Our published studies show that the distribution of the ANG II type 1 (AT1) receptor (AT1R), expressed as a enhanced yellow fluorescent fusion (YFP) protein (AT1R/EYFP), is altered upon cellular treatment with ANG II or coexpression with intracellular ANG II. AT1R accumulates in nuclei of cells only in the presence of ANG II. Several transmembrane receptors are known to accumulate in nuclei, some as holoreceptors and others as cleaved receptor products. The present study was designed to determine whether the AT1R is cleaved before nuclear transport. A plasmid encoding a rat AT1R labeled at the amino terminus with enhanced cyan fluorescent protein (CFP) and at the carboxy terminus with EYFP was employed. Image analyses of this protein in COS-7 cells, CCF-STTG1 glial cells, and A10 vascular smooth muscle cells show the two fluorescent moieties to be largely spatially colocalized in untreated cells. ANG II treatment, however, leads to a separation of the fluorescent moieties with yellow fluorescence accumulating in more than 30% of cellular nuclei. Immunoblot analyses of extracts and conditioned media from transfected cells indicate that the CFP domain fused to the extracellular amino-terminal AT1R domain is cleaved from the membrane and that the YFP domain, together with the intracellular cytoplasmic carboxy terminus of the AT1R, is also cleaved from the membrane-bound receptor. The carboxy terminus of the AT1R is essential for cleavage; cleavage does not occur in protein deleted with respect to this region. Overexpressed native AT1R (nonfusion) is also cleaved; the intracellular 6-kDa cytoplasmic domain product accumulates to a significantly higher level with ANG II treatment.
- nuclear angiotensin II type 1 receptor
our recent published work has shown that intracellular (IC) expression in a rat vascular smooth muscle cell (VSMC) line of ANG II with the type 1A seven-transmembrane G protein-coupled receptor (AT1) results in nuclear accumulation of the AT1 receptor (AT1R), activation of p38 MAPK and cAMP response element-binding protein pathways, and enhanced cell proliferation (8). In those studies, ANG II was expressed as enhanced cyan fluorescent protein (ECFP)/ANG II (fused at the amino terminus to ECFP) and AT1R was expressed as AT1R/enhanced yellow fluorescent protein (EYFP) (fused at the carboxy terminus to EYFP). Our studies, using fluorescent colocalization markers, show that AT1R/EYFP accumulates in the endoplasmic reticulum, Golgi, vesicles, and plasma membrane when expressed exclusively but that the distribution changes upon coexpression with ECFP/ANG II. Whereas <1% of rat A10 VSMCs transfected with pAT1R/EYFP show yellow nuclear fluorescence, 48% of cells that express both ECFP/ANG II and AT1R/EYFP show nuclear yellow fluorescence, indicating nuclear transport of the protein (8). Therefore, nuclear transport of the receptor is temporally linked to several quantifiable cellular changes.
A number of transmembrane receptors, including several receptor tyrosine kinases (e.g., receptors for EGF, insulin, FGF, NGF, IL-1, erbB-4, and Her-2/neu) (25), have been reported to localize to the nucleus, either as holoproteins or protein cleavage fragments. More recently, a limited number of G protein-coupled receptors (GPCRs), including the endothelin-B and GABAB receptor, have also been found associated with cell nuclei (6, 14, 32).
Our previously published studies, in which we report nuclear accumulation of the AT1R when it is expressed with IC ANG II (8, 9), were not designed to differentiate between transport of the AT1R holoprotein versus transport of cleaved IC fragments into the nucleus. We have, therefore, expanded our approaches to investigate that question. In the present study we designed and employed an expression plasmid (pECFP/AT1R/EYFP) that encodes a fusion protein in which the AT1R is labeled at the extracellular amino terminus with ECFP and at the IC carboxy terminus with EYFP. In principle, cleavage within the AT1R leads to dissociation of the two fluorescent moieties and possible spatial separation within the cell or at the plasma membrane. We investigated this construct and related control plasmids using deconvolution microscopy and Western blot analyses to show that the AT1R undergoes cleavage within the cytoplasmic domain and transport of the cytoplasmic fragment into the nucleus. We show that the AT1R cytoplasmic domain is essential for the processing event; cleavage does not occur when the encoded protein is deleted with respect to amino acid residues 306–359. We further demonstrate that the amino-terminal extracellular domain also undergoes cleavage and can be recovered from the tissue culture media.
