Redox Regulation of Phosphatase and Tensin Homolog by Bicarbonate and Hydrogen Peroxide: Implication of Peroxymonocarbonate in Cell Signaling

Abstract

Phosphatase and tensin homolog (PTEN) is a negative regulator of the phosphoinositide 3-kinases/protein kinase B (PI3K/AKT) signaling pathway. Notably, its active site contains a cysteine residue that is susceptible to oxidation by hydrogen peroxide (H₂O₂). This oxidation inhibits the phosphatase function of PTEN, critically contributing to the activation of the PI3K/AKT pathway. Upon the stimulation of cell surface receptors, the activity of NADPH oxidase (NOX) generates a transient amount of H₂O₂, serving as a mediator in this pathway by oxidizing PTEN. The mechanism underlying this oxidation, occurring despite the presence of highly efficient and abundant cellular oxidant-protecting and reducing systems, continues to pose a perplexing conundrum. Here, we demonstrate that the presence of bicarbonate (HCO₃⁻) promoted the rate of H₂O₂-mediated PTEN oxidation, probably through the formation of peroxymonocarbonate (HCO₄⁻), and consequently potentiated the phosphorylation of AKT. Acetazolamide (ATZ), a carbonic anhydrase (CA) inhibitor, was shown to diminish the oxidation of PTEN. Thus, CA can also be considered as a modulator in this context. In essence, our findings consolidate the crucial role of HCO₃⁻ in the redox regulation of PTEN by H₂O₂, leading to the presumption that HCO₄⁻ is a signaling molecule during cellular physiological processes.

1. Introduction

When cells are stimulated by growth factors and cytokines, such as platelet-derived growth factor (PDGF), epidermal growth factor (EGF), insulin, granulocyte–macrophage colony-stimulating factor (GM-CSF), tumor necrosis factor-α (TNF-α), interleukin-1 (IL-1), and interleukin-3 (IL-3), H₂O₂ is produced through the activity of NOXs [1]. The activation of NOXs results in the production of superoxide (O₂^•−), which is subsequently transformed to H₂O₂ by the activity of superoxide dismutase (SOD) [2]. This physiological H₂O₂ influences various intracellular signaling pathways. Its capacity to oxidize the cysteine residues of some proteins, such as protein tyrosine phosphatases (PTPs), results in a functional modification that notably impairs their activities [3]. PTPs play pivotal roles in cellular processes involving cell growth, proliferation, and differentiation [4]. Within their structural framework, all PTPs feature a cysteine residue in their active site [5]. When exposed to H₂O₂, the cysteine residue undergoes oxidation, transforming into cysteine–sulfenic acid (Cys-SOH) and leading to the oxidative inhibition of PTP’s enzymatic activity [6].

PTEN, a member of the PTP family, structurally comprises five functional domains: a short N-terminal phosphatidylinositol (PtdIns)(4,5)P2-binding domain (PBD); a catalytic phosphatase domain; a C2 lipid/membrane-binding domain; a C-terminal tail containing Pro, Glu, Ser, and Thr (PEST) sequences; and a class I PDZ-binding (PDZ-BD) motif. PTEN interacts with phospholipid membranes through the C2 domain located at the C-terminal end. The tight linkage between the C2 and phosphatase domains implies that the C2 domain not only aids in recruiting PTEN to the membrane but also optimizes the orientation of the catalytic domain associated with its membrane-bound substrate. Moreover, the phosphatase domain of PTEN has an enlarged active site to accommodate its phosphoinositide substrate, membrane-bound phosphatidylinositol-3,4,5-triphosphate (PIP3) [7,8]. Hence, PTEN is considered a tumor suppressor owing to its ability to dephosphorylate PIP3 and, consequently, negatively regulate the PI3K/AKT signaling pathway, which is pivotal in controlling cell survival, growth, and proliferation [9,10]. In addition to cancer studies, substantial insights into PTEN’s role in physiological processes are available. Notably, PTEN inhibition has been demonstrated as a promising therapeutic intervention for neurodegenerative diseases, ischemia, infection, and insulin-resistant metabolic disorders [11,12,13].

