点击显示 收起
【关键词】 Mammalian
In the present study, we identified the CUB5 domain of mammalian Tolloid (mTLD) as a novel protein binding to 1A-adrenergic receptor (AR) using the yeast two-hybrid system. Whereas CUB5 did not couple to either 1B-AR or 1D-AR. It was determined that amino acids 322 to 359 of 1A-AR were the major binding region for CUB5. The direct interaction between 1A-AR cytoplasmic tail and CUB5 was discovered by glutathione S-transferase pull-down assay. We confirmed the interaction of mTLD with 1A-AR in human embryonic kidney (HEK) 293 cells by immunoprecipitation, immunofluorescence, and fluorescence resonance energy transfer. Although mTLD did not affect the density and affinity of receptors in crudely prepared membranes from HEK293 cells stably expressing 1A-AR, it significantly altered the subcellular localization of the receptors. Moreover, mTLD reduced the level of cell surface 1A-ARs, delayed the initial rate of agonist-induced receptor internalization, and facilitated agonist-induced calcium transient. We have demonstrated that mTLD interacts with 1A-AR directly, alters the subcellular localization of receptor, and influences agonist-induced 1A-AR internalization and calcium signaling.
1-Adrenergic receptors (ARs) are G-protein-coupled receptors (GPCRs) that respond to norepinephrine and epinephrine. 1-ARs can be classified into at least three subtypes, termed 1A, 1B, and 1D, according to their ligand specificity and their amino acid sequences. Although the three 1-AR subtypes all couple to the Gq/11 signaling pathway, they have distinct functional roles in mediating vascular smooth muscle contraction and cardiomyocyte hypertrophy (Zhu et al., 1997; Autelitano and Woodcock, 1998; Grupp et al., 1998).
As the largest family of membrane proteins, GPCRs respond to extracellular stimulations and convert them to intracellular signals. In general, this transduction is defined as follows: ligand binding changes the conformation of receptor's intracellular region, which induces the interaction of receptor with heterotrimeric G-proteins and the ultimate dissociation of G and G subunits. These activated subunits can stimulate and/or inhibit a variety of intracellular effectors. A large number of intracellular molecules other than G-proteins (non-G-proteins) could also interact with GPCR, mediate the intracellular response, or regulate the receptor's state, distribution, and function (Hall et al., 1999; Heuss and Gerber, 2000; Brzostowski and Kimmel, 2001).
ARs have been shown to associate with several non-G-proteins. The majority of them were identified for 2-AR, such as ARK, -arrestin, AKAP79/150, NHERF, and eIF-2B (Klein et al., 1997; Menard et al., 1997; Hall et al., 1998; Fraser et al., 2000; Cong et al., 2001); only a few proteins are reported to interact with 1-AR. Among them, the µ2 subunit of the AP2 clathrin adaptor complex specifically interacts with 1B-AR (Diviani et al., 2003), whereas tissue transglutaminase II (Gh) and gC1q-R interact with 1B- and 1D-AR (Chen et al., 1996; Pupo and Minneman, 2003). Neuronal nitric-oxide synthase was believed previously to interact with 1A-AR but was found to interact with all three subtypes of 1-AR and 1- and 2-ARs (Pupo and Minneman, 2002). So far, no interacting protein has been identified specifically for 1A-AR, the most important and efficient subtype of 1-AR (Zhong and Minneman, 1999; Zhong et al., 2001).
CUB5 domain, a segment of mammalian Tolloid (mTLD) (Takahara et al., 1994), has been identified as an 1A-AR coupling protein by our group (Xu et al., 2003). In the present study, we have further explored the interaction of 1A-AR and mTLD. In particular, we investigated the differential interactions of CUB5 domain with three 1-AR subtypes and the region of 1A-AR that binds to the CUB5 domain. We also assessed this interaction by glutathione S-transferase (GST)-1A-AR fusion protein pull-down assay, immunoprecipitation, immunofluorescence, and fluorescence resonance energy transfer (FRET). Moreover, the effects of mTLD on pharmacological property, localization, and signaling of 1A-AR were also investigated in this study.
Plasmid DNAs. Flag-tagged human 1A-, 1B-, and 1D-ARs and C-terminally truncated 1A-ARs in mammalian expression vectors PDT-1A,-1B,-1D, and -1ACT, and the GST-tagged 1A-AR cytoplasmic tail, GST-1A, were generous gifts from Dr. Kenneth P. Minneman (Department of Pharmacology, Emory University, Atlanta, GA). The cDNA sequence encoding the 1B- and 1D-AR cytoplasmic tail and various fragments of the 1A-AR carboxyl terminus were generated by PCR and cloned into pGBKT7 with EcoRI and BamHI sites (see Supplemental Data). PCR reactions were carried out using high-fidelity Pfx DNA polymerase (Invitrogen, Carlsbad, CA).
cDNA encoding human mTLD (P13497) was obtained from Invitrogen. For mammalian cell expression, PCR products were subcloned into pcDNA3.1 (+) with NheI and XhoI sites to generate pMT, and other HA-tagged PCR products were also subcloned into pcDNA3.1 (+) with NheI and PmeI sites to generate pMT/HA (supplemental data). The cDNA of prey 3, which encodes aa 867 to 986 of mTLD, was subcloned into pcDNA3.1 (+) with EcoRI and BamHI sites to generate HA-tagged pcP3HA, as we described previously (Xu et al., 2003). Construct EGFP-1A, expressing fused EGFP/1A-AR, was made by cutting 1A-AR cDNA with its flank EcoRI and BamHI restriction sites from PDT-1A and then cloning back into the EGFP-C2 expression vector. All constructs were confirmed with Prism 3700 DNA automatic sequencer (Applied Biosystems Inc., Foster City, CA).
