Evaluation of erythrocyte membrane oxidation due to their exposure to shear flow generated by extracorporeal blood pump

Abstract

Several similarities have been found between shear stress-induced erythrocyte damage and physiological aging of erythrocytes in terms of elevated mechanical fragility, increased erythrocyte aggregation, and decreased membrane surface charge. Accordingly, we hypothesized that blood pump circulation, which generates shear stress, would accelerate erythrocyte aging, manifesting as oxidation. Therefore, the purpose of this study was to investigate the effect of blood pump circulation on erythrocyte oxidation. Fresh porcine blood was acquired from a slaughterhouse and anticoagulated with sodium citrate. About 500 mL of anticoagulated whole blood was circulated for 180 min in an in vitro test circuit comprising a BP-80 blood pump with a pump speed and a pump pressure head of 100–120 mmHg. A blood sample was taken at the start of the circulation and 180 min afterward. The hemolysis level and oxidation amount of the erythrocyte membrane were analyzed and compared between samples. Hemolysis increased with the prolongation of shear exposure inside the pump circuit. After 180 min of blood pumping in circuit, the oxidation level of the erythrocyte membrane showed an increase of 0.1 nmol/mg protein. Moreover, the membrane oxidation levels of sheared erythrocytes were greater than those of control erythrocytes. These results suggest that blood pump circulation accelerates erythrocyte aging and give us a greater understanding of the effects of blood pump perfusion.

Introduction

The main role of erythrocytes is to deliver oxygen to the organs and peripheral tissues using the intracellular hemoglobin (Hb). These cells possess the unique deformability that allows oxygen transport for Hb rich erythrocytes through peripheral blood vessels into the tissue. ¹ Erythrocytes are produced in the bone marrow and released into blood circulation, and those have lifespans of 100–120 days, during which time they are deformed into flexible shape under variety of physiological shear stresses and so many times exposed to oxidative stress in gas exchange procedure. Aged erythrocytes show decreased deformability and increased density,^{2 –4} and these erythrocytes become trapped in the spleen where they are phagocytosed by macrophages.^5,6

Regarding erythrocyte deformability, super-physiological shear stress causes an accumulation of damage to erythrocytes that can result in their destruction or lead to impaired deformability or a density increase.^7,8 Such mechanical stress also occurs inside mechanical circulatory support (MCS) devices, including blood pumps, percutaneous devices, and extracorporeal membrane oxygenators, which are used in patients with severe heart or lung failure. In addition, super-physiological shear stress causes complications such as impaired oxygen transport due to hemolysis.^9–12

In another perspective, erythrocytes are composed of a lipid bilayer that contains unsaturated fatty acids that cover internal Hb.^13,14 Because erythrocytes are constantly exposed to high concentrations of oxygen, unsaturated fatty acids on their lipid membranes are converted to lipid peroxides by reactive oxygen species (ROS). This leads to the primary product lipid peroxide and the secondary product malondialdehyde. ¹⁵ Furthermore, the auto-oxidation of Hb in erythrocytes produces ROS such as hydrogen peroxide (H₂O₂) and superoxide (O₂⁻). ¹⁶ H₂O₂ induces covalent complexes between spectrin and Hb, which increases the rigidity of erythrocyte membranes and reduces their deformability.^17,18

Taking these aspects into consideration, some researchers have reported several similarities between aged, means oxidatively stressed erythrocytes, and those damaged by shear stress in terms of elevated density, surface charge loss, and increased propensity to aggregation.^19–22 In the context of this phenomenon, we hypothesized that the mechanical shear stress generated in the pump circulation would promote oxidation of erythrocytes. Thus, the purpose of this study was to examine whether pump circulation influences erythrocyte membrane oxidation, and to understand the biological effect of such cardiovascular mechanical devices.

Materials and methods

Test blood samples

Porcine whole blood was collected and mixed 1:10 with 3.24 wt% sodium citrate at a slaughterhouse (Shibaura Zoki Co., Shinagawa, Japan). This anticoagulated whole blood was stored at a temperature of 0°C–10°C and transported to the laboratory 1 day after blood collection (n = 6).

