Subtype-specific differences in corticotropin-releasing factor receptor complexes detected by fluorescence spectroscopy

Abstract

G protein-coupled receptors have been proposed to exist in signalosomes subject to agonist-driven shifts in the assembly disassembly equilibrium, affected by stabilizing membrane lipids and/or cortical actin restricting mobility. We investigated the highly homologous corticotropin-releasing factor receptors (CRFRs), CRFR1 and -2, which are different within their hydrophobic core. Agonist stimulation of CRFR1 and CRFR2 gave rise to similar concentration-response curves for cAMP accumulation, but CRFR2 underwent restricted collision coupling. Both CRFR1 and CRFR2 formed constitutive oligomers at the cell surface and recruited beta-arrestin upon agonist activation (as assessed by fluorescence resonance energy transfer microscopy in living cells). However, CRFR2, but not CRFR1, failed to undergo agonist-induced internalization. Likewise, agonist binding accelerated the diffusion rate of CRFR2 only (detected by fluorescence recovery after photobleaching and fluorescence correlation spectroscopy) but reduced the mobile fraction, which is indicative of local confinement. Fluorescence intensity distribution analysis demonstrated that the size of CRFR complexes was not changed. Disruption of the actin cytoskeleton abolished the agonist-dependent increase in CRFR2 mobility, shifted the agonist concentration curve for CRFR2 to the left, and promoted agonist-induced internalization of CRFR2. Our observations are incompatible with an agonist-induced change in monomer-oligomer equilibrium, but they suggest an agonist-induced redistribution of CRFR2 into a membrane microdomain that affords rapid diffusion but restricted mobility and that is stabilized by the actin cytoskeleton. Our data show that membrane anisotropy can determine the shape and duration of receptor-generated signals in a subtype-specific manner.

Fig. 1

Accumulation of cAMP in HEK 293 cells expressing CRF receptors. Bell-shaped concentration-response curve for CRF-induced cAMP accumulation in HEK 293 cells stably expressing the YFP-tagged versions of CRFR1 (A) or CRFR2 (B). The adenine nucleotide pool of stably transfected HEK 293 cells (2 × 10⁵/well) was metabolically prelabeled with [³H]adenine for 16 h; where indicated (triangles), cells were also maintained in the presence of pertussis toxin (100 ng/ml). Thereafter, the cells were preincubated in the presence of rolipram (100 μM) for 1 h and subsequently stimulated with the indicated concentrations of CRF for 30 min. The accumulation of [³H]cAMP was quantified as outlined under Materials and Methods. To normalize for interassay variations, the maximum cAMP accumulation (observed with 6 and 10 nM in cells expressing CRFR1-Y and CRFR2-Y, respectively) was set to 100%. This value corresponded to 6295 ± 688 cpm and 2389 ± 176 cpm for CRFR1-Y and CRFR2-Y, respectively. Basal [³H]cAMP levels were 71 ± 15 and 108 ± 7 cpm for CRFR1-Y and CRFR2-Y, respectively. Data represent means ± S.E.M.; n = 3 to 8 experimental days, each concentration point measured in triplicate.

Fig. 2

FRET microscopy of CRF-receptors. A, HEK 293 cells transiently expressed plasmids encoding CFP- or YFP-tagged proteins as indicated. The columns show CFP and YFP images as indicated, the third column shows a false color rendering of the bleed-through-corrected FRETc image. All images are representative of two to seven experimental days and corrected for background. Scale bar, 10 μm. B, N_FRET values were calculated as described previously (Bartholomaüs et al., 2008); cells expressed the indicated constructs: ECFP + EYFP (n = 106 cells), CRFR2-C + Y-DAT (n = 13), CRFR1-C + Y-DAT (n = 11), CRFR2-C + CRFR2-Y (n = 220), CRFR1-C + CRFR1-Y (n = 90), and C-SERT-Y (n = 124). C, N_FRET values were determined in cells expressing the indicated CRFR-isoforms: CRFR2-C and CRFR2-Y in the absence (n = 220 cells) and presence of CRF (10 nM, n = 103; 100 nM, n = 83); similarly, CRFR1-C and CRFR1-Y in the absence (n = 90) and presence of CRF (10 nM, n = 27; 100 nM, n = 26).

Fig. 3

CRFRs interact with β-arrestin and internalize upon agonist treatment. A, HEK 293 cells transiently expressed plasmids encoding CFP-tagged CRFRs or YFP-tagged β-arrestin1. FRET microscopy was performed in the absence (◻) or presence of agonist CRF (∎; [CRF] 100 nM) as described under Materials and Methods; NFRET values were determined as described previously (Bartholomäus et al., 2008). Data represent means ± S.E.M. (n = 7, two independent transfections). B, confocal images of HEK 293 cotransfected with either CRFR1 or CRFR2 tagged with YFP (displayed in green color) and Rab5 tagged with CFP (displayed in red color). Yellow coloring indicates colocalization of YFP- and CFP-tagged proteins (at control conditions and after incubation with 100 nM CRF at the times indicated). The images are representative of three experimental days. The colocalization of CRFRs and of Rab5 was scored by an observer blinded to the experimental condition. Cells (6–11) from different experiments were scored per time point. The resulting average density of receptor positive vesicle/optical section of a given cell was plotted as a function of time. Error bars represent S.E.M.

