Attenuation of Oxidative Stress in HEK 293 Cells by the TCM Constituents Schisanhenol, Baicalein, Resveratrol or Crocetin and Two Defined Mixtures

- PURPOSE: Our working hypothesis is that single bioactive phytochemicals with antioxidant properties that are important constituents of Traditional Chinese Medicine (TCM) and their defined mixtures have potential as chemoprotective agents for chronic conditions characterized by oxidative and nitrosative stress, including Alzheimer’s. Here we evaluate the ability of baicalein, crocetin, trans -resveratrol or schisanhenol and two defined mixtures of these TCM phytochemicals to attenuate the toxicity resulting from exposure to cell permeant t -butyl hydroperoxide (tBPH) in wild-type and bioengineered (to express choline acetyltransferase) HEK 293 cells. METHODS: Endpoints of tBHP-initiated oxidative and nitrosative stress in both types of HEK 293 cells and its attenuation by TCM constituents and mixtures included cytotoxicity (LDH release); depletion of intracellular glutathione (GSH); formation of S-glutathionylated proteins; oxidative changes to the disulfide proteome; and real-time changes in intracellular redox status. RESULTS: At low µM concentrations, each of the TCM constituents and mixtures effectively attenuated intracellular toxicity due to exposure of HEK 293 cells to 50 or 250 µM tBHP for 30 min to 3 h. Confocal microscopy of HEK 293 cells transfected with mutated green fluorescent protein (roGFP2) showed effective attenuation of tBHP oxidation by baicalein in real time. Three redox-regulated proteins prominent in the disulfide proteome of HEK 293 cells were identified by MALDI-TOF mass spectrometry. CONCLUSIONS: We conclude that single TCM chemicals and their simple mixtures have potential for use in adjunct chemoprotective therapy. Advantages of mixtures compared to single TCM constituents include the ability to combine compounds with varying molecular mechanisms of cytoprotection for enhanced biological activity; and to combine chemicals with complementary pharmacokinetic properties to increase half-life and prolong activity in vivo .


INTRODUCTION
Reactive oxygen species (ROS)(1) and reactive nitrogen species (RNS) (2) are formed as by-products during normal cellular metabolism. In healthy mammalian cells, physiological concentrations of ROS and RNS are detoxified by enzymatic antioxidants including catalase, enzymes of the glutathione (GSH) oxidation-reduction cycle [GSH reductase and GSH peroxidase], the peroxiredoxins (prxs), and superoxide dismutase (SOD) in addition to the common non-enzymatic antioxidants GSH, Vitamins C and E, β-carotene and uric acid (3). Increased, pathological concentrations of ROS and RNS overwhelm these normal detoxication pathways and contribute to several chronic diseases, including AIDS, atherosclerosis, cancer, diabetes, and neurodegenerative disorders such as Alzheimer's and Parkinson's disease (4)(5)(6)(7). Thus, decreased formation and/or increased detoxication of ROS and RNS can attenuate the severity of chronic disease caused or promoted by toxic free radicals. _________________________________________ Their antioxidant activity is one explanation for the chemoprotective and neuroprotective effects of several foods and dietary supplements (8,9), and for some of the therapeutic effects of plant-based TCM (10).
The association between the administration of potent antioxidants and the attenuation of free radical-induced disease (3,9,10) influenced our choice of specific TCM constituents for evaluation both in a recent study of sulfamethoxazole Nhydroxylamine-(SMX-NHOH-) mediated cytotoxicity in Jurkat E6.1 cells (11) and in the present study. Here, we evaluated the chemoprotectant effects of these four major TCM constituents (baicalein (BE), crocetin (Cro), transresveratrol (Res) and schisanhenol (Sal) and two defined mixtures against the well characterized, cellpermeant oxidative and nitrosative stressor, tertiarybutyl hydroperoxide (tBHP) (12,13) in wild-type (WT) transformed human renal epithelial cells (HEK 293) cells and HEK 293 cells bioengineered to express the cholinergic neuron-specific enzyme choline acetyltransferase (ChAT, EC 2.3.1.6) (14,15) (ChAT-HEK 293 cells). Recognizing that TCMs also have cytoprotective effects due to modulation of cell signalling pathways (16)(17)(18)(19), we sought to optimize the opportunity for complementary biological activity in our defined mixtures by selecting four TCM antioxidants of varying structure and chemical class ( Figure 1) with known and diverse mechanisms of antioxidant and cytoprotective activity.
In a recent study (11), we showed these individual TCM constituents and two defined mixtures effectively attenuated the oxidative/nitrosative stress-mediated cytotoxicity of sulfamethoxazole N-hydroxylamine (SMX-NHOH), a toxic reactive metabolite of sulfamethoxazole, in Jurkat E6.1 cells. More specifically, pre-treatment of cells with 5 µM MIX or MIX-S for 30 min prior to incubation with 400 µM SMX-NHOH for 2 h decreased cytotoxicity by 93.4 ± 0.8% (mean ± SEM, N = 3) and 93.1±0.4%, respectively. By comparison, the least effective TCM, 5 µM BE, inhibited cytotoxicity by 57.5 ±1.91% (P<0.05 vs MIX (equimolar amounts of BE, Cro, Res, Sal) or MIX-S (equimolar BE, Cro, Res) and the most effective, Sal attenuated SMX-NHOH toxicity by 82.4 ±1.3% (P<0.05 vs BE). These data encouraged us to investigate these TCM mixtures in further detail.
Low and moderate intracellular oxidative stress can cause reversible post-translational modification of proteins, compromising enzyme activity by altering protein structure and conformation. S-Glutathionylated proteins (PSSG; proteinglutathione mixed disulfides) are formed by both ROS and RNS at reactive (ionized) cysteine thiol residues upon reaction with GSH. As a cell recovers from oxidative/nitrosative stress these disulfides are reduced to GSH and PSH, restoring homeostasis (7,35,36). The conversion of reactive (increased electrophilicity) cysteine thiols in proteins to their GSSP derivatives, and to protein-protein disulfides (PSSP) when GSH is depleted, occurs selectively in response to ROS or the RNS, NO (as NO-GSH). These changes, including the protein disulfides formed, are the disulfide proteome (7,35,36).
Consequently, we compared the Sglutathionylation of intracellular proteins in WT-HEK 293 cells and ChAT-HEK 293 cells treated with tBHP, the latter to determine if ChAT is a redox regulated protein. This is of importance because ChAT is a critical enzyme in cholinergic neurons whose activity decreases early in the development of Alzheimer's (14,15). In addition, we analyzed tBHPmediated changes to the disulfide proteome of HEK 293 cells by reductive two-dimensional (R2D) SDS-PAGE and the attenuation of these changes by pretreatment with Sal, BE, Cro or Res, (Figure 1) alone or as two equimolar TCM mixtures. In supplementary experiments, attenuation of tBHP initiated cytotoxicity (studied by LDH release) and GSH depletion were studied with some but not all TCM treatments. Changes in intracellular redox potential in HEK 293 cells transfected with mutated green fluorescence protein (roGFP2) were followed in real time by confocal microscopy immediately after exposure to tBHP. The ability of BE, a well investigated TCM antioxidant, to attenuate these effects was also evaluated.
Finally, we identified three of the important redox regulated proteins in the disulfide proteome of HEK 293 cells by MALDI-TOF MS analysis of ingel tryptic digests of proteins separated by R2D SDS PAGE.

