Differences in Blood Parameters Associated to Stress Response Between Chilean Rodeo Horses and Chilean Urban Working Horses

Journal of Equine Veterinary Science 73 (2019) 110e114

Contents lists available at ScienceDirect

Journal of Equine Veterinary Science

journal homepage: www.j-evs.com

Short Communication Differences in Blood Parameters Associated to Stress Response Between Chilean Rodeo Horses and Chilean Urban Working Horses

Igor Gonzalez a, Cristian G. Torres b, Ricardo Chihuailaf c, Víctor Neira d, * Tamara A. Tadich e, a Programa de Magíster en Ciencias Animales y Veterinarias, Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, La Pintana, Santiago, Chile b Departamento de Ciencias Clínicas, Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, La Pintana, Santiago, Chile c Departamento de Ciencias Veterinarias y Salud Pública, Universidad Catolica de Temuco, Temuco, Chile d Departamento de Medicina Preventiva Animal, Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, La Pintana, Santiago, Chile e Departamento de Fomento de la Produccion Animal, Facultad de Ciencias Veterinarias y Pecuarias, Universidad de Chile, La Pintana, Santiago, Chile article info abstract

Article history: Physiological measures, such as blood variables, are commonly used to assess the welfare state of ani- Received 5 July 2018 mals. The basal concentration of indicators such as plasma cortisol, white blood cells, cytokines, and Received in revised form indicators of oxidative stress could vary depending on the coping style of individuals. In the case of 22 November 2018 horses, coping styles could be associated to the activity they perform because owners seek different Accepted 6 December 2018 behavioral characteristics. The aim of this study was to compare blood variables frequently used to assess Available online 18 December 2018 welfare, between urban working horses and Chilean rodeo horses, and secondly to determine if horses could be classiﬁed according to activity and blood variable characteristics associated to coping styles Keywords: Welfare (proactive and reactive). A total of 204 horses were sampled and 13 blood variables were assessed. K- ﬁ Blood variable means clusters analysis was used to classify horses in two groups. Signi cant differences between Stress response working horses and Chilean rodeo horses were found for nine of the 13 variables studied. Two clusters Cortisol were formed; cluster 1 grouped most working horses and is characterized by a higher concentration of Coping style cortisol, leukocytes (mainly lymphocytes), and carbonyl groups. On the other hand, cluster 2 groups most rodeo horses and is characterized by horses with lower cortisol concentrations, but a higher antioxidant capacity. Further studies are required to assess if owners are selecting horses by coping style according to the activity they perform (rodeo or draft work) and how these physiological characteristics should be considered when assessing their welfare state. © 2018 Elsevier Inc. All rights reserved.

