Oxidative stress in neurodegeneration: in vitro models for investigating cellular damage and neuroprotective strategies

Oxidative stress in neurodegeneration: in vitro models for investigating cellular damage and neuroprotective strategies Alime Garip1, Denitsa Stefanova1, Virginia Tzankova1 1 Department of Pharmacology, Pharmacotherapy and Toxicology, Faculty of Pharmacy, Medical University of Sofia, Sofia, Bulgaria Corresponding author: Alime Garip ([email protected]) Received 17 October 2025♦ Accepted 24 October 2025♦ Published 24 November 2025 Citation: Garip A, Stefanova D, Tzankova V (2025) Oxidative stress in neurodegeneration: in vitro models for investigating cellular damage and neuroprotective strategies. Pharmacia 72: 1–12. https://doi.org/10.3897/pharmacia.72.e175265 Abstract Neurodegenerative disorders such as Parkinson’s disease (PD) are driven by complex and multifactorial mechanisms, among which oxidative stress plays a central pathogenic role. A sustained imbalance between reactive oxygen species (ROS) production and antioxidant defenses contributes to mitochondrial dysfunction, lipid peroxidation, and dopaminergic neuronal loss. This review focuses on oxidative stress-induced neurodegeneration and explores how in vitro models can be effectively used to study the cellular consequences of oxidative damage. Particular emphasis is placed on toxin-based models, including 6-hydroxydopamine (6-OHDA) and 1-methyl-4-phenylpyridinium (MPP+), as well as cellular systems such as immortalized cell lines, primary neurons, and induced pluripotent stem cell (iPSC)-derived neurons. The applicability, advantages, and limitations of each model are discussed in the context of mimicking PD-related oxidative damage and screening for neuroprotective strategies. Ultimately, this review underscores the importance of selecting appropriate in vitro models for dissecting oxidative stress pathways and advancing neuroprotective research in PD. Keywords in vitro, neuroprotection, oxidative stress, Parkinson’s disease, toxin-based models Introduction Parkinson’s disease (PD) is the second most common neurodegenerative disorder after Alzheimer’s disease, affecting approximately 1–2% of individuals over the age of 65 (Tysnes and Storstein 2017). It is clinically characterized by the progressive loss of dopaminergic neurons in the substantia nigra pars compacta and the accumulation of Lewy bodies, composed mainly of misfolded and aggregated α-synuclein (Dauer and Przedborski 2003; Shulman et al. 2011). Although the exact etiology of PD remains unclear, a combination of genetic predispositions and environmental exposures contributes to disease onset and progression (Warner and Schapira 2003; Simon et al. 2020). Several genetic mutations have been identified in familial and sporadic forms of PD, including those in SNCA, LRRK2, PARK2, PINK1, and DJ-1 (Alexander 2004; Rui et al. 2018). However, monogenic causes account for only a minority of cases, and the majority are considered idiopathic. Increasing evidence suggests that oxidative stress plays a central role in the pathogenesis of both familial and sporadic PD (Zhou et al. 2008; Gaki and Papavassiliou 2014). Oxidative stress arises from an imbalance between reactive oxygen species (ROS) production and the cellular antioxidant defense systems. Dopaminergic neurons are particularly vulnerable due to their high metabolic demand, dopamine autooxidation, abundant mitochondrial activity, and elevated iron levels—all of which promote the Copyright Garip A et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Pharmacia 72: 1–12 DOI 10.3897/pharmacia.72.e175265 Review Article Garip A et al.: In vitro models of oxidative stress in neurodegeneration2 generation of ROS (Ozcan and Ogun 2015; Burbulla et al. 2017). These ROS can induce damage to DNA, proteins, and lipids, contributing to neurodegeneration and neuronal death (Zhang et al. 1999; Floor and Wetzel 2002). Given the complexity of PD pathogenesis, the use of in vitro models has become an essential strategy in experimental neuroscience. These models offer a controlled environment to dissect molecular mechanisms involved in PD, such as mitochondrial dysfunction, α-synuclein aggregation, glutamate excitotoxicity, and redox imbalance (Lopes et al. 2017; Xicoy et al. 2017). Moreover, in vitro systems allow for high-throughput screening of neuroprotective agents and mechanistic studies of cell death and survival, facilitating the development of potential therapeutic strategies (Noraberg et al. 2005; Soldner et al. 2009). This review aims to present an integrated overview of the molecular mechanisms underlying Parkinson’s disease, with an emphasis on oxidative stress, lipid peroxidation, and antioxidant defense. Special attention is paid to in vitro models commonly used in PD research, evaluating their advantages, limitations, and relevance for studying PD-associated neurotoxicity and neuroprotection. Oxidative stress and mechanisms of dopaminergic neurodegeneration in Parkinson’s disease Oxidative stress is recognized as a pivotal upstream event in the pathogenesis of Parkinson’s disease (PD), contributing significantly to the selective degeneration of dopaminergic neurons. It arises from an imbalance between the excessive generation of reactive oxygen species (ROS) and the limited capacity of endogenous antioxidant defenses. The substantia nigra pars compacta (SNpc)—the primary site of dopaminergic neuron loss—is particularly susceptible to oxidative insults due to its high metabolic activity, abundant iron content, and the intrinsic redox vulnerability of dopaminergic neurons (Gaki and Papavassiliou 2014; Burbulla et al. 2017). This oxidative burden promotes a cascade of cellular damage, including mitochondrial dysfunction, protein misfolding, lipid