Angiogenic Host Defense Peptide AG-30/5C and Bradykinin B2 Receptor Antagonist Icatibant Are G Protein Biased Agonists for MRGPRX2 in Mast Cells

Abstract

AG-30/5C is an angiogenic host defense peptide that activates human mast cells (MC) via an unknown mechanism. Using short hairpin RNA-silenced human MC line LAD2 and stably transfected RBL-2H3 cells, we demonstrate that AG-30/5C induces MC degranulation via Mas-related G protein-coupled receptor X2 (MRGPRX2). Most G protein-coupled receptors signal via parallel and independent pathways mediated by G proteins and β-arrestins. AG-30/5C and compound 48/80 induced similar maximal MC degranulation via MRGPRX2, which was abolished by pertussis toxin. However, compound 48/80 induced a robust β-arrestin activation as determined by transcriptional activation following arrestin translocation (Tango), but AG-30/5C did not. Overnight culture of MC with compound 48/80 resulted in reduced cell surface MRGPRX2 expression, and this was associated with a significant decrease in subsequent MC degranulation in response to compound 48/80 or AG-30/5C. However, AG-30/5C pretreatment had no effect on cell surface MRGPRX2 expression or degranulation in response to compound 48/80 or AG-30/5C. Icatibant, a bradykinin B₂ receptor antagonist, promotes MC degranulation via MRGPRX2 and causes pseudoallergic drug reaction. Icatibant caused MC degranulation via a pertussis toxin-sensitive G protein but did not activate β-arrestin. A screen of the National Institutes of Health Clinical Collection library led to the identification of resveratrol as an inhibitor of MRGPRX2. Resveratrol inhibited compound 48/80-induced Tango and MC degranulation in response to compound 48/80, AG-30/5C, and Icatibant. This study demonstrates the novel finding that AG-30/5C and Icatibant serve as G protein-biased agonists for MRGPRX2, but compound 48/80 signals via both G protein and β-arrestin with distinct differences in receptor regulation.

Figure. 1:. AG-30/5C induces MC degranulation via MRGPRX2:

(A) LAD2 cells were stimulated with different concentrations of AG-30/5C (0.01 – 1.5 μM) for 30 min. Percent degranulation (β-hexosaminidase release) was determined. (B) LAD2 cells were stably transduced with scrambled control shRNA or lentivirus shRNA targeting human MRGPRX2. Western blotting was performed to determine MRGPRX2 expression. β-actin was used as a loading control. (C) Control and MRGPRX2 knockdown LAD2 cells were stimulated with AG-30/5C (1 μM) and degranulation was determine. (D) RBL-2H3 and RBL-2H3 cell stably expressing MRGPRX2 (RBL-MRGPRX2) were stimulated with different concentrations of AG-30/5C (0.01 – 1.5 μM) and percent degranulation was determined. All data are expressed as a mean ± SEM of three experiments. Statistical significance was determined by non-parametric t-Test and two way ANOVA (P value<.05). Western blotting data presented are representative of three similar experiments. *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 2:. Pertussis toxin inhibits AG-30/5C and compound 48/80-induced MC degranulation but not calcium mobilization.

(A) RBL-MRGPRX2 cells cultured in absence (-PTx) or presence of pertussis toxin (+PTx, 100 ng/ml, 16h). Cells were stimulated with compound 48/80 (1μg/mL) or AG-30/5C (1 μM) and percent degranulation was determined. (B-C) RBL-MRGPRX2 cells were cultured in the absence or presence of PTx, loaded with Indo-1 (1 μM, 30 min) and stimulated with either compound 48/80 (1 μg/mL) or AG-30/5C (1 μM) as indicated by arrows and time course of calcium mobilization was determined. Representative traces of 3 similar experiments are shown. All data points for degranulation are expressed as mean ± SEM of three experiments. Statistical significance was determined by non-parametric t-Test and two way ANOVA (P value<.05). *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 3:. Compound 48/80 is a balanced agonist whereas AG-30/5C is a β-arrestin-biased agonist for MRGPRX2.

(A) Modular design of Tango constructs. (B) Upon activation (1), β-arrestin is recruited to the C terminus of the receptor (2). This is followed by cleavage of the GPCR fusion protein at the TEV protease–cleavage site (3). Cleavage results in the release of the tTA transcription factor (4), which after transport to the nucleus activates transcription of the luciferase reporter gene (5). (C) Untransfected (HTLA) and stably transfected HTLA-MRGPRX2-Tango (HTLA-MRGPRX2) cells were incubated with anti-FLAG-PE antibody and cell surface receptor expression was determined by flow cytometry. A representative overlay histogram of MRGPRX2 expression is shown. (D) HTLA-MRGPRX2 cells were exposed to compound 48/80 (1.0 or 30 μg/mL) and AG-30/5C (1.0 and 10 μM) respectively and β-arrestin-mediated gene expression was determined. (E) Representative Ca²⁺ measurement traces for compound 48/80 (1 μg/mL and 30 μg/mL) and AG-30/5C (1 μM and 10 μM) are shown. All data points for the β-arrestin activation assay are expressed as a mean ± SEM of three experiments. Statistical significance was determined by non-parametric t-Test and two way ANOVA (P value<.05). *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 4:. Nateglinide is a balanced agonist whereas AG-30/5C is a G protein -biased agonist for MRGPRX4.

