Differences in the signaling pathways of α(1A)- and α(1B)-adrenoceptors are related to different endosomal targeting

Abstract

Aims: To compare the constitutive and agonist-dependent endosomal trafficking of α(1A)- and α(1B)-adrenoceptors (ARs) and to establish if the internalization pattern determines the signaling pathways of each subtype.

Methods: Using CypHer5 technology and VSV-G epitope tagged α(1A)- and α(1B)-ARs stably and transiently expressed in HEK 293 cells, we analyzed by confocal microscopy the constitutive and agonist-induced internalization of each subtype, and the temporal relationship between agonist induced internalization and the increase in intracellular calcium (determined by FLUO-3 flouorescence), or the phosphorylation of ERK1/2 and p38 MAP kinases (determined by Western blot).

Results and conclusions: Constitutive as well as agonist-induced trafficking of α(1A) and α(1B) ARs maintain two different endosomal pools of receptors: one located close to the plasma membrane and the other deeper into the cytosol. Each subtype exhibited specific characteristics of internalization and distribution between these pools that determines their signaling pathways: α(1A)-ARs, when located in the plasma membrane, signal through calcium and ERK1/2 pathways but, when translocated to deeper endosomes, through a mechanism sensitive to β-arrestin and concanavalin A, continue signaling through ERK1/2 and also activate the p38 pathway. α(1B)-ARs signal through calcium and ERK1/2 only when located in the membrane and the signals disappear after endocytosis and by disruption of the membrane lipid rafts by methyl-β-cyclodextrin.

Figure 1. Internalised QAPB ligand-α_1A-AR and α_1B-AR complexes colocalize with the recycling endosomal fluorescent ligand marker Transferrin, (Tfn), -Alexa Fluor546 in unstimulated living R-1Fs stably expressing the human α_1A -AR or α_1B-AR.

In image i, nuclei stained with Hoechst are shown in blue. ii, green punctates represent QAPB (10 nM)-labeled human α_1A -ARs, and iii, recycling vesicles labeled with Tfn-Alexa Fluor546 are represented by the red punctate spots. Overlay image iv shows the partial colocalization (yellow vesicles) detected when unstimulated R-1Fs stably expressing α_1A -AR-QAPB ligand complexes fuse with the recycling red fluorescent endosomal marker. Red-Green pixel scatter intensity plots constructed from the intensity values located within the rectangular region superimposed on each channel image ii–iii in exactly the same x–y position is illustrated in v and r2 represents the Pearson overlap correlation coefficient value quantified from the matched regions defined in image ii and iii. Blue color in image vi represents Hoechst stained nuclei. vii, green punctates represent QAPB (10 nM)-labeled human α_1B-ARs, and viii, recycling vesicles labeled with fluorescent Tfn are represented by the red punctate spots. Merge image ix shows the partial colocalization (yellow vesicles) observed when α_1B-AR-QAPB ligand complexes fuse with the recycling red fluorescent endosomal marker. Pearson correlation coefficient value, (r2), measured from the intensity values located within the rectangular region superimposed on images vii–viii is illustrated in image ix.

Figure 2. Internalised QAPB ligand- α_1A-AR and α_1B-AR complexes colocalize with the late endosomal fluorescent marker Lysotracker Red DND-99 in living R-1Fs stably expressing the human α_1A-AR or α_1B-AR.

(A). In image i, nuclei stained with Hoechst are shown in blue. ii, green punctates represent QAPB (10 nM)-labeled human α_1A -ARs, and iii, acidic late endosomes labeled with Lysotracker Red DND-99 for 80 minutes are represented by the red punctate spots. Overlay image iv shows the extensive colocalization (yellow vesicles) detected when unstimulated R-1Fs stably expressing α_1A -AR-QAPB ligand complexes fuse with the late endosomes fluorescently labeled with Lysotracker Red. Linear red-green pixel scatter intensity plots constructed from the intensity values located within images ii–iii is illustrated in v and r2 represents the Pearson overlap correlation coefficient value quantified from each channel image ii and iii. 3D x-z colocalization maximum projection view of human α_1A -AR-QAPB ligand complexes fusing with late/lytic endosomes is shown in iv (yellow punctates).(B). In image i, nuclei stained with Hoechst are shown in blue. ii, green punctates represent QAPB (10 nM)-labeled human α_1B-ARs, and iii, late endosomes labeled with Lysotracker Red for 80 minutes are illustrated by the red punctate spots. Merge image iv shows the extensive colocalization (yellow vesicles) detected when unstimulated R-1Fs stably expressing α_1B-AR-QAPB ligand complexes fuse with the late endosomes fluorescently labeled with Lysotracker Red. Linear green-red pixel scatter intensity plots constructed from the intensity values located within images ii–iii is illustrated in v and r2 represents the Pearson overlap correlation coefficient value quantified from each channel image ii and iii. A 3D x-z colocalization maximum projection view of human α_1B-AR-QAPB ligand complexes interacting with late/lytic endosomes is shown in iv (yellow punctates).

