Antioxidant Activity of Quercetin in a H2O2-Induced Oxidative Stress Model in Red Blood Cells: Functional Role of Band 3 Protein

During their lifespan, red blood cells (RBCs) are exposed to a large number of stressors and are therefore considered as a suitable model to investigate cell response to oxidative stress (OS). This study was conducted to evaluate the potential beneficial effects of the natural antioxidant quercetin (Q) on an OS model represented by human RBCs treated with H2O2. Markers of OS, including % hemolysis, reactive oxygen species (ROS) production, thiobarbituric acid reactive substances (TBARS) levels, oxidation of protein sulfhydryl groups, CD47 and B3p expression, methemoglobin formation (% MetHb), as well as the anion exchange capability through Band 3 protein (B3p) have been analyzed in RBCs treated for 1 h with 20 mM H2O2 with or without pre-treatment for 1 h with 10 μM Q, or in RBCs pre-treated with 20 mM H2O2 and then exposed to 10 µM Q. The results show that pre-treatment with Q is more effective than post-treatment to counteract OS in RBCs. In particular, pre-exposure to Q avoided morphological alterations (formation of acanthocytes), prevented H2O2-induced OS damage, and restored the abnormal distribution of B3p and CD47 expression. Moreover, H2O2 exposure was associated with a decreased rate constant of SO42− uptake via B3p, as well as an increased MetHb formation. Both alterations have been attenuated by pre-treatment with 10 μM Q. These results contribute (1) to elucidate OS-related events in human RBCs, (2) propose Q as natural antioxidant to counteract OS-related alterations, and (3) identify B3p as a possible target for the treatment and prevention of OS-related disease conditions or aging-related complications impacting on RBCs physiology.


Introduction
Red blood cells (RBCs) are unique, highly specialized, and the most abundant cells in different organisms. Although their primary function is transportation of the respiratory gases O 2 and CO 2 between lungs and tissues, these circulatory cells are equipped with potent endogenous anti-oxidative systems that make them mobile free radical scavengers and provide antioxidant protection, not only to RBCs themselves but also to other tissues and organs in the body [1]. The production of reactive oxygen and nitrogen species (ROS/RNS) generated during cellular metabolism in biological systems is balanced by the ability of the latter to defend through their sophisticated antioxidant machinery [2][3][4][5][6]. Nevertheless, when oxidants are produced in excess, or when the antioxidant defenses regulating them are ineffective, this balance can be perturbed, thus resulting in oxidative stress (OS) [3,[7][8][9][10][11]. In these conditions, organic biomolecules can be altered through oxidation to an extent that exceeds repair capacity [12]. Oxidative modifications lead 3 of 19 those human pathologic conditions having acute OS as hallmark, including hemolytic anemia, vaso-occlusion, and progressive vascular injury affecting multiple organ systems, thus affecting RBC integrity. Since we have demonstrated that B3p function as a sensitive tool to assess the impact of OS on homeostasis of RBCs [48][49][50][51][52][53][54], the anion exchange capability through B3p and the effects of OS on cell membrane structure have been evaluated. represented by human RBCs treated with non-toxic concentrations of H2O2 [47]. Th based model could represent those human pathologic conditions having acute hallmark, including hemolytic anemia, vaso-occlusion, and progressive vascular affecting multiple organ systems, thus affecting RBC integrity. Since we demonstrated that B3p function as a sensitive tool to assess the impact of O homeostasis of RBCs [48][49][50][51][52][53][54], the anion exchange capability through B3p and the of OS on cell membrane structure have been evaluated. Figure 1 reports the hemolysis measurement in human RBCs showed as perce of hemoglobin released in a 0.9% NaCl solution at room temperature (25 °C). Red cells were treated for 1 h at 25 °C with 20 mM H2O2. This exposure did not in hemolysis percentage compared to cells left untreated (control). Quercetin alone (1 °C) did not significantly affect hemolysis. Based on this result, 20 mM H2O2 concent has been used for the experiments shown in the following.

Evaluation of Erythrocyte Cell Shape
In the following, the experimental design shown in Figure 2 applies. As depicted in Figure 3, incubation for 1 h at 25 °C with 20 mM H2O2 in morphological alterations of RBCs. In fact, in this condition we detected 37.

Evaluation of Erythrocyte Cell Shape
In the following, the experimental design shown in Figure 2 applies. Q on RBCs [42][43][44][45]. Because RBCs deliver oxygen to the entire body, the maintenance of a constant amount of RBCs is of pivotal importance for the health of an individual [46]. We hypothesized that Q could exert a beneficial role against oxidative injury in human RBCs. Thus, we explored the potential protective effect of Q (10 µ M) in a model of OS represented by human RBCs treated with non-toxic concentrations of H2O2 [47]. This cellbased model could represent those human pathologic conditions having acute OS as hallmark, including hemolytic anemia, vaso-occlusion, and progressive vascular injury affecting multiple organ systems, thus affecting RBC integrity. Since we have demonstrated that B3p function as a sensitive tool to assess the impact of OS on homeostasis of RBCs [48][49][50][51][52][53][54], the anion exchange capability through B3p and the effects of OS on cell membrane structure have been evaluated. Figure 1 reports the hemolysis measurement in human RBCs showed as percentage of hemoglobin released in a 0.9% NaCl solution at room temperature (25 °C). Red blood cells were treated for 1 h at 25 °C with 20 mM H2O2. This exposure did not increase hemolysis percentage compared to cells left untreated (control). Quercetin alone (1 h at 37 °C) did not significantly affect hemolysis. Based on this result, 20 mM H2O2 concentration has been used for the experiments shown in the following.