(A) Untransfected (HTLA) and stably transfected HTLA-MRGPRX4-Tango (HTLA-MRGPRX4) cells were incubated with anti-FLAG-PE antibody and receptor expression was determined by flow cytometry. A representative overlay histogram of MRGPRX4 expression is shown. (B) Effects of nateglinide (30 μM), AG-30/5C (10 μM), compound 48/80 (30 μg/mL) on β-arrestin-mediated gene expression was determined as described in Figure 3. (C) Representative Ca²⁺ measurements traces in response to high concentrations of nateglinide (30 μM), AG-30/5C (10 μM) and compound 48/80 (30 μg/mL) are shown. All data points for the Tango assay are expressed as a mean ± SEM of three experiments. Statistical significance was determined by non-parametric t-Test and two way ANOVA (P value<.05). *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 5:. Compound 48/80 causes MRGPRX2 internalization and reduced MC degranulation whereas AG-30/5C does not.

(A) RBL-MRGPRX2 cells (0.5 × 10⁶) were cultured in the absence (unstimulated) or presence of compound 48/80 (1 μg/mL) or AG-30/5C (1 μM) for 16 h and cells surface expression of MRGPRX2 was determined by flow cytometry using anti-MRGPRX2-PE antibody. The data is presented as mean fluorescent intensity (MFI) of three experiments (B) A representative histogram of cell surface MRGPRX2 receptor expression is shown. (C) LAD2 cells were cultured in the absence (Control) or presence of in compound 48/80 (1 μg/mL) or AG-30/5C (1 μM) for 16h, washed and plated in a 96 well plate (10,000 cells/well). Cells were stimulated with compound 48/80 (1.0 μg/mL) and AG-30/5C (1 μM) respectively for 30 min and percent degranulation was determined by β-hexosaminidase release assay. All the points are expressed as a mean ± SEM of three experiments. Statistical significance was determined by non-parametric t-Test and two way ANOVA. *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 6:. Resveratrol inhibits β-arrestin-mediated gene expression and MC degranulation.

(A) HTLA cells stably expressing MRGPRX2-Tango were incubated with resveratrol (100 μM, 5 min) followed by overnight stimulation with compound 48/80 (3.0, 10 and 20 μg/mL) or AG-30/5C (10 μM) respectively and relative luminescence was determined. (B) RBL-MRGPRX2 cells and (C) LAD2 cells either left untreated (Control) or incubated with resveratrol (100 μM, 5 min) and stimulated with compound 48/80 (1μg/mL) or AG-30/5C (1 μM) for 30 min and percent degranulation was determined. Statistical significance was determined by non-parametric t-Test and two way ANOVA. *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 7:. Icatibant is a G protein-biased agonist for MRGPRX2 and inhibition of degranulation by resveratrol:

(A and B) RBL-MRGPRX2 cells were cultured in the absence or presence of PTx (100 ng/mL, 16 h) and percent degranulation and Ca²⁺ mobilization in response to Icatibant (20 μg/mL) was determined as described for compound 48/80 and AG-30/5C (Figure 2). (C) RBL-MRGPRX2 and (D) LAD2 cells were incubated with resveratrol (100 μM, 5 min), stimulated with Icatibant (20 μg/mL) for 30 min and percent degranulation was determined. (E) HTLA cells expressing MRGPRX2 (HTLA-MRGPRX2) were exposed to medium (Control) or resveratrol (100 μM of 5 min) followed by overnight stimulation with compound 48/80 (10 μg/mL) or Icatibant (30 μg/mL) and chemiluminescence was measured (F) Representative Ca²⁺ traces for compound 48/80 (30 μg/mL) and Icatibant (30 μg/mL) in HTLA-MRGPRX2 cells are shown. Statistical significance was determined by non-parametric t-Test and two way ANOVA. *** indicates P value<.001,** indicates P value <.01 and * indicates P value <.05.

Figure. 8:. Model for the possible mode of action of AG-30/5C on host defense and wound healing.

We propose that in addition to its direct antimicrobial activity, AG-30/5C contributed to host defense via MC degranulation through MRGPRX2. AG-30/5C contributes to wound healing via its action on vascular endothelial cells keratinocyte proliferation and migration through MRGPRX3 and MRGPRX4.