MATERIALS AND METHODS
pECFP-C1 and pEYFP-N1.
These are otherwise referred to as pECFP and pEYFP in this study and are control vehicles into which desired fusion protein-encoding DNA sequences can be cloned (Clontech, Palo Alto, CA).
This clone is also referred to as pAT1R/EYFP-N1/ΔEYFP. It is the same clone as above except that it is deleted with respect to the EYFP fragment. Thus the AT1R is expressed as a nonfusion protein under control of the cytomegalovirus promoter (pCMV). To produce this expression plasmid, we digested pAT1R/EYFP with AgeI/NotI, both of which digest only once in pEYFP-N1 (at sites directly flanking the EYFP gene) and not at all in AT1R. The large fragment (∼5,050 bp) was purified free of the EYFP fragment (∼731 bp), end filled, and blunt ligated to produce pCMV/AT1R.
pEYFP-N1 was digested with AgeI/NotI, both of which digest only once in pEYFP-N1 (at sites directly flanking the EYFP gene). The large fragment (∼4 kb) was purified free of the EYFP fragment (∼731 bp), end filled, and blunt ligated to produce pEYFP-N1/ΔEYFP. This control plasmid encodes neither a fluorescent protein nor a member of the renin-angiotensin system.
pECFP-C1 was digested with NheI and BglII to liberate a 755-bp ECFP-encoding fragment. The ECFP fragment was inserted into NheI/BglII-digested pEYFP-N1. This vector is designated pECFP/EYFP (note that the fluorescent moieties are not in frame). Upstream primer AT1-DF-U, 5′-GATCGAAAGCTTCTGCCACCATGGCCCTTAACTCTTCTGCT-3′ (ATG start site underlined) and downstream primer AT1-DF-D, 5′-CGAGACCGAGGATCCTGCTCCACCCTCAAAACAAGACGC-3′ were used to amplify AT1R from pAT1R/EYFP. The PCR product was digested with Hind III/BamHI and ligated to HindIII/BamHI-digested pECFP/EYFP, placing the AT1R in proper reading frame with both upstream ECFP and downstream EYFP sequences.
The human IGF-1 receptor signal peptide from plasmid pIGF-I-R.8 (ATCC 59294) was amplified using upstream primer HsIGF1R-U1, 5′-GATCGAGCTAGCGCCACCATGAAGTCTGGCTCCGGAGGA-3′ and downstream primer HsIGF1R-D1, 5′-GATCGAGCTAGCTGTCCACTCGTCGGCCAGAGCGA-3′. The ∼100-bp PCR amplification product was digested with NheI and cloned into NheI-digested pECFP/AT1R/EYFP. The clone was sequenced to confirm orientation of the signal peptide upstream of ECFP.
The rat AT1a receptor-encoding DNA, deleted with respect to the cytoplasmic domain (Gly306–Glu359) was PCR-amplified from AT1R/EYFP using upstream primer AT1-DF-U (see pECFP/AT1R/EYFP) and downstream primer AT1-DF-D2, 5′-CGAGACCGAGGATCCTGCAGAAAGCCGTAGAACAGAGG-3′. The product (∼950 bp) was digested with HindIII/BamHI and ligated to HindIII/BamHI-digested pECFP/EYFP to generate pECFP/AT1RΔCT/EYFP.
The clonal cell line A10 (ATCC, CRL-1476) was derived from the thoracic aorta of DB1X embryonic rat and possesses many of the properties characteristic of smooth muscle cells. The cells produce spontaneous action potentials at the stationary phase of the growth cycle and exhibit an increase in activity of the enzymes myokinase and creatine phosphokinase. In addition, these cells have been shown to express both the AT1 receptor and a number of contractile proteins characteristic of VSMCs (2, 15, 21, 36, 47).
