NETome: A model to Decode the Human Genome and Proteome of Neutrophil Extracellular Traps

Abstract

Neutrophils are the most abundant type of white blood cells in humans with biological roles relevant to inflammation, and fighting off infections. Neutrophil Extracellular Traps (NETs) act as enxogenous agents controlling invasion by bacteria, viruses, fungi, metabolic, and traumatic agents. Traditionally, studies have focused on elucidating molecular and cellular pathways preceding NET formation. Here, we developed a model to decode the human genome and proteome of developted NETs. Via in vitro system to differentiate HL-60 human myeloid cell line into neutrophil extracellular trap (ecTrap) producing cells, we isolated and captured ectrap derived DNA and proteins for shotgun sequencing. The genomic sequences revealed accurate delineation of gene composition including immune response genes and mitochondrial enrichment, while providing a reference database for future interrogation. Shotgun proteomics showed global proteins in differentiated cells with specific immune pathways when compared to undifferentiated counterparts. Coupled with omics' approaches, we validated our system by functional assays and began to dissect host-microbial interactions. Our work provides a new understanding of the genomic and proteomic sequences, establishing the first human database deposition of neutrophil extracellular traps.

Fig. 1

Neutrophil Extracellular Trap Production and Isolation. (a) Schematic of in vitro ecTrap production. Cultured HL-60 cell lines are incubated with DMSO to differentiate into neutrophils (dHL60 cells). After which, PMA stimulation leads to ecTrap production, isolation, and “omics” analysis. (b) Representative 3D holotomographic microscopy images digitally stained based on RI (refractive index) confirm the differentiation of HL-60 cells to neutrophils (dHL60) after 4 days in differentiation media, and successful release of ecTraps after 4-hour incubation in 1,000 nM PMA. (c. i-iii) Scanning electron microscope (SEM) images of isolated DNA samples from (i) HL-60 cells, (ii) dHL60 cells and (iii) released ecTrap of static cells. All samples show the presence of lipid bilayers (indicated by arrowheads). (d) Mitochondrial superoxide generation before - and during - ecTrap release process was measured by MitoSOX assay over 5 hours of incubation with PMA. (i) Doublet discrimination gating strategy was used to ensure accurate MitoSOX-red quantification. Panels shown are negative control (dHL60 in PBS) (ii) Representative panel of flow cytometry analysis shows the generation of superoxide in dHL60 on incubation with PMA over 0.5 hours. Red circle highlights a population shift from MitoSOX negative to MitoSOX positive. (e) DNA quantification by Agilent High-sensitive DNA chip verifies the composition of extracted ecTrap samples. Lane 1 shows an isolated ecTrap DNA sample. Lane 2 shows an isolated ecTrap sample after incubation with DNase to digest all DNA contents. Arrows indicate the electropherogram of each sample in the gel image above. Lane L shows the DNA ladder marker.

Fig. 2

dHL60 induced ecTrap Proteome. (a) Proteome analysis of ecTrap from three representative samples (i.e., rep1, rep2, and rep3) identified a total of 2,364 proteins after Benzonase treatment and 1,711 proteins in untreated samples. Common proteins found among three representative samples in Benzonase treated and untreated ecTrap is 1,358 and 1,008, respectively. (b) Dynamic range of the ecTrap proteome. Data showing (1,722 proteins) here is from Benzonase-treated ecTrap. Median values of the three replicate experiments were used for the plot. Previously reported proteins associated with NET by Urban et. al. denoted by orange dots, most of which ranked among the 100 most abundant proteins found in our ecTrap samples. (Median value of three experimental replicates are plotted here). (c) Hierarchical clustering of the 126 significant proteins (fold change ≥2 or ≤−2; Permutation FDR 0.05) between the two groups. Z-scored LFQ intensities were color-coded as indicated in the scale bar. (d) STRING protein network and Gene Ontology analysis of 101(out of 126) significantly enriched proteins in ecTraps after Benzonase treatment were analysed using embedded STRING app in CytoScape software (version 3.7.2). The confidence score cut-off was set to 0.4. Representative enriched gene ontology (GO) terms (e.g., biological process, molecular function, and cellular compartment) and corresponding FDR values were depicted in the network.

Fig. 3

Gene Enrichment/Depletion Comparison Published Database. A 1.5-fold cut-off enrichment screen was used to determine regions of enrichment/depletion. Using annotated gene coding regions, a comparison to published expression data was used to determine overlap. The resultant circos plot for chromosomes 1–22 is shown. From innermost to outermost tracks: heatmap in order of HL-60, dHL60, and ecTrap enrichments; linkers from heatmap to color coded chromosomes in order from red (chromosome 1) clockwise to pink (chromosome 22); and linkers to gene names. Table of gene names used is available upon request.

Fig. 4

Genomic Enrichment of ecTRapRegions. (a) Comparison of telomere counts (TTAGGGTTAGGG) across all chromosomes for each sample (ANOVA, p < 0.05). (b) Normalized mitochondrial enrichment quantification from two sequencing runs (n = 2 per sample, ANOVA, p < 0.05). (c) Mitochondrial enrichment by position number (sliding window size = 500 nts).