peroxidation, and ultimately, neuronal death. Molecular sources of ROS in PD Mitochondria represent the major source of ROS within neurons, particularly under conditions of impaired oxidative phosphorylation. In PD, complex I (NADH:ubiquinone oxidoreductase) activity is frequently compromised in the SNpc, leading to increased electron leakage and the formation of superoxide anions (O₂•–) (Schapira et al. 1990). These radicals are subsequently dismutated to hydrogen peroxide (H₂O₂), which, in the presence of ferrous iron, undergoes Fenton chemistry to yield hydroxyl radicals (•OH)—among the most cytotoxic ROS (Ozcan and Ogun 2015). Age-related respiratory deficiencies, mitochondrial DNA mutations, and impaired electron transport further exacerbate ROS production, establishing a vicious cycle of oxidative stress and bioenergetic failure. Notably, dopaminergic neurons exhibit heightened dependence on mitochondrial ATP generation, which amplifies their susceptibility to redox-driven injury (Gandhi and Abramov 2012). A summary of the major endogenous sources of ROS, their mechanisms of generation, and associated cellular consequences contributing to dopaminergic neurodegeneration in PD is presented in Table 1. Beyond mitochondrial sources, dopamine metabolism itself constitutes a significant contributor to oxidative stress. Under normal conditions, dopamine is sequestered into synaptic vesicles via vesicular monoamine transporter 2 (VMAT2), minimizing cytosolic oxidation. In PD, impaired VMAT2 function and elevated cytosolic dopamine levels promote dopamine autoxidation, generating reactive quinones and H₂O₂ (Hastings et al. 1996). Dopamine-quinones covalently modify thiol groups on cysteine-rich proteins, disrupting protein function and enhancing α-synuclein aggregation (Burbulla et al. 2017). Additionally, enzymatic degradation of dopamine by monoamine oxidase B (MAO-B) further contributes to H₂O₂ accumulation, compounding the oxidative burden. This dual pathway of dopamine oxidation—both enzymatic and spontaneous—plays a central role in dopaminergic vulnerability. Iron dyshomeostasis in the SNpc represents another critical driver of oxidative neurodegeneration in PD. Postmortem analyses consistently reveal elevated levels of redox-active Fe²+ in PD brains, particularly in regions affected by dopaminergic loss (Dexter et al. 1989; Ward et Table 1. Major molecular sources of ROS and their consequences in Parkinson’s disease. Source of ROS Mechanism of generation Key molecules / enzymes Main consequences Mitochondrial dysfunction Complex I inhibition → electron leakage → O₂•–, H₂O₂ NADH dehydrogenase (Complex I), mitochondrial DNA ↑ROS, energy failure, apoptosis Dopamine oxidation Autooxidation and MAO-B catabolism of cytosolic dopamine Dopamine, MAO-B, VMAT2 Quinone and H₂O₂ production, αsynuclein aggregation Iron accumulation Fenton reaction (Fe²+ + H₂O₂ → •OH) Fe²+, ferritin, transferrin, ferroportin Lipid peroxidation, protein oxidation, αsyn aggregation Microglial activation NADPH oxidase and iNOS → O₂•–, NO•, ONOO– NOX2, iNOS, TNFα, IL-1β Neuroinflammation, oxidative/ nitrosative stress Astrocytic dysfunction ↓EAAT1/2, ↓GSH synthesis EAAT1, EAAT2, glutamate, GCL, GS Glutamate excitotoxicity, impaired neuronal redox balance Pharmacia 72: 1–12 3 al. 2014). Labile iron participates in the Fenton reaction, converting H₂O₂ into highly reactive hydroxyl radicals that damage lipids, nucleic acids, and proteins. Moreover, iron accelerates α-synuclein aggregation through direct binding, facilitating the formation of toxic oligomers and fibrils (Uversky et al. 2001). Thus, iron contributes not only to oxidative injury but also to the proteinopathy characteristic of PD. The homeostatic regulation of iron, normally maintained by proteins such as ferritin (storage), transferrin (transport), and ferroportin (export), is often impaired in PD. Downregulation of ferroportin and insufficient ferritin expression result in cytosolic iron overload (Morris et al. 2017). Targeting iron accumulation has emerged as a promising therapeutic strategy. Iron chelators such as deferiprone have demonstrated efficacy in reducing brain iron levels and improving motor function in both animal models and early-phase clinical trials (Devos et al. 2014). These findings underscore the central role of metal-induced oxidative mechanisms in PD pathogenesis and highlight their relevance as druggable targets. Targets of oxidative damage Lipid peroxidation represents one of the earliest and most detrimental manifestations of oxidative stress in Parkinson’s disease (PD). The abundant presence of polyunsaturated fatty acids (PUFAs) in neuronal membranes— especially in dopaminergic neurons of the substantia nigra—renders them particularly susceptible to peroxidative injury. Hydroxyl radicals (•OH), among the most reactive ROS, initiate this process by abstracting hydrogen atoms from PUFAs, triggering chain reactions that compromise membrane fluidity, permeability, and integrity (Gaki and Papavassiliou 2014). Among the most studied lipid peroxidation by-products are malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE), which serve both as biomarkers and mediators of cellular toxicity. These aldehydes form covalent adducts with nucleic acids, proteins, and membrane lipids, thereby altering their structure and function (Barrera et al. 2018). Elevated concentrations of MDA and 4-HNE have been consistently detected in the substantia nigra of PD patients in postmortem analyses (Yoritaka et al. 1996; Zhou et al. 2008). Notably, 4-HNE modifies mitochondrial proteins and components of the respiratory chain, impairing ATP synthesis and exacerbating energy failure. Furthermore, it promotes the aggregation of α-synuclein into β-sheet-rich oligomers, thus reinforcing