Endotoxin (ie, LPS) plays a pivotal role in the pathogenesis of a number of diseases in a variety of species and appears to be a potent inducer of tissue hypoxia. For example, endotoxemia is associated with decreases in intestinal villus1 and cecal blood flow,2 myocardial ischemia,3 cerebral hypoperfusion,4 and impaired renal perfusion and oxygen uptake,5 whereas skeletal muscle perfusion remains unaffected.6 In addition, laminitis is a recognized complication of clinical endotoxemia of horses7 that is thought to be a consequence of altered digital perfusion. The effect of LPS on perfusion may be direct or indirect through the production of vasoactive mediators by LPS-activated cells.
Marked species differences are found in sensitivity to LPS,8 and humans and horses appear to be exquisitely sensitive. Low-dose LPS infusion in horses9 is associated with maximum plasma LPS concentrations similar to those reported in carbohydrate overload–induced laminitis during the onset of Obel grade 3 lameness10 and in horses with acute abdominal disease11; clinical signs are consistent with endotoxemia and digital hypoperfusion. Digital hypoperfusion is temporally associated with significant increases in 2 platelet-derived vasoactive mediators, TxA2 and 5-HT.9 However, the relative importance of these 2 vasoconstrictor mediators in mediating digital hypoperfusion remains unclear. In addition, PAF appears to play a role in the release of these 2 mediators in vitro.12,13 The aim of the study reported here was to further elucidate the role of TxA2, 5-HT, and PAF in LPS-induced digital hypoperfusion in horses by use of selective inhibitors. Identification of the vasoactive mediators responsible for the effects of LPS on digital perfusion might allow the potentially serious effects of digital ischemia to be avoided with the use of selective antagonists in the clinical situation.
Lipopolysaccharide-activated equine leukocytes and platelets release TxA2 and 5-HT in vitro.13-15 Aspirin prevents equine platelet, leukocyte, and endothelial cell TxA2 synthesis,16,a and unlike leukocytes, platelet TxA2 production does not recover for 7 to 10 days following a single dose of aspirin.16 Thus, aspirin could be used to clarify further the cellular origin as well as the role of TxA2 in LPS-induced digital hypoperfusion in horses in vivo.
Equine digital vessels possess 5-HT1B/D and 5-HT2 receptors that cause vasoconstriction,17 and synergism between 5-HT and TxA2 in mediating digital vasoconstriction in horses has been determined in vitro.b Results of studies18,19 in other species suggest that this synergism may be the result of TxA2 receptor agonists uncovering the actions of 5-HT1 receptors.The compound 3-[3-(dimethylamino) propyl]-4-hydroxy-N-[4-(4-pyridinyl) phenyl] benzamide dihydrochloride (ie, GR55562) is a potent antagonist for 5-HT1B/D receptors20; therefore, GR55562 could be used to investigate the role of 5-HT in LPS-induced digital hypoperfusion in horses. Comparing the effects of simultaneous administration of aspirin and GR55562 with that of aspirin alone would help to establish whether there is synergism between 5-HT acting on 5-HT1 receptors and TxA2.
Platelet-activating factor appears to be involved in the pathogenesis of endotoxemia in horses,21 and its synthesis may be triggered either directly by LPS or indirectly by cytokines released from LPS-activated cells.22 Lipopolysaccharide-induced equine platelet aggregation12 and 5-HT release13 in vitro are dependent on leukocyte-derived PAF. Thus, if PAF mediates LPS-induced TxA2 and 5-HT release by leukocytes and platelets in vivo, a PAF receptor antagonist should decrease the response to LPS. The compound WEB2086 inhibits PAF-induced equine platelet activation in vitro and ex vivo23,24 and could therefore be used to investigate the role of PAF and, potentially, 5-HT and TxA2 synergism in LPS-induced digital hypoperfusion in horses.
Materials and Methods
Animals—Experiments were performed under a home office license approved by the Royal Veterinary College Ethics and Welfare Committee. Six healthy adult Thoroughbreds (3 males, 3 females; mean ± SD age, 8.2 ± 2.5 years; weight, 514 to 600 kg) were used.
Experimental design—Horses were stabled overnight and acclimatized to a partially temperature controlled environment (mean ± SD ambient temperature, 20.5 ± 2.2°C) for 1 hour before the infusions commenced. Bilateral jugular vein catheters were placed.
In the first experiment, to establish an effective GR55562 dose, tryptaminec (1.6 μg/kg/min) was administered for 30 minutes via the left jugular catheter 5 minutes after IV administration of either saline (0.9% NaCl) solution (control treatment) or GR55562d (0.3 mg/kg). Tryptamine is an agonist at 5-HT1 and 5-HT2 receptors in equine digital arteries and veins in vitro25 and causes digital hypoperfusion in vivo.26
In the second part of the study, Escherichia coli type O55:B5 LPSc (30 ng/kg; sterile filtered into 500 mL of saline solution) was administered for 30 minutes via the left jugular catheter. Horses were given one of the following treatment combinations IV: saline solution at 4 days, 2 hours, and 20 minutes before the onset of the LPS infusion and 30 minutes after the onset of the LPS infusion; saline solution at 4 days and 20 minutes before the onset of the LPS infusion and 30 minutes after the onset of LPS infusion and aspirinc (4 mg/kg) at 2 hours before the onset of the LPS infusion; aspirinc (4 mg/kg) at 4 days before the onset of the LPS infusion and saline solution at 2 hours and at 20 minutes before the onset of the LPS infusion and 30 minutes after the onset of the LPS infusion; WEB2086e (3 mg/kg) at 20 minutes before the onset of the LPS infusion and saline solution at 4 days and 2 hours before the onset of the LPS infusion and 30 minutes after the onset of the LPS infusion; saline solution at 4 days, 2 hours, and 20 minutes before the onset of the LPS infusion and GR55562d (0.3 mg/kg) at 30 minutes after the onset of the LPS infusion; and aspirinc (4 mg/kg) at 2 hours before the onset of the LPS infusion, saline solution at 4 days and 20 minutes before the onset of the LPS infusion, and GR55562d (0.3 mg/kg) at 30 minutes after the onset of the LPS infusion.
