2. Preliminary Evaluation of the Neuroprotective Effects of Potential Drugs In Vitro
Currently developed methods for obtaining individual nerve cells allow for high-quality and reliable information about the state of nervous tissue under pathological process modeling conditions, and enable the study of the effects of potential neuroprotectors on them [
15]. For in vitro studies, male white rats aged 4 weeks and weighing 80–100 g are used. Neuron and neuroglia enrichment fractions are isolated in two stages. In the first stage, brain tissue is disintegrated to obtain a cell suspension. In the second stage, differential ultracentrifugation is performed in a sucrose and Ficoll density gradient. The isolated neuronal cells are washed from the sucrose and albumin with cold physiological solution (solution temperature 4°C). Cell lines of nerve cells are also used [
16].
The obtained suspension is divided into series as follows:
- Intact: A suspension of neurons without the addition of initiating agents or potential neuroprotectors under investigation.
-
Control: A suspension of neurons to which agents that induce oxidative and nitrosative stress, glutamatergic excitotoxicity, and glutathione thiol-disulfide system deprivation are added, at concentrations capable of causing the death of 50% of neurons (0.1-5 µM). To initiate
oxidative stress in the neuron suspension, 0.25-1.0 mM H₂O₂ is added to the incubation medium.
Glutamatergic “excitotoxicity” is induced by adding kainate (200-400 µM), glutamate (0.1-10 mM), or N-methyl-D-aspartate (100-150 µM) to the incubation medium.
Glutathione thiol-disulfide system deprivation is achieved by introducing chloro-2,4-dinitrobenzene (CDNB) (50-500 µM), a selective inhibitor of glutathione S-transferase that forms conjugates with glutathione in cytosolic and mitochondrial fractions, into the incubation medium [
16,
17,
18,
19,
20,
21]. Here is a translation suitable for a textbook: Alongside the intact and control samples, samples with the addition of initiating agents and pharmacological agents at various concentrations are prepared, followed by determining their effective concentration. The neuroprotective activity of potential neuroprotectors is assessed by counting neurons exhibiting signs of apoptosis using flow cytometry or histochemical methods. Our studies have shown that the addition of the aforementioned neurotoxins to the incubation medium led to a pronounced disruption of cellular molecular-biochemical processes. These disruptions were consistent in nature but varied in degree of severity—accompanied by a sharp shift in the thiol-disulfide balance towards oxidized thiols (a significant decrease in reduced glutathione concentration and an increase in oxidized glutathione). An increase in the marker of oxidative protein damage—nitrotyrosine—was observed, along with a decrease in the activity of mitochondrial superoxide dismutase (Mn-SOD). We also recorded dynamic changes in the synthesis of endogenous cytoprotective factors—HSP and HIF proteins.
These pathobiochemical changes were detected as early as 15 minutes into incubation, reaching peak values by 60 minutes. Thus, in vitro studies have established that modeling glutamatergic excitotoxicity and depriving the glutathione component of the thiol-disulfide system lead to the development of nitrative stress, resulting in irreversible molecular-biochemical and morphofunctional impairments of neuronal cells. It was found that the extent of these impairments correlated with changes in the expression of HSP and HIF proteins [
8].
Recent research has established the central role of mitochondrial dysfunction in the processes of cell death under conditions of brain ischemia. In this context, it is pertinent to investigate the presence of what is referred to as "mitoprotective activity" in potential neuroprotectors [
8].
For in vitro investigation of the "mitoprotective activity" of potential neuroprotectors, the mitochondrial fraction is isolated from rat brain tissue using differential centrifugation. In the incubation mixture (70 mM sucrose, 5 mM HEPES, 70 mM KCl, 0.5–1 mM KH₂PO₄, pH 7.4), a suspension of mitochondria (1 mg protein per sample) is added, along with the substances to be tested (10–100 µM in the sample), and incubated for 2 minutes. The opening of the mitochondrial pore in the mitochondrial suspension can be initiated by the following agents [
22,
23,
24,
25,
26]:
MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine)-induced opening of mitochondrial pores: add 40–60 µM MPTP to the incubation medium and after 5 minutes, add 50 µM CaCl₂.
