Protein Redox State Monitoring Studies of Thiol Reactivity

Protein cysteine thiol status is a major determinant of oxidative stress and oxidant signaling. The -SulfoBiotics- Protein Redox State Monitoring Kit provides a unique opportunity to investigate protein thiol states. This system adds a 15-kDa Protein-SHifter to reduced cysteine residues, and this molecular mass shift can be detected by gel electrophoresis. Even in biological samples, Protein-SHifter Plus allows the thiol states of specific proteins to be studied using Western blotting. Peroxiredoxin 6 (Prx6) is a unique one-cysteine peroxiredoxin that scavenges peroxides by utilizing conserved Cysteine-47. Human Prx6 also contains an additional non-conserved cysteine residue, while rat Prx6 only has the catalytic cysteine. In cultured cells, cysteine residues of Prx6 were found to be predominantly fully reduced. The treatment of human cells with hydrogen peroxide (H2O2) formed Prx6 with one cysteine reduced. Since catalytic cysteine becomes oxidized in rat cells by the same H2O2 treatment and treating denatured human Prx6 with H2O2 results in the oxidation of both cysteines, non-conserved cysteine may not be accessible to H2O2 in human cells. We also found that untreated cells contained Prx6 multimers bound through disulfide bonds. Surprisingly, treating cells with H2O2 eliminated these Prx6 multimers. In contrast, treating cell lysates with H2O2 promoted the formation of Prx6 multimers. Similarly, treating purified preparations of the recombinant cyclic nucleotide-binding domain of the human hyperpolarization-activated cyclic nucleotide-modulated channels with H2O2 promoted the formation of multimers. These studies revealed that the cellular environment defines the susceptibility of protein cysteines to H2O2 and determines whether H2O2 acts as a facilitator or a disrupter of disulfide bonds.


Introduction
The redox state of protein cysteine residues plays a crucial role in cell regulation and is often the major target of oxidative stress and oxidant signaling conferred by reactive oxygen species [1][2][3][4][5][6][7]. Various labeling probes allow for the determination of the global protein thiol status in the cells [8]. However, the examination of the thiol states of specific cellular proteins has been difficult. In this regard, Dojindo Molecular Technologies, Inc. has developed a novel approach that allows easy determination of the cysteine redox states within the proteins of interest. In the -SulfoBiotics-Protein Redox State Monitoring Kit system, a 15-kDa Protein-SHifter is added to every reduced cysteine residue, allowing To study the redox states in the purified recombinant proteins, samples were subjected to non-reducing sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (PAGE) and visualized by Coomassie Brilliant Blue staining.
To study the protein redox states in the cultured cells, samples were prepared through protein precipitation using trichloroacetic acid (TCA; VWR International, Radnor, PA, USA) and labeled with Protein-SHifter Plus in accordance with the manufacturer's instructions for the -SulfoBiotics-Protein Redox State Monitoring Kit Plus. After cell extracts were subjected to SDS-PAGE, gels were exposed to UV light on a transilluminator to remove Protein-SHifter. Proteins in the gel were then electro-transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA). The membrane was blocked with 5% non-fat dry milk and incubated with rabbit Prx6 IgG antibody (Sigma-Aldrich). Signals were obtained by using a horseradish peroxidase-linked secondary antibody (Bio-Rad) and the Enhanced Chemiluminescence System (GE Healthcare Bio-Sciences, Pittsburgh, PA, USA) and were computed for Western blot band intensities. Protein concentrations of cell lysates were measured using the bicinchoninic acid (BCA) assay (Pierce BCA Protein Assay Kit, Thermo Fisher Scientific, Waltham, MA, USA).

Statistical Analysis
Means and standard errors of mean (SEM) were computed. Two groups were compared by a two-tailed Student's t test, and differences between more than two groups were determined by the analysis of variance (ANOVA). p < 0.05 was defined to be statistically significant.