Evaluation of Erythrocyte Cell Shape
In the following, the experimental design shown in Figure 2 applies. As depicted in Figure 3, incubation for 1 h at 25 °C with 20 mM H2O2 induced morphological alterations of RBCs. In fact, in this condition we detected 37.1% of As depicted in Figure 3, incubation for 1 h at 25 • C with 20 mM H 2 O 2 induced morphological alterations of RBCs. In fact, in this condition we detected 37.1% of acanthocytes (RBCs with surface blebs) by scanning electron microscopy analysis. However, in samples pre-treated with 10 µM Q and then treated with 20 mM H 2 O 2 , the percentage of morphologically altered cells was reduced to 6.5%. Instead, in RBCs treated with 20 mM H 2 O 2 and then exposed to 10 µM Q we still detected 18.4% of acanthocytes. Therefore, the formation of acanthocytes was not completely avoided by Q post-treatment (Table 1).
However, in samples pre-treated with 10 μM Q and then treated with 20 mM H2O2, the percentage of morphologically altered cells was reduced to 6.5%. Instead, in RBCs treated with 20 mM H2O2 and then exposed to 10 μM Q we still detected 18.4% of acanthocytes. Therefore, the formation of acanthocytes was not completely avoided by Q post-treatment (Table 1).  The evaluation of ROS levels was carried out by flow cytometry in RBCs left untreated or, alternatively, exposed to H2O2 with or without pre-exposure to 10 µ M Q for 1 h or pre-treated with 20 mM H2O2 and then exposed to 10 µ M Q. Figure 4A shows the intracellular ROS levels at different time points (0, 15, 30, 45, 60 min) after incubation with H2O2. Samples exposed to 20 mM H2O2 showed a significant increase of ROS levels compared to the untreated samples. After a 30 min treatment, levels of ROS increased by 50% in H2O2-treated samples and remained unchanged over time. In Figure 4A, the effect of Q is also reported. In samples pre-or post-exposed to 10 µ M Q, 20 mM H2O2 failed to significantly increase ROS levels, which remained unchanged when compared to control values.  The evaluation of ROS levels was carried out by flow cytometry in RBCs left untreated or, alternatively, exposed to H 2 O 2 with or without pre-exposure to 10 µM Q for 1 h or pretreated with 20 mM H 2 O 2 and then exposed to 10 µM Q. Figure 4A shows the intracellular ROS levels at different time points (0, 15, 30, 45, 60 min) after incubation with H 2 O 2 . Samples exposed to 20 mM H 2 O 2 showed a significant increase of ROS levels compared to the untreated samples. After a 30 min treatment, levels of ROS increased by 50% in H 2 O 2treated samples and remained unchanged over time. In Figure 4A, the effect of Q is also reported. In samples pre-or post-exposed to 10 µM Q, 20 mM H 2 O 2 failed to significantly increase ROS levels, which remained unchanged when compared to control values.

