Regulation of G-protein coupled receptor traffic by an evolutionary conserved hydrophobic signal

Abstract

Plasma membrane (PM) expression of G-protein coupled receptors (GPCRs) is required for activation by extracellular ligands; however, mechanisms that regulate PM expression of GPCRs are poorly understood. For some GPCRs, such as alpha2c-adrenergic receptors (alpha(2c)-ARs), heterologous expression in non-native cells results in limited PM expression and extensive endoplasmic reticulum (ER) retention. Recently, ER export/retentions signals have been proposed to regulate cellular trafficking of several GPCRs. By utilizing a chimeric alpha(2a)/alpha(2c)-AR strategy, we identified an evolutionary conserved hydrophobic sequence (ALAAALAAAAA) in the extracellular amino terminal region that is responsible in part for alpha(2c)-AR subtype-specific trafficking. To our knowledge, this is the first luminal ER retention signal reported for a GPCR. Removal or disruption of the ER retention signal dramatically increased PM expression and decreased ER retention. Conversely, transplantation of this hydrophobic sequence into alpha(2a)-ARs reduced their PM expression and increased ER retention. This evolutionary conserved hydrophobic trafficking signal within alpha(2c)-ARs serves as a regulator of GPCR trafficking.

Keywords: ER export; ER retention; GPCR trafficking; protein sorting; retention signal.

Figure 1. Differential localization of α_2a- and α_2c-ARs

(A) RAT1 fibroblasts cells were transiently transfected with amino terminus HA epitope-tagged α_2a- or α_2c-ARs and fixed after 48 hours. Cell surface and total cellular receptor distributions were stained with the HA antibody 16B12 in the absence or presence of detergent for permeabilization, respectively. (B) Quantification by flow cytometry of relative surface and total expression of HA epitope- tagged α_2a- and α_2c-ARs was done in transiently transfected HEK293 (see methods). Histograms representing surface and total fluorescent antibody labeling are shown. (C) Cellular localizations α_2a- and α_2c-ARs were determined in transiently transfected RAT1 cells. 48 Hours after transfection the cells were fixed and stained with the HA antiobody 16B12 and either rabbit polyclonal antibody to cis/media-Golgi compartment, giantin, or ER compartment, calnexin. Both α_2a- and α_2c-ARs were noted to have limited amount of intracellular localization that was similar in appearance to the cis/media-Golgi compartment labeled by giantin antibody. By contrast only the α_2c-ARs were noted to have a large intracellular localization similar to ER as labeled with calreticulin antibody. (D) Cellular processing and maturation of wild type α_2a- and α_2c-ARs was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. Membrane preparations obtained for α_2a- and α_2c-ARs were digested with the endoglycosidases, Endo H and PNGase F. After endoglycosidase digestion of α_2a- and α_2c-AR preparations were processed and immunoblotted using the HA epitope antibody, 16B12. Mature, Endo H resistant, α_2a- and α_2c-ARs (filled arrow) were noted at apparent molecular mass of ~60 to ~75kDa and ~55 to ~70kDa, respectively. The immature, Endo H sensitive, form of α_2a- and α_2c-ARs (unfilled arrow) was noted to have an apparent molecular mass of ~53 to 55kDa and ~45 to~50kDa, respectively. As expected, both the mature and immature forms of α_2a- and α_2c-ARs were sensitive to PNGase F digestion. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 1. Differential localization of α_2a- and α_2c-ARs

(A) RAT1 fibroblasts cells were transiently transfected with amino terminus HA epitope-tagged α_2a- or α_2c-ARs and fixed after 48 hours. Cell surface and total cellular receptor distributions were stained with the HA antibody 16B12 in the absence or presence of detergent for permeabilization, respectively. (B) Quantification by flow cytometry of relative surface and total expression of HA epitope- tagged α_2a- and α_2c-ARs was done in transiently transfected HEK293 (see methods). Histograms representing surface and total fluorescent antibody labeling are shown. (C) Cellular localizations α_2a- and α_2c-ARs were determined in transiently transfected RAT1 cells. 48 Hours after transfection the cells were fixed and stained with the HA antiobody 16B12 and either rabbit polyclonal antibody to cis/media-Golgi compartment, giantin, or ER compartment, calnexin. Both α_2a- and α_2c-ARs were noted to have limited amount of intracellular localization that was similar in appearance to the cis/media-Golgi compartment labeled by giantin antibody. By contrast only the α_2c-ARs were noted to have a large intracellular localization similar to ER as labeled with calreticulin antibody. (D) Cellular processing and maturation of wild type α_2a- and α_2c-ARs was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. Membrane preparations obtained for α_2a- and α_2c-ARs were digested with the endoglycosidases, Endo H and PNGase F. After endoglycosidase digestion of α_2a- and α_2c-AR preparations were processed and immunoblotted using the HA epitope antibody, 16B12. Mature, Endo H resistant, α_2a- and α_2c-ARs (filled arrow) were noted at apparent molecular mass of ~60 to ~75kDa and ~55 to ~70kDa, respectively. The immature, Endo H sensitive, form of α_2a- and α_2c-ARs (unfilled arrow) was noted to have an apparent molecular mass of ~53 to 55kDa and ~45 to~50kDa, respectively. As expected, both the mature and immature forms of α_2a- and α_2c-ARs were sensitive to PNGase F digestion. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 1. Differential localization of α_2a- and α_2c-ARs