-SulfoBiotics-Protein Redox State Monitoring Kit to Assess Thiol States of Purified Proteins
The -SulfoBiotics-Protein Redox State Monitoring Kit allows the detection of reduced protein cysteine residues by labeling with Protein-SHifter composed of a maleimide group that covalently binds to a protein thiol group and a DNA moiety that has a molecular mass of~15 kDa [18,19]. Labeled protein samples are subjected to non-reducing SDS-PAGE and visualized by staining the gel ( Figure 1A). If the protein contains one reduced cysteine, the electrophoretic band will be shifted by 15 kDa upward. If the protein contains two reduced cysteines, two Protein-SHifters bind to the protein molecule, causing a total of a 30 kDa shift as described in the schematics of Figure 1B. Figure 1C shows examples of how reduced cysteines may be detected in purified recombinant proteins. With Protein-SHifter, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) with a molecular weight of 36 kDa and containing three cysteine residues exhibited~51,~66, and~81 kDa bands, indicating that one, two, or three cysteine residues can be in the reduced state. In this particular preparation, the GAPDH protein molecules with three cysteine residues in the reduced state were the most prominent. The 28-kDa ethylmalonic encephalopathy 1 (ETHE1) protein molecule with nine cysteines was found to be highly reduced in this recombinant protein preparation.
We also tested the effects of Protein-SHifter on the recombinant CNBD domain of hHCN4 channels (19 kDa) with two cysteines. The results showed two bands corresponding to the modifications with one (34-kDa) and two (49-kDa) Protein-SHifter molecules. The observation that the 49-kDa species of the purified CNBD was predominant indicates that both cysteine residues reacted with the Protein-SHifter, which is consistent with the reducing effect of TCEP present in the protein solution. We also tested the effects of Protein-SHifter on the recombinant CNBD domain of hHCN4 channels (19 kDa) with two cysteines. The results showed two bands corresponding to the modifications with one (34-kDa) and two (49-kDa) Protein-SHifter molecules. The observation that the 49-kDa species of the purified CNBD was predominant indicates that both cysteine residues reacted with the Protein-SHifter, which is consistent with the reducing effect of TCEP present in the protein solution.

Redox Studies of Prx6 in Cultured Cells Using -Sulfobiotics-Protein Redox State Monitoring Kit Plus
In biological samples containing multiple proteins, proteins of interest can be studied using the -SulfoBiotics-Protein Redox State Monitoring Kit Plus. This "Plus" system contains the 15-kDa Protein-SHifter Plus whose DNA moiety can come off after electrophoresis by exposing the gels to UV light, aiding in visualizing the thiol redox states using Western blotting (Figure 2A) [19].
We treated human smooth muscle cells with either platelet-derived growth factor (PDGF) that may elicit oxidant signaling [20] or H2O2. Cells were treated with H2O2 for 15 min in accordance with

