Poster Presentation Multi-Omics Conference 2024

3D tissue scale automated segmentation with alpenglow 3Dai spatial analysis suite (#119)

Jasmine Wilson 1 , Paula Rotger Gonzalez 1 , Caleb Stoltzfus 1 , Alexandra Alvarsson 1
  1. Alpenglow Biosciences, Seattle, WA, United States

Neutrophils undergo rapid morphological changes during migration, activation and phagocytosis. Studies commonly assess these changes using in vitro cell culture environments, which do not capture the full complexity of cell shapes. Techniques, such as atomic force and confocal microscopy, have shown isolated neutrophils transition from spherical cells to disrupted states. Here, we use 3D imaging and analysis of whole tissue samples to characterise neutrophil morphology on a large scale in their native tissue environment. 

Human tonsil FFPE tissue was deparaffinized and processed before being stained with nuclear dye YO-PRO-1 and an anti-Neutrophil Elastase antibody. Samples were optically cleared using a modified iDISCO+ protocol. Samples were imaged with a hybrid open-top light-sheet microscope; the 3Di™. Neutrophils were quantified by 3D spatial analysis using 3Dai™ tools. Automated methods were used to segment neutrophil elastase labeled neutrophils and nuclei rapidly in 3D. We evaluated virtual standard histological sections and compared the variability in individual sections to the whole enumeration provided by the 3D imaging and analysis.  

Our analysis revealed cellular structures associated with neutrophil effector functions, including multivesicular bodies, single cells, and multicellular aggregates. The 295 segmented cells exhibited a range of shapes, from spherical to highly deformed. In 2D representations, multicellular aggregates and multinucleated cells appeared as singular entities. These findings underscore the necessity of 3D imaging for the quantification of the morphological diversity of neutrophils and their potential functional states.

We demonstrated that our pipeline can characterize and enumerate the morphological diversity of neutrophils in whole tissue samples, offering valuable insights into their effector states. This approach provides a significant advancement over conventional histology, which analyses cells in 2D. Moving forward, we plan to correlate features such as shape characteristics and spatial location with disease context. These efforts aim to refine techniques for large-scale, automated characterization of immune cell states.