Abstract
Sickle Cell Disease (SCD) is an inherited red cell disorder caused by the presence of the abnormal hemoglobin S (HbS) resulting from a single base substitution in the beta hemoglobin gene. SCD affects approximately 8 million people globally including 100,000 in the United States. HbS polymerizes under low oxygen conditions in the tissues to stiffen and distort red cells (sickling), that underly the 2 key features of the disease – recurrent acute vaso-occlusive crises and hemolytic anemia. The manifestations of SCD are multisystemic with progressive organ damage and premature death. Gene therapy and allogeneic hemopoietic cell transplantion offer potential cures but these are currently only available to few patients, even in well-resourced countries. Hydroxyurea remains the only proven effective pharmacological treatment. There is a huge unmet need to develop more oral drugs accessible to patients worldwide. Since the formation of abnormal HbS fibers initiates the pathology of SCD, understanding how they form and how they impact red cell membrane at high resolution is vital, as this could provide insights in new therapeutic strategies. Extensive research has been conducted ex vivo to solve the structure of the HbS polymer with only partial success. High resolution in situ data offers a direct physiological view on HbS polymers at near atomic resolution, however, it bears a number of technical challenges.
Here, we established a cryogenic electron tomography (cryo-ET) pipeline to visualize, for the first time, the intracellular molecular organization of sickled red blood cells (RBCs) from patients with sickle cell disease (SCD; HbSS, HbF <10%) in situ. Using a custom-designed chamber for controlled deoxygenation, we induced sickling in leukodepleted RBCs and vitrified them under fully deoxygenated conditions, preserving their native ultrastructure. Cryo-focused ion beam milling coupled with scanning electron microscopy (cryo-FIB-SEM) was employed to prepare 100–150 nm-thick lamellae from intact vitrified sickled RBCs, maintaining spatial context and avoiding structural disruption. Targeted tomographic data collection was performed on a high-end Titan Krios transmission electron microscope under cryogenic conditions. 3D reconstructions revealed densely bundled HbS polymers, 20 nm in width and tightly aligned within the cytoplasm. The cytoplasm is generally devoid of organelles, with an exception of mitochondria, which we validated to be a mixture of functional and dysfunctional forms. Abnormal mitochondria retention in mature RBCs has also been reported in other haemolytic anemias. We then segmented HbS polymers to train a deep learning model (crYOLO) for automated particle detection, facilitating the extraction of ~20,000 subvolumes. Using subtomogram averaging approaches we obtained a preliminary reconstruction at 20 Å resolution that validates key features learnt from in vitro and in silico models. Ongoing efforts aim to enhance resolution and reveal the molecular architecture of HbS polymers within their native cellular environment to provide mechanistic insights into the pathological transformation underlying sickle cell disease. Efforts are also underway to characterize molecular architecture of the red cell cytoskeleton, retained mitochondria in sickle red cells and how small molecule interventional drugs disrupts formation of HbS fiber formation and impacts distortion of the red cell membrane.
