Although moderate exercise provides many health benefits, intense/exhaustive exercise can cause oxidative stress due to massive production of reactive oxygen species (ROS) [1]. A single bout of strenuous exercise induces oxidative stress in circulating human blood cells and in rodent splenocytes and thymocytes, leading to lipid peroxidation, DNA damage, mitochondrial perturbations, and protein oxidation [2-5]. ROS generation during exercise also contributes to inflammatory damage [6].

NF-κB is a master regulator of gene transcription in response to inflammation and oxidative stress in various cells. NF-κB promotes expression of genes encoding pro-inflammatory cytokines and enzymes, such as COX-2, a rate-limiting enzyme in the biosynthesis of prostaglandins and thromboxanes. Inappropriate overexpression of COX-2 has frequently been observed in various premalignant and malignant tissues [7].

Apoptosis, a selective, controlled, and genetically programmed cell death, is characterized by several distinct biochemical and molecular events [8]. In mature animals, apoptosis balances cell division, maintaining the constancy of tissue mass. Removal of cells injured by genetic defects, aging, disease, or exposure to noxious agents is made possible by apoptosis [9]. It has been reported that strenuous exercise can lead to apoptotic cell death [10-12]. A panel of caspases are recognized as an important mediator of apoptosis. Once caspases activated, caspases exert pro-apoptotic activity in part by cleaving PARP, a key enzyme essential for DNA repair and cell survival [13].

In response to oxidative stress, cells tend to initiate protective mechanisms by inducing expression of antioxidant enzymes, such as heme oxygenase-1 (HO-1), a novel stress-responsive enzyme [14]. Numerous studies have identified HO-1 as an important component of cytoprotective defense mechanisms against oxidative and inflammatory insults [15].

Reduced glutathione (GSH) is one of the major antioxidants synthesized in cells and plays a key role in maintaining redox homeostasis, scavenging peroxides, and detoxifying electrophilic xenobiotics. Glutamate cysteine ligase (GCL), which catalyzes the formation of γ-glutamylcysteine, is the rate-limiting enzyme for de novo synthesis of GSH. Catalytic subunit of GCL (GCLC) can be induced in response to various oxidative stressors. This inducible nature suggests that GCL is an essential component of the cellular adaptation machinery under oxidative stress [16].

Expression of HO-1 and GCLC is known to be mediated by electrophilic response elements or antioxidant response elements (ARE), cis-acting enhancer sequences that mediate transcriptional activation of nuclear factor erythroid 2-related factor 2 (Nrf2) in response to oxidative stress and other noxious stimuli [17,18]. Nrf2 is a key mediator of physiological stress response that transcriptionally regulates antioxidant genes including HO-1, GCLC, NAD(P)H: quinone oxidoreductase 1 (NQO1), superoxide dismutase (SOD), catalase, GSH peroxidase, and GSH S-transferase (GST) [19].

Here, we report that exhaustive exercise induces transient expression of HO-1 and GCL via Nrf2-ARE signaling as an adaptive survival response against pro-inflammatory and pro-apoptotic insults in human peripheral blood mononuclear cells (PBMCs).

Materials

Rabbit polyclonal COX-2 and HO-1 antibodies were purchased from Lab Vision Co and Stressgen, respectively. Primary antibodies for caspase-3, cleaved caspase-3, PARP, and cleaved PARP were purchased from Cell Signaling Technology. Anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from Zymed Laboratories. An oligonucleotide probe containing the NF-κB and Nrf2 consensus sequence were purchased from Promega. Enhanced chemiluminescence (ECL) detection kits and [γ³²-³²P] ATP were obtained from Amersham Pharmacia Biotech, and the bicinchoninic acid protein protein assay reagent was supplied by Pierce Biotechnology. All other chemicals were used in the purest form that were commercially available.

Subjects and exercise treatments

Sixteen healthy young (23.1 ± 1.1 years) male volunteers participated in this study, which was approved by the Seoul National University Institutional Review Board (SNU-IRB; approval number: 0709/001-001). Prospective participants were screened to exclude any individuals with chronic illness, acute infection, or current use of prescription drugs. The subjects, who had not been involved in any exercise training for the preceding one month, were directed to make no major changes in their physical activity routines during the study. All procedures were approved by the SNU-IRB. All participants were asked to provide written informed consents to participate in this study before collection of blood samples. Physiological characteristics of the subjects are presented in Table 1.