Ca²⁺-induced opening of mitochondrial pores: add 200 µM CaCl₂ to the incubation medium.
NO-induced opening of mitochondrial pores: add 20–100 µM sodium nitroprusside to the incubation medium and after 2 minutes, add 50 µM CaCl₂.
H₂O₂-induced opening of mitochondrial pores: add 50 mM hydrogen peroxide to the incubation medium and after 2 minutes, add 50 µM CaCl₂.
Mitoprotective activity is assessed based on the ability of the tested substance to prevent the opening of mitochondrial pores (MP) and reduce mitochondrial membrane potential (Ψ), as well as through a range of biochemical parameters [
27].
We have determined that the addition of MPTP (50 µM) to the mitochondrial suspension led to pronounced disturbances in energy processes, characterized by a decrease in succinate dehydrogenase and malate dehydrogenase activities, along with an increase in nitrotyrosine, starting from the 30th minute of incubation. Additionally, mitochondrial dysfunction was observed, including MP opening and changes in mitochondrial membrane potential. The process of MP opening was most actively observed in the mitochondrial suspension at the 60-minute incubation mark [
8]. It is important to emphasize that in vitro studies, both in neuronal suspensions and in mitochondria, are useful for screening a large array of potential neuroprotectors as well as for more detailed investigation of the most promising compounds to elucidate their mechanisms of neuroprotective action.
3. Preliminary Assessment of the Neuroprotective Effects of Potential Pharmaceutical Agents Using Various Models of Cerebral Ischemia
Despite potential drawbacks associated with experimental models of cerebral ischemia, these models are widely used in pharmacological, physiological, and pathomorphological research. This is primarily due to the relative ease of reproducing cerebral ischemia in animals and the ability to control biochemical, immunocytochemical, and histological changes in brain tissue at various time points after ischemia onset (seconds, minutes, days). These factors justify the use of brain ischemia models both for studying the mechanisms of ischemic neurodegeneration and for preclinical evaluation of potential neuroprotective agents [
28,
29,
30].
Cerebral ischemia models can be reproduced in both large animals (such as rabbits, dogs, pigs, and primates) and small animals (such as mice and rats). Using large animals in studies of cerebral ischemia offers several advantages [
31]. In large animals, it is easier to conduct complex physiological (such as electroencephalography, rheography, and gas exchange) and biochemical (including energy metabolism indicators, oxidative stress markers, and molecular markers of neural tissue damage) monitoring of the brain [
32]. All these measurements can be performed simultaneously, at equivalent time intervals, on the same animals. However, modeling ischemia in large animals is a costly and labor-intensive endeavor, and experimental ischemia is not always reproducible. The use of large animals often requires various methods of anesthesia, which can also affect the quality of ischemia reproduction [
33]. Additionally, the ethics committees and animal protection organizations do not support such research [
34,
35].
Using small animals has several advantages. Small animals, especially rodents, are less expensive. Mice, in particular, are genetically homogeneous, and genetic modifications can be relatively easily performed to create transgenic animals [
36]. Another significant advantage, particularly with the use of mice and rats, is the ability to study complex behaviors, neuromotor functions, and memory to assess the severity of ischemia. The small size of the mouse and rat brains allows for procedures such as rapid freezing, which facilitates subsequent biochemical and cytochemical analyses [
37,
38]. Advantages and disadvantages of the most common models of ischemic stroke are represented in
Figure 1.
There are several main experimental models of acute ischemic stroke (AIS), including global, focal, multifocal ischemia, and intracerebral hemorrhage. Modeling intracerebral hemorrhage involves either rupturing cerebral blood vessels or injecting autologous blood [
14,
39].
Most focal ischemia models involve occlusion of one of the major vessels supplying blood to a localized region of the brain, resulting in a reduction of local cerebral blood flow to 30%, which corresponds to the second critical level [
31]. In multifocal ischemia, multiple distinct areas of reduced blood flow are observed. A characteristic feature of focal ischemia is the presence of a central ischemic necrotic core surrounded by a penumbra zone. This penumbra contains neurons that are damaged or dying but still viable, as well as neurons with no pronounced changes [
40]. Most models of focal brain ischemia involve occlusion of the middle cerebral artery (MCA) in either small or large animals. MCA occlusion results in reduced cerebral blood flow in both the striatum and the cortex. The extent and distribution of ischemia depend on the duration of the artery occlusion, the location of the occlusion, and the amount of collateral blood flow to the MCA region [
41,
42].