Doses of aspirin and WEB2086 used are effective in vivo in horses.23,27 The inhibitory effects of WEB2086 were confirmed by collecting blood (5 mL) into heparin (20 U/mL) from 1 Thoroughbred immediately before and 20 and 90 minutes after IV administration of WEB2086e (3 mg/kg). Platelet aggregation in response to PAF (final concentrations, 2nM and 20nM) was measured in platelet-rich plasma as described elsewhere.23
Each horse received every pretreatment combination in a randomized order with at least 28 days between experiments. All horses received flunixin megluminef (1.1 mg/kg, IV) at the end of the experiment to inhibit the inflammatory cascade activated by the LPS infusion as a means to prevent further clinical signs associated with LPS infusion.
Clinical monitoring—Arterial blood pressure and heart rate were monitored by use of a noninvasive oscillometric blood pressure monitorg applied to the middle coccygeal artery. The CBST and HWST were measured every 10 minutes by use of an infrared thermometerh focused on the right forefoot (midline), 1 cm above the skin-hoof junction and a third of the way from the coronary band to the toe, respectively, as previously described.28 Blood flow was measured every 15 minutes for 60 minutes following tryptamine infusion and for 120 minutes and then again at 300 minutes following LPS infusion in the right lateral forelimb digital artery and vein by use of Doppler ultrasonography,i,j as previously described.9 Blood samples (5 mL) were collected into EDTAc from horses infused with LPS every 10 minutes for 60 minutes and then every 15 minutes for 4 hours. Leukocyte and platelet counts were subsequently quantified by use of a Coulter counter.k
Plasma 5-HT and TxA2 concentration following LPS infusion—Blood samples (5 mL) were collected at the same time points into heparinized vacuum tubes containing clomipraminec (1 μmol/L) and phenelzinec (10 μmol/L), which prevent platelet uptake or metabolism of 5-HT, respectively, and kept on ice. Platelet-poor plasma was obtained from blood samples by centrifugation (300 × g for 10 minutes at 4°C and 10,000 × g for 10 minutes), and the concentration of 5-HT was measured by use of high-performance liquid chromatography with electrochemical detection, as previously described.29 Aspirinc (1 mmol/L) was added to heparinized blood samples (5 mL) collected at the same time points, and the separated plasma was stored at −80°C. Plasma TxA2 concentrations were subsequently determined indirectly by radioimmunoassay, as previously described.16
Effect of aspirin on ex vivo A23187-induced TxA2 production—Inhibitory effects of aspirin on platelet and leukocyte TxA2 production were confirmed by collecting blood (5 mL) into EDTAc immediately prior to LPS infusion from horses treated with saline solution or aspirin 4 days previously. Platelets and leukocytes were isolated, resuspended in PBS solutionl to give final concentrations of 1 × 108 platelets/mL and 1 × 107 leukocytes/mL, and incubated at 37°C in a shaking water bath for 15 and 60 minutes, respectively, with dimethyl sulfoxidea (control treatment) or calcium ionophore A23187c (10 μg/mL in dimethyl sulfoxide), as described elsewhere.15 Cell suspensions were centrifuged (5,000 × g for 5 minutes), and the supernatant was removed and stored at −80°C until plasma TxA2 concentrations were determined.
Statistical analysis—Values are expressed as mean ± SEM, and statistical analyses were performed by use of computer software.m Values from tryptamine infusion were compared among treatment groups by a 2-way ANOVA followed by the Bonferroni post hoc test. For heart rate, blood pressure, and hematologic values from the control (pretreatment with saline solution) LPS infusions, values at each time point were compared with time 0 values by use of a 1-way repeated-measures ANOVA followed by the Dunnett post hoc test. For all variables from LPS infusions, the 5 drug pretreated groups (aspirin 2 hours prior and 4 days prior to LPS, GR55562, WEB2086, and aspirin plus GR55562) were compared with each other and with the control group by use of a 1-way repeated-measures ANOVA followed by the Bonferroni post hoc test. Areas under the curves between 0 and 300 minutes of graphs for digital blood flow, CBST and HWST, and plasma mediator concentration for individual LPS-infused horses were calculated. The effect of pretreatment on AUC0–300 was expressed by calculating the percentage difference, compared with the control group, and by comparing the AUC0–300 between pretreatment groups by use of a paired Student t test. The A23187-induced TxA2 production was compared between saline- and aspirin-treated horses by use of a paired Student t test. Values of P ≤ 0.05 were considered significant.
Results
Tryptamine infusions—Tryptamine infusion, regardless of pretreatment combination, did not result in any significant alterations in heart rate or arterial blood pressure over the study period (data not shown). Following treatment with saline solution, tryptamine infusion caused digital arterial and venous blood flow to decrease by maxima of 39.0 ± 7.5% and 26.0 ± 2.5%, compared with time 0 (start of tryptamine infusion) values, after 45 and 30 minutes, respectively (Figure 1). A similar temporal pattern of decreases in CBST and HWST was detected. Treatment with GR55562 (0.3 mg/kg, IV) completely abolished the tryptamineinduced digital hypoperfusion and reductions in CBST and HWST.
The effect of IV administration of saline (0.9% NaCl) solution and GR55562 (0.3 mg/kg) on tryptamine (480 μg/kg, IV, 30 minutes)-induced alterations in digital arterial and venous blood flow (A) and HWST and CBST (B). Horizontal bar represents time of tryptamine infusion. aSignificant (P < 0.05) difference between treatment groups. Values represent mean ± SEM values from 3 horses. CB = Coronary band. HW = Hoof wall.
