Role of the proteasome in modulating native G-CSFR expression

Abstract

The granulocyte colony-stimulating factor receptor (G-CSFR) is a critical regulator of granulopoiesis, but the mechanisms controlling its surface expression are poorly understood. Recent studies using transfected cell lines have suggested the activated G-CSFR is routed to the lysosome and not the proteasome. Here, we examined the role of the ubiquitin/proteasome system in regulating G-CSFR surface expression in both ts20 cells that have a temperature-sensitive E1 ubiquitin-activating enzyme and in primary human neutrophils. We show that the G-CSFR is constitutively ubiquitinated, which increases following ligand binding. In the absence of a functional E1 enzyme, ligand-induced internalization of the receptor is inhibited. Pre-treatment of ts20 transfectants with either chloroquine or MG132 inhibited ligand-induced G-CSFR degradation, suggesting a role for both lysosomes and proteasomes in regulating G-CSFR surface expression in this cell line. In neutrophils, inhibition of the proteasome but not the lysosome was found to inhibit internalization/degradation of the activated G-CSFR. Collectively, these data demonstrate the requirement for a functional ubiquitin/proteasome system in G-CSFR internalization and degradation. Our results suggest a prominent role for the proteasome in physiologic modulation of the G-CSFR, and provide further evidence for the importance of the ubiquitin/proteasome system in the initiation of negative signaling by cytokine receptors.

Fig. 1

Constitutive and ligand-induced ubiquitination of the WT G-CSFR. (A) Whole cell lysates prepared from untransfected ts20 cells (UNT), ts20 cells transfected with V5-tagged WT G-CSFR (left panel) or Myc-tagged Δ716 G-CSFR (right panel), or co-transfected with either tagged receptor form and HA-tagged ubiquitin (HA) were immunoprecipitated with anti-HA antibody, then blotted with antibody to the G-CSFR. (B) ts20 cells co-expressing V5-tagged WT G-CSFR and HA-ubiquitin (left panels) or Myc-tagged Δ716 G-CSFR and HA-ubiquitin (right panels) were stimulated with G-CSF (100 ng/mL) for the indicated times, lysed, then immunoprecipitated with anti-V5 (left) or anti-Myc (right) antibody and blotted with anti-HA antibody (upper panels) or anti-G-CSFR antibody (lower panels). Untransfected cells (UNT) and cells transfected with HA-ubiquitin only (HA) are shown as controls. (C) To control for receptor protein loading, histograms of the density ratio between bands in the upper and lower panels (B) were generated using ImageJ software.

Fig. 2

Ligand binding induces mono-/polyubiquitination of the G-CSFR. ts20 cells expressing V5-tagged WT G-CSFR were incubated at 37°C with G-CSF (100 ng/mL) for the indicated times, lysed, then immunoprecipitated with anti-V5 antibody and blotted with the indicated antibodies. The FK1 antibody (A) recognizes polyubiquitinated proteins whereas the FK2 antibody (B) recognizes both poly- and mono-ubiquitinated proteins. (C) The blot was stripped and re-blotted with anti-V5 antibody. Histograms are shown to the right of each blot depicting the density ratio between bands of FK1 and FK2 to anti-V5, respectively, as a control for receptor protein loading.

Fig. 3

Internalization of the G-CSFR requires an intact ubiquitin-conjugation system. ts20 cells expressing the WT (upper panels) or Δ716 (lower panels) G-CSFR were stimulated with G-CSF (100 ng/mL) for 1h at 4^oC then incubated at 30°C for 0 (A), 30 min (B), and 60 min (C). Cells were fixed, sequentially incubated with biotinylated antibody to the G-CSFR, streptavidin-Cy5, and Hoechst nuclear stain, then examined by confocal microscopy. (A-C) Cellular distribution of WT and Δ716 G-CSFR (red) and nuclei (blue). (D) WT G-CSFR-expressing cells incubated at 42°C (non-permissive temperature) for 60 min. Arrows indicate predominant membrane localization of the WT G-CSFR when ubiquitination is inhibited at 42– and of the Δ716 G-CSFR at the permissive temperature.

Fig. 4

Inhibition of the E1 ubiquitin-activating enzyme blocks G-CSFR internalization. G-CSFR surface expression was quantified from multiple confocal images using LSM5 software to measure the Cy5 fluorescence intensity over the outer 0.35 mM of individual ts20 cells transfected with the WT or Δ716 G-CSFR. The cells were stimulated with G-CSF (100 ng/mL) for 1h at 4°C, then incubated at 30°C or shifted to 42°C to inactivate the E1 enzyme. Cells were fixed and sequentially incubated with biotin-conjugated mouse anti-human G-CSFR antibody (CD114, BD PharMingen) and Streptavidin-Cy5 (Caltag) to detect the G-CSFR. (A) Yellow arrow (left panel) depicts the region used to measure fluorescence intensity from a representative image of a cell expressing the Δ716 G-CSFR incubated for 30 min at 30°C. Histogram (right panel) shows the fluorescence intensity of Cy5 (red) and DAPI (blue) staining in the region corresponding to the yellow arrow. (B) Bar graphs show the mean Cy5 fluorescence intensity corresponding to G-CSFR surface expression on cells stimulated with G-CSF and incubated for the indicated times at either 30°C or 42°C.

Fig. 5

Effect of lysosomal and proteasomal inhibitors on ligand-induced degradation of the G-CSFR. Surface proteins from ts20 cells expressing the WT G-CSFR were labeled by biotinylation. After biotinylation, cells were left untreated, or pre-treated with the following inhibitors: lactacystin (25μM) alone, chloroquine (200μM) alone, MG132 (20μM) alone, or lactacystin plus chloroquine. Cells were then stimulated with or without G-CSF (100ng/ml) at 37ºC for the indicated times, and lysed. (A) Whole cell lysates were immunoprecipitated with anti-V5 antibody and immunoblotted with HRP-labeled streptavidin to detect the biotinylated G-CSFR. Blot with IgG heavy chain (bottom panel) is included to show equal protein loading. (B) Densitometric analysis of bands in (A) for times 0 and 120 min after G-CSF stimulation and at 120 min in the absence of G-CSF.

Fig. 6

Effect of lysosome and proteasome inhibitors on G-CSFR internalization in human neutrophils. Human neutrophils were isolated from whole blood using Ficoll-Paque PLUS, 3% Dextran sulfate, and ACK lysis buffer. After isolation, neutrophils were left untreated, or pre-treated with the following inhibitors for 1 h at 37ºC: lactacystin (100 μM) alone, chloroquine (10 μM) alone, or MG132 (5 μM) alone. Neutrophils were then stimulated with or without G-CSF (100 ng/ml) for 1 h at 4ºC, then transferred to 37ºC for the indicated time points. Cells were then stained using PE-conjugated anti-G-CSFR antibody for 45 min at 4ºC, fixed in 1% paraformaldehyde, and analyzed. Unstained controls are indicated in the solid grey shading. A shift to the left is indicative of decreased G-CSFR expression and consistent with G-CSFR internalization/degradation.