Effects of pentoxifylline on canine platelet aggregation

Abstract Background Pentoxifylline can decrease platelet function in humans, but the anti‐platelet effects of pentoxifylline in dogs is unknown. The addition of a luciferin–luciferase reagent during platelet aggregometry can induce a dose‐dependent potentiation of platelet aggregation. Objective To determine if exposure to pentoxifylline, without the addition of a luciferin–luciferase reagent during aggregometry, causes canine platelet dysfunction. Our hypotheses were that pentoxifylline would inhibit platelet function, and that the addition of a luciferin–luciferase reagent would obscure detection of pentoxifylline‐induced platelet dysfunction as measured via aggregometry. Methods Seven healthy Walker hound dogs. Platelet‐rich plasma (PRP) and whole blood were treated for 30 minutes with pentoxifylline: 0 (control), 1 and 2 μg/mL. The platelet aggregation was determined using optical (maximum amplitude) and impedance (ohms) aggregometry using collagen as the agonists, with and without a luciferin–luciferase reagent. Four samples were analysed per concentration and the results were averaged. Results Based on optical aggregometry, there was no difference (p = 0.964) in the mean maximum amplitude at any pentoxifylline concentration, with and without the luciferin–luciferase reagent. During impedance aggregometry, the addition of a luciferin–luciferase reagent was associated with significantly (p < 0.001) greater platelet aggregation in response to a collagen agonist, regardless of the presence or absence of pentoxifylline. Conclusions Pentoxifylline does not exert an in vitro anti‐platelet effect on canine platelet aggregation when collagen is used as an agonist, but it is unknown if long‐term oral drug administration will inhibit platelet aggregation. The addition of a luciferin–luciferase reagent during platelet aggregometry can artificially enhance canine platelet aggregation.


INTRODUCTION
Pentoxifylline is a methyxanthine derivative that possesses both immunologic and hematologic properties. In dogs, it is most commonly used to treat dermatologic syndromes such as vasculitis and familial dermatomyositis, but in humans it has also been used to treat endotoxemia, cancer and various hypercoagulable states (Marsella et al., 2000;Rees et al., 2003). In addition to its anti-inflammatory properties, pentoxifylline is also capable of influencing hemostasis (Rees et al., 2003). In humans, as measured by optical aggregometry (using plateletrich plasma) and impedance aggregometry (using whole blood), platelet aggregation is significantly decreased during in vitro and in vivo pentoxifylline treatment (De La Cruz et al., 1993;Magnusson et al., 2008).
The cyclic nucleotides, cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), are essential inhibitory intraplatelet second messengers that can interfere with several aspects of platelet function, including degranulation, fibrinogen receptor activation and platelet membrane rearrangement (Gresele et al., 2011;Thomason et al., 2016). Increased levels of platelet cAMP and cGMP will inhibit platelet activation, and increased levels of these nucleotides can be attained either by the binding of endogenous platelet inhibitors (prostacyclin and nitric oxide) to transmembrane receptors or by preventing their breakdown. Phosphodiesterases (PDEs) are enzymes that inactivate cAMP and cGMP, and inhibitors of PDEs can therefore lead to increased levels of intra-platelet cAMP and cGMP, and subsequent platelet inhibition. PDEs are classified by their affinity and rate of degradation of cyclic nucleotides. Platelets express PDE2 and PDE3, which hydrolyse cAMP, and PDE5, which hydrolyses cGMP. Canine platelet function has previously been shown to be inhibited by a range of phosphodiesterase inhibitors (Boudreaux et al., 1986;Tsien et al., 1982). Pentoxifylline is known to be a non-selective inhibitor of PDEs (Ueno et al., 2011).
A previous study by Rees and others performed in dogs showed that a single dose of pentoxifylline did not detectably inhibit platelet aggregation using optical aggregometry (Rees et al., 2003). However, that study also incorporated a luciferin-luciferase reagent during aggregometry as an additional indicator of platelet function. Luciferinluciferase preparations can be added to samples during aggregometry to provide an assessment of platelet activation by evaluating dense granule release from platelets by measuring the luminescence of ATP secretion. Unlike other species, the addition of this reagent during platelet aggregometry in dogs can induce spontaneous, irreversible platelet aggregation (Mehta et al., 1983). In fact, in dogs with hereditary platelet function defects that exhibit weak platelet aggregation during aggregometry, the addition of the luciferin-luciferase reagent causes a dose-dependent potentiation of platelet aggregation (Callan et al., 1998). In the previous study by Rees and others (Rees et al., 2003), it is possible that the addition of the luciferin-luciferase reagent obscured pentoxifylline-induced inhibition of platelet function, a phenomenon that we have previously noted in our laboratory when assess-ing aspirin-associated platelet dysfunction, 1 leading to the erroneous appearance of normal platelet aggregation despite the presence of pentoxifylline.
The objective of our study was to determine if exposure to pentoxifylline, without the addition of a luciferin-luciferase reagent during aggregometry, causes dysfunction in canine platelets. Our hypotheses were that pentoxifylline would inhibit canine platelet function by reducing platelet aggregation, and that the addition of a luciferinluciferase reagent would obscure detection of pentoxifylline-induced platelet dysfunction as measured via aggregometry.