(A) RAT1 fibroblasts cells were transiently transfected with amino terminus HA epitope-tagged α_2a- or α_2c-ARs and fixed after 48 hours. Cell surface and total cellular receptor distributions were stained with the HA antibody 16B12 in the absence or presence of detergent for permeabilization, respectively. (B) Quantification by flow cytometry of relative surface and total expression of HA epitope- tagged α_2a- and α_2c-ARs was done in transiently transfected HEK293 (see methods). Histograms representing surface and total fluorescent antibody labeling are shown. (C) Cellular localizations α_2a- and α_2c-ARs were determined in transiently transfected RAT1 cells. 48 Hours after transfection the cells were fixed and stained with the HA antiobody 16B12 and either rabbit polyclonal antibody to cis/media-Golgi compartment, giantin, or ER compartment, calnexin. Both α_2a- and α_2c-ARs were noted to have limited amount of intracellular localization that was similar in appearance to the cis/media-Golgi compartment labeled by giantin antibody. By contrast only the α_2c-ARs were noted to have a large intracellular localization similar to ER as labeled with calreticulin antibody. (D) Cellular processing and maturation of wild type α_2a- and α_2c-ARs was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. Membrane preparations obtained for α_2a- and α_2c-ARs were digested with the endoglycosidases, Endo H and PNGase F. After endoglycosidase digestion of α_2a- and α_2c-AR preparations were processed and immunoblotted using the HA epitope antibody, 16B12. Mature, Endo H resistant, α_2a- and α_2c-ARs (filled arrow) were noted at apparent molecular mass of ~60 to ~75kDa and ~55 to ~70kDa, respectively. The immature, Endo H sensitive, form of α_2a- and α_2c-ARs (unfilled arrow) was noted to have an apparent molecular mass of ~53 to 55kDa and ~45 to~50kDa, respectively. As expected, both the mature and immature forms of α_2a- and α_2c-ARs were sensitive to PNGase F digestion. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 1. Differential localization of α_2a- and α_2c-ARs

(A) RAT1 fibroblasts cells were transiently transfected with amino terminus HA epitope-tagged α_2a- or α_2c-ARs and fixed after 48 hours. Cell surface and total cellular receptor distributions were stained with the HA antibody 16B12 in the absence or presence of detergent for permeabilization, respectively. (B) Quantification by flow cytometry of relative surface and total expression of HA epitope- tagged α_2a- and α_2c-ARs was done in transiently transfected HEK293 (see methods). Histograms representing surface and total fluorescent antibody labeling are shown. (C) Cellular localizations α_2a- and α_2c-ARs were determined in transiently transfected RAT1 cells. 48 Hours after transfection the cells were fixed and stained with the HA antiobody 16B12 and either rabbit polyclonal antibody to cis/media-Golgi compartment, giantin, or ER compartment, calnexin. Both α_2a- and α_2c-ARs were noted to have limited amount of intracellular localization that was similar in appearance to the cis/media-Golgi compartment labeled by giantin antibody. By contrast only the α_2c-ARs were noted to have a large intracellular localization similar to ER as labeled with calreticulin antibody. (D) Cellular processing and maturation of wild type α_2a- and α_2c-ARs was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. Membrane preparations obtained for α_2a- and α_2c-ARs were digested with the endoglycosidases, Endo H and PNGase F. After endoglycosidase digestion of α_2a- and α_2c-AR preparations were processed and immunoblotted using the HA epitope antibody, 16B12. Mature, Endo H resistant, α_2a- and α_2c-ARs (filled arrow) were noted at apparent molecular mass of ~60 to ~75kDa and ~55 to ~70kDa, respectively. The immature, Endo H sensitive, form of α_2a- and α_2c-ARs (unfilled arrow) was noted to have an apparent molecular mass of ~53 to 55kDa and ~45 to~50kDa, respectively. As expected, both the mature and immature forms of α_2a- and α_2c-ARs were sensitive to PNGase F digestion. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 2. Role of glycosylation in α_2c-AR trafficking