1. Introduction as increases in heart rate, epinephrine secretion, plasma cortisol, and plasma glucose [2] and some long-term ones such as immune The most universally accepted measure of animal welfare is function and oxidative stress status [3e5]. All the abovementioned physical health because injuries, disease, and deformities are physiological welfare indicators can behave differently according to generally acknowledged to be sources of suffering and poor welfare the animal's coping style and have interrelated actions. [1]. Physiological measures are also commonly used in the assess- A coping style has been defined as “a coherent set of behavioral ment of animal welfare. These include short-term responses such and physiological stress responses, which are consistent over time and characteristic for a certain group of individuals” [6]. In other words, coping styles represent an integrated neurophysiological and behavioral response, with low (proactive) and high (reactive) Animal welfare/ethical statement: All procedures performed in animals were response individuals associated to the level of activation of the revised and approved by the Institutional Animal Care and Use Committee of the Universidad de Chile. Authorization N 12-2016. hypothalamic-pituitary-adrenal axis (HPA-axis). Horses can show Conflict of interest statement: The authors have no conflict of interest to declare. specific coping strategies, revealing specific behavioral and physi- * Corresponding author at: Tamara A. Tadich, Departamento de Fomento de la ological patterns to adapt to the everyday challenging situations Produccion Animal, Facultad de Ciencias Veterinarias y Pecuarias, Universidad de [7]. Based on social stress research, two coping styles have been Chile, Santa Rosa, La Pintana, Santiago, Chile. fi E-mail address: [email protected] (T.A. Tadich). suggested according to Henry and Stephens [8]; the rst type, https://doi.org/10.1016/j.jevs.2018.12.005 0737-0806/© 2018 Elsevier Inc. All rights reserved. I. Gonzalez et al. / Journal of Equine Veterinary Science 73 (2019) 110e114 111 known as the proactive (active) response, corresponds to the fight resting day (usually Mondays). The horses had at least 12 hours of or flight response. Behaviorally this response is characterized by rest before sampling was performed. Blood was aliquoted into 3 territorial control, aggression, and routine formation [6]. Proactive tubes; with EDTA for hematology and serum amyloid A (SAA), with individuals respond physiologically with a strong sympathetic heparin for oxidative stress parameters, and with no additives to activation and an increase in noradrenergic stimulation [9]. The obtain serum for cortisol and cytokines determination. After second type is known as the reactive (conservation-withdrawal collection in the field, blood samples were placed in a cooler box response) characterized by behavioral immobility and low levels of with ice packs and transported to the laboratory for analysis or aggression [6]. These individuals respond to challenge with a strong centrifugation as appropriate. HPA-axis reactivity, increasing circulating glucocorticoids [9], and a higher baseline of nonenzymatic antioxidant capacity [10]. 2.3. Analytic Methods for Blood Parameters Although glucocorticoid secretion pattern is the main distinguish- ing feature between both coping styles, other parameters have For hematology, white blood cell (WBC) count, neutrophils (N), been associated with them such as changes in leukocytes and cy- lymphocytes (L), neutrophils-lymphocytes ratio, and eosinophils tokines [4,11] and oxidative status [10,12,13]. Stress response affects were determined by using a Wiener CP-19 counter and confirmed redox physiology, and although it has not been specifically studied by blood-stained smears using a Romanowsky stain (Corzap 1, in horses as a consequence of stress response, experimental in- Hemogram) at 1,000 Å~ (Olympus CX31). For cortisol, radio creases in glucocorticoids have been proved to increase oxidative immunoassay was used. To assess oxidative stress status, gluta- damage in vertebrates (see meta-analysis from Costantini et al. thione peroxidase activity in erythrocytes (EC, 1.11.1.9) was esti- [14]). Most of these physiological parameters associated to proac- mated by a kinetic method with a commercial reagent Ransel tive and reactive coping styles coincide with classical animal-based (Randox), a-tocopherol plasma concentration was determined by welfare measures (indicators). high-performance liquid chromatography reverse phase using a Horses are selected according to the activity they perform, for Perkin Elmer Flexar chromatograph coupled to a UV detector and example, rodeo exercise or urban draft work, for which owners equipped with a Kromasil column (150 4.6 mm and 4.5 mm), prefer certain behavioral characteristics that are usually associated according to an adaptation to the technique indicated by Chihuailaf to a more proactive or reactive coping style. Rodeo horses are et al. [17], malondialdehyde (MDA) plasma concentration was subjected to routines, and owners prefer a horse that responds measured according to the method described by Ohkawa et al. [18] more aggressively, whereas urban draft horses are subjected to a using the thiobarbituric acid reaction, and carbonyl groups con- more flexible environment and owners prefer horses that do not centration was determined by 2,4-dinitrophenylhydrazine as respond when confronted to a stressor (i.e., traffic). Thus, if horses reactive component according to the method suggested by Gua- are being selected according to coping styles, they could have not jardo et al. [19]. The hemoglobin concentration was measured by only behavioral differences according to activity but also differ- the cyanmethemoglobin method. Inflammatory response was ences in their baseline blood parameters. It is hypothesized that assessed through determination of concentration blood cytokines rodeo horses will have a lower baseline concentration of cortisol (Interleukin 6 [IL-6], Interleukin 10 [IL-10], and transforming together with blood variables associated to a proactive coping style, growth factor b-1 [TGF-b1]) by quantitative commercial sandwich whereas urban draft horses will present higher cortisol baselines ELISA tests. Interleukin 6 was detected by the ELISA IL-6 kit (Cloud and blood variables characteristic of a reactive coping style. Clone corp.) Cat No (SEA079Eq 9). Interleukin 10 concentrations were determined using the Equine IL-10 ELISA Kit (Abcam) Cat No 2. Material and Methods (ab155466). The equine TGF-b1 was evaluated by the Nori Equine TGFb1 ELISA Kit (GSI Equine, Genorise) Cat No GR106123. In addi- 2.1. Animals tion, SAA