Dysbindin promotes the post-endocytic sorting of G protein-coupled receptors to lysosomes

Abstract

Background: Dysbindin, a cytoplasmic protein long known to function in the biogenesis of specialized lysosome-related organelles (LROs), has been reported to reduce surface expression of D2 dopamine receptors in neurons. Dysbindin is broadly expressed, and dopamine receptors are members of the large family of G protein-coupled receptors (GPCRs) that function in diverse cell types. Thus we asked if dysbindin regulates receptor number in non-neural cells, and further investigated the cellular basis of this regulation.

Methodology/principal findings: We used RNA interference to deplete endogenous dysbindin in HEK293 and HeLa cells, then used immunochemical and biochemical methods to assess expression and endocytic trafficking of epitope-tagged GPCRs. Dysbindin knockdown up-regulated surface expression of D2 receptors compared to D1 receptors, as reported previously in neurons. This regulation was not mediated by a change in D2 receptor endocytosis. Instead, dysbindin knockdown specifically reduced the subsequent trafficking of internalized D2 receptors to lysosomes. This distinct post-endocytic sorting function explained the minimal effect of dysbindin depletion on D1 receptors, which recycle efficiently and traverse the lysosomal pathway to only a small degree. Moreover, dysbindin regulated the delta opioid receptor, a more distantly related GPCR that is also sorted to lysosomes after endocytosis. Dysbindin was not required for lysosomal trafficking of all signaling receptors, however, as its depletion did not detectably affect down-regulation of the EGF receptor tyrosine kinase. Dysbindin co-immunoprecipitated with GASP-1 (or GPRASP-1), a cytoplasmic protein shown previously to modulate lysosomal trafficking of D2 dopamine and delta opioid receptors by direct interaction, and with HRS that is a core component of the conserved ESCRT machinery mediating lysosome biogenesis and sorting.

Conclusions/significance: These results identify a distinct, and potentially widespread function of dysbindin in promoting the sorting of specific GPCRs to lysosomes after endocytosis.

Figure 1. Dysbindin knockdown increases surface expression of recombinant D2 dopamine receptors in HEK293 cells.

A. Immunoblot showing depletion of endogenous dysbindin in HEK293 cells by three different siRNA duplexes. A representative immunoblot is shown of cell extracts (30 µg/lane) from control-transfected (CTL) cells and cells transfected with one of three siRNA specific duplexes; similar results were obtained with each of the silencing duplexes The immunoreactive band corresponding to endogenous dysbindin is indicated by arrow. A small amount of cell extract (∼2 µg) from cells transfected with HA-DYS was run alongside to verify this assignment (right lane, HA-DYS resolves at a slightly higher apparent molecular mass then that endogenous protein due to the presence of the epitope tag). The high molecular band observed between the 160 and 260 kDa markers represents a nonspecific band useful for verifying comparable loading. GAPDH blot is shown in the panel below to further verify equivalent loading. B. Effect of dysbindin knockdown on relative surface receptor immunoreactivity of stably transfected HEK293 cells expressing either FLAG-D2R or FLAG-D1R (indicated by D2 and D1, respectively). Surface receptor immunoreactivity was determined by anti-FLAG labeling of intact cells and fluorescence flow cytometry, as described in Materials and Methods , comparing mean fluorescence values measured from cells transfected with the non-silencing control RNA duplex (CTL) and a silencing duplex (siRNA #1 from panel A). For each experiment, surface receptor immunoreactivity measured in cells transfected with the dysbindin-silencing duplex (DYS) was normalized to that measured in parallel determination from cells transfected with the non-silencing control duplex (CTL). Bars represent determinations averaged over multiple (≥5) experiments and the error bar indicates S.E.M. p values were calculated from the non-normalized individual data using a paired Student's t-test.

Figure 2. Dysbindin knockdown specifically inhibits proteolysis of internalized D2 receptors without detectably inhibiting receptor internalization.