Measurement of Thiobarbituric Acid Reactive Substances (TBARS) Levels
Thiobarbituric acid reactive substances (TBARS) measurement in RBCs is reported in Figure 4B. As expected, TBARS levels of RBCs treated with 20 mM H 2 O 2 for 1 h were significantly higher than those of RBCs left untreated (control). Importantly, in RBCs pre-treated with 10 µM Q and exposed to 20 mM H 2 O 2 or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q, TBARS levels were significantly reduced compared to those measured in 20 mM H 2 O 2 -treated RBCs. Of note, Q alone did not significantly affect TBARS levels.  Thiobarbituric acid reactive substances (TBARS) measurement in RBCs is reported in Figure 4B. As expected, TBARS levels of RBCs treated with 20 mM H2O2 for 1 h were with or without pre-exposure to 10 µM Q for 1 h at 37 • C or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q. ns, not statistically significant versus control; *** p < 0.001 versus control, one-way ANOVA followed by Bonferroni's post-hoc test (n = 5). (B) Detection of TBARS levels. TBARS levels (µM) in RBCs left untreated (control) or treated for 1 h at 25 • C with 20 mM H 2 O 2 with or without pre-incubation (1 h, 37 • C) with 10 µM Q or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q. ns, not statistically significant versus control; *** p < 0.001 versus control; ••• p < 0.001 versus 20 mM H 2 O 2 , one-way ANOVA followed by Bonferroni's post-hoc test (n = 10). (C) Sulfhydryl group content evaluation. Sulfhydryl group content (µM TNB/µg protein) in RBCs left untreated (control) and in RBCs treated for 1 h at 25 • C with 20 mM H 2 O 2 with or without pre-exposure (1 h, 25 • C) to 10 µM Q or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q. A 2 mM NEM was used as a positive control. ns, not statistically significant versus control; *** p < 0.001 versus control, ANOVA with Dunnet's post-test (n = 10). (D) Methemoglobin (% MetHb) content. Red blood cells were left untreated or incubated with 20 mM H 2 O 2 (1 h at 37 • C) with or without pre-exposure to 10 µM Q (1 h at 37 • C), or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q. NaNO 2 (2 mM for 1 h) was used as positive control. ns, not statistically significant; *,** p < 0.05 p < 0.01 versus control; •• p < 0.01 versus 20 mM H 2 O 2 , one way ANOVA followed by Bonferroni's post-hoc test (n = 10). Figure 4C shows the total content of sulfhydryl groups (µM TNB/µg protein) in RBCs left untreated or treated with either the oxidizing compound NEM (2 mM for 1 h at 25 • C, as the positive control), or 10 µM Q for 1 h, or 20 mM H 2 O 2 for 1 h with or without pretreatment with 10 µM Q, or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q. As expected, exposure to NEM led to a significant reduction in sulfhydryl group content. Sulfhydryl groups in 20 mM H 2 O 2 -treated RBCs were also significantly reduced in respect to control (left untreated). Pre-or post-treatment with 10 µM Q significantly restored the total levels of sulfhydryl groups in 20 mM H 2 O 2 -treated RBCs. Quercetin alone did not significantly affect total sulfhydryl group content. Figure 4D shows MetHb levels (% MetHb) measured in RBCs left untreated or treated with 20 mM H 2 O 2 (1 h at 25 • C) with or without pre-treatment with 10 µM Q (1 h at 37 • C), or pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q, or alternatively, treated with the well-known MetHb-forming agent NaNO 2 (2 mM for 1 h at 25 • C). The levels of MetHb measured after incubation with NaNO 2 were significantly higher than those detected in RBCs left untreated (control). In parallel, MetHb levels measured following exposure to 20 mM H 2 O 2 were significantly higher than those measured in control (left untreated). Pre-exposure to 10 µM Q significantly reduced the MetHb levels in H 2 O 2 -treated RBCs towards values that did not differ from control values. On the contrary, MetHb levels measured in RBCs pre-treated with 20 mM H 2 O 2 and then exposed to 10 µM Q were significantly increased with respect to those of RBCs left untreated (control), but did not differ with respect to those of H 2 O 2 -treated RBCs. Quercetin alone did not significantly affect the % MetHb.

CD47 Expression Level Determination
Flow cytometry analysis has shown a significantly decreased expression of CD47 in RBCs treated with 20 mM H 2 O 2 for 1 h compared to untreated (control) samples ( Figure 5A). Conversely, the expression of this protein was significantly restored by both pre-and posttreatment with 10 µM Q in H 2 O 2 -treated RBCs. Quercetin alone did not significantly affect CD47 expression. Data obtained by flow cytometry were confirmed by immunofluorescence analyses, which reported a remarkable rearrangement and redistribution of this protein ( Figure 5B).

Band 3 Protein Expression Level Determination
B3p protein expression was found significantly decreased in human RBCs treated with 20 mM H 2 O 2 for 1 h with respect to those left untreated (control) ( Figure 6A). Band 3 protein expression was significantly restored in RBCs both pre-and post-treated with 10 µM Q ( Figure 6A). Quercetin alone did not significantly affect B3p expression. In addition, a redistribution of B3p was detected by immunofluorescence. Specifically, B3p was mainly localized in blebs of acanthocytes after treatment with 20 mM H 2 O 2 , with respect to untreated RBCs ( Figure 6B). These changes were attenuated by Q. 4 2− Uptake via B3p Figure 7 reports the SO 4 2− uptake as a function of time in RBCs left untreated (control) and in RBCs treated with 20 mM H 2 O 2 (1 h at 25 • C) with or without Q (pre-or posttreatment, 1 h at 37 • C) . In control conditions, SO 4 2− uptake progressively increased and reached equilibrium within 45 min (rate constant of SO 4 2− uptake = 0.056 ± 0.001 min −1 ). Red blood cells treated with 10 µM Q alone (1 h at 37 • C) showed a rate constant of SO 4 2− uptake that was not significantly different with respect to control. Vice versa, the rate constant value (0.042 ± 0.001 min −1 ) in RBCs treated with 20 mM H 2 O 2 was significantly decreased with respect to control (*** p < 0.001). In RBCs pre-incubated with 10 µM Q (1 h at 37 • C) and then treated with 20 mM H 2 O 2 (1 h at 25 • C), the rate constant (0.062 ± 0.001 min −1 ) was significantly higher than that of RBCs treated with 20 mM H 2 O 2 (0.042 ± 0.001 min −1 ) and was not significantly different with respect to control (Table 2). Similarly, in RBCs incubated with 20 mM H 2 O 2 (1 h at 25 • C) and then exposed to 10 µM Q (1 h at 37 • C), the rate constant (0.047 ± 0.001 min −1 ) was significantly different than that of RBCs treated with 20 mM H 2 O 2 (0.042 ± 0.001 min −1 ) ( Table 2). SO 4 2− uptake was almost completely blocked by 10 µM DIDS applied at the beginning of incubation in SO 4 2− medium (0.017 ± 0.001 min −1 , *** p < 0.001, Table 2). Additionally, the SO 4 2− amount internalized by H 2 O 2 -treated RBCs after 45 min of incubation in SO 4 2− medium was significantly lower compared to control (Table 2), while it was not significantly different in RBCs pre-incubated with 10 µM Q and then exposed to 20 mM H 2 O 2 ( Table 2). Conversely, the SO 4 2− amount internalized by RBCs incubated with 20 mM H 2 O 2 and then treated with 10 µM Q was significantly lower compared to those left untreated (** p < 0.01, Table 1) and was not significantly different with respect to 20 mM H 2 O 2 . In DIDS-treated cells, the SO 4 2− amount internalized (5.49 ± 2.50) was significantly lower than that determined in both control and treated RBCs (*** p < 0.001, Table 2).