Similar to other members of the PTP family, PTEN contains a cysteine residue in the active site of its phosphatase domain, rendering it susceptible to oxidative inhibition by reactive oxygen species (ROS), particularly H₂O₂ [5]. Lee et al. were the first to demonstrate the reversible inactivation of PTEN by H₂O₂ through the oxidation of the Cys124 catalytic residue in the active site and the formation of an intramolecular disulfide bond with Cys71 residue. This inactivation is reversible because oxidized PTEN can be converted back to the functional reduced form by cellular reducing agents, such as the Thioredoxin/Thioredoxin Reductase (Trx/TrxR) system [14]. Additionally, in the cell, Peroxiredoxins (Prx), thiol-specific antioxidants of the peroxidase family, can function as modulators of H₂O₂-induced phosphorylation signaling due to their capacity to sense and eliminate peroxides [15,16,17]. Thus, the mechanism of PTEN oxidation by transient H₂O₂, a signaling molecule produced in response to growth factor receptor stimulation, raises questions about how this small amount of physiological H₂O₂ can function to oxidize PTEN in a cellular environment rich in H₂O₂ scavengers such as catalase, glutathione peroxidase, and Prx, alongside the presence of ubiquitous Trx/TrxR reducing systems.

It has been known since the 1980s that H₂O₂ can react with bicarbonate/carbon dioxide (HCO₃⁻/CO₂) to form HCO₄⁻: H₂O₂ + HCO₃⁻/CO₂ ⇌ HCO₄⁻ + H₂O/H⁺ [18,19]. In a neutral-pH aqueous solution, the HCO₄⁻ formation process occurs rapidly at 25 °C with a half-life t_1/2 of 10 min or below [20,21]. CAs and a zinc model complex can accelerate this reaction [19,21]. HCO₄⁻ is a robust oxidant with a higher catalytic potency comparable to that of H₂O₂. Its interaction with target molecules proceeds at velocities ranging from 100 to 1000 times faster than those observed with H₂O₂ [22]. Sulfide oxidation caused by HCO₄⁻ transcends that caused by H₂O₂ by approximately 300-fold [20]. Notably, HCO₄⁻ also serves as a potent two-electron oxidant primarily accountable for biothiol peroxidation [23]. Hence, the presence of HCO₃⁻ is demonstrated to be a pivotal factor in promoting the oxidative reactivity of H₂O₂ towards PTPs, such as PTP1B, SHP-2, and PTPN22, during signal transduction processes [24,25,26]. When cells are stimulated by growth factors, HCO₃⁻ is produced through the activation of a transmembrane enzyme, CA IX, which catalyzes the hydration of CO₂ (CO₂ + H₂O → HCO₃⁻ + H⁺) [27]. Dagnell et al. demonstrated that HCO₃⁻ not only facilitates but also plays an essential role in the H₂O₂-mediated inactivation of PTP1B, surpassing the protection provided by the Trx/TrxR reducing systems [25]. Given that PTEN belongs to the PTP family, it is worth exploring whether HCO₃⁻ may somehow affect the H₂O₂-mediated oxidative inactivation of PTEN. Based on the available pieces of information, we anticipated that the presence of HCO₃⁻ in combination with H₂O₂, through the formation of HCO₄⁻, would augment the oxidation of PTEN.

2. Materials and Methods

2.1. Materials

Dulbecco’s modified Eagle’s medium (DMEM) D5648 and N-Ethylmaleimide (NEM) were sourced from Sigma-Aldrich (St. Louis, MO, USA), 1 M HEPES from Enzynomics (Daejeon, Republic of Korea), 100X penicillin/streptomycin from Capricorn Scientific (Ebsdorfergrund, Germany), fetal bovine serum (FBS) from Welgene (Gyeongsan, Republic of Korea), 3% hydrogen peroxide from Samchun (Seoul, Republic of Korea), sodium bicarbonate from Amresco (Solon, OH, USA), Pro-prep Protein Extraction Solution from iNtRON Biotechnology (Seongnam, Republic of Korea), a PageRuler Prestained Protein Ladder No. 26616 and BCA Protein Assay Kit from Thermo Scientific (Waltham, MA, USA), and ATZ from MedChemExpress (New York City, NJ, USA). The antibodies used for Western blotting comprised primary PTEN antibody, β-actin, phosphorylated-Serine-473 AKT, total AKT and anti-rabbit immunoglobulin G horseradish peroxidase-linked secondary antibodies. All other chemicals and reagents were of analytical grade.

