Equine herpesvirus type 1 is a highly contagious, double-stranded DNA virus associated with outbreaks of respiratory and neurologic disease, abortion, and neonatal death in horses. Infectious outbreaks cause severe economic losses to the racing industry and have a financial and emotional toll on owners of affected horses.1 When horses are infected with EHV-1, thrombi develop within various blood vessels, including those that supply the spinal cord and placenta.2–4 These thrombi are believed to contribute to the pathogenesis of the clinical syndromes of abortion and neurologic disease through ischemic injury.
The exact mechanism leading to thrombus formation remains unclear; however, we have shown that EHV-1 induces tissue factor expression on equine monocytes ex vivo.5 In addition, equine platelets exposed ex vivo to EHV-1 have α-granule secretion (characterized by P-selectin expression on platelet surfaces) and microvesiculation (characterized by release of small membrane-bound particles expressing phosphatidylserine).6
These EHV-1-induced responses could potentially contribute to creation of a procoagulant and proinflammatory state, priming an infected horse for thrombosis and subsequent fetal and neurologic injury. For instance, P-selectin expression on activated platelets could facilitate direct platelet adhesion to endothelial cells (bound platelets form the physical foundation of the thrombus) or could promote leukocyte recruitment and activation via binding to PSGL-1, which is expressed on endothelial cells and leukocytes.7–10 The simultaneous expression of negatively charged phospholipids and release of PDMPs by activated platelets would accelerate thrombin generation and thrombus formation. Negatively charged phospholipid surfaces, particularly those expressed on PDMPs, serve as binding sites for procoagulant enzyme complexes and amplify the activity of those enzymes over a thousand fold.11 Activated platelets also release coagulation factors, producing localized high concentrations that could then participate in thrombin generation.12,13
Both thrombosis and tissue inflammation can be reduced by inhibiting the interaction of P-selectin with PSGL-1 in mice in vivo.14,15 Because EHV-1 activates platelets ex vivo, inducing expression of P-selectin (a proinflammatory mediator) and microvesiculation (a strong procoagulant stimulus),6 the possibility exists that platelet activation may contribute to the thrombosis and vasculitis identified in the tissues of infected horses. Therefore, inhibition of these platelet activation responses may help prevent or limit these pathological sequelae in horses infected with EHV-1.
Various drugs with potential platelet inhibitory effects are used clinically in horses, including theophylline (the active ingredient of aminophylline), pentoxifylline, acetylsalicylic acid, and clopidogrel bisulfate. Several ex vivo studies16–20 involving equine platelets have shown that these drugs have variable inhibitory effects on platelet activation induced by agonists such as thrombin, convulxin, ADP, platelet-activating factor, and collagen. However, whether any of these drugs can inhibit EHV-1-induced platelet responses is unknown. Therefore, the objective of the study reported here was to evaluate the ability of these drugs to inhibit platelet P-selectin exposure and microvesiculation induced by EHV-1 ex vivo.
Materials and Methods
Animals
Nine healthy adult mares and 3 stallions (5 Thoroughbreds, 4 warmbloods, 2 Quarter Horses, and 1 Oldenburg) belonging to the Cornell University Equine Park were used in the study. Ages ranged from 8 to 25 years (median, 11.5 years), and body weights estimated by use of a body tape measure ranged from 514 to 636 kg (median, 585 kg). Each horse was judged to be healthy on the basis of prior history and results of physical examination and clinicopathologic testing. None had received any medication for at least 3 weeks before treatment sessions began. Horses were kept in small paddocks for the duration of the drug assessment period and for 10 days after the final dose was administered in the final treatment session for monitoring for possible hemorrhagic or other adverse reactions to the drugs. The study was performed in compliance with Institutional Laboratory Animal Care and Use Committee guidelines (protocol No. 2013-0073).
Study design and drug administration
A double-blinded, placebo-controlled crossover study design was used. Horses were randomly allocated by flip of a coin to receive 1 of 5 treatments first: theophyllinea (5 mg/kg, q 12 h), pentoxifyllineb (10 mg/kg, q 12 h), clopidogrel bisulfatec (4 mg/kg, q 24 h), acetylsalicylic acidd (20 mg/kg, q 24 h), or placebo (approx 25 mg of dextrose powder,e q 12 h). All horses received 2 doses/d; horses treated once daily with drugs were given the placebo for the second dose.
The first treatment was administered for 5 days (beginning on day 1), a 3-week washout period was provided, and then the next randomly assigned treatment was administered for 5 days. This process was repeated until all horses had received all 5 treatments. Dosages used were based on standard dosages (theophylline or pentoxifylline) or published reports of these drugs in horses,17,18 although a higher maintenance dosage of clopidogrel was given to overcome interindividual variability and possible drug resistance.18
All drugs were dispensed by the pharmacy of the Cornell University Hospital for Animals as tablets or powder. Because one of the drugs (clopidogrel) was a pink tablet and the others were white powders or tablets, small amounts of red beverage crystalsf were added to each white powder or ground white tablet (to form a powder) to mask their identities. Colored powders were converted into a paste by use of molasses, and these slurries were transferred to 60-mL syringes by 2 personnel who randomized the treatments but did not administer the drugs, perform experimental analyses, or interpret data.
