Segregation of nascent GPCRs in the ER-to-Golgi transport by CCHCR1 via direct interaction

ABSTRACT

G protein-coupled receptors (GPCRs) constitute the largest superfamily of cell surface signaling proteins that share a common structural topology. When compared with agonist-induced internalization, how GPCRs are sorted and delivered to functional destinations after synthesis in the endoplasmic reticulum (ER) is much less well understood. Here, we demonstrate that depletion of coiled-coil α-helical rod protein 1 (CCHCR1) by siRNA and CRISPR-Cas9 significantly inhibits surface expression and signaling of α_2A-adrenergic receptor (α_2A-AR; also known as ADRA2A), without affecting α_2B-AR. Further studies show that CCHCR1 depletion specifically impedes α_2A-AR export from the ER to the Golgi, but not from the Golgi to the surface. We also demonstrate that CCHCR1 selectively interacts with α_2A-AR. The interaction is mediated through multiple domains of both proteins and is ionic in nature. Moreover, mutating CCHCR1-binding motifs significantly attenuates ER-to-Golgi export, surface expression and signaling of α_2A-AR. Collectively, these data reveal a novel function for CCHCR1 in intracellular protein trafficking, indicate that closely related GPCRs can be sorted into distinct ER-to-Golgi transport routes by CCHCR1 via direct interaction, and provide important insights into segregation and anterograde delivery of nascent GPCR members.

Highlighted Article: CCHCR1 differentially regulates the ER-to-Golgi transport, cell surface expression and signaling of two closely related G protein-coupled receptors and this sorting function is mediated through direct interaction.

INTRODUCTION

G protein-coupled receptors (GPCRs) regulate a wide variety of cell functions under physiological and pathological conditions and represent the therapeutic targets of approximately one-third of the drugs on the market (Allen and Roth, 2011; Bradley and Tobin, 2016). One important factor that fine tunes the precise functions of GPCRs is their intracellular trafficking, including anterograde delivery of newly synthesized receptors en route from the endoplasmic reticulum (ER) through the Golgi to the cell surface, internalization of the surface receptors to endosomes after agonist stimulation, recycling of internalized receptors back to the surface, and transport to lysosomes and proteasomes for degradation. These trafficking processes collectively determine the amount of receptors expressed at the cell surface, which in turn affects cellular responses to hormones and drugs at a given time (Kunselman et al., 2021; Marchese et al., 2008; Wang and Wu, 2012; Zhang and Wu, 2019). When compared with well-characterized internalization, the molecular mechanisms underlying the surface targeting of nascent GPCRs are relatively much less well understood. In particular, how distinct GPCR members are sorted from one another during their maturation processing after synthesis in the ER is poorly studied.

Coiled-coil α-helical rod protein 1 (CCHCR1) was identified as a putative candidate gene for psoriasis (Asumalahti et al., 2000, 2002; Elomaa et al., 2004; Oka et al., 1999; Suomela et al., 2003), a common skin disease. The CCHCR1 protein is expressed ubiquitously and localized in multiple subcellular locations, including the cytoplasm, nuclei, mitochondria, cytoplasmic processing bodies (P-bodies) and centrosomes (Andersen et al., 2003; Asumalahti et al., 2002; Corbi et al., 2005; Ling et al., 2014; Pisani et al., 2021; Sugawara et al., 2003; Tervaniemi et al., 2018, 2012; Ying et al., 2022). CCHCR1 has been demonstrated to regulate steroid synthesis (Sugawara et al., 2003; Tiala et al., 2007), cytoskeletal organization (Tervaniemi et al., 2018, 2012), cell proliferation and differentiation (Pacholska-Bogalska et al., 2012; Tiala et al., 2008) and centrosome replication (Ying et al., 2022). Recent studies have also shown that, in addition to psoriasis, CCHCR1 and its naturally occurring mutations are associated with type-2 diabetes (Brenner et al., 2020), cancer (Chang et al., 2018; Shete et al., 2020; Suomela et al., 2009) and chronic obstructive pulmonary disease (Moll et al., 2021). However, virtually nothing is known about its function in intracellular protein trafficking.

In this study, we investigate the possible role of CCHCR1 in GPCR trafficking by focusing on the anterograde transport from the ER through the Golgi to the plasma membrane (PM) and using α₂-adrenergic receptors (α₂-ARs) as models. α₂-ARs are prototypic family A GPCR members with three subtypes, designated as α_2A-AR (also known as ADRA2A), α_2B-AR (ADRA2B) and α_2C-AR (ADRA2C). α_2A-AR and α_2B-AR are robustly expressed at the cell surface at steady state, whereas α_2C-AR surface expression depends on cell types (Daunt et al., 1997). α₂-ARs couple the Gi/Go family G proteins and regulate diverse functions of the sympathetic nervous system, both peripherally and centrally (Gavras et al., 2001; Ruffolo et al., 1993; Xu et al., 2022a). Here, we demonstrate that CCHCR1 depletion selectively affects the ER-to-Golgi transport, surface expression and function of α_2A-AR, but not α_2B-AR, and that this sorting action of CCHCR1 is mediated through direct interaction. These data reveal a novel mechanism by which GPCR members are segregated in the biosynthetic trafficking.

RESULTS

Depletion of CCHCR1 by siRNA and CRISPR-Cas9 inhibits the cell surface expression and signaling of α_2A-AR, but not α_2B-AR

As an initial approach to investigate the function of CCHCR1 in the biosynthetic trafficking of GPCRs, we determined the effects of siRNA-mediated knockdown of CCHCR1 (Fig. 1A,B) on the surface expression of α_2A-AR and α_2B-AR at steady state in HEK293 cells. After the cells were transfected with individual α₂-ARs together with control siRNA (siCtrl) or siRNA targeting CCHCR1 (siCCHCR1), the numbers of α₂-ARs at the surface were quantified by radioligand binding of intact live cells using specific α₂-AR antagonist [³H]RX821002. The surface expression of α_2A-AR was significantly attenuated by ∼40% in cells transfected with CCHCR1 siRNA as compared with that in control siRNA-transfected cells (Fig. 1C). Expression of an siRNA-resistant form of CCHCR1 completely rescued the cell surface transport of α_2A-AR in cells transfected with CCHCR1 siRNA (Fig. 1A–C). Consistent with radioligand binding data, subcellular localization analysis showed that, as expected, α_2A-AR was robustly expressed at the cell surface in control cells, whereas it was extensively accumulated in the perinuclear regions in cells transfected with CCHCR1 siRNA (Fig. 1D). Surprisingly, the surface expression (Fig. 1C) and subcellular localization (Fig. 1D) of α_2B-AR was very similar in cells transfected with control and CCHCR1 siRNA.

Fig. 1.

