Experimental investigation on the factors affecting VWF damage based on an in vitro blood-shearing platform

With the increasing number of people suffering from heart failure, ventricular assist devices have gradually become an effective way to treat end-stage heart failure. However, the blood damage caused by ventricular assist devices has not been effectively solved, which is an obstacle to its clinical promotion. Most research focused on erythrocyte damage under shear stress, while few researches were conducted on the interaction between blood under shear stress and the induction of von Willebrand factor(VWF) damage. This research used a vortex oscillator blood-shearing platform to conduct in vitro experiments and used immunoblotting to quantify VWF damage in sheared samples to study the laws of shear-induced VWF damage under different shear stress, different exposure times, different blood components, and hemolysis conditions. It was found that VWF damage increased with exposure time and shear stress. At the same time, under lower shear stress, other blood components had little effect on VWF damage, while in a higher shear stress, other blood components would accelerate VWF damage. Hemolysis will also affect VWF damage, and the higher the degree of hemolysis, the higher the rate of VWF degradation in the plasma. The results of this research provide a reference for VWF damage evaluation standards and follow-up research and also guide for improving the design of ventricular assist devices to reduce VWF damage.


Background
Because of the increase in aging obesity and population, heart failure is a global disease that affects nearly 26 million people worldwide [1,2].Ventricular assist devices (VADs) as an important treatment option for patients with advanced heart failure are widely used in the clinic [3].Since 2019, more than 600 million have been diagnosed withCOVID-19.Patients with severe COVID-19 sometimes have acute respiratory distress syndrome and require extracorporeal membrane oxygenation (ECMO) support [4,5].Bleeding is a common and severe complication after the implantation of VADs or ECMO, which is caused by acquired von Willebrand syndrome [6][7][8].acquired von Willebrand syndrome develops because of high shear stress inside the VADs and ECMO.The normal physiological shear stress in the human body is less than 10 Pa.However, the non-physiological shear stress (NPSS) in these devices can even reach several times over the normal physiological shear stress, which can cause non-negligible mechanical damage to the blood component [9].
Currently, many studies have shown that bleeding symptoms in clinical patients with VADs or ECMO are relevant to the degradation of high molecular weight von Willebrand factor (HMW-VWF) [10].After patients use VADs or ECMO, HMW-VWF might degrade into low molecular weight fragments and gradually lose primary clotting function as a result of NPSS.10 von Willebrand factor(VWF) is a large multimeric glycoprotein that acts as an intermediary to combine collagen fibers and platelets in hemostasis.According to previous studies, high levels of NPSS could induce the degradation of HMW-VWF and ultimately increase the risk of bleeding.It has also been reported that the loss of HMW-VWF is the result of the combination of mechanical shear stress and enzymatic cleavage [11].Although the development and optimization of VADs or ECMO have successfully avoided the non-physiological environment of high shear stress, the blood is still circulatively exposed to the non-physiological environment of lower shear stress.
In vitro, blood-shearing experiments are a reasonable way to explore the law of blood damage and improve the design of VADs or ECMO [12][13][14].Over the past few decades, researchers have been committed to studying the damage of erythrocytes, platelets, and VWF by different in vitro blood-shearing platforms.Although many researchers have conducted in-depth studies on VWF damage caused by shear stress, the complete mechanism of VWF damage remains unclear.The relationship between shear stress and VWF damage is affected by many factors, such as blood flow speed, blood temperature, and other components in blood.These factors may interact with each other to make the relationship more complex, so this issue needs to be studied in greater depth.Thus, we will explore the effects of shear stress, exposure time, blood composition, and hemolysis on shear-induced VWF damage based on a blood-shearing platform.
This paper is organized as follows: details of the in vitro blood-shearing platform and experimental procedures are provided in the "Materials and Methods" Section.The VWF damage results are described in the "Results" Section, which mainly includes (1) The effect of other blood components on VWF damage was studied by comparing the degradation of HWM-VWF in whole blood and plasma; (2) The effect of hemolysis status on VWF damage was studied by comparing the degradation of HWM-VWF in plasma with different hemolysis.The factors affecting VWF damage and the limitations and future directions are discussed in the "Discussion" Section.Finally, the important works of this study are summarized in the "Conclusion" Section.