Citation: American Journal of Veterinary Research 69, 2; 10.2460/ajvr.69.2.199
LPS infusions—Infusion with LPS following treatment with saline solution caused a significant increase in heart rate, compared with time 0 (start of LPS infusion) values, between 45 and 120 minutes, reaching a maximum of 58.2 ± 9.2 beats/min at 75 minutes, but did not cause significant changes in mean, systolic, or diastolic arterial blood pressure (data not shown). Horses had mild signs of abdominal discomfort and tachypnea from 30 to 60 minutes that abated by 90 to 150 minutes, but returned between 210 and 240 minutes and lasted for approximately 30 minutes. Leukocyte and platelet counts decreased significantly, compared with time 0 values, between 30 and 120 minutes; the total leukocyte count decreased from 6.4 ± 0.3 × 109 cells/L at time 0 to 2.3 ± 0.3 × 109 cells/L at 120 minutes, and the platelet count decreased from 73.3 ± 11.3 × 109 platelets/L to 54.8 ± 14.9 × 109 platelets/L. Drug pretreatment did not significantly alter the effect of LPS on any of these clinical or hematologic variables (data not shown).
Following treatment with saline solution, LPS infusion caused a significant reduction in digital arterial and venous blood flow, which decreased by maxima of 91 ± 3% and 94 ± 2%, respectively, compared with time 0 (start of LPS infusion) values by 75 minutes; blood flow returned to 95 ± 6% and 88 ± 10% of the original values by the end of the study period (Figure 2). Aspirin given 2 hours before LPS significantly attenuated the decrease in blood flow, such that arterial and venous blood flow decreased by maxima of 69 ± 10% and 63 ± 11%, compared with time 0, respectively. Aspirin given 4 days before LPS had no significant effect on the LPS-induced maximum decrease in blood flow. Aspirin given 2 hours before LPS significantly attenuated the overall decrease (AUC0–300) in digital blood flow, whereas aspirin given 4 days before LPS had no significant effect (Table 1). WEB2086 and GR55562 alone did not affect the LPS-induced maximum (data not shown) or overall alterations (AUC0–300) in blood flow at the doses used. In contrast, aspirin plus GR55562 significantly attenuated both of the LPS-induced maximum blood flow decreases, such that arterial and venous blood flow decreased by maxima of 79 ± 5% and 75 ± 5%, compared with time 0 (Figure 3) and the overall decrease (AUC0–300) in digital blood flow.
Mean ± SEM percentage effect of drug pretreatment on LPS-induced overall change in digital blood flow, HWST and CBST, and plasma mediator concentrations in 6 horses.
Effect of IV administration of aspirin (4 mg/kg) alone given either 2 hours or 4 days before LPS infusion and aspirin given with GR55562 (0.3 mg/kg) 2 hours before LPS infusion on LPS (30 ng/kg, IV, for 30 minutes)-induced alterations in digital arterial blood flow. Data from horses pretreated with saline solution (control treatment) are shown. Horizontal bar represents time of LPS infusion. Mean ± SEM values from 6 horses.
Citation: American Journal of Veterinary Research 69, 2; 10.2460/ajvr.69.2.199
A similar temporal pattern of reductions in HWST and CBST was detected following administration of LPS to horses pretreated with saline solution (Figure 3). The maximum reduction in HWST and CBST was significantly attenuated to 63 ± 15% and 42 ± 8% and to 57 ± 11% and 41 ± 10% of control LPS infusion–induced changes by aspirin given 2 hours and 4 days before LPS, respectively. In addition, the overall (AUC0–300) LPS-induced decreases in HWST and CBST were significantly attenuated (Table 1). No significant difference was found between the attenuation of the LPS-induced changes in HWST and CBST caused by aspirin given 2 hours and aspirin given 4 days before LPS. WEB2086 and GR55562 alone had no significant effect on LPS-induced maximum (data not shown) or overall alterations (AUC0–300) in HWST and CBST. Aspirin plus GR55562 significantly attenuated the maximum decreases in HWST and CBST to 19 ± 8% and 33 ± 10%, respectively, of control LPS infusion–induced changes and the overall decrease in HWST and CBST (AUC0–300). The attenuation caused by pretreatment with aspirin plus GR55562 was significantly greater than that caused by aspirin given alone 2 hours prior to LPS.
Effect of IV administration of aspirin (4 mg/kg) alone given either 2 hours or 4 days before LPS infusion and aspirin given with GR55562 (0.3 mg/kg) 2 hours before LPS infusion on LPS (30 ng/kg, IV, for 30 minutes)-induced alterations in HWST. Data from horses pretreated with saline solution (control treatment) are shown. Horizontal bar represents time of LPS infusion. Mean ± SEM values from 6 horses.
Citation: American Journal of Veterinary Research 69, 2; 10.2460/ajvr.69.2.199
Lipopolysaccharide infusion in horses pretreated with saline solution increased plasma TxA2 and 5-HT concentrations to maxima of 10 and 4 times the baseline value at 55 minutes and 60 minutes, respectively. A secondary increase in plasma TxA2 concentration was observed after 210 minutes, peaking at 240 minutes (Figure 4). Plasma 5-HT concentrations returned to baseline values by 180 minutes with no evidence of a secondary increase (data not shown). Aspirin given 2 hours before LPS completely abolished the increase in plasma TxA2. Aspirin given 4 days before LPS caused a significant decrease in AUC0–300, compared with the control LPS infusion value (Table 1). Pretreatment with WEB2086 and GR55562 did not decrease the LPS-induced increase in plasma TxA2 concentration significantly. As with aspirin alone, aspirin plus GR55562 completely abolished the effect of LPS on TxA2 production. The LPS-induced increase in plasma 5-HT concentration remained unaffected by any prior treatments.