Yeast Two-Hybrid. We used pretransformed MATCHMAKER two-hybrid system (Clontech, Mountain View, CA) to perform the screening. Yeast Saccharomyces cerevisiae strains AH109 and Y187 were used as the host for the in vivo interaction studies. Yeast two-hybrid analysis was carried out as described in Clontech's user manual. PGAACT, a bait plasmid containing 1A-AR aa 322 to 466 was transformed into yeast strain AH109 by the LiCl/polyethylene glycol method, and mating with human brain cDNA pretransformed yeast strain Y187. Prey plasmids were isolated from the positive colonies using the Zymoprep Yeast Plasmid Minipreparation Kit (Zymo Research, Orange, CA) and sequenced with Prism 3700 DNA automatic sequencer.
-Galactosidase activity was used to test interactions between different fusion proteins. In brief, both bait and prey plasmids were transformed into the yeast strain Y187; then, -galactosidase activities were measured by colony-lift filter assay and liquid -galactosidase assay with O-nitrophenyl glycoside (Merck and Co., Inc., Whitehouse Station, NJ).
GST Pull-Down Assay. MagneGST Pull-Down System (Promega, San Luis Obispo, CA) was used for direct interaction assays between 1A-AR and mTLD. GST-1A-AR cytoplasmic tail fusion proteins or GST proteins were lysed from 1 ml of bacterial culture, and immobilized on MagneGST Particles. Transcription/translation of aa 867 to 986 of mTLD was undertaken in vitro using TNTT7 Quick-Coupled transcription/translation reactions Master Mix (Promega, Madison, WI). Binding reactions were performed in 200 µl of binding/wash buffer (4.2 mM Na2HPO4,2mMKH2PO4, 140 mM NaCl, and 10 mM KCl, pH 7.2) for 1 h at room temperature. Particles were washed five times with the same buffer, and bound proteins were analyzed by immunoblotting.
Cell Line, Culture, and Transfection. HEK293 cells were obtained from the American Type Culture Collection (Manassas, VA) and cultured in glutamine-containing high-glucose Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate, 100 mg/ml streptomycin, and 100 U/ml penicillin at 37°C under a 95% air/5% CO2 atmosphere. HEK293 cells stably transfected PDT-1A were also cultured in the same conditions except that 200 mg/ml G418 (Geneticin) was added. Lipofectamine 2000 (Invitrogen) was used in all transfection experiments.
Coimmunoprecipitation Assay. HEK293 cells were transfected and cultured for 48 h. Immunoprecipitation was performed as described previously (Vazquez-Prado et al., 2000) with minor modifications. In brief, cells were washed with ice-cold PBS and lysed for 1 h on ice in lysis buffer (10 mM Tris-HCl, pH 7.4, 50 mM NaCl, 5 mM EDTA, 1% Triton X-100, and 0.05% SDS) supplemented with 50 mM sodium fluoride, 100 mM sodium orthovanadate, 10 mM sodium pyrophosphate, 20 mg/ml aprotinin, 20 mg/ml leupeptin, and 100 mg/ml phenylmethylsulfonyl fluoride. Cell lysates were centrifuged at 12,700g for 15 min, and the supernatants were incubated overnight at 4°C with the anti-Flag M2 affinity resin (Sigma-Aldrich.) or anti-c-myc antibodies (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) plus protein G resin (Sigma-Aldrich). After four washes with lysis buffer without protease inhibitors, the immune complexes were denatured by boiling in 2x loading buffer and analyzed by immunoblotting. Anti-1A-AR antibodies (Santa Cruz Biotechnology), anti-mTLD antibodies, and Supersignal West Pico Chemiluminent (Pierce Biotechnology, Inc.) were used to detect proteins.
Fig. 1. A summary of the interaction between CUB5 and 1A-AR receptor cytoplasmic tail in the yeast two-hybrid system. Top, schematic map of mTLD, with the CUB5 domain shown. A yeast two-hybrid assay identified aa 867 to 986 of mTLD (encompassing the CUB5 domain, aa 865 to 931) as a potential partner for the cytoplasmic tail of 1A-AR. Bottom, colony-lift filter assay (X-Gal) indicates that CUB5 interacts only with the cytoplasmic tail of 1A-AR, not that of 1B-AR and 1D-AR. In this assay, yeast cells transformed with the CUB5 cDNA construct and the PGAACT construct turn blue (++) in 3 h after incubation at 30°C on filters, indicating activation of the reporter gene -galactosidase. In contrast, yeast cells transformed with CUB5 cDNA constructs and PGABCT/PGADCT remained white after 12 h.
Immunofluorescence and Fluorescence Resonance Energy Transfer. HEK293 cells were plated on glass coverslips and cotransfected with EGFP-1A and pMT/HA. Twenty-four hours after transfection, cells were fixed in 4% formaldehyde for 10 min and permeated with 0.1% Triton X-100 for 10 min. The fixed cells were incubated with anti-HA monoclonal antibodies (1:100 dilutions; Santa Cruz Biotechnology) overnight at 4°C and then with a TRITC-conjugated anti-mouse antibody (1:200 dilution; Invitrogen) for 1 h. The distributions of 1A-AR and mTLD were examined with a confocal laser-scanning microscope system (TCS SP2; Leica, Wetzlar, Germany) with a 40x oil immersion objective lens. Fluorescence of TRITC was then excited continuously for 5 min to bleach the TRITC signal irreversibly. Images of the EGFP fluorescence were acquired before and after photobleaching. The extent of FRET was assessed by calculating the ratio of the EGFP fluorescence before (DA) and after (D) photobleaching, using the equation E = D/DA. Nine to ten cells from three individual experiments were selected, and the D/DA of intracellular areas was averaged.
