Oxidative stress biomarkers and free amino acid concentrations in the blood plasma of moderately exercised horses indicate adaptive response to prolonged exercise training

Abstract

Oxidative stress caused by routine physical stressors may negatively impact the performance of equine athletes; thus, the present study identifies oxidative biomarkers in the blood plasma of exercising horses. Stock-type horses were subject to a standardized moderate-intensity exercise protocol 3 times per week for 8 wk. Exercise protocol followed NRC guidelines consisting of 30% walk, 55% trot, and 15% canter, with a target heart rate (HR) of 90 BPM. Blood plasma was collected in wk 1, 2, 7, and 8 immediately before and 0, 30, 60, and 90 min after exercise and analyzed for total antioxidant capacity (TAC), thiobarbituric acid reactive substance (TBARS), glutathione peroxidase activity (GPx), and superoxide dismutase activity (SOD). Data were analyzed as repeated measures with wk, d, time, and their interactions as fixed effects. The TAC on day 2 (0.40 mM Trolox) was 7.5% greater than on day 3 (P = 0.013). There were wk × d × time interactions for SOD, TBARS, and GPx (P < 0.001). The TBARS remained at pre-exercise baseline (d-1 wk-1; 2.7 µM malondialdehyde) for most collection times within weeks 1, 7, and 8 (P ≥ 0.058); however, TBARS increased by 0.24 to 0.41 µM on day 2 of week 2 post-exercise (P < 0.001) and remained similarly elevated on day 3 pre- and immediately post-exercise (P < 0.001). The GPx similarly remained at baseline (172.6 µM/min; P ≥ 0.621) but increased by 48.18 to 83.4 µM/min at most collection times on days 1 and 2 of week 2 (P ≤ 0.023). The SOD remained at baseline (167.2 U/ mL; P ≥ 0.055) until increasing by 11.28 to 15.61 U/mL at 30 min post-exercise on day 1, week 1 and at most collection times on day 3, week 8 (P ≤ 0.043). Amino acids with antioxidant properties such as Met, Tyr, and Trp drastically decreased from weeks 2 to 8 (P < 0.001). Met and Tyr also decreased from −60 to 90 min (P < 0.047), whereas there was no time effect on Trp concentration (P = 0.841). The current study indicates the time-dependent nature of oxidative stress concerning persistent stressors such as exercise.

Lay Summary

Performance horses are subjected to numerous stressors. These stressors may subsequently impact their overall performance. The present study measured oxidative stress biomarkers in the blood of exercising horses. Horses were moderately exercised over an 8-wk period and blood plasma was collected to measure total antioxidant capacity (TAC), thiobarbituric acid reactive substance (TBARS), glutathione peroxidase activity (GPx), and superoxide dismutase activity (SOD). Amino acid concentration was also evaluated. The TAC was greater on day 2 vs. day 3. The TBARS remained at pre-exercise (baseline) at most times except for increasing on day 2 of week 2 post-exercise. The GPx also remained at baseline for most times but increased on days 1 and 2 of week 2. The SOD remained at baseline until increasing at 30 min post-exercise on day 1, week 1 and at most collection times on day 3, week 8. Amino acids with antioxidant properties drastically decreased from weeks 2 to 8. Horses are exposed to a variety of physical stressors on a regular basis that may produce similar effects in the equine stress response. Understanding the response in the equine athlete when exposed to new stressors is crucial in determining how to prevent oxidative damage in future athletes.

The current study shows that horses exposed to increased levels of physical stress experience changes in oxidative stress biomarkers following its onset. It was also demonstrated that prolonged exposure to stress can result in adaptation of the stress response.

