(O) subunit Gamma-13)); sustentacular cell markers (SUS-1, Cbr2 (Carbonyl Reductase 2) and Cyp2g1 (Cytochrome P450, family 2, subfamily G, polypeptide 1)) and, microvillar cell marker: (spot-35 proteins). Created with BioRender.com.
The olfactory neuroepithelium is also a source of stem cells, which are capable of self-renewal and can generate neuronal precursors throughout the entire human lifetime. These precursors include neural stem cells known as basal cells. As expected for neural stem cells, basal cells are multipotent and allow the continuous replacement of neuronal and non-neuronal cells such as olfactory receptor neurons (ORNs) and sustentacular cells (of astrocytic lineage), respectively . In addition, the LP contains another less accessible population of stem cells, whose features meet most of the minimum criteria of the mesenchymal and Tissue Stem Cell Committee of the International Society for Cellular Therapy . As such, they are named as olfactory ectomesenchymal stem cells (OE-MSCs) .
Isolation of cells of the olfactory neuroepithelium from patients provides a source of cultured neural stem cells, which has been used to model different brain disorders such as schizophrenia, Parkinson’s disease, autism, ataxia-telangiectasia, hereditary spastic paraplegia (HSP), and AD . These neural stem cells can be frozen and stored for subsequent use and tolerate several passages without significantly losing their main properties. Furthermore, purified cultures obtained by cloning selection through limiting dilution significantly increases cell viability at least until passage 60 . In this work, we will refer to neural stem cells isolated from the olfactory neuroepithelium as olfactory neuronal precursors (ONPs), similar to .
Different strategies have been used to isolate and culture patient-derived ONPs, ranging from biopsies to non-invasive exfoliation of the nasal turbinate. Human ONPs were first isolated by Wolozin et al. from the olfactory neuroepithelium of cadavers or from adult biopsied samples . Another similar isolation approach demonstrated that a significant subpopulation of these cells express markers of mature olfactory neurons such as OMP, Golf, NCAM, and NST and look small and bright to the microscope, in contrast to the remaining “dark phase” cells that do not express OMP, but glial markers . However, a systematic characterization of these cultures has shown that after a few days in vitro, both dark and bright phase cells show an intracellular calcium increase in response to odorants, highlighting the neuronal features of these cells . In addition, cells with features of ONPs have also been obtained from dissociated neurospheres, which have been denominated “olfactory neurosphere-derived” (ONS) cells . Alternatively, ONPs can be non-invasively isolated by an exfoliation of the nasal cavity . These exfoliated cells can be cultured in a modified media to propitiate neural lineage maintenance and proliferation. Notably, these neuronal precursors conserve their capability to differentiate into ORNs in the presence of dibutyryl adenosine 3’,5’-cyclic monophosphate (Db-cAMP) and, strikingly, maintain their electrical response to odorants . Thus, non-invasively isolated ONPs retain neuronal features similar to those obtained by biopsy. A simplified extraction protocol and the molecular characterization of non-invasively isolated ONPs is shown in Figure 2.
Figure 2. Non-invasive isolation of olfactory neuronal precursors (ONPs). (A) Schematic cartoon of the isolation protocol based on the extraction of nasal exfoliate with the subsequent adherent culture and enrichment of ONPs. (B) Left, the nasal exfoliate is directly seeded on adherent plates, showing a mixture of cell morphologies. Right, after 1–2 weeks ONPs dividing colonies are easily observed with their characteristic morphologies. (C) Upper panel, immunofluorescence of cultured ONPs, depicting the stem cell marker Nestin and Ki67 (yellow arrows) to show active cell proliferation. Lower panel, cultured ONPs express neuronal markers such as β3 tubulin. Cell nuclei are shown by DAPI staining. All scale bars = 100 μm. All images were generated in our lab. Created with BioRender.com.
Oxidative stress is the result of an imbalance between oxidant and antioxidant cellular pathways. One of the most studied oxidant compounds are ROS, which are highly reactive molecules, including peroxide (H2O2), superoxide anion radical (O2 • −), and hydroxyl radical (• OH), among others. These molecules may covalently interact with lipids, proteins, and carbohydrates, generating molecular adducts and cumulative damage that, when sensed by cells, may actively trigger different death programs .
It was well established almost three decades ago that oxidative stress damage is linked to AD . Furthermore, it has been proposed that oxidative stress at different brain neuronal and non-neuronal cells might be the earliest event of a pathogenic cascade . Whether oxidative stress is a causative agent or just a consequence in neurodegenerative disorders has been thoroughly debated for several years, but still remains an open question . The most parsimonious interpretation of this evidence is that oxidative stress as well as other potential AD causative agents (such as Aβ accumulation) are part of a highly interconnected vicious cycle rather than a linear chain of events with a unique origin. The molecular mechanisms and implications of oxidative stress on the nervous system and, potentially, during AD pathogenesis have been thoroughly reviewed elsewhere . Here, we focus on evidence showing AD-associated oxidative stress in the peripheral olfactory system rather than reviewing mechanistic explanations.