(A) RAT1 fibroblasts were transiently transfected with cDNAs encoding HA epitope-tagged wild-type α_2c-AR and three α_2c-AR glycosylation deficient mutants, α_2c- (N19Q)-, α_2c- (N33Q)- and α_2c- (N19Q/N33Q)-ARs. 48 hrs post-transfection, steady state distributions of α_2c-AR and the glycosylation deficient mutants were determined as described in the legend to Figure 1C. Note that removal of either one (α_2c-(N19Q)-AR, α_2c- (N33Q)-AR) or both (α_2c-(N19Q/N33Q)) glycosylation sites did not effect membrane trafficking; all α_2c-AR constructs exhibited predominant intracellular staining. Calnexin antibody staining shows the ER compartment for comparison. (B) Cellular processing and maturation of wild type a2c-ARs and three glycosylation deficient mutants, α_2c- (N19Q)-, α_2c- (N33Q)- and α_2c- (N19Q/N33Q)-ARs, was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. The α_2c- (N19Q)-, α_2c- (N33Q)-ARs displayed similar cellular processing of glycosylation sites. Both α_2c- (N19Q)-, α_2c- (N33Q)-ARs were present as both mature, Endo H resistant, forms with apparent molecular masses of ~48 to ~60 kDa (filled arrow) and immature, Endo H sensitive, forms of ~ 42 to ~45 kDa (unfilled arrow). Both the mature and immature forms of these receptors were noted to PNGase F sensitive. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 2. Role of glycosylation in α_2c-AR trafficking

(A) RAT1 fibroblasts were transiently transfected with cDNAs encoding HA epitope-tagged wild-type α_2c-AR and three α_2c-AR glycosylation deficient mutants, α_2c- (N19Q)-, α_2c- (N33Q)- and α_2c- (N19Q/N33Q)-ARs. 48 hrs post-transfection, steady state distributions of α_2c-AR and the glycosylation deficient mutants were determined as described in the legend to Figure 1C. Note that removal of either one (α_2c-(N19Q)-AR, α_2c- (N33Q)-AR) or both (α_2c-(N19Q/N33Q)) glycosylation sites did not effect membrane trafficking; all α_2c-AR constructs exhibited predominant intracellular staining. Calnexin antibody staining shows the ER compartment for comparison. (B) Cellular processing and maturation of wild type a2c-ARs and three glycosylation deficient mutants, α_2c- (N19Q)-, α_2c- (N33Q)- and α_2c- (N19Q/N33Q)-ARs, was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. The α_2c- (N19Q)-, α_2c- (N33Q)-ARs displayed similar cellular processing of glycosylation sites. Both α_2c- (N19Q)-, α_2c- (N33Q)-ARs were present as both mature, Endo H resistant, forms with apparent molecular masses of ~48 to ~60 kDa (filled arrow) and immature, Endo H sensitive, forms of ~ 42 to ~45 kDa (unfilled arrow). Both the mature and immature forms of these receptors were noted to PNGase F sensitive. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 3. Chimeric α_2a/α_2c-ARs

Schematic depiction of the structure and cellular localization of the different chimeric α_2a/α_2c-ARs are shown.

Figure 4. Cellular localization of chimeric α_2a/α_2c-ARs

RAT1 fibroblasts were transiently transfected with HA epitope-tagged chimeric α_2a/α_2c-ARs. At 48 hours post transfection the cells were fixed and stained for receptor with the HA antibody, 16B12.

Figure 5. Analysis of α_2c-AR amino terminus deletion mutants

(A) Alignment of α_2a- and α_2c-AR amino termini shows the consensus sequence and the presence of a non-conserved hydrophobic sequence within the α_2c-ARs (Top). A schematic depicting various α_2c-AR deletion constructs is shown (Bottom). (B) Cellular localization of α_2c-ARs and three α_2c-AR amino terminal deletions mutants, α_2c(Δ5–15)-, α_2c(Δ22–28)-, and α_2c(Δ37–45)-ARs, was determined in transient expression in RAT1 fibroblast cells. At 48 hours post transfection the cells were fixed and stained for receptor with 16B12 antibody and the ER compartment with calreticulin antibody. Enhanced PM expression of receptor is only observed for α_2c(Δ5–15)-AR. (C) Comparison of α_2c-AR amino terminus sequence from various mammalian species shows that the hydrophobic sequence, ALAAALAAAAA, is evolutionary conserved from mouse to man.