A. Flow cytometric analysis of dopamine-induced internalization of D2 dopamine receptors. Stably transfected HEK293 cells expressing FLAG-D2R, exposed for the indicated time periods to 10 µM dopamine, were analyzed by surface antibody labeling and flow cytometry. Loss of surface receptor immunoreactivity was used to assess ligand-induced internalization in cells transfected with dysbindin siRNA (DYS) and compared to cells transfected with a non-silencing control duplex (CTL). Points represent averaged values (normalized to cells not exposed to dopamine (t = 0) from ≥5 experiments and error bars indicate S.E.M. B. Fractional internalization measured after exposure cells to 10 µM dopamine for 30 min, showing the lack of significant difference between CTL and DYS conditions. C. Surface biotinylation experiment showing that dysbindin knockdown inhibits dopamine-induced proteolysis of FLAG-D2Rs. D. Quantification of time-dependent loss of surface-biotinylated FLAG-D2R in cells incubated for the indicated time period after surface biotinylation in the presence of 10 µM dopamine. E. Comparison of D2R degradation measured at the 4 hour time point over multiple experiments (n = 8), verifying the statistical significance (p = 0.001 by Student's t-test) of the observed inhibition. F. Biotin protection-degradation assay showing that dysbindin knockdown specifically inhibits degradation of D2Rs after internalization.

Figure 3. Dysbindin knockdown also inhibits lysosomal proteolysis of DORs.

A. Immunoblot of FLAG-DOR immunoreactivity detected in HEK293 cell lysates (30 µg/lane) 3 days after transfection with dysbindin siRNA (DYS) compared to a non-silencing control RNA duplex (CTL), showing up-regulation of steady-state receptor number. Loading was normalized to equal total protein. B. Flow cytometric analysis showing that DYS knockdown significantly increases surface FLAG-DOR immunoreactivity. Methodology is the same as the experiment described in Fig 1B. C. Immunoblot analysis showing that dysbindin knockdown (DYS) inhibits proteolysis of FLAG-DOR induced by exposure of cells to the opioid agonist etorphine (10 µM) for 3 h compared to cells incubated in the absence of etorphine (t = 0). The immunoreactive species corresponding to the mature receptor is indicated by bracket. Loading was normalized to equivalent receptor levels at t = 0, to allow visual appreciation of differences in ligand-induced degradation. D. Radioligand binding analysis of FLAG-DOR down-regulation induced by the opioid peptide agonist DADLE (10 µM), showing that dysbindin depletion inhibits ligand-induced down-regulation at all time points examined. Statistical analysis of the difference between CTL and DYS conditions was carried out at the 4 hour time point, as indicated, using Student's t-test.

Figure 4. Dysbindin knockdown inhibits FLAG-DOR proteolysis in HeLa cells, but does not affect down-regulation of EGF receptors.

A. Immunoblot analysis of endogenous EGFRs in lysates (30 µg/lane) prepared from HeLa cells that also express FLAG-DOR (introduced by stable transfection), prepared after incubating serum-starved cells for the indicated time with 100 ng/ml EGF. B. Quanitification of ligand-induced proteolysis of EGFRs by scanning densitometry of immunoblots from multiple (n = 5) experiments. C. Quantification of ligand-induced proteolysis of FLAG-DOR in the same HeLa cell clone (n = 5). The statistical significance of the difference between CTL and DYS conditions was calculated at the 1 hour (panel B) or 3 hour (panel C) time point using Student's t-test.

Figure 5. Dysbindin associates with GASP1 and HRS.

A. Co-immunoprecipitation of a GFP-tagged version of full-length GASP1 (GFP-GASP1) with DYS-HA from lysates prepared from transfected HEK293 cells. Asterisk indicates immunoreactive signal representing DYS-HA isolated in the GFP-GASP1 but not control (GFP or GFP-PDZ) immunoprecipitate. Arrow indicates GFP-GASP1 and arrowhead indicates GFP. B. Co-immunoprecipitation of Myc-HRS with DYS-HA from transfected HEK293 cell lysates. Asterisk indicates DYS-HA that copurified specifically with Myc-HRS but not with GFP or with Myc-HRS-ΔCC. C. Colocalization of DYS-GFP with a subset of endosomes labeled by Cherry-HRS. A representative image is shown and scale bar indicates 10 µm. D. Model for the proposed function of dysbindin in specifically promoting the post-endocytic sorting of GPCRs to lysosomes via GASP/HRS connectivity. GASP1 localizes predominantly to the cytoplasm, suggesting that it acts as a transient linker.