By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Reasons for performing study: Intense exercise in horses induces an increase of creatine kinase (CK) and stimulation of neutrophils which release the strong oxidant enzyme, myeloperoxidase (MPO) into the blood. It is not known whether active MPO is found in equine muscles and whether oxidant activity of neutrophils could affect muscular tissues and mitochondrial activity.
Objectives: Specific immuno-extraction followed by enzymatic detection (SIEFED) methods will be employed for the first time to assess both the normal range of MPO and mitochondrial complex I (MCI) activities in equine muscular microbiopsies and to study the variation of these activities induced by endurance races.
Materials and methods: Forty-six microbiopsies were taken from 8 endurance Arabian horses (age: 10 ± 2 years) in the triceps brachii (n = 23) or in the gluteus medius muscle (n = 23). Myeloperoxidase and MCI activities were measured in muscle extracts by enzyme immunocapture assays or SIEFED methods. Further, 7 endurance horses were sampled in the triceps brachii muscle before (T0) and after (T1) a 120 km endurance race (mean speed: 15.4 ± 1.4 km/h).
Results: The 46 microbiopsies from 8 horses revealed mean values for active MPO concentration and MCI activity of 21 ± 14 ng/mg proteins and 0.0172 ± 0.0066 mOD/min/μg proteins, respectively. No significant difference was observed between the 2 muscles. In 3 out of the 7 horses sampled after exercise, the 120 km endurance race induced a severe increase of muscle MPO activity (+118 ± 45% vs. T0), a large decrease of MCI activity (−63 ± 18% vs. T0) associated with a high mean plasma CK level (4642 ± 658 iu/l). In the 4 remaining horses, the 120 km endurance race did not modify the MPO and MCI activities and moderately increased the plasma CK level.
Conclusions: Preliminary observations showed a possible link between MPO activity and mitochondrial functions.
Studies in horses have shown that intense exercise induces a stimulation and degranulation of polymorphonuclear neutrophils which release myeloperoxidase (MPO) and elastase, 2 key enzymes involved in inflammation (Art et al. 2006; Lejeune et al. 2010). The plasma concentrations of these enzymes measured after an endurance race were significantly higher than the values reported before the race. Some of these high values were encountered in acute inflammation in horses e.g. during gastrointestinal diseases or laminitis (Grulke et al. 2008; de la Rebière et al. 2010).
Several studies have demonstrated that intense exercise in man and animals results in neutrophil infiltration into skeletal muscles (Tiidus et al. 2001; Morozov et al. 2006). However, it appears that the methods generally used to detect neutrophil infiltration, such as microscopy, MPO biochemical assay and antibody staining, have been equivocal in demonstrating the presence of neutrophils in muscle following exercise (Schneider and Tiidus 2007).
Recently, Serteyn et al. (2010) measured an increase in total MPO and elastase concentrations after an endurance horse race suggesting a stimulation of neutrophils or an infiltration of the enzymes inside the muscle during intense exercise. Furthermore, the increase of total MPO and elastase contents were correlated to the increase of the plasma creatine kinase (CK) levels suggesting a contribution of these enzymes to the muscle damage(Serteyn et al. 2010). However, the ELISA techniques, which use specific anti-MPO or anti-elastase antibodies, bind and detect the total concentration of theses enzymes, without making the distinction between their active and inactive forms. To determine the true activity of the enzyme, it is necessary to add an adequate chromogenic or fluorogenic substrate to the sample implying the risk that they could react with other components of the sample.
Myeloperoxidase is an enzyme particularly studied during an inflammatory response for its dual activity of peroxidase and chlorination (Deby-Dupont et al. 1999; Klebanoff 2005). Myeloperoxidase uses H2O2 to produce strong oxidant molecules (e.g. HOCl, •NO2) that can participate in secondary oxidation, chlorination and nitration reactions (Davies et al. 2008). The primary role of MPO is the destruction of microorganisms inside the neutrophils (Borregaard and Cowland 1997) but it can exert deleterious oxidant activities on neighbouring cells and tissues when released in the extracellular milieu by dying or highly stimulated neutrophils in pathological conditions of acute and chronic inflammations (Deby-Dupont et al. 1999; Klebanoff 2005). Therefore, an increase of active MPO in muscular tissues or inside muscular cells could contribute to the exacerbation of exercise-induced muscle damage, particularly if the enzyme finds H2O2, the substrate essential for its activity. It is now well established that the mitochondrial dysfunction during intense exercise amplifies the production of ROS including H2O2 (Powers and Jackson 2008).
