Fetal Tissue-Derived Mast Cells (MC) as Experimental Surrogate for In Vivo Connective Tissue MC

Abstract

Bone-marrow-derived mast cells are matured from bone marrow cells in medium containing 20% fetal calf serum (FCS), interleukin (IL)-3 and stem-cell factor (SCF) and are used as in vitro models to study mast cells (MC) and their role in health and disease. In vivo, however, BM-derived hematopoietic stem cells account for only a fraction of MC; the majority of MC in vivo are and remain tissue resident. In this study we established a side-by-side culture with BMMC, fetal skin MC (FSMC) or fetal liver MC (FLMC) for comparative studies to identify the best surrogates for mature connective tissue MC (CTMC). All three MC types showed comparable morphology by histology and MC phenotype by flow cytometry. Heterogeneity was detected in the transcriptome with the most differentially expressed genes in FSMC compared to BMMC being Hdc and Tpsb2. Expression of ST2 was highly expressed in BMMC and FSMC and reduced in FLMC, diminishing their secretion of type 2 cytokines. Higher granule content, stronger response to FcεRI activation and significantly higher release of histamine from FSMC compared to FLMC and BMMC indicated differences in MC development in vitro dependent on the tissue of origin. Thus, tissues of origin imprint MC precursor cells to acquire distinct phenotypes and signatures despite identical culture conditions. Fetal-derived MC resemble mature CTMC, with FSMC being the most developed.

Keywords: IgE; RNA sequencing; in vitro model; mast cell; proteases; transcriptome.

Figure 1

Generation and characterization of MC. (a) MC determined by surface markers using flow cytometry. Cells were cultured in medium with SCF and IL−3 and flow cytometry performed weakly. Mature MC were classified as LD⁻CD45.2⁺FcεRI⁺CD117⁺ cells. Bar graphs represent quantification of maturation from week 2 until week 5 (n = 3, mean with SEM). (b) Morphology of BMMC, FLMC and FSMC analyzed by histology. Semithin sections were stained with Toluidin Blue (top panel, 100× magnification), Giemsa (middle panel, 40× magnification), and Alcian Blue/Safranin O (bottom panel, 40× magnification). (c) Structural analysis of MC was conducted by electron microscopy. Representative images with red arrows representing areas with granules (scalebar 1 µM), (d) quantification of granules per section (n = 70). (e) Proliferation of MC based on the incorporation of ³H−thymidine. Proliferation rate determined as counts/min (n = 3, mean with SEM). Unpaired t-test * p > 0.05, **** p > 0.0001.

Figure 2

MC response to IL−33 stimulation. Flow cytometry analysis of surface receptor ST2 shown as (a) a representative histogram of surface expression on BMMC (light grey), FLMC (medium grey) and FSMC (dark grey) and (b) quantified as mean fluorescent intensity (MFI) (n = 4, mean with SEM). (c) MC were stimulated with 10 ng/mL IL−33 for 1 h and RNA was isolated. mRNA expression level of IL−13 was analyzed by quantitative real-time PCR. Fold change determined as 2^−∆∆ct (n = 3, mean with SEM). (d) MC were stimulated with 10 ng/mL IL−33 for 24 h and concentration of secreted type 2 cytokines were measured in the supernatant via enzyme-linked immunosorbent assay (n = 3, mean with SEM). Color gradient is depicted from 0–50,000 pg/mL.

Figure 3

MC transcriptome analysis. (a) RNA sequencing of BMMC (dark blue), FLMC (light blue) and FSMC (orange) was performed. Principal component analysis showed variance among the MC types (n = 3). (b) MC protease mRNA expression analyzed by quantitative real−time PCR. ∆ct values relative to housekeeping gene β−actin (n = 3, mean with SEM). (c,d) Volcano plot of differentially expressed genes between the groups (c) FLMC vs. BMMC and (d) FSMC vs. BMMC analyzed with the webtool Degust (n = 3, FDR ≤ 0.01, abslogFc > 1). (e) Heatmap showing the top 20 DEG between BMMC (black), FLMC (dark grey) and FSMC (light grey). Upregulated genes are shown in red and downregulated genes in blue (n = 3). (f) Differential expression of genes associated with the MC signature by Dywer et al. [18]. Gene expression of FLMC and FSMC normalized against BMMC. Upregulated genes are shown in red and downregulated genes in blue. (n = 3, FDR ≤ 0.01, abslogFc > 2).

Figure 4

MC response to IgE stimulation. (a) Heatmap of differentially expressed genes relative to BMMC (FDR > 1, abs logFC > 1.5×). Gene list based on the FC_Epsilon_RI_Signaling gene set from KEGG. (b) MC were incubated over night with IgE anti−DNP. Prior to stimulation, cells were loaded with fluorescent Fura−2. Calcium influx was triggered by DNP−HSA. (c) Comparable sustained activation of the K−channel SK4 as shown by patch clamp analysis. (d) Histamine levels detected in supernatant after PMA/ionomycin or IgE−HSA stimulation for 24 h using an enzyme-linked immunosorbent assay (n = 4, mean with SEM), * p > 0.05 ** p > 0.01). (e) MC were stimulated with PMA/ionomycin or IgE−DNP for 30 min. Hexosaminidase activity was calculated as percent released relative to total cell lysis. (n = 4, mean with SEM), * p > 0.05.

Figure 5

MC response towards TLR stimulation. (a) Toll−like receptor mRNA expression levels of MC analyzed by quantitative real-time PCR. ∆ct relative to housekeeping gene HPRT (n = 3, mean with SEM). (b,c) MC were stimulated with various TLR−agonists for 24 h. Concentration of secreted (b) IL−6 and (c) TNF-a were measured in the supernatant via enzyme-linked immunosorbent assay (n = 3, mean with SEM). (d) MC were stimulated with LPS for 24 h. Cytokines in the supernatant were analyzed using an enzyme-linked immunosorbent assay. Concentrations are given in pg/mL. Color gradient is depicted from 0−5000 pg/mL. * p > 0.05, ** p > 0.01.