Introduction

Oxidative stress (OS) is a detrimental consequence of various physiological processes and contributes to a variety of diseases and disorders (Barja, 1999; Fukai and Ushio-Fukai, 2011; Chen and Zhong, 2014). Exercise increases the risk of OS in both humans and rats, leading to damages to muscle components, muscle soreness, and a decrease in force production in affected areas (Reid et al., 1993; Powers and Jackson, 2008; Steinbacher and Eckl, 2015b). Reactive oxygen/reactive nitrogen species are typically produced by the overstimulation of specific enzymes such as nitric oxide synthase and NADPH oxidase (Valko et al., 2007). Oxidative stress occurs when the body fails to neutralize excess reactive oxygen/reactive nitrogen species, causing oxidative damages to cell membranes and other compounds such as proteins, lipids, lipoproteins, and DNA (Valko et al., 2007). These damages negatively affect athletes, decreasing their performance and overall well-being (Davies et al., 1982; Sen, 1995; Reid, 2001; Ji, 2008; Sachdev and Davies, 2008). In the equine model, OS occurs after exposure to increased physical stress such as trailering and increased exercise load, which are commonly seen in the equine show and competition circuits (Niedźwiedź et al., 2013; Andriichuk et al., 2016; Smarsh and Williams, 2017).

Research in human medicine shows that continued exposure to OS can cause adaptation of the antioxidant response through increased production of antioxidants within the body, decreasing the number of free radicals and decreasing damage to body structures (de Moffarts et al., 2004; Traustadóttir et al., 2012). Such responses can be enzymatic or non-enzymatic in nature (Fisher-Wellman and Bloomer, 2009). Some non-enzymatic responses are amino acids, which are used to meet both nutritional and antioxidant requirements during exercise (Bowtell et al., 2007; Hackl et al., 2009). Therefore, the current study was designed to determine the effects of stress induced by moderate-intensity exercise on oxidative biomarkers and free amino acid concentration in blood plasma.

Materials and Methods

The current study was conducted under an approved Mississippi State University Institutional Animal Care and Use Committee protocol (#20-095).

Animals and diets

Stock-type horses (n = 6; 4 mares, 2 geldings; 6–16 ± 3 y; 455.86 ± 43.1 kg; body condition score 5.6 ± 0.60) were housed at the Mississippi State University Horse Unit in 8 × 8 m auxiliary stalls. All horses received 2% BW/d, as fed, of bermudagrass hay (Cyandon dactylon) mix (7.1% CP and 76.7% NDF on a DM basis). Nutrena 12% Stock and Stable pelleted horse grain was fed (0.5% to 0.75% BW/d) at 0830 and 1730 h to maintain a desired body condition score of 5.5 (Henneke et al., 1983). Dietary ration was fed twice daily. Body condition score was evaluated at the beginning of the study and every 2 wk thereafter.

Exercise protocol

All horses entered the trial as sedentary horses (non-exercised). Horses were stalled throughout the project except during exercise bouts. They were subjected to a 2-wk acclimation period for horses to adjust to housing, diet, and exercise, immediately followed by an 8-wk exercise period with 3 d/wk and 1 h each day. During exercise bouts, all horses were exercised at 0600 h. Horses were subject to a standardized moderate-intensity exercise procedure following the NRC guidelines (National Research Council, 2007), consisting of 30% walk, 55% trot, and 15% canter, with a target heart rate (HR) of 90 BPM for a total of 60 min per exercise bout. Acclimation exercise was conducted at 25% and 50% of the experimental intensity in weeks 1 and 2, respectively (Table 1). Heart rates were monitored using Hylofit (Hylonome LLC, Wilton, CT) HR monitors attached to the saddle girth. Monitors were placed on the left side of the horse along the heart girth with electrodes at the base of the heart girth and midway below the withers, as directed by the manufacturer. The electrodes were lubricated with ultrasound gel (Parker Labs, Fairfield, NJ) to ensure an accurate reading. Three HR monitors were used, and HRs were monitored individually to ensure that the minimum HR requirement was met for each gait (walk: 70 BPM; trot: 95 BPM; canter: 115 BPM; Table 1 and Figure 1). Monitors were alternated between horses for each exercise bout. Horses were ridden by the same rider during each exercise bout. Monitoring of gaits and heart rates were performed by the same individual each time to ensure uniformity amongst exercise bouts, riders, and horses.

Table 1.

Moderate intensity exercise protocol adapted from the National Research Council (NRC) requirements for horses

Gait	Duration (min)
Walk	4
Trot	8
Walk	3
Canter	5
Trot	8
Walk	4
Trot	8
Walk	4
Canter	5
Trot	8
Walk	3

Figure 1.

Average heart rate of horses during moderate exercise (N = 6).