The first objective of our study was to determine the basal content of active MPO in 2 types of muscle in horse: the triceps brachii (TB) and gluteus medius (GM). The GM is a major propulsive muscle that contains approximately 70% of type II fibres in endurance breeds (Lopez-Rivero et al. 1992) and the most frequently biopsied muscle (Rivero 1996 and Votion et al. 2007). In this study, the TB muscle was also selected for its mixed composition of types I and II fibres due to its dual role in posture and locomotion and for its easy accessibility and safety during sampling (van den Hoven et al. 1985).
The second objective of our work was to measure the variation of MPO concentration in the TB after an endurance race of 120 km. Simultaneously, we measured the plasma concentration of CK as a marker of muscular cell alteration and activity of the mitochondrial complex I (MCI or NADH-ubiquinone oxidoreductase). Mitochondrial complex I is the primary enzyme complex in the mitochondrial respiratory chain and has been reported to exercise a major control over the oxidative phosphorylation. Moreover, MCI is the most obvious potential target for the ROS when they are produced in muscular cells (Paradies et al. 2004).
We determined the concentration of active MPO and measured the activity of MCI in extracts of muscular microbiopsies by applying 2 original immunological techniques, which measure specifically the activity of the targeted enzyme after its immunocapture from biological fluids or tissues extracts (Franck et al. 2006; Nadanaciva et al. 2007). These techniques were applied for the first time to extracts of equine muscular microbiopsies. Furthermore, both enzymes were assayed on the same extract.
Materials and methods
Horses, blood and microbiopsy samplings and analysis
Horses and study protocol: The study was performed on 8 horses (3 mares and 5 geldings, pure-bred and crossbred Arabians, age: 10 ± 2 years, 418 ± 22 kg) which received the same standardised food without additional electrolytes. All the horses underwent a regular endurance-training programme for 7 consecutive months. Microbiopsies (n = 46) were taken from these 8 endurance horses in the TB (n = 23) or in the GM (n = 23) during a normal training condition period of 10 weeks.
Among these horses, 7 horses participated to a 120 km endurance race event from the ‘Fédération Equestre Internationale’ (mean speed: 15.4 ± 1.4 km/h) and finished the race. In these horses, microbiopsies were sampled in the TB 24 h before (T0) and 2 h after (T1) the race and blood samples for determination of plasma CK activities were taken at the same timing.
Sampling of the microbiopsies: Muscle triceps brachii (TB) and gluteus medius (GM) samples (± 20 mg) were obtained by needle biopsy after administration of local mepivacaine anaesthesia (scandicaine 2%)1 using a 14 gauge biopsy needle (Pro-Mag ultra biopsy needle)2 mounted on an automatic instrument (Pro-Mag ultra biopsy instrument)2. The sampling depth was fixed at 5 cm depth for both muscles. For the TB, muscle biopsy specimens were obtained at a standardised site (Votion et al. 2010) determined as the intersection between a vertical line raised from the tricipital crest and a line running from the point of the shoulder to the elbow. For the GM, the biopsy was made in the dorsal compartment, one third of the distance between the tuber sacrale and the tuber coxae.
Extraction of muscular microbiopsies: Muscular microbiopsies (20 mg) were frozen in liquid nitrogen immediately after sampling and stored at −80°C until use. Microbiopsies were dissected in 20 mmol/l cold PBS buffer, pH 7.4 using forceps with sharp tips to separate muscular fibres further homogenised with a 2.5 ml glass potter homogeniser in 500 µl ice cold 20 mmol/l PBS buffer, pH 7.4. After homogenisation, the sample protein concentration was measured by using bicinchoninic acid assay3. The concentration of the sample was adjusted to 5.5 mg/ml with PBS and then a detergent extraction performed by adding to the sample 1/10 volume of the detergent solution supplied with the Complex I Enzyme Activity Microplate Assay Kit4. The tubes were mixed and place on ice for 30 min before centrifugation at 14,000 g for 15 min at 4°C. The supernatants were aliquoted and frozen at −80°C.
