Ten healthy horses (average age: 3.5 years old): five Jeju horses (Jeju group; n = 5) and five Thoroughbred horses (Thoroughbred group; n = 5) were used in the study. Heights and weights of Jeju horses were 115 to 125 cm and 244 to 249 kg, respectively. Whereas, heights and body weights of Thoroughbred horses were 150 to 173 cm and 450 to 500 kg, respectively. For health status of every horse was monitored for a week until the start date of experiment. Horses were observed without signs of injury, illness and medical records including; vital signs (heart rate, temperature, and respiration rate), ear, eye, nose, skin, coat, feet, limb, appetite, and attitude (bright and alert). Guidelines of horse managements were in compliance with international standards and “Korea Racing Authority” (http://www.kra.co.kr). All procedures used in the experiment were approved by the Pusan National University-Institutional Animal Care and Use Committee (Approval Number: PNU-2015-0864).

Horses were exercised (longeing; circle diameter 11 m) for 30 min without rest. Before and after exercise, rectal temperature of each horse was measured by thermometer. Heart rate was measured by stethoscope in beats per minute. Blood samples (approximately 20 mL) were collected from the jugular vein. Samples were then separated using BD Vacutainer spray-coated K₂ ethylenediaminetetraacetic acid (EDTA) Tubes for hematological analysis. BD Vacutainer SST Tubes contain spray-coated silica and a polymer gel were used to facilitate serum biochemical and immunological analysis. BD Vacutainer Citrate Tubes and a 3.2% buffered sodium citrate solution were used in routine fibrinogen analyses. For each blood sample, the following biochemical parameters were assessed: cortisol level; activities of aspartate aminotransferase, alanine aminotransferase, lactate dehydrogenase, creatine kinase, and alkaline phosphatase; total bilirubin; blood urea nitrogen (BUN); and creatinine (Cr) levels. Hematological parameters, including red blood cell (RBC) count, hemoglobin (Hb), hematocrit (Hct), mean cell (or corpuscular) volume (MCV), mean cell hemoglobin (MCH), mean cell hemoglobin concentration (MCHC), white blood cell (WBC) count, proportion of blood immune cells (percentage of lymphocytes, neutrophils, monocytes, eosinophils, and basophils), platelet count, fibrinogen level and immunological parameters, including immunoglobulin (Ig) G and IgM, were measured using an Auto Analyzer (Seegene, Seoul, Korea). The BUN/Cr ratio was also calculated to assess muscle catabolism.

One drop of blood from each sample was placed on a glass slide and gently spread using the edge of another glass slide. Blood smears were then air-dried, fixed in methanol, and stained according to the standard use of Leishman-Giemsa stain (Sigma-Aldrich, St. Louis, MO, USA).

Plasma samples were centrifuged at 4°C at 16,000×g for 5 min to remove cell debris. The supernatant was diluted in sterile nuclease-free water at a 1:40 ratio for direct quantitative polymerase chain reaction (qPCR) measurement. The diluted plasma samples were stored at −20°C until they were analyzed. The cell free DNA (cfDNA) concentrations of the plasma were quantified using direct qPCR and a primer set (Table 1), producing an 88-bp fragment of the chromosomal myostatin (MSTN) amplicon [13]. Acting as a template, the diluted plasma was added to master mixer 20X Evagreen (SolGent, Seoul, Korea) containing 10 pmol primer set, 25 mM MgCl₂, 10 mM dNTPs, and 0.5 U BIOFACT Taq DNA polymerase (BIOFACT, Seoul, Korea). Amplification was completed over 40 cycles.

Peripheral blood mononuclear cells (PBMCs) and polymorphonuclear cells (PMNs) were isolated by the single-step centrifugation of whole blood samples onto a Polymorphprep application sheet (Axis-Shield, Oslo, Norway) according to the manufacturer’s recommendations. Blood in the EDTA tube was layered onto the application sheet at a ratio of 1:1 and centrifuged at 500×g for 35 min. The PMN (granulocytes) and PBMC (lymphocyte and monocyte) layers were carefully collected and resuspended in 1× phosphate-buffered saline (PBS) and then centrifuged at 14,000×g for 5 min to remove the supernatant. Cell pellets were stored at −80°C while awaiting RNA extraction.

Total RNA was extracted from horse PBMCs and PMNs using TRIzol (Invitrogen, Karlsruhe, Germany) according to the manufacturer’s instructions. The purity of the extracted RNA was confirmed by measuring absorbance at 260 nm and 280 nm using a spectrophotometer (ND-1000, Nanodrop Technologies Inc., Wilmington, DE, USA). RNA of purity (optical density value of 260 nm/280 nm) greater than 1.8 was selected for further analysis and was stored at −80°C until further analysis. To synthesize cDNA, 1 μg of RNA was reverse transcribed for each sample using the SuperScript III First-Strand Synthesis System (Invitrogen, Germany), according to the manufacturer’s instructions.

