The atypical chemokine receptor ACKR3/CXCR7 is a broad-spectrum scavenger for opioid peptides

Abstract

Endogenous opioid peptides and prescription opioid drugs modulate pain, anxiety and stress by activating opioid receptors, currently classified into four subtypes. Here we demonstrate that ACKR3/CXCR7, hitherto known as an atypical scavenger receptor for chemokines, is a broad-spectrum scavenger of opioid peptides. Phylogenetically, ACKR3 is intermediate between chemokine and opioid receptors and is present in various brain regions together with classical opioid receptors. Functionally, ACKR3 is a scavenger receptor for a wide variety of opioid peptides, especially enkephalins and dynorphins, reducing their availability for the classical opioid receptors. ACKR3 is not modulated by prescription opioids, but we show that an ACKR3-selective subnanomolar competitor peptide, LIH383, can restrain ACKR3's negative regulatory function on opioid peptides in rat brain and potentiate their activity towards classical receptors, which may open alternative therapeutic avenues for opioid-related disorders. Altogether, our results reveal that ACKR3 is an atypical opioid receptor with cross-family ligand selectivity.

Fig. 1. Opioid-peptide library screening on ACKR3 and hit activity comparison.

a The ability of 58 compounds, including natural opioid peptides from the four opioid families, variants thereof and small molecule opioid receptor modulators, to induce β-arrestin-2 recruitment to ACKR3, CXCR4 and CXCR3 in U87 cells at a concentration of 5 µM. For full peptide names and sequences, see Supplementary Table 1. Chemokine positive controls (Ctrls) were used at a concentration of 300 nM. Results are expressed as fold change over vehicle-treated cells and presented as mean of two technical replicates for ACKR3 and CXCR4. A single measurement was performed for CXCR3. b–f Comparison of potency and efficacy of ACKR3-activating and other representative opioid peptides in inducing β-arrestin-1 recruitment to the opioid receptors MOR (b), DOR (c), KOR (d), NOP (e), and ACKR3 (f) in U87 cells. Results are expressed as percentage of indicated agonist response. The corresponding EC₅₀ and Emax values are summarized in Table 1. g Binding competition of ACKR3-activating and other representative opioid peptides with Alexa Fluor 647-labeled CXCL12 (5 nM) on U87-ACKR3 cells determined by flow cytometry. Results from b-g are presented as mean ± S.E.M of three or four independent experiments (n = 3 or 4). Peptides from the enkephalin, dynorphin, nociceptin and endorphin families are depicted in blue, green, orange and gray scale, respectively. h Schematic representation of the interactions of the four opioid receptors and chemokine receptors with their respective ligands. The newly identified opioid peptide pairings with ACKR3 are shown, highlighting the cross-family selectivity of ACKR3. Source data are provided as Source Data file.

Fig. 2. Specific activation of ACKR3 by opioid peptides.

Agonist activity of opioid peptides (3 µM) representative of the four opioid families towards 18 classical and three atypical chemokine receptors evaluated in β-arrestin-1 recruitment assay in U87 cells. Results are expressed as percentage of signal monitored with saturating concentration of active chemokines (100 nM) used as positive controls (Supplementary Table 3) and presented as mean of three independent experiments (n = 3). Source data are provided as a Source Data file.

Fig. 3. SAR analysis of adrenorphin variants on ACKR3 and classical opioid receptors.

a Comparison of the impact of substitutions, truncations, extensions, D-amino acid replacement or dimerization on the agonist activity of adrenorphin towards ACKR3, MOR, DOR, and KOR. The agonist activity of each variant was evaluated in β-arrestin-1 recruitment assay in U87 cells and expressed as fold change over the activity of wild-type adrenorphin. None of the peptides showed agonist activity towards NOP. b–e Comparison of potency and efficacy of adrenorphin (b) and its variants bearing mutations Y1F (c), M5L (d) and R6A (e) to induce β-arrestin-1 recruitment to ACKR3 and the opioid receptors KOR, MOR, DOR and NOP in U87 cells. ∼ indicates a similar impact of the modification on the potency of the peptide towards ACKR3 and the indicated opioid receptor. Results represent the mean (a) or mean ± S.E.M (b–e) of three independent experiments (n = 3). Corresponding EC₅₀ values are available in Supplementary Table 4. Source data are provided as a Source Data file.

