Publications

When deformable particles (e.g., drops or living cells) are perfused through a flow channel, they drift into a specific lateral position that depends on their size and mechanical properties. This characteristic can be used for deformability-based particle sorting. Using a fully three-dimensional algorithm for viscoelastic drop dynamics, we study numerically the effects of particle size, bulk shear viscosity and elasticity, interfacial (or cortical) tension, and fluid inertia on lateral migration and deformation of small liquid drops and leukocytes (white blood cells) in a rectangular microfluidic flow chamber. Our numerical data show that there is an almost linear increase in the lateral equilibrium position of liquid drops or leukocytes with the particle diameter-to-channel height ratio increased from 0.1 to 0.5. Excluding the case of drops with high interfacial tension, an increase in bulk viscosity of these particles leads to a substantial decrease in their equilibrium position. Overall, the results of this work indicate that 1) drops with different bulk viscosities can be efficiently separated in a rectangular microchannel if their interfacial tension is low or the flow rate is sufficiently high; and 2) the microfluidic technology is well suited for the separation of leukocytes with different cytoplasmic viscosities and relaxation times, but it is much less sensitive to cortical tension. This investigation opens up the possibility of using microfluidic systems for deformability-based flow cytometry.

Rolling leukocytes deform and show a large area of contact with endothelium under physiological flow conditions. We studied the effect of cytoplasmic viscosity on leukocyte rolling using our novel three-dimensional numerical algorithm that treats leukocyte as a compound droplet in which the core phase (nucleus) and the shell phase (cytoplasm) are viscoelastic fluids. The algorithm includes the mechanical properties of the cell cortex by cortical tension and considers leukocyte microvilli that deform viscoelastically and form viscous tethers at supercritical force. Stochastic binding kinetics describes binding of adhesion molecules. The leukocyte cytoplasmic viscosity plays a critical role in leukocyte rolling on an adhesive substrate. High-viscosity cells are characterized by high mean rolling velocities, increased temporal fluctuations in the instantaneous velocity, and a high probability for detachment from the substrate. A decrease in the rolling velocity, drag and torque with the formation of a large, flat contact area in low-viscosity cells leads to a dramatic decrease in the bond force and stable rolling. Using values of viscosity consistent with step aspiration studies of human neutrophils (5 to 30 Pa·s), our computational model predicts the velocities and shape changes of rolling leukocytes as observed in vitro and in vivo.

In our previous report, we showed that a rheometer equipped with a double concentric cylinder geometry with slotted rotor could effectively reduce wall slip effects and thus it could be used as an alternative to a rheometer with a vane geometry in yield stress measurements. Here, we use three-dimensional CFD simulation to compare these two geometries for rheological measurements of power law and yield stress fluids. Our results indicate that the double concentric cylinder rheometer with slotted rotor (DCCR/SR) is able to accurately measure rheological properties of a wider spectrum of test fluids than a vane rheometer because of significant reduction of the end and secondary flow effects.

Fluid shear stress has been implicated as a regulator of sprouting angiogenesis. However, whether endothelial cells within capillary sprouts in vivo experience physiologically relevant shear stresses remains unclear. The objective of our study is to estimate the shear stress distribution along the length of a capillary sprout through computational modeling of blood flow in a blind ended channel branching off a host vessel. In this model, we use sprout geometries typical for rat mesenteric microvasculature and consider three types of boundary conditions: 1) non-permeable vessel wall, 2) uniformly permeable vessel wall, and 3) a non-permeable vessel wall with open slots (representative of endothelial clefts). Our numerical simulation predicts that for each boundary condition a local maximum shear stress (13.9, 8.9, and 13.3 dyne/cm2 respectively) occurs at the entrance of a 50 um long, 6 um diameter sprout branching at 90 degrees off of a 11 um diameter host vessel. The shear stress drops below 0.2 dyne/cm2, a threshold for endothelial cell activation, within 4.1 um of the entrance for the non-permeable wall case and 4.2 um for the uniformly permeable wall case. Shear stress magnitudes within the sprout are above 0.2 dyne/cm2 for longer sprout scenarios and peaked at 5.9 dyne/cm2 at endothelial cell clefts. These results provide a first estimate of relative fluid shear stress magnitudes along a capillary sprout and highlight the importance of investigating endothelial cell responses to flow conditions during angiogenesis in tumors and other altered microenvironments.

We present a Navier-Stokes/Oldroyd-B immersed boundary algorithm that captures the interaction of a flexible structure with a viscoelastic fluid. In particular, we study the effects of bulk viscoelasticity on freely decaying shape oscillations of an Oldroyd-B fluid droplet suspended in an Oldroyd-B matrix. Our numerical data indicate that if the fluid viscosity is low, viscoelasticity plays a modulating role in the drop shape relaxation; specifically, it increases the oscillation frequency and decreases the decay rate when the fluid relaxation time is above a critical value. In the high-viscosity limit, i.e., when a Newtonian droplet is expected to return to a spherical shape with an aperiodic decay, an increase in the relaxation time eventually results in the reappearance of the oscillations. Both these results are in line with the prediction of small deformation theory for viscoelastic droplet oscillations. The algorithm was also validated by direct comparison with linear asymptotics.