Oxidative stress associated with AD is manifested in the olfactory neuroepithelium. Accordingly, increased immunoreactivity of the antioxidant enzyme manganese and Copper-Zinc superoxide dismutases have been detected in ORNs and basal and sustentacular cells of the olfactory neuroepithelium of AD patients compared with age-matched controls . Analogously, AD patients harbor a higher immunoreactivity against the antioxidant protein Metallothionein both in the olfactory neuroepithelium and the Bowman’s Glands and the LP . Both results suggest that cells from olfactory neuroepithelium elicit an increased antioxidant defense, due to increased oxidative stress during AD. With respect to the direct measurement of oxidation products, post-mortem staining showed an increase in 3-nitrotyrosine (3-NT) in the brain and olfactory neuroepithelium of AD patients . Figure 3 schematizes the antioxidant response and oxidative damage reported in ONPs and OE from AD patients. It would be of interest to uncover whether some AD genetic factors such as the ApoE ε4 allele (ApoE4) (the single most important genetic risk factor for AD) also manifests oxidative stress signatures in the olfactory epithelium. It is plausible that this is the case because deficits in odor fluency, identification, recognition memory, and odor threshold sensitivity have been associated with the inheritance of the ApoE4 genotype in several studies . For a more thorough compiling of evidence showing AD-associated oxidative damage across other domains of the nervous system, readers may refer to the following excellent articles .
Figure 3. Oxidative stress associated with AD in the olfactory neuroepithelium. (a) ONPs and sustentacular cells in the olfactory epithelium (OE) show an increased antioxidant defense with elevated levels of manganese and copper-zinc superoxide dismutases as well as heme oxygenase-1 due to increased oxidative stress in AD patients compared with age-matched controls. Moreover, there is an increase in 3-nitrotyrosine (3-NT) and 4-hydroxynonenal (lipid peroxidation indicator) levels, suggesting AD-associated oxidative damage. (b) The increased generation of superoxide anion activates superoxide dismutases (SOD) as an antioxidant response. The generation of other reactive oxygen species (ROS), such as H2O2, induces the expression of other antioxidant enzymes (heme oxygenase-1). On the other hand, the accumulation of superoxide anion increases the levels of compounds such as 4-hydroxynonenal (4-HNE). Moreover, the increased levels of 3-NT are produced from the interaction of superoxide anion and nitric oxide (NO), whose probable source is located at activated macrophages in the OE of AD patients. Created with BioRender.com.
The relationship between oxidative stress and AD has been extensively studied mainly through cellular and animal models . However, these models may not fully capture key features of the disease. This limitation potentially leads to wrong conclusions about the pathogenic mechanisms and ultimately may dampen the development of effective therapies. Alternatively, patient-derived cells of neuronal lineage such as those from the olfactory epithelium may provide a convenient solution to this problem .
Interestingly, cultured patient-derived ONPs and other peripheral cells also manifest AD-associated oxidative stress. For example, an increase in the level of hydroxynonenal and Nɛ-(carboxymethyl)lysine) (indicating lipid peroxidation), as well as a higher content of heme oxygenase-1, has been found in ONPs isolated from AD patients compared with age-matched controls (Figure 3) . Furthermore, ONPs from AD patients are also more susceptible to oxidative stress-induced cell death . This is strikingly similar to what has been found by our group in blood-derived lymphocytes from AD patients . Indeed, manifestations of oxidative stress associated with AD have been reported in different patient-derived peripheral cells ranging from blood cells to fibroblasts and iPSCs-derived neurons. These changes may include compensatory antioxidant responses and a rise in the concentration of oxidation by-products, as well as increased susceptibility to ROS-induced cell death, which has been demonstrated in different cellular types from AD patients. Many of those findings are summarized in the Table 1. In addition, Table 1 also summarizes similar evidence of other relevant pathogenic mechanisms proposed for AD pathogenesis, including Amyloid/Tau, mitochondria, and ER-stress. Thus, different cells throughout the body show signs of different proposed AD pathogenic mechanisms, including oxidative stress at early stages of the disease continuum. The robustness of this tendency highlights the potential of patient-derived cells, and in particular ONPs, for monitoring oxidative stress associated with AD.
Table 1. Signatures of oxidative stress and other AD mechanistic hypotheses are manifested in patient-derived peripheral cells, iPSCs and ONPs.