Venous blood sampling and creatine kinase analysis: Venous blood was collected from the jugular vein using EDTA vacuum tubes (Vacuette)5. Within 20 min after sampling, the blood was centrifuged at 1000 g for 10 min at room temperature. Plasma was kept frozen in small aliquots at −80°C until analysis. CK levels were measured in plasma using a diagnostic reagent with N-acetylcysteine (CK-NAC, FS, DiaSys)6 for the quantitative in vitro determination of CK in serum or plasma on photometric systems.
Several conditions were tested (choice of extraction buffer, extraction volume, protein concentration, dilution factor) to obtain an optimal response both in the range of the calibration curve made for the measurement of active MPO or in the linear slope corresponding to the rate of mitochondrial complex I activity. Reliability of the assays was evaluated by the intra- and inter-coefficients of variation determined after the measurement of the activities in microbiopsies sampled in 8 TB or in 8 GM of horses received at the Equine Clinic of the University of Liege.
Measurement of MPO activity by SIEFED assay: Immediately prior to performing the specific immuno-extraction followed by enzymatic detection (SIEFED) assay, the supernatants obtained by extraction of microbiopsies were thawed and diluted 1:5 in a dilution buffer prepared with 20 mmol/l PBS solution (pH 7.4) containing bovine serum albumin (5 g/l) and 0.1% Tween 20.
The SIEFED method was developed to measure specifically equine (Franck et al. 2006) or human (Franck et al. 2009) MPO activity in biological fluids, tissue homogenates or aqueous solutions. The SIEFED method consists of the extraction of MPO from biological fluids or solutions by immobilised (microplate-coated) specific antibodies, followed by a washing (to eliminate all molecules except the antibody-enzyme complexes bound to the microplate) and the revelation of the enzyme activity with a detection system composed of a sensitive fluorogenic substrate (Amplex red)7 and a nitrite-based amplifying system.
The primary antibody (rabbit anti-MPO IgG) was coated onto microplate wells (Combiplate 8 EB)8. Equine MPO standards (ranging from 1.56–50 ng/ml and corresponding to 0.25–8 mu/mlin reference to purified equine MPO) and diluted samples containing MPO (200 µl) were added to the microplate and incubated for 2 h at 37°C. After 3 washings, the peroxidase activity of MPO was detected by adding 100 µl of 40 µmol Amplex red7 (10-acetyl-3, 7-dihydroxyphenoxazine), freshly prepared in phosphate buffer (50 mmol/l) at pH 7.5 containing 10 µmol/lH2O2 and 10 mmol/l nitrite. Fluorescence was measured during 30 min at 37°C with a Fluoroscan Ascent8 with excitation and emission wavelengths set at 544 and 590 nm, respectively. Controls (blanks) were prepared with the dilution buffer and each sample run in duplicate. The fluorescence value was directly proportional to the quantity of active MPO in the sample. The active MPO was referred to the total protein concentration measured in the supernatant of muscular microbiopsy extracts using the bicinchoninic acid assay3 and expressed in ng active MPO/mg proteins.
Measurement of the Complex I activity by an immunocapture assay: To measure the activity of Mitochondrial Complex I (NADH dehydrogenase, E.C. 220.127.116.11) we used the microplate assay for human complex I activity4. The technique consisted of immunocapturing the MCI from isolated or crude mitochondria, whole tissues or cell lysates, by immobilised (microplate-coated) specific antibodies followed by a washing and measurement of the activity of MCI by the oxidation of NADH to NAD+ with a simultaneous reduction of a dye which led to an absorbance increase at 450 nm. After determination of the protein content of supernatants by the bicinchoninic acid assay3, the concentration of the samples were adjusted to reach 50 µg/200 µl per well. After incubation (3 h at room temperature), the wells were washed 3 times with the buffer supplied in the kit. The assay solution containing NADH and the dye reagent were added (200 µl) and the increase of absorbance at 450 nm was monitored during 30 min at room temperature with a Multiscan Ascent8. MCI activity in each well was proportional to the increase in absorbance at 450 nm and the activity expressed as the change in absorbance [milliOD (mOD)] /min/μg of protein of the sample added into the well.
T tests on paired data were used to compare the values obtained between TB and GM muscle biopsies in the 8 horses during training conditions and the T0 and T1 values obtained in the 7 horses participating to the endurance race. Linear regressions were calculated between each parameter in the 7 horses before and after the race. All statistical tests were performed with GraphPad Instat (3.05)9. P<0.05 value was considered as significant.