Band 3 Protein Expression Level Determination
B3p protein expression was found significantly decreased in human RBCs treated with 20 mM H2O2 for 1 h with respect to those left untreated (control) ( Figure 6A). Band 3 protein expression was significantly restored in RBCs both pre-and post-treated with 10 µ M Q ( Figure 6A). Quercetin alone did not significantly affect B3p expression. In addition,  Figure 7 reports the SO4 2− uptake as a function of time in RBCs left untreated (control) and in RBCs treated with 20 mM H2O2 (1 h at 25 °C) with or without Q (pre-or posttreatment, 1 h at 37 °C). In control conditions, SO4 2 ⁻ uptake progressively increased and reached equilibrium within 45 min (rate constant of SO4 2− uptake = 0.056 ± 0.001 min −1 ). Red blood cells treated with 10 µ M Q alone (1 h at 37 °C) showed a rate constant of SO4 2− uptake that was not significantly different with respect to control. Vice versa, the rate constant value (0.042 ± 0.001 min −1 ) in RBCs treated with 20 mM H2O2 was significantly decreased with respect to control (*** p < 0.001). In RBCs pre-incubated with 10 µ M Q (1 h at 37 °C) and then treated with 20 mM H2O2 (1 h at 25 °C), the rate constant (0.062 ± 0.001 min −1 ) was significantly higher than that of RBCs treated with 20 mM H2O2 (0.042 ± 0.001 min −1 ) and was not significantly different with respect to control (Table 2). Similarly, in RBCs incubated with 20 mM H2O2 (1 h at 25 °C) and then exposed to 10 µ M Q (1 h at 37 °C), the rate constant (0.047 ± 0.001 min −1 ) was significantly different than that of RBCs treated with 20 mM H2O2 (0.042 ± 0.001 min −1 ) ( Table 2). SO4 2 ⁻ uptake was almost completely blocked by 10 µ M DIDS applied at the beginning of incubation in SO4 2 ⁻ medium (0.017 ± 0.001 min −1 , *** p < 0.001, Table 2). Additionally, the SO4 2 ⁻ amount Conversely, the SO4 2 ⁻ amount internalized by RBCs incubated with 20 mM H2O2 and th treated with 10 µ M Q was significantly lower compared to those left untreated (** p < 0. Table 1) and was not significantly different with respect to 20 mM H2O2. In DIDS-treat cells, the SO4 2 ⁻ amount internalized (5.49 ± 2.50) was significantly lower than th determined in both control and treated RBCs (*** p < 0.001, Table 2). . Time course of SO4 2 ⁻ uptake. RBCs were left untreated (control) or treated with 20 m H2O2 with or without pre-exposure to 10 µM Q for 1 h at 37 °C, or pre-treated with 20 mM H2O2 a then exposed to 10 µ M Q, or exposed to 10 µ M DIDS. ns, not statistically significant versus contr *** p < 0.001 versus control; ** p < 0.01 versus control, one way ANOVA followed by Bonferron post-hoc test. and then exposed to 10 µM Q, or exposed to 10 µM DIDS. ns, not statistically significant versus control; *** p < 0.001 versus control; ** p < 0.01 versus control, one way ANOVA followed by Bonferroni's post-hoc test.

