Sickling kinetics drive genotype-specific impairment of RBC deformability in sickle cell disease Free

Introduction: Sickle cell disease (SCD) is an inherited hemoglobinopathy arising from a single point mutation in the β-globin gene that results in sickle hemoglobin (HbS). Under hypoxic conditions, HbS undergoes polymerization within red blood cells (RBCs), distorting them into rigid, sickled shapes. Repeated cycles of deoxygenation and reoxygenation subject RBCs to cumulative mechanical and oxidative stress, progressively impairing membrane and cytoskeletal integrity, reducing deformability, increasing hemolysis, and promoting microvascular occlusion. Although prior studies have reported that RBCs' past mechanical stress influence future sickling behavior, the relationships between dynamic sickling/unsickling kinetics, RBC mechanical properties, and occlusive potential remain unexplored. Here, we developed a microfluidic assay to assess the sickling/unsickling dynamics across different SCD genotypes, homozygous HbSS (SCD) and heterozygous HbSC (hemoglobin S-C disease), and examined their association with RBC deformability metrics. Additionally, we investigated the relationships between hematological parameters and biophysical measurements.

Methods: Sickling microfluidic chips were fabricated using soft-lithography, assembled, and coated with 1% poly-D-lysine to immobilize RBCs while preserving their native mechanics. Venous blood was collected in EDTA-coated tubes from subjects with HbSS (n = 10), HbSC (n = 5), and healthy HbAA controls (n = 3). Washed RBCs were standardized to 0.5% hematocrit in phosphate-buffered saline and perfused at 50 mbar through microchannels. Hypoxia was induced by switching perfusate gas to 5% CO₂/95% N₂. For each sample, approximately 80–100 individual cells were captured and tracked during a 20-minute time-lapse microscopy session at two frames per second to monitor shape changes. Using a neural network algorithm, we classified sickled versus non-sickled morphologies, yielding time to 50% sickling (T₅₀), recovery times, and maximum reversible sickling cycles. RBC deformability was quantified by the Laser Optical Rotational Red Cell Analyzer (LORRCA, Oxygenscan mode) to derive elongation index (ΔEI), and single-cell occlusion propensity was measured via our previously described hypoxia OcclusionChip assay.

Results: HbSS RBCs sickled significantly faster than HbSC RBCs (T₅₀ = 10.7 ± 0.9 min vs. 15.7 ± 0.9 min; p < 0.001); HbAA RBCs did not sickle. Successive hypoxia–reoxygenation cycles progressively decreased in T₅₀ (mean 20% reduction by cycle six; p < 0.01) and prolongation of recovery time (mean 35% increase; p < 0.01) in both SCD genotypes, indicating cumulative mechanical fatigue. HbSS samples exhibited up to six reversible cycles; HbSC up to five. ΔEI correlated inversely with cycle count (r = –0.81, p < 0.001), demonstrating progressive loss of deformability. Occlusion index likewise declined with cycle number (r = –0.64, p < 0.001). Importantly, shorter initial T₅₀ predicted higher occlusion indices (r = –0.79, p < 0.001), linking rapid early sickling to greater microvascular retention.

Discussion and Conclusion: Our data demonstrate that dynamic sickling behavior is closely linked to the progressive deterioration of RBC mechanical function. The observed “biomechanical memory,” where earlier stress accelerates future sickling and impaired recovery, underscores a feed-forward mechanism driving microvascular obstruction. Genotype-specific differences, notably between HbSS and HbSC, highlight distinct mechanical thresholds that may underlie variable clinical severities and treatment responsiveness. Mechanistically, repeated polymerization–depolymerization cycles likely destabilize membrane–cytoskeleton interactions and disrupt ion homeostasis, exacerbating rigidity. These insights suggest that therapies targeting both HbS polymerization kinetics and membrane stabilization could synergistically preserve RBC deformability. Furthermore, microfluidic sickling assays may serve as sensitive biomarkers for patient stratification and therapeutic monitoring. Implementation of this platform could accelerate preclinical drug screening and support personalized assessment of emerging interventions in SCD. Moreover, combining this assay with clinical metrics such as hemolytic markers and vaso-occlusive crisis frequency may enable comprehensive risk models, informing both prognosis and targeted drug development pipelines.