There are various methods for artery occlusion to reproduce ischemia, either permanent or temporary, at proximal or distal sections of the vessel. These models have been extensively used due to their similarity to human stroke manifestations. The availability of these models has led to their widespread application for evaluating the effectiveness of different neuroprotective agents, understanding the mechanisms of brain damage during ischemia, and determining the roles of genes, proteins, lipids, and messengers involved in the development of ischemia [
43,
44]. A method for permanent proximal artery occlusions in rats has been developed, which includes subtemporal (semi-temporal) craniotomy. Using this model leads to the formation of an infarct zone in the cortex. In the permanent (constant) artery occlusion model, the local cerebral blood flow in cortical areas where histological anomalies occur is approximately 25 ml per 100 g per minute. The ischemic damage area corresponds to the region with reduced local cerebral blood flow [
45,
46,
47]. Different variant of the focal brain ischemia model involves occlusion of the MCA in conjunction with occlusion of the common carotid artery [
48]. Most researchers prefer using electrocoagulation for occluding the middle cerebral artery to reproduce ischemia [
49].
Another model of focal ischemia involves thromboembolism of cerebral vessels in rats and larger animals. This model consists of introducing homologous blood clots directly into the common carotid artery or through a retrograde catheter placed in the external carotid artery [
50]. Several models of brain ischemia-reperfusion have also been developed using a 30-minute temporary occlusion of the MCA. However, this technique requires advanced technical equipment [
51,
52]. There are recommendations for using this model of temporary occlusion in non-human primates, which provides a more objective assessment of behavioral, neurological, and cognitive impairments [
53,
54]. However, focal ischemia models induced by MCA occlusion have several limitations that restrict their widespread use in pharmacological experiments. Primarily, the technique is labor-intensive, requiring specialized equipment and involving the need for skull opening, which often damages the dura mater. This method does not facilitate neurochemical and biochemical studies. Only 30-45% of rats exhibit neurological deficits, and these are typically mild to moderate (up to 3-5 points on the McGraw scale). The recorded mortality of animals, particularly within the first day of the experiment, is usually due to complications associated with the reproduction of this model [
8,
14,
41].
A relatively lesser-known method for reproducing focal permanent transient occlusion in rodents is the use of an intraluminal filament. This method has been widely used since its development in the late 1980s for studying both mechanisms of brain injury and neuroprotection. It involves two stages. In the first stage, the common carotid arteries are occluded, and in the second stage, an occluder is inserted intravascularly into the carotid artery and advanced to the middle cerebral artery. The occluder used is a chromic gut suture, which swells upon contact with blood and adheres firmly to the vessel walls. This ensures high reliability and reproducibility of the method. Due to its minimally invasive nature and the small area of ischemic damage, the mortality rate is 25-40% in animals. Animals that survive the procedure can be used to study post-ischemic recovery, including behavioral and exploratory activity, learning and memory, and mechanisms of neuroprotection [
55,
56,
57].
A variant of focal ischemia models is photo-induced vessel thrombosis, where a constant ischemic focus is formed with respect to both volume and localization. This methodology is based on the principle of photochemical stimulation of thrombus formation in cerebral vessels through the interaction of a light beam with a fluorescent dye that has been previously introduced into the bloodstream. A limitation of this technique is that the photochemical reaction may induce microvascular damage. Focal photo-induced brain ischemia is primarily used in screening studies and behavioral research [
58,
59,
60].
Global brain ischemia occurs due to disruption of blood flow in major arterial vessels (such as the aorta) or acute cardiac arrest. This model can be replicated through techniques such as carotid clamping or decapitation. It is characterized by selective damage to neurons located in various brain structures [
61,
62,
63]. In global cerebral ischemia, the absence of cerebral blood flow leads to neuronal damage. Global brain ischemia can be classified into total (complete) and subtotal (incomplete) ischemia.
