Functional and Phenotypic Characterization of Siglec-6 on Human Mast Cells

Abstract

Mast cells are tissue-resident cells that contribute to allergic diseases, among others, due to excessive or inappropriate cellular activation and degranulation. Therapeutic approaches to modulate mast cell activation are urgently needed. Siglec-6 is an immunoreceptor tyrosine-based inhibitory motif (ITIM)-bearing receptor selectively expressed by mast cells, making it a promising target for therapeutic intervention. However, the effects of its engagement on mast cells are poorly defined. Siglec-6 expression and endocytosis on primary human mast cells and mast cell lines were assessed by flow cytometry. SIGLEC6 mRNA expression was examined by single-cell RNAseq in esophageal tissue biopsy samples. The ability of Siglec-6 engagement or co-engagement to prevent primary mast cell activation was determined based on assessments of mediator and cytokine secretion and degranulation markers. Siglec-6 was highly expressed by all mast cells examined, and the SIGLEC6 transcript was restricted to mast cells in esophageal biopsy samples. Siglec-6 endocytosis occurred with delayed kinetics relative to the related receptor Siglec-8. Co-crosslinking of Siglec-6 with FcεRIα enhanced the inhibition of mast cell activation and diminished downstream ERK1/2 and p38 phosphorylation. The selective, stable expression and potent inhibitory capacity of Siglec-6 on human mast cells are favorable for its use as a therapeutic target in mast cell-driven diseases.

Keywords: FcεRI; ITIM; Siglec-6; degranulation; endocytosis; mast cell; signaling.

Figure 1

Siglec-6 surface expression on human primary mast cells (MCs) and MC lines. Primary human MCs were isolated and cultured from surgical skin specimens. (A) MC purity was determined by flow cytometry based on size, granularity, and expression of the MC markers CD117 and FcεRIα. (B) The number of Siglec-6 receptors per MC was determined by quantitative flow cytometry over time in culture. The expression of Siglec-6 on human skin-derived mast cells (HSMCs) (C) and the MC lines HMC-1.2 (D), LUVA (E), ROSA KIT^WT (F), and ROSA KIT^D816V (G) was assessed by flow cytometry relative to staining with an isotype control mAb. Siglec-8 staining, or lack thereof, is also shown for ROSA KIT^WT (F), and ROSA KIT^D816V (G) cells.

Figure 2

SIGLEC6 expression is restricted to MCs in esophageal tissue biopsies. Gene expression of esophageal cell populations was assessed by single-cell RNAseq from two distinct cohorts at Cincinnati Children’s Hospital Medical Center (A–G) and Lurie Children’s Hospital (H–M). Results were analyzed using an experimental workflow that led to a uniform manifold approximation and projection (UMAP) plot for dimension reduction displaying 39,562 (A) or 44,153 (H) single cells. The plots are colored by cell types derived from the shared nearest neighbor (SNN) clustering and enriched marker genes. (B,I) Dot plots of marker genes for indicated cell types are shown. (C) Heatmap of SIGLEC6 and SIGLEC8 expression level in five donors with active EoE, in which the color for each gene corresponds to the average per-cell gene expression within the given patient in the esophageal cell population. SIGLEC6 expression in all patients is presented in feature plots (D,J), average expression by diagnosis is presented in a bar graph (F), and expression mean and variance by diagnosis is presented in a violin plot (L). SIGLEC8 expression in all patients is presented in feature plots (E,K), average expression by diagnosis is presented in a bar graph (G), and expression mean and variance by diagnosis is presented in a violin plot (M). Cells were isolated from esophageal biopsies derived from patients with active (n = 5 (A–G) or n = 7 (H–M)) or inactive EoE (n = 3 (A–G) or n = 4 (H–M)) or a non-EoE control subject (n = 1 (H–M)).

Figure 3

Siglec-6 is stable at the surface of MCs and is slowly internalized following antibody ligation. (A) Delayed fluorophore-conjugated secondary mAb detection of anti-Siglec-6 mAb was used to assess Siglec-6 levels remaining on the surface of ROSA KIT^D816V cells (Surface Siglec-6) following antibody ligation after various durations. Fluorophore-conjugated anti-Siglec-6 mAb was used to assess both cell-surface and internalized Siglec-6 (Total Siglec-6) on ROSA KIT^D816V cells. Isotype control mAb (Iso) is used as a negative control in both strategies. Because loss of total Siglec-6 was not observed, Siglec-6 endocytosis was calculated based on the loss of cell-surface Siglec-6 on ROSA KIT^D816V cells (A), as well as HSMCs and HMC-1.2 cells (B). Siglec-8 endocytosis was similarly determined on HSMCs and HMC-1.2 cells (C). Data are representative (histograms) or represent the means and standard deviations of three independent experiments.

