Anticipation of subsequent demanding exercise increases the expression of haem oxygenase-1 mRNA in human lymphocytes

Abstract

Oxidative stress induces the expression of the cytoprotective and anti-inflammatory protein haem oxygenase-1 (HO-1). In the present investigation, we show that anticipation of subsequent exercise elevates the expression of HO-1 mRNA in lymphocytes. A between-groups comparison of HO-1 mRNA expression in subjects about to complete a half marathon race vs. subjects who were asked to sit quietly in the laboratory showed an elevated expression of HO-1 mRNA prior to exercise (2.6-fold higher in subjects prior to the half marathon, P < 0.01). This observation led us to examine whether anticipation of subsequent exercise leads to differences in lymphocyte HO-1 mRNA expression within the same subjects. In a second experiment, the same individuals completed two trials, one exercise and one rest, approximately 2 weeks apart in a randomised cross-over design. Lymphocyte HO-1 mRNA expression was greater prior to exercise (1.4 ± 0.3-fold higher in the exercise trial, P < 0.05). These results suggest that knowledge of subsequent demanding exercise may lead to an anticipatory induction of HO-1 mRNA. We tentatively propose that this process has evolved to prepare lymphocytes for subsequent exercise-induced oxidative stress although the mechanism remains to be elucidated.

• Cortisol
• free radicals
• gene expression
• lightcycler
• oxidative stress

Introduction

Haem oxygenase-1 (HO-1) induction is a general response to oxidative stress (Keyse and Tyrrell 1989). The expression of HO-1 protein is controlled at the level of transcription (Keyse et al. 1990). HO-1 mRNA accumulates in many tissues and species and it has been proposed that HO-1 mRNA expression may be a useful marker of cellular oxidative stress (Tyrrell and Basu-Modak 1994). Several studies have shown that lymphocytes have the ability to up-regulate HO-1 in response to oxidative stress. For example, H₂O₂ treatment, hyperbaric oxygen exposure, exercise and heavy metals all induce lymphocyte HO-1 (Ryter et al. 2006). Several investigations have reported that prior up-regulation of HO-1 provides enhanced protection to cells exposed to a subsequent source of oxidative stress (Vile et al. 1994; Rothfuss et al. 2001; McArdle et al. 2004). Anxiety is known to induce oxidative stress (Irie et al. 2001; Sivonova et al. 2004; Hovatta et al. 2005) and our preliminary unpublished observations indicated that knowledge of participation in subsequent demanding exercise increases the expression of HO-1 mRNA in lymphocytes. In order to test this hypothesis, we conducted two independent experiments which show that knowledge of subsequent demanding exercise leads to an anticipatory induction of HO-1 mRNA.

Methods

Healthy volunteers provided written informed consent for each investigation, which had been approved by the local Ethics Committee. In the first investigation, lymphocyte HO-1 mRNA expression was compared in two groups of subjects; one group (five males and three females) was about to perform a half marathon race (mean ± SD age, height, body mass and body mass index (BMI) were 24 ± 2 years, 173 ± 9 cm, 63.4 ± 7.2 kg, 21.1 ± 1.9 kg/m²; N = 8) and the other group (eight males) were asked to rest sitting in the laboratory for the same amount of time (mean age, height, body mass and BMI were 21 ± 1 year, 179 ± 1 cm, 73.9 ± 2.4 kg, 23.1 ± 2.0 kg/m²; N = 8). The half marathon runners reported that they undertook approximately 8.3 ± 7.9 h exercise training per week compared to 5.4 ± 1.5 h for the resting subjects). In a second investigation, in eight different male subjects (mean age, height, body mass, BMI and reported activity levels were 25 ± 4 years, 180 ± 6 cm, 78.8 ± 8.9 kg, 24.3 ± 2.3 kg/m² and 3 ± 1 h exercise per week), we investigated the within-subject expression of HO-1 mRNA during a laboratory-based rest and an exercise trial in a randomised cross-over design.

All subjects completed preliminary exercise tests on a treadmill to determine maximal oxygen uptake ( max) as a measure of fitness as described elsewhere (Thompson et al. 2005). Mean max was 61 ± 7 ml/kg/min in the half marathon runners and 64 ± 2 ml/kg/min in the subjects who came to rest in the laboratory for the equivalent amount of time. In the second cross-over design study, mean max was 58 ± 6 ml/kg/min.