2.2. Cell Culture

HepG2 cells, originally obtained from the American Type Culture Collection (Manassas, VA, USA), were cultured in regular DMEM supplemented with 10% FBS and 1X penicillin/streptomycin.

2.3. Immunoblot Analysis of H₂O₂-Induced Oxidation of PTEN

To assess the redox state of PTEN, we performed mobility shift assays in cells, employing N-Ethylmaleimide (NEM) as an alkylating agent, as described in our previous publication [28]. HepG2 cells were cultured until they reached 90% confluency. Subsequently, they were washed with PBS and switched to media containing 0.1% FBS and 25 mM HEPES, with various sodium bicarbonate or ATZ conditions, and then pre-incubated at 37 °C with 0.1% CO₂. Similar serum-free stimulation media containing 0.5–1 mM H₂O₂ were prepared before being applied to the cells. After pre-incubation for 4 h, the media were removed, and cells were incubated with the prepared stimulation media. Subsequently, the stimulation media were removed at varying time points, and the cells were washed twice using cold PBS. The reaction was stopped by adding Pro-prep lysis buffer containing 10 mM NEM. Cell lysates were collected, and the protein concentrations were quantified using the BCA method. Subsequently, the samples were subjected to non-reducing SDS-PAGE, followed by Western blotting using PTEN, β-Actin, pAKT, and total AKT-specific antibodies (Figure 1).

2.4. Statistical Analysis

PTEN oxidation levels in the samples were quantified by analyzing the oxidized/reduced PTEN states using ImageJ version 1.54d. Data significances were evaluated using Student’s t test. Differences were significant when p < 0.05. Graphical values are presented as the mean ± standard error (SE). The trendlines for PTEN oxidation and AKT phosphorylation represent linear regression.

3. Results

3.1. The Presence of HCO₃⁻ Potentiates the PTEN Oxidation by H₂O₂

HepG2 cells were pre-incubated in HCO₃⁻-free media at 37 °C with 0.1% CO₂ for 4 h. Various stimulation media were made before being administered to the cells: 44 mM HCO₃⁻ alone, 1 mM H₂O₂ alone, a combination of 1 mM H₂O₂ and 22 mM HCO₃⁻, and a combination of 1 mM H₂O₂ and 44 mM HCO₃⁻. The percentages of oxidized PTEN were assessed after 10 min of treatment. The results showed that with the presence of 22 mM or 44 mM HCO₃⁻ in the stimulation media, the oxidations of PTEN in 10 min were significantly higher (p < 0.05) compared to the cells which were treated with H₂O₂ alone. The oxidation did not exhibit an increased magnitude when more HCO₃⁻ was added. In addition, stimulation media with only 44 mM HCO₃⁻ exhibited no oxidation effect (Figure 2).

3.2. The Presence of HCO₃⁻ Facilitates Redox Regulation of PTEN by H₂O₂

To investigate the change in PTEN oxidation, we observed the percentages of oxidized PTEN in different time courses. Treatments of HepG2 cells by H₂O₂, with the presence and absence of 44 mM HCO₃⁻, in 5, 10, 15, 30 60, and 120 min, resulted in PTEN oxidation in various time-dependent manners. In the first group, cells were pre-incubated and treated in stimulation media containing 44 mM HCO₃⁻. In the second group, the condition was similar except that the media was HCO₃⁻-free. Our data indicated that in the HCO₃⁻-containing group, the PTEN oxidation rate soared to its peak at 10 min and then decreased gradually. At 60 min, almost all oxidized PTENs were reduced back, and the recovery was completed by 120 min. In the HCO₃⁻-free group, the oxidation rate increased at a slower speed, reaching its peak at 30 min. At 60 min, there was a considerable amount of oxidized PTEN that was not reduced back, and the amount was much higher than that observed in the HCO₃⁻-containing group. In addition, the maximum oxidation level was higher in the HCO₃⁻-containing group. The linear regression trendlines illustrate that PTEN oxidation was faster in the HCO₃⁻-containing group than in the HCO₃⁻-free group. Comparing the slopes between the trendlines (2.87 and 1.16), the difference in oxidation speed was nearly 2.5 times. During the recovery period, the absence of HCO₃⁻ remarkably impaired the reduction of oxidized PTEN. A comparison of the slopes revealed that the difference in reduction speed was also approximately equivalent (2.3-fold). These results clearly demonstrate that HCO₃⁻ accelerates the H₂O₂-mediated redox regulation of PTEN (Figure 3).