Figure 4

Siglec-6 antibody engagement in parallel with various stimuli inhibits HSMC degranulation. HSMCs were collected, washed, and incubated in the presence or absence of anti-Siglec-6 mAb for the indicated duration of time. HSMCs were then incubated with the indicated concentrations of anti-FcεRI mAb (A,D), rhC5a (B), or compound 48/80 (C) for 30 min. Release of β-hexosaminidase was measured by colorimetric assay, and inhibition was calculated on the basis of the reduction of β-hexosaminidase release relative to treatment with an isotype control mAb. Data represent the means and standard deviations of fifteen (A), twelve (B,C), and six (D) replicates from five (A), four (B,C), and two (D) distinct HSMC cultures. * p < 0.05; **** p < 0.0001 vs. isotype control at same stimulus concentration; ns, not statistically significant, determined by two-way ANOVA with Šidák correction for multiple comparisons.

Figure 5

Co-crosslinking of Siglec-6 and FcεRIα on MCs enhances inhibitory activity. (A,B) CD34+ cell-derived MCs were incubated with the indicated concentration of anti-FcεRIα, as well as anti-Siglec-6 or isotype control mAb. Antibodies were crosslinked using secondary anti-mouse IgG antibody, and cells were incubated at 37 °C for 20 min to permit cell stimulation. The percentage of CD63+ (A) or LAMP-1+ (B) MCs was then determined by flow cytometry. (C–G) CD34+ cell-derived MCs were incubated with 150 ng/mL anti-FcεRIα and either anti-Siglec-6 or isotype control mAb, or were not incubated with mAbs (No Stim). Expression of CD63 (C) was determined by flow cytometry, and tryptase release (D) was measured colorimetrically in cell-free supernatant after 20 min of stimulation. MIP-1β (E), IL-8 (F), and IL-1β (G) were detected in cell-free supernatant after overnight stimulation. Levels were normalized to those measured in the stimulated samples incubated with the isotype control antibody (D–G). (H–J) CD34+ cell-derived MCs were incubated in the presence or absence (No Stim) of 150 ng/mL anti-FcεRIα with either anti-Siglec-6 or isotype control mAb. Antibodies were crosslinked using secondary anti-mouse IgG antibody, and cells were incubated at 37 °C for 15 min in complete medium to permit cell stimulation. Following stimulation, cells were fixed and stained for surface CD63 (H), intracellular phospho-ERK1/2 (I), or intracellular phospho-p38 (J). Data are representative (A,B), represent the means and standard deviations of four (C–E) or three (F,G) independent mast cell cultures and experiments, or represent the individual values, means, and standard deviations of two (H–J: No Stim) or four (H–J: Anti-FcεRIα) replicates. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 by one-way ANOVA with Tukey test to correct for multiple comparisons (C–G). **** p < 0.0001 by two-way ANOVA with Šidák correction for multiple comparisons (H–J).

Figure 6

Inhibition induced by co-engagement of Siglec-6 and FcεRIα does not depend on reduced FcεRIα crosslinking. Streptavidin complexes (SAv) with the indicated ratio of anti-Siglec-6 to anti-FcεRIα or uncomplexed antibody with final concentrations matching that of the 1:1 SAv complexes were used to stimulate HSMCs. After 30 min, the cells were washed, stained, and analyzed by flow cytometry. MCs were pre-gated on live CD117+ cells. The percentage of MCs expressing surface LAMP-1 (A), LAMP-1 normalized to the Iso/Anti-FcεRIα sample in each group (B), and dot plots comparing stimulated samples with anti-Siglec-6 or isotype control mAb with respect to surface LAMP-1 and unbound Siglec-6 expression (C) are shown. Unbound FcεRIα (D) and unbound Siglec-6 (E) normalized to the control are quantified based on sample MFI. (F) Approximately 24 h after initial stimulation as indicated, cells were stimulated with 500 ng/mL anti-FcεRIα and stained for LAMP-1 and MC markers. The percentage of LAMP-1+ HSMCs normalized to that of cells not initially treated with antibody is quantified. Data represent the means and standard deviations (A,B,D–F) or are representative (C) of four independent experiments using three distinct HSMC cultures. *** p < 0.001; **** p < 0.0001; ns, not statistically significant, determined by two-way ANOVA with Tukey test to correct for multiple comparisons.