Effects of CCHCR1 depletion by siRNA and CRISPR-Cas9 on the surface expression, subcellular localization and signaling of α₂-ARs. (A) Western blot analysis of siRNA-mediated knockdown of CCHCR1 and expression of siRNA-resistant CCHCR1. HEK293 cells were transfected with Myc–CCHCR1 or its siRNA-resistant version together with control siRNA or individual siRNA targeting CCHCR1. The expression of Myc-CCHCR1 was measured by immunoblotting using anti-Myc antibodies. (B) Quantitative data for results shown in A. (C) Effects of siRNA-mediated knockdown of CCHCR1 on the cell surface expression of α_2A-AR and α_2B-AR. HEK293 cells were transfected with α_2A-AR or α_2B-AR together with control or CCHCR1 siRNA and the surface expression of α₂-ARs was determined by intact cell ligand binding using [³H]RX821002. The data shown are percentages of specific binding obtained from cells transfected with control siRNA. In a typical experiment, the mean values of specific [³H]RX821002 binding were 15,931±440 and 10,194±477 cpm in control cells transfected with α_2A-AR and α_2B-AR, respectively. (D) Effects of CCHCR1 knockdown by siRNA on the subcellular localization of α_2A-AR and α_2B-AR. HEK293 cells were transfected with GFP-tagged α_2A-AR or α_2B-AR together with control or CCHCR1 siRNA. Images are representatives of three repeats. (E) Western blot analysis of CCHCR1 depletion by CRISPR-Cas9. HEK293 cells were transfected with control or CRISPR-Cas9 KO plasmids together with Myc–CCHCR1. (F) Quantitative data for results shown in E. (G) Effects of CCHCR1 KO by CRISPR-Cas9 on the subcellular distribution of α_2A-AR and α_2B-AR. The cells transfected with control and CCHCR1 KO plasmids were defined by the GFP signal. (H) Quantitative data for results shown in G. The quantitative data are the surface-to-total expression ratio. (I) Effects of CCHCR1 knockdown by siRNA on the surface expression of endogenous α_2A-AR. HT29 cells were transfected with control or CCHCR1 siRNA for 48 h and the numbers of α_2A-AR in membrane fractions were determined by ligand binding using [³H]-RX821002. The data shown are the receptor numbers per µg membrane protein. (J) Effects of CCHCR1 knockdown on ERK1/2 activation by exogenously transfected α₂-AR-GFP in HEK293 cells and by endogenous α_2A-AR in HT29 cells. The cells were stimulated with UK14304 at 1 μM for 5 min. Receptor expression was detected by using GFP antibodies. (K) Quantitative data for results shown in J. All quantitative data are expressed as mean±s.e.m. (n=23–35 cells in three separate experiments in H and n=3 in B, C, F, I and K). *P<0.05, **P<0.01, ***P<0.001 versus respective control (one-way ANOVA with Turkey's multiple comparisons post test in B, C, I and K, and unpaired two-tailed t-test in F and H). Scale bars: 10 μm.

We next measured the effects of CCHCR1 knockout (KO) by CRISPR-Cas9 (Fig. 1E,F) on the subcellular distribution of α_2A-AR and α_2B-AR. For this purpose, we took advantage of GFP-bearing CRISPR-Cas9 KO plasmids of CCHCR1 to analyze receptor subcellular localization in CCHCR1-depleted cells, which were defined by the GFP signal. In control cells, both α_2A-AR and α_2B-AR tagged with RFP were strongly expressed at the cell surface after transient expression for 24 h (Fig. 1G). In CCHCR1 KO cells, α_2A-AR was clearly arrested in the perinuclear regions, whereas α_2B-AR was normally exported to the cell surface (Fig. 1G). The quantitative data showed that α_2A-AR surface expression relative to the total expression (surface/total) was reduced by ∼40% in CCHCR1-depleted cells as compared with control cells, whereas α_2B-AR surface expression was about the same in control and CCHCR1 KO cells (Fig. 1H).

To study the role of CCHCR1 in the cell surface expression of endogenous α_2A-AR, we used HT29 cells which express α_2A-AR, but not α₂-AR and α_2C-AR (Bylund and Ray-Prenger, 1989). As measured in radioligand binding assays, the expression of endogenous α_2A-AR at the PM in HT29 cells was moderately but significantly lower in CCHCR1 siRNA-transfected cells than that in control cells (Fig. 1I).

α₂-ARs regulate the activation of adenylyl cyclases, mitogen-activated protein kinases (MAPKs) and voltage-gated Ca²⁺ channels (Dong et al., 2011; Li et al., 1998; Xu and Wu, 2023). To determine whether CCHCR1 depletion could influence the function of α₂-ARs, the activation of ERK1 and ERK2 (ERK1/2, also known as MAPK3 and MAPK1, respectively) was used as a readout. In parallel with its effects on the surface expression of α_2A-AR and α_2B-AR, siRNA targeting CCHCR1 dramatically suppressed ERK1/2 activation by α_2A-AR, but not α_2B-AR, in response to stimulation with UK14304, an α₂-AR agonist, in HEK293 cells (Fig. 1J,K). Transfection of CCHCR1 siRNA also inhibited ERK1/2 activation by endogenous α_2A-AR after UK14304 stimulation in HT29 cells (Fig. 1J,K). These data demonstrate that CCHCR1 is able to selectively regulate the surface expression and function of α_2A-AR, but not α_2B-AR.

CCHCR1 depletion impedes the ER-to-Golgi transport of nascent α_2A-AR, but not α_2B-AR

To define the transport steps that are regulated by CCHCR1, we determined the effects of CCHCR1 depletion by siRNA and CRISPR-Cas9 on the transport of newly synthesized α_2A-AR and α_2B-AR from the ER to the Golgi by performing ‘retention using the selective hooks’ (RUSH) assays (Boncompain et al., 2012; Xu and Wu, 2022). In RUSH assays, both α₂-ARs are fused to GFP and a streptavidin-binding peptide (SBP) at their N-termini (NT) and the ER retention signal KDEL fused to streptavidin (Str-KDEL) was used as a hook (Fig. 2A). The ER export of nascent receptors was synchronized after addition of biotin and their ER-to-Golgi transport was measured by increase in the Golgi expression relative to the total expression (Golgi/total). Our previous studies have shown that before biotin induction, α_2A-AR and α_2B-AR are retained in the ER; biotin incubation induces their export from the ER to the juxtanuclear region and the receptors become extensively colocalized with the Golgi markers (Xu et al., 2023; Xu and Wu, 2022). Therefore, in our analyses of this study, the receptors that were highly concentrated in the juxtanuclear region after biotin induction were considered to be transported to the Golgi, without using Golgi markers.

Fig. 2.

Effects of CCHCR1 depletion by siRNA and CRISPR-Cas9 on the ER-to-Golgi transport of nascent α_2A-AR and α_2B-AR in RUSH assays. (A) Cartoon of the RUSH system for the transport of newly synthesized α₂-ARs from the ER to the Golgi. Str, streptavidin. (B) Representative images showing the ER-to-Golgi transport kinetics of nascent α_2A-AR and α_2B-AR in RUSH assays. HEK293 cells were transfected with RUSH plasmids expressing α_2A-AR or α_2B-AR for 24 h and then incubated with biotin for 10, 20 and 30 min. After the cells were fixed, receptor expression at the Golgi (area denoted by the white dotted line) and in the whole cell (area denoted by the yellow dotted line) was quantified. (C) Quantitative data for results shown in B. (D,E) Effects of siRNA-mediated knockdown of CCHCR1 on α₂-AR export from the ER to the Golgi. HEK293 cells were transfected with RUSH plasmids expressing α_2A-AR (D) or α_2B-AR (E) together with control or CCHCR1 siRNA and incubated with biotin for 15 and 30 min. (F) Quantitative data for results shown in D and E. (G,H) Effects of CRISPR-Cas9-mediated KO of CCHCR1 on α₂-AR export from the ER to the Golgi. HEK293 cells were transfected with RUSH plasmids expressing mCherry-tagged α_2A-AR (G) or α_2B-AR (H) together with control or CCHCR1 KO plasmids carrying GFP and incubated with biotin for 15 and 30 min. (I) Quantitative data for results shown in G and H. All quantitative data shown are the Golgi-to-total ratio and expressed as mean±s.e.m. (n=20–32, 37–54 and 37–54 cells in C, F and I, respectively, from at least three individual experiments). ***P<0.001 versus control (one-way ANOVA with Turkey's multiple comparisons post test in F and by unpaired two-tailed t-test in I). Scale bars: 10 μm.