Cell Culture and Treatments with TCM Constituents and Mixtures
WT-HEK 293 (ATCC ® CRL-1573™) cells were obtained from ATCC (Manassas, VA, USA) and maintained in a humidified atmosphere at 37°C and 5% CO 2 in Eagle's Minimum Essential Medium (Gibco, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS), 26 mM NaHCO 3 and 10 µg/mL gentamycin sulfate. WT-or ChAT-HEK 293 cells (preparation described in 14) were plated 24 h prior to treatment and then incubated with various concentrations (10 -100 µM) of Sal, BE, Res or Cro individually, or with a mixture of the four antioxidants at equimolar concentrations (MIX), or an equimolar mixture of BE, Res and Cro without Sal (MIX-S) for 30 min. Following this incubation step, cells were exposed to either 50 or 250 µM tBHP for 30 min to 3 h. The effects of MIX-S were tested to determine whether Sal was a significant contributor to the protective effects of the complete mixture (MIX). Solutions of antioxidants were freshly prepared in dimethyl sulfoxide (DMSO), and cells were treated with TCM in no more than 0.2% DMSO. Protein concentrations were determined using the Lowry procedure (37).

Lactate Dehydrogenase (LDH) Activity
Cytotoxicity was examined by measuring release of intracellular LDH activity with a colorimetric assay described in detail elsewhere (11,38). Briefly, WTor ChAT-HEK 293 cells were plated at 8x10 4 cells in a 96-well culture plate for 24 h before treatment. Culture media was replaced with phosphate buffered saline (PBS) before antioxidant treatment (0-100 μM for 30 min). Cells were then treated for 3 h with 0-500 µM tBHP or the detergent 1% Triton X-100 (which released all LDH) for cytotoxicity. After incubation, cell supernatant (100 μL) was removed and incubated with LDH Reaction Mixture (100 μL). Released LDH activity was analyzed by measuring absorbance at 490 nm (Safire F129013, Tecan, Austria) and correction for LDH release in solvent controls.

Determination of Intracellular Glutathione (GSH) Concentration
Cells were plated at 8x10 5 cells/well in 6-well plates and cultured for 24 h. Cells were incubated with TCM constituents (10-50 μM for 30 min), then treated with 0-500 μM tBHP for 30 min and centrifuged for 5 min at 500xg. Cell pellets were resuspended in 250 μL lysis buffer (10 mM Tris, 130 mM NaCl, 10 mM NaF, 10 mM NaH 2 PO 4 , 10 mM Na 4 P 2 O 7 , 2 mM EDTA) and frozen in liquid nitrogen. Cell lysates were thawed at room temperature and aliquots taken for protein analysis (Bradford Bio-Rad Assay Kit). Remaining lysate was centrifuged for 2 min at 14,000xg and supernatant collected for GSH analysis by the GSH S-transferase enzymatic method (39). Supernatant (50 μL) and 0.1 M potassium phosphate buffer (140 μL, pH 7.6) were added to a 96-well plate to which were added 40 mM ethanolic DNCB (5 μl) and GSH S-transferase (0.2 U). Reaction mixtures were incubated in the dark at room temperature for 20 min and the S-2,4-dinitrophenyl GSH formed quantitated by absorbance at 340 nm (39). To quantify GSSG, aliquots of the same incubation mixtures used for GSH analysis were treated with NADPH and GSH reductase to convert all GSSG to GSH; then Total GSH content equalled the amount of S-2,4dinitrophenyl GSH present after reduction of GSSG. Sample GSSG content was calculated as the difference between Total GSH (GSH + GSSG) and Sample GSH content.

Intracellular Redox Status Determined with Transiently Transfected roGFP2
Intracellular redox status of HEK 293 cells was measured in real time by confocal laser scanning microscopy (LSM 510 META, Carl Zeiss Advanced Imaging Microscopy, Germany) using a novel redox-sensitive mutated green fluorescent protein (roGFP2) (40,41) in which two cysteine residues were introduced near the chromophore site (S147C and Q204C). Under oxidizing conditions, a reversible intramolecular disulfide bond is formed between these two cysteine thiols resulting in increased fluorescence excitability at 405 nm and decreased excitability at 488 nm (emission 530 nm) so that oxidized and reduced forms of the transfected roGFP can be quantified. For these experiments, either WT-or ChAT-HEK 293 cells were plated at a density of 1x10 6 cells/100 mm tissue culture dish and cultured for 24 h. Cells were transiently transfected with roGFP2 plasmid DNA at 10 μg/dish for 6 h using ExGen 500 transfection reagent (Fermentas International Inc, Burlington, CA). After 6 h, excess roGFP2 plasmid DNA was removed by changing the culture media, and cells were incubated for a further 24 h. Cells were re-plated at a density of 2x10 5 cells/dish in 35 mm glass-bottomed confocal culture dishes for 24 h to allow cells to adhere. For cell treatment, 50 μM tBHP or 10 μM BE + 50 μM tBHP were added with continuous monitoring of 405 nm/488 nm excitation fluorescence changes by confocal microscopy. roGFP2 plasmid DNA was obtained as a gift from Professor James Remington via a Material Transfer Agreement between Western University and the University of Oregon.