Figure 3. Quantitative analysis of the constitutive internalization of VSV-G α_1A- and VSV-G α_1B-ARs.

Live HEK293 cells stably transfected with each subtype were incubated with CypHer5E Linked anti-VSV-G Antibody at a 5 µg/ml in KRH buffer at 4°C for 1h. After washing with cold KRH Buffer, coverslips were rapidly mounted into a chamber bath, placed on the confocal microscope stage in a 95% air and 5% CO₂ atmosphere at 37°C. At this time, HEK293 were then exposed to prewarmed KRH buffer for 30 min at 37°C and the images were acquired. In some experiments 10 µM prazosin or 250 µg/ml concanavalin A (ConA) were added during the incubation time. (A) Confocal images representatives of the increase of intracellular fluorescence after 30 min of incubation at 37^aC for both VSV-G-α_1A- and VSV-G-α_1B-ARs (B). Changes in intracellular fluorescence intensity were measured in the cells treated or not with prazosin or ConA. Intracellular fluorescence was quantified in the regions close to the cytoplasmic side of the plasma membrane, and in the cytosolic region away from nuclei (Figure S1). Data were expressed as arbitrary units of fluorescence and represent the mean ± S.E.M. of at least 4 independent experiments. Student’s t test: *p<0.05 **p<0.01 vs control.

Figure 4. Spatio-temporal analysis of the constitutive internalization of VSV-G α_1A- and VSV-G α_1B-ARs.

Live HEK293 cells stably transfected with each subtype were treated according to protocol detailed in Figure 3 and coverslips were rapidly mounted into a chamber bath, placed on the confocal microscope stage in a 95% air and 5% CO₂ atmosphere at 37°C. At this time, HEK293 were then exposed to pre-warmed KRH buffer for 30 min. After 30 min of incubation, the images were acquired at zero time, 1 min and then 5 min intervals for 15 min. Confocal images are representatives of the increase of intracellular fluorescence for both VSV-G-α_1A-AR and VSV-G-α_1B-AR. Graphs represent the changes in intracellular fluorescence intensity over time in the regions close to the cytoplasmic side of the plasma membrane and in the cytosolic region away from nuclei in control experiments (black symbols), cells preincubated with 10 µM prazosin (grey symbols) or 250 µg/ml concanavalin A (white symbols). Data were expressed as arbitrary units of fluorescence and represent the mean ± S.E.M. of at least 4 independent experiments. Two-way ANOVA indicates a significant (p<0.001) time dependent change for α_1A- membrane and cytosolic pools.

Figure 5. Spatio-temporal analysis of the agonist-induced internalization of VSV-G α_1A- and VSV-G α_1B-ARs.

Live HEK293 cells stably transfected with each subtype were treated according to protocol detailed in Figure 3 and coverslips were rapidly mounted into a chamber bath, placed on the confocal microscope stage in a 95% air and 5% CO₂ atmosphere at 37°C. After 30 min of incubation PHE 100 µM was added and the images were acquired immediately before PHE addition (zero time) and 1, 5, 10 and 15 min. Confocal images are representatives of the increase of intracellular fluorescence for both VSV-G-α_1B-AR and VSV-G-α_1B-AR. Graphs represent the significant changes in intracellular fluorescence intensity over time in the regions close to the cytoplasmic side of the plasma membrane and in the cytosolic region away from nuclei in control experiments (black symbols), cells preincubated with 10 µM prazosin (grey symbols) or 250 µg/ml concanavalin A (white symbols). Data were expressed as percentage of basal fluorescence determined before agonist addition (0 min) and represent the mean ± S.E.M. of at least 4 independent experiments. Two way ANOVA indicates a significant (p<0.001 or p<0.05) ) time dependent change for α_1A - membrane and cytosolic pools and for α_1B-cytosolic pool. Student’s t test was applied to determine significant differences at a given time vs control, *p<0.05 **p<0.01.

Figure 6. Activation of α₁-ARs in cells stably transfected with the VSV-tagged or untagged α_1A- and α_1B-subtypes increases intracellular Ca²⁺.

Live HEK293 cells stably transfected with each subtype were incubated for 2 h with the fluorescent Ca²⁺ quelant FLUO-3-AM and mounted into a chamber bath placed on the confocal microscopy stage in a 95% air and 5% CO₂ atmosphere at 37°C. After 30 min of stabilization, images were collected before (zero time) and after (1, 5, 10 and 15 min) adding PHE 100 µM. Graphs represent the time-course of intracellular Ca²⁺ increase induced by α₁-AR activation of VSV-tagged or untagged α_1A- and α_1B-ARs. The data were calculated as the increase in fluorescence observed after PHE addition over time, after substraction of the fluorescence observed in parallel experiments performed in the presence of PHE + prazosin (10 µM). Data represent the means ± S.E.M. of 5–8 independent experiments. Two way ANOVA indicates that activation of the VSV tagged and untagged α_1A and α_1B-subtypes elicits a significant (p<0.05) time dependent change in the calcium signal.