Total ischemia is utilized to study the regenerative capacities and delayed neuronal death in selectively vulnerable brain regions. The most straightforward method to induce complete global ischemia without recirculation is decapitation. This technique has been used for many years in small animals to investigate pathobiochemical changes associated with global ischemia. Following decapitation, the head is frozen and homogenized for metabolic and biochemical studies.
In experimental medicine, a cervical inflatable cuff has been used for an extended historical period to reproduce global ischemia in animals [
64]. However, this method is associated with several complicating factors: venous congestion and compression of the vagus nerve, which can exert undesirable (distorting) effects on the development of ischemia. Cervical compression using an inflatable cuff has also been applied in large animals (dogs) to induce global brain ischemia. However, in this case, the vertebral arteries must be occluded separately due to their proximity to the spine. Cervical compression for 2 minutes using this method leads to loss of consciousness, followed by complete recovery. After 4-6 minutes of ischemia induced by this technique, animals remain in a comatose state for 24 hours, after which there is a full recovery of normal neurological functions. After 8 minutes of compression, permanent neurological deficits occur. Similar procedures have been conducted in monkeys, involving a reduction in arterial pressure by 50 mm Hg before compression. Using this brain ischemia model in monkeys, hippocampal disturbances were observed after 15 minutes of ischemia, with neurological deficits persisting for many days following the ischemic event [
65,
66].
Ventricular fibrillation is another method for inducing complete global brain ischemia [
67]. This technology is used to simulate the clinical condition of cardiac arrest, and many researchers further employ resuscitation to continue the state. This method has predominantly been used in large animals. Although it reproduces complete ischemia, it is labor-intensive and costly due to the subsequent need for animal resuscitation. In this method, cerebral perfusion pressure is low, and blood flow levels are significantly reduced compared to the control group, despite attempts to increase perfusion pressure through the administration of adrenaline.
The cardiac arrest model conducted in mice (with 6-10 minutes of cardiac arrest) results in prolonged damage to the hippocampus and the sensorimotor cortex [
68,
69]. Ventricular fibrillation leads to ischemia in most organs. Therefore, models involving occlusion of the cephalic arteries in the neck and thoracic cavity are preferred, as they prevent complete brain ischemia while allowing for targeted ischemia in renal, visceral, and other areas [
31]. A brain ischemia model has also been developed in cats, which involves occlusion of the ring-shaped and left subclavian arteries near their origins in the aortic arch, combined with a reduction in arterial pressure by administering ganglioblockers to levels below 80 mm Hg, and even as low as 50 mm Hg [
8,
70].
In this model, cerebral blood flow was reduced to near zero throughout almost the entire brain volume. As with other ischemic models, the consequences of pathological changes and behavior were proportional to the duration of the induced ischemia. Blood flow decreased to 10% of baseline levels in the neocortex, striatum, and hippocampus. Fifteen to thirty minutes after the removal of occluders, severe cerebral hyperemia developed, lasting up to 15 minutes. This was followed by acute cerebral hypoperfusion, which persisted for 24-36 hours in the affected brain region. Currently, experimental models of complete global ischemia are rarely used in pharmacological research due to their complexity, high mortality rates, and the inability to perform prolonged dynamic monitoring of the animals [
71].
Currently, there is significant interest in models of transient global brain ischemia induced by occlusion of major vessels. These models are used to study the mechanisms of delayed neuronal death in various brain regions and to identify potential neuroprotective agents [
43,
72].
In rats, a model of transient four-vessel occlusion is used. The modeling is performed in two stages. Initially, atraumatic occluders are placed near each common carotid artery and secured to the animal's neck. The vertebral arteries are then electrocoagulated. After three days, the blood flow in the common carotid arteries is temporarily interrupted using the occluders [
73,
74]. This model promotes global ischemia: in 75% of animals, after 10-15 minutes of four-vessel occlusion, remote damage to pyramidal neurons in the CA1 and CA3 regions of the hippocampus was observed, while after 30 minutes of occlusion, neurons in the 3rd and 5th layers of the sensorimotor cortex were affected. Mortality rates for this model range from 20% to 40% during the hyperperfusion period (1-3 days).