Redox Two-Dimensional SDS-Polyacrylamide Gel Electrophoresis (R2D SDS-PAGE)
We recently described this procedure (adapted from Cumming et al (42)) in detail elsewhere (11) so a condensed version appears here. Cells, were treated, lysed and centrifuged for 3 min at 500xg. Cell pellets were washed with PBS, re-centrifuged, resuspended in ice-cold PBS with 40 mM IACD, centrifuged at 500xg for 3 min and resulting pellets resuspended in 60 µL lysis buffer and frozen in liquid nitrogen. At the time of analysis, lysates were thawed in a water bath at room temperature and centrifuged for 10 min at 14,000xg. The supernatant was transferred to a fresh tube and assayed for protein content (37).
An equal volume of Laemmli sample buffer, without reducing agents, was added to solubilized cytosolic extracts. SDS-PAGE gels (10% acrylamide, 1.0 mm thickness) were prepared (11,42), and protein samples (75 or 85 µg) subjected to electrophoresis in the first dimension for 5 h at constant current (25 mA) using a Bio-Rad Protean II apparatus (11). After electrophoresis, the gel lanes containing the separated proteins were cut and immersed in Laemmli sample buffer containing 100 mM DTT for 20 min. Following a brief wash with electrophoresis running buffer, the gel slices were immersed again in Laemmli sample buffer containing 100 mM DTT for 10 min. Each gel strip was then applied horizontally to another gel (10% acrylamide, 1.5 mm thickness), and electrophoresis was performed in the second dimension for 14 h at constant current (10 mA/gel). Gels were fixed in 50% methanol, 5% acetic acid for 20 min, then silver-stained (11,43).

Mass Spectrometry (MS)
Protein spots that exhibited reproducible migration following R2D SDS-PAGE were excised (Ettan Spot Picker, General Electric Healthcare, Piscataway, NJ). In-gel digestion with sequence grade trypsin, and lyophilization were carried out with a Mass Prep Automated Digestor (Waters, Milford, MA). Lyophilized peptide samples were dissolved in 10% acetonitrile and 0.1% trifluoroacetic acid (5 µl), then mixed at a 1:1 ratio with 5 mg/mL α-cyano-4hydroxycinnamic acid in 6 mM ammonium phosphate monobasic solution. Sample (0.75 µL) was spotted on a sample target in duplicate to acquire matrix-assisted laser desorption/ionization (MALDI) MS spectra using a BiosystemsR 4700 Proteomics Discovery System, a matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometer with TOF/TOF optics (MALDI-TOF).

Protein Identification
The MALDI MS spectra were converted into data with 4000 Series Explorer software (Applied Biosystems, Foster City, CA), and processed into a peptide mass fingerprint with Data Explorer software (Applied Biosystems, Foster City, CA). Protein identification using peptide mass fingerprints (PMF) from MALDI MS spectra were obtained using a GPS engine connected to a Mascot server provided by the software manufacturer. PMF were compared to known peptide mass sequences obtained from the NCBI database. Samples with a PMF with a high protein score (typically above 90%) were predicted to identify a particular protein.

Statistics
A one-way analysis of variance (ANOVA) followed by Bonferroni's multiple comparison post-hoc test were used to compare groups. The level of significance was p < 0.05. When appropriate, Student's t-test was used to determine differences between vehicle control and treatment groups. Statistics were performed using GraphPad Prism Version 4.03 (GraphPad Software, Inc).

Protective effects of Sal and TCM Mixtures on tBHP-mediated Cytotoxicity
In these experiments, we compared the effects of Sal with those of our defined TCM mixtures on LDH release from cells as shown in Fig. 2. Treatment of cells for 30 min with Sal alone at either 10 or 20 μM did not alter LDH release compared to vehicle control. Moreover, neither MIX nor MIX-S at 10, 20 or 100 M resulted in LDH release from cells; MIX contains Sal at a concentration of 25 M. As expected, the treatment of cells with the oxidative stressor tBHP (250 µM for 3 h) significantly increased LDH release (38%) (P<0.05). Importantly, when added for 30 min prior to tBHP treatment, all TCM preparations tested were chemoprotective, significantly reducing LDH release from cells (P<0.05). There were no significant differences between MIX, MIX-S or Sal for the attenuation of tBHP-mediated LDH release at the concentrations tested. Acute treatment of cells with 100 µM Sal resulted in moderate cytotoxicity (23% LDH release compared to vehicle-treated control cells, P<0.05) (data not shown).
Of interest, ChAT-HEK 293 cells treated with vehicle (0.2% DMSO) released significantly more LDH than did WT-HEK 293 cells (Fig. 3), suggesting the bioengineered cells may be under greater oxidative stress. Importantly, when ChAT-HEK 293 cells were treated for 30 min with either 5 or 10 µM BE, LDH release was reduced to that from WT-HEK 293 cells treated with either vehicle or BE. Treatment of cells with 250 µM tBHP resulted in a significant increase in LDH from both WT-HEK 293 and ChAT-HEK 293 cells (P<0.05), with the increase being significantly greater in ChAT-HEK 293 cells than WT cells (23% compared to 11%, respectively; P<0.05). This suggests enhanced antioxidant activity in the WT-compared to ChAT-HEK 293 cells. BE pre-treatment at either 5 µM or 10 µM attenuated tBHP cytotoxicity in both WT-and ChAT-HEK 293 cells with the effects being similar at the two concentrations.