Discussion
Red blood cells are mainly known as transporters of metabolic gases and nutrients for tissues and perform additional important biological functions, such as the regulation of redox balance. Moreover, RBCs are deeply sensitive to external stressors and important health indicators [55]. During increase of OS levels, RBCs are exposed to circulating oxidative agents and their redox homeostasis is compromised due to the disequilibrium between pro-oxidant and anti-oxidant species, thus leading to a variety of structural alterations that promptly signal a physiological derangement [1,56]. In the last period, increasing evidence focused on beneficial effects of natural antioxidants and their ability to counteract OS-induced pathological conditions. In fact, dietary supplementation with fruits and vegetables, rich in phytochemicals, has been found to provide several health benefits. Quercetin is a naturally available flavonoid that showed different beneficial activities, including anti-oxidant properties, in different experimental models and therefore represents a good food supplement candidate [39,41,43,57,58].
In this investigation, the ameliorative effect of Q on H 2 O 2 -induced OS in RBCs was assessed. The results shown here indicate that Q could improve both structure and function of RBCs as well as potential pathological events that are closely related to oxidative damage. There are many studies describing the multiple activities of flavonoids, however, the effects of Q on human RBCs following H 2 O 2 treatment (20 mM) have not been evaluated. Firstly, this study tested the H 2 O 2 concentration to exclude a possible hemolytic power. As shown in Figure 1, 20 mM H 2 O 2 did not cause hemolytic events after 1 h incubation compared to control. Based on this finding, 20 mM H 2 O 2 was used to model oxidant conditions in RBCs.
Susceptibility of RBCs to H 2 O 2 exposure was studied in terms of morphological alterations by SEM. The images (Figure 3) show changes in the RBC shape, with the typical biconcave shape lost in a remarkable number of cells, which shows surface blebs, namely acanthocytes. However, pre-treatment with Q attenuated the morphological changes, with a reduction of acanthocytes numbers. Conversely, treatment with Q post-H 2 O 2 still revealed a cell morphology alteration, pointing to irreversible mechanisms underlying the changes in erythrocyte shape ( Table 1). The structure of the circulating cells has a crucial influence on the rheological properties of the blood and changes in morphology can lead to decreased deformability and increased aggregation [15,16,22]. To better clarify the mechanisms underlying these structural changes, some parameters related to OS assessment were monitored. Red blood cells were able to respond to any form of insult by changing their morphology following changes in their membrane or biochemical composition. Different phenomena, including oxidation of sulfhydryl groups of membrane proteins, oxidation of membrane fatty acid residues, or oxidation of hemoglobin, could alter membrane properties and cell shape, thus causing loss of membrane integrity and decreased deformability. Since oxidation of biological macromolecules, such as lipids and proteins, derives from the deleterious effects of ROS generated during cellular metabolism, intracellular ROS levels have been evaluated. Our findings showed that treatment with Q both pre-and postexposure to 20 mM H 2 O 2 induced a decrease of ROS levels ( Figure 4A). This finding is supported by other authors and suggests that dietary supplementation with flavonoids has an effective anti-oxidant activity, which is generally attributed to their ability to directly scavenge ROS [59].
Due to oxidation stress, polyunsaturated fatty acids of RBC membranes are damaged. This results in a steep increase in malondialdehyde (MDA) and thiobarbituric acid (TBA), which are biomarkers currently used to reveal oxidation of lipids under different conditions. Our results showed that exposure to 10 µM Q before or after OS avoided the lipid peroxidation of membranes induced by treatment with 20 mM H 2 O 2 ( Figure 4B). However, the ROS attack on lipids initiates a chain reaction, which leads to generation of more ROS that can harm other cellular components, including proteins. In this regard, RBCs represent a convenient model not only to investigate the degree of lipid peroxidation, but also oxidative damage on a protein level, being the protein component particularly abundant in these peculiar cells. Therefore, the sulfhydryl group content of total proteins was also evaluated. Exposure to Q before or after OS protected erythrocyte proteins from oxidative damage ( Figure 4C), which is in line with what has been previously demonstrated by other authors [60,61].
Elevation of OS has been associated with eryptosis. By analogy with apoptosis of nucleated cells, RBCs may undergo programmed cell death, which is characterized by cell shrinkage and cell membrane phospholipid scrambling. The eryptosis machinery includes activation of redox-sensitive Ca 2+ -permeable cation channels resulting in Ca 2+ entry. Cytosolic Ca 2+ elevation further activates erythrocyte scramblase and calpain resulting in phosphatidylserine (PS) externalization and membrane blebbing, respectively [62,63]. Phosphatidylserine-exposing RBCs are rapidly phagocytosed and, thus, cleared from circu-lating blood. However, no externalization of PS was found in RBCs treated with 20 mM H 2 O 2 , thus denoting that, in this model, RBCs do not enter apoptosis but remain in an early phase of the oxidation process, which is critically important to permit the action of antioxidants. The importance of this time window in the prevention of OS-related alterations has been very recently confirmed by our research group in a model of RBCs aging based on D-Galactose exposure. In this model, where PS scrambling did not occur OS-related derangements could be reversed by pre-treatment with an Açai berry extract, which is particularly rich in flavonoids [64].