Radioligand Binding. HEK293 cells stably transfected PDT-1A were plated in 10-cm dishes and transfected with pcDNA3.1/LacZ or pMT. Forty-eight hours after transfection, cells were washed with PBS, pH 7.6, and harvested. After centrifugation and homogenization with a Polytron homogenizer (Kinematica, Basel, Switzerland), cell membranes were collected by centrifugation at 20,000g for 20 min and resuspended in PBS. Radioligand binding sites were measured by saturation analysis of specific binding of the 1A-AR antag-onist radioligand 125I-BE2254 (15-500 pM). Nonspecific binding was defined as binding in the presence of phentolamine at 100 µM.
Whole-Cell ELISA. HEK293 cells stably transfected with PDT-1A were plated in poly-D-lysine-coated 24-well plates and transfected with pcDNA3.1/LacZ or pMT. Forty-eight hours later, cells were incubated with DMEM containing phenylephrine (10 µM) for 0 to 60 min at 37°C. The transfected cells were then fixed in 2% formaldehyde for 10 min without permeating the cells. Changes in surface receptor density were subsequently determined by ELISA, taking advantage of the FLAG-tag in 1A-AR. Cells were labeled with the anti-FLAG M2 antibody (1:1000 dilution; Sigma-Aldrich) followed by anti-mouse antibody (1:1000 dilution; Jackson ImmunoResearch, West Grove, PA). Antibody labeling was detected by incubation with O-phenylenediamine (Sigma-Aldrich, Inc.), and absorbance was read at 490 nm.
Intracellular Calcium Measurement. PDT-1A stably transfected HEK293 cells were transfected with pcDNA3.1/LacZ or pMT. Twenty-four hours after transfection and a further8hof serum starvation, cells were loaded with calcium indicator, fluo-4AM (In-vitrogen), as described previously (Wang et al., 2001). Fluorescence was visualized by Zeiss LSM510 confocal microscope (Oberkochen, Germany) equipped with an argon laser (488 nm) and 40x, 1.3 numerical aperture oil immersion objective lens. Images were acquired at sampling rates of 10 s/image, and phenylephrine (10 µM) was added after the first three images were acquired. The digital images were analyzed with IDL image-processing software (ver. 5.4; RSI, Boulder, CO). Four to five cells were selected, and the average fluorescence density of intracellular areas was measured to indicate calcium level. All the data were normalized by fluorescent density before stimulation.
Fig. 2. Schematic diagram showing the interaction between CUB5 domain with truncated 1A-AR constructs as determined using the yeast two-hybrid assay. A series of deletion truncations of 1A-AR cytoplasmic tail constructs were cotransfected into Y187 with cDNA of the CUB5 domain. Black block, primary biding region; gray block, subsidiary binding region; white, nonbinding region. The interactions were tested using both X-Gal colony-lift filter assay and liquid -galactosidase assay with O-nitrophenyl glycoside. Results from the colony-lift filter assay are classified as follows: +++, turned blue in 1 h after incubation at 30°C on filters; ++, turned blue in 3 h, +, turned blue in 5 h; -, color unchanged after 12 h. In the liquid -galactosidase assay, results are the mean ± S.E. of four independent experiments.
Statistical Analysis. All data are presented as means ± S.E. Data obtained from radioligand binding and ELISA assay were analyzed with an unpaired t test or analysis of variance (analysis of variance) followed by unpaired t test. P < 0.05 was considered significant.
mTLD CUB5 Domain Interacts with 1A-AR via Yeast Two-Hybrid Assay. To identify proteins interacting with 1A-AR, we screened a human brain cDNA library, using the yeast two-hybrid assay with the cytoplasmic tail (aa 322-466) of 1A-AR as bait. More than 2 x 107 colonies were screened, and positive colonies were identified by monitoring the activation of reporter genes HIS3, ADE2, and MEL1. Plasmids containing prey cDNA from each positive colony were isolated and sequenced. Of the nine positive colonies obtained, six prey plasmids were identified. As shown in Fig. 1, top, the cDNA of prey 3 encodes aa 867 to 986, the CUB5 domain, of protein mTLD.
Fig. 3. Direct interaction of the CUB5 domain with 1A-AR cytoplasmic tail. GST or GST-1A-AR-cytoplasmic-tail fusion proteins were purified from 1 ml of bacterial cultures; HA-tagged CUB5 domains were synthesized in vitro. Then GST or GST-1A-AR-cytoplasmic-tail fusion proteins (60 µg) and CUB5 domains (60 µg) were added together and subjected to GST pull-down. Bound proteins were analyzed by immunoblotting with anti-1A-AR antibodies (top), anti-GST (middle), and anti-HA (bottom).
We then determined whether CUB5 domain could also interact with other subtypes of 1-AR in yeast two-hybrid system. We found that only yeast cells transformed with CUB5 cDNA constructs and cytoplasmic tail of 1A-AR cDNA constructs showed an obvious activation of reporter gene -galactosidase, which was consistent with the yeast two-hybrid assay screening results. No such interaction was detected between the CUB5 domain and cytoplasmic tail of either 1B-AR or 1D-AR (Fig. 1, bottom).
Amino Acid Residues 322 to 359 of 1A-AR Are Important for Interaction with CUB5. As mentioned above, the whole cytoplasmic tail of 1A-AR was used as bait in the yeast two hybrid screen. To localize the CUB5 domain interaction on 1A-AR C terminus, a series of truncations of 1A-AR cytoplasmic tail, illustrated in Fig. 2, were screened against CUB5 domain. Both X-Gal colony-lift filter assay and liquid -galactosidase assay yielded similar results (Fig. 2). Truncations of the carboxyl end of 1A-AR were quite tolerated, displaying positive interactions with CUB5. Only a small fragment, containing residues 360 to 398, did not interact with CUB5. Truncations from the amino terminus of the 1A-AR cytoplasmic tail narrowed the region for interaction with CUB5 domain to residues 322 to 359.