proteotoxic stress and neuronal degeneration (Breen et al. 2013). Protein oxidation constitutes another hallmark of oxidative damage in PD and significantly contributes to neuronal dysfunction. ROS-induced modifications alter the conformation and activity of a wide range of proteins, including mitochondrial enzymes, cytoskeletal elements, and antioxidant defenses (Stadtman and Levine 2000). These changes may lead to loss of enzymatic activity, enhanced degradation, and impaired cellular signaling. Of particular importance is the oxidation of α-synuclein, which accelerates its misfolding and aggregation into toxic oligomers. These aggregates interfere with synaptic function, mitochondrial dynamics, and autophagy–lysosomal pathways, ultimately forming Lewy bodies—a pathological hallmark of PD (Poon et al. 2005; Shulman et al. 2011). Mitochondrial proteins such as complex I are especially vulnerable to oxidative modification, leading to further disruption of electron transport and increased ROS generation (Schapira et al. 1990). Inactivation of key enzymes—including glutathione peroxidase, superoxide dismutase, and ubiquitin ligases—further amplifies redox imbalance and impairs proteostasis (Borre et al. 2018). Nucleic acids are also major targets of oxidative stress. Neurons, due to their high metabolic activity and limited regenerative capacity, are particularly susceptible to oxidative DNA and RNA damage. Hydroxyl radicals attack nucleobases and sugar-phosphate backbones, generating strand breaks and mutagenic lesions. One of the most widely recognized biomarkers is 8-hydroxy-2’-deoxyguanosine (8-OHdG), formed by oxidative damage to guanine residues. Increased levels of 8-OHdG have been observed in the SNpc of PD patients and correlate with disease severity (Zhang et al. 1999). Oxidative damage to RNA, including mRNA, tRNA, and rRNA, disrupts translation fidelity and protein synthesis. Given its single-stranded structure and cytosolic localization near ROS-generating organelles like mitochondria, RNA is particularly prone to oxidation. This results in the production of truncated or misfolded proteins, further aggravating proteotoxic stress (Nunomura et al. 2006). Moreover, impaired DNA repair mechanisms exacerbate oxidative injury. Deficiencies in base excision repair (BER) enzymes such as 8-oxoguanine glycosylase 1 (OGG1) have been reported in PD brains, leading to the persistence of oxidized nucleotides, genomic instability, and apoptotic signaling (Sanders et al. 2014). Antioxidant defenses and their depletion in Parkinson’s disease In healthy neurons, redox homeostasis is maintained by a sophisticated network of enzymatic and non-enzymatic antioxidant systems that neutralize reactive oxygen species (ROS) and prevent damage to lipids, proteins, and nucleic acids. Among these, glutathione (GSH) serves as the most abundant and essential intracellular antioxidant. GSH detoxifies ROS such as hydrogen peroxide (H₂O₂) and lipid hydroperoxides through enzymatic reactions catalyzed by glutathione peroxidase (GPx) and is regenerated from its oxidized form (GSSG) via glutathione reductase (GR) (Perry et al. 1982; Seiler et al. 2008). In Parkinson’s disease (PD), a marked depletion of GSH has been consistently reported in the substantia nigra pars compacta (SNpc), preceding both the degeneration of dopaminergic neurons and the formation of Lewy bodies (Dexter et al. 1989; Sian et al. 1994). This early reduction in GSH impairs the neuronal capacity to detoxify H₂O₂ Garip A et al.: In vitro models of oxidative stress in neurodegeneration4 and lipid peroxides, thereby amplifying oxidative stress and contributing to neuronal vulnerability. Beyond GSH, several antioxidant enzymes—including superoxide dismutases (SOD1 and SOD2), catalase, and glutathione peroxidase 4 (GPx4)—play crucial roles in maintaining redox balance. Dysregulation of these enzymes has been observed in PD brains, characterized by altered expression patterns and decreased enzymatic activity (Seiler et al. 2008; Gaki and Papavassiliou 2014). Notably, GPx4 is indispensable for reducing phospholipid hydroperoxides and inhibiting ferroptosis, an iron-dependent form of cell death increasingly implicated in PD pathology (Stockwell et al. 2017). Furthermore, oxidative stress may disrupt the transcriptional control of antioxidant defense genes. Nuclear factor erythroid 2–related factor 2 (Nrf2) is a master regulator of antioxidant gene expression, including those involved in GSH biosynthesis. Reduced Nrf2 activity in PD has been associated with impaired upregulation of protective genes, further exacerbating oxidative damage (Lastres-Becker et al. 2012). Together, these findings suggest that depletion of endogenous antioxidants in PD is not merely a consequence of disease progression but a primary pathogenic mechanism. This highlights the antioxidant system as a promising therapeutic target for counteracting neurodegeneration in Parkinson’s disease. Glial cell involvement in oxidative stress Microglia, the resident immune cells of the central nervous system (CNS), play a dual role in neuroinflammation and oxidative stress. Under physiological conditions, they contribute to neuronal homeostasis through continuous environmental surveillance. In Parkinson’s disease (PD), however, microglia become chronically activated, adopting a pro-inflammatory M1 phenotype characterized by the release of reactive oxygen species (ROS) and pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β), and interleukin-6 (IL-6) (Block et al. 2007; Zhou et al. 2008). This persistent activation generates a neurotoxic microenvironment that exacerbates dopaminergic neuron degeneration. Activated microglia produce superoxide radicals (O₂•–) via NADPH oxidase and nitric oxide (NO) through the action