Intracellular GSH Status and Depletion in WTand ChAT-HEK 293 Cells
GSH is the main non-enzymatic antioxidant defense within mammalian cells, therefore the concentration of intracellular GSH was determined in ChAT-HEK 293 cells with or without BE treatment, and with and without exposure to tBHP (Figure 4). A concentration-dependent decrease in intracellular GSH was observed after treatment with 50 or 250 µM t-BHP for 1 h (25% and 50% GSH depletion, respectively). The GSH content was higher in cells treated with 50 µM tBHP for 1 h than for 30 min, showing partial recovery from the oxidative stressor by 60 min. At a concentration of 10 µM, BE did not alter cellular GSH content relative to the vehicletreated control cells, nor did it provide protection in cells that were treated with 50 µM tBHP for 30 min (36% vs 35% GSH depletion in control vs BE treated cells). Statistical significance between the groups was tested using one-way ANOVA followed by Bonferroni's multiple comparison test and showed no differences (P>0.05). On the other hand, BE did provide considerable protection against the oxidative stress associated with exposure to 250 µM tBHP for 30 min (25 vs 50% GSH depletion, BE vs control).
Consistent with the greater cytotoxicity of tBHP in ChAT-HEK 293 cells (Figure 3), pre-treatment of WT-HEK 293 cells with 10 µM BE for 30 min effectively prevented subsequent tBHP-mediated oxidation and also increased intracellular GSH content (vs controls) more than in the bioengineered cells. For example, GSH depletion (23%) caused by 50 µM tBHP for 30 or 60 min was completely prevented by 10 µM BE pre-treatment (P<0.05; data not shown). However, treatment with BE 10 µM for 24 h was less effective than treatment for 30 min (30% GSH depletion vs no depletion at 30 min). These data emphasize the importance of repeated administration of BE for effective antioxidant activity due to rapid metabolic conjugation and elimination from mammalian cells in vitro and in vivo.

Detection of Protein-GSH Mixed Disulfides (PSSG) in WT-HEK 293 Cells after Treatment with tBHP
During oxidative stress, proteins with reactive cysteine thiol residues (PS -) can be reversibly oxidized to sulfenic acids (PS-OH) which are converted either to S-glutathionylated proteins (PSSG) in the presence of GSH or to protein-protein disulfides (PSSP) in its absence (7,35,36). Analysis of cellular proteins by one-dimensional nonreducing SDS-PAGE coupled with immunochemical detection using anti-GSH antibodies can be used to identify PSSG as immunoreactive bands following electrophoresis of proteins in lysed cells. As shown in Figure 5, cytosol isolated from WT-HEK 293 cells treated with either 250 μM or 1 mM tBHP for 30 min contained many PSSG formed from redox-sensitive proteins compared to either 0.2% DMSO or BEtreated cells. As verification that PSSG are formed, these bands were absent in aliquots of these samples treated with 100 µM DDT which reduces both PSSG and PSSP. Treatment of cells with 10 μM BE for 30 min did not attenuate PSSG formation initiated by tBHP, emphasizing that these GSH-containing disulfides are part of the detoxication process. They are readily reduced to original PSH and GSH as the cell returns to its (normal) redox potential (35,36).

Detection of PSSG, Including ChAT-SSG, and PSSP Containing ChAT Resulting from Oxidative Stress in ChAT-HEK 293 Cells
Treatment of ChAT-HEK 293 cells with 250 µM tBHP for 30 min also resulted in the formation of many PSSG, as shown in the non-reduced gel stained with anti-GSH antibody ( Figure 6). Of particular interest is the strong, single protein band with an apparent molecular mass near 62 kDa. This protein is probably ChAT-SSG because a prominent immunopositive band of identical molecular mass is present when the blots were stripped and re-probed with anti-ChAT antibody. The fact that each of the bands associated with PSSG virtually disappears upon reduction of cell lysates with 100 mM DTT confirms they are disulfides formed from redox regulated proteins with reactive cysteine thiols. The disappearance of higher molecular mass ChATpositive bands from the reduced lysates of cells treated with tBHP demonstrates that ChAT is a redox regulated protein, as suspected from the cysteine array identified earlier in its crystal structure (44,45). These could be homodimeric or heterodimeric ChAT disulfides.
The greater intensity of the ChAT disulfide band in BE and EtGSH treated cells suggests these chemicals protected against cytotoxicity by preventing further oxidation of ChAT to tri-and poly-sulfides and/or attenuated oxidation of ChAT to its more reactive sulfenic acid derivative (ChAT-S-OH). Ethyl glutathione (EtGSH; 2.5 mM) was used to increase the intracellular GSH content in some experiments.

Imaging Dynamic Redox Changes in WT-HEK 293 Cells with Transfected roGFP2
Dynamic changes in the intracellular redox status of WT-HEK 293 cells transiently transfected to express roGFP2, a novel fluorescent reagent that tracks redox changes in the cytoplasm of living cells (40,41), were monitored in real time by confocal laser microscopy following tBHP treatment. Results are presented as images of cells obtained at an emission wavelength of 530 nm with excitation at both 488 nm (reduced roPFG2) and 405 nm (oxidized roGFP2). Changes were quantified by measuring the ratios of fluorescence from excitation at 488 nm to 405 nm using pixel counting software (ImageJ software, NIH). Changes in the intracellular oxidation state were determined by analyzing roGFP2-expressing WT-HEK 293 cells for 1 h after addition of 50 µM tBHP to medium containing these cells (Figure 7).
Initially, roGFP2 was primarily in its reduced form (preponderance of green compared to red fluorescence at time = 0 min; left panels of Figure 7), an indication of healthy mammalian cells. The ratio of fluorescence intensity at excitation wavelengths of 488 nm/405 nm was 7.35 ± 0.54 (mean ± SEM, n = 8) prior to the addition of tBHP to the cells (time = 0 min). roGFP2 expressed in the WT-HEK 293 cells became oxidized upon the addition of 50 µM tBHP to the medium, and cells showed membrane blebbing as they underwent oxidative stress (discussed below). Thirty min following the addition of tBHP, the 488/405 nm excitation ratio roGFP2 was decreased by about 55%. As a sign that the cells were recovering, oxidative stress-mediated blebbing was markedly less by 40 min and disappeared by 60 min (right panels of Figure 7). However, the cells did not return to their initial redox status at 60 min of treatment with tBHP. Red was selected as the colour for oxidized roGFP2 and green for the reduced form of this protein.

