Specific post-translational histone modifications of neutrophil extracellular traps as immunogens and potential targets of lupus autoantibodies

Abstract

Introduction: Autoreactivity to histones is a pervasive feature of several human autoimmune disorders, including systemic lupus erythematosus (SLE). Specific post-translational modifications (PTMs) of histones within neutrophil extracellular traps (NETs) may potentially drive the process by which tolerance to these chromatin-associated proteins is broken. We hypothesized that NETs and their unique histone PTMs might be capable of inducing autoantibodies that target histones.

Methods: We developed a novel and efficient method for the in vitro production, visualization, and broad profiling of histone-PTMs of human and murine NETs. We also immunized Balb/c mice with murine NETs and profiled their sera on autoantigen and histone peptide microarrays for evidence of autoantibody production to their immunogen.

Results: We confirmed specificity toward acetyl-modified histone H2B as well as to other histone PTMs in sera from patients with SLE known to have autoreactivity against histones. We observed enrichment for distinctive histone marks of transcriptionally silent DNA during NETosis triggered by diverse stimuli. However, NETs derived from human and murine sources did not harbor many of the PTMs toward which autoreactivity was observed in patients with SLE or in MRL/lpr mice. Further, while murine NETs were weak autoantigens in vivo, there was only partial overlap in the immunoglobulin G (IgG) and IgM autoantibody profiles induced by vaccination of mice with NETs and those seen in patients with SLE.

Conclusions: Isolated in vivo exposure to NETs is insufficient to break tolerance and may involve additional factors that have yet to be identified.

Figure 1

Histone-reactive SLE IgG autoantibodies recognize acetyl-H2B. Sera from patients with SLE or healthy subjects were profiled on Human Epigenome Microarray Platform (HEMP) and probed with a secondary anti-human IgG antibody. Samples (rows) are clustered hierarchically based on corresponding reactivity profiles to the depicted peptide epitopes from HEMP arrays (columns). The depicted histone peptides (columns) were selected to match those interrogated by immunoblot assays in Figure 4 and Additional file 1: Supplemental Figures 2 and 4, with specific PTMs epitopes shown at top according to the PTM key. Heatmap tiles reflect magnitude of IgG autoantibody binding reactivity, according to the mean fluorescence intensity (MFI) intensity scale as indicated. The left heatmap panel consists primarily of peptides containing mono-, di- and tri-methyl lysine PTMs, while the right panel consists of peptides containing a variety of PTMs, including citrulline, acetyl, phosphoryl, and both symmetric (s) and asymmetric (a) di-methyl arginine. Blue asterisks (*) mark significant differences in autoantibody binding reactivity between histone positive samples and a combined group of histone negative and healthy control samples, as determined by Significance Analysis of Microarrays (SAM). IgG, immunoglobulin G; K, lysine; MFI, mean fluorescence intensity; nd, not detected; PTM, post-translational modification; R, arginine; SLE, systemic lupus erythematosus.

Figure 2

Generation and visualization of neutrophil extracellular traps (NETs) from myeloid cell lines and human primary neutrophils. A. Three cell line models used to generate neutrophils and NETs. Murine cell lines EPRO and MPRO were induced with ATRA and GM-CSF to differentiate into neutrophils. EPRO cells were generated by differentiating the EML multipotent progenitor cell line with ATRA, GM-CSF and IL-3. NETs were generated by stimulating cell line-derived or primary human neutrophils with hydrogen peroxide. B. Fluorescent (DAPI) images of NETs produced from neutrophils derived from EPRO, HL-60 or MPRO cell lines, or from primary human neutrophils. Color channels are inverted to yield white background for improved contrast. 400× magnification. ATRA, all-trans retinoic acid; DAPI, 4',6-diamidino-2-phenylindole; DMSO, dimethyl sulfoxide; GM-CSF, granulocyte macrophage colony-stimulating factor; IL-3, interleukin-3.

Figure 3

Post-translational histone modifications of neutrophil extracellular traps (NETs) derived from primary human neutrophils. Primary human neutrophils were isolated from peripheral blood, immediately stimulated with 10 mM hydrogen peroxide for 4 hours, and NETs were harvested. NETs or the corresponding unstimulated neutrophils were assayed by MABA. A panel of anti-histone antibody epitopes (top) is shown, with PTMs according to the PTM key. A more limited PTM panel was selected on the basis of available epitopes not subject to proteolysis by considering those enriched or depleted during NETosis of neutrophils derived from human and murine cell lines (Figure 4 and Additional file 1: Supplemental Figure 4). Bottom: asterisks or carats indicate increase or decrease (respectively) of band intensity for indicated lane in NETs compared to unstimulated neutrophils. Band migration is shown for 10-15 kDa range and film exposure time was 20'. MABA, Multiple Antigen Blot Assay; PTM, post translational modification.

Figure 4

EPRO-derived neutrophil extracellular traps (NETs) contain histone post translational modification (PTM) marks consistent with transcriptional silencing and hypercitrullination. Murine cell-line derived neutrophils were stimulated and analyzed as in Figure 3 with additional stimulants including ionomycin and PMA. A panel of anti-histone antibody epitopes (top) matching those in Additional file 1: Supplemental Figure 4 is shown, with PTMs according to the PTM key. A strong and saturated band is notable with H3Cit(2,8,17) in EPRO NETs. Film exposure time for all cells were as described in Figure 3 (additional stimulants: 2' for ionomycin and 90' for PMA). PMA, phorbol myristate acetate.