Table 1 .. Characteristics of the subjects

Characteristic	Subjects (n = 16)
Age (yr)	23.1 ± 1.1
Height (cm)	175.3 ± 3.7
Weight (kg)	70.8 ± 6.6
Body mass index (kg/m2)	22.9 ± 1.6
Body fat (%)	18.0 ± 3.8
VO2max (mL/kg·min)	55.2 ± 5.6
HRmax (freq/min)	194.0 ± 5.6

Values are presented as mean ± standard deviation.

Experimental design and procedures

All subjects performed an incremental running test (Bruce protocol) to determine maximum oxygen uptake (VO_2max) and maximum heart rate (HR_max) by the gas analyzer (Quinton) and the heart rate monitor (Polar Electro), respectively [20]. All participants were subjected to progressive exercise on a treadmill until they were exhausted, followed by resting period. Subjects were instructed to avoid caffeinated beverages and exercise for at least 48 hours before the experiment. They were also fasted from 10:00 PM until the beginning of the exercise on the next day. At the time of the study, all subjects were transported to the laboratory, where they rested for at least 1 hour before the exercise test. After this rest period, each subject’s blood pressure was checked to screen for hypertension. And blood sample of resting venous (10 mL) was taken. Exercise was preceded by a 5-minute warm-up. After the warm-up, the subjects began the incremental running exercise on a treadmill and continued until they were exhausted. A second blood sample was taken immediately after discontinuing exercise and a third blood sample was collected after 1-hour recovery.

Preparation of human PBMCs

Human peripheral blood samples (10 mL) were obtained by venipuncture and transferred to an ethylene-diamine-tetraacetic acid (EDTA) anti-coagulant tube (Green Cross MS Corp.). Immediately after venipuncture, The PBMCs were separated as described previously [21]. We carefully loaded 10-mL whole blood samples and centrifuged them for 10 min at 1,000 g (without brakes) at room temperature [22]. The PBMCs were separated from the whole blood by density-gradient centrifugation in separating solution (GE Healthcare Bio-Sciences AB). The PBMC layer was removed with manual pipetting, washed three times in cold PBS at pH 7.4, and centrifuged for 15 min at 4°C and 2,500 g. Washed PBMCs were collected and stored at –70°C for further analyses.

Preparation of cytosolic and nuclear extracts from PBMCs

Isolated PBMCs were lysed in 200 μL of ice-cold hypotonic buffer A (10 mM HEPES [pH 7.8], 10 mM KCl, 2 mM MgCl₂, 1 mM dithiothereitol [DTT], 0.1 mM EDTA, and 0.1 mM phenylmethylsulfonyl fluoride [PMSF]). After 15-minute incubation on ice, the mixture was centrifuged for 2 minutes at 14,800 g. The supernatant was collected as the cytosolic extract for Western blot analysis. The remaining cell pellets were resuspended in 100 μL of ice-cold buffer C (50 mM HEPES [pH 7.8], 50 mM KCl, 300 mM NaCl, 0.1 mM EDTA, 1 mM DTT, 0.1 mM PMSF, and 10% glycerol) followed by incubation on ice for 20 minutes and centrifuged for 15 minutes at 14,800 g at 4°C. The supernatant was collected as a nuclear extract and stored at –70°C for an electrophoretic mobility shift assay (EMSA).

EMSA

EMSA for measuring the DNA-binding activities of NF-κB and Nrf2 was performed using a DNA protein-binding detection kit (GIBCO BRL), according to the manufacturer’s protocol. Briefly, a synthetic double-stranded oligonucleotide containing the consensus sequence of NF-_κB or Nrf2 was labeled by incubation with [γ-³²P]ATP by T4 polynucleotide kinase and purified on a Nick column (Amersham Pharmacia Biotech). The binding reaction was carried out in 25 μL of a reaction mixture containing 5 μg of incubation buffer (10 mM Tris-HCl [pH 7.5], 100 mM NaCl, 1 mM DTT, 1 mM EDTA, 4% glycerol, and 0.1 mg/mL sonicated salmon sperm DNA), 10 μg of nuclear extract, and 120,000 cpm of [γ-³²P]ATP end-labeled oligonucleotide. After 50-minute incubation at room temperature, 2 μL of 0.1% bromophenol blue was added, and the samples were electrophoresed through 6% non-denaturing polyacrylamide gel at 150 V for 2 hours. Finally, the gel was dried and exposed to an X-ray film.