To mimic clinical dose administration practices, test compounds were given orally via syringe. All personnel involved in drug administration, data acquisition, or experimental analyses were blinded as to the sequence and identities of drugs administered. Before and at completion of each 5-day treatment period, all horses were examined for clinical evidence of platelet dysfunction, specifically for obvious oral mucosal bleeding and general bruising or petechiation.
Sample collection and processing
The day before the first dose of each treatment was administered (day 0) and 4 hours after administration of the final dose on day 5 of each treatment session, a blood sample was collected from each horse via jugular venipuncture by use of an 18-gauge needleg and 12-mL plastic syringeh prefilled with 1.2 mL of 3.8% sodium citrate solution. The needle was inserted first, and blood was allowed to drip for a few seconds before the syringe was attached to gently withdraw 10.8 mL of blood, maintaining an anticoagulant-to-blood ratio of 1:9. The collected sample was gently transferred into a 15-mL polypropylene tube.
To test for potential adverse effects of the drugs, another blood sample was also collected on day 0 for each treatment and 4 hours after the final dose for the last treatment. The same jugular vein from which the first blood sample (above the initial draw site for the citrate tube) was obtained or the contralateral jugular vein was used, and blood was collected into evacuated tubes containing EDTAi or no additional solution by use of a 20-gauge needlej for measurement of specific CBC variablesk (hemoglobin concentration, WBC count, and platelet count) and serum biochemical variablesl (activities of aspartate aminotransferase, sorbitol dehydrogenase, glutamate dehydrogenase, and γ-glutamyltransferase and concentrations of BUN, creatinine, glucose, and total protein).
All samples were transported to the laboratory for further processing within 30 to 45 minutes after collection. Citrated blood was centrifuged at 450 × g for 5 minutes at 21°C to isolate PRP for a platelet count and optical aggregometry and flow cytometry assays. Platelets were allowed to rest at 20° to 23°C before the assays.
Platelet exposure to EHV-1
Equine herpesvirus type 1 strains Ab4 (isolated from a quadriplegic mare4) and RacL11 (isolated from an aborted fetus21) were propagated in RK13 cells and purified from cell lysates on a sucrose cushion, as described in detail elsewhere.6 Lysate of nonvirally exposed RK13 cells was similarly treated and used as a mock-exposure control substance. Platelets in PRP were diluted to a final concentration of 1 × 106/mL in flow buffer (10mM HEPES and 140mM sodium chloride; pH 7.4) with supplemental gly-pro-arg-pro-NH2m (fibrin polymerization inhibitor) and calcium chloride (2.5mM). Platelets (5 × 105) were exposed to the strains of EHV-1 at 1 plaque forming unit/cell or to a positive control substance (mixture of thrombinn [0.15 U/mL] and convulxin° [0.05 μg/mL]) or negative control substance (PBS solution or nonvirally exposed RK13 cell lysate) for 10 minutes at 37°C.
Flow cytometry assays
Platelet activation status was assessed via flow cytometry by quantifying the percentages of P-selectin-expressing platelets and shed PDMPs. The procedure for labeling and quantification is described elsewhere.6 In brief, after exposure to virus or control substances, platelets were triple-labeled by the addition of a phycoerythrin-conjugated antibody against CD41p (a platelet-specific marker; 1:10 final dilution) for 10 minutes in the dark at 20° to 23°C, followed by the addition of fluorescein isothiocyanate–conjugated annexin Vq (final dilution, 1:300) and an allophycocyanin-conjugated anti–P-selectin antibodyr (final concentration, 33.3 ng/mL) for 10 minutes in the dark at 20° to 23°C. The reaction was quenched with 400 μL of flow buffer and analyzed with a flow cytometer.s
Platelets and PDMPs were quantified by use of log settings for forward and side scatter. For measurement of P-selectin, platelets were gated on their characteristic forward and side scatter properties (events between the first and second log decade for both scatter scales), then the percentage P-selectin–positive platelets was quantified by use of histogram plots, with the marker (positive) region defined by an isotype control. For measurement of PDMPs, CD41-positive events were gated and then the percentage of PDMPs was quantified as annexin V–positive events below the first decade of the log forward scatter scale on a dot plot of annexin V fluorescence versus forward scatter.
Platelet aggregometry assay
Platelet aggregation responses were measured by means of light transmission aggregometryt A 1.0-mL sample of PRP was centrifuged for 1 minute at 14.1 × g to yield platelet-poor plasma, which was used as a blank. The ADP agonistu was prepared in accordance with package directions and diluted with saline (0.9% NaCl) solution to achieve a stock concentration of 100μM. Samples of PRP were placed in siliconized glass cuvettes, and reactions were performed at 37°C, with stirring at 1,200 cycles/min. After an initial warming period of 2 minutes, either ADP (10μM final concentration) or saline solution (negative control substance) was added. After 6 minutes, an aggregation reading was performed and the maximal percentage aggregation and AUC were calculated by the equipment manufacturer's software program.
Statistical analysis
Data were tested for normality of distribution by use of the Shapiro-Wilk test.v Because the assumption of normality was not met, all data are reported as median (range). Results obtained from each of the assays (percentage of P-selectin–positive platelets, percentage of PDMPs, maximal percentage aggregation, and AUC) before and after each treatment session for each drug were compared with the paired sample sign test.w Differences among treatments at either measurement point were evaluated by use of the Kruskal-Wallis test followed by the all-pairwise comparison post hoc test (all-pairwise comparisons of mean ranks) when appropriate. Values of P < 0.05 were considered significant.