Fig. 4. Activity of opioid modulators and development of LIH383 as ACKR3-selective agonist.

a Agonist (green color scale) and antagonist (red color scale) activity towards ACKR3 of opioid modulators (10 µM, 1 µM, and 100 nM) commonly used for research purposes or in clinic and comparison with the opioid receptors MOR, DOR, KOR, and NOP monitored in a β-arrestin-1 recruitment assay. Antagonist activity was measured following addition of BAM22 (50 nM), met-enkephalin (70 nM), dynorphin A (50 nM), nociceptin (70 nM) and CXCL12 (4 nM) on MOR, DOR, KOR, NOP, and ACKR3, respectively (b, e) Agonist activity of LIH383 (FGGFMRRK) towards human (b) and mouse (e) ACKR3 and comparison with other endogenous ACKR3 chemokine ligands, and a control peptide (MRRKFGGF), consisting of the eight amino acids building LIH383 in a different arrangement. c, d LIH383 selectivity evaluated by comparison of β-arrestin-1 recruitment to ACKR3 and the opioid receptors MOR, DOR, KOR and NOP (c) or all the other chemokine receptors known to recruit β-arrestin (3 µM) (d). f Selective binding of fluorescently labeled LIH383 (LIH383-Cy5) to ACKR3-expressing cells. All assays were performed in U87 cells. Results represent the mean (a) or mean ± S.E.M (b–f) of three to five independent experiments (n = 3–5) except for LIH383 on hACKR3 (n = 9). All fitted values are available in Supplementary Table 6. For statistical analysis, one-way ANOVA with Bonferroni multiple comparison test was used. ****p < 0.0001. Source data and statistical analysis parameters are provided as Source Data file.

Fig. 5. Absence of ACKR3 signaling in response to opioid and chemokine ligands.

a–d DMR profiles of U87 cells expressing (or not) ACKR3, CXCR4 or classical opioid receptors (KOR and NOP) stimulated by the chemokine CXCL12 (200 nM) (a) or opioid peptides (500 nM) dynorphin A (b), adrenorphin (c) and nociceptin 1–13 (d). Left panels: representative DMR profiles determined over 3000 s of three to five independent experiments. Right panel: area under the curve (AUC) ± S.E.M. of three to five independent experiments (n = 3 to 5). e Comparison of mini Gi recruitment to ACKR3 and classical opioid receptors (MOR, DOR, KOR, or NOP) in response to CXCL12 and opioid peptides monitored in U87 cells. The corresponding EC₅₀ and Emax values are summarized in Supplementary Table 7. f Kinetic analysis of ERK1/2 phosphorylation in U87 cells stably expressing ACKR3 or CXCR4 stimulated by CXCL12 and opioid peptides. EGF was used as positive control. g Comparison of activation of SRE (ERK1/2) and NFAT-RE (Ca²⁺) signaling cascades (green color scale, relative light units, RLU) in U87 cells (-) or U87 expressing ACKR3, CXCR4 or classical opioid receptors (MOR, DOR, KOR, or NOP) in response to chemokines CXCL12 and CXCL11 (200 nM), opioid peptides (500 nM) or positive controls (30 nM PMA, 10% FBS for SRE and 30 nM PMA, 1 μM ionomycin, 10% FBS for NFAT-RE). Results represent the mean ± S.E.M. (e–f) or the mean (g) of three to five independent experiments (n = 3–5). Source data are provided as a Source Data file.

Fig. 6. Uptake of opioid peptides and atypical localization and trafficking of ACKR3.

