Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May;26(9):2413-2434.
doi: 10.1038/s44319-025-00428-2. Epub 2025 Mar 24.

Structural insights into the selective recognition of RF-amide peptides by neuropeptide FF receptor 2

Jeesoo Kim  1   2 , Sooyoung Hong  1 , Hajin Lee  3 , Hyun Sik Lee  1 , Chaehee Park  1   4 , Jinuk Kim  1   5 , Wonpil Im  6 , Hee-Jung Choi  7
Affiliations

Structural insights into the selective recognition of RF-amide peptides by neuropeptide FF receptor 2

Jeesoo Kim et al. EMBO Rep. 2025 May.

Abstract

Neuropeptide FF Receptor 2 (NPFFR2), a G-protein-coupled receptor, plays a role in pain modulation and diet-induced thermogenesis. While NPFFR2 is strongly activated by neuropeptides FF (NPFFs), it shows low activity in response to RF-amide-related peptides (RFRPs), despite the peptides belonging to a shared family. In contrast, NPFFR1, which shares high sequence similarity with NPFFR2, is activated by RFRPs and regulates reproductive hormone balance. The molecular basis for these receptor-specific interactions with their RF-amide peptides remains unclear. Here, we present cryo-electron microscopy structures of NPFFR2 in its active state bound to the agonist RF-amide peptide hNPSF, and in its ligand-free state. Structural analysis reveals that the C-terminal RF-amide moiety engages conserved residues in the transmembrane domain, while the N-terminal segment interacts in a receptor subtype-specific manner. Key selectivity-determining residues in NPFFR2 are also identified. A homology model of NPFFR1 bound to RFRP, supported by mutagenesis studies, further validates this selectivity mechanism. Additionally, structural comparison between the inactive and active states of NPFFR2 suggests a TM3-mediated activation mechanism. These findings provide insights into RF-amide peptide recognition by NPFF receptors.

Keywords: Ligand Selectivity; Neuropeptide FF Receptors; RF-amide Peptide; cryo-EM Structure.

PubMed Disclaimer

Conflict of interest statement

Disclosure and competing interests statement. The authors declare no competing interests.

