Multiscale approach to link red blood cell dynamics, shear viscosity, and ATP release

From Soft-Matter
Jump to: navigation, search
A. M. Forsyth, J. D. Wan, P. D. Owrutsky, M. Abkarian, and H. A. Stone

"Multiscale approach to link red blood cell dynamics, shear viscosity, and ATP release"

Proc. Natl. Acad. Sci. U.S.A. 108 (27), 10986-10991 (2011).

Entry by Meredith Duffy, AP225, Fall 2011

Keywords: rheology, shear thinning, viscosity, mechanotransduction, red blood cell


While it is known that red blood cells (RBCs) release ATP, and that ATP incites endothelial cells lining blood vessels to release nitric oxide which produces vasodilation, the cause and effect chain is not entirely elucidated. Moreover, RBC deformation and a reduction in ATP release have both been linked to several diseases, such as Type II Diabetes and Cystic Fibrosis, but again, causal relationships are not yet agreed upon. Utilizing cell membrane tracking, rheometry and bioluminescence assays, the authors take a multiscale approach to examining the interdependencies of bulk fluid viscosity, ATP release, and single-cell motion and deformation in flow.

Methods and Results


Viscosity and ATP Release

Human blood samples were diluted to 1% hematocrit with physiological salt solution (PSS) mixed with varying percentages of Dextran to produce three different viscosities (measured at 50/s) of suspending media. Assuming a constant viscosity of 12mPa inside the cells, cell-to-media viscosity ratios λ of 1.6, 3.8 and 11.1 were calculated. Using a cone-plate viscometer, the authors then sheared samples of the three RBC solutions over a range of physiologically relevant shear rates from 50 to 5000/s to determine a viscosity parameter η/η(0) at each shear rate (Figure 2B), where η is the effective viscosity of the RBC solution and η(0) is the viscosity of the suspending media at the same shear rate. For the plots below, shear stress σ is used in place of shear rate as the independent variable and is calculated as the product of shear rate and the effective viscosity η.

Immediately after performing a viscosity test on a sample, the authors mixed it with a luciferin/luciferase solution and used a photomultiplier tube to quantify the bioluminescence, I, of the solution, which is proportional to the concentration of extracellular ATP. To obtain normalized ATP release, they calculated [I-I(0)]/I(0) where I(0) is the luminescence I(0) of an unsheared sample (Figure 2A).

From these figures, the authors identify three regions of interest, bounded by the grey lines at σ = .4 Pa and σ = 3 Pa and labeled Regimes I to III from left to right. In Regime I, they assert, the viscosity parameter remains high and constant, meaning the RBC solution stays more viscous than its suspending media; in Regime II, it experiences a rapid drop, a well-known phenomenon of many complex solutions called shear thinning; and in Regime III it approaches unity, i.e. solution viscosity approaches the viscosity of the solvent. Moreover, ATP release remains more or less constant at about 3 times the unsheared rate of release during both Regimes I and II, then jumps for Regime III. This implies that flow on its own increases ATP release, but that ATP release is independent of shear stress until the stress exceeds 3 Pa, or the viscosity parameter approaches unity.

Single-Cell Dynamics

To track the motion of individual cells, RBC solutions were mixed with 1 µm carboxylate microspheres, which randomly attach to cellular membranes, and again subjected the cells to different shear stresses, this time in a microfluidic channel. The authors linked the three different regimes of viscosity and ATP release described above to four
types of cell dynamics: tumbling, tanktreading, swinging, and deformation (or stretching) (See Figure 1). Tumbling, an end-over-end rigid-body rotation in which the cell sweeps out a sphere shape, was most present in Regime I, experienced a sharp decline in Regime II, and was almost absent in Regime II. Over all regimes, more tumbling was present for the RBC solution with a higher cell-to-media viscosity ratio (Fig. 3B), indicating a dependence of cell dynamics on the magnitude of this viscosity contrast. Additionally, as shear stress increased, cell length remained almost unchanged (L/L(0)=1) for both Regimes I and II, then increased rapidly in Regime III (Fig. 3A), possibly indicating that cell deformation (mechanotransduction) is important in generating a greater release of ATP, as is generally accepted, but that contrary to previous research cell deformation is *not* the mechanism behind shear thinning. Rather, their observations indicate that shear thinning is a result of cells transitioning from tumbling to tanktreading, or rotating about their axis of symmetry. Each cell disturbs less surrounding fluid when tanktreading than when tumbling, decreasing the effective solution viscosity.

Rotational frequency of tanktreading cells was also recorded for cells at two of the different viscosity contrast ratios (λ=1.6 and 11.1), and it was determined that tanktreading frequency is dependent on the viscosity ratio of cell to media, supporting the notion that what cell dynamics are exhibited depends on this ratio. Indeed, the fourth type of cell motion, swinging (which is essentially wobbling during tanktreading), was observed for cells at the higher viscosity ratio but not at the lower one. Swinging is possibly explained by dissipation of elastic energy from the cytoskeletal network, and decreases with increasing shear.

ATP Release Mechanisms

To determine an explanation for the change in ATP release rate from Regime II to Regime III, two ATP signaling channels were selectively blocked, one at a time, and the RBC solutions were again subjected to varying shear. First, the hemichannel Pannexin 1, or Px1, was blocked with carbenoxolone; this reduced the ATP release to unsheared levels for the entire range of shear stress (Figure 5, crosshatched bars), indicating that Px1 may be responsible for most flow-dependent (Regimes I and II) and deformation-dependent (Regime III) ATP release. Then, inhibition of the cystic fibrosis transmembrane conductance regulator (CFTR) was accomplished with glibenclamide. (CFTR is not believed itself to release ATP but has been shown before to increase ATP release, presumably by activating another pathway). This did not affect ATP release for Regimes I or II, but severely decreased release at the high shear stresses of Regime III (Figure 5, black bars), implicating the CFTR in deformation-dependent ATP release from RBCs. Thus it is suggested that cell deformation activates CFTR, which in turn may upregulate Px1.


Single-cell dynamics, notably the transition from tumbling to tanktreading, and the viscosity contrast between cell and media were both linked to bulk changes in blood viscosity as a function of shear stress. In addition, a possible mechanism based on mechanotransduction and an interaction between channels was introduced to explain the constant (but higher compared to non-shear) ATP release rate at low shears and rapidly increasing release rate at high shears. These findings add clarity to contradictions between earlier studies, and lead the way to a better understanding of the complex rheology and signalling pathways of blood.