Blood sample collection

Blood samples were collected before- (-60 min), immediately upon completion of (0 min), and at 30, 60, and 90 min after exercise. Blood samples were only collected during weeks 1, 2, 7, and 8, although the exercise procedure was conducted for the entire 8 wk. At each time point, blood samples were collected via jugular venipuncture into 10-mL vacutainer tubes containing potassium heparin or silicone coating for separation of blood plasma and serum (Becton, Dixon and Company, Franklin Lakes, NJ). Blood plasma was centrifuged 10 min immediately after sampling, at 2,000 × g, aliquoted into 1.5-mL microcentrifuge tubes, frozen on dry ice, and stored at −80 °C until analysis.

Oxidative biomarker analysis

Analysis of blood plasma total antioxidant capacity (TAC) was based on the ability of naturally occurring antioxidants in plasma to neutralize a free radical (Re et al., 1999). The free radical solution was created using 2,2ʹ-Azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS•) and potassium persulfate, which was diluted to an absorbance of 0.8 to 0.9 using distilled water. Plasma samples were diluted 1:10 with water before 10 µL of the diluted sample were combined in 200 μL of ABTS to measure absorbance at 734 nm. The TAC value was expressed in mM of trolox equivalence.

Plasma superoxide dismutase activity (SOD) was determined by a Superoxide Dismutase Assay Kit (Cayman Chemical, Ann Arbor, MI) previously utilized in horses (Niedzwiedz and Jaworski, 2014). Plasma was diluted by a factor of 1:5 with provided sample buffer. The diluted sample was reacted with a radical detector and xanthine oxidase. The activity was measured at absorbance 440 nm and compared with SOD standard solution and the results were expressed in U/mL. All assays were performed according to the manufacturer’s instructions within 6 mo of sample collection.

Glutathione peroxidase activity (GPx) was determined using a Glutathione Peroxidase Assay Kit (Cayman Chemical, Ann Arbor, MI) previously utilized in horses (Niedzwiedz and Jaworski, 2014). Plasma was diluted by a factor of 1:2 with the provided sample buffer. The diluted sample was then reacted with nicotinamide adenine diphosphate and GPx co-substrate. The activity was compared with glutathione (GSH) control solution and results were expressed as µM/min.

Lipid oxidation was measured by TBARS using a procedure described by Draper et al. (Draper et al., 1993). Plasma was combined with 10% trichloroacetic acid and 50 ppm of butylated hydroxytoluene to prevent further oxidation. The mixture was incubated in a 90 °C water bath for 30 min, cooled to room temperature by cold water, and centrifuged at 3,000 × g and 4 °C for 10 min. The extract was reacted with 0.02 mM TBA solution at 90 °C for 30 min, cooled to room temperature, and centrifuged at 10,000 × g at room temperature for 10 min. A volume of 0.2 mL of the supernatant was pipetted into a 96-well plate (Costar 3370; Corning Inc., Corning, NY). The absorbance was measured at 532 nm (SpectraMax iD3, Molecular Devices, San Jose, CA) and authentic malondialdehyde solutions. The TBARS value was expressed as μM of MDA.

Free amino acid analysis

Amino acids are nutritional requirements and some acts as naturally occurring antioxidants. Only plasma samples collected on day 1 of week 1 and day 3 of all weeks at −60 and 90 min were analyzed for free amino acid concentration to acquire the baseline free amino acids in blood plasma (week 1, day 1, −60 min) and the endpoint changes every week after exercise (90 min on day 3 each week). Plasma was combined with 200-µM norvaline as an internal standard and derivatized by the propyl chloroformate (Kaspar et al., 2008). The derivatives were extracted in isooctane and transferred to a 2-mL glass vial with a fixed insert (Agilent Technologies, Santa Clara, CA) for GC-MS determination. Amino acid derivatives were injected into an inlet of an Agilent 7890A GC System coupled to an Agilent 5975C inert XL MSD with triple-axis mass detector, an Agilent 7693 Series Autosampler, and a capillary column (Zebron EZ-AAA 10 m × 0.25 mm; Phenomenex, Santa Clara, CA). The inlet was operated at 250 °C and 1:15 split ratio. The helium carrier gas was at a 1 mL/min constant flow rate. The temperature of the transfer line, ion source, and quadrupole was 310, 240, and 180 °C, respectively. The oven was programmed initially at 110 °C and ramped up to 320 °C within 11 min. The MSD was operated in a SIM mode with one target and one qualitative ion being used. Amino acids were quantified by an internal calibration method using authentic standards. The amino acid concentration was expressed as µM.