We first compared the ER-to-Golgi transport kinetics of α_2A-AR and α_2B-AR over 30 min after biotin induction and their transport curves were superimposed (Fig. 2B,C). As such, the effects of CCHCR1 depletion on the ER-to-Golgi export of α₂-ARs were measured at 15 and 30 min after biotin incubation. The ER-to-Golgi transport of α_2A-AR was significantly reduced by 30–40% in cells transfected with siRNA targeting CCHCR1 as compared with that in control cells (Fig. 2D,F). In contrast, the ER-to-Golgi transport of α_2B-AR was very similar in cells transfected with control and CCHCR1 siRNA (Fig. 2E,F). CCHCR1 KO by CRISPR-Cas9 also markedly inhibited the ER-to-Golgi transport of α_2A-AR by more than 40% (Fig. 2G,I), whereas it did not alter the ER-to-Golgi transport of α_2B-AR after biotin incubation (Fig. 2H,I). These data suggest that CCHCR1 depletion specifically affects the ER-to-Golgi traffic of α_2A-AR, but not α_2B-AR.

CCHCR1 depletion does not affect the Golgi-to-PM transport of nascent α_2A-AR and α_2B-AR

We next determined whether CCHCR1 depletion could affect the Golgi-to-PM transport of nascent α_2A-AR and α_2B-AR. For this purpose, the cells transfected with individual α₂-AR RUSH plasmids were incubated with biotin at 20°C for 3 h, which allowed nascent receptors to leave the ER but unable to exit the Golgi. Receptor export from the Golgi to the PM was initiated by incubation at 37°C and reflected by reduction in the ratio of the Golgi expression to total expression at 30 min (Fig. 3A). For both α_2A-AR and α_2B-AR, ∼60% of the total receptors were transported to and accumulated in the Golgi after 3 h biotin induction at 20°C (time 0 min) (Fig. 3B–D). After 30 min incubation at 37°C (time 30 min), ∼15% of the total receptors remained at the Golgi, indicating that more than 70% of the Golgi-accumulated receptors were exported to the cell surface (Fig. 3B–D). No statistical significances were found between two α₂-ARs and between cells transfected with control and CCHCR1 siRNA before (time 0 min) and after incubation at 37°C (time 30 min) (Fig. 3D). Similarly, CCHCR1 KO by CRISPR-Cas9 had no obvious effect on the Golgi-to-surface transport of both α₂-ARs after incubation for 30 min at 37°C following biotin induction for 3 h at 20°C (Fig. 3E–G). These data demonstrate that CCHCR1 does not play a major role in the transport of α₂-ARs from the Golgi to the cell surface.

Fig. 3.

Effects of CCHCR1 depletion by siRNA and CRISPR-Cas9 on the Golgi-to-PM transport of α₂-ARs. (A) Cartoon of the transport of nascent α₂-ARs from the Golgi to the cell surface measured in RUSH assays in combination with temperature-induced block of Golgi export. Str, streptavidin. (B,C) Effects on siRNA-mediated CCHCR1 knockdown on α₂-AR export from the Golgi to the PM. HEK293 cells transfected with RUSH plasmids expressing α_2A-AR (B) or α_2B-AR (C) together with control or CCHCR1 siRNA were treated with biotin at 20°C for 3 h (time 0 min) and then incubated with fresh DMEM without biotin at 37°C for 30 min (time 30 min). (D) Quantitative data for results shown in B and C. (E,F) Effects of CRISPR-Cas9-mediated KO of CCHCR1 on α₂-AR export from the Golgi to the PM. The cells transfected with RUSH plasmids expressing mCherry-tagged α_2A-AR (E) or α_2B-AR (F) together with control or CCHCR1 KO plasmids were incubated at 37°C for 30 min (time 30 min) following biotin induction at 20°C for 3 h (time 0 min). (G) Quantitative data for results shown in E and F. All quantitative data are the Golgi-to-total ratio and expressed as mean±s.e.m. (n=25–52 and 23–36 cells in D and G, respectively, from at least three individual experiments). ***P<0.001 versus respective 0 min (one-way ANOVA with Turkey's multiple comparisons post test). Scale bars: 10 μm.

Selective CCHCR1 interaction with α_2A-AR and identification of CCHCR1-binding motifs

Our above data have demonstrated a selective action of CCHCR1 on the transport of α_2A-AR, but not α_2B-AR. To delineate the possible underlying mechanisms, we measured the interaction of CCHCR1 with both α₂-ARs in co-immunoprecipitation (co-IP) assays. The CCHCR1 isoform, consisting of 756 residues, was tagged with Myc and expressed together with HA-tagged α_2A-AR (Fig. 4A) or α_2B-AR (Fig. 4B) in HEK293 cells. α_2A-AR, but not α_2B-AR, was found to form a complex with CCHCR1 after immunoprecipitation with HA antibodies (Fig. 4A,B).

Fig. 4.

CCHCR1 interaction with α₂-ARs and identification of CCHCR1-binding motifs in the ICL3 of α_2A-AR. (A,B) Co-IP of CCHCR1 and α_2A-AR (A) or α_2B-AR (B). HEK293 cells were transfected with HA–α_2A-AR (A) or HA–α_2B-AR (B) together with empty vectors (Ctrl) or Myc–CCHCR1 and subjected to immunoprecipitation (IP) with HA antibodies. Myc-CCHCR1 and HA-α₂-AR in the IP complex were detected by immunoblotting (IB) using Myc and HA antibodies, respectively. (C) Interaction of the ICL3 and the CT of α_2A-AR and α_2B-AR with Myc–CCHCR1 in GST fusion protein pulldown assays. The ICL3 and the CT of α₂-ARs were generated as GST fusion proteins and incubated with cell lysates prepared from HEK293 cells expressing Myc–CCHCR1. Bound CCHCR1 was detected by immunoblotting using Myc antibodies. (D) Progressive deletion to identify the CCHCR1-binding domain in the α_2A-AR ICL3. Different ICL3 fragments were generated as GST fusion proteins and their interactions with Myc–CCHCR1 were measured. (E) Alignment of the ICL3 of α_2A-AR and α_2B-AR by using EMBOSS Needle. The fragment S296–L332 in the α_2A-AR ICL3 that binds CCHCR1 as demonstrated in D is colored red. Basic residues in this domain are bolded. Acidic residues in the corresponding region of the α_2B-AR ICL3 are colored green. (F) Effects of increasing concentrations of NaCl on CCHCR1 interaction with the fragment S296-L332. (G) Quantitative data for results shown in F. The quantitative data are presented as mean±s.e.m. (n=3). GST fusion proteins used in individual experiments are shown. The results shown in each panel are representatives of at least 3 repeats.

