Differences in human and commercial hybrid pig platelet activation induced by borosilicate glass beads in a modified chandler loop-system

Abstract

 The pig (Sus scrofa) is the most widely used large animal model in Europe, with cardiovascular research being one of the main areas of application.

Adequate refinement of interventional studies in this field, meeting the requirements of Russell and Burch’s 3 R concept, can only be performed if blood-contacting medical devices are hemocompatible.

Because most medical devices for cardiovascular interventional procedures are developed for humans, they are tested only for compatibility with human blood.

The aim of this study was therefore to determine whether there are differences in behavior of human and porcine platelets from commercial hybrid pigs when they come into contact with borosilicate glass, which was used as an exemplary thrombogenic material.

For this purpose, changes in platelet count, platelet volume and platelet expression of the activation markers CD61, CD62P and CD63 were measured using a modified chandler loop-system simulating the fluidic effects of the bloodflow.

Commercial hybrid pig and human platelets showed significant adhesions to borosilicate glass but the commercial hybrid pigs platelets showed a significantly higher tendency to adhere to borosilicate glass.

In contrast to human platelets the platelets of commercial hybrid pigs showed significant activation after 4 to 8 minutes exposure to borosilicate glass and there were differences among the ratios of surface and activation markers in between the platelets of both species.

1. Introduction

Recently, sheep and minipig platelets were shown to differ from human platelets in activability after 4 to 8 minutes of exposure to borosilicate glass as a thrombogenic material, although these differences were small [3, 7].

The study was motivated by the fact that commercial hybrid pigs, as well as minipigs and sheep are an important animal model for interventional cardiovascular studies, and therefore medical devices in these animals are in contact with blood and many of those products and devices have only been tested for hemocompatibility with human blood.

However, for a test refinement, that meets the requirements of the 3 R concept of Russell and Burch [1], it should also be tested whether the hemocompatibility of these medical devices is also given for the blood of the used animal model.

In addition to sheep and minipigs commercial hybrid pigs are an important large animal model for interventional cardiovascular studies.

2020, 6854 pigs were used in basic research in Germany, of which 403 (5.9%) were used for cardiovascular studies and studies on the blood and lymphatic system [2].

This motivated us to investigate in addition to the Göttingen minipigs parameters [3] the platelet count, platelet aggregation, and platelet activation changes of commercial hybrid pigs in comparison to human platelets after glass contact for 4 to 8 minutes.

These studies were performed under dynamic testing conditions in a modified chandler loop-system [4, 5].

2. Materials and methods

2.1. Humans and animals

Human blood samples were collected from healthy adults (2 male and 1 female) who consented to blood collection on a voluntary basis.

Blood sampling from the animals was approved by the Lower Saxony State Office for Consumer Protection and Food Safety (approval number 33.8-42502-05-20A513) and was performed at two adult and healthy commercial hybrid pigs (1 female and 1 castrated male, 4 years old).

The animals were housed under conventional holoxenic hygiene conditions and under care according to the guidelines of the Federation of European Laboratory.

Animal Science Associations [15]. Animals were without clinical signs of disease and monitored by whole blood analysis and general examination by an experienced veterinarian on each day oftesting.

From each individual (commercial hybrid pig and human) were 3 blood samples taken and tested.

2.2. Blood sampling and dynamic test model

For each test at most 83 ml blood were drawn from the external jugular vein using an 18 G sized cannula (human: SR + DM1832WX, Terumo Europe NV, Belgium, pig: 15885680, Vivomed GmbH, Germany) into 8 S-Monovette^® 10 ml 9NC blood collection systems (02.1067.001, Sarstedt AG & Co. KG, Germany) and one S-Monovette^® 1.6 ml K3 EDTA blood collection system (04.1951.001, Sarstedt AG & Co. KG, Germany) to prevent spontaneous coagulation. Each individual had a recovery period of at least 7 days between two blood samplings.

Within 30 minutes, the citrate-anticoagulated whole blood was centrifuged at 120 G for 15 minutes and platelet-rich plasma (PRP)& was carefully removed and placed in 15 ml polypropylen test tubes (62.554.502 PP, Sarstedt AG & Co. KG, Germany).

For the tests a sterilized modified chandler loop-system [5] was filled with the PRP. In brief, the modified chandler loop-system consisted of two pieces of medical-grade platinum-cured silicon tubing (96410-16, Masterflex SE, Germany) with a non-thrombogenic inner surface [16], which were mounted to a closed loop using two three-way valves made of polyvinylidene fluoride (394602, Becton Dickinson Infusion Therapy AB, Sweden).