Measurement of the GSH level

GSH levels were measured using a commercial kit (Bioxytech GSH-400; OxisResearch). The first step of the analysis leads to the formation of substitution products between a patented reagent and all of the mercaptans that are present in the sample. The second step specifically transforms the substitution product obtained with GSH into a chromophoric thione, which has maximal absorbance at 400 nm.

Western blot analysis

Isolated human PBMCs were washed with cold PBS and then lysed at 4°C in 500 μL of ice-cold lysis buffer (150 mM NaCl, 0.5% Triton-X 100, 50 mM Tris-HCl [pH 7.4], 20 mM ethylene glycol tetraacetic acid [EGTA], 1 mM DTT, 1 mM Na₃VO₄, and a protease inhibitor cocktail Tablet [Roche Molecular Biochemicals]) for 1 hour. The lysates were centrifuged at 14,800 g for 15 minutes. The supernatant was collected, and the total protein concentration was quantified using a bovine serum albumin (BSA) protein assay kit (Bio-Rad Laboratories). Cell lysates (40 μg) were boiled in SDS sample loading buffer for 5 min before electrophoresis on 10% SDS-polyacrylamide gel. After electrophoresis for 4 hours, proteins in the SDS-polyacrylamide gel were transferred to polyvinylidene difluoride (PVDF) membrane (Gelman Laboratory) at 300 mA for 4 hours, and the blots were blocked with 5% fat-free dry milk-PBST buffer (PBS containing 0.1% Tween-20) for 1 hour at room temperature and then washed with PBST buffer. The membranes were incubated for 4 hours at room temperature with 1:1,000 dilutions of primary antibodies for GCLC, COX-2, caspase-3, cleaved caspase-3, PARP, cleaved PARP, and HO-1. Equal lane loading was assessed using actin (Sigma-Aldrich). Blots were washed three times with PBST at 10-minute intervals, followed by incubation with 1:5,000 dilutions of the respective HRP-conjugated rabbit secondary antibodies for 1 hour and again washed in PBST three times with PBST buffer at 10-min intervals. The transferred proteins were visualized with an ECL detection kit (Amersham Pharmacia Biotech) according to the manufacturer’s instructions.

Fluorimetric assay of caspase-3 activity

Ac-DEVD-AMC (BD Bioscience) was used as a substrate for the caspase-3 assays based on the manufacturer’s instructions. A fluorescent assay kit was used to measure caspase-3 activity by monitoring the cleavage of a synthetic fluorescent substrate. In brief, cell lysates were prepared in the lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% NP-40, and 1 mM PMSF) supplemented with protease inhibitor mixture (Roche), the lysates were centrifuged at 10,000 g for 10 minutes, and the supernatants were collected with BSA as the standard. Equal amounts of proteins from each sample were reacted with the synthetic fluorescent substrates at 37°C for 1 hour and measured using a SpectraFluor F129003 (Tecan US) at excitation and emission wavelengths of 360 and 465 nm, respectively.

Statistical evaluation

Values were expressed as the mean ± standard deviation of at least three independent experiments. Statistical significance was determined by repeated ANOVA as a post-hoc comparison. P < 0.05 was considered to be statistically significant. Statistical analysis was performed using Windows SPSS version 15.0 (SPSS Inc.).

Cellular adaptive response to oxidative stress caused by exhaustive exercise

GSH levels significantly declined after exhaustive exercise but were partially restored during the 1-hour recovery period (Fig. 1A). GSH restoration was accompanied by induction of GCLC expression after exhaustive exercise (Fig. 1B).

Figure 1. Effects of exhaustive exercise and 1-hour recovery on oxidative stress in human blood cells. (A) Oxidative stress marker levels were measured using a colorimetric assay kit. (B) GCLC expression was measured using Western blot analysis with a specific antibody. Actin levels were measured to ensure equal loading of proteins. The data were quantified using image densitometric analysis. Results are means ± standard deviation; n = 16 per group. GSH, reduced glutathione; GCL, glutamate cysteine ligase; GCLC, catalytic subunit of GCL; IAE, immediately after exercise; 1h-RAE, 1-hour recovery after exercise. *P < 0.05, **P < 0.01, ***P < 0.001.