Statistical analysis

Biomarker data were analyzed as a split-plot design in time (repeated measurement) using a generalized linear mixed model with wk, day, time, and their interactions as fixed effects and animal within a combination of wk × day as random effect. A similar statistical model was used for amino acid data with wk × d combination (wd), time, and wd × time as fixed effects and animal within wd as random effect. The analysis of variance was performed by the GLIMMIX procedure of SAS 9.4 (SAS Institute, Cary, NC). The selection of the appropriate covariance structure for the repeated measurement was based on three default Information Criteria calculated by SAS in the smaller-is-better format (AIC, Akaike’s Information Criteria; AICC, AIC Corrected; and BIC, Bayesian Information Criteria), resulting in a first-order autoregressive structure being used. Means, if differing, were separated by a protected t-test using the LSMEANS statement. When a three-way wk × day × time interaction was significant, all time points were only compared with the baseline values at −60 min on day 1 of week 1 to reduce the total number of pair-wise comparisons and overall experiment-wise error rate. Actual probability values were reported.

Results

Oxidative biomarkers

The main effect of day was significant (P = 0.037) for TAC values. Blood plasma TAC on day 2 were 7.5% (0.03 mM) greater than that on day 3 (P = 0.013). The TAC value did not differ between days 1 and 2 or days 1 and 3 (P = 0.450; Figure 2).

Figure 2.

Total antioxidant capacity (TAC) of blood plasma from moderately exercised stock-type horses on days 1, 2, and 3, averaged across all week and time points. All data are represented as the mean ± standard error. Means without common letters differ (P < 0.05).

A wk × d × time interaction (P < 0.001) for SOD activity was observed in exercised horses (Figure 3). Blood plasma SOD activity ranged from 138.3 to 182.8 U/mL. Horses displayed baseline SOD activity of 167.2 U/mL on day 1 of week 1 at −60 min. Throughout weeks 1 and 2, significant changes in SOD activity were observed. Blood plasma SOD activity increased from baseline to 30 min on day 1 of week 1 (P = 0.003) but dropped back to baseline active afterward (P ≥ 0.253). A decrease by 19.4 and 19.6 U/mL in SOD activity from the baseline was observed on day 2 of week 1 at 60 and 90 min following exercise. Such a decrease continued on day 3 of week 1 at all collection times, by 11.9 to 19.4 U/mL (P ≤ 0.009). This decrease in activity continued into day 1 of week 2 at −60, 0, and 60 min after exercise (P ≤ 0.022), before returning to baseline 90 min following exercise. A decrease in plasma SOD activity was also observed on day 3 of week 2 (P ≤ 0.018) as well as each day of week 7 and days 1 and 2 of week 8 (P ≤ 0.043). However, an increase in SOD activity was observed at most time points on day 3 of week 8 (P ≤ 0.029), except at 30 min.

Figure 3.

Superoxide dismutase (SOD) activity in blood plasma from stock-type horses (N = 6) 60 min before (−60 min) and 0, 30, 60, and 90 min after moderate exercised. All data are represented as the mean ± standard error. Open circles denote a decrease from baseline, and closed circles denote an increase from baseline (P ≤ 0.05).

A wk x d x time interaction (P < 0.001) for GPx activity was observed The GPx activity ranged from 107.5 to 255.9 µM/min with a baseline activity of 172.6 µM/min. The GPx activity increased by 74.7 and 63.4 µM/min at 60 and 90 min on day 1 of week 2 (P ≤ 0.003) and remained at 77.0, 83.4, and 48.2 µM/min greater than the baseline at −60, 0, and 30 min on day 2 of week 2, respectively (P ≤ 0.023) before returning to baseline at 60 min on day 2 of week 2 (Figure 4). Plasma GPx activity was either lower or at baseline level during weeks 7 and 8 (P ≥ 0.102).

Figure 4.

