Sphingosine-1-Phosphate/Sphingosine-1-Phosphate Receptor 2 Axis Can Promote Mouse and Human Primary Mast Cell Angiogenic Potential through Upregulation of Vascular Endothelial Growth Factor-A and Matrix Metalloproteinase-2

Abstract

Mast cells (MC) are present in most vascularized tissues around the vasculature likely exerting immunomodulatory functions. Endowed with diverse mediators, resident MC represent first-line fine-tuners of local microenvironment. Sphingosine-1-phosphate (S1P) functions as a pluripotent signaling sphingolipid metabolite in health and disease. S1P formation occurs at low levels in resting MC and is upregulated upon activation. Its export can result in type 2 S1P receptor- (S1PR2-) mediated stimulation of MC, further fueling inflammation. However, the role of S1PR2 ligation in proangiogenic vascular endothelial growth factor- (VEGF-) A and matrix metalloproteinase- (MMP-) 2 release from MC is unknown. Using a preclinical MC-dependent model of acute allergic responses and in vitro stimulated primary mouse bone marrow-derived MC (BMMC) or human primary skin MC, we report that S1P signaling resulted in substantial amount of VEGF-A release. Similar experiments using S1pr2-deficient mice or BMMC or selective S1P receptor agonists or antagonists demonstrated that S1P/S1PR2 ligation on MC is important for VEGF-A secretion. Further, we show that S1P stimulation triggered transcriptional upregulation of VEGF-A and MMP-2 mRNA in human but not in mouse MC. S1P exposure also triggered MMP-2 secretion from human MC. These studies identify a novel proangiogenic axis encompassing MC/S1P/S1PR2 likely relevant to inflammation.

Figure 1

Role of S1P/S1PR2 signaling in VEGF-A secretion. (a) Blood was collected from allergenically challenged WT (open bar, n = 5-6 mice) and S1PR2-null (black bar, n = 5-6 mice) mice, euthanized 2 hours after Ag challenge, and serum VEGF-A levels were measured in duplicate determination for each animal. (b) Bone marrow-derived mast cells (BMMC, three independent populations) from both genotypes were stimulated for 24 hours with vehicle (DMSO/PBS/4 mg/mL fatty acid-free BSA), S1P (100 nM), IgE/Ag, or ionomycin (Iono) and VEGF-A was measured in MC supernatants, in duplicate determinations. Error bars show standard error of means. ^∗ p < 0.05, ^∗∗ p < 0.005.

Figure 2

Human MC secrete proangiogenic factors upon exposure to S1P in a S1PR2-dependent manner. (a) VEGF-A levels were measured in the supernatants of activated human MC in the absence (open bars, vehicle (DMSO/PBS/4 mg/mL fatty acid-free BSA)) or presence of a 30-minute pretreatment with CYM-5442 (filled bars, S1PR1 agonist, 1 μM) or with JTE-013 (gray bars, S1PR2 antagonist, 1 μM), followed by the indicated stimuli similar to Figure 1(b). (b) Matrix metalloproteinase- (MMP-) 2 levels were measured in supernatants of MC activated exactly as described in Figure 1(b). All supernatants were collected 24 hours after stimulation. Activation experiments were conducted using five independent human skin MC populations generated from five donors. Activation was conducted in triplicate determinations and measurements were conducted in duplicate determinations for each individual determination. When reaching significance, statistics are indicated in each figure. Error bars show standard error of means. ^∗∗∗ p < 0.0001.

Figure 3

Effect of S1P on VEGF-A and MMP-2 mRNA expression. Human (a, b) and mouse (c, d, BMMC) MC were stimulated in the presence of S1P (100 nM) for different periods of time at which VEGF-A (a, c) and MMP-2 (b, d) mRNA levels were measured by QPCR. mRNA was prepared from 3 independent BMMC and 5 independent human MC populations, in duplicate. Kinetics of S1P stimulation was repeated 3 to 4 times; each QPCR experiment was performed in duplicate determination for each individual sample. When reaching significance, statistics are indicated in each figure. Error bars show standard error of means. ^∗∗∗ p < 0.0001.