Effect of IV administration of aspirin (4 mg/kg) alone given either 2 hours or 4 days before LPS infusion and aspirin given with GR55562 (0.3 mg/kg) 2 hours before LPS infusion on LPS (30 ng/kg, IV, for 30 minutes)-induced alterations in plasma TxA2 concentration. Data from horses pretreated with saline solution (control treatment) are shown. Horizontal bar represents time of LPS infusion. Mean ± SEM values from 6 horses.
Citation: American Journal of Veterinary Research 69, 2; 10.2460/ajvr.69.2.199
PAF receptor antagonist activity of WEB2086—Pretreatment with WEB2086, at the concentration used, completely inhibited ex vivo PAF (2 or 20nM)-induced aggregation for at least 90 minutes (37.8 ± 1.6 mV vs 0 ± 0 mV for 20nM PAF at time 0 and time 90 minutes, respectively, n = 1).
Effect of aspirin on ex vivo A23187-induced TxA2 production—Use of A23187 caused a 12.0 ± 1.7 fold increase and 12.8 ± 2.1 fold increase in TxA2 production by platelets and leukocytes from horses treated with saline solution, compared with the control solution (0.1% dimethyl sulfoxide), respectively. Aspirin administration 4 days previously significantly decreased ex vivo A23187-induced platelet TxA2 production (4.1 ± 0.9 fold increase), whereas leukocyte TxA2 release was unaltered (15.3 ± 5.8 fold increase).
Discussion
The aim of the present study was to elucidate further the roles of the vasoactive mediators 5-HT, TxA2, and PAF in LPS-induced digital hypoperfusion in horses by use of selective inhibitors of their formation or action. All 3 mediators are implicated in endotoxininduced tissue hypoperfusion in a variety of species including horses,9,30 and synergism between TxA2 and 5-HT in causing digital vasoconstriction in horses is suggested.b However, it must be acknowledged that experimental endotoxin infusion and clinical endotoxemia are associated with activation of the systemic inflammatory cascade and the generation of various other vasoactive mediators that may also play a role, as well as hemodynamic changes that were not evaluated in the study reported here.
In vitro, tryptamine is a potent nonselective 5-HT receptor agonist that inhibits equine platelet and endothelial cell 5-HT uptake and releases 5-HT from equine platelets.24,25,31 When infused into horses at the dosage used in the present study, tryptamine causes an increase in the plasma concentration of 5-HT, which is temporally related to the digital vasoconstriction that ensues.26 Thus, infusion of tryptamine can be used as a method to evaluate the in vivo effectiveness of potential 5-HT receptor antagonists. GR55562 is a selective 5-HT1B/D receptor antagonist in other species,20 and equine digital vessels possess 5-HT1B/D and 5-HT2 receptors mediating vasoconstriction.17 In the present study, the fact that GR55562 completely abolished tryptamine-induced digital hypoperfusion suggested that 5-HT1 receptors mediate the vasoconstriction, thereby negating the requirement to investigate additionally 5-HT2 receptor antagonism. In addition, these findings also suggested that GR55562 at this dose is suitable for use in further evaluating the role of 5-HT acting on 5-HT1B/D receptors in LPS-induced hypoperfusion.
Lipopolysaccharide-activated equine platelets release substantial amounts of TxA2 and 5-HT in vitro,13,14 and LPS-induced platelet activation is through an indirect mechanism involving calcium, leukocytes, adenine nucleotides, and PAF.12 Lipopolysaccharide-induced platelet 5-HT release in vitro is inhibited by the PAF receptor antagonist WEB2086 at a concentration of 100nM.24 Use of WEB2086, when given IV at 3 mg/kg, achieves a concentration in equine plasma of > 100nM for 3 hours23 and therefore should inhibit PAF-induced platelet 5-HT release in vivo following LPS infusion. In addition, Use of WEB2086 (3 mg/kg, IV) inhibits PAFinduced platelet aggregation ex vivo and inflammatory responses to PAF in vivo.23,32 This inhibition of ex vivo platelet aggregation was also detected in the present study, confirming that WEB2086 at this dose was suitable for investigation of the role of PAF in LPS-induced vasoactive mediator production in vivo and in digital hypoperfusion in horses.
The 2 techniques used in the present study to quantify digital perfusion were chosen because they are noninvasive techniques that have been used in other studies. Doppler ultrasonography has been used to measure the effect of acepromazine on metatarsal artery blood flow in standing horses,33 forelimb blood flow in unsedated horses,34 and arterial digital blood flow during weight bearing and non–weight-bearing conditions.35 In anesthetized horses, Doppler ultrasonographic measurement of femoral blood flow allows comparison of the hemodynamic effects of inhaled anesthetic agents36 and determination of the effect of dobutamine infusion of hind limb perfusion.37 Peripheral blood flow measurements calculated by use of Doppler ultrasonography correlate well with other invasive techniques including electromagnetic flowmetry,38 and Doppler ultrasonographic measurements of volumetric flow provide accurate estimates of in vitro blood flow.39 The repeatability of Doppler ultrasonographic measurement of digital blood flow in horses has also been evaluated.40 The within and between horse variability was found to be acceptable (coefficients of variation < 11%), and the repeatability was excellent (intraclass correlation coefficients > 0.71) for all variables measured. In addition, power calculations revealed that differences of 0.005 mL/min and 0.01 mL/min in digital arterial and venous blood flow, respectively, would be detected with measurements from 6 horses.40 Thus, the technique is sufficiently repeatable and sensitive to be able to detect changes in blood flow during different physiologic or pathologic states or following pharmacologic intervention.