Activation of pro-inflammatory signaling upon exhaustive exercise

We also determined whether exhaustive exercise and 1 hour of recovery after exercise differentially modulated expression of COX-2 as an inflammatory response in human PBMCs. All subjects who participated in exhaustive exercise exhibited significantly elevated COX-2 expression during the 1-h recovery period (Fig. 2A). Because NF-κB is known to play a critical role in regulating induction of COX-2 expression, we also evaluated whether exhaustive exercise activated this transcription factor in nuclear extracts obtained from human PBMCs. As illustrated in Fig. 2B, NF-κB DNA-binding activity was significantly enhanced immediately after exhaustive exercise but declined during the 1-hour resting period.

Figure 2. Comparison of the inflammatory signaling pathway during exhaustive exercise and 1-hour recovery in human PBMCs. (A) Protein extracts (40 μg) were loaded onto a 10% SDS PAGE, electrophoresed, and subsequently transferred onto a polyvinylidene difluoride membrane. Western blot analysis was probed with a rabbit polyclonal COX-2 antibody. (B) The nuclear extract (10 μg) was incubated with radiolabeled oligonucleotides containing the NF-κB consensus sequence for analysis by electrophoretic mobility shift assay. (C) Cytosolic p-IκBα levels in human PBMCs. Results are means ± standard deviation; n = 16 per group. PBMC, peripheral blood mononuclear cell; IAE, immediately after exercise; 1h-RAE, 1-hour recovery after exercise; p, phospho. *P < 0.05, **P < 0.01, ***P < 0.001.

A critical event in the activation of NF-κB is its dissociation from IκBα, which requires phosphorylation and subsequent degradation of this inhibitory subunit. To determine whether the enhancement by exhaustive exercise of NF-κB DNA binding was due to increased phosphorylation of IκBα, the cytoplasmic levels of phospho (p)-IκBα were measured by western blot analysis. Consistent with transient induction of NF-κB DNA binding, the cytoplasmic levels of p-IκBα were substantially elevated upon exhaustive exercise, but completely returned to the baseline level during the 1-hour recovery period (Fig. 2C).

Induction of apoptosis by exhaustive exercise

One of the well-defined hallmarks of apoptosis is cleavage of PARP by pro-apoptotic signaling. Thus, we examined whether exhaustive exercise induced cleavage of PARP in human PBMCs. As illustrated in Fig. 3A, PARP cleavage was evident upon exhaustive exercise. However, initial induction was almost completely abolished after 1 hour of recovery. In another experiment, we determined whether exhaustive exercise induced cleavage and catalytic activity of caspase-3, a key enzyme involved in cleavage of PARP, in human PBMCs. Cleavage (Fig. 3B) and activity (Fig. 3C) of caspase-3 were both significantly elevated after exhaustive exercise. This activation of caspase-3 through proteolytic cleavage returned nearly to the baseline level during the 1-hour recovery period.

Figure 3. Exhaustive exercise-induced apoptosis in human PBMCs. (A) PARP and cleaved PARP levels in human PBMCs. (B) Caspase-3 activity (arbitrary units) was measured based on proteolytic cleavage of fluorogenic Ac-DEVD-AMC peptide as a substrate. (C) Caspase-3 and cleaved caspase-3 levels in human PBMCs. Protein extracts (40 μg) (A) and (C) were resolved by SDS PAGE and immunoblotted using specific antibodies. Results are means ± standard deviation; n = 16 per group. PBMC, peripheral blood mononuclear cell; IAE, immediately after exercise; 1h-RAE, 1-hour recovery after exercise. *P < 0.05, ***P < 0.001.

Activation of Nrf2 and induction of HO-1 by exhaustive exercise

The onset of pro-inflammatory response was accompanied by transient induction of an antioxidant response. As shown in Fig. 4A, the initial increase in HO-1 expression was followed by a decline during the 1-hour recovery period. Nrf2 is known to play a critical role in regulating the induction of HO-1 expression. In parallel with transient HO-1 induction, the Nrf2-ARE binding activity was significantly enhanced after exhaustive exercise and returned to the baseline level during 1 hour of recovery (Fig. 4B).

