Biomechanics of phagocytosis of red blood cells by macrophages in the human spleen

Abstract

The clearance of senescent and altered red blood cells (RBCs) in the red pulp of the human spleen involves sequential processes of prefiltration, filtration, and postfiltration. While prior work has elucidated the mechanisms underlying the first two processes, biomechanical processes driving the postfiltration phagocytosis of RBCs retained at interendothelial slits (IES) are still poorly understood. We present here a unique computational model of macrophages to study the role of cell biomechanics in modulating the kinetics of phagocytosis of aged and diseased RBCs retained in the spleen. After validating the macrophage model using in vitro phagocytosis experiments, we employ it to probe the mechanisms underlying the kinetics of phagocytosis of mechanically altered RBCs, such as heated RBCs and abnormal RBCs in hereditary spherocytosis (HS) and sickle cell disease (SCD). Our simulations show pronounced deformation of the flexible and healthy RBCs in contrast to minimal shape changes in altered RBCs. Simulations also show that less deformable RBCs are engulfed faster and at lower adhesive strength than flexible RBCs, consistent with our experimental measurements. This efficient sensing and engulfment by macrophages of stiff RBCs retained at IES are expected to temper splenic congestion, a common pathogenic process in malaria, HS, and SCD. Altogether, our combined computational and in vitro experimental studies suggest that mechanical alterations of retained RBCs may suffice to enhance their phagocytosis, thereby adapting the kinetics of their elimination to the kinetics of their mechanical retention, an equilibrium essential for adequately cleaning the splenic filter to preserve its function.

Keywords: biomechanics; erythrophagocytosis; hypersplenism; macrophages; splenic sequestration.

Fig. 1.

(A) Sequential snapshots of the engulfment of a microbead with a diameter of 5 $\mathrm{\mu }$m by a human-derived THP-1 macrophage. (B) Schematics adapted from ref. illustrating the key events involved in particle uptake through phagocytosis: (B, I) binding of macrophage surface receptors to ligands on the surface of phagocytic targets; (B, II) generation of membrane protrusion and formation of phagocytic cup; (B, III) wrapping of the targets; (B, IV) complete enveloping of the target. (C) A computational model is developed to simulate the engulfment of microbeads or RBCs by macrophages through a sequence of processes analogous to (A) and (B). The green region in (C, II) signifies the activated portion of the macrophage membrane that can stretch out toward the target. We model this process to be driven by the effective protrusion forces, mimicking the actin-induced protrusion force. The red region signifies the membrane of the macrophage in contact with the target cells, where the membrane adheres to the target. In (C, III–IV), the particles constituting the activated region (green) are presented in a smaller size to better illustrate the microbead wrapped by the macrophage.

Fig. 2.

Comparison of the mechanical properties of macrophages determined by micropipette aspiration experiments and simulations. (A) Experimental (Left) and computational (Right) setup for micropipette aspiration of a macrophage in suspension. (B) Experimental (Left) and computational (Right) setup for micropipette aspiration of a macrophage attached to the substrate. The inner radius of the pipette (R) is 1 $\mathrm{\mu }$m in (A) and (B). (C) The aspirated length (L) of macrophages normalized by R is plotted as a function of aspiration pressure for both experimental measurements and computational results. The mean and SD of the experimental results are calculated based on measurements of 10 macrophages. The simulation results of the macrophages attached to the substrate are computed for five different values of adhesive strength between the macrophage and substrates: 105, 120, 135, 150, and 165 pN/$\mathrm{\mu }$m².

Fig. 3.

Phagocytosis of spherical polystyrene microbeads by macrophages. Simulations of a macrophage phagocytoing a spherical microbead with a diameter of 5 $\mathrm{\mu }$m (A) mimic the characteristics associated with the engulfment of microbeads evident from in vitro phagocytosis assays (B). Similar simulations (C) and experiments (D) are performed for a spherical microbead with a diameter of 1$\mathrm{\mu }$m. In the images shown in (A and C, III, and IV), a portion of the macrophage membrane is not shown to enable visualization of microbeads wrapped by the internalized membrane (red). Videos of all the results presented in (A–D) can be found in Movies S1–S4. (E) The effects of macrophage membrane stiffness and the diameter of the target sphere on the critical adhesive strength for full engulfment are summarized as a mechanism map. The brown zone represents states where the simulations show partial engulfment of microparticles because of insufficient membrane surface area. The olive color zone signifies states where simulations illustrate full engulfment of the microparticles, whereas the yellow zone denotes partial engulfment because of insufficient adhesive strength.