Skin-surface temperature is the thermal equilibrium between the external temperature and the exothermic metabolic and mechanical activity and vascular supply of the underlying viable tissues. If measured under conditions of constant moderate ambient temperature and minimal mechanical activity, it is a noninvasive indirect method to assess underlying local perfusion. The technique has been applied to the equine digit in several previous studies9,26,28,41,42 as an indirect indicator of digital perfusion. Results of these studies indicate that the technique is sensitive enough to detect alterations in digital perfusion following induction of pathologic states9,28 or pharmacologic intervention.41,42 In addition, the technique is sufficiently repeatable that the values obtained for HWST and CBST do not change over time following saline solution infusion.9
Thus, Doppler ultrasonography was chosen to quantify changes in total blood flow to the equine foot, and HWST and CBST were measured to reflect the underlying microvascular blood flow. Simultaneous use of the 2 techniques should allow effects on total and microvascular blood flow to be differentiated. However, it must be acknowledged that the validity of this assumption has not been investigated and that neither technique distinguishes between blood flow through the arteriovenous anastomoses, sublaminar circulation, or laminar microcirculation.
Following control (pretreatment with saline solution) LPS infusion, digital blood flow decreased, reaching minimum values at 75 minutes and almost returning to pretreatment values by 300 minutes. These decreases in blood flow were similar to those reported in other studies43,44 and were corroborated by a similar temporal pattern of measurable decreases in surface temperatures, suggesting that total and microvascular blood flow were decreased. In the absence of significant changes in systemic arterial blood pressure, these changes in blood flow may reflect selective digital vascular resistance changes.9 However, in view of the LPSinduced tachycardia, which is potentially mediated by increased sympathetic tone, the possibility of a generalized systemic response cannot be excluded. Simultaneous measurement of blood flow in other peripheral vessels would be required to determine whether the changes detected in the digital circulation are selective. In agreement with findings of another report,9 a temporal relationship between LPS-induced increases in plasma 5-HT and TxA2 concentrations and digital hypoperfusion was evident in the present study.
Pretreatment with GR55562 alone did not affect the LPS-induced alterations in digital perfusion or plasma mediator concentrations. The lack of effect of the 5-HT1B/D receptor antagonist on TxA2 production and 5-HT release is unsurprising. The lack of an effect on LPS-induced digital hypoperfusion suggests that 5-HT (acting at 5-HT1B/D receptors) alone is not responsible for the decrease in digital blood flow. The dose of GR55562 used in the present study was sufficient to abolish the digital hypoperfusion resulting from tryptamine infusion. However, infusion of tryptamine caused plasma 5-HT concentrations to approximately double, compared with the baseline value,26 whereas LPS infusion caused a 4-fold increase in plasma 5-HT concentration. It is possible, therefore, that the dose of GR55562 used in the present study was not sufficient to prevent 5-HT induced vasoconstriction. It is possible that 5-HT2 receptors could be involved in mediating the responses to 5-HT and tryptamine. However, although tryptamine is a nonselective agonist that stimulates 5-HT1 and 5-HT2 receptors, the 5-HT2 receptor antagonist ritanserin does not attenuate tryptamine-induced digital hypoperfusion,a suggesting that 5-HT1 receptors are more important in mediating the digital vasoconstriction in vivo.
Aspirin administration significantly attenuated the LPS-induced alterations in plasma TxA2 concentration, but did not affect plasma 5-HT concentration. Although TxA2 is a powerful auto-activator of platelets in other species, it does not appear to have this effect on equine platelets,14 such that aspirin would not be expected to affect LPS-induced platelet activation and, hence, 5-HT release. Indeed, no effect of aspirin is found on 5-HT release from platelets stimulated with LPS in vitro.45 Two hours after administration in the study reported here, aspirin caused complete inactivation of platelet and leukocyte cyclooxygenase, as evidenced by complete attenuation of the LPS-induced increases in plasma TxA2 concentration. This finding is in agreement with findings of other studies16,27,46 evaluating the effect of aspirin on TxA2 production in horses. The abolition of the LPS-induced increases in plasma TxA2 concentration in the present study was accompanied by significant, but not complete, inhibition of LPS-induced digital hypoperfusion and hypothermia. This suggests that leukocyte- and platelet-derived TxA2 are partially responsible for the LPS-induced reduction in total and microvascular blood flow.
Four days after aspirin administration, leukocyte cyclooxygenase activity had recovered, whereas platelet cyclooxygenase remained partially inhibited, as evidenced by the significant reduction in ex vivo platelet, but not leukocyte, responses to the calcium ionophore A23187 and the partial return of the first LPS-induced peak in plasma TxA2 concentration in vivo. Thus, platelets appear to be responsible for the second peak in plasma TxA2 concentration. The LPS-induced total digital hypoperfusion was not changed, but there was a significant attenuation of the LPS-induced hypothermia. If HWST and CBST reflect microvascular blood flow, then this suggests that the microvessels, composed of the laminar microcirculation and the sublaminar circulation separated by numerous arteriovenous anastomoses, are more affected by platelet-derived TxA2 than the larger vessels. However, neither noninvasive technique used in the present study to assess perfusion can distinguish between blood flow through the arteriovenous shunts, sublaminar vessels, or laminar vessels. Thus, the attenuation of the digital hypothermia may be a consequence of alteration of blood flow within any of these vessels. If the microvascular blood flow was decreased, total blood flow would also be expected to be decreased. One explanation for this discordance is that the ratio between arteriovenous anastomoses shunt and laminar blood flow is altered, thus attenuating the decrease in HWST and CBST, without an influence on LPS-induced decrease in total blood flow. More direct measures of nutrient versus shunt blood flow, such as microsphere studies, would be necessary to address this issue.