CCF-STTG1 cells (ATCC, CRL-1718) are derived from a human grade IV astrocytoma and express the glial marker GFAP (4). These are alternatively referred to CCF cells in this study.
COS-7 is an African green monkey kidney fibroblast-like cell line suitable for transfection by vectors requiring expression of SV40 T antigen. The line was derived from the CV-1 cell line (ATCC, CCL-70) by transformation with an origin-defective mutant of SV40 that codes for wild-type T antigen. We typically use this line as an initial test line for transfections (9) because it is readily transfected using commercial cationic lipids and expresses proteins from recombinant constructs at high levels. This line is alternatively referred to as COS cells in this study.
Cell Growth Measurements
A10 VSMCs, passages 15–20, were plated in DMEM high glucose at 105 cells/well on 12-well plates and permitted to attach overnight before transfection. Cells were transfected, six wells per clone or combination of clones, with various plasmids (0.5 μg DNA and 1.25 μg of lipofectamine, 2,000/well) for 4 h. Media were replaced at 24 h. At 42 h, bromodeoxyuridine (BrdU, 10 μmol/l) labeling in 1% FBS-containing medium was initiated, and labeling proceeded for 6 h (based on our pilot studies, the optimum labeling time for these cells). Where appropriate, candesartan (10−9 mol/l) and/or ANG II (10−9 mol/l) were added during the BrdU-labeling period. The labeling reagent was removed, and anti-BrdU staining was conducted using the Oncogene Research Products BrdU kit (Calbiochem-Novabiochem, San Diego, CA). Approximately 100 cells/field × 10 fields were counted (1,000 for each treatment) by a technician blinded to the study identifiers.
Transfections/Western Blot Analysis
Line A10, passages 15–20, was seeded at 2 × 105 cells/35-mm dish in DMEM and 10% FBS. Twenty-four hours after being plated, cells were transfected overnight using 0.5 μg of DNA and 1.25 μl of Lipofectamine 2000 (1 μg/μl; Invitrogen, Carlsbad, CA) in Optimem low-serum medium. COS cells were similarly transfected. CCF cells were seeded at 2 × 105 cells/35-mm dish in RPMI and 10% FBS. Twenty-four hours after being plated, cells were transfected with 3 μl of Fugene-6 (Roche Diagnostics) and 0.5 μg of DNA for 4 h in serum-free medium.
Where indicated, at ∼45 h posttransfection, cells were rendered quiescent in 0.5% FBS-containing media for 3 h and then treated with ANG II (Sigma, St. Louis, MO) for 30 min before harvest. EDTA (Fisher Scientific, Fair Lawn, NJ) and 1,10-ortho-phenanthroline monohydrate (OPA) were used as indicated. Total cell extract was collected (1, 8), and protein was measured by using the Bio-Rad Protein Assay (Hercules, CA). Protein was electrophoresed on a 4%-12% NuPAGE Bis-Tris gradient gel or an 18% Novex Tris-glycine gel (Invitrogen) and transferred to Hybond-P:PVDF membrane. Specific immunoreactive proteins were detected by using ECL Plus Western blotting detection reagents (Amersham Biosciences, Piscataway, NJ). AT1R (COOH terminus) was detected by using sheep polyclonal (ab1886) to AT1R (rat protein, amino acids 297–359) and rabbit anti-sheep IgG (both from Abcam, Cambridge, MA). AT1R (NH2 terminus) was detected by using rabbit polyclonal IgG to AT1 (N-10, sc-1173) from Santa Cruz and donkey anti-rabbit IgG (Amersham). EYFP was detected by using rabbit polyclonal antibodies to GFP (FL, sc-8334, Santa Cruz Biotechnology, Santa Cruz, CA) and donkey anti-rabbit horseradish peroxidase (NA934V, Amersham Biosciences). Mouse monoclonal anti-histone and anti-α-tubulin antibodies (Santa Cruz Biotechnology) were used as controls for nuclear and cytosolic fractions, respectively. High and low molecular weight markers were High- and Low-Range Rainbow Molecular Weight Markers (Amersham Biosciences).