Results
Treatment-related adverse effects
Except for 1 horse, which was euthanized because of an unrelated episode of severe colic midway through the study, all horses appeared healthy with unremarkable vital signs throughout the study period. None of the horses developed obvious clinical signs associated with platelet dysfunction. Data from the euthanized horse were excluded from the analyses, leaving 11 horses in the study.
No significant differences in values of selected CBC variables were identified between days 0 and 5 (before and after each treatment session) for any treatment (data not shown). Comparisons among treatments revealed no significant differences in CBC values on either measurement day. Comparison between days 0 and 5 revealed a significant difference in median values of some measured serum biochemical variables for some treatments. A moderate decrease in serum glutamate dehydrogenase and aspartate aminotransferase activities was detected in horses on day 5 of the pentoxifylline and clopidogrel treatment sessions, respectively, whereas a mild increase in serum creatinine concentration was identified in horses on day 5 of the acetylsalicylic acid treatment session. However, these changes were small and unlikely to have been clinically relevant. No significant differences were detected in median serum biochemical values between days 0 and 5 for the other treatments, nor were any differences identified among treatments on day 0 or 5 (data not shown).
Effect of drugs on platelet P-selectin expression and microvesiculation
After exposure of PRP samples to either strain of EHV-1 or the negative control substances, no significant differences in median percentage of P-selectin–positive platelets were identified between days 0 and 5 for each treatment or among treatments on day 0 or 5. A significant (P = 0.02) increase from day 0 in median percentage of P-selectin–positive platelets was identified for PRP samples obtained on day 5 of the clopidogrel treatment session that were subsequently exposed to thrombin-convulxin (positive control substance; Table 1). This finding was attributed to the response of 4 horses, in which unexpectedly low P-selectin expression was identified with the thrombin-convulxin control substance for PRP samples obtained on day 0. In particular, 1 horse had a marked increase from 20% on day 0 to 90% on day 5 in percentage of P-selectin–expressing platelets with the positive control reaction during the clopidogrel treatment session.
Median (range) percentage of P-selectin–positive platelets in PRP samples obtained from 11 horses the day before (D0) and 4 hours after 5 days (D5) of treatment with various platelet inhibitors or placebo after exposure of samples to 2 strains of EHV-1 (RacL11 and Ab4) or positive (thrombin-convulxin) or negative (PBS solution or lysate of nonvirally exposed RK13 cells) control substances.
| Exposure substance | Point | Pentoxyfilline (10 mg/kg, q 12 h) | Theophylline (5 mg/kg, q 12 h) | Clopidogrel bisulfate (4 mg/kg, q 24 h) | Acetylsalicylic acid (20 mg/kg, q 24 h) | Placebo |
|---|---|---|---|---|---|---|
| PBS solution | D0 | 0.9 (0.2–4.7) | 1.1 (0.2–2.7) | 1.1 (0.3–4.2) | 0.9 (0.1–2.6) | 0.7 (0.1–5.5) |
| D5 | 0.7 (0.2–2.9) | 0.7 (0.2–3.2) | 0.6 (0.1–2.6) | 0.9 (0.1–7.2) | 1.0 (0.1–5.8) | |
| RK13 cell lysate | DO | 0.6 (0.4–16.8) | 0.8 (0.0–5.5) | 0.9 (0.2–8.8) | 0.7 (0.3–4.1) | 0.6 (0.6–3.4) |
| D5 | 0.6 (0.1–1.4) | 0.5 (0.2–2.3) | 0.6 (0.2–2.2) | 0.4 (0.1–3.9) | 0.6 (0.1–5.2) | |
| Thrombin-convulxin | D0 | 78.5 (53.0–96.4) | 90.8 (39.6–96.4) | 83.2 (21.6–95.4) | 92.8 (21.6–95.7) | 87.5 (37.6–94.8) |
| D5 | 91.4 (41.5–94.7) | 92.7 (88.7–95.4) | 93.7* (87.7–95.5) | 91.1 (66.7–94.3) | 87.0 (69.5–95.2) | |
| RacL11 | D0 | 56.1 (0.7–72.8) | 51.9 (9.6–66.2) | 50.6 (31.7–68.6) | 51.1 (10.2–72.2) | 50.7 (12.9–74.4) |
| D5 | 61.3 (24.2–76.6) | 57.7 (14.4–80.5) | 62.5 (15.9–81.9) | 57.4 (18.8–81.1) | 55.6 (18.3–80.8) | |
| Ab4 | D0 | 83.5 (0.6–91.7) | 73.7 (10.8–88.9) | 77.0 (57.4–90.9) | 65.2 (1.1–90.7) | 86.2 (1.9–91.3) |
| D5 | 82.3 (2.6–92.1) | 83.0 (2.4–90.0) | 82.5 (22.7–91.7) | 81.9 (4.5–88.8) | 82.0 (3.4–89.4) |
Value differs significantly (P = 0.01) between measurement points.
Horses received all treatments in random order, with each 5-day treatment session separated by a 3-week washout period.