Another major feature of RBC oxidation is the clustering and/or the breakdown of B3p and the binding of oxidized hemoglobin (MetHb) to high affinity sites on B3p. To better investigate the role of B3p, the expression of this protein during OS was also investigated. In particular, data showed that, following 20 mM H 2 O 2 treatment, B3p re-arranged in surface blebs ( Figure 6C). The N-terminal cytoplasmic domain of B3p contains binding sites for cytoskeletal and cytoplasmic proteins, including hemoglobin [65][66][67]. In RBCs, ROS production led to hemoglobin denaturation and further release of heme iron. This process can be autocatalytic, leading to an ever-increasing OS once it is initiated by the release of threshold amounts of free iron. Not surprisingly, healthy RBCs are equipped with multiple mechanisms to inactivate potent oxidants (e.g., MetHb). To better elucidate the molecular interaction between B3p and oxidized hemoglobin, MetHb levels were also evaluated. Our data indicated that exposure to 20 mM H 2 O 2 for 1 h increased the levels of MetHb in RBCs ( Figure 4D). These modifications can start a cascade of biochemical and structural transformations, including the release of microparticles containing hemicromes and clustering of B3p regions, as previously demonstrated by other groups [23,24,68]. When the oxidation processes are advanced, these clusters could provide a recognition site for antibodies directed against aging cells, thus triggering the premature removal of senescent RBCs from the circulation at the end of their 120-day life span. Interestingly, only pre-treatment with Q prevented H 2 O 2 -induced MetHb formation ( Figure 4C), which was still observed when Q was applied after OS.
One of the most interesting and still poorly investigated implications of OS is its impact on membrane transport systems. To define the possible effects of 20 mM H 2 O 2 on RBC functional activity, SO 4 2− uptake was measured by means of a validated method to assay anion exchange capability through B3p. In RBCs treated with 20 mM H 2 O 2 , the rate constant for SO 4 2− uptake was decreased compared to control ( Figure 7) and, in parallel, the amount of internalized SO 4 2− was significantly reduced (Table 2). However, 1 h pretreatment with 10 µM Q completely restored the rate constant of SO 4 2− uptake, as well as the amount of internalized SO 4 2− , thus demonstrating a beneficial effect of Q on B3p function. On the contrary, Q application after exposure to H 2 O 2 restored neither the rate constant of SO 4 2− uptake (Figure 7) nor the amount of SO 4 2− internalized ( Table 2). In this regard, we suggest that the mechanisms by which high levels of H 2 O 2 , or eventually other oxidants, induce changes in the RBC function are due to reactions taking place in the cell interior and involving oxidation of heme proteins, which may result in cross-linking with the cytoplasmic domain of B3p.
This, and our former work, reported that B3p exhibits changes in the rate constant for SO 4 2− uptake after exposure of RBCs to OS. In particular, non-hemolytic concentrations of H 2 O 2 induced OS and provoked a decrease in the rate constant for SO 4 2− uptake through B3p [53,54]. However, melatonin pre-treatment ameliorated the reduction in the rate constant for SO 4 2− uptake, as well as the reduction in B3p expression that was observed following treatment with H 2 O 2 [50]. Furthermore, the reduction of B3p anion exchange efficiency caused by a mild OS was prevented or attenuated by a short-time pre-incubation of RBCs with low H 2 O 2 doses. This pre-incubation encourages RBCs to adapt to a mild and transient OS and favors an increased tolerance to a successive stronger oxidant condition. Such adaptation response, termed pre-conditioning, was monitored by measuring B3p activity, did not involve B3p-related Tyr-phosphorylation pathways, but was mediated by an increased activity of catalase [52]. In summary, a reduction of the transport rate following H 2 O 2 exposure is most likely linked to the formation of oxidizing hemoglobin following OS. Therefore, the different effect on the SO 4 2− transport kinetics observed in a former study is not surprising. However, in different models of OS obtained by exposing RBCs to high concentrations of D-Glucose or D-Galactose, we measured an acceleration of the anion exchange through B3p instead of a reduction. It is tempting to speculate that such a two-sided effect on anion exchange velocity depends on the specific structure targeted by the stressors and on the possible underlying pathways [49,69,70].
At last, our focus has been addressed to assay CD47 protein expression, which is a specific marker of "self" [71]. In fact, expression of CD47 on RBCs' surface could be one of the mechanisms that regulate the removal of senescent cells from the bloodstream by phagocytosis. Treatment with 20 mM H 2 O 2 for 1 h did not induce a remarkable loss of CD47 protein, but rather its rearrangement and redistribution ( Figure 5C), which could be linked to the progressive plasma membrane blebbing and vesiculation, respectively ( Figure 3). In this model, CD47 expression and distribution were completely re-established in RBCs pre-treated with Q as well as RBCs treated with Q after exposure to H 2 O 2 . During their lifetime, RBCs release plasma membrane-derived ectosomes. These structures are generated by outward budding of the plasma membrane, followed by vesicle shedding [72,73]. Vesiculation is a way to remove dangerous molecules, such as oxidized proteins or denatured hemoglobin. Changes in lipids and proteins of RBC membranes induce a decrease in erythrocyte deformability and unfavorable changes in blood flow, which promote additional OS, proneness to atherosclerotic lesions, and increase in blood viscosity. As a result of ectosome formation, the protein composition of RBCs varies among circulating RBCs, with new RBCs being larger with a full equipment of membrane proteins and old RBCs smaller and denser with significantly lower content in membrane proteins, including CD47 and B3p [74]. In our model, the vesiculation process is missing, thus demonstrating that RBCs remain in an acute and early phase of the oxidation process.

