Circulating angiotensin converting enzyme in endurance horses: effect of exercise on blood levels and its value in predicting performance



    Corresponding author
    1. Faculty of Veterinary Medicine, The University of Melbourne, Werribee, Australia; and Waikato Institute of Technology, Department of Science and Primary Industries, New Zealand.
    Search for more papers by this author

    1. Faculty of Veterinary Medicine, The University of Melbourne, Werribee, Australia; and Waikato Institute of Technology, Department of Science and Primary Industries, New Zealand.
    Search for more papers by this author
  • H. M. DAVIES,

    1. Faculty of Veterinary Medicine, The University of Melbourne, Werribee, Australia; and Waikato Institute of Technology, Department of Science and Primary Industries, New Zealand.
    Search for more papers by this author
  • C. M. EL-HAGE,

    1. Faculty of Veterinary Medicine, The University of Melbourne, Werribee, Australia; and Waikato Institute of Technology, Department of Science and Primary Industries, New Zealand.
    Search for more papers by this author

    1. Faculty of Veterinary Medicine, The University of Melbourne, Werribee, Australia; and Waikato Institute of Technology, Department of Science and Primary Industries, New Zealand.
    Search for more papers by this author


Reasons for performing study: Investigate angiotensin converting enzyme (ACE) activity in equine plasma as a predictor of performance in endurance competitions and the effect of endurance exercise on ACE activity.

Hypothesis: Precompetition values of ACE activity in equine blood are correlated with performance results and with heart rates pre- and post competition used as indicators of fitness. Endurance exercise increases ACE activity.

Methods: Nineteen horses participating in an 80 km endurance competition had venous blood samples collected before and after the ride. ACE activity and total protein were measured in the blood samples and heart rates and finishing positions were recorded. Statistical analysis included paired t tests and Spearman's rank correlation coefficient.

Results: Of the 19 horses enlisted, only 16 horses completed the ride. Of these 16, another 2 were disqualified at the last veterinary check. When the 16 horses were considered, precompetition heart rate, but not ACE, was correlated with finishing position. When only the 14 horses that were classified were considered, the association disappeared. ACE activity was similar before and after competition.

Conclusions: Precompetition ACE activity in endurance horses competing in an 80 km event was not associated with either finishing position or heart rates before or after competition, indicating that the enzyme is not a good predictor of performance in this form of equestrian competition. Endurance competition did not significantly alter ACE activity in this group of horses.


The association between a polymorphism of the angiotensin converting enzyme (ACE) gene and the response elicited by exercise was first addressed in the 1990s (Montgomery et al. 1997). An I/D ACE polymorphism was associated with better endurance performance, with human subjects homozygous for the insertion of a 286 base pairs fragment (I allele) having an 11-fold increase in weightlifting capacity in comparison to deletion (D allele) homozygous subjects. In extreme endurance sports, such as high altitude mountaineering, an absolute predominance of the II genotype in subjects who successfully ascended over 8000 m was reported (Montgomery et al. 1998). Elite athletes competing at long distance running and rowing also demonstrated a positive association between the ACE genotype and better performance results (Gayagay et al. 1998; Myerson et al. 1999). The possible reasons for the findings are improved metabolism, better utilisation of fat substrates, increased capillary density and improved cardiac output. While the I allele and lower ACE activity seem to produce better anabolic response and hypertrophy in muscle tissue (Montgomery et al. 1999; Williams et al. 2000), it also causes vascular adaptations that are vasoprotective and allow further endothelium derived vasodilation (Day et al. 2004). There have been no studies yet in horses specifically exploring the possible association between plasma ACE levels and suitability for certain types of competition, and although Coomer et al. (2003) speculated that such an association might exist, their study was limited to investigating the effects of age and gender on equine ACE activity. Few previous studies have investigated the association between equine ACE activity and performance, and the effects of exercise on ACE (Costa et al. 2009; Costa 2010). ACE genotype and phenotype are associated, and the close proximity between the site of the genetic polymorphisms of the ACE gene and a putative regulatory site for the ACE protein transcription is believed to be the source of this association (Ellis 2005). While genetic polymorphism explains part of the variation seen in the circulating levels of ACE (47% in man [Rigat et al. 1990] and 10% in horses [Ellis 2005]), environmental factors, such as season of the year and training, exert a much more meaningful influence over circulating ACE activity in horses (Costa 2010). Therefore, circulating ACE activity might provide a better representation of the status of the renin-angiotensin-aldosterone system (RAS) in horses than ACE genotyping. This paper aims to investigate ACE activity as a predictor of endurance performance in horses competing in 80 km rides and its association with heart rates as a marker of fitness and onset of fatigue.