Glutathione peroxidase (GPx) activity in blood plasma from stock-type horses (N = 6) 60 min before (−60 min) and 0, 30, 60, and 90 min after moderate exercised. All data are represented as the mean ± standard error. Open circles denote a decrease from baseline, and closed circles denote an increase from baseline (P ≤ 0.05).

A wk × d × time interaction (P < 0.001) for TBARS was observed (Figure 5). Blood plasma TBARS values ranged from 2.8 to 4.9 µM of MDA. Horses had a baseline (−60 min on day 1 of week 1) TBARS value of 2.7 µM. Although there were variations among weeks, TBARS values mostly remained at the baseline (P > 0.05), except for a few time points in weeks 1 and 7 and most time points on days 1 and 2 of week 2. TBARS decreased by 0.5 to 0.6 µM compared to baseline throughout week 1, at 30 and 60 min on day 1, at 60 min on day 2, and at 90 min on day 3 (P ≤ 0.037). An increase of 1.2 to 2.2 µM in TBARS from baseline was observed on day 2 of week 2 at 0, 30, and 90 min (P < 0.001). Such an increase remained significant at 2.0 and 1.3 µM at −60 and 0 min on day 3 of week 2 (P ≤ 0.001). Except for day 1 of week 7 when TBARS values decreased by 0.5 to 0.6 µM until 60 min (P ≤ 0.047), TBARS values returned to baseline at most time points of weeks 7 and 8 (P ≥ 0.063).

Figure 5.

Thiobarbituric acid reactive substances (TBARS) in blood plasma from stock-type horses (N = 6) 60 min before (−60 min) and 0, 30, 60, and 90 min after moderate exercised. All data are represented as the mean ± standard error. Open circles denote a decrease from baseline, and closed circles denote an increase from baseline (P ≤ 0.05).

Free amino acids

Twenty-nine amino acids were quantified in the blood plasma of exercised horses. There was no wd × time interaction (P ≥ 0.075). Thirteen amino acids, including Ala, Val, Thr, Ser, Pro, Asn, Asp, Met, Phe, Glu, His, Tyr, Php, and Cys2 had both wd and time main effects (P ≤ 0.051). Most amino acids in this group, especially essential amino acids such as Val, Thr, and Met, showed a similar decreasing pattern from weeks 2 to 8 and from −60 to 90 min (Table 2). However, the concentration of Cys2 increased by 34% from weeks 1 to 8 and 6% from −60 to 90 min (Table 2). Thirteen other amino acids, including Sar, Gly, Aba, βAiB, allo-Ile, Ile, Tpr, Glu, Phe, Apa, Orn, Lys, and Trp, had only wd main effects (P ≤ 0.001). Most amino acids in this group, again, decreased from weeks 2 to 8, except that Orn and Lys increased by wd. Additionally, Glu decreased until week 7 but increased by week 8 (Table 2).

Table 2.

Free amino acid concentration (µM) of blood plasma from stock-type horses (N = 6), collected before (−60 min) and 90 min after moderate exercise on day 1 of week 1 and day 3 of weeks 1, 2, 7, and 8