Figures

Figure 1
Figure 1. Structural analysis and downstream signaling of NPFFR2.
(A) The concentration-dependent response of various agonists on NPFFR2 is measured using a forskolin-stimulated cAMP assay. Data points represent mean values from three independent experiments (each with technical triplicate), with error bars indicating SEM (standard error of the mean). Raw data were normalized to the vehicle-treated (0%) and 0.5 μM forskolin-treated (100%) responses. Emax and EC50 values along with their respective errors are summarized in Table EV1. (B) The overall cryo-EM electron density map and structural model of the hNPSF–NPFFR2–Gαiβγ–scFv16 complex are presented, with the following color scheme: hNPSF (yellow), NPFFR2 (green), Gαi (cyan), Gβ (orange), Gγ (purple), and scFv16 (gray). (C) The extracellular loop 2 (ECL2, pink) and N-terminus (Nt, blue) of NPFFR2 are depicted using a cartoon model. They form stable β-sheets. Source data are available online for this figure.
Figure EV1
Figure EV1. Sample preparation and cryo-EM data analysis of the hNPSF–NPFFR2–Gi complex.
(A, B) Size exclusion chromatography and SDS-PAGE profile of hNPSF–NPFFR2–Gi complex used in the cryo-EM sample preparation. (C) Representative image of cryo-EM micrograph (scale bar: 140 nm) and 2D classification results. (D) Cryo-EM data processing workflow using CryoSPARC. (E) Gold standard Fourier shell correlation (GSFSC) curve and direction distribution profile. (F) Representative image of electron density map and model of the hNPSF–NPFFR2–Gi complex. Source data are available online for this figure.
Figure EV2
Figure EV2. Comparison of the agonist binding pockets of NPFFR2 and NPY2R.
(A) Superposition of the hNPSF–NPFFR2–Gi complex and the NPY–NPY2R–Gi complex (PDB: 7YOO). hNPSF is colored yellow, NPFFR2 is green, NPY is light purple, and NPY2R is pink. The structures are aligned based on the receptors, with an overall RMSD of 0.9 Å. (B) A detailed view at the bottom of the ligand-binding pocket of NPFFR2 and NPY2R reveals that two C-terminal residues of hNPSF and NPY share similar binding poses. NPY2R residues involved in the interaction toward Arg-Tyr-NH2 and the corresponding residues in NPFFR2 are shown in sticks.
Figure 2
Figure 2. Interaction of hNPSF with NPFFR2.
(A) Residues participating in the interactions between hNPSF and NPFFR2 in the transmembrane (TM) binding pocket are shown as sticks. Dashed lines indicate polar interactions and hydrogen bonds between residues. Zoomed-in views highlighting interactions of the C-terminal segment (residues −1 to −4, PQRF), middle segment (residues −5 to −8, FLFQ), and the N-terminal segment (residues −9 to −11, SQA) of hNPSF are enclosed in red, blue, and yellow boxes, respectively. (B) NPFFR2 residues interacting with hNPSF were mutated to alanine, and the impact of each mutation on the EC50 value (ΔpEC50) was investigated using a forskolin-stimulated cAMP assay, after confirming surface expression of each mutant by an ELISA-based surface expression assay (Appendix Fig. S5). Three independent experiments (with technical triplicates) were conducted for each mutant. The bar graphs represent the mean of ΔpEC50 values from each experiment, with error bars indicating SEM. Emax and EC50 values along with their respective errors are summarized in Table EV1. Statistical analyses were performed using ordinary one-way ANOVA followed by Dunnett’s test, comparing to the wild-type response. ns (not significant, P  >  0.05); *P  <  0.05; **P  <  0.01; ***P  <  0.001. The exact P values are provided in Table EV1. Source data are available online for this figure.
Figure EV3
Figure EV3. Water-mediated interaction between hNPSF Q(−5) and NPFFR2.
(A) Water molecules near the extracellular regions of TMD are shown as red spheres in the all-atom MD simulation. (B, C) Water-mediated interactions involving Q(−5)SF, based on model structures from MD simulation frames. The ligand and receptor residues (R2165.35, S2976.58, N3117.31 and Y3157.35) form hydrogen bonds with water molecules, with N and O atoms positioned within 3.5 Å.
Figure 3
Figure 3. Structural comparison between RF-amide receptors.
(A) The endogenous ligand-bound structures of RF-amide receptors, including hNPSF-bound NPFFR2, QRFP26-bound QRFPR (PDB: 8ZH8), Kisspeptin-10-bound KISS1R (PDB: 8ZJD), and PrRP20-bound PrRPR (PDB:8ZPT) are shown. The unique ECL2 structures and ligand-binding modes are depicted with cartoon models: NPFFR2 (green), hNPSF (yellow), QRFP26 (blue-purple), QRFPR (light brown), Kisspeptin-10 (light cyan), KISS1R (salmon), PrRP20 (pale green), and PrRPR (orange). (B) Detailed views of the TM binding pocket of each RF-amide receptor show that these receptors share common binding mode while also exhibiting unique binding interactions with the C-terminal RF-amide motif of their respective ligands. (C) This table illustrates the residues of NPFFR2 and other RF-amide receptors (NPFFR1, QRFPR, KISS1R, and PrRPR) that interact with the RF-amide motif at the C-terminus of their respective ligands. For NPFFR1, the residues shown correspond to those interacting residues in NPFFR2. The shading represents the conservation score of each residue among the RF-amide receptors, calculated using the Valdar method, with darker gray indicating higher conservation. Negatively charged residues are shown in red, positively charged residues in blue, and uncharged residues in white or black.
Figure 4
Figure 4. RF-amide ligand selectivity of NPFFR2 and NPFFR1.
(A) Sequence alignment between human NPFFs and RFRPs, the endogenous ligands of NPFFRs, which share PXRF-NH2 motif at their C-terminus (shown in green letters). Key residues at the fifth and sixth positions from the C-terminus, are highlighted in bold. The alignment clearly shows that RFRPs have hydrophobic residues (black) at the fifth position and hydrophilic residues (red) at the sixth position, while NPFF and hNPSF have reversed hydrophobicity at these key positions. (B) The concentration-dependent response of various agonists on NPFFR1 was measured using a forskolin-stimulated cAMP assay. Data points represent the mean values from three independent experiments (each with technical triplicate), with error bars indicating SEM. Raw data were normalized to the vehicle-treated (0%) and 0.5 μM forskolin-treated (100%) responses. (C) Structural comparison between the binding modes of hNPSF to NPFFR2 and GnIH to NPFFR1. An AlphaFold2 prediction model of the GnIH–NPFFR1 complex was generated using the hNPSF–NPFFR2 complex structure as a template. The hydrophobicity of NPFFRs is shown with surface representation, where purple indicates hydrophilic regions and bronze indicates hydrophobic regions. The upper left panel shows a hydrophobic patch on the NPFFR2 surface where F(−6) of hNPSF binds, while the upper right panel depicts the corresponding hydrophilic surface region of NPFFR1, where N(−6) of GnIH is located. The lower left panel illustrates the hydrophilic surface near the water pocket close to the binding site of Q(−5) of hNPSF, while the lower right panel displays the corresponding hydrophobic region in NPFFR1 where L(−5) of GnIH binds, which is hydrophobic. (D) The cAMP response to the ligands hNPSF and GnIH was measured for both wild-type NPFFR1 and a mutant form of NPFFR1 using a forskolin-stimulated cAMP assay. The NPFFR1 mutant (Q37NtL, I2936.58S, and T3077.31N) shows an increased EC50 for hNPSF while exhibiting a decreased response to GnIH. Symbols represent mean values from three independent experiments, each conducted in technical triplicates, with error bars indicating SEM. Responses were normalized to vehicle-treated controls (0%) and 0.5 μM forskolin-treated controls (100%). Source data are available online for this figure.
Figure 5
Figure 5. Structural comparison between ligand-free and active states of NPFFR2.
(A) The overall cryo-EM density map and structure model of the NPFFR2-BRIL–Fab–Nb complex are displayed with the following color scheme: ligand-free NPFFR2 (orange), BRIL (cyan), heavy chains of anti-BRIL Fab (FabBRIL) (yellow), light chains of FabBRIL (purple), and anti-Fab nanobody (NbFab) (gray). (B) The superposition of active (green) and ligand-free NPFFR2 (orange) structures clearly shows conformational changes upon ligand binding. The movements of TM6 and TM7 at the extracellular region, along with the shifts of ECL2 and the N-terminus (Nt) toward the ligand, are indicated by arrows. The movements of TM6 and TM7 at the intracellular region, resulting from G-protein coupling, are also depicted with arrows. (C) The conformational changes in TM7, ECL2, and the Nt that trap the ligand are illustrated with arrows. (D) The movements of the residues at the bottom of the ligand-binding pocket including toggle switch (W6.48) and Q3.32 are indicated by arrows. (E) The conformational changes of residues near the DRY motif and NPxxY motif upon activation are indicated by arrows. The hydrogen bonds are indicated with dashed lines.
Figure EV4
Figure EV4. Sample preparation and cryo-EM data analysis of the NPFFR2-BRIL–FabBRIL –NbFab complex.
(A, B) Size exclusion chromatography and SDS-PAGE profile of the NPFFR2-BRIL–FabBRIL–NbFab complex used in the cryo-EM sample preparation. (C) Representative image of cryo-EM micrograph (scale bar: 180 nm) and 2D classification results. (D) Cryo-EM data processing workflow using CryoSPARC. (E) Gold standard Fourier shell correlation (GSFSC) curve and direction distribution profile. (F) Representative image of electron density map and model of the NPFFR2-BRIL–FabBRIL–NbFab complex. Source data are available online for this figure.
Figure 6
Figure 6. Molecular mechanism of subtype-specific RF-amide peptide recognition by NPFFRs.
Schematic representations of the hNPSF-bound NPFFR2 and GnIH-bound NPFFR1 are shown. While NPFFR2 and NPFFR1 share high sequence similarity, they exhibit distinct preferences for RF-amide peptide subfamilies, namely NPFFs and RFRPs. Structural comparison between the cryo-EM structure of hNPSF-bound NPFFR2 and the AlphaFold2-predicted structure of GnIH-bound NPFFR1 shows that both receptors share similar overall architecture, including a characteristic β-sheet formed by the N-terminus and ECL2, as well as conserved interactions with the C-terminal RF-amide motif of each peptide. However, a key difference is observed in the hydrophobicity of the receptor surface at the binding site for the fifth and sixth residues from the C-terminus of each peptide. This difference determines whether these NPFFRs prefer NPFFs or RFRPs.
None
None
See this image and copyright information in PMC