3.3. The Activation of PI3K/AKT Pathway via PTEN Oxidation by H₂O₂ and HCO₃⁻

PTEN oxidation inhibits its phosphatase function and subsequently leads to the activation of the PI3K/AKT signaling pathway [10]. We have shown that PTEN can be oxidized at a faster speed by the combination of HCO₃⁻ and H₂O₂ than by H₂O₂ alone. To investigate how this PTEN oxidative inhibition affects the phosphorylation of AKT, we conducted the same protocol; cell lysates were subjected to SDS-PAGE with antibodies specific for pAKT and total AKT. With the addition of 44 mM HCO₃⁻, AKT phosphorylation started to rise significantly following 10 min of exposure to H₂O₂, reaching the maximum value at 15 min before gradually declining. Meanwhile, the absence of HCO₃⁻ delayed this process, with a slower increase to its peak at 30 min. Furthermore, the HCO₃⁻-containing group also exhibited a higher proportion of pAKT across all timepoints. Comparing the slopes between the trendlines (0.0704 and 0.0263), the disparity in phosphorylation velocity was approximately 2.7-fold. Overall, under stimulation by H₂O₂, the presence of HCO₃⁻ potentiates the phosphorylation of AKT in a time-dependent manner (Figure 4).

3.4. Oxidation of PTEN Is Decreased in the Presence of CA Inhibitor ATZ

CAs catalyze the following reaction: CO₂ + H₂O ⇌ HCO₃⁻ + H⁺ [29]. We utilized a CA inhibitor, ATZ, to further investigate PTEN oxidation by H₂O₂ in condition with 0.1% CO₂ and lacking HCO₃⁻. Cells were pre-incubated for 4 h with 0.1% CO₂ in HCO₃⁻-free media and supplemented with various concentrations of ATZ such as 0, 0.01, 0.1, 1, and 10 µM. After the pre-incubation period, cells were exposed to 0.5 mM H₂O₂ in stimulation media under the same condition for 10 min. The results showed 10 µM of ATZ alone did not cause any PTEN oxidation. The presence of 1 or 10 µM of ATZ in combination with 0.5 mM H₂O₂ was significantly effective compared to H₂O₂ alone (p < 0.05). As the ATZ concentration increased, the PTEN oxidation decreased in a dose-dependent manner. At 10 µM, the percentage of oxidized PTEN was considerably lower than when ATZ was absent (Figure 5).

4. Discussion

An elevated PI3K/AKT signaling pathway has been shown to be advantageous in specific physiological processes relating to tissue regeneration; thus, inhibiting PTEN, a suppressor of this pathway, has emerged as a potential therapeutic approach for ischemia, neurodegenerative diseases, infection, and insulin-resistant metabolic disorders [11,12,13]. In addition, the activation of receptor tyrosine kinases (RTKs) is a pivotal event in transmitting phosphorylation signals upon stimulation by growth factors, therefore holding substantial importance in cell physiology [30]. In this situation, a transient amount of H₂O₂ is generated by membrane NOXs [31]. Subsequently, H₂O₂ causes the reversible oxidative inhibition of PTEN, leading to an increase in PI3K/AKT downstream cascades and, consequently, eliciting cellular effects [32,33]. The oxidative inactivation of PTEN by ROS has also been demonstrated to play a significant positive role in those physiological processes demanding cell growth [34]. It can promote cardiac remodeling [35], neuronal regeneration [36], immune response [37], glucose metabolism [38], and myogenesis [39]. However, within the cellular environment, H₂O₂ can be eliminated by thiol proteins from the Prx family. Furthermore, oxidized PTEN and Prx can be rapidly converted back to their active reduced forms via the Trx/TrxR/NADPH system [40,41]. Thus, PTEN phosphatase catalytic activity is conserved by Trx/TrxR and Prx systems [16,17,40,42]. Yet, in this context, PTEN is inhibited during PI3K/AKT signaling pathway activation. Additionally, under RTK stimulation, CA IX undergoes phosphorylation and engages in catalyzing the conversion of CO₂ to HCO₃⁻ [27]. Through our experiment results, we presume the factor that facilitates PTEN oxidation by H₂O₂ in the presence of HCO₃⁻ is the formation of HCO₄⁻ (Figure 6).

