Sensing of red blood cells with decreased membrane deformability by the human spleen

Abstract

The current paradigm in the pathogenesis of several hemolytic red blood cell disorders is that reduced cellular deformability is a key determinant of splenic sequestration of affected red cells. Three distinct features regulate cellular deformability: membrane deformability, surface area-to-volume ratio (cell sphericity), and cytoplasmic viscosity. By perfusing normal human spleens ex vivo, we had previously showed that red cells with increased sphericity are rapidly sequestered by the spleen. Here, we assessed the retention kinetics of red cells with decreased membrane deformability but without marked shape changes. A controlled decrease in membrane deformability (increased membrane rigidity) was induced by treating normal red cells with increasing concentrations of diamide. Following perfusion, diamide-treated red blood cells (RBCs) were rapidly retained in the spleen with a mean clearance half-time of 5.9 minutes (range, 4.0-13.0). Splenic clearance correlated positively with increased membrane rigidity (r = 0.93; P < .0001). To determine to what extent this increased retention was related to mechanical blockade in the spleen, diamide-treated red cells were filtered through microsphere layers that mimic the mechanical sensing of red cells by the spleen. Diamide-treated red cells were retained in the microsphilters (median, 7.5%; range, 0%-38.6%), although to a lesser extent compared with the spleen (median, 44.1%; range, 7.3%-64.0%; P < .0001). Taken together, these results have implications for understanding the sensitivity of the human spleen to sequester red cells with altered cellular deformability due to various cellular alterations and for explaining clinical heterogeneity of RBC membrane disorders.

Graphical abstract

Figure 1.

Dimensional and morphological characteristics of dRBCs. Projected surface area (A), diameter (B), circularity (C), aspect ratio (D), and shape ratio (E) (sphericity) of untreated RBCs and RBCs treated with increasing concentrations of diamide (10-500 µM) and analyzed with an ImageStream imaging cytometer. RBCs treated with LPC to induce a control loss of cell surface area were used as a positive control for imaging. Unlike LPC treatment, exposure to diamide had no effect on RBC dimensions (surface area and diameter [A-B]) and only a slight effect on cell morphology (circularity [C] and sphericity [D-E]).

Figure 2.

Effect of diamide exposure on RBC membrane protein–protein associations and deformability. (A) Sodium dodecyl sulfate–polyacrylamide gel electrophoresis analysis of dRBCs, run in nonreducing (A1) or reducing (A2) conditions. (B) Deformability profile of dRBCs or untreated control RBCs before the spleen perfusion (shown is a representative experiment, n = 8), as assessed by Lorrca. (C) EI (RBC deformability parameter) is plotted as a function of the logarithm of the shear stress. dRBCs required significantly greater applied shear stress to reach equivalent deformation as RBCs with normal membrane. Because the lines are parallel, one can see that RBC membranes treated with 10 µM diamide required 1.3-fold greater shear stress than normal membranes to reach equivalent deformation at all points along the curve, indicating that the diamide-treated membranes had 0.79 times the normal deformability. Treatment with 125, 300, and 500 mM diamide resulted in membranes that had 0.29-, 0.14-, and 0.07-fold the normal deformability, respectively. (D) Linear regression fit of the correlation between diamide concentrations and the relative rigidity of dRBCs. (E) Osmotic fragility of dRBCs (a representative experiment).

Figure 3.

Retention of dRBCs by the isolated-perfused human spleen and the microsphiltration system. (A) Kinetics of dRBC clearance in the perfused spleen, monitored by measuring PKH-labeled counts in the perfusate (a representative experiment, n = 8). (B) Linear regression fit of the correlation between levels of dRBC retention within the spleen and the relative rigidity. (C) Retention rates of dRBCs in the isolated-perfused human spleen 2 hours after the onset of perfusion or in the microspheres filter. (D) Linear regression fit of the correlation between levels of dRBC retention within the microsphilters and the relative rigidity.