Reliability of the assays
Table 1 shows intra- and inter-reproducibility of the SIEFED technique for active MCI and MPO measurements in muscular microbiopsies sampled in 8 TB and in 8 GM. The intra-assay precision of the SIEFED techniques estimated by 2 or 4 repeated measurements within the same run showed a mean CV value of 6.7 ± 3.8% for MCI activity and a mean CV value of 3.8 ± 2.7% for MPO activity.
Table 1. Intra- and inter-assay reproducibility of the SIEFED technique for active MCI and MPO measurements in extracts of equine microbiopsies sampled in 8 TB and 8 GM muscles.
Mean MCI activity (mOD/min/μg proteins x 100)
Mean MPO activity (ng/mg proteins)
n(a): number of sample measurements during the same assay
n(b): number of assay with the sample measured in duplicate or in triplicate in each assay. CV: coefficient of variation, s.d.: standard deviation.
Inter-assay precision was estimated by duplicate or triplicate assays of the extract performed in different run over a period of 3 days. The mean CV value obtained for MCI activity was 14.0 ± 3.7% while it was 6.9 ± 4.3% for MPO activity.
The 46 microbiopsies (from TB and GM) sampled in 8 horses during the training condition period of 10 weeks showed a mean value of 21 ± 14 ng/mg proteins (range 6.0–72.5 ng/proteins) for the active MPO content and a mean value of 0.0172 ± 0.0066 mOD/min/μg proteins (range 0.0073–0.0373 mOD/min/μg proteins) for MCI activity.
No significant difference was observed between the 2 muscles. The mean concentration of active MPO and the mean MCI activity in the TB were 20.98 ± 11.44 ng/mg proteins and 0.0180 ± 0.0066 mOD/min/μg proteins, respectively, while they were 20.76 ± 15.71 ng/mg and 0.0164 ± 0.0066 mOD/min/μg proteins in the GM.
In the 7 endurance horses, a significant increase of plasma CK was observed after the race (P<0.05) but this increase was not accompanied by significant changes of MPO and MC1 activities in the TB in comparison to the mean values observed before the race (Table 2).
Table 2. Mean values ± s.d. and ranges (in brackets) of plasma creatine kinase (CK), MPO and mitochondrial complex I (MCI) activities in 7 endurance horses 24 h before (T0) and 2 h (T1) after a 120 km race
The distribution of CK values for the 7 endurance horses show a strong narrow distribution at T0 while at T1 the value distribution is large and showed two subgroups: one subgroup of 4 horses with CK values <4000 (iu/l) and another subgroup of 3 horses (H1, H2, H4) with CK values >4000 iu/l (Fig 1).
In these 7 endurance horses, the value distributions of active MPO content (Fig 2) and MC1 activity (Fig 3) in TB were quite large before the race (T0). All the active MPO values were distributed in the range of ± 8 –18 ng/mg proteins, except for one horse (H3). After the race there was a shift towards higher values for the active MPO content and towards lower values for the MC1 activity.
Interestingly, the 3 horses that presented the highest CK values (H1, H2, H4) after the race were also those that showed the highest MPO contents and lowest MC1 activities in the TB after the race: there was a severe increase in active MPO content (+ 118 ± 45% vs. T0) and a large decrease of the MC1 activity (−63 ± 18% vs. T0) associated to a high mean plasma CK level of 4642 ± 658 iu/ml (Table 3). In the remaining 4 horses, the race did not change the active MPO content and slightly increased the MC1 activity (Table 3), except for Horse 3 which presented an important decrease of MPO activity (Fig 2) while MCI activity decreased (Fig 3).