In the between-groups comparison, both groups of subjects arrived at the laboratory in the morning at around the same time of day (between 0800 and 0900 h) following an overnight fast and having refrained from demanding exercise for at least 48 h. Subjects competing in the half marathon attended the laboratory on the morning of the race and subjects completing the rest trial attended the laboratory at approximately the same time of day for a similar amount of time on a different occasion. Subjects recruited for the rest trial knew that they would be asked to sit quietly in the laboratory.

In the within-subject comparison, subjects completed one exercise trial and one rest trial in a randomised order approximately 2 weeks apart. Subjects knew which trial they would be performing on each occasion. Subjects refrained from strenuous physical activity for at least 48 h before each trial. Subjects completed a weighed food and fluid record for three days prior to their first trial and were asked to repeat this diet for the same period prior to their second trial. Subjects were familiarised with the exercise protocol for 15 min approximately seven days prior to the exercise trial. The demanding exercise consisted of four, 15 min blocks of running on a treadmill comprised of repeated cycles of activity equivalent to 65, 85 and 100% max. This protocol alternated between 65% (3 min) and 85% (2 min) max with the last 2 min of each 15 min block being performed at 100% max. The average intensity of this protocol was 75% max. For the resting trial, subjects sat quietly in the laboratory for 60 min.

On the morning of a rest or exercise trial, subjects reported to the laboratory between 0700 and 0900 h following an overnight fast. Subjects were asked to adopt a supine position and a baseline venous blood sample was taken from a forearm vein. 25 ml of whole blood was collected by venepuncture and used to harvest lymphocytes and obtain plasma as described previously (Thompson et al. 2005). Lymphocytes were lysed in TRIzol, with the volume added dependent on cell concentration (1 ml per 5 × 10⁶ cells) and then frozen at − 70°C until subsequent analysis.

A two-step RT-PCR approach using Superscript first-strand synthesis kit was used to analyze HO-1 mRNA accumulation as described previously (Basu-Modak et al. 2003). HO-1 mRNA was expressed relative to GAPDH mRNA. Each subject's baseline exercise and rest trial samples were analyzed in one run for HO-1 mRNA and a separate run for GAPDH mRNA. Plasma cortisol was determined by a commercially available ELISA (DRG Diagnostics, Germany) using a microplate reader (Anthos HT-III, Labtech International Ltd, Ringmer, East Sussex).

An independent Student's t-test was used to determine whether HO-1 mRNA and cortisol were different in the subjects about to complete the half marathon or rest trials. A paired t-test was used for the within-subject comparison. In both studies, the data represent the mean ± SD. Significance was accepted at the 5% level.

Results

The baseline expression of lymphocyte HO-1 mRNA in subjects taking part in a half marathon race was 2.6-fold greater than in subjects who reported to the laboratory for a rest trial (Table I). In order to confirm that the increase in lymphocyte HO-1 mRNA expression was due to knowledge of subsequent exercise and not just inherent differences between individuals or differences due to lifestyle (e.g. training), we assessed lymphocyte HO-1 mRNA in a different group of individuals on two separate occasions (prior to exercise and rest trials). Lymphocyte HO-1 mRNA expression was greater in the exercise trial (1.4 ± 0.3-fold greater than rest; Table I). In seven out of the eight subjects, HO-1 mRNA expression was greater in the exercise trial than in the rest trial (data not shown). The trial order was randomised and there was no significant difference by trial order (data not shown). In both investigations, there was no difference in the baseline blood lymphocyte count (data not shown).

Table I.  Summary of data from the between-subjects comparison and the within-subjects comparison of basal HO-1 mRNA expression in human lymphocytes and plasma cortisol concentration.