Figure 7. α_1A- and α_1B-AR stimulation shows a subtype-specific pattern of ERK1/2 phosphorylation.

HEK293 cells stably transfected with VSV-tagged or untagged α_1A- and α_1B-subtypes were serum-starved for 4 hours and stimulated with PHE (100 µM) for a 15 minute time-course at 37°C. In some experiments, prazosin (10 µM) or 5-methylurapidil (10 µM) were added for 30 min. After stimulation, cellular extracts were prepared as described under the “Experimental procedures”. Equal amounts (15 µg) of each sample were used to visualize the p-ERK1/2 expression Western blots from representative experiments were shown. The lower panels show total ERK1/2 loaded on each sample. The graph quantifies the p-ERK1/2 signal at different times. Data represent means ± S.E.M. of 4-6 independent experiments. Student’s t test was applied to determine significant differences at a given time vs control, *p<0.05 **p<0.01.

Figure 8. α_1A- and α_1B-AR mediated ERK1/2 activation is modulated by methyl-β-cyclodextrin, filipin and concanavalin A.

HEK293 cells stably transfected with α_1A- and α_1B-AR subtypes were serum-starved for 4 hours and stimulated with PHE (100 µM) for a 15 minute time-course at 37°C. In some experiments, methyl-β-cyclodextrin 10 mM (mβCD) was added for 30 or 60 min, filipin 1 µg/ml and concanavalin A 250 µg/ml for 30 min. After stimulation, cellular extracts were prepared as described under the “Experimental procedures”. Equal amounts (15 µg) of each sample were used to visualize the p-ERK1/2 expression Western blots from representative experiments are shown. The lower panels show total ERK1/2 loaded on each sample. The graph quantifies the p-ERK1/2 signal at different times. Data represent means ± S.E.M. of 3-6 independent experiments. Student’s t test was applied to determine significant differences at a given time vs control, *p<0.05 **p<0.01 ***p<0.001.

Figure 9. α_1A- and α_1B-AR stimulation shows a subtype-specific pattern of p38-MAPK phosphorylation.

HEK293 cells stably transfected with VSV-tagged or untagged α_1A- and α_1B-subtypes were serum-starved for 4 hours and stimulated with PHE (100 µM) for a 15 minute time-course at 37°C. In some experiments, prazosin (10 µM) or 5-methylurapidil (10 µM) were added for 30 min. After stimulation, cellular extracts were prepared as described under the “Experimental procedures”. Equal amounts (15 µg) of each sample were used to visualize the phosphorylated p-38 MAPK expression Western blots from representative experiments are shown. The lower panels show total p38 loaded on each sample. The graph quantifies the p-p38 MAPK signal at different times. Data represent means ± S.E.M. of 3-6 independent experiments. Student’s t test was applied to determine significant differences at a given time vs control, *p<0.05 **p<0.01 ***p<0.001.

Figure 10. α_1A- and α_1B-AR mediated p-38 activation is modulated by methyl-β-cyclodextrin, filipin and concanavalin A.

HEK293 cells stably transfected with each α₁-AR subtype were serum-starved for 4 hours and stimulated with PHE (100 µM) for a 15 minute time-course at 37°C. In some experiments, methyl-β-cyclodextrin 10 mM (mβCD) was added for 30 or 60 min, filipin 1 µg/ml and concanavalin A 250 µg/ml for 30 min. After stimulation, cellular extracts were prepared as described under the “Experimental procedures”. Equal amounts (15 µg) of each sample were used to visualize the p-p38-MAPK expression Western blots from representative experiments are shown. The lower panels show amounts of p38-MAPK loaded on each sample. The graph quantifies the p-p38-MAPK signal at different times after agonist addition. Data represent means ± S.E.M. of 3–4 independent experiments. Student’s t test was applied to determine significant differences at a given time vs control, ***p<0.001.

Figure 11. Combined analysis of the temporal patterns of calcium signal, pERK1/2 signal and internalization of VSV-α₁-ARs following stimulation.

Schematic diagram representing the kinetic of the internalization patterns of VSVG-α_1A and VSVG-α_1B ARs in endosomes located near the plasma membrane (continuous line) or in the cytosol (discontinuous line), and the kinetic of the intracellular calcium signal (green), or the pERK1/2 signal (blue) elicited by activation of these receptors or by activation of the receptors after disruption of lipid rafts with mβCD (red). In order to facilitate the comparison, the data were calculated as a percentage of the maximal response in each case, and represent the mean of the experiments previously described