Erythrocyte Preparation
This study was prospectively reviewed and approved by a duly constitute Ethics Committee (prot. 52-22, 20 April 2022). Upon informed consent, whole human blood samples from healthy volunteers were collected in test tubes containing ethylenediaminetetraacetic acid (EDTA). Plasma concentration of glycated hemoglobin (A1c) was less than 5%. Red blood cells were washed in isotonic solution (NaCl 150, 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid (HEPES) 5, Glucose 5, pH 7.4, osmotic pressure 300 mOsm/kgH 2 O) and centrifuged thrice (Neya 16R, 1200× g, 5 min) to delete plasma and buffy coat, respectively. Red blood cells were then suspended at specific hematocrits in isotonic solution and addressed to downstream analysis.

Analysis of Cell Shape by Scanning Electron Microscopy (SEM)
Samples, which were left untreated or exposed to 20 mM H 2 O 2 (1 h at 25 • C) with or without 10 µM Q (1 h at 37 • C, before or after treatment with H 2 O 2 ) were collected, plated on poly-l-lysine-coated slides, and fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) at 25 • C for 20 min. Then, samples were post-fixed with 1% OsO 4 in 0.1 M sodium cacodylate buffer and dehydrated through a graded series of ethanol solutions (from 30% to 100%). Absolute ethanol was gradually substituted by a 1:1 solution of hexamethyldisilazane (HMDS)/absolute ethanol and successively by pure HMDS. Afterwards, HMDS was completely removed, and samples were dried in a desiccator. Dried samples were mounted on stubs, coated with gold (10 nm), and analyzed by a Cambridge 360 scanning electron microscope (Leica Microsystem, Wetzlar, Germany) [76]. Altered erythrocyte shape was evaluated by counting ≥500 cells (50 RBCs for each different scanning electron microscopy (SEM) field at a magnification of 3000×) from samples in triplicate.

Total Sulfhydryl Group Content Determination
Measurement of total -SH groups was carried out according to the method of Aksenov and Markesbery [79], with some modifications. Red blood cells (35% hematocrit), left untreated or treated with to 20 mM H 2 O 2 (1 h at 25 • C) with or without 10 µM Q (1 h at 37 • C before or after treatment with H 2 O 2 ), were centrifuged (Neya 16R, 1200× g, 5 min) and a sample of 100 µL was hemolyzed in 1 mL of distilled water. A 50 µL aliquot was added to 1 mL of phosphate-buffered saline (PBS, pH 7.4) containing EDTA (1 mM). 5,5 -Dithiobis (2-nitrobenzoic acid) (DTNB, 10 mM, 30 µL) was added to start the reaction and the samples were incubated for 30 min at 25 • C protected from light. Control samples, without cell lysate or DTNB, were processed concurrently. After incubation, sample absorbance was measured at 412 nm (Onda spectrophotometer, UV-21) and 3-thio-2-nitro-benzoic acid (TNB) levels were detected after subtraction of blank absorbance (samples containing only DTNB). To achieve a complete oxidation of -SH groups, an aliquot of red blood cells (positive control) was incubated with 2 mM NEM for 1 h at 25 • C [53,80]. Obtained values were normalized to protein content and results reported as µM TNB/mg protein.