Attenuation of tBHP-mediated Oxidation by BE in ChAT-HEK 293 Cells Transfected with roGFP2 and Imaged by Confocal Microscopy
The treatment of roGFP2-expressing ChAT-HEK 293 cells with 10 µM BE for 30 min prior to their exposure to 50 µM tBHP significantly attenuates intracellular oxidation of roGFP2 ( Figure 8). Changes in cellular roGFP2 fluorescence (488/405 nm excitation ratio, emission at 505 nm) demonstrate divergence from BE-treated cells as early as 6 min after exposure to the oxidative stressor, although the differences did not become significant until 15 min after tBHP addition. Relatively little oxidation of roGFP2 is associated with tBHP exposure in the BEtreated cells over the 30 min course of dynamic monitoring. This was not the case with the tBHPtreated cells. There was no indication of roGFP2 reduction in the cells treated only with tBHP over 30 min. More specifically, treatment with 50 µM tBHP caused a 44% decrease of reduced/oxidized roGFP2 fluorescence ratio by 14 min which increased to 55% by 30 min. In cells pre-treated with BE, the decrease in reduced/oxidized roGFP2 fluorescence ratio was 16% at 30 min (P< 0.05 compared to tBHP treatment only). Confocal microscopy also showed that BE effectively attenuated the cell membrane blebbing induced by tBHP exposure, consistent with its chemoprotectant properties (data not shown)..

Identification of Important Redox-Regulated Proteins in WT-HEK 293 Cells by Mass Spectrometry and Peptide Mass Fingerprinting (MS-PMF) Following Separation by R2D SDS-PAGE
We chose to use post-translational modifications of reactive cysteine thiol residues of redox-regulated proteins in WT-HEK 293 cells as an endpoint to compare the ability of TCM constituents and defined mixtures to attenuate oxidation of the disulfide proteome by cell permeant tBHP. Initially, we selected 9 protein spots from the redox 2D protein profile of WT-HEK 293 cells for potential identification and "biomarker" use. However, we subsequently focused on 2 prominent spots (labelled 2 and 3 in Figure 9A) and a minor spot (labelled 1) that were present in all samples from control cells (Fig. 9A). These 3 spots showed dramatic changes upon tBHP treatment, suggesting their importance during oxidative/nitrosative stress.

Brightfield
We positively identified two of these proteins [spots # 2 and 3] by MS-PMF (Table 1 for spot # 2) following spot picking and in-gel digestion with sequence grade trypsin, and have tentatively identified the third protein [spot # 1] ( Table 2). The PMF generated for each trypsin-digested protein spot was compared to known peptide mass sequences available on the NCBI database using a search engine (MASCOT, Matrix Science, Boston, MA), and the protein score confidence (PSC) indicated as a percentage for each protein PMF was reported. The PSC is a statistical calculation that compares how closely the acquired PMF compares to known protein peptide sequences and is a measure of the certainty for which the PMF identifies a specific protein. MS-PMF analysis of spots 2 (human prx 2) and 3 (human prx 3) in WT-HEK 293 cells yielded PSC percentages >90%, whereas that of spot 1 (human prx 1) was only 77%. Six other protein spots that we analyzed did not yield acceptable PSC data demonstrating that in our hands protein identification by R2D-SDS-PAGE-MS is limited to proteins that are expressed in relatively large amounts.

Attenuation of Oxidation of Prx 2 and Prx 3 Thiols by BE Pre-treatment of WT-HEK 293 Cells Exposed to tBHP
Cytosolic prx 2 and prx 3, and to a lesser extent mitochondrial prx 1, are clearly visible upon R2D SDS-PAGE analysis of cytosol from vehicle-treated WT-HEK 293 control cells ( Figure 9A). This is the case for both lysate and cytosol from all untreated or solvent treated WT-HEK 293 cell preparations we  Figure 9B) increases the size of the spot for each of these proteins slightly, indicative of more prx disulfide (vs solvent controls). The disulfide-linked dimeric intermediary prx (prx-SS-prx) is present in each case because these disulfides are part of the 2-cysteine prx catalytic recycling mechanism that occurs with prx 1-prx 4 (46)(47)(48)(49). The presence of prx 1, prx 2 and prx 3 in their disulfide-linked dimeric form in untreated WT-HEK 293 cells is related to their physiological role, the reduction of µM concentrations of intracellular H 2 O 2 (49). Prx protein sulfinic acids are not reduced by DTT so they are not visible as individual spots following R2D SDS-PAGE, but instead run on the prominent diagonal line in the gels that contains all proteins that are not in a disulfide form. Prx 1, 2 and 3 occur as homodimeric disulfides (45)(46)(47)(48) that are reduced by DTT in the second dimension of R2D SDS-PAGE. As their molecular weight is halved (protein disulfide to 2 molecules protein thiol), these prxs are visible as single spots below and to the right of the prominent diagonal line that is formed by the mass of cellular non-disulfide, non-redox regulated proteins. Cytosol was isolated from cells and 75 µg protein analyzed for each treatment group. All gels were run in triplicate and sets of gels were silver stained simultaneously, with the experiment repeated three separate times with different cell cultures (i.e. a total of 9 gels were analyzed for each treatment group). All spots marked were present on each gel in the respective treatment groups.  Treatment of cells with 50 µM tBHP for 30 min ( Figure 9C) resulted in the disappearance of prx 1 and prx 3 and a marked diminution of prx 2. This is due to the increased formation of prx sulfinic acids at the expense of prx disulfide-linked dimers (48), a reaction termed hyperoxidation. Pretreatment of the cells with 10 µM BE for 30 min before tBHP exposure partially attenuated the effects of this oxidative stressor. Thus, the prx 2 spot is more intense than in cells treated with tBHP alone, and prx 3 becomes visible in BE-pretreated cells treated with tBPH but not in cells treated with tBPH alone.