Fig. 4. Binding of mTLD and 1A-AR in transfected HEK293 cells as shown by immunoprecipitation. HEK293 cells transfected with PDT-1A and/or pMT were lysed, and 500 µg of total proteins were immunoprecipitated with anti-FLAG M2 affinity resin (A) or anti-c-myc anti-body plus protein-G resin (B). Cell lysates (50 µg) and immunoprecipitates were immunoblotted with anti-mTLD (top) or anti-1A-AR anti-bodies (bottom). C, HEK293 cells transfected with PDT-1A or PDT-1ACT and pcP3HA were lysed, and 500 µg of total proteins were immunoprecipitated with anti-FLAG M2 affinity resin. Cell lysates (50 µg, top) and immunoprecipitates (middle and bottom) were immunoblotted with anti-HA, anti-FLAG, or anti-1A-AR antibodies. Control, nontransfected HEK293 cells; 1A, HEK293 cells transfected with PDT-1A; C, HEK293 cells transfected with pcP3HA; 1A + C, HEK293 cells cotransfected with PDT-1A and pcP3HA; 1ACT+ C, HEK293 cells cotransfected with PDT-1ACT and pcP3HA; T, HEK293 cells transfected with pMT; 1A + T, HEK293 cells cotransfected with PDT-1A and pMT. The results are representative of three independent experiments.
Association of mTLD with 1A-AR in Vitro and Vivo. To examine whether mTLD can directly interact with 1A-AR, purified GST-1A-AR-cytoplasmic-tail fusion proteins from bacterial cells and in vitro synthesized CUB5 domains were subject to GST pull-down assay. The bound proteins were fractionated by SDS-polyacrylamide gel electrophoresis and detected by Western blotting. As shown in Fig. 3, the CUB5 domains bound to GST-1A-AR-cytoplasmic-tail fusion proteins but not to GST alone.
To confirm the interaction between 1A-AR and mTLD in vivo, 1A-AR and mTLD were transiently expressed in HEK293 cells, lysed, and immunoprecipitated by their antibodies, respectively. In immunoblots of cell lysate, bands corresponding to 1A-AR (50 kDa) and mTLD were revealed by their respective antibodies in transfected but not in untransfected cells (Fig. 4). Immunoprecipitating 1A-AR with anti-Flag M2 affinity resin resulted in coimmunoprecipitation of mTLD, as revealed by the specific 150-kDa band detected in immunoprecipitates of cotransfected cells. This band was not present in immunoprecipitates from HEK293 cells expressing 1A-AR or mTLD alone. Conversely, reverse immunoprecipitating mTLD also showed coimmunoprecipitation of 1A-AR (Fig. 4). Thus, in HEK293 cells, mTLD interacted specifically with 1A-AR.
Likewise, in HEK293 cells transiently expressing 1A-AR and CUB5 domain, immunoprecipitating 1A-AR with antiFlag M2 affinity resin also resulted in coimmunoprecipitation of CUB5 (Fig. 4C). Moreover, the CUB5 coimmunoprecipitated by the C-terminally truncated 1A-AR was less than 5% of that by full-length 1A-AR, confirming that the binding to CUB5 domain was mostly contributed by the C terminus of 1A-AR.
Fig. 5. Colocalization and FRET of 1A-AR and mTLD in transfected HEK293 cells. Images were taken of fixed HEK293 cells using a confocal laser scanning microscope system. A, expressing EGFP-1A-AR individually; B-D, coexpressing the EGFP-1A-AR and mTLD/HA; E-G, coexpressing the EGFP-1A-AR and mTLD/HA, and incubated with 10 µM phenylephrine for 5 min. mTLD was detected using an anti-HA monoclonal antibody and an anti-mouse TRITC-conjugated antibody. Images were taken after excitation at 488 nm for EGFP (A, B, and E) or 543 nm for TRITC (C and F). Merge (D and G) represents the superimposition of the EGFP and TRITC images. Scale bars, 20 µm. The results shown are representative of observations of three independent transfection experiments. H, quantitative analysis of the FRET efficiency. Nine to ten cells from three individual experiments were selected, and the D/DA value of intracellular areas was averaged.
Colocalization and Association of HA-Tagged mTLD and 1A-AR. To characterize the mTLD-1A-AR association in vivo, EGFP-fused 1A-AR and HA-tagged mTLD were transfected into HEK293 cells to allow the subcellular localization of these proteins to be determined using an immunofluorescence assay. As shown in Fig. 5, in cells expressing EGFP-fused 1A-AR alone, fluorescence was observed both on the cell surface and intracellularly (Fig. 5A), which was consistent with previous reports (Chalothorn et al., 2002). When mTLD and 1A-AR were coexpressed, the subcellular localization of 1A-AR was significantly altered. The EGFP fluorescence signal of 1A-AR accumulated predominantly in certain cytoplasmic compartments colocalizing with mTLD (Fig. 5, B, C, and D). However, if the cells were stimulated with 10-5 M phenylephrine for 5 min before fixation, colocalization of 1A-AR with mTLD was evident on the cell surface and the cytoplasmic compartment, although the fluorescence signal of 1A-AR was still concentrated in the cytoplasm (Fig. 5, E, F, and G). To validate whether a direct interaction happens between these two molecules, FRET between EGFP-fused 1A-AR (donor fluorescence) and immunofluorescence stained mTLD (acceptor fluorescence) was measured by acceptor photobleaching method. As shown in Fig. 5H, in the absence of mTLD, EGFP signal declined 16.9 ± 3.6% after 5 min of continuous activation of TRITC signal, because EGFP fluorescence could also be excited and inactivated during the process. In the presence of mTLD, however, the fluorescence of the donor in the cells increased 26.7 ± 5.4% after photobleaching of acceptor fluorescence, which indicates a tight intermolecular association between 1A-AR and mTLD in HEK 293 cells.