The closed tube was filled with the PRP via one of both three-way valves. The second three-way valve ensured that the air in the system could escape.

After insertion of the silicone tubing into a roller pump (Masterflex^® L/S^®, Masterflex SE, Germany) a defined flow was maintained (7.5 ml/h).

Filled only with PRP the system was regarded as the negative-reference-system (non-thrombogenic control).

For tests under thrombogenic conditions the tube system was partly filled with borosilicate glass beads with an outer diameter of 0.75-1.0 mm (A554.1, Carl Roth GmbH + Co. KG, Germany).

The borosilicate glass beads (n = 1084; calculated total surface 2221 mm²) were prevented to follow the blood by a medical-grade V2A-steel-based mesh with a mesh width of 0.5 mm, which was integrated in the tube.

Directly after the initial filling of the tube system with PRP was completed, baseline values were determined. PRP volumes removed were always replaced isovolemically by autologous PRP.

The PRP was exposed to the tube system in a total of 40 consecutive test cycles. Platelet analysis was performed after the PRP within the closed tubing had circulated through the total tube system 20 and 40 times.

This corresponded to a circulation time of 4 and 8 minutes.

2.3. Platelet count and mean platelet volume

The platelets that were free to move in the tube system and that did not became adherent to inner surfaces of the platinum cured-silicone tube system or the borosilicate glass beads were analyzed in terms of total number and mean platelet volume (as parameter for platelet aggregation).

A double measurement was performed on all samples. These parameters were measured using an automated hematology analyzer (ProCyte Dx^®, IDEXX, Germany).

2.4. FACS analysis

Platelet activation was evaluated using flow cytometry (MACSQuant^® 10 Analyzer, Miltenyi Biotec B.V. & Co. KG, Germany). Platelets were defined as CD42a (= GPIX) positive cells within the PRP-sample.

To assess the platelet activation state, the platelet markers, CD62P (P-selectin) and CD63 (GP53) were used.

In preliminary concentration tests the antibodies used (recombinant human IgG1 antibodies; PE-conjugated anti-CD42a (130-100-966), FITC-conjugated anti-CD61 (130-110-748), APC-conjugated anti-CD62P (130-118-980), PE-Vio 771-conjugated anti-CD63 (130-118-081); Miltenyi Biotec B.V. & Co. KG, Germany) were titrated to the endpoint so that measurements could always be conducted at saturation level, thereby excluding any dependence between the fluorescent intensity and the concentration of the antibodies.

Between 50,000 and 100,000 particles were analyzed in every flow cytometric measurement to ensure reliable analysis.

2.5. Statistics

Data analysis was performed with SPSS Statistics 28 (IBM Deutschland GmbH, Germany).

For each measurement, two PRP samples (silicone + glass beads silicone) were available.

Each of these PRP samples was measured three times, before circulation and after passing through the modified Chandler loop system 20 and 40 times, respectively.

For each individual the means of all single measuring points per measurement were calculated and used for further calculations.

Data sets consisting out of the means from each individual were tested with Kruskal-Wallis tests for Gaussian distribution and Levene's tests for homoscedasticity. Data were expressed as means and standard deviations (±SD) and graphs were created with Origin 2022 (OriginLab Corporation, United States).

For pairwise comparisons t-tests for independent samples were conducted and for multiple comparisons one-way-ANOVA with Tukey’s tests.

3. Results

In human and commercial hybrid pig PRP, platelet counts decreased within 40 test cycles or 8 minutes more when they were in contact with the borosilicate glass beads than when they were just in contact with the silicone tubing as the control (see Fig. 1A and B).

Fig. 1.

Total number of non-adherent platelets from platelet rich plasma (PRP) of adult commercial hybrid pigs (n = 2) and humans (n = 3); each PRP sample was measured three times, before and when it has passed 20 or 40 times through a modified chandler loop-system with or without filled with glass beads; means and standard deviation; * p < 0.05, ** p < 0.01.

In this regard, the decrease in platelet count in the silicone tubing without contact to borosilicate glass beads was greater for human PRP (-5.27±1.66%) than for commercial hybrid pig PRP (-1.85±0,37%) after 40 test cycles or 8 minutes.

Exposure to borosilicate glass beads in comparison to just the silicone tubing caused a stronger decrease in the platelet counts in PRP of both species over time, whereby the decrease in platelet count after 40 test cycles or 8 minutes was also greater in human PRP (-20.73±3.82%) than in comparison to commercial hybrid pig PRP (-3.40±1.45%) (see Fig. 1C and D).