Concomitant administration of aspirin and GR55562 virtually abolished the LPS-induced alterations in plasma TxA2 concentration and unsurprisingly had no effect on plasma 5-HT concentration. The inhibitory effect on LPS-induced digital hypothermia was significantly greater than that detected with aspirin alone when given 2 hours prior to LPS. In addition, LPS-induced digital hypoperfusion was evident only when plasma TxA2 and 5-HT concentrations were increased (ie, 40 to 120 minutes) and not during the second plasma TxA2 concentration peak when the plasma 5-HT concentration was returning to baseline values. Thus, these data might suggest that 5-HT, acting on 5-HT1 receptors, synergizes with TxA2 to mediate LPS-induced alterations in digital perfusion. Synergism between these 2 mediators is found in human platelet aggregation,47 LPS-induced rat cerebral artery vasoconstriction,48 vascular smooth muscle cell proliferation,49 and human19 and equineb digital vasoconstriction in vitro. Results of studies indicate that this synergism is the result of TxA2 receptor agonists uncovering the actions of 5-HT1 receptors18,19 and may involve activation of multiple signaling pathways, including G-protein–coupled receptor activation of phospholipase C and calcium sensitization.47 In addition, as evident with aspirin pretreatment alone, a greater effect on surface temperatures was found in the present study, compared with total blood flow, again suggesting that differential sensitivity of the large and microvascular blood vessels exists, with microvascular blood flow being most affected by platelet-derived vasoactive mediators.
Lipopolysaccharide-induced alterations in plasma mediator concentrations were unaffected by WEB2086, suggesting that LPS-induced 5-HT and TxA2 release in vivo is not mediated by PAF, in contrast to other in vitro findings.12 Alternative explanations include direct platelet activation by LPS, a different leukocyte-derived mediator released in response to LPS activation, or adhesion molecule interactions and subsequent intracellular signaling. Some in vitro evidence exists and suggests that equine platelets are directly activated by LPS,50 as are human platelets,51 which express Toll-like receptors.52 Lipopolysaccharide-stimulated monocyte tissue factor is a potent inducer of platelet activation,53 although the mechanism remains unclear. At sites of vascular injury, platelets adhere to activated endothelial cells via endothelial P- and E-selectin and the subendothelium and become activated, binding leukocytes via selectins.54 Upon binding, intracellular signaling is initiated, leading to modulation of several cellular biological properties55 including, potentially, thromboxane synthase activity. Further studies are required to determine the mechanism of LPS-induced 5-HT and TxA2 release in horses. The lack of effect of WEB2086 on the increase in plasma TxA2 and 5-HT concentrations caused by LPS accounts for the lack of any reduction in digital hypoperfusion. This would suggest that PAF does not play a role in LPS-induced digital hypoperfusion.
Other vasoconstrictor mediators that may be induced by LPS and be responsible for the changes seen in the large vessel blood flow (directly or indirectly) include tumor necrosis factor-A; interleukin-1; prostaglandin F2α; endothelin-1; and the leukotrienes C4, D4, and E4.56 Aspirin would be expected to inhibit prostaglandins, and it has been determined that equine plasma endothelin-1 concentrations do not increase following low-dose LPS infusion in horses.9 However, systemic endothelin-1 concentrations may not accurately reflect local production and activity of endothelin-1 in the microcirculation because > 80% of endothelin-1 released from endothelial cells is secreted abluminally.57 Further evaluation of the role of the leukotrienes and of endothelial dysfunction in LPS-induced digital hypoperfusion in horses is required. In addition, because of the widespread actions of endotoxin, the possibility that the alterations in blood flow are a consequence of vasoactive mediator generation, neurogenic factors released from sympathetic nerves such as norepinephrine, or hemodynamic pressure–flow factors affecting the systemic circulation must also be considered.
In conclusion, results of the present study on selective inhibitors of mediator formation or action suggest that platelet-derived TxA2 plays a role in LPS-induced digital hypoperfusion and hypothermia in horses and that the sensitivity of the large and microvascular blood vessels to platelet-derived mediators may vary. Inhibition of 5-HT1 receptors alone had no significant effect on the response to LPS but increased the inhibition caused by aspirin, indicating that 5-HT may additionally be involved in mediating vasoconstriction. Platelet-activating factor does not appear to be required for the release of these vasoactive mediators, and further studies are warranted to determine the mechanisms by which LPS affects total digital blood flow.
ABBREVIATIONS
| LPS | Lipopolysaccharide |
| TxA2 | Thromboxane A2 |
| 5-HT | 5-hydroxytryptamine |
| PAF | Platelet-activating factor |
| CBST | Coronary band surface temperature |
| HWST | Hoof wall surface temperature |
| AUC0–300 | Area under the curve between 0 and 300 minutes |
Menzies-Gow NJ. Endotoxin: role in equine digital hypoperfusion. PhD thesis, Department of Science, Royal Veterinary College, University of London, 2004.
Bailey SR. A study of the vascular effects and factors regulating the concentration of 5-hydroxytryptamine in the equine digital circulation. PhD thesis, Department of Science, Royal Veterinary College, University of London, 1998.
Sigma, Poole, Dorset, England.
Tocris Cookson Ltd, Avonmouth, Bristol, England.
Boehringer Ingleheim, Rhein, Germany.
Schering-Plough Animal Health, Uxbridge, Middlesex, England.
Datascope Accutorr 3/4 series Datasette, Datascope Corp, Paramus, NJ.
Model IT-304, Horiba Inc, Kyoto, Japan.
Vingmed CFM 800A ultrasound system 5 ultrasound machine, GE Ultrasound, Bedford, England.
Echopac image analysis and archiving program, Vingmed Sound, Horten, Norway.
Model ZM, Coulter Counter Electronic Inc, Fullerton, Calif.
Gibco, Invitrogen Corp, Paisley, Scotland.
GraphPad Prism, version 3.00 for Windows, GraphPad Software, San Diego, Calif.
References
- 1.↑
Schmidt W, Tinelli M, Secchi A, et al. Influence of amrinone on intestinal villus blood flow during endotoxemia. J Crit Care 2000;15:97–102.