The four-vessel occlusion model is complex to perform and requires specific skills from the experimenter.
Models of brain ischemia that induce moderate reductions in cerebral blood flow (subtotal ischemia) through irreversible occlusion of major vessels are more widely used in pharmacological research [
14,
41,
75]. The most commonly used models involve unilateral or bilateral occlusion of the common carotid arteries, depending on the species-specific anatomical features of cerebral blood supply and collateral circulation capabilities in the animals. Mongolian gerbils or gerbils are particularly preferred for this purpose due to their unique and convenient vascular anatomy, making them ideal for studying neuroprotection under brain ischemia conditions. A distinctive feature of gerbils is the disconnection of the circle of Willis in cerebral circulation. To reproduce persistent global brain ischemia in gerbils, irreversible unilateral occlusion of the common carotid artery is performed. In this model, 30-40% of gerbils develop severe neurological deficits (15-20 points on the C. P. McGraw scale), while others experience moderate deficits and unilateral infarction 3-5 days after occlusion. The mortality rate in this model reaches 40-60% within the first 7 days post-occlusion [
76,
77,
78].
In the absence of gerbils, white rats, both inbred and outbred strains, are commonly used to induce a similar level of ischemia. In these cases, ischemia is induced by bilateral occlusion of the common carotid arteries. This procedure results in a 50-60% reduction in cerebral blood flow, which gradually recovers to 85-90% and eventually returns to baseline levels within 2-3 days due to the compensatory activation of collateral circulation. The ischemic process involves sequential stages of brain perfusion disturbances: initial post-ischemic hyperemia is followed by a phase of post-ischemic hypoperfusion. Incomplete reperfusion limits the survival of ischemic tissue and leads to additional oxidative and osmotic brain damage. In 40-50% of the operated animals, severe neurological deficits develop (15-20 points on the C. P. McGraw scale), with persistent cognitive deficits observed up to 21 days post-occlusion. The mortality rate in this model reaches 60-80% within the first 7 days of observation.
To reduce mortality, a modified model involves complete irreversible occlusion of one common carotid artery and reduction of blood flow by 50-70% in the second common carotid artery. In this modified model, mortality is 40-60%. Bilateral ligation of the common carotid arteries is performed under general anesthesia through a surgical approach involving dissection of the carotid arteries and simultaneous placement of silk ligatures. It is important to note that this model is traumatic, and animal mortality, particularly in the first few hours, often results from errors made by experimenters, such as inadvertently ligating both the common carotid artery and the vagus nerve. Therefore, a properly executed procedure for modeling cerebral ischemia through bilateral occlusion of the common carotid arteries results in a mortality rate of 60-80%, depending on the strain of rats.
Bilateral occlusion of the common carotid arteries is associated with typical ischemia-induced biochemical disturbances in the brain: activation of glycolysis with hyperproduction of lactate, suppression of Krebs cycle enzymes and the electron transport chain, and ATP depletion accompanied by reduced expression of HSP70 and HIF-1α on days 1 and, at most, day 4 post-occlusion. By days 7 and, most notably, day 15 of the experiment, some improvement in energy metabolism in the brain was observed, but even by day 21, these metrics had not returned to control levels (sham-operated rats). This model of cerebral ischemia has demonstrated that inhibition of malate production and NAD-MDH activity correlates with decreased ATP, HSP70, and HIF-1α levels, as well as the severity of neurological impairments [
8,
79].