In mean platelet volume (MPV) no significant changes was measurable in human and commercial hybrid pig PRP after 20 and 40 test cycles with or without exposure to borosilicate glass (seeFig. 2).

Fig. 2.

Volume of non-adherent platelets from platelet rich plasma (PRP) of adult commercial hybrid pigs (n = 2) and humans (n = 3); each PRP sample was measured three times, before and when it has passed 20 or 40 times through a modified chandler loop-system filled with glass beads; means and standard deviation.

The proportion of platelets (CD42a + cells) within the PRP samples that were also positive for CD61 remained relative constant over the time of 20 or 40 test cycles regardless whether the PRP was exposed borosilicate glass beads or not (see Fig. 3).

Fig. 3.

Percentage of CD61 + /CD42a⁺ cells in platelet rich plasma (PRP) of adult commercial hybrid pigs (n = 2) and humans (n = 3); each PRP sample was measured three times, before and when it has passed 20 or 40 times through the modified chandler loop-system filled with glass beads; means and standard deviation.

In contrast to human platelets the contact of commercial hybrid pig platelets with the platinum-cured silicone tubing led to a significant activation which was indicated by an increased ratio of CD62P + and CD63 + cells within the population of the CD42a + platelets after passing though the modified chandler loop-system 20 or 40 times (see Figs. 4 and 5).

Fig. 4.

Percentage of CD62P⁺/ CD42a⁺ cells in platelet rich plasma (PRP) of commercial hybrid pigs (n = 2) and humans (n = 3); each PRP sample was measured three times, before and when it has passed 20 or 40 times through the modified chandler loop-system filled with glass beads; means and standard deviations; * p < 0.05, ** p < 0.01.

Fig. 5.

Percentage of CD63 + /CD42a⁺ cells in platelet rich plasma (PRP) of commercial hybrid pigs (n = 2) and humans (n = 3); each PRP sample was measured three times, before and when it has passed 20 or 40 times through the modified chandler loop-system filled with glass beads; means and standard deviations; * p < 0.05, ** p < 0.01.

Where human platelets just displayed a minor mean increase in CD62P + /CD42a + cell ratio, which did not significantly differed between the platelets that had contact with the borosilicate glass beads and those who had not, there was an significant activation of the commercial hybrid pig platelets after contact with the borosilicate glass beads (9.8±0.14%) in contrast to the just the platinum-cured silicone tubing as a control (7.48±0.8%) after 40 times passing through the modified chandler loop-system (p < 0.05).

Regarding the species-specific expression of the surface antigens among the CD42a + platelets there were major differences before passing through the modified chandler loop-system between human and commercial hybrid pig platelets detectable.

Whereby the baseline expression of CD61 on platelets in commercial hybrid pigs (99.96±0.02%) was slightly higher than humans (96.6±4.13%) the mean expression of CD62P among the CD42a + platelets in humans (26.81±11.3%) exceeds the level of CD62P in commercial hybrid pigs platelets (7.31±0,18%) by the factor 3,67.

A similar tendency in comparison to the ratio of CD62P + / CD42a + platelets is observable regarding the ratio of the CD63 + cells among the CD42a + fraction.

Here the mean human platelets level of CD63 (13.62±7.29%) exceeds the commercial hybrid pig ones (9.72±0,01%) by the factor 1.4 (see Fig. 6).

Fig. 6.

Percentage of CD63 + /CD42a⁺, CD62P⁺/CD42a⁺ and CD63 + /CD42a⁺cells in platelet rich plasma (PRP) of commercial hybrid pigs (n = 2) and humans (n = 3); without passing through the modified chandler loop-system filled without borosilicate glass beads; means and standard deviations.

4. Discussion

This study was performed to investigate how the behavior of commercial hybrid pig platelets after borosilicate glass contact differs from the behavior of human platelets in terms of counts, volume and activation.

An established modified chandler loop-system was used for testing, which allows the behavior of platelets to be studied ex vivo in a dynamic mode.

The test results showed partly significant differences between the platelets of the two species. In human commercial hybrid pig PRP, glass contact resulted in a decrease in platelet count, which was more pronounced in human PRP than in commercial hybrid pig PRP, and the changes in MPV were neglectable in human and commercial hybrid pig PRP.

This suggests that human platelets show stronger adherence when getting in contact with borosilicate glass than those of commercial hybrid pigs.