In vitro Blood-shearing platform
As shown in Figure 1A, 1B. the vortex oscillator was used as the in vitro blood-shearing device in this study to simulate the effect of VADs-related mechanical stress on VWF.Vortex oscillator made the liquid in the centrifuge tubes produce eddy currents using Rotation so that the liquid was thrown to the tube wall under the action of centrifugal force, forming a liquid layer.The wall shear stress was generated under the action of relative movement between the liquid layer and the tube wall.Shear stress on blood can be controlled by changing the rotation speed of the vortex oscillator.Therefore, the shear situation of blood under different shear stress can be simulated through this device.In this experiment, the centrifuge tube with a rotating radius of 2 mm was used as a blood container, a schematic diagram of the centrifuge tube is shown in Figure 1C.Sodium chloride injection was used to flush the internal tube, and 3 mL of whole blood or plasma was filled into each centrifuge tube without any contact with outside air before the test.

Blood flow parameters
The magnitude of the wall shear stress increases as the rotational speed increases.The trends in the radial distribution change as a function of rotational speed and the liquid fill volume.A two-dimensional extension of Stokes' second problem for an infinite, immersed plate undergoing circular motion has been used extensively to estimate the magnitude of the wall shear stress at the bottom of the tube.For a constant rotational speed, Stokes' approximation yields that the magnitude of the wall shear stress is constant over the entire tube surface.
The wall shear stress in the tube can be determined as where represents the inner radius of the centrifuge tube, represents the rotational speed of the vortex oscillator, is blood density, is blood viscosity [15].According to the equation: In this study, the rotational speed of the vortex oscillator was set to 1,000 rpm to 2,500 rpm.Represents blood density, set as 1056 kg/m 3 .Represents the dynamic viscosity coefficient, set as 0.0035 Pa•s.According to the formula calculation, the corresponding shear stress range was from 3.48 to 21.05 Pa, which had a positive correlation with the rotational speed.

Computational fluid dynamic simulations
The mesh of the flow field in the vortex oscillator blood-shearing devices was generated by Fluent meshing and solved by Fluent v6.3.26.The software was commercially available from Ansys Fluent, Inc. (Lebanon, NH, USA).Three different mesh sizes were fabricated by the tetrahedral mesh method to study the mesh independence.The calculation grid is equipped with ten layers of grids on the wall surface to correctly analyze the boundary layer, and the mesh is shown in Figure 2. The wall shear stress corresponding to different mesh quantities is shown in Table 1.When the change rate of the wall Figure 2 The mesh used for CFD analysis at 2,500 rpm.Similar mesh are obtained for other rotational speed.CFD, Computational fluid dynamic. is the wall stress of 2.2 million mesh stress is less than 1%, the calculation result is independent of the number of mesh.Finally, the total number of fluid domain grids is 825,621.Blood was treated as an incompressible Newtonian fluid with a dynamic viscosity of 0.0035 Pa•s and density of 1056 kg/m 3 in simulation.According to the Reynolds number (< 2,300) at the maximum cylinder speed (2,500 rpm), a Laminar flow model solver was applied.The simulations were done with a second-order upwind discretization scheme, and the pressure-velocity coupling algorithm SIMPLE was utilized with the second order accuracy for pressure.

Mesh
Taking the condition of 2,500 rpm as an example, the simulation result is shown in Figure 3.The simulation results are in agreement with the calculated results.

Sample preparation
Commercial porcine blood was used in the tests because of its availability and similarity to human blood.500 mL Porcine whole blood was collected and mixed 1:20 with 3.24 wt% sodium (Yuanye Bio-Technology, Shanghai, China) from a healthy pig.This anticoagulated whole blood was stored at a temperature of 0 °C-10 °C and transported to the laboratory 2 h after blood collection.Activated clotting time of blood was controlled to be greater than 300 s to avoid blood clotting during experiment [16].
The following four kinds of samples were used in this experiment, namely, whole blood, pure plasma, 25 mg/dL hemoglobin plasma, and 100 mg/dL hemoglobin plasma, to study the shear-induced VWF damage under different rotation speeds, different exposure time, different blood components and different hemolysis.25 mg/dL of hemoglobin plasma and 100 mg/dL of hemoglobin plasma were prepared using pig red opal powder and pure plasma, respectively.Thirty-two centrifugal tubes were prepared before the experiment, of which sixteen were used to study VWF damage at 1,000 rpm, and the other sixteen were used to study VWF damage at 2,500 rpm.Four centrifugal tubes were prepared for each blood sample, and the shear conditions under different exposure times of 0 min, 20 min, 40 min, and 60 min at the same rotational speed were studied, respectively.By adjusting the rotation speed of the vortex oscillator, different shear stresses were controlled.Sheared blood samples under each parameter condition were collected.Each sample was then centrifuged at 1,500 g for 10 min.Plasma was isolated from the centrifuged sample and subjected to VWF damage measurement.

