Redox Regulation of PTEN by Bicarbonate and Hydrogen Peroxide: Implication of Peroxymonocarbonate

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), hydrogen peroxide (H₂O₂) is produced through the activity of NADPH oxidases (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].

Phosphatase and tensin homolog (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]. Besides 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].

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 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. 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 [12]. 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 [13,14,15]. 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 1980s that H₂O₂ can react with bicarbonate/carbon dioxide (HCO₃^-/CO₂) to form peroxymonocarbonate (HCO₄^-)^-: H₂O₂ + HCO₃^-/CO₂ ⇌ HCO₄^- + H₂O/H⁺. In neutral pH aqueous solution, HCO₄^- formation process occurs rapidly at 25 ℃ with half-life t_1/2 of 10 min or below. Carbonic anhydrase and a Zinc model complex can accelerate this reaction [16,17]. 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 [18]. Sulfide oxidation caused by HCO₄^- surpasses that caused by H₂O₂ by approximately 300-fold [19]. Notably, HCO₄^- also serves as a potent two-electron oxidant primarily accountable for biothiol peroxidation [20]. 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 [21,22,23]. When cells are stimulated by growth factors, HCO₃^- is produced through the activation of a transmembrane enzyme, carbonic anhydrase IX, which catalyzes the hydration of CO₂ (CO₂ + H₂O → HCO₃^- + H⁺) [24]. Dagnell et al. demonstrated that HCO₃^- not only facilitates but also plays an essential role in H₂O₂-mediated inactivation of PTP1B, surpassing the protection provided by the Trx/TrxR reducing systems [22]. 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 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, 1 M HEPES from Enzynomics, 100X Penicillin/Streptomycin from Capricorn Scientific, fetal bovine serum (FBS) from Welgene, 3% hydrogen peroxide from Samchun, sodium bicarbonate from Amresco, Pro-prep Protein Extraction Solution from iNtRON Biotechnology, PageRuler Prestained Protein Ladder No. 26616 and BCA Protein Assay Kit from Thermo Scientific. The antibodies used for western blotting comprised primary PTEN antibody, β-actin antibody, 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 were cultured in DMEM supplemented with 5% 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 [25]. 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 or without sodium bicarbonate, and then pre-incubated at 37 °C with 0.1% CO₂. Serum-free stimulation media containing 0.5-1 mM H₂O₂, with various bicarbonate conditions, were prepared before applying 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. Data significances were evaluated using unpair 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 represent linear regression.

3. Results

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 administering 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 treatment. The result showed that with the presence of 22 mM or 44 mM bicarbonate 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 increase magnitude when adding more HCO₃^-. In addition, stimulation media with only 44 mM HCO₃^- exhibited no oxidation effect. (Figure 2)

The Presence of HCO₃^- Accelerates the Redox Regulation of PTEN by H₂O₂

To investigate the change in PTEN oxidation, we observed the percentages of oxidized PTEN in different time period. 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 varying 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 PTEN 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)

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 PI3K/AKT signaling pathway [10]. We have shown that PTEN can be oxidized at faster speed by the combination of HCO₃^- and H₂O₂ than by H₂O₂ alone. To investigate how this PTEN oxidative inhibition affect 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 exposure to H₂O₂, reaching maximum 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 stimulating by H₂O₂, the presence of HCO₃^- potentiates the phosphorylation of AKT in a time-dependent manner. (Figure 4)

4. Discussion

The activation of receptor tyrosine kinases (RTK) is a pivotal event in transmitting phosphorylation signals upon stimulation by growth factors and, therefore, holds substantial importance in cell physiology [26]. In this situation, a transient amount of H₂O₂ is generated by membrane NOXs [27]. Subsequently, H₂O₂ causes reversible oxidative inhibition of PTEN, leading to an increase in PI3K/AKT signaling pathway and, consequently, eliciting cellular substantial effects [28,29]. The activation of AKT and its downstream cascades are proved to be related to variety of cancers [30]. Besides, studies also indicate that the elevated AKT activity through oxidative inactivation of PTEN can yield benefits in particular physiological processes that require cell growth such as cardiac remodeling following ischemia [31], neuronal regeneration [32], immune response [33], insulin-related metabolism [34], and myogenesis [35]. 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 [36,37]. Thus, PTEN phosphatase catalytic activity is conserved by Trx/TrxR and Prx systems [14,15,36,38]. However, PTEN is inhibited during PI3K/AKT signaling pathway activation. Through our experiment result, we presume the factor that facilitates PTEN oxidation by H₂O₂ in the presence of HCO₃^- is the formation of HCO₄^- (Figure 5).

The addition of 44 mM sodium bicarbonate to DMEM D5648, which lacks HCO₃^-, replicates the common condition of culture media. 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 [12]. 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 [25,38]. Therefore, we initially tested the role of HCO₃^- in PTEN oxidation by H₂O₂ through evaluating the proportions of oxidized PTEN following 10 min exposure to stimulation media. The pre-incubation period aimed to diminish the HCO₃^- in both extracellular and intracellular environments. Our findings indicated HCO₃^- alone did not cause any PTEN oxidation. Meanwhile, H₂O₂ alone has significant lower percentage than H₂O₂-HCO₃^- combination. Additionally, there was no notable difference between 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 each H₂O₂ and HCO₃^-_. The plausible explanation is that HCO₃^- can empower the catalytic activity of H₂O₂, supporting the hypothesis there might be a formation of HCO₄^-, a more reactive oxidant.

Given that HCO₄^- was demonstrated to exhibit a greater oxidation rate towards sulfide when compared to H₂O₂ [19], 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 points to investigate the redox regulation of HCO₃^- and H₂O₂. Additionally, we reduced the concentration of H₂O₂ to 0.5 mM and expect 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 observed at 10 min, then reduced back to 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 minutes, typically peaking at 10 min [12,39,40]. 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 [22]. 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 the PTEN redox regulation, therefore exerts function as a signaling molecule.

PTEN acts as a negative regulator of PI3K/AKT signaling pathway, which induces various cellular processes including protein synthesis, cell survival, proliferation, and migration. This mechanism plays important role in human physiological and pathological conditions [7,11,41]. Following the result that H₂O₂-mediated oxidative inhibition of PTEN is potentiated and accelerated by the HCO₃^-, we examined its subsequent impact on phosphorylation of AKT. The elevating trends of pAKT/AKT reflect similar patterns observed in PTEN oxidation, both in bicarbonate-containing and bicarbonate-free status. Our findings are consistent with the previous evidence suggesting that H₂O₂ can activate PI3K/AKT pathway by oxidizing PTEN [10]. Additionally, the time-dependent increase of pAKT/AKT is equivalent to the rise of PTEN oxidation in the presence of bicarbonate, 2.7-fold and 2.5-fold, correspondingly. In another words, without bicarbonate, PTEN oxidation is impaired, leading to the delay in AKT phosphorylation. Thus, in this context, the increase in AKT activation definitely 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 PI3K/AKT signaling pathway via oxidative inactivation of PTEN.

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 PI3K/AKT signaling pathway. In addition, the reduction of oxidized PTEN by antioxidant systems is also be equivalently affected. The plausible explanation can be the formation of HCO₄^-. In the future, further experiments would be conducted to consolidate the role of HCO₄^- in the redox regulation of PTEN.