Materials and methods

This study was approved by The University of Melbourne Ethics Committee (AEC 0704151.1) and owners of the horses included in this study gave informed written consent on the eve of the event. A total of 19 horses (73% of the total) competing in an 80 km ride participated in this study. The only inclusion criterion was informed owner consent. On the day before the event a precompetition venous blood sample (10 ml) was obtained from simple jugular venipuncture of each horse, immediately after the veterinarian inspection was carried out. The horses were held by their owners or riders after the inspection trot up, and samples were taken into vacuum tubes containing lithium heparin. The samples were centrifuged for 10 min at 3000 rpm (1700 g) immediately after collection using a bench centrifuge (Orbital 325)1. The resulting plasma was divided into 1 ml aliquots and transferred to labelled Eppendorf tubes, which were placed in dry ice. This processing technique has been validated previously for other components of equine RAS (Guthrie et al. 1982). On the following day, as the horses finished the 80 km ride, a post competition blood sample was collected and processed in the same way as the precompetition samples. These samples were collected throughout the day, at different times, after the horses went through their final veterinary check. All samples were transported in dry ice to the laboratory and were transferred to a -70°C freezer where they were kept until analysis. FAPGG was the substrate used for ACE measurement, with assays conducted at a commercial laboratory2. In addition to enzymatic activity, total plasma protein was also measured. This was carried out immediately after centrifugation with a portable refractometer. The same researcher (M.F.M.C.) was responsible for all readings to minimise observer variation. Additional data collected included heart rate at each of the veterinary examinations (pre-, during and post competition), time to complete the ride, number of competitors in the category and weight of the rider. In regards to finishing position, determined by the time taken to complete the ride, some animals tied. An adjusted finishing position was created by untying according to the heart rate at the last veterinary check (the lower the heart rate, the better the position). Paired t tests were used to compare mean values obtained before and after the competition. Spearman's rank correlation coefficients were calculated using the values of ACE activity (pre- and post competition) and heart rate (precompetition and at the last veterinary check), and adjusted finishing position. Descriptive statistics are presented as mean ± s.e. The results of the paired t tests are presented as mean difference ± s.e. and 95% CI of the difference between the post and prerace results for each parameter. A 2-sided P value <0.05 was considered to be statistically significant. Stata 11.03 and Minitab were used for analysis.


Twenty-six horses participated in the 80 km endurance ride and 19 were enlisted for the study. The additional 7 horses in the competition were not included in the study because owner consent was not given. Of the horses enlisted, 3 were excluded because they did not complete the competition for various reasons. Mean ACE activity precompetition for the horses that successfully completed the race (n = 14) was 111.3 ± 5.3 u/l, while post competition values were 112.6 ± 7.0 u/l. The range of ACE activity precompetition was 79–144 u/l. ACE activity post race was not significantly different from prerace activity in those 14 horses (difference 1.3 ± 6.8 u/l; 95% CI -13.4–16.0, P = 0.85). Plasma protein post race (71 ± 3 g/l) was significantly higher than prerace (56 ± 1 g/l; difference 15 ± 3 g/l, 95% CI 8–21, P<0.001) and heart rate at the last veterinary check (52 ± 1 beats/min) was significantly higher than the heart rate before the race (37 ± 1 beats/min; n = 14; difference 15.1 ± 1.1 beats/min, 95% CI 12.7–17.4, P<0.001). Mean ACE activity for the 5 horses that did not complete the race was 101.6 ± 11.5 u/l prerace and 112.6 ± 8.4 at the time of disqualification/withdrawal from the race. When precompetition ACE measurements from horses that completed the 80 km ride (n = 14) were compared with those that did not complete the race (n = 5), there was no statistically significant difference (difference 9.7; 95% CI -13.7–33.1; P = 0.39). Spearman's rank correlation coefficient demonstrated no association between ACE activity (n = 14) before the competition and heart rate before the competition (rho = 0.29, P = 0.31), or heart rate at the last veterinary check (rho = 0.28, P = 0.33). No association was found between ACE activity (n = 14) precompetition and plasma protein precompetition (rho = 0.04, P = 0.89), plasma protein post competition (rho = -0.34, P = 0.24) or adjusted finishing position (rho = 0.40, 95% CI -0.16–0.77, P = 0.15). ACE activity (n = 14) post competition was associated with neither heart rates pre- and post competition (rho = 0.47 and 0.41, P = 0.09 and 0.15, respectively) nor adjusted finishing position (rho = 0.42, P = 0.14). The same method showed that the heart rate precompetition was associated with the finishing position only if the 2 horses that completed the 80 km ride but were disqualified at the last veterinary check (heart rates of 62 and 63) were included (rho = 0.59, P = 0.016, n = 16). If only the 14 classified horses are considered, the association disappeared (rho = 0.50, P = 0.069).