was evaluated with a semiquantitative method using the commercial kit EquiChek from Accuplex Diagnostics (Maynooth, A total of 204 horses, 102 Chilean rodeo horses and 102 Chilean Ireland). urban draft horses were sampled prior informed consent of their owners and approval by the Institutional Animal Care and Use 2.4. Statistical Analysis Committee (N12e2016). Individual information of horses was registered: name, sex, age, general attitude (alert or depressed), All data were organized in an Excel spread sheet. First, outliers presentation of stereotypic behaviors, and number of working days were identified by the Thompson Tau method and eliminated. Then per week. Horses that presented any abnormality at general clinical descriptive statistics (median, mean standard deviation, and min- examination were excluded from the study, only sound horses aged imum and maximum values) were determined. Normality of data above 3 years, either mares, geldings, or stallions, were included. was checked using ShapiroeWilk test. To check for differences Horses used for Chilean rodeo and for urban draft work are very between age ranges within each group of horses, KruskaleWallis or similar in terms of physical conformation. Chilean rodeo horses ANOVA was used (Minitab 14). To compare each parameter behave average weights between 390e394 and height to the withers tween Chilean rodeo horses and Chilean urban draft horses, the between 138e148 cm with an ideal between 140e142 cm [15] Wilcoxon Rank Sum test or two-sample t-test was used according whereas Chilean urban draft horses have live weights between data distribution (Minitab 14). Finally, the nonhierarchical K-means 370e415 kg and a height to the withers of 143 cm (±5.4) [16]. clustering method was used to classify horses in two groups ac- Chilean rodeo horses sampled in this study are kept in individual cording to blood variables, for this standardized value of the blood boxes, with a fixed ration based on hay and oats, whereas urban variables were used (IBM SPSS, Statistics v. 22). A significant value draft horses are also kept in individual boxes but with a diet based of P .05 was established. on hay and residues from vegetables (carrots, lettuce, spinach). 3. Results and Discussion 2.2. Sampling Equal number of rodeo and urban working horses was sampled. Blood samples were obtained through a jugular puncture, at For both groups, the mean age was 8 years. No significant differ- rest, during morning time (between 09:00 and 12:00) of their ences (P > .05) were found between age ranges for any of the 112 I. Gonzalez et al. / Journal of Equine Veterinary Science 73 (2019) 110e114 parameters assessed. In the urban draft group, most horses were Table 2 mares (63.7%), similar to prior studies in Chilean working horses Relative weight given to each variable (F score) to determine the cluster in which they were allocated and their significance P-value according to the ANOVA test reporting a preference for mares by owners [16,20]. The proportion included in the K-means cluster analysis. of stallions was low and similar in both groups (17.6% and 15.6% for rodeo and working horses, respectively). According to number of Blood Variable ANOVA (F Score) P-Value working days, most rodeo horses (98%) worked 6e7 days per week, Leukocytes 23.306 <.001 whereas most working horses (61.7%) only worked 1e3 days per Neutrophils 0.004 .947 < week. The low number of working days for Chilean urban draft Lymphocytes 99.123 .001 Eosinophils 4.547 .035 horses is consistent with previous studies [16] and differs with N:L 38.056 <.001 working equids in other countries [21]. Stereotypic behavior was Cortisol 25.296 <.001 only recorded for two horses of the rodeo group. GPx 1.491 .225 In the present study, significant differences between working Tocopherol 0.345 .558 MDA 7.420 .007 horses and Chilean rodeo horses were found for nine of the 13 Carbonyl groups 3.298 .072 variables studied (Table 1). When forming two clusters, significant differences are found between the two (Table 2), cluster 1 being Abbreviations: ANOVA, analysis of variance; GPx, glutathione peroxidase; MDA, malondialdehyde; N:L, neutrophils-lymphocytes ratio. represented by higher concentrations of serum cortisol, leukocytes (mainly lymphocytes), and carbonyl groups, including most working horses (57%). On the other hand, cluster 2 represents those cytokines could also be playing a key role in this response [36]. horses with lower cortisol concentrations, but a higher antioxidant Response differs from that occurring after an acute stressor, for capacity, including most Chilean rodeo horses (58%) (Fig. 1). There example, after high-intensity exercise, changes in leukocyte count was no significant association between belonging to a cluster and are likely because of catecholamine and splenic contraction [37,38]. horses' activity (P ¼ .097). In the present study, possible obesity or insulin resistance was Working horses had a higher mean concentration of cortisol not considered as a factor influencing inflammatory profile. One than rodeo horses (P ¼ .02). On the other hand, rodeo horses had a study in 113 horses reported that 79.6% of them were overweight higher total leukocyte count associated to a higher concentration of [39], whereas for working horses, over 59% [16] and 80% [40] were neutrophils but with lower concentrations of lymphocytes and reported in good body condition. Obesity-related disease metabolic eosinophils (Table 1). Although significant differences were found syndrome should be considered in future studies as a possible between groups, both groups have higher basal cortisol concen- source of a distinctive inflammatory phenotype. For example, Burns trations than those described in the literature for fit adult horses et al. [41] described a higher inflammatory signaling from the (25e155 nmol/L) [33]. It is noteworthy that these variables, except nuchal ligament in obese horses but were not able to support a for neutrophils, explain the clusters in which they were allocated direct link between the degree of insulin resistance and inflam- (Table 2). Interestingly, chronic intermittent stress associated to matory mediator gene expression by adipose tissue in obese horses. unpredictability of stressors has been associated to increases in On the other hand, an obesity-prone phenotype associated to circulating cortisol [34]. Nevertheless, cortisol cannot account by reactive coping styles (high cortisol response) has been described itself for all the changes in the WBCs [35] and glucocorticoid re- in ewes [42] and rats [43]; it would be interesting to further study ceptors resistance together with an increase in proinflammatory this relationship in horses.

