Background Erythrocyte aggregation is a phenomenon that is commonly found in several pathological disease states: stroke, myocardial infarction, thermal burn injury, and COVID‐19. Erythrocyte aggregation is characterized by rouleaux, closely packed stacks of cells, forming three‐dimensional structures. Healthy blood flow monodisperses the red blood cells (RBCs) throughout the vasculature; however, in select pathological conditions, involving hyperthermia and hypoxemia, rouleaux formation remains and results in occlusion of microvessels with decreased perfusion. Objectives Our objective is to address the kinetics of rouleaux formation with sudden cessation of flow in variable temperature and oxygen conditions. Methods RBCs used in this in vitro system were obtained from healthy human donors. Using a vertical stop‐flow system aligned with a microscope, images were acquired and analyzed for increased variation in grayscale to indicate increased aggregation. The onset of aggregation after sudden cessation of flow was determined at proscribed temperatures (37–49°C) and oxygen (0%, 10%), and in the presence and absence of 4, 4′‐Diisothiocyano‐2,2′‐stilbenedisulfonic acid (DIDS). Both autologous and homologous plasma were tested. Results RBCs in autologous plasma aggregate faster and with a higher magnitude with both hyperthermia and hypoxemia. Preventing deoxyhemoglobin from binding to band 3 with DIDS (dissociates the cytoskeleton from the membrane) fully blocks aggregation. Further, RBC aggregation magnitude is greater in autologous plasma. Conclusions We show that the C‐terminal domain of band 3 plays a pivotal role in RBC aggregation. Further, aggregation is enhanced by hyperthermia and hypoxemia.

Our purpose was to identify the kinetics of erythrocyte (RBC) reversible aggregation (Agg) during stop flow (SF) experiments as a function of temperature and O2 for RBC from healthy subjects self‐identified as Black (B, n=8), Hispanic (H, 4) or Caucasian (C, 4) (48+/‐ 9 y). Whole blood (EDTA, BioIVT, Westbury, NY), vendor verified to be viral free, and used within 1 week of donation, was centrifuged and the platelet poor plasma (PPP) was recovered. All RBCs were resuspended in autologous PPP, and incubated at specified temperatures (37, 41, 45, 49 C) and O2 (0, 5, 10%, saturating). Flow was initiated in a microchannel system at 0.1‐0.2 Re, then stopped while recording the entire microchannel width. The greyscale range of the images showed a significant Fahraeus effect with flow (cell free layer at wall). With SF, as Agg proceeded, the aggregates accumulated within the centerline of the microchannel, increasing the variability of the greyscale range from white at the microchannel wall to black in the centerline. The change in variability indicated kinetics of Agg that occurred over 3 min of recording. The primary data of greyscale vs time (custom software) was fit to a sigmoid (dose response) curve, where the fitted EC50 was the T1/2 for Agg and with the fitted slope (p) indicated the kinetics of Agg. At 37 C the T1/2 was 93 s in 0% O2, 75 s with venous (5%) or arterial (10%) and was not different by group. As temperature increased, the T1/2 became significantly faster in blood from B or H (to 65 s), but not C. Correlation between temperature and the fitted T1/2 were significant in B (slope, R2; ‐2.9, 0.92), H (‐2.7, 0.97) and C (‐1.1, 0.6), but the F‐test on the slope was significant for B (p=0.02) and H (0.04) not C (0.6). This coincided with a significantly steeper p for B and H. At venous, arterial or saturating O2, the effect of increasing temperature to decrease the T1/2 was much less pronounced, and not different between groups. This data shows that in only hypoxic conditions, with temperatures of 41 C (105.8 F) or higher, RBC from self‐identified B or H subjects aggregates more readily than C. These results pertain to our focus of microvascular compromise in thermal burn injury progression, but also clinical conditions of localized cautery to control bleeding, high fevers or malignant hyperthermia.