Study population
Blood from seven healthy Walker hound dogs (5 intact females and 2 castrated males) was used for the study.

Sample collection and preparation
Blood samples were collected via jugular venipuncture with a 20-gauge needle into a 4.5 mL vacutainer tube containing 3.2% sodium citrate anticoagulant. 2 Using a previously published technique (Haines et al., 2016), platelet-rich plasma (PRP) was created from whole blood via centrifugation. Briefly, whole blood was centrifuged at 1200 × g at room temperature for 3 minutes, the PRP supernatant was removed, and the remaining blood sample was centrifuged 1800 × g at room temperature for 8 minutes to create platelet-poor plasma (PPP).

Pentoxifylline incubation
Previously published protocols (Kornreich et al., 2010;Shipley et al., 2013) were modified to evaluate canine platelet function via aggregometry following the in vitro exposure of PRP to two different concentrations of pentoxifylline, 3 plus a saline control. Exposure concentrations were selected based on approximate plasma concentrations of pentoxifylline attained in dogs with standard oral dose rates (Marsella et al., 2000). Briefly, two concentrations of pentoxifylline working solution, 1 and 2 µg/mL, were created from a stock solution containing 10 mg/mL of pentoxifylline in 0.9% sodium chloride. To determine the dose effect on platelet aggregation, 10 µL from each working solution was added to 990 µL of whole blood, inverted three times and incubated, at room temperature, for 30 minutes. For the control sample, 10 µL of 0.9% sodium chloride was added to the PRP and whole blood, respectively. Following incubation, the treated samples were transferred to a cuvette for optical and impedance aggregometry analysis.

Optical aggregometry
A 2-channel optical platelet aggregometer 4 was used to analyse platelet aggregation. Aggregation was assessed using collagen 5 (10 µg/mL) as the agonist, with a temperature of 37 • C, and a stirring speed of 1200 rpm. Prior to the study, the collagen concentration was optimised to achieve consistent platelet activation. Samples were analysed based on the manufacturer's standard guidelines. 6 Briefly, 450 µL of pentoxifylline-exposed PRP was placed into a glass cuvette with a stir bar, and 500 µL of PPP was placed into a cuvette without a stir bar. Samples were incubated for 1 minute at 37 • C, placed into the aggregometer, and stable baseline values corresponding to 0% and 100% aggregation were obtained using PRP and PPP, respectively.
Collagen was added to the PRP, and platelet aggregation was mon- ization Committee, the platelet count in the PRP was not adjusted to a standardised count by dilution with PPP prior to analysis (Cattaneo et al., 2007;Linnemann et al., 2008;Mani, 2005 transferred to a plastic cuvette containing a magnetic stir bar. Samples were incubated at 37 • C for 5 minutes, placed into the aggregometer, and a reusable impedance probe was inserted into each cuvette. Collagen, 10 µg/mL, was then added to the sample and aggregation was monitored for 12 minutes. The maximal amplitude, measured in ohms, was calculated and recorded as an indicator of maximal aggregation. Four samples per pentoxifylline concentration were analysed, and the results were averaged to yield a single value. For the samples containing luciferin-luciferase, 100 µL (16 µg luciferin, 1760 Units d-luciferase) was added to the whole blood prior to the addition of collagen. All samples were analysed within 4 hours of blood collection.