When the data from these 4 horses were examined, substantial microvesiculation was noticed in the day 0 PRP samples, which we speculated was attributable to collection- or handling-induced sensitization of platelets (prestimulation). This microvesiculation can be associated with a decrease in P-selectin expression6,18 and would explain the relatively impaired response to thrombin-convulxin in pretreatment samples. Removal of the data from these 4 horses resulted in the difference between day 0 (median value, 85.8%) and day 5 (94.1%) values for clopidogrel no longer being significant (P = 0.13).
No significant differences in median percentage of PDMPs were identified between days 0 and 5 for each treatment or among treatments on day 0 or 5 when PRP samples were exposed to either strain of EHV-1 or the positive and negative control substances (Table 2). In contrast to findings for P-selectin, treatments involving the standard platelet inhibitors (with the exception of acetylsalicylic acid), but not the placebo, generally resulted in a decrease in median percentage of PDMPs, but the difference between days 0 and 5 was quite small, possibly owing to biological variation rather than a true drug effect. Exposure of PRP samples to EHV-1 strain RacL11 consistently yielded more PDMPs than did exposure to EHV-1 strain Ab4, with subsequently lower percentages of P-selectin–positive platelets (Table 1).
Median (range) percentage of PDMPs in PRP samples obtained from 11 horses the day before (D0) and 4 hours after 5 days (D5) of treatment with various platelet inhibitors or placebo after exposure of samples to the same conditions as in Table 1.
| Exposure substance | Point | Pentoxyfilline (10 mg/kg, q 12 h) | Theophylline (5 mg/kg, q 12 h) | Clopidogrel bisulfate (4 mg/kg, q 24 h) | Acetylsalicylic acid (20 mg/kg, q 24 h) | Placebo |
|---|---|---|---|---|---|---|
| PBS solution | D0 | 1.7 (1.0–3.6) | 2.0 (0.2–5.6) | 2.1 (1.2–7.8) | 1.8 (0.5–3.3) | 1.9 (0.3–17.9) |
| D5 | 1.5 (0.4–4.9) | 2.4 (0.2–4.1) | 1.9 (0.3–4.9) | 2.3 (0.5–7.6) | 2.6 (0.5–4.7) | |
| RK13 cell lysate | D0 | 1.6 (0.6–22.8) | 1.4 (0.1–3.7) | 2.0 (0.4–19.1) | 1.7 (0.9–24.3) | 1.3 (0.3–12.5) |
| D5 | 1.8 (0.7–3.6) | 2.1 (0.4–9.7) | 2.1 (0.9–4.8) | 1.7 (0.4–3.9) | 1.8 (0.4–2.5) | |
| Thrombin-convulxin | D0 | 12.7 (1.6–39.6) | 19.8 (5.9–34.8) | 10.3 (1.1–50.0) | 9.2 (0.9–47.5) | 12.3 (4.8–36.3) |
| D5 | 16.6 (3.5–44.9) | 11.3 (2.9–33.7) | 12.2 (1.6–48.0) | 12.3 (5.5–54.3) | 11.5 (5.2–49.4) | |
| RacL11 | D0 | 16.8 (1.6–34.9) | 15.3 (6.8–37.3) | 15.5 (5.5–31.0) | 13.3 (3.7–49.3) | 12.7 (6.5–39.5) |
| D5 | 13.4 (4.5–22.0) | 12.6 (8.1–29.8) | 11.3 (7.6–28.0) | 16.4 (8.7–32.1) | 15.0 (6.3–52.2) | |
| Ab4 | D0 | 7.1 (3.1–19.7) | 4.8 (3.5–16.8) | 5.4 (3.0–23.8) | 6.0 (2.8–15.9) | 5.2 (3.4–11.0) |
| D5 | 6.2 (3.1–8.8) | 4.6 (2.6–10.2) | 5.8 (2.8–11.2) | 6.9 (2.7–18.5) | 5.9 (2.8–20.2) |
No significant differences were identified between measurement points for each treatment or among treatments at each measurement point.
See Table 1 for remainder of key.
Effect of standard platelet inhibitors on platelet aggregation
An inhibitory effect on ADP-stimulated platelet aggregation (maximal percentage aggregation and AUC) was identified on day 5 of the clopidogrel treatment session (compared with day 0), whereas no inhibitory effect on platelet aggregation was identified following the other treatment sessions. Accordingly, day 5 median values for maximal percentage aggregation and AUC after ADP stimulation were significantly (P < 0.01) lower for clopidogrel than for the other drugs or placebo on day 5 of treatment.
On the other hand, no significant difference was identified in day 0 aggregation responses to ADP for any treatment (Table 3). An inhibitory effect on platelet aggregation was detected in PRP samples of all horses after receiving clopidogrel, although the inhibitory response was weak (< 50% inhibition) in 1 horse (Figure 1).
Degrees of ADP-induced platelet aggregation (maximal aggregation percentage [A] and AUC [B]) in PRP samples from individual horses (n = 11) the day before and 4 hours after 5 days of treatment with clopidogrel bisulfate (4 mg/kg, PO, q 24 h). In all horses, aggregation was inhibited by clopidogrel, with a weak inhibitory response (< 50% inhibition) identified in 1 horse (white squares).