Redox Studies of Prx6 in Cultured Cells Using -Sulfobiotics-Protein Redox State Monitoring Kit Plus
In biological samples containing multiple proteins, proteins of interest can be studied using the -SulfoBiotics-Protein Redox State Monitoring Kit Plus. This "Plus" system contains the 15-kDa Protein-SHifter Plus whose DNA moiety can come off after electrophoresis by exposing the gels to UV light, aiding in visualizing the thiol redox states using Western blotting (Figure 2A) [19].
We treated human smooth muscle cells with either platelet-derived growth factor (PDGF) that may elicit oxidant signaling [20] or H 2 O 2 . Cells were treated with H 2 O 2 for 15 min in accordance with the product instructions of the -SulfoBiotics-Protein Redox State Monitoring Kit Plus. We prepared cell lysates in accordance with the -SulfoBiotics-Protein Redox State Monitoring Kit Plus protocol using the cell lysate buffer that was included in the kit after the TCA protein precipitation. We electrophoresed the Protein-SHifter Plus-labeled samples under the non-reducing condition, eliminated the DNA moiety of the Protein-SHifter Plus by UV light and blotted onto a nitrocellulose membrane, followed by immunoblotting with the Prx6 antibody.
The human peroxiredoxin 6 (Prx6) protein has a molecular weight of 25-kDa and contains two cysteine residues: a catalytic cysteine at position 47 responsible for the peroxidase activity and a non-conserved cysteine at position 91 ( Figure 2B). Our results showed that after labeling with Protein-SHifter Plus, Prx6 molecules from untreated human smooth muscle cells predominantly contained the 55-kDa species. This was consistent with the protein having two Protein-SHifters incorporated ( Figure 2C), suggesting that both of the cysteine residues were mostly in the reduced state in the cells ( Figure 2D). The band patterns were not altered by changing the amounts of Protein-SHifter Plus ( Figure 2E). While the treatment of cells with PDGF did not influence the thiol state of Prx6, the treatment of cells with H 2 O 2 as low as 200 µM for 15 min dramatically decreased the 55-kDa band and increased the 40-kDa band ( Figure 2D,F), indicating that only one sulfhydryl group was oxidized. This event does not appear to be Fenton reaction-dependent since the pre-treatment of cells with deferoxamine (50 µM) for 30 min did not influence the H 2 O 2 effects (data not shown). the product instructions of the -SulfoBiotics-Protein Redox State Monitoring Kit Plus. We prepared cell lysates in accordance with the -SulfoBiotics-Protein Redox State Monitoring Kit Plus protocol using the cell lysate buffer that was included in the kit after the TCA protein precipitation. We electrophoresed the Protein-SHifter Plus-labeled samples under the non-reducing condition, eliminated the DNA moiety of the Protein-SHifter Plus by UV light and blotted onto a nitrocellulose membrane, followed by immunoblotting with the Prx6 antibody. The human peroxiredoxin 6 (Prx6) protein has a molecular weight of 25-kDa and contains two cysteine residues: a catalytic cysteine at position 47 responsible for the peroxidase activity and a non-conserved cysteine at position 91 ( Figure 2B). Our results showed that after labeling with Protein-SHifter Plus, Prx6 molecules from untreated human smooth muscle cells predominantly contained the 55-kDa species. This was consistent with the protein having two Protein-SHifters incorporated ( Figure 2C), suggesting that both of the cysteine residues were mostly in the reduced state in the cells ( Figure 2D). The band patterns were not altered by changing the amounts of Protein-SHifter Plus ( Figure 2E). While the treatment of cells with PDGF did not influence the thiol state of Prx6, the treatment of cells with H2O2 as low as 200 μM for 15 min dramatically decreased the 55-kDa band and increased the 40-kDa band ( Figure 2D,F), indicating that only one sulfhydryl group was oxidized. This event does not appear to be Fenton reaction-dependent since the pre-treatment of cells with deferoxamine (50 μM) for 30 min did not influence the H2O2 effects (data not shown). Since only one cysteine out of two in human Prx6 was found to be oxidized by H2O2, we examined rat Prx6, which contains only one cysteine that is the conserved catalytic cysteine at position 47 ( Figure 3A). Using cultured rat smooth muscle cells, we performed the same Protein-SHifter Plus experiments as those performed in human cells. Results showed that the untreated rat cells contained Prx6 molecules that predominantly exhibited the 40 kDa band, which was consistent with one cysteine being reduced ( Figure 3B  Since only one cysteine out of two in human Prx6 was found to be oxidized by H 2 O 2 , we examined rat Prx6, which contains only one cysteine that is the conserved catalytic cysteine at position 47 ( Figure 3A). Using cultured rat smooth muscle cells, we performed the same Protein-SHifter Plus experiments as those performed in human cells. Results showed that the untreated rat cells contained Prx6 molecules that predominantly exhibited the 40 kDa band, which was consistent with one cysteine being reduced ( Figure 3B  To test the hypothesis that one cysteine residue of human Prx6 is not accessed by H2O2 in cells, we treated cell lysates with H2O2 after the proteins were denatured by SDS and before the addition of Protein-SHifter Plus. In contrast to the cell treatment with H2O2, which increased the 40 kDa Prx6 band (lane 1 vs. lane 2 in Figure 4), the ex vivo treatment with H2O2 dramatically decreased the 40 kDa Prx6 band (lane 2 vs. lane 3 in Figure 4). These results suggest that, perhaps due to the three-dimensional structure of Prx6 defined by the cellular environment, oxidation of non-conserved cysteine with H2O2 is blocked in the human Prx6 protein.  Figure 4). These results suggest that, perhaps due to the three-dimensional structure of Prx6 defined by the cellular environment, oxidation of non-conserved cysteine with H 2 O 2 is blocked in the human Prx6 protein.  Protein-SHifter Plus was then added and the samples subjected to SDS-PAGE without BME and immunoblotted with the Prx6 antibody. The bar graph represents the means ± SEM of the intensity of the 40 kDa bands (n = 3). The symbol * represents the value significantly different from each other at p < 0.05.