Measurement of shear-induced VWF damage
Thaw the refrigerated plasma sample and centrifuge at 15,000 g for 15 min.The upper plasma from each sample was collected for immunoblotting.The VWF multimers in plasma samples were separated by gel-electrophoresis, and transferred onto a 0.45 µm polyvinylidene difluoride membrane (Immobilon-P; Millpore Corporation, Bedford, MA, USA).Then, primary antibody (Polyclonal Rabbit Anit-human Von Willebrand Factor, Cell Signaling Technology, Boston, USA) and secondary antibody (Polyclonal Goat Anti-rabbit lgG HRP-linked Antiboby, Massachusetts, USA) were performed to detect the VWF multimers molecular weight distribution on the film.Then Image-J software was used to process the VWF bands image obtained from the Kodak film (Rayco Medical Products Company Limited for Carestream Health, Xiamen, China) development.The section of high molecular weight VWF multimers was selected to acquire the gray value corresponding to each sample.The amount of VWF damage was calculated using the following formula and the acquired gray value， Where the Hsample is the gray value of high molecular weight bands corresponding to the sheared sample, and Hcontrol is the gray value of high molecular weight bands corresponding to the initial sample.

Statistical analysis
Each experiment represents a minimum of three replicate tests.All statistical analyses were performed using JMP Pro ver.14 (SAS Institute, Cary, NC).Data are reported as the mean ± standard deviation (SD).Statistical analysis was carried out using a Student t-test (unpaired) among different samples, and a P value < 0.05 was considered statistically significant.

Figure 3
Shear stress distribution of the device at 2,500 rpm Submit a manuscript: https://www.tmrjournals.com/bmec

The loss of HMW-VWF multimers
The image of VWF multimer bands obtained by immunoblotting are shown in Figure 4 and Figure 5.The section in the dotted box represented the HMW-VWF multimer bands.
Based on the image, the loss of HMW-VWF multimers (%) was calculated and shown in Table 2 and Table 3.The loss of HMW-VWF increased with exposure time in different shear stress.When other conditions are the same, the higher the shear stress is, the more HMW-VWF loss is.

The effect of other blood components on VWF damage
As shown in Figure 6, the loss of HMW-VWF multimers (%) in whole blood and pure plasma samples at the rotation speed of 2,500 rpm (Shear stress 21.05 Pa) were compared.The results showed that the loss of HMW-VWF multimers (%) in whole blood and pure plasma was 35.88% and 13.72% after 60 min, respectively.The loss of HMW-VWF multimers (%) in whole blood was 2.6 times that in pure plasma.That is, VWF damage in whole blood was significantly higher than that inpure plasma (P < 0.05).At high shear stress, the loss of HMW-VWF multimers (%) in whole blood was more affected by shear stress.Other blood components in the blood, such as red blood cells, can exacerbate VWF damage.

The effect of the degree of hemolysis on VWF damage
As shown in Figure 7, the loss of HMW-VWF multimers (%) in pure plasma, 25 mg/dL hemoglobin plasma, and 100 mg/dL hemoglobin plasma at the rotation speed of 2,500 rpm (Shear stress 21.05 Pa) were compared.The results showed that after 60min, the loss of HMW-VWF multimers (%) in pure plasma was the lowest, at 13.72%; the loss of HMW-VWF multimers (%) in 25 mg/dL hemoglobin plasma was 15.83%, which was 2.11% higher than that in pure plasma; the loss of HMW-VWF multimers (%) in 100 mg/dL hemoglobin plasma was the highest.It was 23.21% higher than that of pure plasma by 9.49%.This indicates that the degree of hemolysis will have a certain effect on the loss of HMW-VWF multimers (%) when the same shear stress is destroyed.The higher the degree of hemolysis, the higher the loss of HMW-VWF multimers (%).