References

    1. Bepler T, Morin A, Rapp M, Brasch J, Shapiro L, Noble AJ, Berger B (2019) Positive-unlabeled convolutional neural networks for particle picking in cryo-electron micrographs. Nat Methods 16:1153–1160 - PMC - PubMed
    1. Courteix C, Coudore-Civiale MA, Privat AM, Zajac JM, Eschalier A, Fialip J (1999) Spinal effect of a neuropeptide FF analogue on hyperalgesia and morphine-induced analgesia in mononeuropathic and diabetic rats. Br J Pharm 127:1454–1462 - PMC - PubMed
    1. De Neve J, Elhabazi K, Gonzalez S, Herby C, Schneider S, Utard V, Fellmann-Clauss R, Petit-Demouliere N, Lecat S, Kremer M et al (2024) Multitarget mu-Opioid receptor agonists horizontal line neuropeptide FF receptor antagonists induce potent antinociception with reduced adverse side effects. J Med Chem 67:7603–7619 - PubMed
    1. Dion M, Rydberg H, Schroder E, Langreth DC, Lundqvist BI (2004) van der Waals density functional for general geometries. Phys Rev Lett 92:246401 - PubMed
    1. Elshourbagy NA, Ames RS, Fitzgerald LR, Foley JJ, Chambers JK, Szekeres PG, Evans NA, Schmidt DB, Buckley PT, Dytko GM et al (2000) Receptor for the pain modulatory neuropeptides FF and AF is an orphan G protein-coupled receptor. J Biol Chem 275:25965–25971 - PubMed

Substances

LinkOut - more resources