Table 1 Mean, median, minimum, and maximum values for the thirteen variables studied in rodeo horses (n ¼ 102) and working horses (n ¼ 102).

Blood Variables Reference Values Activity Mean (SD) Median Min Max P-Value

Cortisol nmol/L 32e240 [22] Rodeo 237 (60) 236 94 381 .0265 Work 258 (66) 261 111 398 Leukocytes 103 mL5e11 [22] Rodeo 8.503 (1.607) 8.500 5.100 12.000 .0387 Work 8.024 (1.596) 7.950 5.000 11.300 Neutrophils 103 mL2e6.1 [22] Rodeo 4.908 (1.228) 4.836 2.491 8.260 .0102 Work 4.410 (1.417) 4.386 1.360 7.410 Lymphocytes 103 mL 1.5e6.5 [22] Rodeo 3.028 (.945) 2.933 0.972 5.394 .423 Work 3.201 (1.022) 3.100 1.554 5.330 Eosinophils 103 mL 0.1e0.8 [22] Rodeo 0.130 (137) 0.091 0 0.528 .002 Work 0.226 (186) 0.190 0 0.640 N:L 0.8e2.8 [23] Rodeo 1.67 (0.75) 1.52 0.11 3.5 .0481 Work 1.42 (0.68) 1.38 0.13 3.33 GP U/g Hb >300 [22] Rodeo 241 (98) 260 39 423 < .001 Work 134 (61) 123 30 261 a-Tocopherol mg/ml 2e5.4 [24,25] Rodeo 10.09 (0.52) 10.08 9.23 10.9 .796 Work 9.55(3.48) 8.76 2.25 16.51 Carbonyl groups nmol/mg protein 0.00002e0.00099 [26,27] Rodeo 0.0004 (0.0003) 0.0003 0 0.001 .0168 Work 0.0005 (0.0004) 0.0005 0 0.001 MDA nmol/ml 0.07e15 [28] Rodeo 21.14 (14.23) 16.46 7.51 82.43 .019 Work 16.37 (9) 14.89 3.57 49.85 IL-6 pg/mL 2,000e618,000 [29] Rodeo 6,302 (5,749) 4,232 0 24,430 .325 Work 5,958 (6,071) 3,493 0 26,000 IL-10 pg/mL 4e6.66 [30,31] Rodeo 202 (605) 0.5 0 4,267 .741 Work 100 (285) 0.6 0 2,271 TGF b 1 pg/mL 6,000e27,000 [32] Rodeo 350 (202) 298 175 1,228 < .001 Work 812 (318) 815 161 1,648

Abbreviations: GPx, glutathione peroxidase; IL-6, interleukin 6; IL-10, interleukin 10; MDA, malondialdehyde; TGF-b1, transforming growth factor b-1; SD, standard deviation. The P-values for differences between groups are also shown for each variable. I. Gonzalez et al. / Journal of Equine Veterinary Science 73 (2019) 110e114 113

Fig. 1. The histogram shows the division of the 204 horses into two clusters based on 10 speciﬁed blood variables. Each column represents the standardize value of the ten variables included in the iterative k-means analysis and their average distance from the cluster's center.