This model of cerebral ischemia is associated with persistent disruptions in the brain's thiol-disulfide system, characterized by a reduction in the concentration of reduced glutathione, an increase in its oxidized form, and decreased expression of glutathione reductase and glutathione peroxidase 1 and 4. The most significant decrease in activity of these enzymes on day 4 of the experiment correlates with diminished levels of glutathione and the overall thiol pool. This hypothesis is supported by two factors: firstly, the difference in the reduction of glutathione peroxidase (GPx) and glutathione reductase (GR) activity, particularly on day 1, and secondly, the maximal reduction in thiol-containing free amino acids in brain tissues on day 4 of the experiment [
8,
80]. The maximum accumulation of oxidative stress markers (such as nitrotyrosine, malondialdehyde, and carbonylated proteins) in brain tissues (CA1 hippocampus and cortex) occurred between days 1 and 4 of the experiment. Starting from day 15 of observation, the oxidative stress response was attenuated, but even on day 21, concentrations of nitrotyrosine and malondialdehyde (MDA) remained significantly higher than control levels (sham-operated rats). We found that on days 1 and, most notably, day 4 after bilateral occlusion of the common carotid arteries, there was an increase in the levels of Fe, Cu, and Mn in the cortex of the rats compared to the control group. The oxidative-reduction pair Fe²⁺/Fe³⁺ plays a crucial role in the activation of free radical reactions. Fe²⁺ is essential in all systems that generate reactive oxygen species, particularly in the formation of hydroxyl radicals through Fenton and Haber-Weiss reactions [
8].
The disruption of the overall histostructure of the brain following bilateral irreversible occlusion of the common carotid arteries was characterized by pronounced perivascular and pericellular edema, significant vascular congestion, and accumulation of blood beneath the pia mater. There was also marked ischemic pathology in the ependymal cells of the brain ventricles. For the first 3 days after the bilateral occlusion, diffuse gliocyte hyperplasia predominated. Starting from day 5 post-occlusion, there was an increased intensity of nuclear staining in pyramidal neurons of the cortex, and somewhat less so in the cytoplasm of cortical neurons.
Pathomorphological changes in the ependymal lining of the ventricles, such as the appearance of scalloped edges and desquamation of cells and cell fragments into the lumen of the lateral and third ventricles, were most pronounced between days 3 and 5 of the experiment. During days 1 to 3 of the experiment, there was an increase in nuclear size, with pyknotic nuclei observed on day 1. Changes in neuronal morphology were noted from day 1 through day 15, with the most significant alterations occurring on day 3 of observation. A decrease in the size of pyramidal neurons was evident on days 1 to 3, and by days 5 to 7, cell vacuolation in the tissue was observed, indicating brain edema. Morphometric studies revealed a reduction in neuronal density in the CA1 region of the hippocampus and layers IV-V of the cortex, decreased RNA concentration in the nuclei and cytoplasm, and increased density of apoptotic neurons, with the most pronounced changes occurring between days 1 and 5 of the experiment [
8].
Among all the aforementioned experimental models of cerebral ischemia, only the subtotal ischemia model (bilateral irreversible occlusion of the common carotid arteries) is associated with pronounced pathobiochemical disturbances and activation of oxidative stress reactions in brain tissues. This makes it a suitable model for evaluating agents with antioxidant and anti-ischemic activity. Additionally, this model is preferred for studying cerebroprotective drugs that exhibit vasotropic, anti-edematous, antioxidant, and other protective activities. It allows for the assessment of the drug's efficacy on brain regions with varying degrees of ischemia and the identification of potential side effects of vasotropic agents, such as the "steal" phenomenon.
The generalized methods for modeling AIS are presented in
Scheme 1, while the biochemical, cytochemical, and morphological consequences of AIS are detailed in
Scheme 2.
5. Determination of Oxidative Stress Markers and Antioxidant System Status
Oxidative stress reactions, throughout their various stages (initial, free-radical, and peroxide stages), produce a range of products resulting from the interactions between free radicals and biological macromolecules. The levels of these products can indicate the intensity of oxidative stress within different regions of the brain, including neurons, glial cells, and subcellular structures. Therefore, these products can serve as markers of neurodestruction and the efficacy of potential neuroprotective agents [
81,
82]. The most important markers of oxidative stress are products of polyunsaturated fatty acid (PUFA) oxidation. These include short-chain alkanes and alkenes, as well as alkanals such as 2,4-alkadienals, alkatrienals, hydroxyalkenals, 4-hydroxyalkenals, and their peroxides, MDA, and normal aliphatic ketones. Isoprostanes, which are products of the interaction between arachidonic acid and free radicals, can also serve as markers of oxidative stress.