Amino acids	Wk1d1		Wk1d3		Wk2d3		Wk7d3		Wk8d3		SE	P wd	P time	P x
	−60	90	−60	90	−60	90	−60	90	−60	90
ALA	235.50	206.76	253.57	194.95	222.42	164.84	200.34	149.32	187.14	131.47	22.29	0.019	0.001	0.960
SAR	209.81	213.86	211.40	210.16	204.22	187.26	180.85	183.21	156.16	141.17	6.31	< 0.001	0.186	0.310
GLY	723.73	768.42	727.16	717.82	669.06	596.91	531.28	522.55	457.64	413.91	46.20	< 0.001	0.544	0.772
ABA	9.37	9.73	9.09	8.65	8.18	7.76	7.67	7.06	7.76	7.08	0.33	< 0.001	0.092	0.530
VAL	268.12	257.23	285.69	254.63	274.09	244.65	252.84	233.82	215.75	195.37	15.22	0.001	0.026	0.964
βAiB	13.09	13.11	11.55	12.02	10.81	11.05	9.79	10.44	10.27	10.15	0.50	< 0.001	0.427	0.937
LEU	147.62	136.98	144.13	125.16	141.36	132.23	149.24	131.86	139.60	135.90	10.96	0.961	0.091	0.956
aILE	5.75	5.36	5.69	5.34	5.71	5.45	5.44	5.22	3.29	3.59	0.57	0.001	0.613	0.975
ILE	62.68	57.04	53.52	50.12	58.41	55.00	45.85	43.74	41.52	42.31	4.15	0.000	0.300	0.957
THR	118.82	115.06	95.11	90.78	90.66	72.16	75.62	66.65	58.95	48.05	6.26	< 0.001	0.023	0.768
SER	151.97	125.44	147.58	127.67	148.47	100.56	109.13	88.97	93.59	67.18	10.22	< 0.001	< 0.001	0.642
PRO	156.58	132.00	181.97	125.38	151.06	100.03	129.22	88.72	121.37	77.67	12.31	0.000	< 0.001	0.743
ASN	36.31	29.88	42.28	32.03	73.96	34.12	70.90	36.82	53.36	26.60	6.88	0.007	< 0.001	0.075
TPR	3.01	2.97	2.99	2.97	2.98	3.02	2.92	2.94	2.92	2.94	0.02	0.000	0.763	0.220
ASP	8.08	7.93	7.85	7.75	8.40	7.65	7.69	7.53	7.87	7.26	0.19	0.051	0.005	0.299
MET	64.37	60.78	69.00	56.39	62.67	54.80	51.17	45.05	55.04	48.45	2.53	< 0.001	< 0.001	0.494
GLU	28.07	25.58	22.17	23.36	23.44	21.96	19.65	20.99	24.13	22.03	1.08	< 0.001	0.307	0.233
PHE	136.51	124.51	122.81	109.83	100.82	96.30	102.62	100.21	97.80	96.00	6.37	< 0.001	0.101	0.841
AAA	9.39	9.96	9.47	9.94	9.65	10.10	9.53	9.94	10.01	10.41	0.21	0.083	0.001	0.994
APA	6.87	6.89	6.79	6.87	6.80	6.79	6.74	6.77	6.79	6.77	0.02	< 0.001	0.184	0.373
GLN	284.09	244.54	210.59	241.29	242.61	185.48	210.45	197.08	196.75	134.23	20.37	0.000	0.032	0.152
ORN	61.32	56.93	68.00	55.42	69.69	67.16	82.09	67.56	79.51	85.40	5.72	0.001	0.126	0.398
GPR	7.73	7.74	7.74	7.73	7.74	7.73	7.73	7.72	7.76	6.44	0.41	0.434	0.299	0.401
LYS	86.98	83.48	78.87	72.83	84.86	83.73	101.29	91.22	110.22	116.73	7.34	< 0.001	0.543	0.839
HIS	188.52	158.20	194.48	162.34	175.77	154.21	160.89	144.53	160.10	148.97	9.50	0.034	0.001	0.775
TYR	134.17	120.26	125.46	114.62	101.47	93.29	103.55	98.56	85.16	87.08	5.58	< 0.001	0.047	0.671
PHP	13.79	13.95	13.28	13.79	13.29	13.39	13.04	13.34	13.09	13.16	0.18	0.001	0.048	0.730
TRP	157.68	155.82	151.83	145.36	127.40	120.72	114.17	118.04	92.44	98.77	7.53	< 0.001	0.841	0.870
CYS2	63.71	72.10	83.73	89.16	106.35	116.35	108.37	114.33	116.40	114.72	3.60	< 0.001	0.017	0.547

Discussion

This study aimed to identify patterns of expression of acute and chronic OS in performance horses when exposed to extended periods of exercise training, thereby gaining insight into how the body adapts to OS exposure immediately after each exercise bout as well as over an extended period. The oxidative responses to exercise in horses vary among numerous forms of exercise (such as treadmill, race, endurance, and jumping), as well as other parameters such as breed and fitness level. There is evidence that training has a positive effect on the oxidant/antioxidant equilibrium in humans and horses (de Moffarts et al., 2004; Traustadóttir et al., 2012), but there is little information about the effects of prolonged moderate-intensity exercise training on unfit horses. This information helps equine trainers and riders better understand the conditioning process for incoming or recovering performance horses, and aid in future research in the development of techniques for reducing OS created by numerous stressors experienced by equine athletes.