Citation: American Journal of Veterinary Research 77, 12; 10.2460/ajvr.77.12.1366
Median (range) values for ADP-induced platelet aggregation responses in PRP samples obtained from 11 horses the day before (D0) and 4 hours after 5 days (D5) of treatment with various platelet inhibitors or placebo.
| Variable | Point | Pentoxyfilline (10 mg/kg, q 12 h) | Theophylline (5 mg/kg, q 12 h) | Clopidogrel bisulfate (4 mg/kg, q 24 h) | Acetylsalicylic acid (20 mg/kg, q 24 h) | Placebo |
|---|---|---|---|---|---|---|
| Maximal | D0 | 66 (33–78) | 68 (19–88) | 65 (27–90) | 68 (22–83) | 62 (2–82) |
| percentage | D5 | 64 (10–8) | 65 (46–80) | 18*† (10–50) | 59 (31–78) | 62 (17–88) |
| AUC | D0 | 307 (143–362) | 314 (89–390) | 305 (126–419) | 320 (111–372) | 283 (11–372) |
| D5 | 295 (41–400) | 289 (201–374) | 67*† (19–274) | 291 (147–356) | 277 (75–374) |
Value differs significantly (P ≤ 0.001) between measurement points.
Value differs significantly (P ≤ 0.001) from values on the same day for other treatments.
See Table 1 for remainder of key.
Discussion
The clinical relevance of EHV-1–induced platelet activation is as yet unknown. We speculate that platelets could be exposed to and activated by EHV-1 during the viremic phase of infection (when released from peripheral blood mononuclear cells harboring the virus) or from virus-infected endothelial cells within the lungs (the route of infection) or distal infected sites in horses. Platelet expression of P-selectin induced by EHV-1 could then promote the formation of platelet-leukocyte aggregates via interactions between P-selectin and PSGL-1,8 leading to leukocyte activation and, potentially, subsequent endothelial infection.22,23 Alternatively, EHV-1-activated platelets may bind directly to endothelial cells expressing PSGL-1, thereby transporting any bound virus to these cells or promoting leukocyte recruitment, leading to vasculitis.24,25
Microvesiculation of activated platelets would also provide a large surface area for assembly of coagulation factor complexes, thereby promoting thrombin generation. Thrombin generation by procoagulant PDMPs may be further potentiated by the EHV-1–induced expression of tissue factor on monocytes, resulting in enhanced activation of coagulation proteases.5 Thus, inhibition of these EHV-1-induced platelet responses would seem worthwhile to pursue as a means of limiting EHV-1-associated thrombosis or virus dissemination in vivo.
Findings of the present study suggested that the standard platelet inhibitors evaluated failed to inhibit P-selectin expression or release of annexin V-positive PDMPs induced by platelet exposure to either strain of EHV-1. The lack of any inhibitory effect of the drugs on these markers of platelet activation may have been related to their mechanisms of action, which do not directly target the platelet thrombin receptor. Indeed, platelets exposed to EHV-1 had an activation response similar to that of platelets exposed to the positive control substance (thrombin-convulxin), corroborating our previously reported findings of EHV-1-induced α-granule secretion and microvesiculation by equine platelets.6
Thrombin activates platelets by binding to protease-activated receptors, which then trigger downstream signaling cascades that lead to rapid α-granule secretion in most of the exposed platelets. Thrombin combined with convulxin induces microvesiculation in equine platelets to a similar extent as that induced by EHV-1.6,18 Convulxin is an agonist of glycoprotein VI, which is a surface receptor that binds to and is activated by collagen. Microvesiculation requires sustained high intraplatelet calcium concentration and does not occur in the absence of exogenous calcium to support store-operated calcium entry. As in a previous study,6 we found that exposure of equine platelets to EHV-1 induced robust microvesiculation similar to that of calcium ionophore, which induces robust microvesiculation at the expense of P-selectin expression.
Of the drugs evaluated in the present study, only clopidogrel inhibited ADP-induced platelet aggregation. This could be attributed to its active metabolite acting as an antagonist of the P2Y12 ADP receptor,17,18 thereby preferentially inhibiting ADP-induced aggregation without affecting other activation markers, such as P-selectin expression or microvesiculation. Platelet aggregation in PRP samples from all horses was inhibited by clopidogrel, in contrast to findings of our previous study18 in which some horses appeared to be nonresponders. This could have been due to the higher doses and longer treatment periods used in the present study or interindividual variability caused by genetic differences in drug metabolism, as described for humans.26,27 The higher-dose hypothesis could not be directly tested because we did not use the same horses in these 2 studies. We have not yet investigated whether EHV-1 can stimulate platelet aggregation in horses (via thrombin-activated receptors or thrombin-mediated ADP secretion) nor whether any virus-induced aggregation responses would be inhibited by clopidogrel.
Acetylsalicylic acid acts by irreversibly blocking cyclooxygenase production of thromboxane from arachidonic acid. The lack of an inhibitory effect of acetylsalicylic acid on any of the platelet activation markers used in the present study was likely attributable to the fact that equine platelets do not require thromboxane generation to undergo aggregation, α-granule secretion, or microvesiculation.28,29 Two nonspecific PDE inhibitors, theophylline and pentoxifylline, were also evaluated in the present study. Although these drugs are more commonly used as bronchodilators in horses with respiratory disease (aminophylline)30,31 or as rheological and anti-inflammatory treatments for horses with endotoxemia, laminitis, or navicular disease (pentoxifylline),32–34 they have been used anecdotally as antithrombotic agents. Neither of these drugs inhibited EHV-1- or thrombin-convulxin–induced α-granule secretion or ADP-induced platelet aggregation.