Comparison	Lymphocyte HO-1 mRNA (Au)	Plasma cortisol concentration (nmol/l)
Between-subjects
Half marathon	146 ± 53*	551 ± 160**
Rest	57 ± 24	396 ± 96
Within-subjects
Exercise	72 ± 21***	435 ± 111
Rest	56 ± 20	431 ± 92

Values are group means ± SD (N = 8 in each group). HO-1 mRNA is expressed relative to GAPDH in arbitrary units (Au). *Denotes a significant difference between the half marathon runners and resting subjects in HO-1 mRNA levels (t (14) = 4.312; P = 0.001; P < 0.01). **Denotes a significant difference between half marathon runners and resting subjects in plasma cortisol concentration (t (14) = 2.371; P = 0.033; P < 0.05). ***Denotes a significant difference between exercise and rest trials in HO-1 mRNA levels (t (7) = − 2.470; P = 0.043; P < 0.05).

Plasma cortisol concentration was measured in both the between-subject comparison and the within-subject experiment. Subjects about to take part in a half marathon race had a significantly higher concentration of plasma cortisol, when compared to those subjects asked to rest in the laboratory (Table I). In the within-subject comparison, there was no difference in plasma cortisol concentration between exercise and rest trials (Table I).

Discussion

In the present study, both a between-subjects and within-subjects comparison of lymphocyte HO-1 mRNA expression showed that prior to performing either competitive or laboratory-based exercise the expression of HO-1 mRNA was significantly increased relative to resting conditions.

Collectively, these results suggest that the knowledge and anticipation of participation in demanding exercise increases the expression of the oxidant-responsive gene HO-1 in lymphocytes. Several earlier investigations suggest that psychological stress leads to oxidative stress (Irie et al. 2001; Sivonova et al. 2004; Hovatta et al. 2005). Although the precise mechanisms remain to be determined, it is possible that these changes were induced by the anxiety associated with preparation for demanding exercise. For example, catecholamines induce HO-1 mRNA accumulation and HO-1 protein (Schmidt et al. 1999; Berger et al. 2000; Khorchid et al. 2002). Interestingly, the increase in HO-1 expression prior to a competitive half marathon was greater than the increase observed prior to laboratory-based exercise. Plasma cortisol was also elevated prior to a competitive half marathon, whereas it was not affected by laboratory-based exercise. It is possible that anticipation of participation in a competitive event may provoke greater stress hormone secretion and subsequent oxidative stress than non-competitive exercise although this speculative suggestion requires further investigation. Because, the half-life of HO-1 mRNA is measured in hours (Keyse et al. 1990), whereas the half-life of cortisol is measured in minutes (Kraan et al. 1998) we cannot preclude the possibility that cortisol and other stress hormones were different at some time point prior to laboratory-based exercise but not at the time of sampling. In addition, it should be acknowledged that although HO-1 expression is regulated at the level of transcription (Keyse et al. 1990), without direct measurement of HO-1 protein it is not possible to state whether the observed changes in HO-1 mRNA would have produced a greater HO-1 protein content. This does not undermine the main observation that the oxidant-responsive gene HO-1 was up-regulated by anticipation of subsequent demanding exercise.

The subjects about to complete a half marathon had been training specifically for this race in the preceding months and it is possible that this, or some other intra-individual difference, may explain the findings. For example, it is possible that the greater lymphocyte HO-1 mRNA expression in these individuals is related to training-induced oxidative stress in the weeks and months preceding the competitive event. This adaptive response to regular exercise has been shown in skeletal muscle with a 2-fold increase in HO-1 mRNA found in an exercised leg when compared to a control leg following five days of training (Pilegaard et al. 2000). However, it is noteworthy that the subjects recruited for the rest trial were also habitually active and in fact max was higher in these individuals than in the half marathon runners and so fitness and training status, which were not different between-groups, are unlikely to explain the observed differences prior to a half marathon. The difference reported in the subsequent crossover experiment confirms that knowledge of subsequent demanding exercise does lead to an increase in HO-1 mRNA. We tentatively suggest that the different magnitude of the effect reported in the two experiments is explained by the anxiety about competitive exercise being greater than that associated with laboratory-based exercise and perhaps anxiety is in someway proportional to the perceived magnitude of the subsequent exercise challenge.

In summary, we observed a significant increase in the expression of HO-1 mRNA in lymphocytes relative to resting conditions prior to performing either competitive or laboratory-based exercise. We propose that this anticipatory induction of HO-1 is a mechanism that has evolved in order to help the subsequent transition to participation in demanding exercise and associated exercise-induced oxidative stress.