H2O2 Can Decrease Disulfide Bonds in the Cells
As the -SulfoBiotics-Protein Redox State Monitoring Kit Plus requires running non-reducing gels, in addition to the 25-, 40-, and 55-kDa bands that were expected to occur in the Prx6 monomer with or without the Protein SHifter, we also noticed higher molecular weight bands in these gels ( Figure 5A). These bands included 50-, 65-, 75-, 80-, and 105-kDa bands consistent with the occurrence of dimers and trimers of various redox states with different numbers of Protein-SHifters interacting with Prx6 ( Figure 5B). Interestingly, the treatment of cells with H2O2 for 15 min resulted in the elimination of these multimers, indicating that H2O2 may have disrupted the protein disulfide bonds in the cells ( Figure 5A,C).

H 2 O 2 Can Decrease Disulfide Bonds in the Cells
As the -SulfoBiotics-Protein Redox State Monitoring Kit Plus requires running non-reducing gels, in addition to the 25-, 40-, and 55-kDa bands that were expected to occur in the Prx6 monomer with or without the Protein SHifter, we also noticed higher molecular weight bands in these gels ( Figure 5A). These bands included 50-, 65-, 75-, 80-, and 105-kDa bands consistent with the occurrence of dimers and trimers of various redox states with different numbers of Protein-SHifters interacting with Prx6 ( Figure 5B). Interestingly, the treatment of cells with H 2 O 2 for 15 min resulted in the elimination of these multimers, indicating that H 2 O 2 may have disrupted the protein disulfide bonds in the cells ( Figure 5A,C). These Prx6 multimers were also detected in samples without the addition of Protein-SHifter. Figures 6A,B show that, even without Protein-SHifter, the 50 kDa dimer formation of Prx6 decreased in response to the H2O2 treatment of human cells in a dose-dependent fashion. The inclusion of β-mercaptoethanol (BME) in the loading buffer for gel electrophoresis also decreased the 50-kDa band ( Figure 6A,C), confirming that the 50-kDa species resulted from protein-protein interactions through disulfide bonds. An iron chelator, deferoxamine, had no effects ( Figure 6A,D), suggesting that the H2O2-induced multimer formation is not dependent on the Fenton reaction or hydroxyl radicals. A redox cycling reactive oxygen generator, 1,4-naphthoquinone, did not mimic the H2O2 effects (data not shown). In untreated rat cells, we also detected the Prx6 dimers. Figure 7   These Prx6 multimers were also detected in samples without the addition of Protein-SHifter. Figure 6A,B show that, even without Protein-SHifter, the 50 kDa dimer formation of Prx6 decreased in response to the H 2 O 2 treatment of human cells in a dose-dependent fashion. The inclusion of β-mercaptoethanol (BME) in the loading buffer for gel electrophoresis also decreased the 50-kDa band ( Figure 6A,C), confirming that the 50-kDa species resulted from protein-protein interactions through disulfide bonds. An iron chelator, deferoxamine, had no effects ( Figure 6A,D), suggesting that the H 2 O 2 -induced multimer formation is not dependent on the Fenton reaction or hydroxyl radicals. A redox cycling reactive oxygen generator, 1,4-naphthoquinone, did not mimic the H 2 O 2 effects (data not shown). In untreated rat cells, we also detected the Prx6 dimers.    In contrast to the effects of H2O2 that occur in cells, the treatment of human cell lysates with H2O2 resulted in the increased 50-kDa Prx6 dimer formation ( Figure 8A). The dimer formation decreased in the presence of BME, suggesting that dimer formation is mediated by disulfide bonds ( Figure 8A). To test if this effect could also be observed in other proteins, we used the purified CNBD domain of hHCN4 channels. Consistent with the multimeric assembly of the intact HCN channels and previous studies of the isolated CNBD domains [16], non-reducing gels showed the formation of CNBD dimers ( Figure 8B). Similar to the cell lysate studies of Prx6, the treatment of the purified CNBD domain with H2O2 increased the formation of 38-kDa dimers and 57-kDa trimers ( Figure 8B). In contrast to the effects of H 2 O 2 that occur in cells, the treatment of human cell lysates with H 2 O 2 resulted in the increased 50-kDa Prx6 dimer formation ( Figure 8A). The dimer formation decreased in the presence of BME, suggesting that dimer formation is mediated by disulfide bonds ( Figure 8A). To test if this effect could also be observed in other proteins, we used the purified CNBD domain of hHCN4 channels. Consistent with the multimeric assembly of the intact HCN channels and previous studies of the isolated CNBD domains [16], non-reducing gels showed the formation of CNBD dimers ( Figure 8B). Similar to the cell lysate studies of Prx6, the treatment of the purified CNBD domain with H 2 O 2 increased the formation of 38-kDa dimers and 57-kDa trimers ( Figure 8B).