of inducible nitric oxide synthase (iNOS). These reactive species can interact to form peroxynitrite (ONOO– ), a highly reactive nitrogen species that induces lipid peroxidation, mitochondrial dysfunction, and protein nitration (Gao and Hong 2008). Furthermore, microglial activation is perpetuated by feedback mechanisms involving danger-associated molecular patterns (DAMPs), such as extracellular α-synuclein aggregates released by degenerating neurons (Kim et al. 2013). Astrocytes also play a key role in maintaining redox equilibrium by supplying glutathione (GSH) precursors to neurons and regulating extracellular glutamate levels via excitatory amino acid transporters (EAAT1 and EAAT2) (Dringen 2000). In PD, astrocytes exhibit morphological atrophy and functional impairment, collectively referred to as astroglial dystrophy, which compromises their antioxidant capacity and neuroprotective support (Booth et al. 2017). Moreover, astrocytes may undergo a phenotypic shift toward a neurotoxic A1 state, losing neurotrophic functions and releasing pro-inflammatory mediators. This transition is associated with downregulation of glutamate transporters and reduced expression of GSH-synthesizing enzymes, promoting glutamate excitotoxicity and oxidative stress (Liddelow et al. 2017). As such, astrocytic dysfunction contributes to non-cell-autonomous mechanisms of dopaminergic neurodegeneration in PD. The interrelated processes of mitochondrial dysfunction, dopamine oxidation, iron accumulation, and glial activation that collectively drive oxidative stress and dopaminergic cell death in Parkinson’s disease are illustrated in Fig. 1. Figure 1. Mechanisms of oxidative stress–induced dopaminergic neurodegeneration in Parkinson’s disease. Pharmacia 72: 1–12 5 Consequences: signaling pathways and cell death Reactive oxygen species (ROS) are not merely passive agents of molecular damage; they also serve as active modulators of intracellular signaling cascades that culminate in neuronal dysfunction and cell death. Among the most extensively studied ROS-responsive pathways in Parkinson’s disease (PD) is the mitogen-activated protein kinase (MAPK) cascade. Oxidative stress activates upstream kinases such as apoptosis signal-regulating kinase 1 (ASK1), which phosphorylate downstream effectors like c-Jun N-terminal kinase (JNK) and p38 MAPK, resulting in pro-apoptotic gene expression, mitochondrial dysfunction, and activation of cell death pathways (Wang et al. 2004; Choi et al. 2012). Another critical pathway is the nuclear factor kappa-lightchain-enhancer of activated B cells (NF-κB), a transcription factor that integrates signals from oxidative and inflammatory stimuli. In PD models, chronic NF-κB activation leads to transcriptional upregulation of pro-inflammatory cytokines, promoting glial activation and contributing to progressive neurodegeneration (Ghosh et al. 2007). In parallel, the tumor suppressor protein p53 is activated by ROS and DNA damage, orchestrating mitochondrial outer membrane permeabilization (MOMP), release of cytochrome c, and subsequent caspase activation (Duan et al. 2002). Beyond apoptosis, ferroptosis has emerged as an alternative and highly relevant form of regulated cell death in PD. This iron-dependent process is characterized by glutathione depletion, inactivation of glutathione peroxidase 4 (GPx4), and accumulation of lipid peroxides (Dixon et al. 2012; Do Van et al. 2016). Morphologically, ferroptosis is distinguished by shrunken mitochondria, absent chromatin condensation, and loss of plasma membrane integrity. The high iron content and abundance of polyunsaturated fatty acids in substantia nigra pars compacta (SNpc) neurons make them particularly vulnerable to ferroptotic death (Stockwell et al. 2017). Cumulatively, oxidative stress serves as a central driver of dopaminergic neurodegeneration in PD through its impact on redox-sensitive signaling pathways and cell death programs. ROS derived from dysfunctional mitochondria, dopamine metabolism, and iron overload inflict widespread damage to cellular macromolecules—including lipids, proteins, and nucleic acids. This damage triggers downstream signaling cascades such as MAPK, NFκB, and p53, ultimately resulting in apoptosis, ferroptosis, and sustained neuroinflammation. Compounding this process is the depletion of endogenous antioxidants, including glutathione (GSH), superoxide dismutase (SOD), catalase, and GPx4, which reduces the neuronal capacity to neutralize ROS. Additionally, glial dysfunction—manifested by microglial activation and astrocytic failure—further amplifies oxidative stress and contributes to a toxic microenvironment. Thus, understanding the interplay between ROS, signaling networks, and regulated cell death pathways provides a mechanistic foundation for developing neuroprotective strategies targeting oxidative stress in PD. Collectively, oxidative stress acts as both a trigger and an amplifier of dopaminergic neurodegeneration in Parkinson’s disease, initiating a cascade of mitochondrial dysfunction, lipid peroxidation, and protein misfolding that culminates in neuronal death. However, neurons possess intrinsic adaptive mechanisms designed to counteract redox imbalance and to restore cellular homeostasis. These endogenous neuroprotective systems, encompassing both enzymatic and non-enzymatic antioxidant defenses, play a decisive role in mitigating oxidative injury and maintaining neuronal viability. Understanding these defense pathways provides essential insight into the molecular basis of neuronal resilience and offers a foundation for the development of targeted neuroprotective strategies. Mechanisms of neuroprotection The neuronal capacity to withstand oxidative stress relies on an intricately coordinated hierarchy of defense mechanisms that