Research ArticleBiochemistry

FZD5 is a Gαq-coupled receptor that exhibits the functional hallmarks of prototypical GPCRs

See allHide authors and affiliations

Science Signaling  04 Dec 2018:
Vol. 11, Issue 559, eaar5536
DOI: 10.1126/scisignal.aar5536
  • Fig. 1 Agonist-induced conformational dynamics in FZD5 measured by FRET.

    (A) Models of FZD5 in the active and inactive conformations. The model of the inactive conformation (gray) is based on the structure of SMO in an inactive state (PDB ID: 5V56). The model of active FZD5 was generated using inactive SMO combined with information on the conformation of transmembrane helix 6 (TM6, green) from the Gs-bound structure of the β2AR (PDB ID: 3SN6). The predicted movement of TM6 from the inactive to the active conformation is highlighted with a red arrow. (B) Illustration depicting FZD5 constructs used in FRET experiments. All constructs were cloned with an N-terminal V5 tag into CFP N1 vectors. The black line represents unedited FZD5 sequence. For V5-FZD5-FlAsH436-CFP, the FlAsH-binding motif (CCPGCC) was inserted into IL3 at position 436. In the absence of a ligand, there is a basal energy transfer from CFP to FlAsH. Conformational changes in the receptor upon ligand binding alter the relative distance between FlAsH and CFP resulting in a change in FRET. (C) Representative confocal images showing CFP fluorescence in HEK293 cells expressing V5-FZD5-CFP or V5-FZD5-FlAsH436-CFP. n = 3 independent experiments. Scale bars, 25 μm. (D) FRET efficiency measurements from single cells expressing only V5-FZD5-FlAsH436-CFP or both V5-FZD5-FlAsH436 and V5-FZD5-CFP. Addition of the compound BAL displaces FlAsH from its binding site. Fluorescence of liganded FlAsH is shown in yellow, CFP fluorescence in blue, and the FRET ratio (FlAsH/CFP) in red. Data are representative of n = 3 independent experiments. a.u., arbitrary units. (E) The FRET efficiency for V5-FZD5-FlAsH436-CFP for individual cells was calculated and shown as a scatter plot. Data are represented as means ± SEM. n = 16 individual cells from three independent experiments. (F) The FRET ratio (FlAsH/CFP) from a representative cell is shown normalized to the first measured data point. Single cells were stimulated with saturating concentrations of recombinant WNT-5A approximately 10 to 15 s after beginning the recording. (G) FRET ratio (FlAsH/CFP) in HEK293 cells expressing V5-FZD5-FlAsH436-CFP in a microplate format. Cells were stimulated with the indicated concentrations of recombinant WNT-5A. The FRET change induced by each concentration has been corrected for the signal obtained from vehicle-treated cells. n = 3 independent experiments. Data are represented as means ± SEM.

  • Fig. 2 FZD5 in the inactive state forms a complex with Gαq.

    (A) Representative image of a HEK293 cell expressing V5-FZD5-mCherry and Gαq-Venus. These cells were used for dcFRAP experiments. Scale bar, 5 μm. (B and C) Recovery of mCherry (B) and Venus (C) fluorescence after photobleaching (FRAP) in the absence (gray) and presence (red) of surface cross-linking. The first measurement after photobleaching is time = 0. (D) Bar graph showing the fluorescence intensity averages of the mobile fraction of V5-FZD5-mCherry and Gαq-Venus before and after cross-linking (CL) within the time frame of 85 to 101 s (including 15-s prebleach measurements) as obtained by dcFRAP. White bars, V5-FZD5-mCherry; gray bars, Gαq-Venus; hatching, cross-linking. n = 52 regions of interest (ROIs) before cross-linking; n = 66 ROIs after cross-linking from three independent experiments. (E) Representative image of a HEK293 cell expressing V5-FZD5-mCherry and Gαi1-GFP. (F and G) mCherry (F) and GFP (G) FRAP in the absence (gray) and presence (red) of surface cross-linking. (H) Bar graph showing the fluorescence intensity averages of the mobile fraction of V5-FZD5-mCherry and Gαi1-GFP before and after cross-linking. White bars, V5-FZD5-mCherry; gray bars, Gαi1-GFP; hatching, cross-linking. n = 41 ROIs before cross-linking; n = 41 ROIs after cross-linking from three independent experiments. Data are represented as means ± SEM. ***P < 0.0001; ns, not significant (two-tailed t test).

  • Fig. 3 WNT-5A induces an FZD5-dependent structural rearrangement of the Gαq-Gβ1γ1 interface.

    (A and B) BRET experiments in HEK293 cells coexpressing either Gαq-118-RlucII and GFP10-Gγ1 (A) or Gαi1-118-RlucII and GFP10-Gγ2 (B) in the presence or absence of SNAP-FZD5. Cells were stimulated with the indicated concentrations of recombinant WNT-5A for 5 min before measuring the ratio between GFP10 and RlucII. Data are represented as means ± SEM of n = 3 to 6 independent experiments conducted at least in duplicate. (C and D) FRET in HEK293 cells expressing Gαq-127-mTqΔ6, Gβ1, and cpVenus-Gγ2 with or without V5-FZD5 and stimulated with recombinant WNT-5A approximately 15 to 20 s beginning of the recording. FRET ratios (Venus/Turquoise) from a representative single cell stimulated with recombinant WNT-5A (1000 ng/ml) (C) and from 96-well plates (D) are shown. Data are represented as means ± SEM of n = 3 independent experiments conducted in quadruplicate. (E and F) FRET experiments in HEK293 cells expressing Gαi1-121-mTq2, Gβ1, and cpVenus-Gγ2 with and without V5-FZD5 and stimulated with recombinant WNT-5A approximately 15 to 20 s at the beginning of the recording. FRET ratios (Venus/Turquoise) from a representative single cell stimulated with recombinant WNT-5A (1000 ng/ml) (E) and from 96-well plates stimulated with the indicated concentrations of WNT-5A. Data are represented as means ± SEM of n = 3 independent experiments conducted at least in quadruplicate.