Chinese hamster ovary (CHO)-K1 and A10 cells were plated at 2 × 106/100-mm dish, serum deprived at 24 h of posttransfection, and (where indicated) treated with ANG II for 30 min before extract collection. Extracts were fractionated by a modification of the method of Maloney et al. (26). Specifically, plates were washed with 4-ml cold PBS and (step 1) scraped in 500 μl of CSK buffer [(in mM) 10 Tris·HCl (pH 6.8), 3 MgCl2, 100 NaCl, 1 EGTA, 300 sucrose, 1 DTT, 1 PMSF, 10 NaF, and 1 NaVO3 and phosphatase inhibitor]. Samples were transferred to a Dounce homogenizer and lysed with 30 strokes of a B-type pestle. Lysis was confirmed by microscopy using trypan blue staining. Lysate was transferred to an eppendorf and centrifuged for 10 min at 1,000 g. The pellet from this step was (step 2) resuspended in 500 μl of CSK buffer and centrifuged for 5 min at 4°C (2,000 g). The resulting pellet was then resuspended in 200 μl CSK-1% TX-100 buffer, incubated on ice for 5 min, and then centrifuged at 10,000 g for 5 min at 4°C. The supernatant was diluted with Laemmli sample buffer and represents the nuclear fraction. The supernatant from step 1 was (step 3) centrifuged at 8,000 g for 10 min at 4°C. The supernatant was (step 4) removed and centrifuged in a SW50.1 rotor (35,000 rpm for 30 min at 4°C). The resulting supernatant was diluted with 4× Laemmli sample buffer to 1× for cytoplasmic fraction.
The pellet from step 4 was resuspended in CSK-1% TX-100 buffer, incubated on ice for 30 min, and then centrifuged at 100,000 g for 30 min. The supernatant was diluted with Laemmli sample buffer (represents membrane fraction). Western blot analyses were performed (see Transfections/Western Blot Analysis; antibodies are also described) using membrane, cytosolic, and nuclear fractions.
VSMCs were grown to 85%-90% confluence, after which media were changed to serum-free media for 24 h before ANG II treatment. Cells were treated with ANG II (10−7 mol/l) for 1 h (with fresh additions every 15 min). Conditioned media were collected and centrifuged to remove cells and particulates. An equal volume of 15% TCA and carrier BSA to a final concentration of 2 μg/ml were added to precipitate protein (17). Samples were incubated at 4°C overnight and then centrifuged at 11,000 rpm at 4°C for 30 min. Pellets were resuspended in 50 μl of lysis buffer, diluted with 20 μl of sample buffer, and electrophoresed as described in Transfections/Western Blot Analysis for Western blot assays.
Transfections of plasmid DNA:polycationic lipid complexes were as described previously (see Transfections/Western Blot Analysis). Transfections were performed on glass bottom dishes (No. 1.5, poly-d-lysine coated) from MatTek (Ashland, MA).
Three-dimensional Deconvolution Microscopy
Deconvolution microscopy was performed using a Zeiss Axiovert 200 M microscope, xenon light source with automated Z-axis, and appropriate filters. Constrained iterative and nearest neighbor algorithms were performed by using Slidebook 4.1.0 software (Intelligent Imaging Innovations, Denver, CO). All images were captured with a 63X PlanApo oil lens (1.4 numerical apperature) in three dimensions at 0.5-μm steps through the vertical z-axis (∼10–15 z-axis planes captured per image) and deconvolved to render confocal images. Images shown here are approximately midnuclear in the z-axis.
Statistics and Correlation Coefficients
Pearson correlation coefficients were determined by using Slidebook 4.1.0 software. Images were deconvolved, segment masks were applied for yellow and cyan fluorescence, intensities and cross-channel statistical analyses were performed, and correlation coefficients were calculated.