The results for pentoxifylline were similar to those of a previous study19 in which this drug, when added to whole blood ex vivo, did not inhibit platelet aggregation stimulated by ADP or collagen. In contrast, other studies have shown that theophylline and IBMX, which is another nonspecific PDE inhibitor, suppress ADP-induced platelet aggregation20 and thrombin-induced P-selectin expression,35 respectively, in equine PRP samples.
The discordance between results of the present study and the 2 previous studies20,35 can potentially be explained by interindividual variability among horses in the inhibitory effects of the drugs or by differences in experimental design. For instance, the drugs were added to platelets ex vivo in both previous studies,20,35 excluding biological factors that might influence their effects. In addition, unstimulated platelets appeared activated in the IBMX study,35 with a mean of 25% of unstimulated platelets expressing P-selectin (compared with a median of < 2% when PBS solution was used in the present study), which may have influenced the results. Indeed, the addition of thrombin only mildly increased P-selectin expression (to a mean of 33%) in the previous study35 involving PDE inhibitors, versus the robust response achieved in the present study (medians > 70%), despite use of similar thrombin concentrations (0.1 U/L vs 0.15 U/L, respectively). In support of this assessment, 1 horse on day 0 of the clopidogrel treatment session had a low percentage of P-selectin–positive platelets (20%) after thrombin-convulxin stimulation, which we attributed to a preactivated state in those platelets. The lack of an inhibitory effect of theophylline on P-selectin expression in the present study could have also been explained by the fact that theophylline is a far weaker inhibitor of PDE activity than IBMX.36
A stronger nonspecific PDE blocker such as IBMX or a more specific PDE3 inhibitor may suppress both α-granule secretion and microvesiculation in EHV-1–activated equine platelets. More than 60 isoforms of PDEs have been described, of which several appear to be expressed in equine platelets (potentially PDEs 1 through 5) on the basis of inhibitory studies.36 Phosphodiesterases hydrolyze cAMP and cGMP, which halt platelet activation. Inhibition of PDEs leads to increases in intracellular concentrations of cAMP and cGMP, which then block platelet signaling pathways and cytoskeletal changes responsible for degranulation, aggregation, and intracellular calcium release.36,37,38 Drugs that specifically target PDE3 (eg, cilostazol, trequinsin, and quazinone) appear to be more effective at inhibiting platelet aggregation and P-selectin expression by equine and human platelets than nonspecific inhibitors such as theophylline.20,36,39 Additional experiments are needed to determine whether PDE3 inhibitors or the stronger nonspecific PDE inhibitor IBMX can block EHV-1–induced activation of equine platelets.
After the present study was completed, we discovered that virus particles (in the form of aggregates) were likely inadvertently included in the gating logic to discriminate platelet events and therefore influenced data analyses of P-selectin and PDMPs. Because we have also found that EHV-1 expresses phosphatidylserine (likely acquired from the rabbit kidney cells used for propagating the virus) but is negative for P-selectin, we believe that inclusion of some virus events in the analysis would have caused a relative decrease in the percentage of P-selectin–positive events and an increase in the percentage of PDMPs. However, any minor changes in the calculated values would not have altered our overall conclusions.
The results of the study reported here indicated that clopidogrel, acetylsalicylic acid, and the nonspecific PDE inhibitors theophylline and pentoxifylline failed to inhibit EHV-1–induced activation of equine platelets ex vivo. Only clopidogrel inhibited ADP-induced platelet aggregation, to variable degrees, in all treated horses. Although the drugs and doses used did not inhibit EHV-1–induced platelet activation responses that were measured ex vivo, this does not mean that these drugs would be ineffective at preventing or reducing clot formation in EHV-1–infected horses. For instance, clopidogrel may still inhibit ADP-stimulated platelet aggregation in vivo in infected horses, whether induced by EHV-1–triggered thrombin generation with subsequent ADP release from platelet activation or by other platelet stimuli such as endothelial injury exposing subendothelial collagen. Findings support the need for testing of additional anticoagulant drugs, in particular ones that specifically target thrombin generation (such as heparin), for their potential to inhibit platelet activation directly caused by EHV-1.
Acknowledgments
Supported by the Grayson-Jockey Club Research Foundation Incorporated.
Presented in abstract form at the 2015 American College of Veterinary Pathologists, American Society for Veterinary Clinical Pathology, and Society of Toxicologic Pathology Combined Annual Meeting, Minneapolis, October 2015.
The authors thank Alexa Fland, Scott Baxendell, Steven Elser, Joe McLain, Ian Barrie, Samantha Lovering, Brooke Wilson, Katherine Frazier, Lauren Witter, and Laura Espinosa for their assistance in treating, collecting blood samples from, and taking care of the horses. The authors also thank Amelia Frye, Gail Babcock, and Jessica Waffle for technical assistance.