The suboptimal environment and management conditions, oxidative stress, welfare, and personality in horses requires further many times unpredictable, under which working horses are kept studies because most literature is associated to inflammatory pa- could correspond to this type of physiological response to chronic thologies, mainly articular disease. It is also worth studying if these intermittent stress and are worth studying in more detail because differences in baseline variables result in differences in the similar responses have been reported in humans with major response to specific stressors associated to coping styles and how depression. Working horses also tend to show a higher frequency of they should be considered during welfare assessments. depression-like behaviors [16,21], a characteristic that also associated to a more reactive coping style, which is also characterized by a Acknowledgments higher cortisol response, as in cluster 1 mainly represented by working horses (57.1%), whereas cluster two groups most rodeo The authors would like to thank the Chilean Comision Nacional horses (58.2%). Differences between groups for oxidative stress de Investigacion Científica y Tecnologica (CONICYT) for funding this variables were found, especially for glutathione peroxidase, which study through the Fondo Nacional de Desarrollo Científico y Tec- fi was signi cantly higher in the rodeo group. Rodeo horses are nologico Project FONDECYT N 1161136. usually supplemented with vitamin E and selenium due to the selenium deficiency found in Chile's pastures, whereas working References horses are not. Both groups reported high levels of alpha tocopherol (Vit E), compared with the results from the study by Steiss et al. [24] [1] Dawkins M. Behaviour as a tool in the assessment of animal welfare. Zoology for Quarter Horses, Thoroughbreds, and mixed horses; this could be (Jena) 2003;106:383e7. explained by dietary differences, and the high levels detected in [2] Broom DM, Johnson K. Stress and animal welfare. London, UK: Chapman and Hall; 1993. working horses could be associated to the green forage obtained [3] Moberg GP. Biological responses to stress: implications for animal welfare. In: from vegetables leftovers from farmer's markets that they are fed Moberg GP, Mench JA, editors. The biology of animal stress, basic principles with. Chronic exposure to glucocorticoids can induce oxidative and implications for animal welfare. Cambridge, UK: CABI Publishing; 2000. [4] Aschbacher K, O’Donovan A, Wolkowitz OM, Dhabhar FS, Su Y, Epel E. Good stress, increasing oxidative damage [14]; thus, oxidative stress is a stress, bad stress and oxidative stress: insights from anticipatory cortisol potential index of animal welfare [44]. In the present study, reactivity. Psychoneuroendocrinology 2013;38:1698e708. carbonyl groups and MDA were used as indicators of oxidative [5] Passantino A, Quartarone V, Pediliggeri MC, Rizzo M, Piccione G. Possible fi application of oxidative stress parameters for the evaluation of animal welfare damage of protein and lipids, respectively, MDA being signi cantly in sheltered dogs subjected to different environmental and health conditions. higher in rodeo horses, but both groups higher than reports from J Vet Behav 2014;9:290e4. other studies [45,46]. The role of antioxidants and oxidative dam- [6] Koolhas JM, Korte SM, de Boer SF, Van Der Vegt BJ, Van Reenen CG, Hopster H, age markers in chronic stress and personality of horses requires De Jong IC, Ruis MAW, Blokhuis HJ. Coping styles in animals: current status in behaviour and stress physiology. Neurosci Biobehav Rev 1999;23:925e35. further studies because most literature has been related to nutri- [7] Budzynska M. Stress reactivity and coping in horse adaptation to environ- tional status and exercise intensity in the case of horses [45,46].In ment. J Equine Vet Sci 2014;34:935e41. this regard, interesting associations between oxidative stress and [8] Henry JP, Stephens PM. Stress, health and the social environment: a socio- biological approach to medicine. Berlin: Springer; 1977. personality in marmots have been reported [47], and between [9] Koolhas JM, de Boer SF, Coppens CM, Buwalda B. Neurobiology of coping oxidative stress and welfare in dogs [5]. In the present study, two styles: towards understanding the biology of individual variation. Front profiles associated to blood parameters were described, which Neuroendocrinol 2010;31:307e21. [10] Costantini D, Carere C, Caramaschi D, Koolhaas JM. Aggressive and non- could be related to coping styles. Nevertheless, the link between aggressive personalities differ in oxidative status in selected lies of mice dysregulated function of inflammation associated to cytokines and (Mus musculus). Biol Lett 2008;4:119e22. 114 I. Gonzalez et al. / Journal of Equine Veterinary Science 73 (2019) 110e114