Fig. 4

FRAP microscopy of CRF-receptors. A, confocal images of stably transfected HEK 293-CRFR1-Y or HEK 293-CRFR2-Y cells, representative of three to six experimental days. After an initial cell scan (0 s), a region of interest (arrow) was photobleached, and the recovery of the fluorescence was monitored over 90 s. The normalized fluorescence recovery was plotted versus time as an example of 1 experimental day and fitted by nonlinear regression (see Materials and Methods; basal: n = 9, 10 nM CRF: n = 7 and 100 nM CRF: n = 7, 10 μM concentration of antagonist αhCRF[9–41]: n = 11). The length of the photobleached strip was 5.5 ± 0.4 μm. B, scatter plots of the diffusion coefficients (n = 20–30 cells for CRFR1; n = 21–88 cells for CRFR2). C, scatter plots of the mobile fractions for each receptor subtype under the same conditions described before. Horizontal lines represent the mean. **, p < 0.01 and ***, p < 0.001; n.s., not significant. Scale bar, 10 μm.

Fig. 5

FCS recordings of CRFRs stably expressed in HEK 293 cells. A, representative recordings of the intensity fluctuations (top) of CRFR1-Y and CRFR2-Y in the absence (black curve, a) and presence of 100 nM CRF (gray curve, b). Autocorrelation curves (bottom) calculated from the intensity fluctuations (in kilohertz) were fitted to a two-component model and normalized (raw count rates; control conditions: CRFR1-Y = 238 kHz, CRFR2-Y = 239 kHz; 100 nM CRF: CRFR1-Y = 131 kHz, CRFR2-Y = 170 kHz). Under control conditions, the two components have diffusion times of τ1 ~0.13 ms and τ2 ~15.6 ms for CRFR1-Y cells and τ1 ~0.15 ms and τ2 ~20.5 ms for CRFR2-Y cells. B, scatter plots represent the diffusion coefficient of both CRFRs after incubation (15 min) in the absence and presence of CRF or 10 μM antagonist (αhCRF[9–41]); 3 to 13 experimental days, n = 41 to 140 cells. Horizontal lines represent the mean; *, p < 0.05; **, p < 0.01; and ***, p < 0.001; n.s., not significant.

Fig. 6

FCS recordings of CRFRs expressed in hippocampal neurons. A, representative measurements of the intensity fluctuations (top two) of YFP-tagged CRFR1 and CRFR2 in transiently transfected hippocampal neurons. The measurements were done in the absence (black curve, a) and presence of 100 nM CRF (gray curve, b) and are representative of three experimental days. Autocorrelation curves (bottom) were calculated from the intensity fluctuations (in kilohertz); the curves were fitted to a two component-model and normalized (raw count rates; control conditions: CRFR1-Y = 11 kHz, CRFR2-Y = 8 kHz; 100 nM CRF: CRFR1-Y = 4 kHz, CRFR2-Y = 2.5 kHz). The two components have diffusion times of τ1 ~0.013 ms and τ1 ~8.1 ms for CRFR1 control cells and τ1 ~0.018 ms and τ2 ~10.6 ms for CRFR2 control cells. B, bar chart represents the diffusion coefficient of each receptor subtype at different concentrations of CRF; n = 30 to 62.

Fig. 7

FCS recordings in HEK 293 cell lines expressing different levels of CRFR2. A, representative measurements of the intensity fluctuations (top) of three cell lines stably expressing CRFR2-Y (denoted 1 to 3). Autocorrelation curves (bottom) were calculated from the intensity fluctuations (in kilohertz) and fitted to a two-component model. Right, the curves were superimposable. B, receptor levels visualized by immunoblotting; lysates (~7 μg of protein) were prepared from the three cell lines (1–3) used in A and from the CRFR1-expressing cell (4) for comparison; after separation on a denaturing SDS-polyacrylamide gels, proteins were electrophoretically transferred onto nitrocellulose membranes, stained with Ponceau-S (right), and the immunoreactivity was subsequently visualized with an anti-GFP antibody (1:5000) by enhanced chemiluminescence using three different exposure times (indicated under each blot) to account for the large difference in expression levels. Note that differences in loading do not account for the different levels of immunoreactivity. The closed and open arrows indicate immunoreactivity for CRFRs and free YFP, respectively. Data are from one representative experiment that was replicated with identical results.

