A G protein-coupled receptor and the intracellular synthase of its agonist functionally cooperate

Abstract

Export of newly synthesized G protein-coupled receptors (GPCRs) remains poorly characterized. We show in this paper that lipocalin-type prostaglandin D2 (PGD2) synthase (L-PGDS) interacts intracellularly with the GPCR DP1 in an agonist-independent manner. L-PGDS promotes cell surface expression of DP1, but not of other GPCRs, in HEK293 and HeLa cells, independent of L-PGDS enzyme activity. In addition, formation of a DP1-Hsp90 complex necessary for DP1 export to the cell surface is dependent on the interaction between L-PGDS and the C-terminal MEEVD residues of Hsp90. Surprisingly, PGD2 synthesis by L-PGDS is promoted by coexpression of DP1, suggesting a possible intracrine/autocrine signaling mechanism. In this regard, L-PGDS increases the formation of a DP1-ERK1/2 complex and increases DP1-mediated ERK1/2 signaling. Our findings define a novel cooperative mechanism in which a GPCR (DP1) promotes the activity of the enzyme (L-PGDS) that produces its agonist (PGD2) and in which this enzyme in turn acts as a cofactor (of Hsp90) to promote export and agonist-dependent activity of the receptor.

Figure 1.

L-PGDS colocalizes intracellularly with DP1. (A) HEK293 cells were transiently transfected with a pcDNA3-Flag-DP1 construct for 48 h. The cells were fixed and prepared for confocal microscopy as described under Materials and methods. DP1 was visualized using Flag-specific monoclonal and Alexa Fluor 488–conjugated anti–mouse IgG antibodies (green). Endogenous PDI was detected using a PDI-specific polyclonal antibody and Alexa Fluor 546–conjugated anti–rabbit IgG (red). An overlay of staining patterns of green-labeled DP1 and red-labeled PDI (merge) and the corresponding fluorogram are shown. (B) HEK293 cells were transiently transfected with pcDNA3-Flag-DP1, pcDNA3–L-PGDS–HA, or both for 48 h. The cells were then fixed and prepared for confocal microscopy as in A. L-PGDS was visualized using HA-specific monoclonal and Alexa Fluor 633–conjugated anti–mouse IgG antibodies (red). DP1 was visualized using a Flag-specific polyclonal and Alexa Fluor 488–conjugated anti–rabbit IgG antibodies (green). An overlay of staining patterns of the green-labeled DP1 and red-labeled L-PGDS (merge) and the corresponding fluorogram are shown. Bars, 10 µM.

Figure 2.

Both L-PGDS and DP1 localize to the TGN. HEK293 cells were transiently transfected with pcDNA3-Flag-DP1 or pcDNA3–L-PGDS–HA for 48 h. The cells were then fixed and prepared for confocal microscopy as described under Materials and methods. L-PGDS was visualized using HA-specific monoclonal and Alexa Fluor 546–conjugated anti–mouse IgG antibodies (orange). DP1 was visualized using Flag-specific polyclonal and Alexa Fluor 633–conjugated anti–mouse IgG antibodies (red). Endogenous TGN was detected using a TGN46-specific polyclonal antibody and Alexa Fluor 488–conjugated anti–rabbit IgG (green). An overlay of staining patterns of the green-labeled TGN46 and orange-labeled L-PGDS or red-labeled DP1 (merge) and the corresponding fluorogram are shown. Bars, 10 µM.

Figure 3.

L-PGDS interacts with DP1. (A) Binding assays were performed using purified glutathione-Sepharose–bound GST-DP1–carboxyl terminal (CT) and intracellular loops (ICL) incubated with purified His₆–L-PGDS. The binding of L-PGDS to the receptor domains was detected by immunoblotting (IB) using an L-PGDS–specific polyclonal antibody, and the GST fusion proteins present in the binding reaction were detected using an anti-GST antibody. Blots shown were spliced but came from the same gels. (B) HEK293 cells transiently transfected with Flag-DP1 and L-PGDS–HA were stimulated or not stimulated with 1 µM PGD₂ for 5 or 30 min. Immunoprecipitation (IP) of the receptor was performed using a Flag-specific monoclonal antibody, and immunoblotting was performed with Flag-specific polyclonal or peroxidase-conjugated anti-HA antibodies. Black lines indicate that intervening lanes have been spliced out. (C) Immunoprecipitation was performed in HT-29 cells using L-PGDS–specific monoclonal or rat isotypic control IgG antibodies, and immunoblotting was performed using L-PGDS–specific polyclonal or DP1-specific polyclonal antibodies. Blots shown are representative of three independent experiments.

Figure 4.