Goodman et al. [6] postulated that human and porcine platelets showed comparable adherence to low-temperature isotropic PYC valve leaflets, polyethylene, silicone rubber and Formvar.

Using a dynamic test model, they studied the adherence behavior of platelets from three pigs (2 to 3 months old Yorkshire pigs), three sheep (two Suffolk sheep, one crossbred between a Suffolk sheep and a Dorset sheep), and adult humans on the mentioned biomaterials. Interestingly, in Goodman’s tests, the pig platelets showed much greater adherence to silicone than the sheep platelets. This contrasts with results of our previous study, which showed a significant 39.5% decrease in ovine platelet count after only 4 minutes (20 test cycles) of silicon contact (w/o borosilicate glass bead exposure) compared to our previous study on minipigs with a decrease of 0.4% [7].

The differences from the results of Goodman et al. [6] could be traced, at least in part, to the different assays approaches, the anesthetic medication the animal received prior blood sampling or the farm animal species varying genetic backgrounds. Goodman’s group exposed platelets to the biomaterials for 45 minutes [5], whereas in the study presented here the maximum exposure time was 8 minutes.

In addition, Goodman et al. [6] used a static model, whereas our results are based on tests with a dynamic model. In this model, shear forces become effective, which might have influenced platelet adhesion.

In addition, the different roughness of the foreign surfaces used probably also played a role [8–10]. Hecker and Edwards [9] investigated how the roughness of polyvinyl chloride affects the thrombogenicity of the material.

They studied thrombus formation on tubing of varying roughness (OD 3.0 mm) in 40 adult Merino sheep after insertion of the tubing into the great saphenous vein and aortic artery. They found a positive correlation between roughness and thrombus formation.

In the test runs where platelets were exposed to the tubing system with and without borosilicate glass beads, the mechanical action of the roller pump might also have influenced platelet activation as well [8–10].

Jung et al. [11] studied the thrombogenicity of a coronary stent using the same dynamic test system as the one presented here. They found an increase in CD62P + platelets even in the tests in which the PRP was not exposed to the stents and attributed this to the mechanical action on the platelets by the roller pump.

Species-specific differences in platelet behavior on plasma protein composition must also be considered. Pelagalli et al. [12] compared the adherence of sheep, porcine, and human platelets to autologous fibrinogen and found that sheep platelets could not adhere to autologous fibrinogen without pre-activation by adenosine 5-phosphate, whereas human and porcine platelets could, and the ability of human platelets to adhere to autologous fibrinogen was more pronounced than that of porcine platelets.

It was suggested that these differences were due to structural differences in platelets and to differences in the availability of the fibrinogen (GPIIb/IIIa) receptor on the platelet surface. Species-specific differences in the ability to adhere to materials were also found by Grabowski et al. [13]. Using a dynamic closed loop model, they compared platelets from different animal species with those from humans with respect to their adhesiveness to a cellulose-based hemodialysis membrane. After 10 minutes of blood flow at a shear rate of 986 sec^-1, they found that < 100/mm² platelets each from sheep, pig, and human adhered to the surface of the membrane. In the tests with platelets from dogs and rabbits, this value was several times higher (dog 27,400±4600; rabbit 78,400±6400).

However, with regard to commercial hybrid pig platelet activation, it was found that only a small proportion of platelets were activated by glass contact. In this respect, the results were comparable to those obtained with minipig [3], sheep [7] and human blood.

If platelets from commercial hybrid pigs could be activated to a greater extent than those from humans requires further investigation because our data in combination with those from Goodmann et al. [6] suggest, that there might be not just interspecies differences but intraspecies differences as well regarding the animals genetic background like investigated different mouse strains by Paigen et al. [14].

Regardless, however, the sheep platelets in our previous study [7] seem to be more adherent to borosilicate glass than the commercial hybrid pigs platelets, as number of platelets circulating through the modified chandler loop-system plummets in contrast to our experience with porcine platelets in general after exposure to borosilicate glass and silicone.

If these observed interspecies and potentially even intraspecies or breed dependent differences in animal’s platelet biology are caused by the different ratios of surface molecules or the plasma protein composition this requires further investigation.

5. Conclusion

Although the results are limited in their significance by the small number of animals used, there were differences in the tendency to adhere after borosilicate glass contact between the human and commercial hybrid pig platelets.

Human platelets tend to adhere more than those of commercial hybrid pigs, which indicates that that platelet activation after contact with borosilicate glass is lower in commercial hybrid pigs than in humans.