To identify the CCHCR1-binding domains in α_2A-AR and determine whether CCHCR1 could differentially interact with the intracellular domains of α_2A-AR and α_2B-AR, the third intracellular loops (ICL3) and the C-termini (CT) of α_2A-AR and α_2B-AR were generated as GST fusion proteins and their abilities to bind CCHCR1 were measured. The ICL3 and the CT of α₂-ARs have been demonstrated to mediate receptor interaction with a number of proteins involved in regulation of receptor trafficking and signaling (Duvernay et al., 2011; Li et al., 2012, 2017; Pao and Benovic, 2005; Taylor et al., 1994; Wang et al., 2004; Wu et al., 1998, 1997; Xu and Wu, 2022; Zhang et al., 2016a). GST fusion proteins containing the ICL3 or CT of α_2A-AR strongly interacted with CCHCR1 (Fig. 4C). In contrast, GST alone and GST fusion proteins containing the ICL3 or CT of α_2B-AR had no or much weaker binding to CCHCR1 (Fig. 4C). The progressive deletion strategy was used to identify the ICL3 subdomain binding CCHCR1 and the shortest fragment identified was S296–L332, with further shortening abolishing the interaction with CCHCR1 (Fig. 4D). These data are consistent with the amino acid sequence alignment of α_2A-AR and α_2B-AR in which the greatest variability was found in their ICL3 (Fig. 4E). In particular, the fragment S296–L332 of α2A-AR possesses 10 positively charged residues, whereas α_2B-AR contains multiple acidic residues in the corresponding region (Fig. 4E). Therefore, we determined the effect of increasing salt concentrations on S296–L332 interaction with CCHCR1. Increasing NaCl concentrations inhibited S296–L332 interaction with CCHCR1 and the inhibitory effects were in a dose-dependent fashion (Fig. 4F,G).

To identify the CCHCR1-binding residues in the α_2A-AR CT, we focused on the RKR motif, which is absent in the α_2B-AR CT (Fig. 5A). Mutating the RKR motif to AAA or EEE abolished the interaction of the CT with CCHCR1 in the GST fusion protein pulldown assays (Fig. 5B,C). In addition, similar to its effects on S296–L332 interaction with CCHCR1, enhancing NaCl concentrations dose dependently suppressed the interaction of the CT with CCHCR1 (Fig. 5D,E). These data demonstrate that CCHCR1 selectively interacts with α_2A-AR, specifically its ICL3 and CT, and CCHCR1 interactions with the ICL3 and CT are most likely ionic.

Fig. 5.

Identification of CCHCR1-binding residues in the CT of α_2A-AR. (A) Alignment of the CT of α2A-AR and α2B-AR. The membrane-proximal CT of both receptors form helix 8. The RKR motif in α_2A-AR is colored red. (B) Effects of mutating the RKR motif to AAA and EEE on the interaction of the α_2A-AR CT with Myc–CCHCR1 in GST fusion protein pulldown assays. (C) Quantitative data shown for results shown in B. (D) Effects of increasing concentrations of NaCl on CCHCR1 interaction with the α_2A-AR CT. (E) Quantitative data for results shown in D. GST fusion proteins used in individual experiments are shown. The quantitative data are presented as mean±s.e.m. (n=3). ***P<0.001 versus WT CT.

Identification of α_2A-AR-binding domains of CCHCR1

The structure of CCHCR1 contains a relatively long non-structural region in the NT and 5 α-helixes as predicated by AlphaFold (Fig. 6A). To identify the domains mediating CCHCR1 interaction with α_2A-AR, CCHCR1 was split into five fragments containing different helical domains and their abilities to interact with the ICL3 and the CT of α_2A-AR were tested in GST fusion protein pulldown assays (Fig. 6B). The fragment 1–276 strongly bound both the ICL3 and the CT. The fragment 666–756 also strongly bound the CT, but its interaction with the ICL3 was much weaker. In contrast, the fragments 58–276, 277–471 and 472–665 did not interact with the ICL3 and the CT (Fig. 6C). These data suggest that the N-terminal 57 residues and the C-terminal 92 residues containing helix 5 are important domains for CCHCR1 interaction with α_2A-AR.

Fig. 6.

Identification of α_2A-AR-binding domains in CCHCR1. (A) The structure of CCHCR1 predicted by AlphaFold, containing 5 helixes. (B) The strategy to identify specific α_2A-AR-binding domains in CCHCR1. The domains are depicted to scale. The right panel summarizes the results shown in C. +++, strong interaction; +, weak interaction; –, no detectable interaction. (C) Interactions of Myc-tagged fragments of CCHCR1 with the ICL3 and the CT of α_2A-AR in GST fusion protein pulldown assays. Similar results were obtained in at least three repeats. (D) Sequence and structure of helix 5 of CCHCR1. Three acidic residues which were mutated to alanine residues (DDE-3A) are colored red. (E) Effect of mutating the three acidic residues on the interaction of the CCHCR1 fragment 666–756 with the α_2A-AR CT. (F) Quantitative data for results shown in E. The data are the bound CCHCR1 fragment relative to the input (lysate) and presented as mean±s.e.m. (n=3). ***P<0.001 versus WT 666–756 (unpaired two-tailed t-test).

To further identify specific residues in CCHCR1 responsible for interaction with α_2A-AR, we focused on acidic residues. The NT fragment 1–57 possesses two acidic residue-rich domains, DVSERRLD and ERDVSSDRQE. However, the CCHCR1 mutant in which all seven acidic residues in these two domains were mutated to alanine residues was almost undetectable in the total cell lysates after transient expression (data not shown), suggesting that these acidic residues play a crucial role in stabilizing CCHCR1. The CT fragment 666–756 has one acidic residue-rich domain, DDLQDLSEAISKEE, in helix 5 in which the side chains of three acidic residues (D723, D727 and E730) project from the same face of the helix (Fig. 6D). Mutation of these 3 acidic residues to alanine residues (DDE-3A) remarkably reduced the interaction of the fragment 666–756 with the α_2A-AR CT by ∼70% as measured in GST fusion protein pulldown assays (Fig. 6E,F). These data demonstrate that the acidic domain in helix 5 mediates at least partially CCHCR1 interaction with α_2A-AR.

We next determined whether α_2A-AR interaction and export from the ER could affect the subcellular localization of CCHCR1. Consistent with previous studies (Andersen et al., 2003; Tervaniemi et al., 2018, 2012; Ying et al., 2022), CCHCR1 was highly concentrated in centrosomes (Fig. S1A). CCHCR1 was also partially colocalized with the ER marker DsRed-ER, which was not affected by mutation of the motif DDA (Fig. S1B). In RUSH assays, CCHCR1 and α_2A-AR were partially colocalized in the absence of biotin (Fig. S1C, upper panel), and biotin incubation to induce α_2A-AR export from the ER to the Golgi did not apparently affect the overall subcellular distribution of CCHCR1 (Fig. S1C).

Mutation of CCHCR1-binding sites blocks the ER-to-Golgi export, surface expression and signaling of α_2A-AR

To determine the functional importance of CCHCR1 interaction, we generated the α_2A-AR mutant Δ37/3E in which the fragment S296–L332 (37 residues) of the ICL3 was deleted and simultaneously the RKR motif in the CT was mutated to EEE, and compared its ER-to-Golgi transport, surface expression and signaling with wild-type (WT) α_2A-AR. As measured in RUSH assays, the ER-to-Golgi transport of the mutant Δ37/3E was clearly attenuated by ∼40% as compared with its WT counterpart (Fig. 7A,B). Neither the subcellular localization nor the ER-to-Golgi transport of the mutant Δ37/3E was affected by CCHCR1 siRNA (Fig. S2A–C). We next used RUSH-based bioluminescence resonance energy transfer (BRET) assays (Xu and Wu, 2022) to measure the surface expression of α_2A-AR and its mutant Δ37/3E. In this assay, the RUSH plasmids with Rluc8 fused to the CT of SBP-α_2A-AR were co-transfected with the PM marker Venus–kRas, and the surface expression of α_2A-AR was measured by BRET after biotin induction for 2 h (Fig. 7C). The cell surface expression of Δ37/3E was significantly lower than that of WT α_2A-AR (Fig. 7D). The surface expression of Δ37/3E was also markedly decreased as compared with WT α_2A-AR as measured in both intact cell radioligand binding (Fig. 7E) and β-lactamase reporter assays (Fig. 7F). Consistent with the reduction of surface expression of Δ37/3E, ERK1/2 activation by Δ37/3E was significantly suppressed as compared with that by WT α_2A-AR (Fig. 7G,H). These data demonstrate that α_2A-AR interaction with CCHCR1 is required for its export from the ER and subsequent transport to the cell surface where it activates the MAPK pathway.

