**7. Perspectives and Future Directions**

Label-free monitoring of oxidative stress in patient-derived ONPs may accelerate the discovery of molecules for effectively targeting AD. In this sense, imaging the dynamics of NAD(P)H intrinsic fluorescence (e.g., by FLIM) may offer a readily available, less toxic, and comparatively richer lecture of drug effects compared to classic proteomic and cell-fixation methods. Interestingly, patient-derived ONPs have already been used for drug screening. In particular, these cells were used to test drugs that restored acetylated tubulin patientderived stem cells with a variety of *SPAST* mutations in Hereditary Spastic Paraplegia (HSP) [48] and to perform a multidimensional phenotypic screening with different natural products in Parkinson's disease [148].

Different cellular AD models have been used for high-throughput screening (HTS) of therapeutic molecules [149–152]. For example, a search for inhibitors of calpain activity (to prevent Aβ-induced neurotoxicity) was performed on a library of approximately 120,000 compounds and tested on differentiated SH-SY5Y cells [153]. In another approach, the motility and proliferation of PC12 cells was assessed to test multiple drugs based on Chinese herbal compounds targeting Aβ42-induced apoptosis [154]. Similarly, it has been found that a combination of bromocriptine, cromolyn, and topiramate has a potent anti-Aβ effect on patient-derived iPSCs neurons [155]. Thus, it seems plausible to perform HTS of therapeutic molecules in patient-derived ONPs coupled to label free microscopy.

A potential therapeutic strategy, which could be monitored in patient-derived ONPs, is to delay the AD-associated depletion of free NADH. This is supported by the recent observation that imposed manipulation of cysteine/cystine (Cys/CySS) redox state was able to restore mitochondrial levels of free NADH to normal ranges in neurons from triple transgenic AD-like mice [28]. Given the relevance of free NADH inside cells—not only for redox management but also for metabolic supply, to sustain ATP levels—testing for antioxidant compounds capable of modulating free NADH deserves to be further studied. It is surprising to realize that the use of patient-derived ONPs to study the role of oxidative stress during AD has been, to some extent, neglected during the past decade. At least two reasons may have contributed to this delay; the first is that culturing patient-derived ONPs from biopsies is relatively more difficult and the second is the potential lack of technologies efficient enough to detect subtle changes of oxidative stress. Nevertheless, as highlighted in this article, both reasons can no longer be sustained.

Antioxidant therapies directed against AD have shown limited success; however, they still hold great promise and room for improvement. Some clinical trials in which AD patients were supplemented with antioxidants such as vitamins C and E, either alone or in combination with cholinesterase inhibitors, have failed to improve cognitive function [156,157]. However, other attempts have shown to be moderately effective; as is the case for polyphenols, a group of phytochemicals that showed a great antioxidant and antiinflammatory potential together with neuroprotective properties [158,159]. As such, clinical trials have suggested that polyphenolic compounds such as curcumin, resveratrol, and

green tea catechins may prevent and treat some forms of dementia [160–163]. Nevertheless, other reports show poor effects of antioxidants on cognitive function, which could be related to their low bioavailability [164–166]. Emerging evidence suggests that the combined intervention of different antioxidants may improve therapeutic efficacy. For example, some clinical trials have reported cognitive improvements in AD patients treated with a mix of antioxidant compounds harboring α-tocopherol, NAC, folate, acetyl-L-carnitine, vitamin B12, and S-adenosyl methionine [167]. In line with these findings, fibroblasts derived from AD patients have shown decreased mitochondrial oxidative stress after treatment with lipoic acid and N-acetyl-cysteine (NAC) [168]. It would be extremely interesting to monitor the intrinsic fluorescence of NADPH (reflecting the antioxidant capacity of the cell) in patient-derived ONPs in response to different mixes, proportions, and doses of these antioxidant compounds. Table 3 resumes some candidate natural and chemical compounds that could be successful in clinical trials evaluating them with AD-derived ONPs.