Discussion
Here, we used the -SulfoBiotics-Protein Redox State Monitoring Kit to report on the cysteine redox state of two structurally and functionally unrelated proteins, Prx6 and CNBD of hHCN4 channels.
To our knowledge, there are currently no other methods that can easily detect thiol status, especially the number of reduced cysteine residues, of a given protein in the complex biological samples. The -SulfoBiotics-Protein Redox State Monitoring Kit Plus allowed us to easily assess the protein redox states in the cells as defined by the biological system, without the necessity of enriching (purifying) the protein of interest. Although there are remarkable advances in the methods of protein cysteine state detection, as summarized in the excellent reviews by Thamsen and Jakob, and Yang et al. [21,22], the methods that can easily detect thiol status and the number of reduced cysteine residues for a given protein in the complex biological samples are limited. For example, DTNB is frequently used for the detection of reduced cysteines [8]. In the biological samples, DTNB would label all reduced cysteines in the sample. Since the molecular weight of DTNB is ~400 Da, labeling of cysteines with DTNB does not introduce a significant enough shift in the molecular weight to be easily detected with Western blot using antibodies for the protein of interest. One could purify the protein of interest after the cell treatment, then measure thiol status by various methods.

Discussion
Here, we used the -SulfoBiotics-Protein Redox State Monitoring Kit to report on the cysteine redox state of two structurally and functionally unrelated proteins, Prx6 and CNBD of hHCN4 channels.
To our knowledge, there are currently no other methods that can easily detect thiol status, especially the number of reduced cysteine residues, of a given protein in the complex biological samples. The -SulfoBiotics-Protein Redox State Monitoring Kit Plus allowed us to easily assess the protein redox states in the cells as defined by the biological system, without the necessity of enriching (purifying) the protein of interest. Although there are remarkable advances in the methods of protein cysteine state detection, as summarized in the excellent reviews by Thamsen and Jakob, and Yang et al. [21,22], the methods that can easily detect thiol status and the number of reduced cysteine residues for a given protein in the complex biological samples are limited. For example, DTNB is frequently used for the detection of reduced cysteines [8]. In the biological samples, DTNB would label all reduced cysteines in the sample. Since the molecular weight of DTNB is~400 Da, labeling of cysteines with DTNB does not introduce a significant enough shift in the molecular weight to be easily detected with Western blot using antibodies for the protein of interest. One could purify the protein of interest after the cell treatment, then measure thiol status by various methods. However, these procedures are cumbersome and the thiol status of the protein could be altered during the protein purification process. Another frequently used method is a biotin switch assay, where the free cysteines are first alkylated with methyl methanethiosulfonate, then the oxidative thiol modifications are reduced and switched for biotin [23]. To determine the cysteine status of a given protein, the proteins with biotinylated cysteines have to be purified with avidin-affinity chromatography and the protein of interest has to be visualized with the specific antibody. Therefore, an additional purification step has to be involved. Perhaps the most precise method for determining the cysteine status of a given protein is the isotope coded affinity tag (ICAT) technology that involves labeling of the