restrict the generation of reactive oxygen species (ROS), neutralize pre-existing radicals, and repair oxidatively modified biomolecules before irreversible dysfunction occurs. These protective systems function within both aqueous and lipid compartments and are broadly divided into enzymatic and non-enzymatic antioxidant networks (Godic et al. 2014). In the central nervous system (CNS), their importance is amplified by the brain’s disproportionately high oxygen consumption, enrichment in polyunsaturated lipids, and limited regenerative potential. Among all neuronal populations, dopaminergic neurons of the substantia nigra pars compacta are particularly susceptible to oxidative stress due to their elevated metabolic activity, reliance on mitochondrial respiration, and relatively low antioxidant buffering capacity. Dysregulation of essential antioxidant enzymes—including superoxide dismutase (SOD), catalase, and glutathione peroxidases (GPx)—has been documented in multiple brain regions and correlates with mitochondrial impairment and progressive neuronal loss in Parkinson’s disease (Migliore and Coppedè 2009; Zhang et al. 2018). Enzymatic antioxidant systems Enzymatic antioxidants constitute the first line of defense against ROS. Superoxide dismutases (SODs) catalyze the dismutation of the superoxide anion (O₂•–) into hydrogen peroxide (H₂O₂), thereby preventing its interaction with nitric oxide and subsequent formation of peroxynitrite (ONOO–). Three human isoforms are distinguished by metal cofactors and localization: Cu/Zn-SOD (SOD1) in the cytosol, Mn-SOD (SOD2) in mitochondria, and extracellular SOD (SOD3) in the interstitial space (Pisoschi and Pop 2015). Catalase subsequently decomposes H₂O₂ into water and molecular oxygen, preventing its involvement in Fenton chemistry that yields highly reactive hydroxyl radicals (•OH). Peroxiredoxins and thioredoxin reductases act synergistically to reduce organic peroxides and to maintain protein thiol homeostasis, ensuring dynamic equilibrium between oxidation and reduction across cellular compartments. Garip A et al.: In vitro models of oxidative stress in neurodegeneration6 The glutathione system The glutathione (GSH) system represents the cornerstone of intracellular antioxidant defense. It encompasses reduced glutathione, glutathione reductase (GR), glutathione peroxidases (GPx), and glutathione S-transferases (GSTs). GR maintains the cellular pool of reduced GSH by catalyzing the NADPH-dependent reduction of oxidized glutathione (GSSG) (Meister and Anderson 1983). GPx enzymes, characterized by their selenium-containing active sites, catalyze the reduction of both hydrogen and lipid hydroperoxides. GPx1 efficiently detoxifies soluble H₂O₂, whereas GPx4 uniquely reduces complex lipid peroxides embedded within phospholipid bilayers (Lobo et al. 2010). GSTs contribute to cytoprotection by conjugating electrophilic xenobiotics and lipid peroxidation products to GSH, facilitating their neutralization and removal (Josephy 2010). In Parkinson’s disease (PD), early and profound depletion of GSH in the substantia nigra (Perry et al. 1982) leads to a decline in GPx activity, mitochondrial instability, and redox disequilibrium. Notably, impairment of GPx4 promotes ferroptosis—an iron-dependent, lipid peroxide–driven cell death pathway increasingly recognized as a major contributor to dopaminergic neurodegeneration (Seiler et al. 2008). Thus, the glutathione system not only detoxifies reactive intermediates but also determines the threshold between neuronal survival and ferroptotic vulnerability. Non-enzymatic antioxidant defenses Non-enzymatic antioxidants complement enzymatic defenses by directly scavenging free radicals and regenerating oxidized enzyme cofactors. Glutathione itself, through its cysteinyl thiol group, acts as a ubiquitous intracellular redox buffer. Melatonin, an amphiphilic indoleamine, neutralizes a broad spectrum of ROS and reactive nitrogen species (RNS)—including hydroxyl radicals, superoxide anions, and nitric oxide—while upregulating the transcription of antioxidant enzymes within the GSH and thioredoxin systems (Reiter et al. 2017). Vitamin E (α-tocopherol), the predominant lipid-soluble antioxidant, localizes within cellular membranes, where it interrupts lipid peroxidation chain reactions by donating hydrogen atoms to lipid peroxyl radicals. Vitamin C (ascorbic acid), a water-soluble antioxidant, acts synergistically by regenerating α-tocopherol from its oxidized form, thereby sustaining membrane protection (Birben et al. 2012; Pisoschi and Pop 2015). Experimental and clinical studies have shown that melatonin and vitamin E exert neuroprotective effects in PD models by attenuating oxidative damage, preserving dopaminergic neurons, and enhancing endogenous antioxidant activity (Shoulson 1998; GonzálezGonzález et al. 2018). Collectively, these small-molecule antioxidants serve as a secondary, rapid-response system that complements enzymatic detoxification processes and maintains redox homeostasis under stress conditions. Inhibition and detoxification of lipid peroxidation Suppression of lipid peroxidation constitutes a pivotal neuroprotective mechanism against oxidative stress–induced neuronal damage. Polyunsaturated fatty acids (PUFAs), highly abundant in neuronal membranes, are prone to radical-mediated peroxidation, yielding cytotoxic aldehydes such as 4-hydroxynonenal (4-HNE) and malondialdehyde (MDA). These reactive by-products propagate oxidative damage and covalently modify proteins, lipids, and nucleic acids, thereby promoting α-synuclein aggregation and mitochondrial dysfunction (Yoritaka et al. 1996; Niki et al. 2005). Pharmacological inhibition of lipoxygenase (LOX) enzymes, particularly 5-LOX, mitigates enzymatic lipid hydroperoxide formation. The