The NCBI (http://www.ncbi.nlm.nih.gov) and the Ensembl Genome Browser (www.ensembl.org) were used to retrieve gene sequence information. The primers used in the amplification of the genes (Table 1) were synthesized using the PRIMER3 software (http://bioinfo.ut.ee/primer3-0.4.0/). Reverse transcriptase (RT)-PCR and real-time qPCR were carried out using a C1000 Thermal Cycler (Bio Rad, Hercules, CA, USA) to measure the relevant expression of target genes. PCR products were separated using agarose gel electrophoresis and detected under UV light. Real-time qPCR was performed using master mixture Evagreen for 40 cycles. All measurements were carried out in triplicate for each sample, and the 2^−ΔΔCt method was used to determine relative gene expression. Gene expression was normalized using glyceraldehyde 3-phosphate dehydrogenase.

Statistical analyses were performed using GraphPad Prism 6. Data are represented as the means±standard error of the mean of three or five independent samples. The results were analyzed using analysis of variance and Student’s t-test and were considered significant when * p<0.05, ** p<0.01 or *** p<0.001.

To determine whether 30 min of exercise differentially alters the blood metabolic parameters of Jeju and Thoroughbred horses, blood biochemical analyses were performed. Blood metabolite concentrations that reflect the enzymatic functions of the liver, bones and kidneys, also, immunological parameter including IgG, IgM level, WBC count and proportion of blood immune cells were not significantly affected by exercise in either of the two breeds (Table 2). However, 30 min of exercise significantly elevated rectal temperature (Figure 1A), heart rate (Figure 1B), and cortisol levels (Figure 1C) in both breeds. Although BUN and Cr levels were not significantly altered by exercise, the BUN/Cr ratio, which is negatively correlated with muscle catabolism, significantly decreased in Jeju horses but not in Thoroughbred horses (Figure 1D). Taken together, the results indicate that 30 min of exercise is sufficient to induce a stress response in horses that is characterized by increased body temperature, blood cortisol level, heart rate and muscle catabolism. We also note that the muscle catabolism-related responses of the Jeju and Thoroughbred horses differed.

During exercise, the cardiovascular system increases blood flow to facilitate O₂ transportation to muscle tissues [14]. We assessed exercise induced any changes in the RBC count, Hb level, and Hct of the blood of the two breeds in order to compare their oxygen transportation-related responses. After exercise, RBC count, Hb, and Hct tended to increase in Thoroughbred horses, resulting in Thoroughbred horses having significantly higher RBC counts and Hb and Hct levels than those in Jeju horses (Figure 2A). Next, we observed the RBC characteristics of blood samples from both breeds using light microscopy and found that RBCs of both breeds formed a rouleaux formation and were similar in size with lack of a central pallor. Then, we evaluated the RBC indices, including MCV as a marker for RBC size, MCH as an average amount of Hb per RBC, and MCHC as an indicator of the amount of Hb per unit volume. The MCV and MCHC did not differ between the two breeds. Therefore, the results from the blood parameter analysis and the blood smear assessment were consistent. Nevertheless, MCH, which represents the average amount of oxygen-carrying Hb within one RBC, was significantly higher in Thoroughbred horses than that of Jeju horses in both before and after exercise (Figure 2B). Collectively, these results suggest that Thoroughbred horses may have an increased capacity for RBC-based oxygen delivery during exercise.

According to a study of humans, exercise can induce the release of DNA from neutrophils into extracellular space, forming neutrophil extracellular traps (NETs) [15]. NETosis, a process of NET, can be triggered by various stimuli in response to infection, tissue damage, and inflammation. In sterile inflammation, increased high mobility group box 1, citrullinated histones, and platelet-thrombin activation potentiate NETosis [16]. During NETosis, the nuclei of neutrophils swell abnormally, a process indicative of the release of decondensed chromatin and granular contents. We examined whether NETosis can be induced by exercise in horses. Blood smear samples from before and after exercise were Giemsa stained and observed under a microscope. Blood smear analysis revealed that NETosis occurred in horse blood after exercise, more so in Jeju horses (Figure 3A). Continued exercise causes tissue damage, which stimulates increases in cell free (cf) DNA in the blood [17], with one mechanism of cfDNA release being suicidal NETosis. Therefore, we measured cfDNA in the plasma both before and after exercise. Exercise did not affect cfDNA levels in either breed. However, Thoroughbred horses showed higher cfDNA levels than Jeju horses both before and after exercise. In addition to being triggered by microorganisms and tissue damage, NET formation is also triggered by platelets. NETs further stimulate platelet release and thrombosis due to their externalized chromatin, and initiate the coagulation cascade [16]. Therefore, we assessed the platelet count and fibrinogen content of blood samples, both of which are central mediators of hemostasis during coagulation. We found that platelet count was significantly reduced in Jeju horses after exercise, while platelet count in Thoroughbred horses remained unchanged (Figure 3C). Unlike platelet count, fibrinogen level was not significantly different between the breeds after exercise (Figure 3D). Therefore, NETs in Jeju horses respond more rapidly and sensitively to exercise-induced stresses than Thoroughbred horses, likely associated with differential platelet responses (cfDNA level remained unchanged and thus likely had no effect on the response of NETs).

