Multiscale computational framework for predicting viscoelasticity of red blood cells in aging and mechanical fatigue

Abstract

Red blood cells (RBCs) experience significant cyclic deformation through large elastic stretching and relaxation as they circulate in the bloodstream. Such hundreds of thousands times cell deformation causes cumulative fatigue damage to the circulating RBCs, leading to alterations in cell deformability and membrane viscoelasticity. Here, we employ a two-step multiscale computational framework based on coarse-grained molecular dynamics (CGMD) and dissipative particle dynamics (DPD) to probe the viscoelastic properties of the surface-altered RBCs in aging and mechanical fatigue, using experimental information on the molecular structures of both RBC membrane and cytosolic Hb as well as Hb concentration of RBC cytosol. We perform CGMD simulations to compute the shear modulus and membrane viscosity of a small RBC patch with altered horizontal connectivity within spectrin network; meanwhile, we compute the shear viscosity of RBC cytosol under different Hb concentrations. We then use these computed parameters as inputs for a more coarse-grained DPD-based RBC model at the whole-cell level to probe cell deformation, dynamic relaxation, viscoelastic performance of the altered RBC membrane. We find that the RBC membrane rigidity increases with the enhanced horizontal connectivity within spectrin network, but by a factor less than the effective membrane viscosity, resulting in an elevated characteristic relaxation time with increasing fatigue cycles. In addition, we quantify the two-dimensional storage and loss moduli of RBC membrane under time-dependent loading, revealing that the loss modulus is much larger than the storage modulus throughout the entire frequency range covered by the simulations, especially under enhanced connectivity in the cytoskeletal network, which is analogous to the performance of typical viscoelastic materials. These quantitative findings provide unique insights into the progressive damage of circulating RBCs, and demonstrate the capability of the two-step multiscale framework in effectively predicting the altered viscoelastic properties of RBCs influenced by in vivo aging and mechanical fatigue.