Fig. 7.

Mutation of CCHCR1-binding motifs impedes the ER-to-Golgi export, surface expression and signaling of α_2A-AR. (A) The ER-to-Golgi transport of α_2A-AR and its mutant Δ37/3E lacking the CCHCR1-binding sites in the ICL3 and the CT. HEK293 cells were transfected with RUSH plasmids expressing mCherry-tagged α_2A-AR or Δ37/3E and incubated with biotin for 15 min. (B) Quantitative data for results shown in A. The quantitative data are the Golgi-to-total expression ratio. (C) Schematic diagram showing RUSH-based BRET assays to measure the cell surface expression of α_2A-AR. Str, streptavidin. (D) The surface expression of α_2A-AR and Δ37/3E as measured in RUSH-based BRET assays in live cells. HEK293 cells were transfected with RUSH plasmids expressing α_2A-AR or Δ37/3E together with Venus–kRas. After incubation with biotin for 2 h, the surface expression of α_2A-AR was measured by BRET assays. (E) The surface expression of α_2A-AR and Δ37/3E as measured by intact cell radioligand binding after transient expression for 24 h. (F) The surface expression of α_2A-AR and Δ37/3E as measured by β-lactamase reporter assays. HEK293 cells were transfected with β-lactamase-tagged α_2A-AR or Δ37/3E and then incubated with nitrocefin. The absorbance at 486 nm was measured for 30 min and the rate of reaction was used as the readout of receptor surface expression. (G) ERK1/2 activation by α_2A-AR and Δ37/3E. HEK293 cells were transfected with GFP-tagged α_2A-AR or Δ37/3E and then stimulated with UK14304 at 1 μM for 5 min. Receptor expression was detected by using GFP antibodies. (H) Quantitative data for results shown in G. All quantitative data are presented as mean±s.e.m. (n=31–47 cells from at least three repeats in B; n=3 in D, E, F and H). *P<0.05, ***P<0.001 versus α_2A-AR (unpaired two-tailed t-test). Scale bars: 10 μm. (I) A model depicting the function and mechanism of CCHCR1 in the sorting of newly synthesized α_2A-AR and α_2B-AR in their ER-to-Golgi transport. α_2A-AR interacts with CCHCR1 and the interaction is mediated through positively charged residues (+) in the ICL3 and the CT of α_2A-AR and negatively charged residues (–) in both termini of CCHCR1. In particular, the RKR motif in α_2A-AR and the DDE motif in CCHCR1 have been demonstrated to be important for the interaction.

DISCUSSION

The GPCR superfamily has more than 800 members that all share a common structural topology. A fundamental but poorly addressed question in studies of GPCR trafficking is how they are sorted from one another during their targeted delivery to the functional destinations after synthesis in the ER. In this paper, we have investigated the function of CCHCR1 in the anterograde transport of α_2A-AR and α_2B-AR. The most important finding presented here is the soring action of CCHCR1 on these two α₂-ARs and this function of CCHCR1 is mediated through direct interaction with the receptor it regulates.

We have demonstrated that CCHCR1 depletion by siRNA and CRISPR-Cas9 clearly inhibits the surface expression of exogenously transfected and endogenously expressed α_2A-AR at steady state, as well as its signaling to ERK1/2. In marked contrast, CCHCR1 depletion has no effect on the surface expression and signaling of α_2B-AR. We have further shown that CCHCR1 depletion markedly compromises the ER-to-Golgi transport of nascent α_2A-AR as measured in RUSH assays, without affecting the ER-to-Golgi transport of α_2B-AR and the Golgi-to-PM traffic of both α₂-ARs. These data suggest that CCHCR1 may control the surface expression and function of α_2A-AR through regulating the ER-to-Golgi transport dynamics. These data also indicate that the selection of α_2A-AR or segregation of the α₂-ARs by CCHCR1 occurs at the level of the ER (Fig. 7I). Given that the ER-to-Golgi transport was measured by the amount of the receptors that reached the Golgi, the function of CCHCR1 in the ER-to-Golgi transport of α_2A-AR might be mediated through regulating receptor folding and assembly in the ER, its recruitment to COPII vesicles at the ER exit sites, and/or forward moving of vesicles carrying the receptors.

Our data described here have also provided direct evidence indicating that the sorting function of CCHCR1 is directed by its selective interaction with the receptors (Fig. 7I). We have demonstrated that CCHCR1 specifically interacts with α_2A-AR, but not α_2A-AR, in both co-IP and GST fusion protein pulldown assays and that the interaction between α_2A-AR and CCHCR1 is mediated via both termini of CCHCR1 and the ICL3 and CT of the receptor. As the CCHCR1–α_2A-AR interaction is markedly attenuated by increasing NaCl concentrations, mutation of the RKR motif in the CT of α_2A-AR and mutation of the DDE motif in the helix 5 of CCHCR1, their interaction is ionic in nature (Fig. 7I).

More importantly, we have shown that the α_2A-AR mutant Δ37/3E lacking the CCHCR1-binding sites identified in both the ICl3 and the CT is clearly detective in the transport from the ER to the Golgi and the cell surface and in signaling to MAPKs. These data strongly imply that CCHCR1 interaction is a prerequisite for the normal export of nascent α_2A-AR from the ER. However, we cannot exclude the possibilities that the removal of the CCHCR1-binding sites might affect α_2A-AR interaction with other regulatory proteins involved in maturation processing, which could also contribute to the observed defective trafficking. Indeed, the CCHCR1-binding domain S296–L332 binds C1orf27, an ER-anchored protein that was recently shown to regulate α_2A-AR export from the ER to the Golgi (Xu and Wu, 2022). In addition, α_2A-AR export from the ER has been shown to be regulated by Rab43, Sar1, TBC domain-containing proteins (Li et al., 2017; Wei et al., 2019; Xu et al., 2023). We also cannot rule out the possible contribution to the reduced ERK1/2 activation by defective coupling of Δ37/3E to downstream signaling molecules and/or its defective binding to agonists. Nevertheless, these data, together with previous studies, demonstrate that α_2A-AR export from the ER to the Golgi is under control by interactions with multiple regulatory proteins and provide important evidence to support a notion that the cargo GPCRs can physically associate with regulatory proteins or transport machinery to control their own export trafficking to the cell surface after synthesis in the ER (Li et al., 2012, 2017; Xu et al., 2022b; Xu and Wu, 2022; Zhang et al., 2016a).

CCHCR1 was identified more than 2 decades ago; its functions, however, remain largely undefined. Here, we have revealed a novel function for CCHCR1 in α_2A-AR export along the early secretory pathway, thus the data presented in this paper represent the first demonstration of CCHCR1 having a role in intracellular protein trafficking. CCHCR1 is structurally predicted to form coiled-coil domains in the first four α-helixes (Asumalahti et al., 2002; Ling et al., 2014). Although the functions of coiled-coil domains can vary depending on the specific protein context in which they are present, they have been well defined to regulate protein–protein interaction, structural stabilization, protein localization and filament formation. A recent study has demonstrated that CCHCR1, via its CT coiled-coil domains, interacts with astrin, a centrosome-related protein, leading to its recruitment to centrosomes where it controls centriole duplication and microtubule assembly (Ying et al., 2022). We have demonstrated here that the NT non-structural region and the CT portion including helix 5 which does not form coiled-coil motifs mediate CCHCR1 interaction with α_2A-AR, suggesting that the coiled-coil motifs of CCHCR1 are not involved in interaction with α_2A-AR. These data imply that CCHCR1 might use different domains to physically associate with distinct proteins to carry out diverse functions in cells.