  • Fig. 4 WNT-5A–mediated stimulation of FZD5 induces Gαq-dependent downstream signaling.

    (A) BRET experiments in HEK293 cells expressing the DAG biosensor (Mem-GFP10-RlucII-C1b) alone or with SNAP-FZD5. Cells were stimulated with the indicated concentrations of recombinant WNT-5A before measuring the ratio between GFP10 and RlucII fluorescence. Data are represented as means ± SEM of n = 3 to 6 independent experiments conducted at least in duplicate. (B) BRET experiments in HEK293 cells expressing the PKC biosensor (GFP10-FHA1-FHA2-pPKC2-pPKC1-RlucII-C1b) alone or with SNAP-FZD5. Cells were stimulated with the indicated concentrations of recombinant WNT-5A before measuring the ratio between GFP10 and RlucII fluorescence. Data are represented as means ± SEM of n = 3 to 6 independent experiments conducted at least in duplicate. (C) HEK293 cells expressing the PKC biosensor and SNAP-FZD5 were pretreated with the Gαq inhibitor YM-294890 or the PKC inhibitor Gö 6983 before being stimulated with WNT-5A (2000 ng/ml). Data are expressed as the difference in BRET ratio with and without ligand (ΔBRET). Data are represented as means ± SEM of n = 3 to 10 independent experiments conducted at least in duplicate. **P < 0.01 and ***P < 0.0001 [one-way analysis of variance (ANOVA) with Fisher’s least significant difference (LSD) post hoc analysis]. DMSO, dimethyl sulfoxide. (D) BRET experiments in parental and Gαq/11 KO HEK293 cells expressing the PKC biosensor and SNAP-FZD5. Cells were stimulated with WNT-5A (2000 ng/ml) in the presence or absence of a transgene encoding Gαq. Mock treatment transfection with empty vector DNA instead of Gαq. Data are expressed as the fold change in BRET over vehicle. Data are represented as means ± SEM of n = 5 independent experiments conducted at least in duplicate. *P < 0.05 and **P < 0.01 (two-tailed t test).

  • Fig. 5 WNT-5A–induced, FZD5-mediated changes in cellular DMR depend on Gαq activation.

    DMR was measured by changes in the wavelength of light reflected from cells (pm) after stimulation of parental or transgenic HEK293 cells with the indicated concentrations of WNT-5A after at least 200 s of baseline read. (A and B) DMR recordings from HEK293 cells expressing V5-FZD5-CFP in the absence (A) and presence (B) of the Gαq inhibitor FR900359. (C and D) Control experiments in parental HEK293 cells in the absence (C) and presence (D) of FR900359. Representative traces that were buffer-corrected and measured in triplicate are shown + SEM. (E) AUC data are represented as means ± SEM of n = 4 to 5 independent experiments from 0 to 2000 s. *P < 0.05, **P < 0.01, and ***P < 0.0001 (two-way ANOVA with Dunnett’s multiple comparisons test post hoc analysis).

  • Fig. 6 WNT-5A–FZD–Gαq signaling in PDACs.

    (A) Intracellular Ca2+ in HPAF-II cells was measured after the application of recombinant WNT-5A in the presence and absence of the Gαq inhibitor YM-254890. RFU, relative fluorescence units. (B) Quantification and statistical analysis of AUC from experiments in (A). Data are represented as means ± SEM of n = 3 independent experiments conducted in duplicate. **P < 0.01 (two-tailed t test). (C and D) Cell viability assays in HPAF-II and PANC-1 cells using the indicated concentrations of the PORCN inhibitor C59 and the Gαq inhibitor FR900359. Data are represented as means ± SEM of n = 3 to 4 independent experiments conducted in triplicate. *P < 0.05 (two-tailed t test).

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/11/559/eaar5536/DC1

    Fig. S1. Alignment of mouse FZD5 with the SMO and β2AR sequences corresponding to the crystal structures used for modeling.

    Fig. S2. Validation of V5-FZD5-FlAsH439-CFP.

    Fig. S3. Individual traces from single-cell data and kinetic experiments using V5-FZD5-FlAsH436-CFP.

    Fig. S4. Validation of Gαq and Gαi1 FRET probes with FZD5.

    Fig. S5. The FZD5 FRET sensors activate Gαq but do not recruit DVL.

    Fig. S6. Control experiments for DMR in HEK293 cells.

    Fig. S7. WNT-5A does not activate the WNT–β-catenin pathway in HPAF-II cells.

    Fig. S8. Pharmacological properties of receptor activation and stimulation of downstream signaling.

  • This PDF file includes:

    • Fig. S1. Alignment of mouse FZD5 with the SMO and β2AR sequences corresponding to the crystal structures used for modeling.
    • Fig. S2. Validation of V5-FZD5-FlAsH439-CFP.
    • Fig. S3. Individual traces from single-cell data and kinetic experiments using V5-FZD5-FlAsH436-CFP.
    • Fig. S4. Validation of Gαq and Gαi1 FRET probes with FZD5.
    • Fig. S5. The FZD5 FRET sensors activate Gαq but do not recruit DVL.
    • Fig. S6. Control experiments for DMR in HEK293 cells.
    • Fig. S7. WNT-5A does not activate the WNT–β-catenin pathway in HPAF-II cells.
    • Fig. S8. Pharmacological properties of receptor activation and stimulation of downstream signaling.

    [Download PDF]

Stay Connected to Science Signaling

Navigate This Article