Neutrophils actively swell to potentiate rapid migration

While the involvement of actin polymerization in membrane protrusion is well-established, we have a more limited understanding of the role of transmembrane water flow in cell motility. Here we investigate the role of water influx in neutrophil migration. These cells undergo directed movement to sites of injury and infection. Chemoattractant exposure increases cell volume and potentiates neutrophil migration, but the causal link between these processes is not known. Using a genome-wide CRISPR screen, we identify the regulators of the chemoattractant-induced neutrophil swelling, including NHE1, AE2, PI3K-gamma, and CA2. Through NHE1 inhibition in primary human neutrophils, we show that cell swelling is both necessary and sufficient for rapid migration following chemoattractant stimulation. Our data demonstrate that cell swelling complements cytoskeletal inputs for chemoattractant-induced potentiation of migration.

Methods
In these experiments, primary human neutrophils were imaged using Fluorescence Exclusion Microscopy (FxM). FxM is a highly accurate method for measuring cell volume that relies on cells excluding a dye in a microfluidic chip with a flat ceiling held up by pillars. The height of chamber is known and can be used to determine the volume displaced by the cells. The FxM image is taken with a low magnification objective (in this case a 20x 0.75NA Plan Apo) with a depth of field larger than the height of the chamber to capture all of the fluorescence in the chip. For all experiments in this dataset, the excluded dye is a 10,000 MW dextran conjugated to AlexaFluor-647 and the light source is a Sutter Lambda LS with a Xenon Arc lamp. Two other channels were taken for segmentation purposes including a nuclear channel (a Hoechst stain imaged with a BFP filter set) and a cytoplasmic channel (Calcein Red-Orange imaged with a RFP filter set) also taken in epifluorescence mode. After 15 minutes of imaging, a high powered transmitted UV LED was flashed for 30-45 seconds to uncage a caged chemoattractant present in the media to acutely activate the cells. The cellular response was then recorded for an additional 30 minutes.

Both the raw movies and the processed FxM channel are included in this dataset. The raw data was processed using a custom Julia script (https://github.com/tlnagy/fxm-processing) that denoises, flatfield corrects, segments, extracts the cell volumes, and links the cell tracks. This script relies on a license-encumbered denoising algorithm from Boulanger, J. et al (2010) and therefore for ease of reproducibility the processed FxM channel is also included. The git hash of the script version used to process the data is included.