Statistical analysis
Based on the mean and standard deviation values (53.8% ± 9.9) for optical aggregometry from a previous study (Mclewee et al., 2018), with a power of 90% and significant level of 0.05, 6 dogs were calculated to be needed to identify a 25% decrease in platelet aggregation. Separate linear mixed models using PROC MIXED were fit for median optical and median impedance aggregometry outcome in a statistical computer program. 9 Pentoxifylline concentration, inclusion of luciferin-luciferase, and pentoxifylline concentration by luciferinluciferase interaction were included as fixed effects. Dog identification was included as a random effect with variance components covariance structure. Differences in least squares means were determined for outcomes with significant main effect or interaction terms. The distribution of the conditional residuals was evaluated for each outcome to ensure the assumptions of normality and homoscedasticity for the statistical method had been met. An alpha level of 0.05 was used to determine statistical significance for all methods.

Optical aggregometry
The optical aggregometry results for all pentoxifylline concentrations are represented in Figure 1. There were no significant differences in the maximum amplitude at any pentoxifylline concentration (p = 0.964). The addition of luciferin-luciferase also had no significant effect on aggregometry results.

Impedance aggregometry
The impedance aggregometry results for all pentoxifylline concentrations are represented in Figure 2. The addition of a luciferin-luciferase reagent was associated with significantly (p < 0.001) greater platelet aggregation in response to a collagen agonist, regardless of the presence or absence of pentoxifylline. 9 SAS for Windows 9.4 SAS Institute, Inc., Cary, NC, USA.

F I G U R E 1
Maximum amplitude (percentage) of aggregation via optical aggregometry in canine blood exposed to saline control (0.9% sodium chloride) or pentoxifylline (1 and 2 µL/mL), both with and without concurrent exposure to luciferin-luciferase [Chronolume ® (CL)]. The box and whiskers plot demonstrates the median (line), interquartile range (box) and total range (whiskers)

F I G U R E 2
Maximum amplitude (ohms) of aggregation via impedance aggregometry in canine blood exposed to saline control (0.9% sodium chloride) or pentoxifylline (1 and 2 µL/mL), both with and without concurrent exposure to luciferin-luciferase [Chronolume ® (CL)]. The box and whiskers plot demonstrates the median (line), interquartile range (box) and total range (whiskers). '*' illustrate significant differences from non-Chronolume ® values