Effect of Treatment of WT-HEK 293 Cells with Schisanhenol (Sal) or Two Defined Equimolar TCM Mixtures on the Disulfide Proteome, With or Without Subsequent Exposure to tBHP
The  Figures 10E vs 10D). In addition, the prx 2 spot appears slightly larger in 20 µM MIX-S vs 20 µM MIX treated cells (Figures 10F vs 10D) but is not as intense. Thus, not even a slight difference can be inferred between these two treatment regimens. In common, each of the TCM treatments evaluated reduces the cytosolic compartment of HEK 293 cells, an observation previously made with BE, Cro, Res and Sal and these two TCM mixtures in Jurkat E6.1 cells (11). MIX-S 20 μM. All gels were run in triplicate from three different cell cultures. Proteins spots marked 1, 2 and 3 were present on each run and were identified by mass spectrometry to be peroxiredoxins (prx) 1, 2 and 3, respectively ( Table 2).
As opposed to the R2D SDS-PAGE gels that were prepared for the analysis of cytosol from WT-HEK 293 cells treated only with TCM constituents that contained prominent spots for prx 2 and prx 3 ( Figure 10B-10F), gels from WT-HEK 293 cells treated in the same experiments with 50 µM tBHP contained a much less intense spot for prx 2 and no spots for prx 3 or prx 1 ( Figure 11A), equivalent to data with a different cell culture shown in Figure 9C.
In cells treated with TCM regimen first and then with tBHP, minor chemoprotection was observed with some ( Figures 11C-11F) but not all of the treatments ( Figure 11B). The prx 2 spot is somewhat more intense in cells treated with 100 µM ( Figure  11C) compared to 20 µM Sal ( Figure 11B) and 100 µM ( Figure 11E) compared to 20 µM MIX ( Figure  11D). Of significance to our interest in defined TCM mixtures as chemoprotectants, the prx 2 spot is more intense with 20 µM MIX ( Figure 11D) and 20 µM

D E
MIX-S ( Figure 11F) than with 20 µM Sal ( Figure  11B). Although these changes are not dramatic, they were observed in each of the separate experiments (n=9 gels for each treatment we conducted.