Neutrophils roll by reversible binding and unbinding of P-selectin Glycoprotein Ligand-1 (PSGL-1) to P-selectin. The average lifetime of P-selectin—PSGL-1 bonds may increase with applied force until the force reaches a critical value of 11 pN. Here, we apply this "catch bond" behavior and the transition to slip bonds at supercritical forces to study their effect on neutrophil rolling. Leukocyte adhesion significantly depends on cellular deformability, viscoelastic extension of leukocyte microvilli, and membrane tether pulling. We use two custom computational models: ETMA in which the leukocyte is a rigid sphere and Visco-Elastic Cell Adhesion Model (VECAM) in which the leukocyte is a deformable particle. Both models incorporate microvillus viscoelasticity and tether pulling and describe receptor-ligand binding kinetics using a stochastic Monte Carlo approach, with on- and off-rates determined according to the spring model of Dembo. Bonds are characterized by bound state and transition state spring constants, with the latter being higher of the two in the case of catch bonds. Our simulations show that catch bonds lead to firm adhesion of a rigid cell at wall shear stress of 0.5 dyn/cm2, but the transition of these bonds to slip bonds induces stable rolling. Catch bonds favor tether pulling in a deformable cell that eventually results in cell detachment at this subthreshold shear stress. We conclude that catch bonds, tether dynamics and cell deformation all contribute to leukocyte rolling.

Cellular deformability may lead to lateral migration of cells toward the centerline during their perfusion in a microfluidic flow chamber. This property can be used for separation of cells of different deformabilities, such as, for example, red and white blood cells. Here, we study the effect of bulk viscoelasticity on lateral migration of cells and particles in shear flow using custom computational fluid dynamics code. The cells and particles are modeled as multiphase (a nucleus surrounded by a layer of cytoplasm) and single-phase viscoelastic drops, respectively. The numerical algorithm is based on the volume-of-fluid continuous-surface-force (VOF-CSF) method and the semi-implicit solvers for the Navier-Stokes equations and the Giesekus constitutive equation for bulk viscoelasticity. Our simulations show the cell/particle with larger deformability moves at a lower translational velocity than the fluid flow. At the same time, it migrates with a relatively constant velocity toward the channel centerline until reaching the equilibrium position. A more deformable cell is characterized by a higher lateral migration velocity, especially when its cytoplasmic viscosity drops to the value of less than 10 P.

Shear stress has been implicated as a modulator of angiogenesis. However, a full understanding of how shear stresses influence endothelial cell phenotype and function during angiogenesis requires identification of local stress distribution along capillary sprouts. The objective of this study was to investigate the influence of vessel permeability on shear stress distribution along a capillary sprout. Using Fluent, the shear stress was computed for Newtonian flow (1.2 cP viscosity) through a 2-D blind-ended vessel (5 ?m diameter, 200 ?m length) originating at a 90° angle off a host vessel (10 ?m diameter). Vessel walls were modeled as a solid wall 1) without or 2) with holes of 0.5 ?m in diameter spaced 5 ?m apart or 3) as a 0.3 ?m thick porous layer. The average inflow velocity was equal to 1 mm/s. For the last two models the hydraulic conductivity was matched to 0.018 ?m/(s mmHg). Independently of the model used, the entrance to the sprout experiences a local maximum in wall shear stress. The stress is reduced from the upstream value of 7 dyne/cm2 by four orders of magnitude within 8 ?m of the entrance in the solid wall model and by 2.5 orders in permeable models. For the solid wall with pores, the stress locally peaks at pore sites. Simulations for 45° & 135° sprout angles indicate similar stress magnitudes but with asymmetric distribution along upstream & downstream walls of the sprout. The presented data give insights into the shear stress distribution in capillary sprouts that can be linked to endothelial cell phenotypic changes during angiogenesis.

Coiling embolization in which platinum wires are inserted into the aneurysm through an artery is a common treatment in therapy of cerebral aneurysms. This procedure, however, may result in non-uniform and incomplete closure of the aneurysm, thereby reducing therapeutic effects. In this work, we propose to use a yield stress fluid material instead of platinum wires to close the aneurysm. This material behaves like solid when the applied shear stresses are less than the critical yield stress but it starts to flow when the shear stresses are larger than this critical value. Our objective is to investigate the viability of this approach using computational fluid dynamics. The giant aneurysm geometry used in our simulation was modeled from deceased patient’s CT images. The velocity and shear stress fields in the blood circulation surrounding the aneurysm were computed before and after embolization with a yield stress fluid using the incompressible and time-steady computational fluid dynamics solver. Our simulation results indicate that the shear stress distribution along the aneurysm surface becomes more uniform after filling it with the yield stress fluid. This reduces the risk of aneurysm rupture. We also determine the minimum yield stress at which the embolizing material will not leak out of the aneurysm. Overall, our analysis shows that yield stress fluids can be potentially used for treatment of cerebral aneurysms.

Histamine plays an important role in both normal physiology and various pathophysiological conditions. It upregulates the expression of inflammatory cytokines, including TNF-?, and endothelial cell adhesion molecules such as P-selectin and integrin ligands. These factors may trigger monocyte adhesion to arterial endothelium and subsequent accumulation of monocytes/macrophages in the arterial intima, leading to development of an atherosclerotic plaque. In this work, we target to investigate the effect of histamine on interactions of THP-1 (human acute monocytic leukemia cell line) with HUVEC monolayer in vitro. This analysis is conducted under flow conditions using the Bioflux 200 microfluidic shear flow system. According to our data, activation of HUVEC with histamine increases the rolling flux and decreases the rolling velocity of THP-1. Flow cytometric analysis indicates that P-selectin is the primary cell adhesion molecule involved in histamine-induced monocyte rolling. The histamine effect on cell rolling becomes more pronounced when it is used in combination with TNF-?. Histamine+TNF-? also lead to a significant increase in the number of firmly adherent monocytes. Parallel studies with OxLDL-stimulated endothelium show a less effect of OxLDL on monocyte rolling and adhesion than that of histamine+TNF-?. Overall, this study suggests that histamine may be an important regulator of atherogenesis.