a Uptake of dynorphin A (1–13) by ACKR3-expressing cells visualized by imaging flow cytometry. Left panels: U87, U87-ACKR3 or U87-ACKR3 cells pretreated with LIH383 (3 µM) stimulated with (FAM)-labeled dynorphin A (1–13) (250 nM, dA13-FAM, green channel). Five representative cells per condition are shown. Scale bar: 10 µm. Right panel: Percentage of cells with a given number of distinguishable vesicle-like structures (spots) and the geometrical mean fluorescence intensity (MFI, green channel) (inset). Data are representative of three independent experiments. b Uptake of opioid peptides (dynorphin A (1–13)-FAM (250 nM), big dynorphin-Cy5 (400 nM), BAM22-Cy5 (400 nM) or nociceptin-FAM (1 µM)) by U87 cells (NT) or U87 cells transfected with ACKR3 or classical opioid receptors analyzed by imaging flow cytometry as described in a. c ACKR3-mediated depletion of extracellular dynorphin A. U87 or U87-ACKR3 cells pretreated with LIH383 (400 nM) or LIH383ctrl were incubated with dynorphin A. Cell supernatant was added on U87 cells expressing KOR-SmBiT and LgBiT-beta-arrestin-1. (inset): EC₅₀ values. d Cellular localization of ACKR3 and classical opioid receptors fused to Neongreen fluorescent protein (green) stimulated or not by opioid peptides (1 µM) monitored by fluorescent confocal microscopy. Nuclear DNA was Hoechst-stained (blue). Pictures are representative of 10 acquired images from two independent experiments. Scale bar: 5 µm. e Ligand-induced receptor-arrestin delivery to endosomes monitored by β-galactosidase complementation assay in U2OS cells stably expressing ACKR3. Results are expressed as mean ± SD of three technical replicates. f–h Kinetics of ligand-induced internalization of ACKR3 and classical opioid receptors (f and h, HiBiT technology) and (g, flow cytometry). f U87 cells expressing N-terminally HiBiT-tagged receptors were stimulated with opioid peptides (1 µM) or CXCL12 (300 nM) for indicated times. Remaining membrane receptors were quantified with soluble LgBiT protein. g ACKR3 and KOR internalization and recycling in U87 cells after ligand stimulation (1 µM for opioid peptides, 300 nM for CXCL12) followed by acid wash, monitored by flow cytometry. h Effect of bafilomycin A1 (1.5 µM) on endosomal trafficking/cycling of ACKR3 and KOR following ligand stimulation (1 µM) monitored by HiBiT technology. If not otherwise indicated, results are presented as mean ± S.E.M of three to five independent experiments (n = 3 to 5). *p < 0.05, **p < 0.01, ***p < 0.001 by one-way ANOVA with Bonferroni correction (b), by two-way ANOVA: interaction between cell line and LIH383 treatment with Tukey’s post hoc test (c), and by two-tailed unpaired t-test (h). Source data and statistical analysis parameters are provided as Source Data file.

Fig. 7. ACKR3-mediated regulation of opioid peptide availability for classical receptors.

a–c Relative gene expression of ACKR3 and classical opioid receptors in smNPCs and different brain regions corresponding to centers important for opioid peptide activity. Box plots encompass 25th to 75th percentile with median as central mark and whiskers extending to most extreme data points. a RNA-Seq RPKM (reads per kilobase per million) values from the open-source brainspan.org database (n = 16 for amygdaloid complex and mediodorsal nucleus of thalamus, n = 15 for hippocampus and n = 17 for anterior cingulate cortex). b, c mRNA expression determined by qPCR on five human adult brains (b) or two preparations of smNPCs (c) normalized to the arithmetic mean of PPIA and GAPDH as housekeeping genes. d Extracellular and intracellular expression of ACKR3 monitored by flow cytometry using ACKR3-specific mAb antibody (11G8, red) or a matched isotype control (MG1-45, gray) and a PE-conjugated secondary antibody in comparison to unstained cells (filled light gray), representative of two experiments with 10,000 gated cells. e Uptake of 250 nM (FAM)-labeled dynorphin A (1–13) by smNPCs pretreated with LIH383 or LIH383ctrl (3 µM) analyzed by imaging flow cytometry. f SRE (ERK1/2) signaling cascade activation in smNPC in response to various ligands (500 nM) or 10% FBS as positive control. g ACKR3-mediated depletion of extracellular dynorphin A. SmNPCs, pretreated with LIH383, LIHctrl (1.5 µM), CXCL12 or CXCL10 (300 nM), were incubated with dynorphin A (3 µM) and the activity of dynorphin A remaining in the cell supernatants was probed on U87 cells expressing SmBiT-tagged KOR and LgBiT-tagged mini Gi. Representative 30x supernatant dilution is shown. h, i Ex vivo rat locus coeruleus inhibition of neuronal firing induced by LIH383 alone, dynorphin A in the absence or presence of naloxone (h) or increasing concentrations of dynorphin A in the presence or absence of LIH383 (i) (inset): EC₅₀ values. For e–i, data are presented as mean ± S.E.M of independent experiments (n = 3 to 5 except in g for LIH and LIHctrl (n = 8) and all animal experiments (h and i) (n = 6)). *p < 0.05, **p < 0.01 by one-way ANOVA with Bonferroni correction (e and g) and Kruskal–Wallis with two-sided Dunn’s test (i). Source data and statistical analysis parameters are provided as Source Data file.