[11] Hessing MJC, Coenen GJ, Vaiman M, Renard C. Individual differences in cell [30] Burton A, Wagner B, Erb H, Ainsworth D. Serum interleukin-6 (IL-6) and IL-10 mediated and humoral immunity in pigs. Vet Immunol Immunopathol concentrations in normal and septic neonatal foals. Vet Immunol Immuno- 1995;45:97e113. pathol 2009;132:122e8. [12] Lin H, Decuypere E, Buyse J. Oxidative stress induced by corticosterone [31] Curto E, Messenger K, Salmon J, Gilger B. Cytokine and chemokine profiles of administration in broiler chicken (Gallus gallus domesticus) 1. Chronic expo- aqueous humor and serum in horse with uveitis measured using multiplex sure. Comp Biochem Physiol B Biochem Mol Biol 2004;139:737e44. bead immunoassay analysis. Vet Immunol Immunopathol 2016;182:43e51. [13] Sahin E, Gümüslu S. Stress-dependent induction of protein induction of pro- [32] Argüelles D, Carmona J, Prades M. Evaluacion de los niveles del factor de tein oxidation, lipid peroxidation and anti-oxidants in peripheral tissues of crecimiento transformante beta 1 y beta 3 en plasma y líquido peritoneal de rats: comparison of three stress models (immobilization, cold and immobili- caballos con enfermedad abdominal aguda. Arch Med Vet 2009;41:213e9. zation cold). Clin Exp Pharmacol Physiol 2007;34:425e31. [33] Knottenbelt D. Vital signs, normal values. In: Knottenbelt D, editor. Saunders [14] Costantini D, Marasco V, Moller AP. A meta-analysis of glucocorticoids as equine formulary. Philadelphia, USA: W.B. Saunders Co; 2006. p. 7. modulators of oxidative stress in vertebrates. J Comp Physiol B 2011;181: [34] Ladewig J. Chronic intermittent stress: a model for the study of long term 447e56. stressors. In: Moberg GP, Mench JA, editors. The biology of animal stress, basic [15] García M, Cabezas I, Guzman R, Valenzuela S, Merino V, Perez R. Hypometric principles and implications for animal welfare. Cambridge, UK: CABI Pub- characteristic, body weight and loading capacity of the native chilean horses lishing; 2000. in rodeo competition. Avances Em Ciencias Veterinarias 1997;12:1. [35] Minton JE, Apple JK, Parsons KM, Blecha F. Stress-associated concentrations of [16] Tadich T, Pearson RA, Escobar A. Husbandry and welfare aspects of urban plasma cortisol cannot account for reduced lymphocyte function and changes draught horses in the south of Chile. Arch Med Vet 2008;40:267e73. in serum enzymes in lambs exposed to restraint and isolation stress. J Anim [17] Chihuailaf R, Gonzalez C, Wittwer F, Contreras P. Concentraciones plasmaticas Sci 1995;73:812e7. de retinol en vaquillas a pastoreo: primeros valores obtenidos en un rebano~ [36] Pace TWW, Hu F, Miller AH. Cytokine-effects on glucocorticoid receptor del sur de Chile. Arch Med Vet 2008;40:65e8. function: relevance to glucocorticoid resistance and the pathophysiology and [18] Ohkawa H, Orishi N, Yagi K. Assay for lipid peroxides in animal tissues by treatment of major depression. Brain Behav Immun 2007;21:9e19. thiobarbituric acid reaction. Anal Biochem 1979;92:337e42. [37] Vazzana I, Rizzo M, Dara S, Niutta PP, Giudice E, Piccione G. Haematological [19] Guajardo MH, Terrasa AM, Catala A. Lipideprotein modifications during changes following reining trials in quarter horses. Acta Sci Vet 2014;42:1e5. ascorbate-Fe2þ peroxidation of photoreceptor membranes: protective effect [38] Rizzo M, Arfuso F, Gianneto C, Giudice E, Longo F, Di Pietro S, Piccione G. of melatonin. J Pineal Res 2006;41:201e10. Cortisol levels and leukocyte population values in transported and exercised [20] Lanas R, Luna D, Tadich T. The relationship between working horse welfare horses after acupuncture needle stimulation. J Vet Behav 2017;18:56e61. and their owners’ socio economic status. Anim Welf 2018;27:47e54. [39] Soto