ABBREVIATIONS
| AUC | Area under the aggregation curve |
| EHV-1 | Equine herpesvirus type 1 |
| IBMX | 3-Isobutyl-1-methylxanthine |
| PDE | Phosphodiesterase |
| PDMP | Platelet-derived microparticle |
| PRP | Platelet-rich plasma |
| PSGL-1 | Platelet surface glycoprotein ligand-1 |
| RK13 | Rabbit kidney 13 |
Footnotes
Theophylline, PCCA, Houston, Tex.
Pentoxifylline, PCCA, Houston, Tex.
Clopidogrel bisulfate, Plavix, Bristol-Myers Squibb, New York, NY.
Aspirin, PCCA, Houston, Tex.
Dextrose USP anhydrous, PCCA, Houston, Tex.
Cherry-flavored Kool-Aid, Kraft Foods Inc, Northfield, Ill.
Monoject veterinary hypodermic needle, Covidien Ltd, Mansfield, Mass.
Monoject syringe, Covidien Ltd, Mansfield, Mass.
Vacutainer, Becton-Dickinson, Franklin Lakes, NJ.
Vacuette, Greiner Bio-One International, Frickenhausen, Germany.
Advia 2120, Siemens Healthcare Diagnostics, Tarrytown, NY.
Roche/Hitachi Modular P chemistry analyzer, Roche Diagnostics, Indianapolis, Ind.
Gly-pro-arg-pro-NH2 acetate, Sigma-Aldrich Corp, St Louis, Mo.
Bovine thrombin, Sigma-Aldrich Corp, St Louis, Mo.
DSM Nutritional Products, Herleen, The Netherlands.
CD41-phycoerythrin (clone P2), Beckman Coulter Inc, Brea, Calif.
TACS AV-FITC apoptosis detection kit, Tevigen, Gaithersburg, Md.
CD62P-allophycocyanin (clone Psel.KO.2.7), Novus Biologicals LLC, Littleton, Colo.
BD FACSCalibur, BD Biosciences, San Jose, Calif.
Platelet Aggregation Profiler (Model PAP-8E), Bio/Data Corp, Horshman, Pa.
ADP, Bio/Data Corp, Horshman, Pa.
Analyse-it for Microsoft Excel, version 3.90.7, Analyse-it Software Ltd, Leeds, West Yorkshire, England.
Statistix, version 9, Analytical Software, Tallahassee, Fla.
References
1. Lunn DP, Davis-Poynter N, Flaminio MJ, et al. Equine herpesvirus-1 consensus statement. J Vet Intern Med 2009; 23: 450–461.
2. Smith KC, Whitwell KE, Blunden AS, et al. Equine herpesvirus-1 abortion: atypical cases with lesions largely or wholly restricted to the placenta. Equine Vet J 2004; 36: 79–82.
3. Edington N, Bridges CG, Patel JR. Endothelial cell infection and thrombosis in paralysis caused by equid herpesvirus-1: equine stroke. Arch Virol 1986; 90: 111–124.
4. Smith KC, Borchers K. A study of the pathogenesis of equid herpesvirus-1 (EHV-1) abortion by DNA in-situ hybridization. J Comp Pathol 2001; 125: 304–310.
5. Yeo WM, Osterrieder N, Stokol T. Equine herpesvirus type 1 infection induces procoagulant activity in equine monocytes. Vet Res 2013; 44: 16.
6. Stokol T, Yeo WM, Burnett D, et al. Equid herpesvirus type 1 activates platelets. PLoS ONE 2015; 10: e0122640.
7. Bournazos S, Rennie J, Hart SP, et al. Monocyte functional responsiveness after PSGL-1-mediated platelet adhesion is dependent on platelet activation status. Arterioscler Thromb Vasc Biol 2008; 28: 1491–1498.
8. Passacquale G, Vamadevan P, Pereira L, et al. Monocyte-platelet interaction induces a pro-inflammatory phenotype in circulating monocytes. PLoS ONE 2011; 6: e25595.
9. Kuckleburg CJ, Yates CM, Kalia N, et al. Endothelial cell-borne platelet bridges selectively recruit monocytes in human and mouse models of vascular inflammation. Cardiovasc Res 2011; 91: 134–141.
10. Lam FW, Burns AR, Smith CW, et al. Platelets enhance neutrophil transendothelial migration via P-selectin glycoprotein ligand-1. Am J Physiol Heart Circ Physiol 2011; 300: H468–H475.
11. Sinauridze EI, Kireev DA, Popenko NY, et al. Platelet microparticle membranes have 50- to 100-fold higher specific procoagulant activity than activated platelets. Thromb Haemost 2007; 97: 425–434.
12. de Witt SM, Verdoold R, Cosemans JM, et al. Insights into platelet-based control of coagulation. Thromb Res 2014; 133(suppl 2): S139–S148.
13. Heemskerk JW, Mattheij NJ, Cosemans JM. Platelet-based coagulation: different populations, different functions. J Thromb Haemost 2013; 11: 2–16.
14. Oostingh GJ, Pozgajova M, Ludwig RJ, et al. Diminished thrombus formation and alleviation of myocardial infarction and reperfusion injury through antibody- or small-molecule-mediated inhibition of selectin-dependent platelet functions. Haematologica 2007; 92: 502–512.
15. Slotta JE, Braun OO, Menger MD, et al. Capture of platelets to 1373 the endothelium of the femoral vein is mediated by CD62P and CD162. Platelets 2009; 20: 505–512.