- 2.↑
Eades SC, Moore JN. Blockade of endotoxin-induced cecal hypoperfusion and ileus with an alpha 2 antagonist in horses. Am J Vet Res 1993;54:586–590.
- 3.↑
Avontuur JA, Bruining HA, Ince C. Inhibition of nitric oxide synthesis causes myocardial ischemia in endotoxemic rats. Circ Res 1995;76:418–425.
- 4.↑
Ando M, Takashima S, Mito T. Endotoxin, cerebral blood flow, amino acids and brain damage in young rabbits. Brain Dev 1988;10:365–370.
- 5.↑
Gullichsen E. Renal perfusion and metabolism in experimental endotoxin shock. Acta Chir Scand 1991;560:7–31.
- 6.↑
van Lambalgen AA, van den Bos GC, Thijs LG. Blood flow and plasma extravasation in skeletal muscle during endotoxemia. A study in rats. Int J Microcirc Clin Exp 1989;8:217–232.
- 7.↑
Hunt JM, Edwards GB, Clarke KW. Incidence, diagnosis and treatment of postoperative complications in colic cases. Equine Vet J 1986;18:264–270.
- 8.↑
Olson NC, Hellyer PW, Dodam JR. Mediators and vascular effects in response to endotoxin. Br Vet J 1995;151:489–522.
- 9.↑
Menzies-Gow NJ, Bailey SR, Katz LM, et al. Endotoxin-induced digital vasoconstriction in horses: associated changes in plasma concentrations of vasoconstrictor mediators. Equine Vet J 2004;36:273–278.
- 10.↑
Sprouse RF, Garner HE, Green EM. Plasma endotoxin levels in horses subjected to carbohydrate induced laminitis. Equine Vet J 1987;19:25–28.
- 11.↑
Steverink PJ, Salden HJ, Sturk A, et al. Laboratory and clinical evaluation of a chromogenic endotoxin assay for horses with acute intestinal disorders. Vet Q 1994;16 (suppl 2):S117–S121.
- 12.↑
Jarvis GE, Evans RJ. Endotoxin-induced platelet aggregation in heparinised equine whole blood in vitro. Res Vet Sci 1994;57:317–324.
- 13.↑
Bailey SR, Andrews MJ, Elliott J, et al. Actions and interactions of ADP, 5-HT, histamine and PAF on equine platelets. Res Vet Sci 2000;68:175–180.
- 14.↑
Jarvis GE, Evans RJ. Platelet-activating factor and not thromboxane A2 is an important mediator of endotoxin-induced platelet aggregation in equine heparinised whole blood in vitro. Blood Coagul Fibrinolysis 1996;7:194–198.
- 15.
Bottoms GD, Johnson M, Ward D, et al. Release of eicosanoids from white blood cells, platelets, smooth muscle cells, and endothelial cells in response to endotoxin and A23187. Circ Shock 1986;20:25–34.
- 16.↑
Lees P, Ewins CP, Taylor JB, et al. Serum thromboxane in the horse and its inhibition by aspirin, phenylbutazone and flunixin. Br Vet J 1987;143:462–476.
- 17.↑
Bailey SR, Elliott J. Evidence for different 5-HT1B/1D receptors mediating vasoconstriction of equine digital arteries and veins. Eur J Pharmacol 1998;355:175–187.
- 18.
Yildiz O, Tuncer M. Amplification of responses to sumatriptan by various agonists in rabbit isolated iliac artery. J Cardiovasc Pharmacol 1995;25:508–510.
- 19.↑
Young MS, Iwanov V, Moulds RF. Interaction between plateletreleased serotonin and thromboxane A2 on human digital arteries. Clin Exp Pharmacol Physiol 1986;13:143–152.
- 20.↑
Moorecroft I, Heeley RP, Prentice HM, et al. 5-hydroxytryptamine receptors mediating contraction in human small muscular pulmonary arteries: importance of the 5-HT1B receptor. Br J Pharmacol 1999;128:730–734.
- 21.↑
Carrick JB, Morris DD, Moore JN. Administration of a receptor antagonist for platelet-activating factor during equine endotoxaemia. Equine Vet J 1993;25:152–157.
- 22.↑
Valone FH, Epstein LB. Biphasic platelet-activating factor synthesis by human monocytes stimulated with IL-1β, tumour necrosis factor, or IFN-G. J Immunol 1988;141:3945–3950.
- 23.↑
Foster AP, Lees P, Andrews MJ, et al. Effects of WEB 2086, an antagonist to the receptor for platelet-activating factor (PAF), on PAF-induced responses in the horse. Equine Vet J 1992;24:203–207.
- 24.↑
Bailey SR, Cunningham FM, Elliott J. Endotoxin and dietary amines may increase plasma 5-hydroxytryptamine in the horse. Equine Vet J 2000;32:497–504.
- 25.
Elliott J, Berhane Y, Bailey SR. Effects of monoamines formed in the cecum of horses on equine digital blood vessels and platelets. Am J Vet Res 2003;64:1124–1131.
- 26.↑
Bailey SR, Menzies-Gow NJ, Marr CM, et al. The effects of vasoactive amines found in the equine hindgut on digital blood flow in the normal horse. Equine Vet J 2004;36:267–272.
- 27.
Cambridge H, Lees P, Hooke RE, et al. Antithrombotic actions of aspirin in the horse. Equine Vet J 1991;23:123–127.
- 28.
Hood DM, Wagner IP, Brumbaugh GW. Evaluation of hoof wall surface temperature as an index of digital vascular perfusion during the prodromal and acute phases of carbohydrate-induced laminitis in horses. Am J Vet Res 2001;62:1167–1172.
- 29.
Bailey SR, Elliott J. Plasma 5-hydroxytryptamine constricts equine digital blood vessels in vitro: implications for pathogenesis of acute laminitis. Equine Vet J 1998;30:124–130.
- 30.