Figure 5. Analysis of α_2c-AR amino terminus deletion mutants

(A) Alignment of α_2a- and α_2c-AR amino termini shows the consensus sequence and the presence of a non-conserved hydrophobic sequence within the α_2c-ARs (Top). A schematic depicting various α_2c-AR deletion constructs is shown (Bottom). (B) Cellular localization of α_2c-ARs and three α_2c-AR amino terminal deletions mutants, α_2c(Δ5–15)-, α_2c(Δ22–28)-, and α_2c(Δ37–45)-ARs, was determined in transient expression in RAT1 fibroblast cells. At 48 hours post transfection the cells were fixed and stained for receptor with 16B12 antibody and the ER compartment with calreticulin antibody. Enhanced PM expression of receptor is only observed for α_2c(Δ5–15)-AR. (C) Comparison of α_2c-AR amino terminus sequence from various mammalian species shows that the hydrophobic sequence, ALAAALAAAAA, is evolutionary conserved from mouse to man.

Figure 5. Analysis of α_2c-AR amino terminus deletion mutants

(A) Alignment of α_2a- and α_2c-AR amino termini shows the consensus sequence and the presence of a non-conserved hydrophobic sequence within the α_2c-ARs (Top). A schematic depicting various α_2c-AR deletion constructs is shown (Bottom). (B) Cellular localization of α_2c-ARs and three α_2c-AR amino terminal deletions mutants, α_2c(Δ5–15)-, α_2c(Δ22–28)-, and α_2c(Δ37–45)-ARs, was determined in transient expression in RAT1 fibroblast cells. At 48 hours post transfection the cells were fixed and stained for receptor with 16B12 antibody and the ER compartment with calreticulin antibody. Enhanced PM expression of receptor is only observed for α_2c(Δ5–15)-AR. (C) Comparison of α_2c-AR amino terminus sequence from various mammalian species shows that the hydrophobic sequence, ALAAALAAAAA, is evolutionary conserved from mouse to man.

Figure 6. Single charged amino acid substitution disrupts the function of the hydrophobic trafficking signal

(A) Cellular localizations of α_2c-ARs, α_2c(A7D)-, α_2c(L10D)-, α_2c(A13D)-ARs were determined in transiently transfected RAT1 fibroblast cells as described in figure 5B. The α_2c(A7D)-AR displays increased PM expression compared to α_2c-AR.

Figure 7. The α_2c-AR hydrophobic trafficking signal functions within the ER

Cellular processing and maturation of a2c-AR, α_2c(Δ5–15)-, and α_2c(A13D)-ARs was examined by biochemical analysis of N-linked glycosylation of these receptors expressed in HEK293 cells. Membrane preparations were digested with the endoglycosidases, Endo H and PNGase F. Mature, Endo H resistant, form at ~55 to ~70 kDa (filled arrow) and immature, Endo H sensitive, form at ~45 to ~50 kDa (unfilled arrow) are evident, though in different amounts for the various α_2c-ARs. Both α_2c(Δ5–15)- or α_2c(A13D)-AR displayed increase expression of mature, Endo H resistant, protein compared to _2c-ARs. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 8. Transplantation of the α_2c-AR hydrophobic trafficking signal into the α_2a-AR causes ER retention

(A) Schematic depicting various α_2a-AR insertion constructs is shown. The eleven amino acid hydrophobic sequence was inserted utilizing two replacement strategies to create α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. (B) Cellular localization of α_2a-AR and two α_2a-AR insertion mutants, α_2a-I(α_2c[5–15])- and α_2a- IR(α_2c[5–15])-ARs, was examined in transiently transfected RAT1 fibroblast cells. Cell preparations were processed as described in figure 5B. Transplantation of the hydrophobic trafficking signal into the a2a-ARs (α_2a-I(α_2c[5–15])-and α_2a- IR(α_2c[5–15])-ARs) causes ER retention and apparent decreased PM expression of these receptors compared to a2a-ARs. (C) Biochemical analysis of N-linked glycosylation of wild-type α_2a-AR and α_2a-ARs possessing the hydrophobic trafficking signal, α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. Membrane preparations were obtained for HEK293 cells expressing α_2a-, α_2a-I(α_2c[5–15]) and α_2a-IR(α_2c[5–15])-ARs. Mature, Endo H resistant, at ~60–75 kDa (filled arrow), and immature, Endo H sensitive, at ~53 to ~55 kDa (unfilled arrow) forms of receptor were found. Immature forms of both α_2a-I(α_2c[5–15]) and α_2a- IR(α_2c[5–15])-ARs were expressed in increased amounts compared to _2a-ARs. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 8. Transplantation of the α_2c-AR hydrophobic trafficking signal into the α_2a-AR causes ER retention