Groups were compared using a one-way ANOVA with Tukey-Kramer or Bonferroni multiple-comparison post hoc tests.
Imaging: Effects of ANG II on ECFP/AT1R/EYFP
Our results show that the fluorescent moieties of double-labeled AT1R (ECFP/AT1R/EYFP) are coincidental when expressed in transfected untreated COS cells (Fig. 1, row 1). However, ANG II treatment alters the distribution of fluorescence. When expressed in ANG II-treated COS cells, double-labeled AT1R was found to be cleaved with corresponding transport of the yellow fluorescent moiety to the nucleus (Fig. 1, row 2). The cyan fluorescent moiety does not accumulate in cell nuclei. Moreover the periphery of cells is devoid of cyan fluorescence, suggesting that the cyan fluorescent moiety may be cleaved from the plasma membrane and released into the medium, perhaps with a portion of the extracellular AT1R domain. The same fluorescent distribution is observed as early as 48 h and as late as 96 h posttransfection.
To confirm these results in a biologically pertinent cell line, we tested CCF-STTG1, a glial cell line that is reported to express angiotensinogen mRNA and protein and to exhibit nuclear localization of an angiotensinogen fusion protein (36). In untreated CCF cells, as in COS cells, ECFP/AT1R/EYFP demonstrates little nuclear fluorescence (Table 1). However, in ANG II-treated transfected cells, nuclear yellow fluorescence accumulates in 39% of cells (Table 1 and Fig. 1, row 3), and there is a corresponding loss of cyan fluorescence from the cell periphery. In addition, some transfections were performed with a construct encoding a double-fluorescent moiety modified by the addition of the human IGF-1 signal peptide at the amino terminus. This method has been previously used to increase the efficiency of plasma membrane transport of integral membrane proteins, like the AT1R, which have no true signal peptide (25). No significant difference was observed for cells transfected with pIGF1SS/ECFP/AT1R/EYFP compared with pECFP/AT1R/EYFP and imaged at 72 h of posttransfection (not shown).
In A10 cells, as in COS and CCF cells, the yellow and cyan fluorescent moieties of ECFP/AT1R/EYFP are coincidental in untreated cells (Tables 1 and 2). ANG II administration in low-serum medium induces a rapid accumulation of yellow fluorescence in the nucleus; the nucleus remains free of cyan fluorescence (Tables 1 and 2; and Fig. 1, row 4). Importantly, in all tested cell lines, ANG II treatment frequently causes changes in cellular morphology, leading to flattening, ruffling, enhanced focal accumulation of yellow fluorescence (clustering), and extension of lamellapodia or filapodia.
To determine the role of the carboxy terminus in receptor processing, we constructed an AT1a-receptor mutant, deleted with respect to the complete cytoplasmic domain, amino acids 306–359. Following ANG II-treatment of ECFP/AT1RΔCT/EYFP-transfected cells, we find that the YFP domain is not visibly processed and there is no accumulation of yellow fluorescence in the nucleus (Fig. 1, row 5; and Table 2). Therefore, the AT1R carboxy terminus is necessary for processing and transport of the yellow fluorescent moiety to the nucleus; the yellow fluorescent domain is not itself cleaved in the absence of the AT1R carboxy terminus; the proteolytic activity is not nonspecific.
Transient Transfections, Cell Proliferation: ANG II Stimulates pECFP/AT1R/EYFP-Transfected Cells
Since the distribution of yellow fluorescence from pECFP/AT1R/EYFP follows closely that which we have observed in previous studies with pAT1R/EYFP, we suspected that ECFP/AT1R/EYFP retains biological function, as does AT1R/EYFP. Nevertheless, we performed a series of experiments to verify that ECFP/AT1R/EYFP stimulates A10 VSMC proliferation in the presence of ANG II. Exogenous ANG II treatment significantly increases proliferation (1.44-fold) of untransfected A10 VSMCs, consistent with the existence of endogenous receptor in these cells (Fig. 2). ANG II increases proliferation of both pAT1R/EYFP- and pECFP/AT1R/EYFP-transfected cells ∼1.7-fold over corresponding untreated cells and in a candesartan-sensitive fashion. Receptor transfection in no case increases baseline proliferation in the absence of ANG II treatment. These studies show that ECFP/AT1R/EYFP, like AT1R/EYFP, retains native biological properties.