There is evidence that ACE genotype in man and therefore phenotype are correlated to athletic aptitude, with individuals displaying lower ACE activity due to the II genotype showing better aptitude towards endurance-related sports. In Thoroughbreds, ACE activity can be associated with successful racing, with horses displaying low ACE activity competing successfully at longer distances, while horses with higher ACE compete better at shorter distances (Costa et al. 2009). Therefore, the original hypothesis was that horses with lower ACE activities would compete better at endurance rides, with the horses displaying the lowest values being the best placed. This was not in accordance with our findings, since ACE activity was not significantly associated with finishing position in a 80 km endurance ride. There are many possible reasons for this lack of agreement between the findings in human athletes competing in endurance sports and the inability to identify a similar association for the equine endurance athletes involved in this event. The most obvious one is the inherent metabolism of each species. Although both human athletes and horses would be most likely to utilise aerobic metabolic pathways during long competitions, horses have higher reserves and capacity for accumulation of larger amounts of lactate (Poso et al. 2008). Secondly, horses are capable of achieving much higher systemic blood pressure; therefore blood regulation requirements are slightly different (Poole and Erickson 2008). The third reason for the difference is that even in endurance riding there might be a sprint component in the last kilometres of the race, up to 200 m before the finish line, which might induce sudden vasoconstriction and therefore a rise in ACE activity. ACE activity was also not associated with total protein measurements, indicating that it was not dependent on hydration status. It may also be that horses for this event were highly selected already, based on their ability for the sport. We demonstrated an association between finishing position and heart rate precompetition. This is well established in the literature, with heart rate measurements being one of the most common methods to assess fitness and showing a linear relationship with running speed (Holloszy and Coyle 1984; Evans 1994; Aerts et al. 2008; Poole and Erickson 2008). This is especially common in endurance competitions, where heart rates immediately after the finish of the ride, and the time to recovery of the heart rate, are determinants of fitness and completion of the ride. However, the association between finishing position and heart rate precompetition disappeared when the 2 horses eliminated at the last veterinary check due to heart rates of 62 and 63 were excluded. It might be that a correlation between ACE activity and the parameters studied here was missed since only 14 horses actually finished the race. Studies with a higher number of animals would be required to improve statistical power. We have used the heart rate at the last veterinary check as the parameter to untie horses classified at the same finishing position. Other methods, rather than finishing heart rate, might also be employed for untying the finishing positions in future studies, although heart rate is the main factor in deciding which horses actually finish the race successfully and which horses require disqualification. We also found that endurance exercise under the conditions described here did not cause a significant rise in ACE activity, with similar values found pre- and post competition. This was unexpected since we have previously demonstrated that acute exercise and training cause a significant elevation in ACE activity (Costa 2010). It would be valuable to measure ACE activity in endurance horses of a similar age and at similar training stages to verify the use of ACE as an indicator of quality of endurance animals, as has been demonstrated to be true for racing Thoroughbreds (Costa 2010).

In conclusion, ACE activity in the endurance horses studied does not correlate with heart rate, plasma protein or finishing position, and does not appear to be a suitable predictor of performance for horses engaged in endurance events. ACE activity was similar before and after the competition, under the conditions studied.


The authors wish to acknowledge the funding from the Pathology section of the Faculty of Veterinary Science (The University of Melbourne) and the contributions of Gribbles Veterinary Pathology, VERA (Victorian Endurance Riding Association) and the owners and riders of the animals engaged in the study.

Conflicts of interest

The authors have declared no potential conflicts.

Manufacturers' addresses

1 Clements, Somersby, New South Wales, Australia.

2 Gribbles Veterinary Pathology, Clayton, Victoria, Australia.

3 StataCorp, College Station, Texas, USA.

4 Minitab Inc, State College, Pennsylvania, USA.