Fig. 6. mTLD changes the number of cell surface 1A-AR and internalization in response to phenylephrine. HEK293 cells expressing 1A-AR were transfected with pcDNA3.1/LacZ or pMT (A). 48 h later, cells were incubated without (B) or with (C) phenylephrine for 5 to 60 min at 37°C. Then transfected cells were fixed, and changes in surface receptor density were subsequently determined by whole-cell ELISA. Data are mean ± S.E. of eight independent experiments, each performed in duplicate and normalized to surface receptor density of control group (B) or untreated cell (C).
mTLD Changes The Number of Cell Surface Receptors and the Time Course of Internalization. To verify the finding of subcellular localization of 1A-AR in immunofluorescence assay, we compared the number of cell surface 1A-AR in HEK293 cells stably transfected with PDT-1A in the presence or absence of mTLD (Fig. 6A) using whole-cell ELISA assays. As shown in Fig. 6B, the presence of mTLD reduced the number of cell surface 1A-AR by 44 ± 6% (P < 0.001, n = 8) compared with cells transfected with a control vector, pcDNA3.1/LacZ. In addition, we examined the time course of 1A-AR internalization in response to phenylephrine (10 µM). As shown in Fig. 6C, in the control group (pcDNA3.1/LacZ transfected), there was a rapid agonist-induced internalization of cell surface receptor with a 63 ± 2% (n = 8) reduction after 1 h; in the mTLD group, the time course of 1A-AR internalization was somewhat delayed. The number of cell surface receptors initially increased after 5 min of agonist stimulation and decreased by 58 ± 5% (n = 8) between 5 and 60 min of agonist exposure.
Fig. 7. mTLD does not alter the pharmacological properties of 1A-AR. HEK293 cells expressing 1A-AR were transfected with pcDNA3.1/LacZ or pMT. Membranes were collected 48 h after transfection. Radioligand binding sites were measured by saturation analysis of specific binding of the 1A-AR antagonist radioligand 125I-BE2254 (15-500 pM). Nonspecific binding was defined as binding in the presence of phentolamine (100 µM). Results are the mean ± S.E. of four independent experiments. There was no significant difference in Bmax and KD between LacZ-coexpressing and mTLD-coexpressing groups.
mTLD Does Not Alter Density and Affinity of 1A-AR. To study the effect of mTLD on the pharmacological properties of 1A-AR, crude membranes were prepared from HEK293 cells expressing 1A-AR with and without concomitant expression of mTLD. The density and affinity of 1A-AR binding sites were then determined by saturation analysis of the specific binding of the antagonist radioligand 125I-BE2254. As shown in Fig. 7, the presence of mTLD had no effects on the density of 1A-AR. There was no significant difference in Bmax between LacZ-coexpressing and mTLD-coexpressing groups (1030 ± 65 versus 1048 ± 119 fmol/mg, respectively; both n = 4). Furthermore, there was no significant difference in the affinity of 1A-AR to the radioligand 125I-BE2254 in the presence of LacZ or mTLD (272 ± 26 versus 352 ± 69 pM, respectively; both n = 4). These results indicated that the interaction of mTLD with 1A-AR does not affect the pharmacological properties of the receptor, although mTLD pulled considerable receptors into intracellular compartment.
Fig. 8. Time course of intracellular calcium in response to PE stimulation in the presence of LacZ or mTLD. HEK293 cells expressing 1A-AR were transfected with pcDNA3.1/LacZ (A) or pMT (B). Twenty-four hours after transfection, cells were starved with serum-free media for 8 h. Intracellular Ca2+ was visualized by confocal microscopy using the calcium indicator fluo-4. Phenylephrine was added to stimulate 1A-AR (final concentration, 10 µM) after three images were acquired. A, a sharp, high-amplitude, and phasic [Ca2+]i change was observed 2 to 3 min after receptor activation in the control group (LacZ); B, a phasic [Ca2+]i response occurred immediately after receptor activation in the presence of mTLD; C, quantification of [Ca2+]i demonstrated that maximum calcium transient of the mTLD group was smaller than control group. D and E, the calcium transient reductions in control and mTLD group display a similar exponential decay with a time constant of 4.5 and 4.6 min, respectively.
Effect of mTLD on Agonist-Induced Changes of Intracellular Calcium Concentration. To further understand the influence of mTLD on 1A-AR signaling function, the dynamics of intracellular calcium concentration ([Ca2+]i) in response to 1A-AR activation was examined with a calcium indicator, fluo-4. In 1A-AR and LacZ cotransfected cells, we detected a sharp, high-amplitude, and phasic [Ca2+]i change approximately 2 to 3 min after 10 µM phenylephrine stimulation (Fig. 8, A and C). The calcium transient diminished slowly and displayed an exponential decay with a time constant of 4.5 min (Fig. 8D). In contrast to LacZ cotransfected control cells, in the presence of mTLD, the phasic [Ca2+]i response was initiated immediately after receptor activation. However, the maximum amplitude of the [Ca2+]i response was relative smaller (Fig. 8, B and C), only approximately 58% of control group, although the attenuation of the calcium transient was also shown as exponential decay with a time constant of 4.6 min (Fig. 8E). We also studied the effect of mTLD on intracellular calcium response to angiotensin-II type 1 receptor (AT1R) and 1-AR activation. Quantification of [Ca2+]i demonstrated that in response to 1-AR activation, intracellular calcium increased slowly, and progressively (Supplemental Data, Fig A). Intracellular calcium in response to AT1R activation occurred immediately after receptor activation (Supplemental Data, Fig B). In both experiments, calcium response was not affected by mTLD.
Using techniques for detecting protein-protein interactions, such as protein overlay, yeast two-hybrid assay, phage display, and surface plasmon resonance, increasing numbers of GPCR-associated intracellular partners have been discovered. The intracellular carboxyl terminus cytoplasmic domain is one of the popular regions mediating receptor/non-G protein interactions (Hall et al., 1998; Fraser et al., 2000; Cong et al., 2001). The interactions of GPCRs with various cellular proteins are likely to mediate or regulate the intracellular signaling of receptor, providing explanation for the distinct functions by close-related subtypes within the same family.