Table 3. Mean values ± s.d. of plasma creatine kinase (CK), MPO and mitochondrial complex I (MCI) activities in 7 endurance horses divided in 2 subgroups based on their CK activities 2 h after a 120 km endurance race (at T1). T0: 24 h before the race. In brackets: relative percentage calculated versus the corresponding T0
Horses (n = 4) with CK <4000 iu/l after the race (T1)
Horses (n = 3) with CK >4000 iu/l after the race (T1)
186 ± 30
1187 ± 667
198 ± 87
4642 ± 658
Active MPO (ng/mg proteins)
18.16 ± 9.59
15.71 ± 5.37
13.12 ± 5.30
28.57 ± 5.84
(+118 ± 45%)
MCI activity (mOD/min/μg proteins x100)
1.81 ± 0.14
2.59 ± 0.65
2.19 ± 1.05
0.80 ± 0.38
(−63 ± 18%)
Before the race, no significant correlations were observed between CK levels and either MPO contents (r2: 0.021), or MC1 activities (r2: 0.220), and between active MPO contents and MC1 activities (r2: 0.066). Interestingly, after the race, a trend towards a positive but not significant (P = 0.11) correlation was observed between CK an MPO activity (Fig 4a), a significant negative correlation was measured between CK levels and MC1 activity (r2: 0.622; P = 0.0350) (Fig 4b) and a trend to nearly significant negative correlation was observed between active MPO contents and MC1 activities (r2: 0.492; P = 0.079) (Fig 4c).
Assays of MPO and MCI activities by immunocapture methods
This study is the first one to demonstrate that the activity of neutrophil MPO and the activity of mitochondrial respiratory chain complex I can be measured in equine muscle microbiopsies obtained from conscious horses. To do that, we used original immunoenzymatic techniques, the SIEFED technique for myeloperoxidase (Franck et al. 2006) and immunocapture technique for MCI (MitoSciences company, USA)4.
In complex biological fluids, it is difficult to measure the activity of a targeted enzyme while avoiding artefacts due to the presence of other enzymes or compounds, which interfere with the enzymatic reaction. The enzymatic reactions used to measure targeted enzymes are performed with unspecific substrates that require a partial purification of the sample or the addition of inhibitory compounds to neutralize interfering enzymes. The physiochemical characteristics (pH, viscosity, etc.) of the biological sample, its redox status can also interfere with the colorimetric or fluorescence response (Franck et al. 2009). Immunological methods such as ELISA are specific for measuring an enzyme in biological fluids, but they quantify the total concentration of the enzyme without reflecting its true enzymatic activity. New immunological techniques have been designed recently to measure rapidly and specifically the activity of selected enzymes in complex media. These techniques have 3 major advantages: (i) the enzyme present in a complex solution (biological samples or extracts) is captured by specific immobilised antibodies (ii) washes after the immuno-capture step allow the elimination of the sample, discarding so compounds with potential adverse effects on the enzymatic detection step (iii) the specificity of the immobilised antibodies and the washing after the enzyme capture assure that the detection system discloses only the activity of the targeted enzyme. We designed the SIEFED technique to measure the activity of equine (Franck et al. 2006) or human (Franck et al. 2009) MPO and applied it to the measurement of MPO in complex fluids (Riggs et al. 2007; de la Rebière et al. 2008; Kohnen et al. 2007). Similar immunoenzymatic techniques were developed by the MitoScience company and validated for the measurement of the activity of mitochondrial oxidative phosphorylation (OXPHOS) complexes (Nadanaciva et al. 2007; Willis et al. 2009) in bovine and human fluids and tissue extracts. The Complex I microplate kit from MitoScience was found to be applicable to the horse, owing to the conservation of the major epitopes, supported by a high amino acid identity and sequence homology of Complex I proteins among species (Schilling et al. 2005). The intra- and inter-assays showed a good reproducibility of the 2 tests with intra-assay CVs generally below 10% and acceptable inter-assay CVs below 20%.
MPO and MCI activities during training
In the 8 horses sampled during their training condition period of 10 weeks, no significant differences in mean MPO and MCI activities were observed between the 2 types of muscles. For its easy and safe access for microbiopsy sampling, the TB was chosen to study the variation of MPO and MCI activities during an endurance race.