It is worth noting that our previous studies have demonstrated that GGA3 modulates the post-Golgi transport of α_2B-AR, but not α_2A-AR. More intriguingly, similar to CCHCR1 interaction with the RKR motif in the CT of α_2A-AR as demonstrated here, GGA3 interacts with the RRR motif in the ICL3 of α_2B-AR (Zhang et al., 2016a). These data indicate that these two α₂-ARs use similar motifs in different intracellular domains to interact with distinct regulatory proteins to control their sorting in the transport both from the ER to the Golgi and from the Golgi to the cell surface. In addition, the small GTPase Rab43 is able to sort a subset of GPCRs, including α_2A-AR and α_2B-AR, from non-GPCR PM proteins in their export from the ER (Li et al., 2017), implicating multiple routes for the ER-to-Golgi and Golgi-to-PM trafficking of GPCR members.

It is increasingly apparent that GPCR forward delivery to the functional destinations is a highly coordinated, sophisticated process that is mediated through multiple pathways and distinct mechanisms and in a cell type- and receptor-specific manner (Filipeanu et al., 2006; Wu et al., 2003; Zhang and Wu, 2019). As the first step in intracellular trafficking of GPCRs, export from the ER has the most remarkable effect on the kinetics of GPCR maturation and cell surface targeting and is under control by a number of regulatory proteins interacting with the receptors (Bermak et al., 2001; Chen et al., 2014; Colley et al., 1991; Doly et al., 2016; Dwyer et al., 1998; McLatchie et al., 1998), specific export motifs (Dong et al., 2012; Dong and Wu, 2006; Duvernay et al., 2004; Shiwarski et al., 2019; Zhang et al., 2011) and transport machinery (such as Rab and ARF GTPases, Sec24 and GGA) (Dong et al., 2010; Li et al., 2012, 2017; Wei et al., 2021; Wu et al., 2003; Zhang et al., 2016b, 2013). Selective action of CCHCR1 on the ER export of two GPCRs in the same subfamily as demonstrated here further indicates the complexity of transport and segregation of nascent GPCRs. The importance of GPCR export from the ER is also indicated by the facts that a number of naturally occurring mutations that cause receptor misfolding and ER accumulation are directly associated with human diseases, and that using pharmacological chaperones to rescue the surface export of the misfolded receptors has beneficial effects in animal models (Dong et al., 2007; Morello et al., 2000; Rene et al., 2021; Ulloa-Aguirre et al., 2022). Therefore, a thorough elucidation of the regulatory mechanisms governing the export and sorting of nascent GPCRs may open new therapeutic avenues for the diseases involving defective GPCR biosynthesis.

MATERIALS AND METHODS

Materials

Antibodies against β-actin (1:1000; cat. no 47778), GFP (1:1000; cat. no 9996), Myc (1:1000; cat. no 40) and phospho-ERK1/2 (1:1000; cat. no 7383) were from Santa Cruz Biotechnology; antibodies against ERK1/2 (1:1000; cat. no 4695) and rabbit host antibodies against the HA epitope tag (1:1000; cat. no 3724) were from Cell Signaling Technology; mouse host antibodies against HA (cat. no 11583816001) were from Roche. Lipofectamine 2000, D-biotin, cycloheximide (CHX) and dynabeads™ protein G were from Thermo Fisher Scientific. UK14304 was obtained from Sigma-Aldrich. Linear polyethyleneimine (PEI, MW 25,000) was from Polysciences. MagneGST™ glutathione particles were from Promega. All other materials were obtained as described elsewhere (Xu and Wu, 2022).