However, there are few reports about coupling of 1A-AR with non-G-proteins. In this study, aa 867 to 986 of mTLD, which is located in the CUB5 domain of mTLD (Takahara et al., 1994), was identified as an 1A-AR-interacting protein. We confirmed the interaction of full-length mTLD with 1A-AR in HEK293 cells and showed that the presence of mTLD alters the localization, agonist-induced internalization, and calcium signaling of 1A-AR without affecting its pharmacological properties and intracellular signaling function.
In this study, the ability of the CUB5 domain to bind to two other 1-AR subtypes was also investigated. Results indicate that the CUB5 domain specifically interacts with the C tail of 1A-AR but not that of 1B- and 1D-AR. This may arise from the poor homology in this region (less than 10%) among the three subtypes (Weinberg et al., 1994). Our results also indicate that a discrete sequence in the cytoplasmic tail of 1A-AR, as demonstrated by the tight binding activity of aa 322 to 359 and halved binding ability of aa 399 to 467, interacted with the CUB5 domain.
So far, the function of the CUB domain remains poorly understood. Some investigators have proposed that CUB domains may be involved in protein-protein interactions (Bond and Beynon, 1995). Recent studies have shed some light on this field; for instance, the CUB1 domain of mTLD was critical for secretion, CUB2 domain was essential for the enzyme activity of the protein (Hartigan et al., 2003), and in the present studies, CUB5 domain was identified to bind to 1A-AR.
Our results demonstrated that 1A-AR not only interacted with the CUB5 domain directly but also associated with full-length mTLD. mTLD, the longer splice variant of BMP-1 (Takahara et al., 1994; Kessler et al., 1996), and BMP-1 share 702 identical amino acids, but mTLD has an additional EGF-like domain and two CUB domains, including the CUB5 domain. To date, most studies on mTLD or BMP-1 have concentrated on identifying their substrates. mTLD (or BMP-1) is able to cleave several precursors of extracellular matrix proteins including procollagens, biglycan, prolysyl oxidase and the chain of laminin-5 (Amano et al., 2000; Scott et al., 2000; Sasaki et al., 2001; Colombo et al., 2003). Little has been known about intracellular function of either mTLD or BMP-1. Only one study showed that intracellular BMP-1 is first synthesized as an inactive pro-BMP-1 and stored in the endoplasmic reticulum and early Golgi compartment (Leighton and Kadler, 2003). Because of the high homology with BMP-1, we hypothesized that mTLD may be processed in a manner similar to that of BMP-1. It is tempting to speculate that mTLD may be synthesized as an inactive form and stored in the endoplasmic reticulum and early Golgi compartments before secretion. Therefore, the intracellular abundance of mTLD made it possible to interact with 1A-AR inside the cells. According to a recent study (Morris et al., 2004), 1A-AR internalizes constitutively, and the surface receptor density is maintained by receptor recycling. This led to speculation that an internal pool of 1A-ARs recycles to allow for continuous agonist-induced signaling. In the same study, this constitutive trafficking was shown to be partially blocked by the Golgi-disturbing agent monensin, implying that the Golgi system is also a potential trafficking system. However, whether mTLD is involved in 1A-AR trafficking through the Golgi system remains unclear.
Using laser-scanning confocal microscopy and whole-cell ELISA, we showed that 1A-AR localized on the cell surface and intracellular compartment in HEK293 cells, an observation consistent with earlier studies (Hirasawa et al., 1997; Chalothorn et al., 2002). In mTLD and 1A-AR cotransfected HEK293 cells, the interaction between 1A-AR and mTLD induced the internalization and accumulation of receptors. Moreover, mTLD influenced the time course of 1A-AR inter-nalization after agonist exposure. In the absence of mTLD, there was a rapid agonist-induced internalization of cell surface receptor, a finding in agreement with previous studies in HEK293 cells (Chalothorn et al., 2002). In the presence of mTLD, the time course of 1A-AR internalization in response to agonist was different. The number of cell surface receptors initially increased during the first 5 min of agonist stimulation and then decreased by 58% by 60 min. Therefore, the results were consistent with the internalization pattern of 1A-AR in rat-1 fibroblast cells (Price et al., 2002). Price et al. (2002) observed that although the number of cell-surface wild-type 1-AR showed little increase in the early stages of agonist stimulation, the number of carboxyl-terminally truncated mutant 1-AR T348 increased significantly during the same period. Given that fibroblasts express mTLD naturally, whereas there was no mTLD expression at the mRNA level in HEK293 cells (data not shown), and considering our results that amino acid residues 322 to 359 are the primary binding region to CUB5 domain of mTLD, conflicting results of 1A-AR internalization from different groups are expected. The naturally expressed mTLD participates in the internalization of 1A-AR in fibroblasts, similar to what we observed in mTLD-transfected HEK293 cells. Partial deletion of the receptor's cytoplasmic tail would expose the primary binding region and tighten the interaction of mTLD and 1A-AR, thereby enhancing the influence of mTLD on the localization of 1A-AR.
On the other hand, phosphorylation, desensitization, internalization, and down-regulation of the receptor in response to agonist exposure would occur within seconds to hours after stimulation. Whereas arrestins have been implicated in mediating the internalization of a variety of GPCRs, the effects of -arrestin on 1A-AR are not clear. gC1q-R is a cellular protein that interacts with 1B- and 1D-AR, and regulates the cellular localization and expression of 1B-AR. Thus, it is interesting that mTLD expression alters the subcellular localization and internalization of 1A-AR as shown in the present study.
Our results suggest that mTLD did not alter the density or affinity of 1A-AR, although it altered the subcellular localization of the receptor. Thus, receptor internalization does not alter the conformation of 1A-AR, or at least does not alter the conformation of the ligand-binding region. It has been reported that 1D-ARs mainly localize intracellularly in a perinuclear orientation (Chalothorn et al., 2002) and that this fraction of the receptors can still respond well to agonist stimulation (McCune et al., 2000; Waldrop et al., 2002). Therefore, it is possible that internalized 1A-AR is still localized in certain intracellular membrane structures and preserves the receptor structure and ligand binding properties.
