Proadrenomedullin N-Terminal 20 Peptides (PAMPs) Are Agonists of the Chemokine Scavenger Receptor ACKR3/CXCR7

Abstract

Adrenomedullin (ADM) and proadrenomedullin N-terminal 20 peptide (PAMP) are two peptides with vasodilative, bronchodilative, and angiogenic properties, originating from a common precursor, proADM. Previous studies proposed that the atypical chemokine receptor ACKR3 might act as a low-affinity scavenger for ADM, regulating its availability for its cognate receptor calcitonin receptor-like receptor (CLR) in complex with a receptor activity modifying protein (RAMP). In this study, we compared the activation of ACKR3 by ADM and PAMP, as well as other related members of the calcitonin gene-related peptide (CGRP) family. Irrespective of the presence of RAMPs, ADM was the only member of the CGRP family to show moderate activity toward ACKR3. Remarkably, PAMP, and especially further processed PAMP-12, had a stronger potency toward ACKR3 than ADM. Importantly, PAMP-12 induced β-arrestin recruitment and was efficiently internalized by ACKR3 without inducing G protein or ERK signaling in vitro. Our results further extend the panel of endogenous ACKR3 ligands and broaden ACKR3 functions to a regulator of PAMP-12 availability for its primary receptor Mas-related G-protein-coupled receptor member X2 (MrgX2).

Figure 1

ADM is the only CGRP-family member with limited ACKR3 activity not influenced by RAMPs. (A) Schematic overview of preproADM processing into at least two active peptides (PAMP and ADM) and adrenotensin (ADT), whose bioactivity has to be confirmed. Amino acid (AA) motifs recognized by pro-protein convertases potentially involved during pro-ADM maturation are indicated. MR-proADM: midregional proadrenomedullin, SP: signal peptide. (B–D) Efficacy and potency of different CGRP family members in inducing β-arrestin recruitment toward ACKR3 (B and C) or CLR (D) in HEK cells in the absence of regulatory proteins (B), or in the presence of one of the three RAMPs or CD8 used as negative control protein (C and D) using NanoBiT technology. Results are expressed as percentage of full agonist response and represent the mean ± SEM of three independent experiments (n = 3).

Figure 2

PAMP peptides have comparable activity toward ACKR3 and MrgX2 and no activity toward any other chemokine receptor. (A) Sequences of ADM and the three PAMP variants tested: full-length PAMP, comprising AAs 1–20; PAMP-12, comprising AAs 9–20; and PAMP(12–20), comprising the last 9 AAs of PAMP. For ADM, cysteine residues involved in a disulfide bridge forming a 4-residue intrapeptide arch are depicted in red. (B–C) Comparison of potency and efficacy of different active PAMP variants, ADM, ADT, and substance P in inducing β-arrestin-2 toward ACKR3 (B) or MrgX2 (C) in HEK cells, normalized to percent activity of their respective full agonists. (D) Binding competition of CGRP, ADM, and PAMP variants (9 μM) with AlexaFluor647-labeled CXCL12 (5 nM) on HEK-ACKR3 cells determined by flow cytometry. (E) Agonist activity of ADM and different PAMP variants (3 μM) toward all chemokine receptors, as well as the MrgX2 and GPR182 monitored in a β-arrestin-2 recruitment assay. Results are expressed as fold change over vehicle. For each receptor, an agonist chemokine (100 nM) listed in the IUPHAR repository of chemokine receptor ligands was used as the positive control. Results from B–E are represented as mean ± SEM of three independent experiments. **, p < 0.01 by one-way ANOVA with Bonferroni multiple comparison test.

Figure 3

SAR analysis of PAMP-12 variants on ACKR3 and MrgX2. (A) Comparison of the impact of substitutions or truncations on the agonist activity of PAMP-12 toward ACKR3 and MrgX2. The agonist activity of each variant was evaluated in a β-arrestin-2 recruitment assay in HEK cells and expressed in a heat map as fold change in EC₅₀ values with respect to wild-type PAMP-12. Four variants of PAMP were also included: PAMP(1–17), PAMP(16–20), PAMP(13–20), and PAMP(4–20). (B–D) Comparison of potency and efficacy of PAMP-12 (B) and its variants bearing mutations F1Y or K7R in inducing β-arrestin-2 recruitment to ACKR3 (C) and MrgX2 (D) in HEK cells. Results represent the mean (A) or mean ± SEM (B−D) of three independent experiments (n = 3). The corresponding pEC₅₀ values are available in Supplementary Table 2.

Figure 4

ACKR3 internalizes PAMP-12 without inducing G protein or ERK-signaling. (A, B) Comparison of miniGi and miniGq recruitment to ACKR3 (A) and MrgX2 (B) in response to ADM, substance P, and PAMP variants monitored in HEK cells using NanoBiT technology. (C) Kinetic analysis of ERK1/2 phosphorylation in HEK cells transfected with ACKR3- or MrgX2-encoding plasmids treated with vehicle or PAMP-12 (3 μM). EGF (100 nM) was used as positive control. (D) Activation of ERK (SRE) and Ca²⁺ (NFAT) signaling cascades in HEK cells expressing ACKR3 or MrgX2 in response to CXCL12 (300 nM), ADM, substance P, or PAMP variants (3 μM) or positive control (30 nM PMA, 10% FBS for SRE; 30 nM PMA, 1 μM ionomycin, 10% FBS for NFAT). (E) ACKR3 delivery to endosomes induced by peptides (3 μM) and chemokines (300 nM) monitored in HEK cells by NanoBRET using nanoluciferase-tagged β-arrestin-2 and mNeonGreen-tagged FYVE domain of endofin, which binds phosphatidylinositol 3-phosphate (PI3P) in early endosomes. Results are expressed in miliBRET units (mBU). (F–I) Uptake of fluorescently labeled PAMP-12 (PAMP-12-FAM, 3 μM). (F) Uptake of PAMP-12-FAM by ACKR3-positive or -negative HEK cells pretreated (or not) with CXCL12 or CXCL10 (200 nM) visualized by imaging flow cytometry. Five representative HEK or HEK-ACKR3 cells are shown. BF: brightfield. Scale bar: 7 μm. (G) PAMP-12-FAM uptake for conditions described in (F), quantified by mean fluorescence intensity (MFI) and normalized to signal obtained for nontransfected HEK cells. (H) Percentage of cells with a given number of distinguishable vesicle-like structures (spots) for conditions determined in (F). (I) PAMP-12-FAM uptake by HEK cells, transiently transfected with equal amounts of ACKR3 or MrgX2 encoding plasmids or an empty vector (NT) quantified by MFI. For all panels, results represent the mean ± SEM of at least three independent experiments (n ≥ 3) except for F and H, where one representative experiment of three independent repetitions is shown. *, p < 0.05; **, p < 0.01; ***, p < 0.001; and ****, p < 0.0001 by one-way ANOVA with Bonferroni (E) or Tuckey’s multiple comparison test (G and I).