clinically approved 5-LOX inhibitor zileuton functions by disrupting iron coordination within the enzyme’s catalytic center, thereby curtailing lipid peroxide synthesis. More recently, supplementation with deuterated PUFAs has emerged as an innovative strategy. The substitution of bis-allylic hydrogen atoms with deuterium strengthens the C–D bond, slowing radical-mediated hydrogen abstraction and consequently inhibiting both enzymatic and non-enzymatic lipid peroxidation (Gaschler and Stockwell 2017). Glutathione peroxidase 4 (GPx4) remains indispensable for detoxifying pre-formed lipid hydroperoxides, including hydroperoxyeicosatetraenoic and hydroperoxyoctadecadienoic acids. The enzyme’s selenocysteine residue catalyzes nucleophilic attack on the peroxide oxygen, forming a transient selenic acid intermediate that is subsequently reduced by GSH. This catalytic cycle regenerates the active enzyme and oxidizes GSH to GSSG, completing a critical antioxidant loop that maintains neuronal membrane stability. The dynamic interplay between oxidative injury and antioxidant defense ultimately dictates dopaminergic neuron survival under metabolic and redox stress. When the enzymatic and non-enzymatic defense systems falter, redox homeostasis collapses toward oxidative disequilibrium, precipitating lipid peroxidation, ferroptosis, and neurodegeneration. Deciphering the molecular crosstalk among these protective pathways has greatly refined our understanding of neuronal vulnerability and resilience in Parkinson’s disease. Nevertheless, translating these mechanistic insights into actionable neuroprotective strategies demands experimental platforms that accurately recapitulate the oxidative microenvironment of the human brain. In vitro neuronal models—spanning immortalized neuroblastoma lines, primary neuronal cultures, and stem cell–derived dopaminergic systems—provide controllable, reproducible frameworks for dissecting redox-regulated signaling, screening antioxidant compounds, and modeling oxidative neurotoxicity. These systems bridge the conceptual gap between molecular redox biology and translational therapeutics, forming the experimental foundation for the next stage of research: the development and validation of in vitro models for studying oxidative stress–driven neuronal injury and protection. Pharmacia 72: 1–12 7 In vitro models for studying neuronal toxicity To experimentally validate the molecular mechanisms of oxidative stress and neuroprotection described above, reliable in vitro systems are indispensable. These models enable mechanistic exploration of dopaminergic vulnerability, allow for quantitative evaluation of antioxidant interventions, and provide a high-resolution platform for screening neuroprotective agents. In contrast to complex in vivo environments, in vitro approaches offer precise control over experimental conditions, enhanced reproducibility, and direct manipulation of redox parameters at the cellular and subcellular levels (Freshney 2001). Such systems have become the cornerstone of contemporary neurodegeneration research, particularly for elucidating the oxidative mechanisms that underlie Parkinson’s disease (PD). Principles and rationale for in vitro modeling of Parkinson’s disease Effective in vitro modeling of PD requires mimicking the distinctive features of dopaminergic neurons from the substantia nigra pars compacta (SNpc)—cells that are uniquely susceptible to oxidative stress due to their high metabolic activity, dopamine auto-oxidation, and abundant mitochondrial content. Dopamine metabolism itself generates hydrogen peroxide (H₂O₂), superoxide (O₂•–), and dopamine quinones, thereby reinforcing a pro-oxidant intracellular milieu. Consequently, appropriate cell models must preserve the metabolic, enzymatic, and signaling characteristics of dopaminergic neurons, including expression of tyrosine hydroxylase (TH), the dopamine transporter (DAT), and vesicular monoamine transporter 2 (VMAT2), as well as functional mitochondrial respiration (Lopes et al. 2017). In vitro modeling provides unique opportunities to dissect the cascade of oxidative injury—from ROS formation and lipid peroxidation to mitochondrial dysfunction and cell death. These systems also facilitate mechanistic studies of protective agents such as polyphenols, caffeine derivatives, and synthetic antioxidants under tightly regulated oxidative environments. By integrating pharmacological and biochemical endpoints, in vitro studies bridge the gap between molecular hypothesis and translational neuroprotection. Neurotoxin-based in vitro models of oxidative stress The most widely used models of PD-related oxidative stress are those employing specific dopaminergic neurotoxins, including 6-hydroxydopamine (6-OHDA) and 1-methyl-4-phenylpyridinium (MPP+). These compounds selectively target catecholaminergic neurons and reproduce essential hallmarks of PD, such as mitochondrial dysfunction, ROS overproduction, and apoptotic cell death. 