We observed an enhanced intracellular calcium response to 1A-AR activation in the presence of mTLD, although the time course of calcium transient decay was not altered. In addition, the effect of mTLD on intracellular calcium response to 1A-AR activation was not observed on two other GPCRs, AT1R and 1-AR. Although the physiological mechanism of calcium signaling flux remains to be investigated, it seems clear that mTLD plays an important role in the early response of 1A-AR to agonist by ensuring a rapid reaction after receptor activation.
In conclusion, we have demonstrated that mTLD is a novel and specific intracellular partner of 1A-AR, and the interaction of these two proteins influences the localization, internalization, and calcium signaling function of 1A-AR. The precise mechanism of mTLD's effects on 1A-AR is an important target for further investigation.
Acknowledgements
We thank Prof. K. P. Minneman for the generous gift of 1-AR cDNAs. We also thank Prof. K. P. Minneman and Dr. Xiaojun Du (Baker Heart Research Institute, Melbourne, Australia) for review and advice on this article.
ABBREVIATIONS: AR, adrenergic receptor; GPCR, G-protein-coupled receptor; mTLD, mammalian Tolloid; FRET, fluorescence resonance energy transfer; GST, glutathione S-transferase; PCR, polymerase chain reaction; HA, hemagglutinin; aa, amino acids; AT1R, angiotensin II type 1 receptor; EGFP, enhanced green fluorescent protein; HEK, human embryonic kidney; TRITC, tetramethylrhodamine B isothiocyanate; PBS, phosphate-buffered saline; BE2254, 2-[[2-(4-hydroxy-3-iodo-phenyl)ethylamino]methyl]tetralin-1-one; ELISA, enzyme-linked immunosorbent assay.
The online version of this article (available at http://molpharm.aspetjournals.org) contains supplemental material.
【参考文献】
Amano S, Scott IC, Takahara K, Koch M, Champliaud MF, Gerecke DR, Keene DR, Hudson DL, Nishiyama T, Lee S, et al. (2000) Bone morphogenetic protein 1 is an extracellular processing enzyme of the laminin 5 2 chain. J Biol Chem 275: 22728-22735.[Abstract/Free Full Text]
Autelitano DJ and Woodcock EA (1998) Selective activation of alpha1A-adrenergic receptors in neonatal cardiac myocytes is sufficient to cause hypertrophy and differential regulation of alpha1-adrenergic receptor subtype mRNAs. J Mol Cell Cardiol 30: 1515-1523.
Bond JS and Beynon RJ (1995) The astacin family of metalloendopeptidases. Protein Sci 4: 1247-1261.
Brzostowski JA and Kimmel AR (2001) Signaling at zero G: G-protein-independent functions for 7-TM receptors. Trends Biochem Sci 26: 291-297.
Chalothorn D, McCune DF, Edelmann SE, Garcia-Cazarin ML, Tsujimoto G, and Piascik MT (2002) Differences in the cellular localization and agonist-mediated internalization properties of the 1-adrenoceptor subtypes. Mol Pharmacol 61: 1008-1016.[Abstract/Free Full Text]
Chen S, Lin F, Iismaa S, Lee KN, Birckbichler PJ, and Graham RM (1996) Alpha1-adrenergic receptor signaling via Gh is subtype specific and independent of its transglutaminase activity. J Biol Chem 271: 32385-32391.[Abstract/Free Full Text]
Colombo M, Brittingham RJ, Klement JF, Majsterek I, Birk DE, Uitto J, and Fertala A (2003) Procollagen VII self-assembly depends on site-specific interactions and is promoted by cleavage of the NC2 domain with procollagen C-proteinase. Biochemistry 42: 11434-11442.
Cong M, Perry SJ, Lin FT, Fraser ID, Hu LA, Chen W, Pitcher JA, Scott JD, and Lefkowitz RJ (2001) Regulation of membrane targeting of the G protein-coupled receptor kinase 2 by protein kinase A and its anchoring protein AKAP79. J Biol Chem 276: 15192-15199.[Abstract/Free Full Text]
Diviani D, Lattion AL, Abuin L, Staub O, and Cotecchia S (2003) The adaptor complex 2 directly interacts with the 1b-adrenergic receptor and plays a role in receptor endocytosis. J Biol Chem 278: 19331-19340.[Abstract/Free Full Text]
Fraser ID, Cong M, Kim J, Rollins EN, Daaka Y, Lefkowitz RJ, and Scott JD (2000) Assembly of an A kinase-anchoring protein-2-adrenergic receptor complex facilitates receptor phosphorylation and signaling. Curr Biol 10: 409-412.
Grupp IL, Lorenz JN, Walsh RA, Boivin GP, and Rindt H (1998) Overexpression of alpha1B-adrenergic receptor induces left ventricular dysfunction in the absence of hypertrophy. Am J Physiol 275: H1338-H1350.
Hall RA, Premont RT, Chow CW, Blitzer JT, Pitcher JA, Claing A, Stoffel RH, Barak LS, Shenolikar S, Weinman EJ, et al. (1998) The beta2-adrenergic receptor interacts with the Na+/H+-exchanger regulatory factor to control Na+/H+ exchange. Nature (Lond) 392: 626-630.
Hall RA, Premont RT, and Lefkowitz RJ (1999) Heptahelical receptor signaling: beyond the G protein paradigm. J Cell Biol 145: 927-932.[Free Full Text]
Hartigan N, Garrigue-Antar L, and Kadler KE (2003) Bone morphogenetic protein-1 (BMP-1). Identification of the minimal domain structure for procollagen C-proteinase activity. J Biol Chem 278: 18045-18049.[Abstract/Free Full Text]
Heuss C and Gerber U (2000) G-protein-independent signaling by G-protein-coupled receptors. Trends Neurosci 23: 469-475.