MPO and MCI activities before and after endurance race
The 7 horses participating in an endurance race of 120 km did not show the same intensity of muscle inflammatory response. This could be partly explained by the fact that all the horses did not provided the same effort during the competition and because of their innate qualities or level of training. The horses could be divided into 2 subgroups when considering the plasma CK activity measured after the race: a high CK subgroup of 3 horses (CK values >4000 iu/l) and a subgroup of 4 horses with CK values <4000 iu/l. In the high CK subgroup, an increase of the activity of MPO and a strong decrease of the MC1 activity were measured. Our study highlighted that after an endurance race, active MPO is present in horse skeletal muscle and that its activity increases in association with an increase of plasma CK, a witness of muscle damage. These results are to be connected to recent studies which demonstrated that the concentration of 2 inflammatory markers, MPO and elastase, increased both in plasma and muscle in correlation with an increase of the plasmatic CK levels after an endurance race in horses (Lejeune et al. 2010; Serteyn et al. 2010). Furthermore, in all the horses participating in the endurance race, we found a significant negative correlation between CK and MCI and a nearly significant negative correlation between MPO and MCI activities suggesting a possible link between MPO activity, mitochondrial function and muscle damage.
Recently, Votion et al. (2010), by using a high-resolution respirometry technique on muscular biopsies, suggested a possible relationship between impaired mitochondrial function and exercise-induced muscle alterations. Active MPO will contribute to exacerbate the exercise-induced muscle damage by increasing membrane alterations and by oxidising muscular proteins involved in cell functions and respiratory pathway (Suzuki et al. 2004; Davies et al. 2008). MCI is the first protein complex involved in the mitochondrial electron transport chain leading to oxidative phosphorylation and a potential source of ROS production in muscular cells (Paradies et al. 2004), but it is also a target for ROS, mainly at the level of its Fe-S clusters.
With the immunocapture technique that binds only the complex I proteins, all the other elements of the muscle extracts (substrates, associated lipids, other mitochondrial complexes) were eliminated by washing, so that the activity of MCI was only dependent on the functional state of the complex. The decrease of MCI activity that we observed after the endurance race is thus attribute to a decrease of MCI content or to an alteration of the enzymatic complex by a structure modification or a lack of subunits association resulting from protein modification. Moreover, during intense exercise, reactive nitrogen and oxygen species (RNOS) are produced inside muscular cells (Powers and Jackson 2008). Mitochondria themselves can be the source of RNOS as a consequence of muscle hypoxia induced by the endurance race, or may further respond to an RNOS production by an increased production of these active species (Zorov et al. 2006). RNOS play a pivotal role in the inhibition of MCI by their derived nitration or nitrosylation reactions (Brown and Borutaite 2004) and the alteration of mitochondrial complexes will restrict energy output and further increase the production of O2 active species such as superoxide anion (O2) and H2O2 the substrate needed for MPO activity. MPO is by itself a source of oxidant species, mainly hypochlorous acid active on Fe-S clusters, but also nitrogen species responsible of nitration or nitrosation reactions (Suzuki et al. 2004). It was shown that nitrogen species and nitrosation reactions could inhibit the activity of MCI (Galkin and Moncada 2007). Therefore, the presence of MPO inside muscular cells will amplify the mitochondrial dysfunction and the muscular damage provided that the enzyme finds H2O2. In vitro studies are in progress to confirm the potential oxidant effect of active MPO on muscular cells.
In conclusion, SIEFED techniques used for the first time to measure MPO and MCI activities in muscular biospsies evidenced a reverse relationship between MPO and MCI activities in horses after endurance race. However, further studies are needed to confirm these preliminary results in a larger number of horses and to better understand the relationship between the increase of MPO content in muscle and the mitochondrial respiratory chain dysfunction and the muscle damage. Moreover, the question remains open whether the accumulation of MPO inside the muscle comes from the stimulation of neutrophils inside the muscle or whether MPO infiltrates the muscle via its excessive accumulation in the blood stream and transcytosis of the endothelial barrier.
The authors wish to express their sincere gratitude to the riders and owner of the horses. This work was supported by the COST (Comité d'Orientation Scientifique et Technique) from the ‘Haras nationaux’ of France and the ‘Ministère de l'Agriculture et de la Ruralité de la Région Wallonne’ of Belgium. We also thank the ‘the Aliments Etienne’, Audevard, Michel Vaillant and Virbac that contributed to the maintenance of horses.
Conflicts of interest
The authors have declared no potential conflicts.
1 Astra Zeneca, Brussels, Belgium.
2 Angiotech, Gainesville, USA.
3 Sigma-Aldrich, Bornem, Belgium.
4 MitoSciences, Eugene, Oregon, USA.
5 Greiner-Bio, Kremsmünster, Austria.
6 DiaSys Diagnostic Systems GmbH, Holzheim, Germany.