Figure 4. Effects of exhaustive exercise on Nrf2-driven HO-1 expression during exhaustive exercise and the 1-hour recovery period as well as resting conditions in human PBMCs. (A) HO-1 expression levels in human PBMCs. Protein levels were measured using Western blot analysis with a rabbit polyclonal HO-1 antibody. (B) Nrf2-ARE binding activity in human peripheral blood mononuclear cells. Nuclear proteins (10 μg) were resolved by electrophoretic mobility shift assay. Results are means ± standard deviation; n = 16 per group. Nrf-2, NF-E2 related factor 2; PBMC, peripheral blood mononuclear cell; HO-1, heme oxygenase-1; ARE, antioxidant response elements; IAE, immediately after exercise; 1h-RAE, 1-hour recovery after exercise. ***P < 0.001.

There is substantial evidence for the effectiveness of periodic exercise in primary and secondary prevention of various disorders including cancer and premature death [23]. However, exhaustive or strenuous exercise can cause injuries and chronic fatigue, leading to overtraining syndrome, which occurs partly due to oxidative stress via production of ROS [3,24,25]. Davies et al. [26] reported increased production of ROS in the muscle tissue of rats subjected to strenuous exercise. A study by Gomez-Cabrera et al. [27] revealed the protective role of xanthine oxidase against oxidative damage that occurs as a consequence of vigorous exercise in rats and humans. The present study clearly demonstrates that exhaustive exercise increases oxidative stress, as revealed by diminished GSH in erythrocytes. Oxidative stress is essentially an imbalance between pro- and antioxidants. Increased oxidative stress can be harmful to all cellular macromolecules, such as lipids, proteins, and DNA [28]. In contrast, transiently elevated ROS often mediate the activation of distinct intracellular signaling pathways which regulate transcription of a number of genes encoding antioxidant proteins, DNA-repair proteins, stress-regulated chaperones, and anti-apoptotic proteins [29].

NF-_κB has a pleiotropic function in regulation of many genes involved in cellular immunity and inflammatory responses [30]. In particular, NF-κB promotes expression of COX-2, a rate-limiting enzyme in the biosynthesis of prostaglandins that mediate inflammatory processes. We noted that exhaustive exercise did induce expression of COX-2 as well as NF-κB activation, as an inflammatory response in human PBMCs. Although some studies have reported a reduction in NF-κB DNA binding activity after exercise [31,32], we found that NF-κB binding activity was markedly elevated following exhaustive exercise. This finding is also in good agreement with the results of other studies demonstrating that exercise is a powerful activator of NF-κB in blood, brain, kidney, and muscle cells [33-35].

We also observed that NF-κB activation, initially induced by exhaustive exercise, almost completely returned to the baseline level during 1-hour of recovery, while COX-2 expression continuously increased during the recovery period. This sustained increase in COX-2 expression even during the recovery period may reflect the interval between transactivation of NF-κB and expression of the COX-2 gene. Considering the important role of NF-κB in inflammatory responses, the activation of this transcription factor is likely to contribute to COX-2 expression in response to exhaustive exercise-induced oxidative stress in human PBMCs. Classical regulation of NF-κB activity occurs via phosphorylation of IκBα and its subsequent ubiquitination and proteasomal degradation, which allow NF-κB to migrate to the nucleus and bind to κB regulatory elements [36]. In parallel with NF-κB DNA binding, the phosphorylation of IκBα peaked immediately after the exhaustive exercise.

It has been reported that strenuous/vigorous exercise can cause apoptotic cell death [10-12]. Caspases are recognized as important mediators of apoptosis. Several previous studies have shown that exercise causes DNA damage in leukocytes [37,38]. Using single-cell gel electrophoresis, a sensitive measurement of DNA strand damage and alkali-labile damage, appearance of leukocytes with damaged DNA after exhaustive exercise has been demonstrated [37]. Our results reveal that exhaustive exercise does induce apoptosis in human PBMCs, as evidenced by elevation of caspase-3 activity and proteolytic cleavage of caspase-3 and PARP. Although exhaustive exercise significantly induced the aforementioned hallmarks of apoptosis, there was an almost complete decline in these markers during the 1-hour recovery period. This decline in apoptosis during the resting period may be attributable to induction of antioxidant or other defense/repair mechanisms as an adaptive survival response to the oxidative and inflammatory stresses caused by exhaustive exercise.