Experimental design and blood pump circuit

A blood pump circuit was constructed to simulate the effect of MCS-related mechanical stress on erythrocytes. The circuit comprised a BP-80 Bio-Pump (Medtronic, Dublin, Ireland) a reservoir, a clamp to act as a resistor, and polyvinyl chloride tubing with an inner diameter of 9.5 mm and length of 2 m. The reason for the selection of BP-80 was due to its conventional actual achievement as clinical use and standard reference to evaluate the anti-hemolytic performance of the newly developed pump. The tubing contained ports for obtaining the inlet and outlet pressures and for blood sampling (Figure 1). Priming porcine blood (540 mL) was circulated through this simulated circuit without any contact with outside air. The pump was operated at a rotational speed of 2000 rpm against a pressure head of 100–120 mm Hg, then it resulted in a flow rate of about 5 L/min based on the characteristics of the BP-80. ²³ The flow rate was selected referring ASTM F1841-97 showing the method to evaluate anti-hemolytic performance of continuous flow pumps that expose erythrocytes to mechanical stress. The sample blood was corrected from the test circuit at the first and 180 min afterwards. The duration of 180 min was defined as a half time of ASTM F1841-97 (360 min) to examine the erythrocyte deformability and the lipid peroxide levels of erythrocyte before all of the cells is ruptured. Each sample was then centrifuged at 1500×g for 10 min and the plasma layer was subjected to hemolysis measurement. The remaining erythrocytes after plasma removal were washed with phosphate-buffered saline (Fujifilm Wako Pure Chemical Co., Osaka, Japan) twice at 1500×g for 10 min. Aged erythrocytes have a higher density, and it is widely known to be less deformable and contain higher amounts of lipid peroxide corresponding to aging degree. Thus, 1 mL of the upper and lower layers in the erythrocyte suspension after centrifugation were collected individually and defined as young and old cells, respectively, after washing procedure.

Figure 1.

Setup of a pump circulation loop for exposing blood to mechanical stress.

Measurement of shear-induced hemolysis

Plasma free hemoglobin (pfHb) values were determined using a Harboe assay. ²⁴ Briefly, 15 µL of plasma was diluted 100 times with 1485 µL of 0.1% sodium carbonate (Fujifilm Wako Pure Chemical Co.). ²⁵ The absorbance was then made at 380, 415, and 450 nm using a spectrophotometer (UVmini-1240; Shimadzu, Kyoto, Japan). The amount of pfHb was assessed using the following formula:

$$\begin{array}{l}\mathrm{p}\mathrm{f}\mathrm{H}\mathrm{b}\left(\frac{\mathrm{m}\mathrm{g}}{\mathrm{d}\mathrm{L}}\right)=(167.5\times A_{415}-83.6\times A_{450}-83.6\times A_{380})\\ \times \frac{V_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}}{V_{\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{m}\mathrm{a}}}\times \frac{1}{10}\end{array}$$

Here, A415, A450, and A380 represent the absorbance at each wavelength while V_total (µL) and V_plasma (µL) show the total volume of the measurement sample and the plasma volume, respectively.

Measurement of erythrocyte deformability

In the present study, a custom-built counter-rotating shear generator was deployed as previously described.^26,27 The system comprises a shearing chamber and an inverted microscope (Figure 2(a)). The chamber generates shear stress by counter-rotating the inner and outer cylinders. Morphological changes in erythrocytes under shear stress were obtained through a 40× objective lens (LUCPlanELN; Olympus, Tokyo, Japan) on an inverted microscope (IX-73; Olympus; Figure 2(b)). Briefly, 1490 µL of 10% polyvinylpyrrolidone (Sigma-Aldrich Co., Burlington, MA) was added to 10 µL of young or old cells for a 150-fold dilution. The shear stress to evaluate deformability was adjusted to 30 Pa and loaded for 60 s. A high-speed camera (MEMRECAM GX-1; NAC Image Technology Inc., Tokyo, Japan) was equipped with a recording speed of 100 fps and a pixel count of 512 × 512 pixels. Erythrocyte morphology was extracted from the obtained images, and the elongation index, which is described elsewhere,^28–30 was calculated by measuring the long and short axes using ImageJ (ver. 1.8.0_112; National Institute of Health, Bethesda, MD; Figure 3).