Plasmids and constructions

α_2A-AR tagged with HA at its NT or GFP at its CT was generated as described previously (Li et al., 2017). α_2A-AR tagged with RFP in the pTagRFP-N vector or Rluc8 in the pRluc8-N1 vector was generated by PCR using the primers (forward, 5′-GATCCTCGAGATGGGCTCCCTGCAGCCGGACG-3′; and reverse, 5′-GATCGGTACCGTCACGATCCGCTTCCTGTCCC-3′). The RUSH plasmids, Str-KDEL_SBP-GFP-α_2A-AR, Str-KDEL_SBP-mCherry-α_2A-AR and Str-KDEL_SBP-α_2A-AR-Rluc8, were generated as described previously (Xu and Wu, 2022). To generate Str-KDEL_SBP-mCherry-α_2A-AR-Δ37/3E, α_2A-AR-Δ37 was obtained by two-step PCR using Str-KDEL_SBP-mCherry-α_2A-AR as template and primers (forward #1, 5′-GATCGGCCGGCCAGGCTCCCTGCAGCCG-3′, reverse #1, 5′-GGCCCGCGCCGCGGCTCCTCCAGGTCCAGCG-3′; forward #2, 5′-CGCTGGACCTGGAGGAGCCGCGGCGCGGGCC-3′ and reverse #2, 5′-GATCTCTAGATCACACGATCCGCTTCCTGTCCCCC-3′), digested with Fsel and Xbal, and inserted into Str-KDEL_SBP-mCherry-α_2A-AR after release of α_2A-AR, which was then used as template to mutate RKR to EEE by mutagenesis using primers (forward, 5′-GAAGATCCTCTGTCGGGGGGACGAGGAGGAGATCGTGTGATCTAGATAACTG-3′ and reverse, 5′-CAGTTATCTAGATCACACGATCTCCTCCTCGTCCCCCCGACAGAGGATCTTC-3′). To generate Str-KDEL_SBP-α_2A-AR-Δ37/3E-Rluc8, α_2A-AR-Δ37-Rluc8 was generated by two-step PCR using Str-KDEL_SBP-α_2A-AR-Rluc8 as template and primers (forward #1, 5′-GATCCCTGCAGGTATGGGCTCCCTGCAGCCG-3′, reverse #1,5′- GGCCCGCGCCGCGGCTCCTCCAGGTCCAGCG-3′; forward #2, 5′-CGCTGGACCTGGAGGAGCCGCGGCGCGGGCC-3′ and reverse #2, 5′-GATCTCTAGATTACTGCTCGTTCTTCAGCACGCG-3′), digested with Sdal and Xbal and inserted into Str-KDEL_SBP-α_2A-AR-Rluc8 after release of α_2A-AR-Rluc8, which was then used as template to mutate RKR using primers (forward, 5′-AGAAGATCCTCTGTCGGGGGGACGAGGAGGAGATCGTGACGGTACCGCGGGCC-3′ and reverse, 5′-GGCCCGCGGTACCGTCACGATCTCCTCCTCGTCCCCCCGACAGAGGATCTTCT-3′). To generate α_2A-AR-Δ37/3E in the pEGFP-N1 vector, α_2A-AR-Δ37/3E was amplified using KDEL_SBP-mCherry-α_2A-AR-Δ37/3E as template and primers (forward, 5′-GATCGAATTCTGATGGGCTCCCTGCAGCCGGACG-3′ and reverse, 5′-GATCGGTACCGT CACGATCTCCTCCTCGTCC-3′). To generate α_2A-AR-Δ37/3E fused with β-lactamase at its NT, β-lactamase in pcDNA3.1 was amplified by using primers (5′-GATCGCTAGCATGAGTATTCAACATTTCC-3′ and reverse, 5′-GATCAAGCTTCCAATGCTTAATCAGTGAG-3′), and inserted into α_2A-AR-Δ37/3E-GFP after digestion with NheI and HindIII. To generate the RUSH plasmids Str-KDEL_SBP-EGFP-α_2B-AR and Str-KDEL_SBP-mCherry-α_2B-AR, α_2B-AR was amplified by PCR using primers (forward, 5′-GATCGGCCGGCCATCCGGCCCCACCATGGAC-3′ and reverse, 5′-GATCTCTAGATCACCAGCCAGTCTGGGTCCACGG-3′), digested with Fsel and Xbal and then ligated to the plasmids Str-KDEL_SBP-EGFP-Ecadherin (Addgene #65286) and Str-KDEL_SBP-mCherry-Ecadherin (Addgene #65287) (Boncompain et al., 2012), respectively, which were digested with the same enzymes to release E-cadherin. GST fusion protein constructs coding different lengths of the ICL3 and the CT of α_2A-AR and α_2B-AR were generated in the pGEX-4T-1 vector as described previously (Li et al., 2012; Xu and Wu, 2022). GST-α_2A-AR-CT mutations were generated by QuickChange site-directed mutagenesis using primers (RKR to AAA: forward, 5′-GATCCTCTGTCGGGGGGACGCGGCGGCGATCGTGTGACTCGAGCGG-3′; and reverse, 5′-CCGCTCGAGTCACACGATCGCCGCCGCGTCCCCCCGACAGAGGATC-3′; RKR to EEE: forward, 5′-ATCCTCTGTCGGGGGGACGAGGAGGAGATCGTGTGACTCGAGCG-3′; and reverse, 5′-CGCTCGAGTCACACGATCTCCTCCTCGTCCCCCCGACAGAGGAT-3′). CCHCR1 and its fragments tagged with Myc at their NT were generated by using primers (M1–S756: forward, 5′-GATCGAATTCGGATGGCTCCCACCTGGCTC-3′, reverse, 5′-GGGGTACCTTAGCTGCTCATCTGGGG-3′; M1–T276: forward, 5′- GATCGAATTCGGATGGCTCCCACCTGGCTC-3′, reverse, 5′-GATCGGTACCTTATGTGAGGCTCTGCACCC-3′; S58-T276: forward, 5′-CGGAATTCGGTCACAGGCCCTGAGCCAG-3′, reverse, 5′-CCGCTCGAGTTATGTGAGGCTCTGCACCCG-3′; H277–P471: forward, 5′-CGGAATTCGGCACATCCTCGCCCTGCAG-3′, reverse, 5′-GGGGTACCTTACGGTGGTGGTAGGGGACAG-3′; V472-G665: forward, 5′-CGGAATTCGGGTCACAGACGTGAGCCTT-3′, reverse, 5′-GGGGTACCTTAACCTTCCTGCTGCAAGGT-3′; L666–S756: forward, 5′-CGGAATTCGGCTCCTCTCCCGTTACAAG-3′, reverse, 5′-GGGGTACCTTAGCTGCTCATCTGGGG-3′). The L666–S756 mutant DDE-3A was generated by using primers (forward, 5′-CTCTGTCCTGCTCGCTGACCTGCAGGCCCTGAGTGCAGCCATTTCCAAAG-3′ and reverse, 5′-CTTTGGAAATGGCTGCACTCAGGGCCTGCAGGTCAGCGAGCAGGACAGAG-3′). GFP-tagged CCHCR1 and its mutant DDE-3A in the pEGFP-N1 vector were generated by using primers (forward, 5′-CGGAATTCCGATGGCTCCCACCTGGCTC-3′ and reverse, 5′-CGGGATCCCGGCTGCTCATCTGGGGATT-3′). All mutants were generated by using QuikChange ll XL Site-Directed Mutagenesis kits (Agilent Technologies) and all constructs used in the present study were verified by nucleotide sequence analysis.

Cell culture and transfection

HEK293 (ATCC; cat. no CRL-1573) and HT29 (ATCC; cat. no HTB-38) cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS), 100 units/ml penicillin and 100 μg/ml streptomycin. For the experiments involving siRNA, transient transfection of cells was carried out by using Lipofectamine 2000 as described previously (Wu et al., 2003). For all other experiments, the cells were transfected by using PEI.

siRNA-mediated knockdown of CCHCR1

Two siRNAs targeting human CCHCR1 sequences (#1: CCUUCCCUACUGGAUAAGAAGAAAU and #2: CCAAGAGCAGCUGUCCUCUUUGACA), as well as negative control med GC duplex, were purchased from Thermo Fisher Scientific. siRNA-mediated depletion was carried out as described previously (Wu et al., 2003). To generate siRNA#1-resistant CCHCR1, two primers (5′-GACAGTTCTTCCTTCACTCCTAGACAAAAAGAAATCTGTGGTGT-3′ and 5′-ACACCACAGATTTCTTTTTGTCTAGGAGTGAAGGAAGAACTGTC-3′) were used to mutate five bases in Myc–CCHCR1 to achieve siRNA resistance without changing the encoded amino acid sequence. Briefly, cells were cultured on six-well plates overnight and then transfected with 60 pmol of control or CCHCR1 siRNA per well by using Lipofectamine 2000 for 24 h. The cells were transfected again with the same amount of siRNA together with 1 µg of α₂-AR plasmids per well for 6 h and split into 12-well plates for 20–24 h. For measurement of the PM expression of endogenous α_2A-AR, HT29 cells were cultured on 10-cm dishes and transfected with 360 pmol of siRNA twice for a total of 48 h since the first transfection.

CRISPR-Cas9-mediated KO of CCHCR1

The CRISPR-Cas9 CCHCR1 KO plasmids targeting CCHCR1, as well as control plasmids, were purchased from Santa Cruz Biotechnology and the experiments were carried out essentially as described previously (Wei et al., 2021). The CCHCR1 KO plasmid consists of a pool of three plasmids, each encoding the Cas9 nuclease and a target-specific 20 nt single guide RNA (sgRNA). Three sgRNA sequences in CCHCR1 are AACGGGATGTTTCCAGTGAC, GGACAGCCTGCATGCCACCG and GGAAACATCCCGTTCCCACA. To study the effect of CCHCR1 KO on the transport of α₂-ARs in RUSH assays, HEK293 cells were cultured on 12-well plates and transfected with control or CCHCR1 KO plasmids (1.0 µg) for 24 h. The cells were then transfected again with control or CCHCR1 KO plasmids (0.75 µg) together with mCherry-tagged α₂-ARs in RUSH plasmids (0.25 µg) for 24 h before incubation with biotin. As control and CCHCR1 KO plasmids carry GFP, transfected cells were defined by the GFP signal and receptor transport was analyzed only in GFP-expressing cells.

Radioligand binding

The cell surface expression of transiently expressed α_2A-AR and α_2B-AR was measured by ligand binding of intact live cells using the radioligand [³H]RX821002 as described previously (Xu and Wu, 2020). Briefly, HEK293 cells were cultured on six-well dishes and transfected with individual α₂-ARs together with control siRNA or siRNA targeting CCHCR1. The cells were than incubated with [³H]RX821002 at a concentration of 20 nM for 90 min at room temperature. After the cells were washed twice with 1 ml of ice-cold PBS, the cells were treated with 1 M NaOH for 1 h. The non-specific binding was determined in the presence of rauwolscine (10 μM). The radioactivity was counted by liquid scintillation spectrometry in 4 ml of Ecoscint A scintillation solution in the AccuFLEX LSC-8000 counter (Hitachi Aloka). All intact cell radioligand binding assays were performed in triplicate.