6-OHDA is a synthetic catecholamine that is selectively taken up by dopaminergic and noradrenergic neurons via DAT and NAT transporters. Once internalized, it undergoes enzymatic (MAO-A) and auto-oxidation, producing hydrogen peroxide, superoxide radicals, and reactive quinones (Simola et al. 2007). These quinones covalently modify cysteine residues on cellular proteins, leading to enzyme inactivation and oxidative damage. Moreover, 6-OHDA oxidation yields neuromelanin-like pigments, whose extracellular release upon neuronal death activates microglia and initiates secondary neuroinflammation (Varešlija et al. 2020). In vitro, 6-OHDA is primarily applied to SH-SY5Y, PC12, and primary mesencephalic cultures. Its advantages include high selectivity for catecholaminergic neurons and reproducible dose-dependent toxicity, allowing for consistent assessment of antioxidant and neuroprotective compounds. However, at excessive concentrations, 6-OHDA induces non-specific cytotoxicity due to overwhelming ROS formation. Despite this limitation, it remains an invaluable model for investigating oxidative stress–induced mitochondrial dysfunction, glutathione depletion, and apoptotic signaling (Blum et al. 2001). The neurotoxin MPTP and its metabolite MPP+ represent another classical system for reproducing dopaminergic neurodegeneration in vitro. Following metabolic conversion by glial monoamine oxidase B (MAO-B), MPP+ is selectively transported into dopaminergic neurons through the dopamine transporter (DAT). Within mitochondria, it inhibits complex I (NADH:ubiquinone oxidoreductase), resulting in ATP depletion, ROS generation, and activation of apoptotic cascades involving p38 MAPK and JNK pathways (Gao et al. 2003). MPP+ toxicity has been extensively characterized in SH-SY5Y, LUHMES, and iPSC-derived dopaminergic neurons, offering high reproducibility and mechanistic relevance. While it faithfully recapitulates mitochondrial oxidative damage, it does not reproduce α-synuclein aggregation or Lewy body formation. Therefore, it is most effective for dissecting redox imbalance, mitochondrial permeability transition, and ferroptotic pathways under controlled oxidative conditions (Kalkman and Loetscher 2003; Zeng et al. 2006). Cellular systems for modeling dopaminergic neurodegeneration The choice of cellular model is critical for ensuring biological relevance, reproducibility, and translational value. Below are the most widely used systems, categorized by origin, phenotype, and applicability to oxidative stress research. SH-SY5Y Cells Derived from human neuroblastoma, SH-SY5Y cells exhibit catecholaminergic properties and express tyro- Garip A et al.: In vitro models of oxidative stress in neurodegeneration8 sine hydroxylase (TH) and dopamine-β-hydroxylase. These cells can be differentiated into neuron-like phenotypes using retinoic acid, brain-derived neurotrophic factor (BDNF), or phorbol esters, which enhance dopaminergic markers such as DAT and VMAT2 (Agholme et al. 2010). They are a mainstay of in vitro PD research owing to their human origin, cost-effectiveness, and ease of genetic and pharmacological manipulation. Their limitations include a tumor-derived genotype, limited maturation, and gradual loss of dopaminergic traits at high passage numbers. Nevertheless, they remain ideal for oxidative stress assays, neuroprotection screening, and mechanistic studies on mitochondrial dysfunction and apoptosis. LUHMES Cells LUHMES (Lund human mesencephalic) cells originate from the human embryonic mesencephalon and represent an advanced model of post-mitotic dopaminergic neurons. Upon differentiation with tetracycline, cAMP, and GDNF, they express a complete dopaminergic profile (TH, DAT, VMAT2, and D2 receptors) and exhibit spontaneous electrical activity (Zhang et al. 2014). Their homogeneity, stable phenotype, and human origin make them superior to immortalized lines for longterm and mechanistic toxicity studies. They are particularly suitable for MPP+ and 6-OHDA models, as well as siRNA-mediated modulation of redox enzymes. The main challenge is their strict differentiation requirement and higher sensitivity to oxidative stress, necessitating meticulous culture conditions. PC12 Cells PC12 cells, derived from rat pheochromocytoma, provide a robust system for studying neurite outgrowth, neurotrophin signaling, and oxidative stress responses (Malagelada and Greene 2008). They synthesize dopamine and norepinephrine but lack dopamine transporters (DAT), which limits their susceptibility to 6-OHDA and MPP+. Their metabolism relies heavily on glycolysis, resulting in reduced sensitivity to mitochondrial inhibitors. Nonetheless, they remain highly valuable for evaluating neurotrophic protection, NGF-induced differentiation, and antioxidant response pathways. Primary dopaminergic cultures Primary midbrain dopaminergic neurons isolated from rodent embryos most closely replicate the native neuronal microenvironment. They exhibit authentic morphology, expression of TH, DAT, and VMAT2, and maintain neuron–glia interactions that influence oxidative homeostasis. These cultures are instrumental for studying neuroinflammation, microglial activation, and the interplay between neuronal and glial oxidative mechanisms (Kalkman and Loetscher 2003). However, ethical limitations, variability among preparations, and interspecies differences in redox regulation restrict their widespread use. Despite this, they provide unparalleled physiological fidelity and serve as a critical reference for validating findings from immortalized lines. iPSC-derived dopaminergic neurons Induced pluripotent stem cells (iPSCs), reprogrammed from adult somatic cells through expression of Oct4, Sox2, Klf4, and c-Myc, represent the most advanced human-based in vitro model. When differentiated into dopaminergic neurons, they recapitulate human-specific transcriptional and metabolic features, expressing TH, DAT, and α-synuclein (Soldner et al. 2009). iPSC-derived neurons provide an unprecedented opportunity to model both sporadic and familial PD, examine patient-specific oxidative susceptibilities, and evaluate targeted antioxidant therapies. Limitations include labor-intensive protocols, high cost, and batch-to-batch variability in differentiation efficiency. Nevertheless, they constitute a vital tool for precision medicine and personalized neuroprotection research. Synaptosomes Synaptosomes—isolated synaptic terminals derived from rodent or human brain tissue—offer a metabolically active preparation enriched in mitochondria and neurotransmitter machinery. They are ideal for studying