Fig. 4.

In vitro phagocytosis of biomechanically altered RBCs. (A and B) Two examples of phagocytosis of stiff sickle RBCs by macrophages are presented. (B) is adopted from ref. . (C and D) Two examples of phagocytosis of flexible sickle RBCs by macrophages are displayed. While only slight morphological changes are observed for granular-shaped sickle RBCs in (A) and (B), flexible RBCs are seen to undergo severe deformation during engulfment in (C) and (D). Compared to (A–C), where the uptake of RBCs is complete, the RBC in (D) is only partially engulfed within a two-hour observation period. (E and F) are examples of the uptake process of HS and heated RBCs, respectively. Videos for the results presented in (A) and (C–F) can be found in Movies S5–S9.

Fig. 5.

Effect of RBC stiffness on the kinetics of phagocytosis illustrated by simulation results in (A–D) and combined experimental and simulation results in (E). Sequential images of macrophage are shown as they engulf sickle RBCs with stiffness values equivalent to (A) healthy RBCs, (B) 10 times stiffer, and (C) 20 times stiffer than healthy RBCs. In (A–C, IV, and V), a portion of the macrophage membrane is not plotted to enable visualization of internalized RBCs. Videos of the results presented in (A–C) can be found in Movies S10–S12. Variations of the critical adhesive strength for full engulfment (D) and the phagocytic time (E) as functions of RBC stiffness are plotted. Sickle RBC models with biconcave, crescent, and granular shapes are examined. Color bars starting from the Right side of (D) highlight the reported ranges of shear modulus values for sickle RBCs (HbSS) and HE, as well as RBCs invaded by P. falciparum malaria parasite representing intraerythrocytic parasite maturation stages “ring” (less than 24 h of parasite invasion inside the RBC), “trophozoite” (24 to 36 h), and “schizont” (36 to 48 h) (55, 60, 64). In (E), data measured from phagocytosis experiments of sickled, nonsickled, heated, and HS RBCs are plotted as a bar graph for comparison. The width of the bars in (E) signifies the SD of the measured RBC stiffness. The adhesive strength of 180 $pN/\mathrm{\mu }$m² is applied in all the simulations to ensure full engulfment of the RBCs. The mean and SD of experimental phagocytic time in (E) are calculated based on 10 phagocytosis experiments. The simulation results are computed based on five macrophage models with stiffness values of 0.1, 0.2, 1, 5, and 10 times that of healthy RBCs (4.73 $\mathrm{\mu }$N/m). The experimental and simulated phagocytic times in (E) are normalized by their mean values. Part of the data plotted in (E) was adapted from ref. .

Fig. 6.

Cooperative processes involving the IES and macrophages in retaining and eliminating abnormal RBCs. (A) Progressive shape changes of RBCs as the S/V is reduced from 1.44 (healthy RBCs) to 0.93. (B) Variations in critical adhesive strength for full engulfment of RBCs for different values of S/V and membrane stiffness by macrophages. The dotted vertical lines in (B, III) highlight the two critical S/V reported from refs. (brown) and (purple) that determine the passage or retention of RBCs by splenic IES, as illustrated by the schematics in (B, I) and (B, II), respectively.

Fig. 7.

Effect of RBC morphology on the kinetics of phagocytosis. Simulations of engulfment of relatively undeformable sickle RBCs with different shapes: granular (A), biconcave (B), crescent (C), and elongated (D). In (A–D, III and IV), particles that constitute the activated region (green) are presented in a smaller size to better illustrate the sickle cell wrapped by the internalized membrane (red). Videos for the results presented in (A–D) can be found in Movies S13–S16. (E) Schematic illustration of a macrophage as it engulfs sickle RBCs of different morphologies. Two different patterns are observed depending on the adhesive strength between the macrophage and sickle RBCs, as well as the cell shape. The orange zone represents simulations of full engulfment of the sickle RBCs, whereas the blue zone denotes the simulations of partial engulfment. The red biconcave RBC represents RBCs with normal stiffness, while the purple biconcave RBC represents undeformable RBCs. (F) Predicted variation of phagocytic time as a function of adhesive strength for sickle RBCs of different morphologies. The mean and SD of the simulation results in (F) are computed based on five macrophage models with stiffness values of 0.1, 0.2, 1, 5, and 10 times that of the control value. The values are normalized using the phagocytic time of spherical microbeads with a diameter of 5 $\mathrm{\mu }$m.