DISCUSSION
Experiments described in this paper were designed to evaluate the ability of single TCM constituents (Sal, BE, Cro or Res, with emphasis on understudied Sal and well-studied BE (24)) with known chemoprotectant properties at nontoxic concentrations (16)(17)(18)20,24,25,28), and of two novel defined mixtures with lower cytotoxicity than individual constituents in experiments performed in our laboratory with Jurkat E6.1 cells (11), to attenuate tBHP-mediated oxidative stress in WT-HEK 293 cells and ChAT-HEK 293 cells.
The individual chemicals were chosen because their structures and chemical classes are divergent ( Figure 1A -1D) as are some of their mechanisms of chemoprotection against oxidative and nitrosative stress (20,21,(24)(25)(26)(27)(28). These experiments have limitations because chemoprotection in vitro can vary markedly from that in vivo because of differences in absorption, distribution, metabolism and elimination (i.e. pharmacokinetics). This is also an area where carefully chosen TCM mixtures can provide superior chemoprotection compared to single TCM constituents while minimizing the adverse effects often associated with administration of complex TCM prescriptions (50).
tBHP is a membrane penetrable alkyl hydroperoxide reagent widely used to induce oxidative stress in cells (51)(52)(53)(54). The toxicity of tBHP is attributed to the generation of butoxyl radicals due to the Fenton reaction which can subsequently initiate lipid peroxidation, modify cellular lipids and proteins (Figures 9-11), affect cell integrity and signalling, and form covalent bonds with cellular molecules resulting in cell injury. Alternatively, the free radicals produced from tBHP (butoxyl and other ROS, including H 2 O 2 and NO, an RNS) can directly deplete intracellular thiols and GSH (Figure 4), thereby decreasing the capacity of nonenzymatic antioxidant systems and enzymatic radical detoxication (for example by GSH transferases and GSH peroxidases), eventually leading to cell death. An interesting phenomenon of tBHP toxicity in hepatocytes and other cells is plasma membrane bleb formation (blebbing). We observed this phenomenon by confocal microscopy in WT-HEK 293 cells (Figure 7) that had been treated with tBHP. Bleb formation is due to the oxidative modification of free thiol groups in cytoskeletal proteins (54), including actin which contains a reactive cysteine thiol moiety (Cys374), forms a S-glutathionylated derivative and is redox-regulated (55,56). Of importance, tBHPmediated blebbing was prevented in WT-HEK 293 cells (Figure 7) by pre-treatment with 10 µM BE, an excellent example of chemoprotection.
LDH is a cytosolic enzyme whose release into the cell medium (in vitro) or blood (in vivo) is a sensitive assay for cytotoxicity. Loss of cellular LDH due to impaired cell membrane integrity is an early indicator for cell death by both apoptosis and necrosis. Of the TCM antioxidants evaluated here (Figures 2 and 3), only treatment with 100 µM Sal for 30 min caused significant cytotoxicity to WT-HEK 293 cells (23% LDH release). Treatment of cells with 10 or 20 µM Sal did not cause cytotoxicity ( Figure 2).
All TCM treatments attenuated cytotoxicity due to exposure to 250 µM tBHP but there were no significant differences between Sal and the two mixtures, MIX (equimolar Sal, Cro, BE, Res) and MIX-S (equimolar Cro, BE, Res) in this regard.
ChAT is a critical enzyme that catalyzes the synthesis of the neurotransmitter acetylcholine. It has an unusually high number of cysteine amino acid residues (3.2% of total) and is a probable site for oxidative stress-mediated neurotoxicity (14,15,44,45). The availability of stably-transfected ChAT-expressing HEK 293 cells (14) allowed us to compare tBHP toxicity in ChAT-HEK 293 to WT-HEK 293 cells (Figure 3). Vehicle-treated ChATexpressing cells released more LDH into the medium than did WT-HEK 293 cells, but this cytotoxicity was decreased to control values by treatment of ChAT-HEK 293 cells with either 5 or 10 µM BE. Upon exposure to 250 µM tBHP for 3 h, the ChAT-HEK 293 cells released twice as much LDH into the medium as did the WT-HEK 293 cells. Attenuation of tBHP-mediated cytotoxicity by pre-treatment of the cells with 10 µM BE was significantly less effective in ChAT-expressing than it was in WT-HEK 293 cells (Figure 3).
The tripeptide, GSH (γ-Glu-Cys-Gly) is present at high concentrations (0.5-10 mM) in mammalian cells. It represents approximately 95% of total nonprotein sulfhydryls and is the major modulator of cellular redox equilibrium (57). The cytoplasmic ratio of reduced to oxidized glutathione (GSH/GSSG) in healthy cells approaches 100/1, and steady state is maintained by NADPH-dependent GSH reductase (57). Under conditions of moderate oxidative stress, a limited amount of GSH is converted to GSSG which alters the GSH/GSSH ratio as well as cellular redox status. In this study, we observed rapid, concentrationdependent GSH depletion after exposure of WT-HEK 293 cells to increasing concentrations (10 to 250 µM) of tBHP (data not shown). A similar concentration-dependent depletion of GSH was noted in ChAT-HEK 293 cells (20% at 50 µM, 50% at 250 µM tBHP for 60 min; Figure 4). A significant difference between the HEK cell types studied is that 10 µM BE was very effective at attenuating GSH depletion in WT cells but not in ChAT-HEK 293 cells, which appear to be under more oxidative stress (Figure 4). Of importance, treatment with BE (10 µM) for 24 h was less effective than 30 min treatment in preventing GSH depletion in WT cells. This emphasizes the importance of pharmacokinetics in selection of TCM agents and mixtures for in vivo (including human) studies. As mentioned above, BE is rapidly eliminated due to conjugation and active transport into urine. Thus, repeated administration is required for rapidly eliminated chemicals such as BE to be effective chemoprotectants in vivo although preparation of more-slowly metabolized derivatives, such as methoxyl ethers increases the half-life of polyphenols (58).
Proteins with ionized cysteine thiol residues at physiological pH (approximately 400 such proteins currently known) are more readily oxidized to their PS-OH derivatives than unionized protein thiols under conditions of mild and moderate oxidative or nitrosative stress (8,29,30). Post-translational modification at reactive cysteine thiols forms products that are biologically reducible (PSSG, PSSG-NO; PSSP; P'SSP). However, further oxidation leads to products that cannot be readily reduced enzymatically, including protein sulfinic acids (PSO 2 H), protein sulfonic acids (PSO 3 H) and polymeric protein disulfides ((PSSPSSP'SSPS'')x). Formation of these latter products frequently occurs concomitantly with severe cytotoxicity and cell death. Post-translational reactions at ionized cysteine thiol are the basis for selective redox regulation of many proteins including the prxs (7,47,48). We monitored S-glutathionylation of proteins subsequent to treatment with tBHP in both WT-( Figure 5) and ChAT-HEK 293 cells ( Figure 6) by non-reductive SDS-PAGE with detection of PSSG by a monoclonal antibody to GSH. In short, many Sglutathionylated proteins were formed during exposure to tBHP, verified by reduction of aliquots of cell lysates or cytosol with 100 mM DTT upon which anti-GSH staining was obliterated ( Figure 6). A more sophisticated approach was possible with tBHP-oxidized ChAT-HEK 293 cells ( Figure 6) because of the availability of anti-ChAT antibody (15). Rylett and her colleagues postulated that ChAT is a redox-regulated protein from their resolved crystal structure of human ChAT (44,45). Analysis of the protein crystals identified a cysteine tetrad at the active site of the enzyme. As ionized cysteine thiol is a common component of the active site of enzymes, they predicted that ChAT is redoxregulated (45). The experiments reported here confirm this fact. Under conditions of oxidative stress ( Figure 6) ChAT is converted to both ChAT-SS-G, a prominent band at 62 kDa which stained both with anti-GSH and anti-ChAT antibodies, and protein disulfides, demonstrated by disappearance of higher molecular wt bands than 62 kDa that stained with anti-ChAT upon reduction with DTT. Whether these are monomeric (ChAT-SS-ChAT) or dimeric ChAT disulfides (ChAT-SS-P) is unknown.