16. Brainard BM, Epstein KL, LoBato D, et al. Effects of clopidogrel and aspirin on platelet aggregation, thromboxane production, and serotonin secretion in horses. J Vet Intern Med 2011; 25: 116–122.
17. Brainard BM, Epstein KL, LoBato DN, et al. Treatment with aspirin or clopidogrel does not affect equine platelet expression of P selectin or platelet-neutrophil aggregates. Vet Immunol Immunopathol 2012; 149: 119–125.
18. Brooks MB, Divers TJ, Watts AE, et al. Effects of clopidogrel on the platelet activation response in horses. Am J Vet Res 2013; 74: 1212–1222.
19. Kornreich B, Enyeart M, Jesty SA, et al. The effects of pentoxifylline on equine platelet aggregation. J Vet Intern Med 2010; 24: 1196–1202.
20. Rickards KJ, Andrews MJ, Waterworth TH, et al. Differential effects of phosphodiesterase inhibitors on platelet activating factor (PAF)- and adenosine diphosphate (ADP)-induced equine platelet aggregation. J Vet Pharmacol Ther 2003; 26: 277–282.
21. Reczko E, Mayr A. [On the fine structure of a virus of the herpes group isolated from horses (short report)]. Arch Gesamte Virusforsch 1963; 13: 591–593.
22. Goehring LS, Hussey GS, Ashton LV, et al. Infection of central nervous system endothelial cells by cell-associated EHV-1. Vet Microbiol 2011; 148: 389–395.
23. Smith D, Hamblin A, Edington N. Equid herpesvirus 1 infection of endothelial cells requires activation of putative adhesion molecules: an in vitro model. Clin Exp Immunol 2002; 129: 281–287.
24. Dole VS, Bergmeier W, Patten IS, et al. PSGL-1 regulates platelet P-selectin-mediated endothelial activation and shedding of P-selectin from activated platelets. Thromb Haemost 2007; 98: 806–812.
25. da Costa Martins P, Garcia-Vallejo JJ, van Thienen JV, et al. P-selectin glycoprotein ligand-1 is expressed on endothelial cells and mediates monocyte adhesion to activated endothelium. Arterioscler Thromb Vasc Biol 2007; 27: 1023–1029.
26. Desta Z, Zhao X, Shin JG, et al. Clinical significance of the cytochrome P450 2C19 genetic polymorphism. Clin Pharmacokinet 2002; 41: 913–958.
27. Mega JL, Close SL, Wiviott SD, et al. Cytochrome p-450 polymorphisms and response to clopidogrel. N Engl J Med 2009; 360: 354–362.
28. Bailey SR, Andrews MJ, Elliott J, et al. Differential activation of platelets from normal and allergic ponies by PAF and ADP. Inflamm Res 2000; 49: 224–230.
29. Heath MF, Evans RJ, Poole AW, et al. The effects of aspirin and paracetamol on the aggregation of equine blood platelets. J Vet Pharmacol Ther 1994; 17: 374–378.
30. McKiernan BC, Koritz GD, Scott JS, et al. Plasma theophylline concentration and lung function in ponies with recurrent obstructive lung disease. Equine Vet J 1990; 22: 194–197.
31. Pearson EG, Riebold TW. Comparison of bronchodilators in alleviating clinical signs in horses with chronic obstructive pulmonary disease. J Am Vet Med Assoc 1989; 194: 1287–1291.
32. Barton MH, Ferguson D, Davis PJ, et al. The effects of pentoxifylline infusion on plasma 6-keto-prostaglandin F1α and ex vivo endotoxin-induced tumour necrosis factor activity in horses. J Vet Pharmacol Ther 1997; 20: 487–492.
33. Léguillette R, Désévaux C, Lavoie JP. Effects of pentoxifylline on pulmonary function and results of cytologic examination of bronchoalveolar lavage fluid in horses with recurrent airway obstruction. Am J Vet Res 2002; 63: 459–463.
34. Mitchell JD, Elliott J. Towards a new treatment for equine acute laminitis: the importance of signalling pathways. Vet J 2012; 192: 258–259.
35. Dunkel B, Rickards KJ, Werling D, et al. Evaluation of the effect of phosphodiesterase on equine platelet activation and the effect of antigen challenge on platelet phosphodiesterase activity in horses with recurrent airway obstruction. Am J Vet Res 2010; 71: 534–540.
36. Dunkel B, Rickards KJ, Page CP, et al. Phosphodiesterase isoenzymes in equine platelets and their influence on platelet adhesion. Am J Vet Res 2007; 68: 1354–1360.
37. Hung DT, Wong YH, Vu TK, et al. The cloned platelet thrombin receptor couples to at least two distinct effectors to stimulate phosphoinositide hydrolysis and inhibit adenylyl cyclase. J Biol Chem 1992; 267: 20831–20834.
38. Zhang W, Colman RW. Thrombin regulates intracellular cyclic AMP concentration in human platelets through phosphorylation/activation of phosphodiesterase 3A. Blood 2007; 110: 1475–1482.
39. Ito H, Miyakoda G, Mori T. Cilostazol inhibits platelet-leukocyte interaction by suppression of platelet activation. Platelets 2004; 15: 293–301.