(A) Schematic depicting various α_2a-AR insertion constructs is shown. The eleven amino acid hydrophobic sequence was inserted utilizing two replacement strategies to create α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. (B) Cellular localization of α_2a-AR and two α_2a-AR insertion mutants, α_2a-I(α_2c[5–15])- and α_2a- IR(α_2c[5–15])-ARs, was examined in transiently transfected RAT1 fibroblast cells. Cell preparations were processed as described in figure 5B. Transplantation of the hydrophobic trafficking signal into the a2a-ARs (α_2a-I(α_2c[5–15])-and α_2a- IR(α_2c[5–15])-ARs) causes ER retention and apparent decreased PM expression of these receptors compared to a2a-ARs. (C) Biochemical analysis of N-linked glycosylation of wild-type α_2a-AR and α_2a-ARs possessing the hydrophobic trafficking signal, α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. Membrane preparations were obtained for HEK293 cells expressing α_2a-, α_2a-I(α_2c[5–15]) and α_2a-IR(α_2c[5–15])-ARs. Mature, Endo H resistant, at ~60–75 kDa (filled arrow), and immature, Endo H sensitive, at ~53 to ~55 kDa (unfilled arrow) forms of receptor were found. Immature forms of both α_2a-I(α_2c[5–15]) and α_2a- IR(α_2c[5–15])-ARs were expressed in increased amounts compared to _2a-ARs. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 8. Transplantation of the α_2c-AR hydrophobic trafficking signal into the α_2a-AR causes ER retention

(A) Schematic depicting various α_2a-AR insertion constructs is shown. The eleven amino acid hydrophobic sequence was inserted utilizing two replacement strategies to create α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. (B) Cellular localization of α_2a-AR and two α_2a-AR insertion mutants, α_2a-I(α_2c[5–15])- and α_2a- IR(α_2c[5–15])-ARs, was examined in transiently transfected RAT1 fibroblast cells. Cell preparations were processed as described in figure 5B. Transplantation of the hydrophobic trafficking signal into the a2a-ARs (α_2a-I(α_2c[5–15])-and α_2a- IR(α_2c[5–15])-ARs) causes ER retention and apparent decreased PM expression of these receptors compared to a2a-ARs. (C) Biochemical analysis of N-linked glycosylation of wild-type α_2a-AR and α_2a-ARs possessing the hydrophobic trafficking signal, α_2a-I(α_2c[5–15]) or α_2a- IR(α_2c[5–15])-ARs. Membrane preparations were obtained for HEK293 cells expressing α_2a-, α_2a-I(α_2c[5–15]) and α_2a-IR(α_2c[5–15])-ARs. Mature, Endo H resistant, at ~60–75 kDa (filled arrow), and immature, Endo H sensitive, at ~53 to ~55 kDa (unfilled arrow) forms of receptor were found. Immature forms of both α_2a-I(α_2c[5–15]) and α_2a- IR(α_2c[5–15])-ARs were expressed in increased amounts compared to _2a-ARs. C – Control; E – Endo H digestion; P – PNGase F digestion.

Figure 9. Neuronal expression of α₂-ARs possessing hydrophobic trafficking signal

The cellular localization of wild type α_2a-ARs and α₂-ARs possessing the hydrophobic trafficking signal, α_2a-I(α_2c[5–15])-, α_2a-IR(α_2c[5–15])- and α_2c-ARs, was examined in transiently transfected PC12 cells. 48 Hours after transfection the cells were fixed and stained with the HA antibody 16B12 and the rabbit polyclonal antibody to cis/media-Golgi compartment, giantin. In the PC12 cell line, the α₂-ARs possessing hydrophobic trafficking signal were primarily expressed in the PM and in a small intracellular compartment that localized to the cis/media Golgi compartment with giantin.