Figure 2.

Photograph (a) and Schematic view (b) of the shearing visualization system developed to determine erythrocyte morphology under shear flow.

Figure 3.

Representative image of elongated erythrocytes under shear flow (young cells after circulation in the pump circuit for 180 min) and measurement of the elongation index.

Protocol for the analysis of lipid peroxide

We measured the amount of lipid peroxidative products to assess the oxidation level of erythrocyte membranes. The reaction of malondialdehyde with thiobarbituric acid (TBA) forms reactive substances of thiobarbituric acid (TBARS), which have a maximum absorption wavelength around 532 nm. Consequently, the absorbance at 532 nm was measured using a spectrophotometer (UV-1200; Shimadzu). ³¹ The erythrocytes collected from the pump circuit were mixed with 10 µL of butylated hydroxytoluene (BHT), 1% phosphoric acid, and 0.7% TBA reagent. The mixture was then incubated at 100°C for 60 min using a block heater. Subsequently, the sample was returned to room temperature and 2000 µL of 1-butanol was added to extract TBARS. After centrifugation at 3000 rpm for 10 min, the butanol fraction was extracted with sodium hydroxide and phosphoric acid was added to remove Hb and neutralize the sample. To quantify the amount of malondialdehyde per 1 mg of protein, the amount of protein was measured using a Bio-Rad protein assay (#500-006JA; Bio-Rad Japan, Tokyo, Japan) according to the manufacturer’s procedure.

Statistical analysis

Normality testing was investigated for each dataset using an F test. When variances were considered equal, group comparisons were performed using the t-test with multiple comparisons. All data was analyzed with JMP Pro ver. 14 (SAS Institute, Cary, NC). Data are reported as mean ± standard deviation (SD) and the p < 0.05 was considered statistically significant.

Results

Shear-induced hemolysis

The amount of pfHb significantly increased from <50 mg/dL before pump circulation to >200 mg/dL after 180 min of pump circulation (p < 0.01, Figure 4).

Figure 4.

Levels of plasma free hemoglobin (pfHb) assessed by Harboe assay using spectrophotometry after 180 min of the pump circuit. Results are reported as the mean ± SD, n = 6, *p < 0.01.

Deformability of erythrocytes after shear exposure in the pump circuit

The elongation index of young cells decreased significantly from 0.56 before pump circulation to 0.51 after 180 min of pump circulation (p < 0.05, Figure 5). On the contrary, the elongation index of the old cells was 0.51 before pump circulation and 0.47 after 180 min.

Figure 5.

Elongation indices of young and old erythrocytes after 180 min of the pump circulation measured by microscopy with a shearing visualization system. Results are reported as the mean ± SD, n = 6, *p < 0.05.

Changes in the lipid peroxide levels of erythrocyte membranes

In young cells, the amount of lipid peroxide increased from 5.98 nmol/mg protein before pump circulation to 7.66 nmol/mg protein after 180 min of pump circulation (Figure 6). For old cells, the amount of lipid peroxide was 9.10 nmol/mg protein before circulation and 9.33 nmol/mg protein after 180 min. Although an overall tendency for increased oxidation of the cell membrane was observed after pump circulation, the difference was not significant.

Figure 6.

Lipid peroxide levels of young and old erythrocytes after 180 min of the pump circulation. Results are reported as the mean ± SD, n = 6.

Discussion

In the present study, pfHb levels, erythrocyte deformability, and lipid peroxide levels were quantified in the first and 180 min after the blood pump circulation. Our results showed an elevated level of pfHb, a decreased value of deformability, and an increased level of lipid peroxide.

Many researchers have evaluated pfHb levels in order to evaluate erythrocytes response to mechanical stress.^22,32,33 Hemolysis appears when the shear stress exceeds 150 Pa as inside of blood pumps. In this study, 180 min of pump circulation increased pfHb levels to greater than 200 mg/dL. Previously, Naito et al. ³⁴ examined the level of hemolysis after the use of various blood pumps, and an increased amount of pfHb, approximately 9.0 mg/dL, was confirmed with the BP-80 pump. Although the higher level of pfHb was higher in this study than in Naito et al., this can be explained by differences in pump rotational speed.