The data in the current study revealed that OS occurred; however, the antioxidant system in the animal’s body responded. During OS, free radicals attack lipid molecules first. As a result, lipid oxidation products such as malondialdehyde increase. Blood plasma concentrations of TBARS stabilized at baseline following an initial response to exercise within the first week of sampling. The initial decrease in TBARS concentrations following exercise during the first week is consistent with the previous findings of an immediate response to a single bout of exercise (Andriichuk et al., 2016). However, these results can only be compared to the first day of exercise in the current study. A similar study on trained and untrained yearlings had similar lipid oxidation results in the untrained yearling group (Smarsh and Williams, 2017), as demonstrated in the current study. The untrained yearling group had the largest spike of plasma MDA concentration during the second wk of exercise, which is consistent with the current findings. This spike in week 2 could be attributed to a gradual buildup of peroxides through lipid peroxidation after several exercise bouts before the decomposition of peroxides to secondary oxidation products, such as MDA. Although these authors reported an increase in MDA in the 7th week of exercise, horses in the current study exhibited a decrease in TBARS values on day 1 of week 7 with no significant changes for the remainder of week 7 or 8. These differences may be due to dissimilarities in exercise protocol, as Smarsh and Williams (2017) used a graded exercise test and adjusted throughout the program. Exercise protocols varied in time and intensity and were adjusted multiple times throughout their project. These adjustments might not allow the horses to adapt to the administered exercise as in the current study.

The body responds to OS by first quenching free radicals, such as H₂O₂ through the activity of SOD and GPx. Small molecules with antioxidant capacity such as Met or Trp—the precursor of melatonin—are also recruited to join the antioxidant system. The current study indicated that although TBARS values, indicative of lipid oxidation, peaked early in week 2, the antioxidant system (TAC, GPx, SOD, Met, and Trp) responded to OS events much earlier. This response continued throughout the experimental period, especially at TBARS peaks (GPx activity, loss of Met and Trp), showcasing the horse’s ability to adapt to OS when exposed to exercise training. As previously mentioned, TAC increased on day 2 across every week and returned to the baseline value on day 3. The increase in TAC corresponded to the increase in TBARS and GPx in the same period. Increased TAC was most likely a response to the OS events in the body caused by multiple days of exercise. The reduction back to baseline on day 3 indicates recovery from the OS overload experienced on day 2 and may be evidence of the horses adjusting to the effects of exercise at the end of each week. Andriichuk et al. (2016), however, found no change in blood plasma TAC in Ukrainian warmblood and Holsteiner horses after undergoing an exercise test. TAC responses may require an incubation period during which more obvious oxidation occurs, providing feedback to the synthesis and recruitment of antioxidants.

The SOD and GPx activity acts almost in a synergistic manner. The SOD activity peaked in the first and last weeks when the GPx activity subsided. The GPx activity peaked only in week 2 when SOD activity remained at the baseline. There are conflicting reports on the behavior of SOD and GPx in exercising horses. Multiple studies have indicated that elevated antioxidant activity is common following exercise. Both Williams and Carlucci (2006) and Lamprecht and Williams (2012) reporting an increase in antioxidant activity. These authors investigated the OS response to a single exercise bout in horses, therefore, did not provide insights into the effects of training on the adaptation of the OS response. De Moffarts et al. (2004) implemented an 8-wk interval training program following an initial 4-wk period of adaptation to exercise and observed a 100% increase in SOD plasma concentrations over a 12-wk period but no response within 60 min of a single exercise bout as reported by the previously mentioned authors (Williams and Carlucci, 2006; Lamprecht and Williams, 2012). Others documented no difference (Marlin et al., 2002) or a decrease in antioxidant activity following exercise (Ono et al., 1990; Hargreaves et al., 2002). These authors, however, studied horses fit for endurance racing at the time of sample collection. This could contribute to the contrasting results, as these horses would most likely be adapted to the high-intensity exercise expected in competition. These horses might already be adapted to an increase in OS associated with an exercise routine, in contrast to untrained or unfit horses in the current study.