Using the terminal deoxynucleotidyl transferase deoxyuridine triphosphate nick end-labeling assay, Mars et al. [39] demonstrated that apoptosis occurred in 63% of lymphocytes immediately after high-intensity exercise. At 24 hours after exercise, 86.2% of the cells still exhibited an apoptotic profile of DNA distribution [39]. According to the results of our present study, however, initial induction of caspase-3 activity and cleavage of PAPR were almost completely abolished during the 1-hour recovery period. This discrepancy between the two results may be due to the different exercise protocols and cell types examined. The mechanisms responsible for exercise-related apoptosis of human blood cells remain poorly understood. Alterations in the cellular redox status may influence apoptotic signaling. In numerous cells, ROS are involved as both initiators and mediators of apoptosis [40]. Treatment with antioxidants has been shown to attenuate exercise-induced DNA damage, indicative of a pro-apoptotic function of ROS [41]. In contrast to intensive exercise, moderate exercise attenuates lymphocyte apoptosis induced by oxidative stress, possibly by potentiating intracellular anti-oxidative capabilities [42]. Thus, excessive production of ROS during intensive exercise is considered responsible for a series of biochemical and physiological changes that account for adaptive response to oxidative stress [24-26].

The human body maintains distinct antioxidant defense systems against oxidative stress. Nrf2 up-regulates a battery of antioxidant enzymes and other cytoprotective proteins [21]. One of the principal stress-responsive cytoprotective proteins whose gene transcription is regulated by Nrf2 is HO-1 [18]. HO-1 has potent anti-inflammatory, anti-oxidative, and anti-apoptotic functions [14,43,44]. HO-1-mediated cytoprotection has been shown to be critical for tissues that are vulnerable to oxidative stress [45]. Notably, an acute bout of exercise has been found to increase expression of HO-1 in human lymphocytes [46]. Although Nrf2 and HO-1 are well known to have cytoprotective functions, there is a paucity of data on the induction of Nrf2 activation and HO-1 expression by exercise. Consistent with previous studies on exercise-induced expression of HO-1 in human lymphocytes and leukocytes [45-47], our study documents exhaustive exercise-related upregulation of HO-1 expression in conjunction with Nrf2 activation in human PBMCs.

HO-1 is an important stress-response enzyme induced by oxidative and inflammatory stimuli. Previous studies have shown that a competitive half-marathon [47], treadmill test until exhaustion [48], and prolonged treadmill running [45], but not a short run or eccentric exercise [47], increase HO-1 expression. Although our exercise trial was not a long-duration type, there was pronounced induction of HO-1 expression and Nrf2-ARE binding activity immediately after exercise. High-intensity exercise may disrupt the fragile balance between oxidative stress and anti-oxidative defense, and as a result, the mitochondrial respiratory chain and other metabolic pathways generate ROS [49-51]. This, in turn, provokes adaptive cellular antioxidant responses to resolve oxidative and inflammatory stresses [24,25].

In addition to HO-1, exhaustive exercise induced induction of other antioxidant enzymes, such as NQO1, SOD, and GST (Figure S1), which also constitute an important part of the cellular defense against pro-oxidative/pro-inflammatory insults. Therefore, moderate exercise may offset inflammation via ROS generation, likely by timely induction of antioxidant gene expression through activation of Nrf2-ARE signaling.

In conclusion, exhaustive exercise triggers oxidative and inflammatory responses and induces expression of cellular antioxidant enzymes, including HO-1 in human PBMCs. In addition, exhaustive exercise-induced apoptosis in human PBMCs appears to facilitate elimination of cells that cannot survive oxidative and inflammatory injuries caused by vigorous exercise. Further studies will be necessary more precisely assess the effects of exhaustive exercise on human health and disease.

Supplementary materials can be found via https://doi.org/10.15430/JCP.2023.28.1.3.

This work was supported by the grant (NRF-2010-G00134) funded by the National Research Foundation of Korea.

No potential conflicts of interest were disclosed.