The deformability of erythrocytes decreases with longer exposure to supra-physiological shear stress,^35,36 and we found a significant decrease in the elongation index of young cells from 0.56 to 0.51 after 180 min of pump circulation. In contrast, no significant decrease in the elongation index was found for old cells. The cause is limited durability against shear stress. In other words, the erythrocytes extracted from the bottom layer of the tube after centrifugation are already the least deformable erythrocytes of the bulk of erythrocytes. Therefore, it is possible that the erythrocytes, whose deformability was further reduced, were broken by the shear stress generated in the pump circuit. Young cells showed a slight increase in lipid peroxide levels after 180 min of pump circulation. The auto-oxidation of oxyhemoglobin explains this reaction between pump circulation and erythrocyte membrane oxidation as the starting point for continuous lipid peroxidation reactions. The auto-oxidation of the oxyhemoglobin present in the blood is shown below ¹⁶ :

$${\mathrm{Hb}}^{2+}{\mathrm{O}}_2\to {\mathrm{Hb}}^{3+}+{\mathrm{O}}_2^{-}$$

$${\mathrm{2H}}^{+}+{\mathrm{O}}_2^{-}+{\mathrm{Hb}}^{2+}{\mathrm{O}}_2\to {\mathrm{Hb}}^{3+}+{\mathrm{O}}_2^{-}+{\mathrm{H}}_2{\mathrm{O}}_2$$

This auto-oxidation produces H₂O₂. Consequently, the formula for the H₂O₂ and Hb²⁺O₂ generated is as follows:

$${\mathrm{H}}_2{\mathrm{O}}_2+{\mathrm{2Hb}}^{2+}{\mathrm{O}}_2\to {\mathrm{2Hb}}^{3+}{\mathrm{2O}}_2+{\mathrm{2OH}}^{-}$$

Then, the presence of O₂ and H₂O₂ results in a Haber-Weiss reaction ³⁷ :

$${\mathrm{O}}_2^{-}+{\mathrm{H}}_2{\mathrm{O}}_2\to \mathrm{OH}\cdot +{\mathrm{OH}}^{-}+{\mathrm{O}}_2$$

These reactions explain the phenomenon that occurred after 180 min of pump circulation. The high shear stress generated in the pump circulation harms erythrocytes and causes hemolysis. As a result of auto-oxidation of oxyhemoglobin that is released into the blood during hemolysis, O₂⁻ was ultimately generated and the amount of active oxygen increased. Previous research suggested that ROS generated in erythrocyte membranes by the auto-oxidation of Hb are present in optimal locations to react with proteins that generate lipid peroxides and thus modify membrane proteins. ³⁸ The reason for this arrangement of ROS was believed to be that proteins that generate lipid peroxides and modify membrane proteins affect membrane lipids and membrane structures, supporting the results of this study. ³ . In other words, we concluded that the level of oxidation of the cell membrane was accelerated by the increased level of plasma free hemoglobin caused by the shear force generated during acceleration of pump circulation, not by the shear force itself.

A limitation of this study is the large measurement error in change in amount of the lipid peroxide. The reason for the large error may be because the TBA method is time-dependent and therefore potentially includes the human error. In future work, the measurement of lipid peroxide with a highly accurate measurement method that is less time-dependent will better elucidate the hemolysis mechanism that can occur in MCS devices. In addition, our experiment was performed at room temperature without any controlling to mimic clinical condition. However, this study successfully provided new knowledge that the levels of erythrocyte membrane oxidation certainly increased after their exposure to the blood pump circuit. This knowledge may lead to the damage of elucidation of the blood cells in MCS devices and their senescence in vivo.

Conclusions

In this study, we exposed whole porcine blood to shear stress using a pump circuit and measured changes in hemolysis, erythrocyte deformability, and lipid peroxide levels before and after the pump circulation. After 180 min of pump circulation, erythrocytes showed elevated hemolysis, decreased erythrocyte deformability, and slightly increased lipid peroxide content. This phenomenon is presumed to be due to the auto-oxidation of oxyhemoglobin caused by shear stress-induced hemolysis, which increased the amount of lipid peroxides and thereby reduced erythrocyte deformability.