The addition of 44 mM sodium bicarbonate to DMEM D5648, which lacks HCO₃⁻, replicates the common condition of culture media. Sodium bicarbonate can alter the pH of the tumor microenvironment, which has been demonstrated to critically affect cancer cell apoptosis and survival [43]. Therefore, we supplemented the media with HEPES as a pH buffer. Previous investigations on NIH 3T3 cells exposed to H₂O₂ in standard DMEM revealed reversible PTEN oxidation in a time-dependent manner. Incubation with 0.5 mM H₂O₂ resulted in increased levels of oxidized PTEN, peaking at 10 min before declining gradually, suggesting that PTEN was initially inactivated by H₂O₂ and then reactivated by cellular reductants as the H₂O₂ declined. This reversible PTEN inactivation was also observed in HeLa cells exposed to H₂O₂ [14]. In our recent studies, cells were treated with H₂O₂ for specific durations, extending the time points to 120 min to ensure the completion of the reduction process. Similarly, oxidized PTEN exhibited a time-dependent increase followed by a decline. The PTEN oxidation rate reached its peak after exposure for 10 min, and the oxidized PTEN was completely converted to the reduced form within 120 min of exposure. Based on these results, it is evident that treatments with H₂O₂ in standard HCO₃⁻-containing media yield maximum PTEN oxidation after 10 min of exposure, followed by the conversion of oxidized PTEN to its reduced state by the cellular Trx/TrxR system within approximately 60 min after exposure [28,42]. Therefore, we initially tested the role of HCO₃⁻ in PTEN oxidation by H₂O₂ through evaluating the proportions of oxidized PTEN following 10 min of exposure to stimulation media. The pre-incubation period aimed to diminish the HCO₃⁻ in both the extracellular and intracellular environments. Our findings indicated HCO₃⁻ alone did not cause any PTEN oxidation. Meanwhile, H₂O₂ alone had a significantly lower percentage than the H₂O₂-HCO₃⁻ combination. Additionally, there was no notable difference between the HCO₃⁻ concentration of 22 mM and 44 mM. This implies that the greater percentages observed in the H₂O₂-HCO₃⁻ combination groups were not merely the cumulative effect of both H₂O₂ and HCO₃⁻_. A plausible explanation is that HCO₃⁻ can empower the catalytic activity of H₂O₂, supporting the hypothesis that there might be the formation of HCO₄⁻, a more reactive oxidant.

Given that HCO₄⁻ was demonstrated to exhibit a greater oxidation rate towards sulfide when compared to H₂O₂ [20], we suppose that it may also influence PTEN oxidation. This implies that the peak of PTEN oxidation in each group may not be solely at the 10 min timepoint. Therefore, we extended experiments to additional time courses to investigate the redox regulation of HCO₃⁻ and H₂O₂. Additionally, we reduced the concentration of H₂O₂ to 0.5 mM and anticipated it could make the difference more visible. A similar pattern to results of our previous studies was observed this time on HepG2 cells in 44 mM HCO₃⁻-supplemented media. PTEN was predominantly oxidized within 5–15 min, with the highest oxidation rate being observed at 10 min, then reduced back to the active form. Thus, the PTEN oxidation observed in the HCO₃⁻-containing group in the present study follows a trend consistent with our former studies.