The PM expression of endogenous α_2A-AR in HT29 cells was measured by radioligand binding of membrane preparations essentially as described previously (Duvernay et al., 2004). Briefly, HT29 cells were cultured on 10-cm dishes and transfected with control siRNA or siRNA targeting CCHCR1 for 48 h as described above. The cells were homogenized in hypotonic buffer containing 5 mM Tris-HCl pH 7.4, 5 mM EGTA and 5 mM EDTA. After centrifugation at 13,000 g for 10 min at 4°C, the pellet was washed once and re-suspended in 200 µl per plate of membrane buffer containing 50 mM Tris-HCl, pH 7.4, 0.6 mM EDTA and 5 mM MgCl₂. The membrane suspension containing 250 µg protein was incubated with [³H]RX-821002 at 2 nM in a total volume of 100 µl for 1 h at room temperature with constant shaking. Non-specific binding was determined in the presence of rauwolscine (100 µM). The reaction was terminated by vacuum filtration. After washing with 100 mM Tris-HCl pH 7.4 (4×4 ml), the retained radioactivity was measured by liquid scintillation spectrometry.

Fluorescence microscopy

To measure the effect of CCHCR1 depletion on the subcellular localization, ER-to-Golgi transport and Golgi-to-PM transport of α_2A-AR and α_2B-AR in fixed cells, the cells were transfected with RUSH plasmids and then fixed with 4% paraformaldehyde for 15 min. To analyze the subcellular distribution of CCHCR1, the cells were transfected with CCHCR1–GFP with or without co-transfection with the ER marker DsRed-ER or mCherry-α_2A-AR. All images were captured with a 63× objective on a Leica Stellaris 5 confocal microscope using LAS X software as described previously (Xu and Wu, 2022). The receptor expression at the Golgi, cell surface and the whole cell was quantified by using NIH ImageJ to measure the fluorescence intensities in a single confocal plane.

Measurement of ERK1/2 activation

ERK1/2 activation by α₂-ARs was determined by measuring their phosphorylation by immunoblotting as described previously (Wu et al., 2003). To measure the effects of siRNA-mediated knockdown of CCHCR1 on ERK1/2 activation by α₂-ARs, HEK293 cells were transfected with α_2A-AR or α_2B-AR together with control or CCHCR1 siRNA for 48 h. To study ERK1/2 activation by α_2A-AR and its mutant Δ37/3E, HEK293 cells were transfected with the receptor for 24 h. After the cells were starved for 8 h, the cells were stimulated with the α₂-AR agonist UK14304 at a concentration of 1 µM for 5 min. To measure ERK1/2 activation by endogenous α_2A-AR in HT29 cells, the cells were transfected with control or CCHCR1 siRNA and then starved for 24 h before stimulation with UK14304.

RUSH assays

RUSH assays were essentially carried out as described (Boncompain et al., 2012; Xu and Wu, 2022). Briefly, HEK293 cells were transfected with α₂-AR RUSH plasmids, together with siRNA or CRISPR-Cas9 KO plasmids as described above. To measure receptor transport from the ER to the Golgi, the cells were incubated with biotin at a final concentration of 40 µM plus CHX (400 µg/ml) for different periods of time as indicated in figure legends. To measure receptor transport from the Golgi to the surface, the cells were first incubated with biotin plus CHX at 20°C for 3 h. Under this condition, the nascent receptors were normally exported from the ER but accumulated in the Golgi. After replacing cell culture medium with DMEM without biotin, the cells were incubated at 37°C for 30 min to induce receptor export from the Golgi to the cell surface. The Golgi area of individual cells was defined by the region with highly concentrated receptors after biotin induction. The ER-to-Golgi and Golgi-to-surface transport were expressed as the Golgi-to-total expression ratio. Enhancement in this ratio reflects receptor transport from the ER to the Golgi, whereas reduction in this ratio indicates receptor transport out of the Golgi to the surface.

Co-IP

Co-IP assays were carried out as described previously (Li et al., 2017). Briefly, HEK293 cells were cultured on 10-cm dishes and transfected with HA-α₂-AR together with Myc-CCHCR1 plasmids (10 µg each) for 24 h. The cells were harvested and lysed with 500 µl of lysis buffer containing 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate and 0.1% SDS. After centrifugation, the supernatants were incubated with 2 µg of anti-HA antibodies (Roche, mouse host) overnight at 4°C, followed by incubation with 30 µl of protein G dynabeads for 1 h at 4°C. The beads were collected and washed three times with lysis buffer. Immunoprecipitated proteins were solubilized with SDS gel loading buffer and detected by immunoblotting.

GST fusion protein pulldown assays

GST fusion protein pulldown assays to measure the interaction between α₂-AR and CCHCR1 were carried out using the MagneGST pulldown system (Promega) as described previously (Li et al., 2012, 2017). Briefly, purified GST fusion proteins were incubated with HEK293 cell homogenates expressing Myc-tagged CCHCR1 or its fragments in a total volume of 400 µl binding buffer containing 20 mM Tris-HCl pH 7.4, 140 mM NaCl, 1% Nonidet P-40 and 10% glycerol overnight at 4°C. After washing three times with binding buffer, the bound proteins were solubilized in SDS gel loading buffer and detected by immunoblotting using anti-Myc antibodies.

BRET assays

The live cell-based BRET assay, in combination with the RUSH system, was used to measure the cell surface expression of α_2A-AR in HEK293 cells as described previously (Lan et al., 2012; Xu and Wu, 2022). Briefly, cells were seeded on six-well plates and transfected with 250 ng of Str-KDEL_SBP-α_2A-AR-Rluc8 and 750 ng of Venus–kRas or pcDNA3.1 for 24 h. The cells were then incubated with biotin at 40 µM for 2 h and split onto black 96-well plates. After addition of coelenterazine h (5 mM), luminescence was immediately measured using a Mithras LB940 photon-counting plate reader (Berthold Technologies GmbH, Bad Wildbad, Germany). The BRET signals were calculated by dividing the emission intensity at 520–545 nm by the emission intensity at 475–495 nm. Net BRET was this ratio minus the same ratio measured from cells expressing Str-KDEL_SBP-α_2A-AR-Rluc8 and pcDNA3.1.

β-lactamase reporter assays

The β-lactamase reporter assay to measure α_2A-AR expression at the cell surface was carried out as described previously (Lam et al., 2013). HEK293 cells cultured on 12-well plates were transfected with β-lactamase-tagged α_2A-AR for 6 h and then split into 48-well plates. After the cells were washed with PBS, 200 µl of the nitrocefin solution at a final concentration of 200 µM were added to each well. Immediately following nitrocefin addition, the absorbance at 486 nm was read once every min for 30 min on a microplate reader (Spectramax M2, Molecular Devices). The rate of reaction (slope of the curve in the linear range) was used as the readout of receptor surface expression.

Statistical analysis

Statistical differences were analyzed by using one-way ANOVA with a Turkey's multiple comparisons post test for more than two groups or by unpaired two-tailed t-test for two groups. All data were expressed as mean±s.e. Significance levels are *P<0.05, **P<0.01 and ***P<0.001.

Peer review history

The peer review history is available online at https://journals.biologists.com/jcs/lookup/doi/10.1242/jcs.261685.reviewer-comments.pdf