Discussion
For the problem of circulating blood in the body iteratively exposed to the non-physiological environment after using VADs and ECMO, a vortex oscillator blood-shearing platform was designed to explore the relationship between VWF damage and two flow-parameters (high shear stress and iterative exposure time)and blood components.Compared to other blood-shearing device, a vortex oscillator blood-shearing platform could ensure that the shear stress generated in the centrifuge tube can be controlled quantitatively by changing the rotation speed in laminar flow and no additional serious blood Submit a manuscript: https://www.tmrjournals.com/bmecdamage caused by other factors, such as the sealing structure and mechanical bearing.According to the formula, the NPSS provided by the device is in the range of 3.48 to 21.05 Pa, and the computational fluid dynamics simulation is used to verify the results.This indicates that the vortex oscillator blood-shearing device can be used to study the VWF damage caused by blood contact medical devices in vitro.
The data collected from the experiment showed that VWF damage increased with exposure time under the same conditions.The higher the shear stress, the higher the VWF damage under the same exposure time.In addition, under high shear stress, the loss of HMW-VWF multimers in whole blood was the fastest in 0-20 min.By comparing the whole blood and pure plasma samples, it was found that VWF damage in whole blood was much greater than that in pure plasma.It is known that during hemostasis, VWF can expand into chains to capture platelets in the blood, thus adhesion to collagen fibers to form thrombosis. Excess long-chain VWF polymers will be degraded to a low molecular state by von Willebrand factor lyase (ADAMTS-13) and cleared by the body.In addition, the VWF polymers will also unfold into chains under NPSS, exposing the ADAMTS-13 binding site, resulting in the loss of a large number of HMW-VWF to a low molecular state.It can be inferred that in the environment of high shear stress, some components in whole blood may develop from clumps into chains in a short time due to the intensification of adhesion and then degrade.It makes it impossible to play the clotting effect when the body needs to stop bleeding, resulting in clotting disorders [17,18].By comparing plasma samples with different hemolysis degrees under the same shear stress, it was found that the hemolysis produced in the blood was correlated with VWF degradation, and the higher the degree of hemolysis, the higher the degradation rate of VWF in the blood.
Through this study, it is found that other components in the blood and the degree of hemolysis of the blood have certain effects on the damage of VWF in the blood, which may be helpful for future research and improvement of VADs and ECMO.Because VWF damage may be not only related to shear stress and exposure time, but also other influencing factors.Since VADs often need to be used together with anticoagulant drugs to reduce the formation of thrombus, and anticoagulant drugs often cause hemolysis to a certain extent, experiments have shown that hemolysis will further increase VWF damage.Therefore, in the future VAD optimization, how to reduce the occurrence of hemolysis while reducing the damage of VWF should be evaluated and tested.
There were also some study limitations and future directions.Due to the limitation of the platform, the device speed can only reach 2,500 rpm, and it is planned to obtain more samples under more working conditions with the help of a vortex oscillator with better performance in subsequent studies.In the future, the device could be used to conduct more in vitro simulations to explore the effects of blood flow parameters on other components of the blood, such as white blood cells and platelets.In addition, the literature suggested that blood damage may differ from species to species.We planned to do more research with human blood and ovine blood in the future.

Conclusion
In this study, a vortex oscillator blood-shearing platform was used to conduct in vitro experiments and used immunoblotting to quantify VWF damage in sheared samples to study the laws of shear-induced VWF damage under different shear stress, different exposure times, different blood components, and hemolysis conditions.The loss of HMW-VWF was considered as the result of the accumulation over exposure time under non-physiological shear stress.Under lower shear stress, other blood components had little effect on VWF damage, while in higher shear stress, other blood components would accelerate VWF damage.Hemolysis will also affect VWF damage, and the higher the degree of hemolysis, the higher the rate of VWF degradation in the plasma.The laws of VWF damage derived from this study could provide useful guidance for the design and optimization of VADs and ECMO.

Figure 1
Figure 1 (a) Schematic diagram of vortex oscillator platform; (b) Physical diagram of vortex oscillator platform; (c) Schematic diagram of centrifuge tube.

Figure 4
Figure 4 The image of VWF multimers for different exposure time at the rotation speed of 1,000 rpm (Shear stress 3.48 Pa).VWF, von Willebrand factor.

Figure 5 Figure 6
Figure 5 The image of VWF multimers for different exposure time at the rotation speed of 2,500 rpm (Shear stress 21.05 Pa).VWF, von Willebrand factor.