|Pathogenic Mechanism||Main Finding||Cellular Type||Lineage||References|
|Amyloid/Tau||Platelets from AD patients reproduce the increased amyloidogenic processing of AβPP||Platelets||Non-neuronal|||
|Amyloid/Tau||AD platelets harbor increased levels of a higher molecular weight tau isoform||Platelets||Non-neuronal|||
|Amyloid/Tau||Alteration of AβPP, BACE, and ADAM 10 levels in early stages of the disease||Platelets||Non-neuronal|||
|Amyloid/Tau||It is suggested a decreased non-amyloidogenic processing of AβPP by a lack of nicastrin mRNA expression in samples obtained from AD patients||Lymphocytes||Non-neuronal|||
|Amyloid/Tau||Altered balance between Aβ-oligomers and PKCε levels in AD. Loss of PKCε-mediated inhibition
|Amyloid/Tau||Higher Aβ42/Aβ40 ratio compared to control cells||PSEN1 iPSC-derived
|Amyloid/Tau||Mutation alters the initial cleavage site of γ-secretase, resulting in an increased generation of Aβ42, in addition to an increase in the levels of total and phosphorylated tau||Neuron-derived iPSCs from patients harboring the London FAD AβPP
|Amyloid/Tau||Oligomeric forms of canonical Aβ impairs
|Cortical neurons from three genetic forms of AD —PSEN1 L113_I114insT, AβPP duplication (AβPPDp), and Ts21— generated from iPSCs||Neuronal|||
|Amyloid/Tau||Increase in the content and changes in the subcellular distribution of t-tau and p-tau in cells from AD patients compared to controls||Non-invasively isolated ONPs||Neuronal|||
|Mitochondria||Compromise of mitochondrial COX from
|Mitochondria||Platelets isolated from AD patients show decreased ATP levels||Platelets||Non-neuronal|||
|Mitochondria||AD lymphocytes exhibit impairment of total OXPHOS capacity||Lymphocytes||Non-neuronal|||
|Mitochondria||AD skin fibroblasts show increased production of CO2 and reduced oxygen uptake suggesting that mitochondrial electron transport chain might
|Mitochondria||AD fibroblasts present reduction in mitochondrial length and a dysfunctional mitochondrial bioenergetics profile||Fibroblasts||Non-neuronal|||
|Mitochondria||SAD fibroblasts exhibit aged mitochondria, and their recycling process is impaired||Fibroblasts||Non-neuronal|||
|Mitochondria||Patient-derived cells show increased levels of oxidative phosphorylation chain complexes||Human induced pluripotent stem cell-derived neuronal cells (iN cells) from
|Mitochondria||Mitophagy failure as a consequence of
|iPSC-derived neurons from FAD1 patients harboring PSEN1 A246E mutation||Neuronal|||
|Mitochondria||Neurons exhibit defective mitochondrial
|iPSC-derived neurons from an AD patient carrying AβPP -V715M mutation||Neuronal|||
|Oxidative Stress||Increased activity of the antioxidant enzyme catalase in probable AD patients||Erythrocytes||Non-neuronal|||
|Oxidative Stress||Increased production and content of thiobarbituric acid-reactive substances (TBARS), superoxide dismutase (SOD), and nitric oxide
|Erythrocytes and Platelets||Non-neuronal|||
|Oxidative Stress||Increase in the content of the unfolded version of p53 as well as reduced SOD activity||Peripheral blood mononuclear cells (PBMCs)||Non-neuronal|||
|Oxidative Stress||Exacerbated response to NFKB pathway||PBMCs||Non-neuronal|||
|Oxidative Stress||Increased ROS production in response to H2O2||PBMCs||Non-neuronal|||
|Oxidative Stress||AD lymphocytes were more prone to cell death after a H2O2 challenge||Lymphocytes||Non-neuronal|||
|Oxidative Stress||Reduced antioxidant capacity of FAD lymphocytes and fibroblasts together with increased lipid peroxidation on their plasma membrane||Lymphocytes and Fibroblasts||Non-neuronal|||
|Oxidative Stress||Aβ peptides were better internalized and generated greater oxidative damage in FAD fibroblasts||Fibroblasts||Non-neuronal|||
|Oxidative Stress||Aβ peptide caused a higher increase in the oxidation of HSP60||Fibroblasts||Non-neuronal|||
|Oxidative Stress||Reduction in the levels of Vimentin in samples from AD patients||iPSCs-derived neurons from AD patient||Neuronal|||
|Oxidative Stress||Increased levels of hydroxynonenal, Nɛ-(carboxymethyl)lysine), and heme oxygenase-1 in samples from AD patients||Biopsy-derived ONPs||Neuronal|||
|Oxidative Stress||Increased susceptibility to oxidative-stress-induced cell death||Biopsy-derived ONPs||Neuronal|||
|ER-Stress||Impaired ER Ca2+ and ER stress in PBMCs from MCIs and mild AD patients||PBMCs||Non-neuronal|||
|ER-Stress||Accumulation of Aβ oligomers induced ER and oxidative stress||iPSC-derived neural cells from a patient carrying APP-E693Δ mutation and a sporadic AD patient||Neuronal|||
|ER-Stress||Aβ-S8C dimer triggers an ER stress response more prominent in AD neuronal cultures where several genes from the UPR were upregulated||iPSC-derived neuronal cultures carrying the AD-associated TREM2 R47H variant||Neuronal|||
|ER-Stress||Accumulation of Aβ oligomers in iPSC-derived neurons from AD patients leads to increased
|iPSC-derived neurons from patients with an AβPP-E693Δ mutation||Neuronal|||