Hirasawa A, Sugawara T, Awaji T, Tsumaya K, Ito H, and Tsujimoto G (1997) Subtype-specific differences in subcellular localization of alpha1-adrenoceptors: chlorethylclonidine preferentially alkylates the accessible cell surface 1-adrenoceptors irrespective of the subtype. Mol Pharmacol 52: 764-770.[Abstract/Free Full Text]
Kessler E, Takahara K, Biniaminov L, Brusel M, and Greenspan DS (1996) Bone morphogenetic protein-1: the type I procollagen C-proteinase. Science (Wash DC) 271: 360-362.
Klein U, Ramirez MT, Kobilka BK, and von Zastrow M (1997) A novel interaction between adrenergic receptors and the alpha-subunit of eukaryotic initiation factor 2B. J Biol Chem 272: 19099-19102.[Abstract/Free Full Text]
Leighton M and Kadler KE (2003) Paired basic/Furin-like proprotein convertase cleavage of Pro-BMP-1 in the trans-Golgi network. J Biol Chem 278: 18478-18484.[Abstract/Free Full Text]
McCune DF, Edelmann SE, Olges JR, Post GR, Waldrop BA, Waugh DJ, Perez DM, and Piascik MT (2000) Regulation of the cellular localization and signaling properties of the 1B- and 1D-adrenoceptors by agonists and inverse agonists. Mol Pharmacol 57: 659-666.[Abstract/Free Full Text]
Menard L, Ferguson SS, Zhang J, Lin FT, Lefkowitz RJ, Caron MG, and Barak LS (1997) Synergistic regulation of 2-adrenergic receptor sequestration: intracellular complement of -adrenergic receptor kinase and -arrestin determine kinetics of internalization. Mol Pharmacol 51: 800-808.[Abstract/Free Full Text]
Morris DP, Price RR, Smith MP, Lei B, and Schwinn DA (2004) Cellular trafficking of human 1a-adrenergic receptors is continuous and primarily agonist-independent. Mol Pharmacol 66: 843-854.[Abstract/Free Full Text]
Price RR, Morris DP, Biswas G, Smith MP, and Schwinn DA (2002) Acute agonist-mediated desensitization of the human 1a-adrenergic receptor is primarily independent of carboxyl terminus regulation: implications for regulation of alpha 1aAR splice variants. J Biol Chem 277: 9570-9579.[Abstract/Free Full Text]
Pupo AS and Minneman KP (2002) Interaction of neuronal nitric oxide synthase with alpha1-adrenergic receptor subtypes in transfected HEK-293 cells. BMC Pharmacol 2: 17.
Pupo AS and Minneman KP (2003) Specific interactions between gC1qR and alpha1adrenoceptor subtypes. J Recept Signal Transduct Res 23: 185-195.
Sasaki T, Gohring W, Mann K, Brakebusch C, Yamada Y, Fassler R, and Timpl R (2001) Short arm region of laminin-5 gamma2 chain: structure, mechanism of processing and binding to heparin and proteins. J Mol Biol 314: 751-763.
Scott IC, Imamura Y, Pappano WN, Troedel JM, Recklies AD, Roughley PJ, and Greenspan DS (2000) Bone morphogenetic protein-1 processes probiglycan. J Biol Chem 275: 30504-30511.[Abstract/Free Full Text]
Takahara K, Lyons GE, and Greenspan DS (1994) Bone morphogenetic protein-1 and a mammalian tolloid homologue (mTld) are encoded by alternatively spliced transcripts which are differentially expressed in some tissues. J Biol Chem 269: 32572-32578.[Abstract/Free Full Text]
Vazquez-Prado J, Medina LC, Romero-Avila MT, Gonzalez-Espinosa C, and Garcia-Sainz JA (2000) Norepinephrine- and phorbol ester-induced phosphorylation of 1a-adrenergic receptors. Functional aspects. J Biol Chem 275: 6553-6559.[Abstract/Free Full Text]
Waldrop BA, Mastalerz D, Piascik MT, and Post GR (2002) 1B- and 1D-Adrenergic receptors exhibit different requirements for agonist and mitogen-activated protein kinase activation to regulate growth responses in rat 1 fibroblasts. J Pharmacol Exp Ther 300: 83-90.[Abstract/Free Full Text]
Wang SQ, Song LS, Lakatta EG, and Cheng H (2001) Ca2+ signalling between single L-type Ca2+ channels and ryanodine receptors in heart cells. Nature (Lond) 410: 592-596.
Weinberg DH, Trivedi P, Tan CP, Mitra S, Perkins-Barrow A, Borkowski D, Strader CD, and Bayne M (1994) Cloning, expression and characterization of human alpha adrenergic receptors alpha 1a, alpha 1b and alpha 1c. Biochem Biophys Res Commun 201: 1296-1304.
Xu Q, Zhang T, Han QD, and Zhang YY (2003) Binding between alpha 1A-adrenergic receptor and segment of bone morphogenetic protein-1 in human embryonic cell 293. Sheng Li Xue Bao 55: 692-698.
Zhong H, Lee D, Robeva A, and Minneman KP (2001) Signaling pathways activated by alpha1-adrenergic receptor subtypes in PC12 cells. Life Sci 68: 2269-2276.
Zhong H and Minneman KP (1999) Differential activation of mitogen-activated protein kinase pathways in PC12 cells by closely related alpha1-adrenergic receptor subtypes. J Neurochem 72: 2388-2396.
Zhu W, Zhang Y, and Han C (1997) Characterization of subtype of alpha1-adrenoceptor mediating vasoconstriction in perfused rat hind limb. Eur J Pharmacol 329: 55-61.
作者单位：Institute of Vascular Medicine, Peking University Third Hospital and Key Laboratory of Molecular Cardiovascular Sciences, Ministry of Education, Beijing, China