Markers of oxidative stress also include reactive oxygen species and stable metabolites of nitric oxide. The latter are formed from the metabolism of nitrite and peroxynitrite radicals, as well as from nitrosylated S-, N-, and O-macromolecules. Additionally, products of nucleic acid oxidation, such as 8-hydroxy-2'-deoxyguanosine, can be used as indicators of oxidative stress [
83,
84]. Each group of oxidative stress products reflects the intensity of oxidative stress and the degree of oxidative modification of lipids, amino acids, nucleic acids, and specific functional groups of proteins, including SH- and NH- groups, methionine SCH3 groups, and L-lysine amino groups. These modifications can lead not only to changes in proteins and enzyme activity but also to the destruction of bioantioxidants (such as vitamins, ubiquinone, steroid hormones, etc.), alterations in the phospholipid composition of biomembranes, and the appearance of oxidation products in the hydrophobic regions of the cell. These products can inhibit ion transport processes, alter the conformation of lipids and proteins, and consequently affect the structural and functional properties of membranes. However, the precise quantitative determination of oxidative stress products is an exceedingly complex process. This complexity arises primarily because, in biological systems, products of oxidative stress reactions are simultaneously metabolized (through oxidation, reduction, isomerization, polymerization, etc.). Additionally, a significant portion of these products is unstable and has a short lifespan. Furthermore, the low concentrations of these products necessitate highly sensitive physicochemical methods for identification. Moreover, the transformation of oxidative stress reaction products can occur during the processing of biological material, especially when examining ischemic tissues. For example, the free radical processes may be intensified when the homogenate or extracted system comes into contact with atmospheric oxygen. Therefore, the determination of oxidative stress products formed in ischemic tissues should be performed under conditions that prevent contact with air, to avoid prolonging chain reactions [
85,
86].
Oxidative stress reaction products can be conditionally categorized into the following groups represented in
Table 1 [
8,
87]:
The most widely used biomarkers for oxidative stress are stable products formed as a result of the oxidation of PUFAs, arachidonic acid, nucleic acids, tyrosine, and arginine. As previously noted, the peroxidation of PUFAs leads to the formation of peroxide compounds (hydroperoxides, endoperoxides, dialkylperoxides, and epoxides) and conjugated diene products. These conjugated dienes are formed through the redistribution of electronic density in the molecules of linoleic, linolenic, and arachidonic acids (but not oleic acid) [
88]. Since diene conjugation appears during the formation of free radicals, its presence in lipid extracts from animal brains indicates the generation of free radicals and thus confirms the free radical mechanism of PUFA oxidation.
Secondary products of oxidative stress arise from the breakdown of PUFA hydroperoxides, leading to the formation of a wide range of carbonyl compounds (n-alkenals, 2-alkenals, 2,4-alkadienals, alkatrienals, hydroxyalkenals, hydroperoxyalkenals, 4-hydroxyalkenals, 4-hydroperoxyalkenals, MDA, ketones, alkanes, and alkenes). Due to their chemical nature (stability), these compounds serve as primary markers of oxidative stress [
89,
90].
In addition to MDA, the primary carbonyl compounds formed from the peroxidation of ω-6 PUFAs include hexanal and 4-hydroxy-2,3-trans-nonenal; ω-3 PUFAs form propanal and 4-hydroxy-2,3-trans-hexenal. Beyond these compounds, free radical–peroxidative oxidation also produces 4-hydroxy-2,3-octenal, 4-hydroxydecenal, 4-hydroxyundecenal, 4,5-dihydroxydecenal, 4-hydroxy-2,5-nonanal, 2-hydroxyheptenal, 2-hydroxyhexenal, butanal, pentanal, octenal, and nonenal, though these are present in smaller amounts. Some of these products exhibit cytotoxic and mutagenic effects and can interact with biomolecules (such as proteins, nucleic acids, etc.), altering the structure of receptors, ion channels, and enzymes. They can also inhibit the synthesis of intracellular mediators and induce DNA and RNA damage. Unlike free radicals, carbonyl compounds are more stable and can exist both inside and outside the cell [
91,
92,
93].
To determine secondary products, physical-chemical methods are commonly used. These methods are based on the absorption of energy by carbonyl compounds or their interaction products with analytical reagents in the ultraviolet region of the spectrum [
94].