No abstract

Erythrocyte (RBC) aggregation in blood or in plasma is a reversible phenomenon with an unknown mechanism. However there is evidence that it is induced, at least in part, by fibrinogen. RBC aggregation leads to vascular occlusion or flow diversion and shunting of oxygenated blood. If fibrinogen/fibrin ultimately becomes the major occlusive moiety, as one of us has observed in peri‐burn tissue at 48h post‐burn, the occlusion is likely to be irreversible. Previous studies exploring fibrinogen induced aggregation used Millipore fibrinogen. However, based on our preliminary observation of fibrinogen‐induced aggregation, we propose that the same concentration of fibrinogen across different company brands will induce aggregation of variable extent. Millipore fibrinogen was compared with isolated soluble fibrin‐depleted fibrinogen (FgD) and with soluble fibrin‐rich fibrinogen (FgR) (human). Using a spectrophotometer (Bio‐Rad SmartSpec 3000), focused on wavelengths of 540, 560, and 580nm, corresponding to a peak in oxy, deoxy, and oxygenated hemoglobin state in the RBCs, each fibrinogen isolated was tested at varying concentrations (1, 2, and 4mg/mL) and temperatures (32, 37, and 45°C) with washed human RBC. Across all temperatures and wavelengths, FgR fibrinogen dissolved in phosphate buffered saline (PBS) had a higher percent transmission, which correlates to an increased RBC aggregation, than both Millipore fibrinogen and FgD fibrinogen. Additionally, FgR induced the most aggregation at the highest tested concentration, 4mg/mL. Using this increased aggregation at 4mg/mL of FgR, we then compared to PBS and autologous platelet poor plasma. Across all temperatures, the magnitude of RBC aggregation is higher in plasma than FgR, which induces higher aggregation than PBS alone. The results demonstrated that fibrinogen‐induced RBC aggregation varies with the levels of its soluble fibrin/fibrinogen complex. By extension, these complexes clearly enhance RBC aggregation, intimating an in vivo role particularly in pathologic conditions that increase their levels.

Thermal burn injury is associated with an increase in erythrocyte (RBC) aggregation (rouleaux) prior to fibrin clot formation; unresolved rouleaux occlude vessels and contribute to thermal burn injury progression (Clark, 2013). RBC aggregation is reversible, yet the mechanism is unknown. Current models do not include all biophysical phenomena, including the role of the cytoskeleton, and sudden oxygen release from hemoglobin. We previously proposed that thermally induced membrane deformation creates nucleation sites for excess oxygen to coalesce, i.e., form nanobubbles, which may contribute to surface tension induced aggregation. To begin to address the role of the cytoskeleton in membrane deformation through changes in the band 3‐ankyrin/adducin bridge and spectrin‐actin interactions, we have disrupted actin polymerization with cytochalasin D (cytoD). We hypothesized that disrupting the cytoskeleton would alter the apparent presence of oxygen nanobubbles on singlet RBCs. Our experimental model examines both size and greyscale appearance of RBCs with sudden changes in temperature. Temperatures of 41–49 C cause shifts in brighter greyscale and larger apparent RBC size, consistent with light diffraction through concentrated gas bubbles that dissipates with time or return to 37 C (Blasco, 2017). Washed murine RBCs (obtained through a tissue sharing program) were suspended in saline (PBS) containing 1% EtOH, 0.5 uM CytoD (in 0.1% EtOH) or control, and incubated at 37°C prior to use. RBCs settled to a monolayer within an enclosed microchannel at 37°C (heated stage). The stage temperature rapidly heated from 37°C to test temperature (37, 39, 41, 43, 45, 47, or 49°C) and we imaged at 2 Hz for 70 seconds (60X, SWI). Singlet RBCs were followed over time, measuring greyscale and apparent size of the cell (outer rim) and core (central biconcave) (ImageJ). The initial changes were seen during the temperature shift and included increased core greyscale and apparent size of 20–40% between 41–47°C, with corresponding changes in cell size. Over the next several seconds there was temperature dependent fluctuations in size and greyscale of up to 50%. Cells in EtOH were similar overall to controls. CytoD prevented changes in greyscale and in size, and suppressed further fluctuations until higher temperatures where those treated cells had formed echinocytes. The data are consistent with gas bubbles forming at the RBC membrane, causing diffraction of the light, in a temperature dependent fashion. As destabilization of the cytoskeleton prevented these changes, we conclude that the cytoskeleton contributes to membrane shape change necessary for oxygen nanobubble formation with heat.This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.