MR. Determinacion de la condicion corporal en caballos pura raza chileno [21] Pritchard JC, Lindberg AC, Main DCJ, Whay HR. Assessment of the welfare of de la region de los ríos, Chile. Memoria de Título, Facultad de Ciencias Vet- working horses, mules and donkeys, using health and behavior parameters. erinarias. Valdivia, Chile: The Universidad Austral de Chile; 2013. p. 25. Prev Vet Med 2005;69:265e83. [40] Luna D, Vasquez RA, Rojas M, Tadich T. Welfare status of working horses and [22] Wittwer F. Manual de patología clínica veterinaria. Valdivia, Chile: Segunda owners’ perceptions of their animals. Animals 2017;7:56. edicion editorial America; 2012. p. 200. [41] Burns TA, Geor RJ, Mudge MC, McCutcheon LJ, Hinchcliff KW, Belknap JK. [23] Aros K, Carrasco J, Briones R, Tadich T. Haematological and serum biochemical Proinflammatory cytokine and chemokine gene expression profiles in sub- reference values for urban working equines in Chile. Austral J Vet Sci 2017;49: cutaneous and visceral adipose tissue depots of insulin-resistant and insulin- 27e33. sensitive light breed horses. J Vet Intern Med 2010;24:932e9. [24] Steiss JE, Traber MG, Williams MA, Kayden HJ, Wright JC. Alpha tocopherol [42] Lee TK, Lee C, Bischof R, Lambert GW, Clarke IJ, Henry BA. Stress-induced behav- concentrations in clinically normal adult horses. J Equine Vet Sci 1994;26: ioral and metabolic adaptations lead to an obesity-probne phenotype in eweswith 417e9. elevated cortisol responses. Psychoneuroendocrinology 2014;47:166e77. [25] Williams C, Kronfeld D, Hess T, Saker K, Waldron J, Crandell K, Hoffman R, [43] Boersma GJ, Benthem L, van Dijk G, Steimer TJ, Scheurink AJW. Coping style Harris P. Antioxidant supplementation and subsequent oxidative stress of predicts the (in)sensitivity for developing hyperinsulinemia on a high fat diet horses during an 80-km endurance race. J Anim Sci 2004;82:588e94. in rats. Physiol Behav 2010;100:401e7. [26] Moffarts B, Kirschvink N, Van Erck E, Art T, Pincemail J, Lekeux P. Assessment [44] Fazio F, Arfuso F, Rizzo M, Giannetto C, Giudice E, Zanghi E, Piccione G. of the oxidant-antioxidant blood balance in a field exercise test in standard- Livestock handling and road transport influence somo oxidative stress pa- bread an eventing horses. Comp Exerc Physiol 2005;2:253e61. rameters in ewes. J Vet Behav 2018;26:5e10. [27] Mcgorum B, Wilson R, Pirie R, Mayhew G, Kaur H, Aruoma O. Systemic con- [45] Jagric MS, Nemec SA, Zrimsek P, Kramaric P, Ko KV, Vovk T, Kobal S. Plasma centrations of antioxidants and biomarkers of macromolecular oxidative malondialdehyde, biochemical and haematological parameters in Standard- damage in horses with grass sickness. Equine Vet J 2003;35:121e6. bred horses during a selected field exercise test. Acta Vet Beogr 2012;62:53e68. [28] Chiaradia E, Avellini L, Rueca F, Spaterna A, Porciello F, Antonioni MT, Gaiti A. [46] Yonezawa LA, de Sousa Barbosa T, Jun Watanabe M, Knaut JL, Marinho CL, dos Physical exercise, oxidative stress and muscle damage in racehorses. Comp Santos Michima LE, Kohayagawa A. Oxidative and cardiac metabolism of Biochem Physiol B Biochem Mol Biol 1998;119:833e6. horses subjected to low intensity exercise before and after antioxidant sup- [29] Suagee JK, Corl BA, Crisman MV, Pleasant RS, Thatcher CD, Geor RJ. Re- plementation. Cienc Rural 2014;14:1060. lationships between body condition score and plasma inflammatory cyto- [47] Costantini D, Ferrari C, Pasquaretta C, Cavallone E, Carere C, von Hardenberg A, kines, insulin, and lipids in mixed population of light-breed horses. J Vet Reale D. Interplay between plasma oxidative status, cortisol and coping styles Intern Med 2013;27:157e63. in wild alpine marmots, marmota marmot. J Exp Biol 2012;215:374e83.