DISCUSSION
The results of our study indicate that, contrary to our hypothesis, the in vitro exposure of platelets to pentoxifylline does not inhibit platelet aggregation in dogs when using collagen as an agonist. Our results are in contrast to similar studies using human platelets, where exposure to pentoxifylline in vitro inhibited platelet aggregation (De La Cruz et al., 1993;Magnusson et al., 2008). In contrast, a previous ex vivo study performed in dogs treated with pentoxifylline did not detect drug-associated inhibition of platelet aggregation, a finding that is now supported by the findings of our in vitro study (Rees et al., 2003).
Our study evaluated the effects of pentoxifylline on platelet function both with and without the use of a luciferin-luciferase reagent, a reagent that can artificially enhance platelet aggregation and potentially obscure any pentoxifylline-associated inhibition of platelet function. Luciferin-luciferase preparations allow for additional assessment of platelet function by evaluating dense granule release following platelet activation by measuring the luminescence of ATP secretion.
Measurement of luminescence provides a more direct measure of platelet activation compared to light scattering properties associated with platelet aggregation within a PRP sample, and it is common laboratory practice to utilise aggregometers that concurrently measure both aggregation and luminescence following the addition of a platelet agonist.
The luciferin-luciferase preparation used in our study, and a previous ex vivo pentoxifylline study in dogs (Rees et al., 2003), however, has been shown to potentiate platelet aggregation, as evaluated by optical aggregometry, in both normal dogs and thrombocytopathic dogs when stimulated with ADP, collagen and thrombin, suggesting that the use of this product in dogs could obscure congenital and acquire thrombocytopathic conditions in dogs (Callan et al., 1998). In fact, previous work in our laboratory has confirmed that the addition of luciferinluciferase can artefactually obscure aspirin-associated inhibition of platelet aggregation. 1 The reason this luciferin-luciferase preparation appears to induce platelet activation in dogs, but not humans, is unknown, but the in vitro addition of magnesium sulfate, an ingredient in the luciferin-luciferase product, prior to analysis has been proposed as a potential mechanism (Callan et al., 1998). Our study confirmed that the presence of luciferin-luciferase significantly increased the magnitude of platelet aggregation in response to a collagen agonist, independent of the presence of pentoxifylline. Interestingly, this effect was only observed with impedance aggregometry, and not with optical aggregometry. The reason for the difference in the platelet responses to luciferin-luciferase with the two different types of aggregometry is unclear. Nevertheless, our study confirmed the findings of the previous study by Rees and others (Rees et al., 2003), that pentoxifylline does not appear to inhibit canine platelet aggregation when using collagen as an agonist, and that this observation is not erroneously affected by the presence of luciferin-luciferase.
In both human and veterinary medicine, platelet aggregometry is considered the gold standard for evaluation of platelet function (Lordkipanidzé et al., 2009;Nielsen et al., 2007). Aggregometry measures the ability of platelets to aggregate following the use of specific agonists to activate platelets in either PRP or whole blood (Lordkipanidzé et al., 2009). Similar to previous studies (Rees et al., 2003), we used collagen as the agonist to activate the platelets. Although collagen provides consistent platelet activation and is commonly used to Although pentoxifylline has not been shown to inhibit platelet function in dogs, other PDE inhibitors have been associated with druginduced platelet dysfunction. Dipyridamole, for example, inhibits both PDE3 and PDE5, and may potentiate the inhibitory effects of prostacyclin and nitric oxide (Gresele et al., 2011). Dipyridamole is an ineffective inhibitor of platelet function when used as a single agent anti-platelet therapy in dogs, but enhances platelet dysfunction when combined with low dose aspirin (Weselcouch et al., 1987). Pimbobendan, another inhibitor of PDE3 and PDE5 isoforms, does not inhibit platelet function at clinically applicable doses in dogs (Shipley et al., 2013). Sildenafil also inhibits PDE5 but, in humans, does not inhibit platelet aggregation when used as a single agent (Gresele et al., 2011;Wallis et al., 1999).
The potential inhibitory effects of pimobendan or sildenafil when combined with an antiplatelet agent in dogs are unknown, and similarly it is also unknown if the combination of pentoxifylline with other antiplatelet agents will enhance inhibition of canine platelet function.  (Marsella et al., 2000). Exposure of platelets to higher concentrations of pentoxifylline, although such concentrations would be unlikely to be clinically relevant, may have revealed concentration-dependent alterations in platelet aggregation. Fifth, the user manual for the aggregometer used in this study recommends performing platelet analysis within 3 hours of venipuncture. In our study, a majority of samples were analysed within 3 hours of venipuncture, but with the addition of a pentoxifylline incubation period prior to analysis, a few samples were analysed between 3 and 4 hours after venipuncture. It is possible that the extended period of time could have adversely affected our results, although this slightly extended time period has been used previously in dogs (Blois et al., 2010). Finally, the samples in our study were exposed to pentoxifylline for only a single fixed time duration of 30 minutes.
This exposure duration was selected to coincide with peak plasma concentrations since, when given with food, peak plasma concentrations of pentoxifylline occur 30 minutes after the initial oral administration (Marsella et al., 2000). It is possible that with a longer in vitro platelet exposure to pentoxifylline, the results of our study would have been different.
The results of our study confirm that, unlike in humans, pentoxifylline does not exert an in vitro inhibitory effect on canine platelet aggregation using collagen as an agonist. Our study also confirms that the addition of a luciferin-luciferase reagent during platelet aggregometry can artificially enhance platelet aggregation, which could potentially have an impact on interpretation of studies performed using this reagent. Our study mimics the effects on platelet function of a single dose exposure to pentoxifylline, and it is still unknown if long-term oral pentoxifylline administration will inhibit platelet aggregation in dogs.