Exercise induces endoplasmic reticulum, oxidative, and inflammatory stress in the muscle tissue [18]. Our previous study of horse muscle cells showed that heat stress increases the expression of various stress-related genes [19]. Although, it is not clear whether stresses induced by exercise can affect WBCs, neutrophils and monocytes are mainly associated with the recognition of stress-induced molecules via various receptors. As immune cell function, such as NET and platelet levels, differ between Jeju and Thoroughbred horses, their WBCs may differ in their functions and responses to exercise-induced stress. Thus, we determined whether exercise-induced stress-related gene expression took place in blood immune cells. We investigated the gene expression of stress-related genes in two populations of WBCs; PBMCs and PMNs including heat shock protein 72 (HSP72), heat shock protein family A (Hsp70) member 6 (HSPA6), and nuclear factor erythroid 2 (Nfe2)-related factor 2 (NRF2), which are prosurvival molecules to protect against variety of stressors. The results showed that after 30 min of exercise, HSPA6 expression in the PBMCs of Jeju horses was higher than in Thoroughbred horses. In Jeju horses, HSPA6 expression in PBMCs was significantly higher than in PMNs. However, the expression of HSP72 and NRF2 was not significantly different between the two breeds, and WBC subpopulations also did not vary significantly (Figure 4A–B). In our previous study of horse muscle cells, the ratio of the splicing form of X-box binding protein 1 (sXBP1) gene to total XBP1, which is a common indicator of endoplasmic reticulum (ER)-stress, was increased after exposure to heat-stress [19]. We investigated whether sXBP1 in PBMCs may be affected by exercise-induced stress. However, the ratio of sXBP1 to XBP1 in the PBMCs in both breeds was not affected by exercise (Figure 4C). The results suggest that PBMCs and PMNs in both breeds have unique responses to exercise-induced stress. Additionally, HSPA6 was upregulated more sensitively in Jeju horse PBMCs than in the PMBCs of Thoroughbred horses.

In the blood, cytokine and chemokine receptors are mainly expressed in immune cells, and the upregulation of cytokines and chemokines involve crosstalk between immune cell functions and muscle regeneration after stress-induced tissue damage. In our previous study, interleukin 6 (IL-6), IL-8, chemokine (C-C motif) ligand 4 (CCL4), and chemokine (C-X-C motif) ligand 2 (CXCL2) expression was significantly increased after exercise in the skeletal muscles of both Thoroughbred and Jeju horses [20]. As 30 min of exercise can induce stress-related gene expression in PBMCs, we hypothesized that exercise-induced stress could alter immunogenic markers in horse lymphocytes and monocytes. Thus, we examined the expression of various genes including cytokines, chemokines, and inflammasome signaling genes (IL-6, IL-8, IL-1B, CCL2, nucleotide-binding oligomerization domain, leucine rich repeat and pyrin domain containing 3, and caspase 1). We found that exercise did not affect the expression of any of the assessed genes (data not shown). Next, we assessed whether exercise could induce changes in the expression of various chemokine receptors, including C-C chemokine receptor 2 (CCR2), CCR3, CCR9, CCR10, CXC chemokine receptor 1 (CXCR1), CXCR2, CXCR4, purinergic receptor P2X 7, and beta-2 adrenergic receptor in PBMCs. The results showed that the expression of only CXCR4 was significantly altered after exercise in the PBMCs of Thoroughbred horses (Figure 5A). Furthermore, CXCR4 expression in Thoroughbred horses was significantly higher than in Jeju horses. (Figure 5B). Overall, the upregulation of CXCR4 after 30 min of exercise indicates that Thoroughbred horse lymphocytes may respond directly to CXCR4-ligands produced by muscle injury more rapidly than Jeju horse lymphocytes.