Analysis of Analytical Cytology
Red blood cells, which were left untreated or exposed to 20 mM H 2 O 2 (1 h at 25 • C) with or without 10 µM Q (1 h at 37 • C before or after treatment with H 2 O 2 ), were fixed with 3.7% formaldehyde in PBS (pH 7.4) for 10 min at 25 • C and then washed in the same buffer. Red blood cells were then permeabilized with 0.5% Triton X-100 in PBS for 5 min at 25 • C. After washing with PBS, samples were incubated with monoclonal anti-Band 3 protein (Sigma, St. Louis, MI, USA) or monoclonal anti-CD47 (Santa Cruz Biotechnology, Dallas, TX, USA) antibodies for 30 min at 37 • C, washed, and then incubated with a fluorescein isothiocyanate (FITC)-labeled anti-mouse antibody (Sigma) for 30 min at 37 • C [33]. Cells incubated with the secondary antibody given alone were used as negative control. Samples were analyzed by an Olympus BX51 Microphot fluorescence microscope or by a FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) equipped with a 488 nm argon laser. At least 20,000 events have been acquired. The median values of fluorescence intensity were used to provide a semiquantitative analysis. Fluorescence intensity values were normalized for those of untreated RBCs and expressed in %. 4 2− Uptake Measurement 4.7.1. Control Condition SO 4 2− uptake measurement was used to evaluate the anion exchange through B3p, as described elsewhere [51,[82][83][84]. In short, after washing, red blood cells were suspended to 3% hematocrit in 35 mL SO 4 2− medium (Na 2 SO 4 118, HEPES 10, glucose 5, pH 7.4, osmotic pressure 300 mOsm/kgH 2 O) and incubated at 25 • C in this medium. After 5, 10, 15, 30, 45, 60, 90, and 120 min, DIDS (10 µM), which is an inhibitor of B3p activity [85,86], was added to 5 mL sample aliquots, which were kept on ice. Afterwards, cells were washed three times in cold isotonic solution and centrifuged (Neya 16R, 4 • C, 1200× g, 5 min) to remove SO 4 2− from the external medium. Distilled water (1 mL) was added to induce osmotic lysis of cells and perchloric acid (4% v/v) was used to precipitate proteins. After centrifugation (Neya 16R, 4 • C, 2500× g, 10 min), the supernatant containing SO 4 2− trapped by cells was directed to the turbidimetric analysis. Supernatant (500 µL from each sample) was sequentially mixed to 500 µL glycerol diluted (1:1) in distilled water, 1 mL 4 M NaCl, and 500 µL 1.24 M BaCl 2 •2H 2 O. At last, the absorbance of each sample was measured at 425 nm (Onda Spectrophotometer, UV-21). The absorbance was converted to [SO 4 2− ] L cells × 10 −2 by means of a standard curve previously obtained by precipitating known SO 4 2− concentrations. The rate constant of SO 4 2− uptake (min −1 ) was derived from the following equation: C t = C ∞ (1 − e −rt ) + C 0 , where C t , C ∞ , and C 0 indicate the intracellular SO 4 2− concentrations measured at time t, ∞, and 0, respectively, e represents the Neper number (2.7182818), r indicates the rate constant accounting for the process velocity, and t is the specific time at which the SO 4 2− concentration was measured. The rate constant is the inverse of the time needed to reach~63% of total SO 4 2− intracellular concentration [82] and [SO 4 2− ] L cells × 10 −2 reported in figures represents SO 4 2− micromolar concentration internalized by 5 mL RBCs suspended at 3% hematocrit.

Experimental Conditions
Red blood cells (3% hematocrit), which were left untreated or treated with 20 mM H 2 O 2 (1 h at 25 • C) with or without 10 µM Q (1 h at 37 • C before or after treatment with H 2 O 2 ), were centrifuged (Neya 16R, 4 • C, 1200× g, 5 min) to replace the supernatant with SO 4 2− medium. The rate constant of SO 4 2− uptake was determined as described for the control condition.

Experimental Data and Statistics
All data are expressed as arithmetic mean ± standard error of the mean. For statistical analysis and graphics, GraphPad Prism (version 8.0, GraphPad Software, San Diego, CA, USA) and Excel (Version 2019, Microsoft, Redmond, WA, USA) software were used. Data normality was verified with the D'Agostino and Pearson Omnibus normality test. Significant differences between mean values were determined by one-way analysis of variance (ANOVA), followed by Student's t-test, Bonferroni's multiple comparison posttest, or ANOVA with Dunnet's post-test, as appropriate. Statistically significant differences were assumed at p < 0.05; (n) corresponds to the number of separate measurements.

Conclusions
We conclude that exposure to 20 mM H 2 O 2 induced OS in RBCs, which was manifested as morphological changes, as well as alteration of B3p and CD47 protein surface expression. Pre-treatment with Q avoided the formation of acanthocytes observed after exposure to 20 mM H 2 O 2 , prevented H 2 O 2 -induced OS damage, including ROS production, lipid peroxidation as well as protein oxidation, and restored the distribution of B3p and CD47 expression on the plasma membrane. Moreover, H 2 O 2 exposure was associated with a reduction of the rate constant of SO 4 2− uptake through B3p, as well as MetHb formation. Both alterations have been attenuated by pre-treatment with Q. Conversely, Q post-treatment prevented H 2 O 2 -induced OS damage, including ROS production, lipid peroxidation as well as protein oxidation, restored the distribution of B3p and CD47 on the plasma membrane, but only partially avoided the formation of acanthocytes. Importantly, post-treatment with Q did not restore the rate constant of SO 4 2− uptake through B3p and did not inhibit the MetHb formation. These findings reveal that some, but not all, OSrelated alterations are reversible, and that early application of antioxidants is envisioned to counteract the majority of OS-induced derangements. The present investigation provides mechanistic insights into the higher number of benefits deriving from the use of naturally occurring flavonoids against OS on a cellular level. The results obtained here confirm that measurement of B3p anion exchange capability remains a suitable tool for monitoring the impact of OS on RBC homeostasis. Considering the involvement of OS in a wide range of pathologies, new OS biomarkers, with both diagnostic and monitoring potential, are needed. Blood can be obtained from patients with minimally invasive procedures, reflects the physiological status of peripheral tissues, and therefore might represent a convenient source of OS biomarkers. We propose monitoring of B3p expression and function as a novel OS biomarker. In this light, future investigations are needed to clarify the signaling underlying the protective activity of Q on normal anion exchange ability, including possible effects on the interaction between B3p and the cytoskeletal proteins ankyrin and spectrin, their potential post-translation modifications, as well as a possible influence on the endogenous antioxidant machinery. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.

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