L-PGDS regulates the cell surface expression of DP1 but not of other tested GPCRs. (A) Cell surface expression of the receptors was measured by ELISA in cells transiently transfected for 48 h with Flag-tagged receptors in combination with pcDNA3, pcDNA3–L-PGDS–HA, or pcDNA3–H-PGDS–HA. Results are shown as the percentage of cell surface expression of the receptors when cells were transfected with L-PGDS or H-PGDS compared with cell surface expression of the receptors when they were transfected with pcDNA3. (B) HeLa cells were transfected with negative control or L-PGDS siRNAs for 72 h. Immunoprecipitation (IP) of L-PGDS was performed using an L-PGDS–specific monoclonal antibody, and immunoblotting (IB) was performed with an L-PGDS–specific polyclonal antibody. (C) Cell surface expression of stably transfected HA-DP1 was measured by ELISA in HeLa transfected with negative control or L-PGDS siRNAs for 72 h. (D) Agonist-induced internalization of DP1 was studied in HEK293 cells after 15 and 30 min of stimulation with 1 µM PGD₂. (E) HEK293 cells were transiently transfected with the indicated combinations of pcDNA3, Flag-DP1, L-PGDS–HA, or Flag-CRTH2 for 48 h. Cells were then incubated with 5 µM PGH₂ for 15 min. Supernatants were assessed for PGD₂ production by commercial enzyme-linked immunoassays as described under Materials and methods. (F) Cell surface expression of stably expressed Flag-DP1 in HeLa cells transiently transfected with L-PGDS–HA alone, in combination with dynamin-K44A, or treated with BWA868C was measured by ELISA. All values are the means ± SE from at least three separate experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

The enzyme activity of L-PGDS is not required for regulation of DP1 export. Cell surface expression of DP1 was measured by ELISA in HEK293 cells transiently transfected for 48 h with Flag-DP1 in combination with pcDNA3, pcDNA3–L-PGDS–HA, or pcDNA3–L-PGDS–C65A-HA. Results are shown as the percentage of cell surface expression of DP1 when cells were transfected with L-PGDS or L-PGDS–C65A compared with cell surface expression of DP1 when transfected with pcDNA3. All values are the means ± SE from at least three separate experiments. ***, P < 0.001.

Figure 6.

L-PGDS interacts with Hsp90. (A) HEK293 cells were transiently transfected with pcDNA3, L-PGDS–Myc, Hsp90-HA, or Hsp90ΔMEEVD-HA mutant constructs. Immunoprecipitation (IP) of L-PGDS was performed using a Myc-specific monoclonal antibody, and immunoblotting (IB) was performed using peroxidase-conjugated anti-Myc or anti-HA antibodies. (B) Immunoprecipitation was performed in HeLa cells using Hsp90-specific monoclonal or mouse isotypic control IgG antibodies, and immunoblotting was performed using Hsp90-specific polyclonal or L-PGDS–specific polyclonal antibodies. (C) His pull-down assays were performed using purified His-Hsp90 bound to Ni²⁺-agarose beads incubated with purified GST–L-PGDS. The binding of L-PGDS to Hsp90 was detected by immunoblotting using an L-PGDS–specific polyclonal antibody, and the His-Hsp90 present in the binding reaction were detected using an Hsp90-specific polyclonal antibody. (D) Illustration of the funnel-shaped β barrel structure of L-PGDS (available in the Protein Data Bank under accession no. 2E4J). On the outside of the funnel near the bottom, a surface was identified that had basic residues (Arg42, Lys66, Arg151, Lys156, and Lys160) surrounding a sunken hydrophobic site (Trp43, Tyr44, Ala46, and Gly47) under a C-terminal loop. (E) HEK293 cells were transiently transfected with pcDNA3, L-PGDS–HA, or its W43A/G47A-HA mutant construct. Immunoprecipitation of L-PGDS was performed using a HA-specific monoclonal antibody, and immunoblotting was performed using a peroxidase-conjugated anti-HA antibody. Endogenous Hsp90 was detected using an Hsp90-specific polyclonal antibody. Blots shown are representative of three independent experiments. NTA, nitrilotriacetic acid.

Figure 7.

L-PGDS promotes the interaction between DP1 and Hsp90. (A) HEK293 cells were transiently transfected with the indicated combinations of pcDNA3, Flag-DP1, L-PGDS-HA, Hsp90-HA, or Hsp90ΔMEEVD-HA mutant constructs. Immunoprecipitation (IP) of the receptor was performed using a Flag-specific monoclonal antibody, and immunoblotting (IB) was performed with Flag-specific polyclonal or peroxidase-conjugated anti-HA antibodies. (B) HEK293 cells were transiently transfected with Flag-DP1 together with pcDNA3, L-PGDS–HA, or L-PGDS–W43A/G47A-HA. Immunoprecipitation of the receptor was performed using a Flag-specific monoclonal antibody, and immunoblotting was performed using peroxidase-conjugated anti-HA or polyclonal anti-Flag antibodies. Endogenous Hsp90 was detected using an Hsp90-specific polyclonal antibody. Blots shown are representative of three separate experiments. Blots shown were spliced but came from the same gels. (C) Cell surface expression of DP1 was measured by ELISA in HEK293 cells transiently transfected for 48 h with Flag-DP1 together with pcDNA3, pcDNA3–L-PGDS–HA, or pcDNA3–L-PGD–W43A/G47A-HA. Results are shown as the percentage of cell surface expression of DP1 when cells were transfected with L-PGDS or L-PGDS–W43A/G47A compared with cell surface expression of DP1 when transfected with pcDNA3. (D) HEK293 cells were transiently transfected for 48 h with Flag-DP1 together with pcDNA3 or pcDNA3–L-PGDS–HA, treated with 10 µM geldanamycin over a time course of 2 h, and cell surface expression of DP1 was measured by ELISA. All values are the means ± SE from at least three separate experiments. *, P < 0.05; **, P < 0.01.