The exclusion of HCO₃⁻ from the experimental conditions considerably prolonged the PTEN oxidation period, requiring up to 30 min to reach its peak. However, the H₂O₂-mediated oxidative inactivation of intracellular PTPs in signaling events typically happens swiftly within the timeframe of 5–15 min, typically peaking at 10 min [14,44,45]. This disparity suggests that the absence of HCO₃⁻ adversely affects the PTEN oxidation speed. Our result aligns with the findings of Dagnell et al., indicating that HCO₃⁻ can potentiate the H₂O₂-dependent oxidative inactivation of PTP [25]. Illustrative data analysis suggests that the combination of HCO₃⁻ and H₂O₂ accelerated the rate of PTEN oxidation in the cellular environment by approximately 2.5-fold in a time-dependent manner. Notably, this was also strongly correlated to the similar decrease in the PTEN reduction speed in the HCO₃⁻-free group. Based on growing evidence about the role of HCO₃⁻ in H₂O₂-mediated oxidation under cell signaling processes, we suppose the formed HCO₄⁻ promotes PTEN redox regulation and therefore exerts function as a signaling molecule.

PTEN acts as a negative regulator of the PI3K/AKT signaling pathway, which induces various cellular processes including protein synthesis, cell survival, proliferation, and migration. This mechanism plays an important role in human physiological and pathological conditions [7,12,46]. Following the result that H₂O₂-mediated oxidative inhibition of PTEN is potentiated and accelerated by HCO₃⁻, we examined its subsequent impact on the phosphorylation of AKT. The elevating trends of pAKT/AKT reflect similar patterns observed in PTEN oxidation, both in HCO₃⁻-containing and HCO₃⁻-free status. Our findings are consistent with previous evidence suggesting that H₂O₂ can activate the PI3K/AKT pathway by oxidizing PTEN [10]. Additionally, the time-dependent increase in pAKT/AKT is equivalent to the rise in PTEN oxidation in the presence of HCO₃⁻, which is 2.7-fold and 2.5-fold, correspondingly. In another words, without HCO₃⁻, PTEN oxidation is impaired, leading to the delay in AKT phosphorylation. Thus, in this context, the increase in AKT activation arises from the surge in PTEN oxidation. Considering HCO₄⁻ is formed through the reaction between H₂O₂ and HCO₃⁻, we propose that HCO₄⁻ can positively regulate the PI3K/AKT signaling pathway via the oxidative inactivation of PTEN.

However, we also concede a possibility that a minor amount of HCO₃⁻ might be produced in the media through the reaction of CO₂ and H₂O, despite the CO₂ level being minimized to 0.1% in our experiments. The CO₂ hydration reaction to form HCO₃⁻ is reversible and catalyzed by CAs [29]. Notably, CA IX, a transmembrane isoform, is often over-expressed in various types of cancer cells as it is associated with hypoxia-inducible factor 1 (HIF1) [47]. Thus, we had expected utilizing ATZ, a CA inhibitor, to prevent the production of HCO₃⁻. Our prior data indicated that PTEN oxidation by H₂O₂ in HCO₃⁻-free media was significantly lower than that in HCO₃⁻-containing media. However, when 0.01–10 mM ATZ was added, the oxidation decreased further in a dose-dependent manner. Given that no oxidized PTEN bands were detected in the presence of 10 mM ATZ alone, we acknowledged that ATZ did not directly cause PTEN oxidation in these conditions. A possible explanation could be that a higher dose of ATZ more effectively impedes the conversion of CO₂ to HCO₃⁻. Consequently, the formation of HCO₄⁻ from HCO₃⁻ and H₂O₂ was also reduced. This supports our presumption that HCO₄⁻ plays a critical role in PTEN oxidation under stimulation by H₂O₂.

A limitation of our study is the lack of a direct evaluation regarding the actual concentration of formed HCO₄⁻ when combining H₂O₂ and HCO₃⁻. We could only adjust indirectly through conducting different conditions of CO₂, HCO₃⁻, and CA inhibitor.

In summary, the results of our experiment support the hypothesis that the presence of HCO₃⁻ promotes the H₂O₂-mediated inactivation of PTEN. This subsequently promotes the activation of the PI3K/AKT signaling pathway. In addition, the reduction of oxidized PTEN by antioxidant systems is also equivalently affected. A plausible explanation could be the formation of HCO₄⁻. CAs are prospective targets for modulating this process as they are involved in HCO₃⁻ production. In the future, further experiments would be conducted to consolidate the role of HCO₄⁻ in the redox regulation of PTEN.