accessible cysteines with the thiol-trapping reagent iodoacetamide and biotin affinity tag, and then high pressure liquid chromatography-based protein purification, followed by mass spectrometry identification of the precise protein mass [24]. Although ICAT can provide valuable and precise information on the state of cysteines in a given protein, it is much more difficult to implement than the use of the -SulfoBiotics-Protein Redox State Monitoring Kit Plus described here. Thus, we believe that the -SulfoBiotics-Protein Redox State Monitoring Kit Plus that relies on novel SHifter use is revolutionary technology that should gain attention from the research community.
While studying the redox states of cysteine residues using the -SulfoBiotics-Protein Redox State Monitoring Kit, we encountered novel redox events. We found that in the cells, Prx6 proteins are subjected to a very complex situation where free cysteine residues are largely in the reduced state, but some of them are oxidized forming disulfide bonds. Remarkably, the exposure to H 2 O 2 resulted in further complex redox events including the oxidation of a specific cysteine residue without influencing others and the elimination of disulfide bonds that form multimers.
Our results indicate that, in the cellular environment, human Prx6 may be folded in a way that non-conserved Cys91 is not subjected to H 2 O 2 -dependent oxidation. Our conclusion is based on our experiments on rat Prx6, where catalytic Cys47 is the sole cysteine. This cysteine was found to be oxidized by H 2 O 2 in rat cells under the conditions identical to experiments using human cells. Such a conclusion was further supported by our observations that the exposure of denatured human cellular proteins to H 2 O 2 resulted in the oxidation of both cysteines in human Prx6. It is still unclear how Cys91 can stay reduced and yet inaccessible to H 2 O 2 . Further work is needed to address the issue of H 2 O 2 accessibility to protein thiols. In this regard, the -SulfoBiotics-Protein Redox State Monitoring Kit and the -SulfoBiotics-Protein Redox State Monitoring Kit Plus should serve as useful tools for such investigations.
It is intriguing that H 2 O 2 seems to be able to disrupt disulfide bonds in both human and rat smooth muscle cells. While it is known that H 2 O 2 can be both an oxidant and a reductant depending on the pH, it typically triggers the formation of disulfide bonds at physiological pH [25,26]. Consistently, our experiments using two unrelated proteins, Prx6 and hHCN4 CNBD, demonstrated that the treatment of denatured proteins with the same H 2 O 2 reagent as that used in our studies using living cells indeed promoted H 2 O 2 -induced oxidation and disulfide formation. Therefore, we propose that there are cellular mechanisms that allow for H 2 O 2 to exhibit its disruptive capacity toward disulfide bonds. Disulfide bonds may be dynamically oxidized and reduced in the cells and, when in a reduced state, H 2 O 2 may oxidize the thiols to alternative states, thus preventing the reformation of disulfide bonds. Alternatively, H 2 O 2 may further oxidize disulfides to thiosulfinates and thiosulfonates [27]. Further work is needed to define the mechanism of this event.

Conclusions
In summary, the -SulfoBiotics-Protein Redox State Monitoring Kit and the -SulfoBiotics-Protein Redox State Monitoring Kit Plus allowed us to provide new information concerning protein thiol biology in the present study. It is likely that new important information will be generated using these tools in the future. Funding: This work was supported in part by the NIH (R01HL072844, R21AI142649, R03AG059554, and R03AA026516) to Y.J.S. and the NIH (R01GM124020) to T.I.B. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.