Ewer AK, Al-Salti W, Coney AM, et al. The role of platelet activating factor in a neonatal piglet model of necrotising enterocolitis. Gut 2004;53:207–213.
- 31.
Bailey SR, Wheeler-Jones C, Elliott J. Uptake of 5-hydroxytryptamine by equine digital vein endothelial cells: inhibition by amines found in the equine caecum. Equine Vet J 2003;35:164–169.
- 32.
Fairbairn SM, Marr KA, Lees P, et al. Effects of platelet activating factor on the distribution of radiolabelled leucocytes and platelets in normal horses and asymptomatic horses with chronic obstructive pulmonary disease. Res Vet Sci 1996;61:107–113.
- 33.↑
Walker M, Geiser D. Effects of acetylpromazine on the hemodynamics of the equine metatarsal artery, as determined by twodimensional real-time and pulsed Doppler ultrasonography. Am J Vet Res 1986;47:1075–1078.
- 34.↑
Cochard T, Toal RL, Saxton AM. Doppler ultrasonographic features of thoracic limb arteries in clinically normal horses. Am J Vet Res 2000;61:183–190.
- 35.↑
Hoffmann KL, Wood AK, Griffiths KA, et al. Doppler sonographic measurements of arterial blood flow and their repeatability in the equine foot during weight bearing and non-weight bearing. Res Vet Sci 2001;70:199–203.
- 36.↑
Raisis AL, Young LE, Blissitt KJ, et al. A comparison of the haemodynamic effects of isoflurane and halothane anaesthesia in horses. Equine Vet J 2000c;32:318–326.
- 37.↑
Raisis AL, Young LE, Blissitt KJ, et al. Effect of a 30-minute infusion of dobutamine hydrochloride on hind limb blood flow and hemodynamics in halothane-anesthetized horses. Am J Vet Res 2000;61:1282–1288.
- 38.↑
Gill RW. Measurement of blood flow by ultrasound: accuracy and sources of error. Ultrasound Med Biol 1985;11:625–641.
- 39.↑
Lewis P, Psaila JV, Davies WT, et al. Measurement of volume flow in the human common femoral artery using a duplex ultrasound system. Ultrasound Med Biol 1986;12:777–784.
- 40.↑
Menzies-Gow NJ, Marr CM. The repeatability of Doppler ultrasonographic measurement of equine digital blood flow. Vet Radiol Ultrasound 2007;48:281–285.
- 41.
Hood DM, Brumbaugh GW, Wagner IP. Effectiveness of a unique dihydropyridine (BAYTG 1000) for prevention of laminitis in horses. Am J Vet Res 2002;63:443–447.
- 42.
Hoff TK, Hood DM, Wagner IP. Effectiveness of glyceryl trinitrate for enhancing digital submural perfusion in horses. Am J Vet Res 2002;63:648–652.
- 43.
Ingle-Fehr JE, Baxter GM. Evaluation of digital and laminar blood flow in horses given a low dose of endotoxin. Am J Vet Res 1998;59:192–196.
- 44.
Hunt RJ, Allen D, Moore JN. Effect of endotoxin administration on equine digital hemodynamics and starling forces. Am J Vet Res 1990;51:1703–1707.
- 45.↑
Bailey SR, Elliott J. The role of prostanoids and nitric oxide in endotoxin-induced hyporesponsiveness of equine digital blood vessels. Equine Vet J 1999;31:212–218.
- 46.
Baxter GM, Moore JN. Effect of aspirin on ex vivo generation of thromboxane in healthy horses. Am J Vet Res 1987;48:13–16.
- 47.↑
Saeed S, Rasheed H, Gilani A. Synergism interaction between arachadonic acid by 5-hydroxytryptamine in human platelet aggregation is mediated through multiple signalling pathways. Acta Pharmacol Sin 2003;24:958–964.
- 48.↑
Hernanz R, Alonso MJ, Briones AM, et al. Mechanisms involved in the early increase of serotonin contraction evoked by endotoxin in rat middle cerebral arteries. Br J Pharmacol 2003;140:671–680.
- 49.↑
Pakala R, Willerson JT, Benedict CR. Effect of serotonin, thromboxane A2 and specific receptor anatagonists on vascular smooth muscle cell proliferation. Circulation 1997;96:2280–2286.
- 50.↑
Brooks A, Menzies-Gow NJ, Bailey SR, et al. Endotoxin-induced activation of equine platelets: evidence for direct activation of p38 MAPK pathways and vasoactive mediator production. Inflamm Res 2007;56:154–161.
- 51.↑
Nakamura M, Honda Z, Waga I, et al. Endotoxin transduces Ca2+ signalling via platelet-activating factor receptor. FEBS Lett 1992;314:125–129.
- 52.↑
Cognasse F, Hamzeh H, Chavarin P, et al. Evidence of Tolllike receptor molecules on human platelets. Immunol Cell Biol 2005;83:196–198.
- 53.↑
Nemerson Y, Giesen PL. Some thoughts about localisation and expression of tissue factor. Blood Coagul Fibrinolysis 1998;9:S45–S47.
- 54.↑
Diacovo TG, Roth SJ, Bucola JM. Neutrophil rolling, arrest, and transmigration across activated, surface adherent platelets via sequential action of P-selectin and the Bβ-integrin CB11b/CD18. Blood 1996;88:146–157.
- 55.↑
Celi A, Lorenzet R, Furie B, et al. Platelet-leucocyte-endothelial cell interaction on the blood vessel wall. Semin Hematol 1997;34:327–335.
- 56.↑
Olson NC, Salzer WL, McCall CE. Biochemical, physiological and clinical aspects of endotoxemia. Mol Aspects Med 1988;10:511–629.
- 57.↑
Wagner OF, Christ G, Wojta J, et al. Polar secretion of endothelin-1 by cultured endothelial cells. J Biol Chem 1992;267:16066–16068.