The superoxide radical is one of the most prevalent and damaging free radicals found within the body. Superoxide dismutase catalyzes the dismutation of superoxide radicals to hydrogen peroxide and oxygen (Powers et al., 1999; Halliwell and Gutteridge, 2015), whereas GPx catalyzes the reduction of hydrogen peroxide to water and oxygen. The SOD activity in the current study increased immediately on day 1 of week 1, indicative of an immediate increase in superoxide radicals within the body, typical of OS events. This phenomenon is consistent with the literature (de Moffarts et al., 2004; Williams and Carlucci, 2006; Lamprecht and Williams, 2012; Andriichuk et al., 2016), although a large portion of the literature on SOD activity in horses is reported for one exercise bout (de Moffarts et al., 2004; Williams and Carlucci, 2006; Lamprecht and Williams, 2012; Andriichuk et al., 2016). Glutathione peroxidase manages hydrogen peroxide homeostasis following the breakdown of the superoxide anion by SOD by using reduced glutathione to reduce hydrogen peroxide resulting in oxidized glutathione and water (Fukai and Ushio-Fukai, 2011). Glutathione reductase is responsible for reducing oxidized glutathione to reduced glutathione, ensuring an adequate level for GPx activity (Austin et al., 1988). Therefore, a peak of GPx activity in week 2 coincided well with a SOD activity peak in week 1. These findings are consistent with research on equine exercise (Balogh et al., 2001; de Moffarts et al., 2004), but is inconsistent with several reports in human medicine (Leeuwenburgh et al., 1994; Somani et al., 1995; Deaton and Marlin, 2003), which show increased blood plasma GPx activity after prolonged exercise training. The return of plasma GPx activity to baseline possibly indicated horses had adapted to the exercise protocol and the recruitment of other antioxidants.

Evidence of improvement to the body’s response to oxidative damage has been reported. This is supported by the findings within the current study, as lipid peroxidation levels remain mostly at baseline throughout the last 2 wk of sampling, following an extended period of physical training. Stabilization of TBARS levels and GPx activity in the final 2 wk of exercise training indicates horses were adapted to the exercise protocol. The SOD activity, similarly, only peaked at the last day of week 8. In addition to immediate responses by the redox enzymatic system, small molecules such as amino acids also possess antioxidant activity (Xu et al., 2017).

Amino acids with electron-rich aromatic rings such as Trp and Tyr, or those with sulfur atoms in their side chains, such as Met, are easily oxidized, therefore, act as excellent antioxidants (Xu et al., 2017). Both Trp and Met as well as Tyr drastically decreased throughout the current study. Methionine started decreasing in week 7 but Trp had already decreased since week 2. By week 8, Trp decreased by 40%. These findings agree with Hackl et al. (2009) who found that both Met and Trp decreased after exercise. These authors also reported that Trp decreased more as exercise intensity increased. Other major amino acids with stable alkyl or phenyl groups in their side chains such as Ala, Val, Thr, Ser, Asp, Asn, Gly, Gln, Phe, and Pro, although not being susceptible to oxidation (Xu et al., 2017), also decreased as exercise progressed. These amino acids might be deaminated to provide intermediate metabolites for the tricarboxylic acid cycle (Wagenmakers, 1998; Bowtell et al., 2007).

The current study shows that horses exposed to increased levels of physical stress experience changes in OS biomarkers following its onset. It was also demonstrated that prolonged exposure to stress can result in adaptation of the stress response. This study used exercise as a model to create physical stress; however, horses are exposed to a variety of physical stressors on a regular basis that may produce similar effects in the equine OS response. Understanding the initial and extended response in the equine athlete when exposed to new stressors is crucial in determining how to prevent oxidative damage in future athletes. Changes in TAC, SOD, GPx, and TBARS and in almost all free amino acids especially those with antioxidant capacity such as Met and Trp, indicate that there should be an emphasis on nutritional and antioxidant requirements during these OS events. Further research identifying methods to reduce OS is necessary and has the potential to improve performance as well as the overall well-being of the equine athlete.

Glossary

Abbreviations

BPM

beats per minute

GPx

glutathione peroxidase

GSH

glutathione

HR

heart rate

MDA

malondialdehyde

OS

oxidative stress

TAC

total antioxidant capacity

TBARS

thiobarbituric acid reactive substances

SOD

superoxide dismutase

Conflict of Interest Statement

The authors have no conflicts of interest to disclose.