Among the stable products of oxidative stress reactions, MD) is commonly used as a marker. It is assessed through its reaction with thiobarbituric acid (TBA) and quantified spectrophotometrically [
95].
This method is relatively simple to perform and has thus found widespread application. However, it lacks specificity because TBA reacts not only with MDA but also with other aldehydes, amino acids, carbohydrates, and bilirubin. These limitations of the TBA test have led to the development of numerous modifications. These include selective extraction of the azomethine complex using n-butanol and performing the reaction in the presence of iron (II) salts, which leads to the breakdown of hydroperoxides, among other adjustments [
96]. Currently, fundamentally new methods have been developed for the determination of MDA and carbonyl compounds (such as formaldehyde, acetaldehyde, acetone, and both saturated and unsaturated aldehydes). These methods involve the use of specific reagents, including:
Pentafluorophenylhydrazine (PFPH);
Methylhydrazine (MH);
4-(2-phthalimidyl)benzohydrazine (FBH);
2,4-Dinitrophenylhydrazine (DNPH);
o-(2,3,4,5,6-Pentafluorobenzyl)hydroxylamine hydrochloride (PFBH);
tert-Butyldimethylchlorosilane (BDMCS);
N,O-Di-(trimethylsilyl)-trifluoroacetamide (DTSFA);
2-Hydrazinobenzothiazole (HBT).
These methods involve the quantitative determination of reaction products using techniques such as Thin-Layer Chromatography (TLC), Gas Chromatography (GC), High-Performance Liquid Chromatography (HPLC), Ultraviolet-Visible (UV-Vis) Spectrophotometry, Infrared (IR) Spectroscopy, and Mass Spectrometry (MS) [
97,
98,
99].
Recently, stable products of oxidative reactions that can be used as markers of oxidative stress include oxidative modification products of arachidonic acid, specifically isoprostanes. 8-epi-Isoprostane (8-epi-PGFa2) can be accurately quantified in the daily urine of experimental animals [
100].
Cyclic endoperoxides (PgG2-PgH2), which are formed from arachidonic acid through the action of cyclooxygenase, are unstable products and thus are not used as markers. Arachidonic acid hydroperoxides, which are produced by lipoxygenase, are inactivated by glutathione peroxidase (GPx) or leukotriene-A-synthase, converting into leukotrienes. It should be noted that arachidonic acid hydroperoxides are formed in relatively low concentrations, making their use as markers of oxidative stress less feasible [
101].
Oxidative modification of nucleic acids also results in products that serve as markers of oxidative stress. For instance, the damaging effects of hydroxyl radicals and singlet oxygen on nucleic acids lead to the formation of products such as 5-hydroxymethyluracil, 8-hydroxyadenine, thymidine glycol, and 8-hydroxyguanine. 8-Hydroxyguanine is chemically stable, making it a useful marker for oxidative damage to the genome. Recent studies have shown that the level of 8-hydroxyguanine is significantly elevated in urine in various neurodegenerative diseases and after experiencing an acute stroke. Additionally, it is detected in urine much earlier than these diseases are clinically diagnosed [
102]. Another significant biomarker of genome damage is thymidine glycol. However, it has not found widespread use as a marker of oxidative stress due to the more complex methods required for its isolation and detection [
103]. During the activation of oxidative stress reactions in body tissues, the sequential breakdown of lipid hydroperoxides and the cleavage of alkoxyl radicals produce lower hydrocarbons. These hydrocarbons are exhaled, and their concentration in exhaled air serves as a specific and highly sensitive indicator of the intensity of oxidative stress processes. Quantitative analysis of hydrocarbons in exhaled air is conducted using GC. The primary end product of oxidative modification of PUFAs is pentane, which constitutes approximately 90% of the total, while the remaining 10% is made up of heptane and hexane [
104]. Since methods for determining lower hydrocarbons are non-invasive, these markers are recommended for dynamic monitoring in larger animals during chronic experiments, as well as to help preserve the animal's life.
Schiff bases, as products of the interaction between carbonyl compounds and the amino groups of proteins, amino acids, and nucleic acids, are also extracted and identified as markers of oxidative stress [
105].