Real-time visualization of intracellular redox status of tBHP treated WT-HEK 293 cells (Figure 7) and ChAT-HEK 293 cells (Figure 8) via confocal microscopy was made possible by monitoring oxidation-reduction of the disulfide-dependent reporter protein, roGFP2 (40,41). This novel technology allowed us to monitor oxidation of the cell cytosol and blebbing concomitantly. The fact that blebbing disappeared before the cells returned to their initial redox equilibrium indicates there is a threshold for oxidative stress-induced blebbing in HEK 293 cells transiently transfected with roGFP2. Of importance, BE (10 µM) is able to prevent tBHP-(50 µM) mediated oxidation of cytosolic roGFP2 in HEK 293 cells over the 30 min these effects were monitored ( Figure 8). Part of this BE effect may result from its ability to decrease intracellular H 2 O 2 concentrations (20).
Because PSSG and PSSP were formed in abundance during tBHP treatment of WT-( Figure 5) and ChAT-HEK 293 cells (Figure 6) we required a more sensitive technique for the analysis of PSSP, i.e., the disulfide proteome (7,35,36,42). We utilized reduced two-dimensional SDS-PAGE (R2D SDS-PAGE) for this purpose (Figure 9). This technique is also called non-reducing/reducing "diagonal" PAGE and is an established method to resolve cytoplasmic proteins that undergo disulfide bond formation following oxidative stress (42). After staining the gel, proteins that do not contain any disulfide bond form a prominent diagonal line; proteins that are covalently linked by an intermolecular disulfide bond are found below the diagonal line (right side) and proteins that form intramolecular disulfide bonds are found slightly above the diagonal line (left side) formed by unmodified proteins. We noted more than 25 homodimeric or heterodimeric protein disulfides formed in WT-HEK 293 cells exposed to tBHP throughout the course of our study. Important for our purposes, tBHP always resulted in the disappearance or decreased size of two prominent cytosolic proteins, prx 2 and prx 3 upon R2D SDS-PAGE analysis. We identified these proteins using MALDI-MS FSM ( Table 1). The prxs exist, in part, as homodimeric protein disulfides, the reason they are resolved upon R2D SDS-PAGE. We selected prx 2 and prx 3 as our primary biomarkers for monitoring change in the disulfide proteome because they: i) are prominent protein spots present in every R2D SDS-PAGE analysis of WT-HEK 293 cell lysate we have ever run; ii) show dramatic changes (i.e. disappear) as a result of oxidative stress-mediated hyperoxidation; iii) are well characterized and known to be localized in the cytosol of cells (46,48); and iv) we confirmed their identity by MS-FSM analysis of WT-HEK 293 cells.
Based on changes in the size or intensity of silver-stained spots for prx 2 and prx 3 (and to a more limited extent, prx 1) in the R2D SDS-PAGE analysis of the 50 µM-tBHP oxidized disulfide proteome of WT-HEK-293 cells, we conclude that at equimolar concentrations, MIX-S (equimolar BE, Cro, Res) and MIX (equimolar Sal, BE, Cro, Res) are as protective or even slightly more cytoprotective than the single TCM constituents, Sal ( Figure 11) or BE ( Figure 9). Of significance, each TCM treatment evaluated did provide chemoprotection from oxidation of the disulfide proteome which is often a critical early step in cytotoxicity (e.g. cell blebbing) (54)(55)(56).
We are especially interested in the chemoprotection potential of Sal, a poorly studied dibenzocyclooctene lignin isolated from Schisandra rubriflora and characterized in the Chen laboratory (57,58). Here, we directly evaluated several properties of pure (>99%) Sal in HEK 293 cells in culture. In this regard, Sal was cytotoxic at 100 µM (caused release of approximately 20% intracellular LDH) but not at 10 or 20 µM (Figure 2). It significantly attenuated the cytotoxicity of 250 µM tBHP at 10, 20 or 100 µM showing optimal protection at 20 µM ( Figure 3). Sal (20 µM) attenuated tBHP-mediated oxidation of the disulfide proteome but not dramatically. We also evaluated the effectiveness of Sal as part of an equimolar mixture (Sal, BE, Cro, Res) compared to an equimolar mixture containing only BE, Cro and Res in this study. These two mixtures were almost equivalent in their ability to attenuate the cytotoxicity, lipid peroxidation, protein carbonylation, and hyperoxidation of prx 2 initiated by 400 µM SMX-NHOH in Jurkat E6.1 cells (11).
In a previous investigation (11), we showed that MIX and MIX-S are slightly more effective in attenuating hyperoxidation of intracellular prx 2 in Jurkat E6.1 cells exposed to SMX-NHOH than are equal concentrations of Sal, BE, Cro or Res.
In concert the results from these two studies suggest that defined mixtures of well characterized TCM constituents might serve as chemoprotectants for diseases caused or promoted by ROS/RNS. In this regard, such mixtures might best serve as complementary or adjunct therapy.
There are two potential advantages of welldefined TCM mixtures over single TCM constituents in therapy. First, potent compounds that exert their chemoprotective properties via different molecular mechanisms are likely to be more active in combination that alone. Second, compounds with different pharmacokinetic properties might be combined to increase half-life and prolong desired pharmacodynamic effect(s). Sal is an example of a TCM constituent with prolonged half-life, and prolonged biological activity. Sal contains 6 phenolic hydroxyl groups, which are responsible for antioxidant activity. However, 5 of these phenols are masked as methyl ethers in Sal ( Figure 1A, 57,58). These 5 methoxyl groups result in more efficient absorption due to increased lipophilicity of Sal (vs polyphenols with six free phenols), as shown by its short half-life (approximately 1 h) in plasma (30). To be efficiently eliminated, the methoxyl groups of Sal must be O-demethylated by cytochrome P450dependent oxidation, a slower Phase 1 reaction which also unmasks antioxidant phenolic groups. Once formed, phenols are rapidly conjugated by Phase 2 enzymes and the sulfate or glucuronide conjugates rapidly eliminated by active transporters in the liver (bile) and kidney (urine), We believe compounds similar to Sal can serve as important components of defined TCM mixtures used for chemoprotection in vivo due to their prolonged antioxidant activity and delayed metabolic clearance. Another approach to achieve the same pharmacokinetic and pharmacodynamic goals is by synthesis of lipophilic derivatives of antioxidant phenolic hydroxyls in TCM ingredients. Thus, five novel alkyl ether (ethyl, butyl, pentyl, hexyl and octyl) derivatives of Res were recently prepared which increased bioavailability compared to Res. Significantly, these derivatives retained the pharmacodynamic (melanogenesis inhibition) effect of Res evaluated (60) so that a mixture of Res and Alkyl-O-Res could be formulated to have both immediate and more prolonged biological activity based on rate of metabolic elimination and unmasking of antioxidant phenolic groups over time. In this respect, the metabolism of the ethyl ether would be more rapid than with any of the other alkyl ethers, with O-deoctylation being the slowest P450catalyzed reaction in this series.