presynaptic events such as dopamine uptake, vesicular storage, and calcium-dependent neurotransmitter release (Jhou and Tai 2017). While they lack nuclear and postsynaptic elements, their simplicity enables high-resolution analysis of oxidative stress at the synaptic level, as well as pharmacological testing of antioxidant and transporter-targeted compounds. Synaptosomes are particularly valuable when human brain material is available, bridging the translational gap between cell cultures and human pathology. A comparative overview of commonly employed models is presented in Table 2, summarizing their origin, major features, and experimental applicability. The diversity of in vitro systems reflects the complexity of oxidative stress in PD and provides complementary insights at molecular, cellular, and synaptic levels. Each model captures a distinct dimension of dopaminergic vulnerability—from mitochondrial dysfunction in SH-SY5Y cells to redox regulation in LUHMES and synaptosomal preparations. Far from being oversimplified representations of the brain, these systems serve as precise analytical tools for dissecting redox-dependent signaling, evaluating antioxidant efficiency, and elucidating neuroprotective Pharmacia 72: 1–12 9 mechanisms. Collectively, they form the experimental foundation for advancing translational strategies that target oxidative stress in Parkinson’s disease. Conclusion In vitro models constitute indispensable platforms for deciphering the cellular and molecular mechanisms underlying Parkinson’s disease. By enabling precise control over the extracellular milieu and intracellular redox balance, these systems allow for high-resolution analysis of dopaminergic neuron biology, oxidative injury, and neuroprotective signaling. Each experimental model—ranging from immortalized cell lines such as SH-SY5Y and PC12 to advanced human-based systems like LUHMES and iPSC-derived neurons—provides complementary insights into the multifaceted nature of neurodegeneration. Toxin-based paradigms employing 6-hydroxydopamine (6-OHDA) and 1-methyl-4-phenylpyridinium (MPP+) remain fundamental for recapitulating key pathogenic processes, including mitochondrial impairment, glutathione depletion, and ROS-driven apoptosis. Meanwhile, iPSC-derived dopaminergic neurons bridge the translational gap by capturing patient-specific genetic and metabolic susceptibilities, thereby facilitating individualized drug discovery and redox-modulatory research. Although no single model can reproduce the full pathophysiological spectrum of PD, the strategic combination of multiple in vitro systems substantially enhances experimental fidelity and translational relevance. Future advancements will depend on the refinement of differentiation protocols, the incorporation of glia– neuron co-culture systems, and the development of three-dimensional and organoid platforms that more accurately emulate the cytoarchitecture and oxidative microenvironment of the human midbrain. Integration of such complex models with high-throughput screening and omics-based analytics will further strengthen the predictive value of preclinical investigations. Ultimately, the evolution of in vitro modeling stands as a critical bridge between mechanistic neuroscience and therapeutic innovation, accelerating the discovery of effective interventions for Parkinson’s disease. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statements The authors declared that no clinical trials were used in the present study. The authors declared that no experiments on humans or human tissues were performed for the present study. The authors declared that no informed consent was obtained from the humans, donors or donors’ representatives participating in the study. The authors declared that no experiments on animals were performed for the present study. The authors declared that no commercially available immortalised human and animal cell lines were used in the present study. Use of AI No use of AI was reported. Funding This study was financed by the European Union—NextGenerationEU through the National Recovery and Resilience Plan of the Republic of Bulgaria, project No. BG-RRP-2.004-0004-C01. Author contributions In vitro models of oxidative stress in neurodegeneration Data availability All of the data that support the findings of this study are available in the main text. Table 2. Summary of widely used in vitro models for Parkinson’s disease research, including their origin, main features, and application domains. Model Origin Advantages Limitations Application in PD 1. SH-SY5Y Human neuroblastoma Easy to culture, widely used, can be differentiated Phenotype loss over time, tumor origin Testing antioxidants, 6OHDA and MPP+ toxicity 2. LUHMES Human embryonic mesencephalon Homogeneo us population, stable dopaminergic markers Requires complex differentiation, toxinsensitive Genetic manipulation, chronic toxicity studies 3. PC12 Rat pheochromocytoma Robust, suitable for neurotrophin assays Lacks DAT, low mitochondrial sensitivity Neurotrophininduced differentiation studies 4. Primary dopaminergic cultures Embryonic rodent (midbrain) High physiologica l relevance, glial interaction Variability, ethical considerations Neuroinflammation, mitochondrial dysfunction 5. Organotypic brain slices Neonatal rodents (in situ brain) 3D architecture, functional connectivity Short in vitro viability, technically demanding Connectivity betwen brain regions 6. iPSC-derived neurons Reprogrammed from patient somatic cells Human origin, models familial mutations Expensive, differentiation variability Modeling αsynuclein, mitochondria, screening 7. Synaptosoms Human brain tissue (postmortem) Useful for studying presynaptic function Lacks complete neuronal architecture Presynaptic activity, transporters and receptors