**Table 3.** Natural and chemical compounds that may target ONPs.


Human embryonic stem cells (ESCs) and subsequently human induced pluripotent stem cells (iPSCs) have emerged as powerful tools due to their ability for modeling neurodegenerative diseases [179]. For instance, three-dimensional (3D) organoids using patientderived induced pluripotent stem (iPS) cells can recapitulate microcephaly that has been difficult to model in mice [180]. On the other hand, 3D advanced culture models of the brain including blood–brain barrier (BBB) allow a precise study of candidate drugs by recapitulating the brain environment [181]. In this sense, the implementation of a human brain microvessel-on-a-chip that is amenable for quantitative live 3D fluorescence

analysis with high-resolution will facilitate the monitoring of NADPH movement and permeability during oxidative stress [182]. Moreover, 3D models can be harnessed to perform cutting-edge super-resolution microscopy, including high resolution volumetric imaging using Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) and also with the novel modality of expansion microscopy, which integrates with lattice light-sheet microscopy (Ex-LLSM) [183]. Thus, these emerging systems to model BBB will significantly improve drug discovery.

The olfactory system, the olfactory ensheathing cells (OECs), has been cultured in three dimensions to understand their behavior in a hampered environment, such as a spinal cord injury [184]. Although there is no evidence of ONPs cultured in 3D, the findings in other cells of the olfactory system suggest ONPs will have the same outcome. Thus, it would be extremely interesting to generate a 3D model for AD with ONPs from patients, incorporating a BBB microfluidic platform and analyzing cell metabolism by label-free microscopy in response to drug treatment like that which was reported in the organotypic microfluidic breast cancer model [185]. This approach will enable us to evaluate both the effect and the efficiency to traverse BBB of the drug candidate in an AD model.

In all, non-invasively isolated ONPs from AD patients coupled to real-time monitoring of relevant metabolic intermediaries may provide an unprecedent platform for early diagnosis and drug discovery. Furthermore, cellular models derived from patients might be sensitive enough to even develop personalized therapies, as has been suggested [186]. Proposed innovations are schematized in Figure 5. We envision that these strategies may generate large improvements required for the timely diagnosis and treatment of this devastating disease.

**Figure 5.** Isolation of patient-derived ONPs coupled to label-free microscopy offers relevant translational outcomes. (**A**) Schematic drawing of the high-throughput screening platform to study different

antioxidant molecules. (**B**) Cultured ONPs isolated from control patient-derived neurospheres were analyzed by label-free microscopy, using fluorescence hyperspectral analysis and the intensity of different intrinsic fluorophores was determined [187]. These fluorophores included: bound flavins, fluorescent retinoid derivative bis-retinoid N-retinylidene-N-retinyl ethanolamine (A2E), proteinbound NADH (bound NADH), free NADH, and lipofuscin. The original images from the publication of Gosnell et al. [187] (Figure 2) were adapted (cropped) with permission, following the guidelines of the creative commons license (CC BY 4.0, https://creativecommons.org/licenses/by/4.0/ ). (**C**) Analysis of intrinsic fluorophores such as protein-bound NADH or free NADH could provide relevant translational outcomes such detecting oxidative/metabolic signatures for early AD diagnosis and monitoring. In addition, those the subtle molecular profiling could settle the base for development of personalized therapies to treat AD. Created with BioRender.com.

**Author Contributions:** Writing—original draft preparation: L.G.-V., A.L.; writing—review and editing: L.G.-V., A.L., D.P.P., B.A.B., M.I.B.; supervision: M.I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the grant: "Fondo Nacional de Desarrollo Científico y Tecnológico—Fondecyt" Grant Number: 1190958 (M.I.B.).

**Acknowledgments:** L.G.-V. was supported by Secretaría de Educación, Ciencia, Tecnología e Innovación de la Ciudad de México (SECTEI).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