Fig. 8

Mobility of CRFR2-Y after disruption of cortical actin and cholesterol depletion/clustering assessed by FCS. A, representative measurement of the intensity fluctuations (top) of CRFR2-Y cells and after treatment with latrunculin A (1 μM), MβCD (8 mM), and filipin3 (5 μg/ml). Autocorrelation curves (bottom) calculated from the intensity fluctuations (in kilohertz) were fitted to a two-component model yielding diffusion times of CRFR2 τ2: ~20.5, ~22, ~31.6, and ~92.5 ms for control, latrunculin A-, MβCD-, and filipin3-treated cells, respectively. B, scatter plots represent the diffusion coefficient of CRFR2-Y after incubation with CRF, latrunculin A (n = 70), MβCD (n = 20), or filipin3 (n = 19). Horizontal lines represent the mean; **, p < 0.01; and ***, P < 0.001; n.s., not significant.

Fig. 9

CRFR complexes examined by FIDA. A, representative measurements of the intensity fluctuations (in kilohertz) detected by FCS (top) of CRFR1-Y and CRFR2-Y were used to calculate the probability of the photons detected by the microscope per 40 μs. A photon-counting histogram is shown for CRFR1-Y and CRFR2-Y in the presence or absence of 100 nM CRF. All measurements were best fitted with a single component model, and the brightness of each was determined as described under Materials and Methods. B, bar chart representing the brightness calculated for CRFR1-Y and CRFR2-Y at control conditions and 100 nM CRF of (two experimental days). Bars represent mean, S.E.M., n = 18 to 25 cells.

Fig. 10

Time course of ¹²⁵I-sauvagine binding to membranes prepared from CRFR1 and CRFR2 expressing cells at different temperatures. A, membranes from HEK 293 cells expressing CRFR1 (2–4 μg/assay) were incubated for the indicated time intervals and at the indicated temperatures in the presence of 0.2 nM ¹²⁵I-sauvagine. A parallel incubation contained 0.1 mM GTPγS (△), which reduced binding to a similar extent at all temperatures studied. Data are means ± S.D. from three independent experiments carried out in duplicate. A similar experiment was also done at 30°C but the points were omitted for the sake of clarity. The solid lines were drawn by fitting the data points to an equation describing a monoexponential association. B, Arrhenius plot. Apparent on rates determined were obtained as outlined in A for CRFR1.

Fig. 11

cAMP accumulation induced by CRFR1 and CRFR2 expressed at different levels (A and B) and after pretreatment of HEK 293 cells with latrunculin (C and D). A and B, HEK 293 cells (6 × 10⁶ cells) were transiently transfected with 0.8 (closed symbol) or 4 μg of plasmid (open symbol) encoding YFP-tagged CRFR1 (A) or CRFR2 (B). Cells (3 × 10⁵ cells/six-well dish) were seeded 24 h after transfection and allowed to adhere for 8 h. Subsequently, the cells were incubated in medium containing pertussis toxin (100 ng/ml) and [³H]adenine (1 μCi/ml) overnight. Cells were subsequently stimulated with CRF for 20 min, and the formation of [³H]cAMP was quantified as outlined in the legend to Fig. 1. EC₅₀ values were (means ± S.D.): 0.56 ± 0.11 and 0.12 ± 0.07 nM for low and high CRFR1 expression, respectively (p = 0.0027; t test for paired data); 0.29 ± 0.10 and 0.21 ± 0.05 for low and high CRFR2 expression, respectively (p = 0.10; t test for paired data). C and D, stably transfected HEK 293 cells expressing CRFR1 (C) or CRFR2 (D) were prelabeled with [³H]adenine overnight and then pretreated with latrunculin A as in Fig. 7 for 1 h before the addition of agonist for 20 min. EC₅₀ values for CRF were 0.21 ± 0.05 and 0.22 ± 0.08 (CRFR1; p = 0.92) and 0.21 ± 0.04 and 0.14 ± 0.03 nM (CRFR2; p = 0.04). Data points represent mean values ± S.D. from 3 (A and B) and 5 (C and D) independent experiments (done in triplicate); maximal [³H]cAMP accumulation in each experiment was set at 100% to account for interassay variation. These 100% values varied from 600 cpm (low receptor expression) to 5000 cpm (stable cell lines expressing CRFR1).

Fig. 12

Agonist-induced internalization of CRFR1 and of CRFR2 in the presence or absence of latrunculin A. HEK 293 cells stably expressing CRFR1 or CRFR2 tagged with YFP were seeded on cover slips and pretreated with vehicle or latrunculin (row labeled + Latr.) as in Fig. 8. Images were captured under basal conditions (time point 0 min) using the 514-nm laser line of a Zeiss 501 confocal laser scanning microscope (see Materials and Methods). Subsequently, CRF was added in a concentration of 100 nM, and images were captured in intervals of 10 min in the presence or absence of latrunculin A, laser power was set to 6% with the exception of row “CRF2R Basal”: here, laser intensity was set to 11% to visualize all possible intracellular fluorescent particles. Shown are representative images captured in parallel in the same experiment (and replicated with three coverslips each); the experiment was repeated twice.