Figure 8.

Intracellular distribution and PGD₂ synthase activity of the L-PGDS–W43A/G47A mutant. (A) HEK293 cells were transiently transfected with pcDNA3–L-PGDS–HA or pcDNA3–L-PGDS–W43A/G47A-HA for 48 h. The cells were then fixed and prepared for confocal microscopy as described under Materials and methods. The L-PGDS constructs were visualized using HA-specific monoclonal and Alexa Fluor 546–conjugated anti–mouse IgG antibodies (orange). An overlay of staining patterns of the orange-labeled L-PGDS or its mutant and the nuclei are shown (merge). Bars, 10 µM. (B) PGD₂ production by purified GST, GST–L-PGDS, and GST–L-PGDS–W43A/G47A was measured in vitro in the presence of 0.5 µM PGH₂ for 1 min. The reactions were stopped with 0.4 mg/ml SnCl₂, and PGD₂ was measured with commercial enzyme-linked immunoassays as described under Materials and methods. (C) HEK293 cells were transiently transfected with pcDNA3, L-PGDS, or L-PGDS–W43A/G47A for 48 h. Cells were then incubated with 5 µM PGH₂ for 15 min. Supernatants were assessed for PGD₂ production by commercial enzyme-linked immunoassays as described under Materials and methods. All values are the means ± SE from at least three separate experiments. wt, wild type. ***, P < 0.001; ****, P < 0.0001.

Figure 9.

L-PGDS promotes ERK1/2 activation. (A) HeLa cells were transfected with the indicated siRNA for 48 h, serum starved overnight, and stimulated or not stimulated with 1 µM PGD₂ (left) or 10 µM isoproterenol (right) for the indicated times. siCTL, control siRNA. (B) HEK293 cells were transfected with pcDNA3, L-PGDS, L-PGDS–C65A, DP1 (left), or β2AR (right) for 24 h, serum starved overnight, and stimulated or not stimulated with 1 µM PGD₂ (left) or 10 µM isoproterenol (right) for 5 min. All assays were performed as described in Materials and methods. Protein levels were assessed by Western blotting using p-ERK1/2 and ERK1/2 antibodies. Bar graphs show densitometry analyses performed on four different experiments. Phospho-ERK1/2 pixels were normalized to total ERK1/2 pixels, and results are presented as the fold of these values (means ± SE) over that of the first lane, which was arbitrarily set as 1. *, P < 0.01; **, P < 0.005.

Figure 10.

L-PDS is part of a DP1–ERK1/2 signaling complex in the perinuclear region. (A) HeLa cells were transiently transfected with pcDNA3 or pcDNA3–L-PGDS–HA for 24 h, serum starved overnight, and treated with 100 µM AT-56 or 1 µM BWA868C for 60 min. The cells were then prepared for confocal microscopy as described under Materials and methods. L-PGDS was visualized using HA-specific monoclonal and Alexa Fluor 488–conjugated anti–mouse IgG antibodies (green). Endogenous p-ERK1/2 was detected using a p-ERK1/2–specific polyclonal antibody and Alexa Fluor 546–conjugated anti–rabbit IgG (orange). An overlay of staining patterns of the orange-labeled p-ERK1/2 and green-labeled L-PGDS and the corresponding fluorograms are shown. Bars, 10 µM. (B) Mean pixel integration of p-ERK1/2 of ≥10 different cells was determined using the FluoView 2.0 software (Olympus). (C) The same L-PGDS immunoprecipitation (IP) reaction as in Fig. 3 C performed using L-PGDS–specific monoclonal or isotypic control IgG antibodies was used, and immunoblotting (IB) was performed using p-ERK1/2– or ERK1/2-specific polyclonal or L-PGDS–specific polyclonal antibodies. (D) HEK293 cells were transiently transfected with pcDNA3, L-PGDS–HA, or Flag-DP1 constructs. Immunoprecipitation of DP1 was performed using a Flag-specific monoclonal antibody, and immunoblotting was performed using peroxidase-conjugated anti-HA– or p-ERK1/2–specific polyclonal antibodies. Black lines indicate that intervening lanes have been spliced out. Bar graph shows densitometry analyses performed on four different experiments. ERK1/2 pixels were normalized on the receptor pixels, and results are presented as the fold of these values over that of the third lane (means ± SE), which was arbitrarily set as 1. **, P < 0.005; ***, P < 0.0005.