Western Blot Analysis: Cleavage of AT1R/EYFP and ECFP/AT1R/EYFP
To determine whether the AT1 receptor processing evident in deconvolved fluorescent images could be corroborated by electrophoretic protein analyses, extracts from transfected CCF cells were evaluated. Western blots were probed with anti-GFP antibody (polyclonal, generated to full-length GFP), which reacts with EYFP and ECFP proteins (Fig. 3). Extracts from pEYFP-N1-transfected cells, as predicted, produce a band of 29–30 kDa. pAT1R/EYFP-transfected cells produce a predominant broad band between 66 and 97 kDa (predicted molecular weight based on linear regression of protein markers is 85 kDa). This is in good agreement with the minimum theoretical molecular weight of 71 kDa. Of course, the AT1R typically has a molecular weight between 41 and 58 kDa (with as high as 75 kDa documented), depending on posttranslational modifications (16, 32). pECFP/AT1R/EYFP-transfected CCF cells show a predominant cluster of immunoreactive proteins of >97 kDa (consistent with the theoretical minimum size of 101 kDa). Cells transfected with either pAT1R/EYFP or pECFP/AT1R/EYFP likewise demonstrate a smaller immunoreactive fragment of ∼36 kDa. This fragment size is consistent with the size of the IC EYFP domain together with the IC cytoplasmic carboxy terminus of the AT1R, suggesting that the receptor is cleaved either 1) within the IC receptor domain and close to the membrane or 2) within the seventh transmembrane-spanning region.
Results from Western blot analyses (Fig. 4) of pAT1R/EYFP- and pECFP/AT1R/EYFP-transfected A10 cells, using the polyclonal GFP antibody made to full-length GFP and a polyclonal AT1R antibody generated against the cytoplasmic carboxy terminus of the AT1R (residues 297–359), reveal the existence of a small carboxy-terminal cleavage fragment (CT-CF) that cross-reacts with the GFP antibody and the carboxy-terminal AT1R antibody (consistent with the cytoplasmic domain of the AT1R fused to EYFP). The relative amount of CT-CF increases 5.5–6.4-fold (P < 0.05, 5 measurements) following ANG II treatment.
No product conforming to the predicted size of ECFP/AT1R (less the carboxy-terminal cytoplasmic fragment) is detected, suggesting that the extracellular ECFP moiety may be cleaved off and released into the culture medium.
Processing of ECFP/AT1R/EYFP: Effects of Metalloprotease Inhibitors
The canonical receptor tyrosine kinase model (e.g., ErbB-4) involves cleavage of the extracellular receptor domain by a metalloprotease, such as TNF-α-converting enzyme (TACE) followed by cleavage of the IC domain [receptor intramembrane proteolysis (RIP)] by a secretase enzyme (6). To determine whether metalloproteases might be involved in cleavage of the AT1R, we transfected A10 cells with pAT1R/EYFP or pECFP/AT1R/EYFP, treated with the metal-chelating metalloprotease inhibitors EDTA or OPA, and evaluated extracts for accumulation of the cleavage fragment CT-CF. Both EDTA and OPA inhibit accumulation of the CT-CF (Fig. 5). Furthermore, both EDTA and OPA inhibit accumulation of nuclear yellow fluorescence by image analyses (Table 1).
Processing of Native AT1R
To determine whether native, unmodified AT1R is cleaved, we sought to identify a low molecular weight fragment on high percentage polyacrylamide gels following transfection of CHO-K1 cells, which lack detectable endogenous AT1R (10, 21, 24, 39, 40), with a plasmid that encodes the native AT1R protein. Cells were transfected with pEYFP-N1/ΔEYFP (control) or pCMV/AT1R. Where indicated, cells were treated with ANG II (10−6 mol/l for 30 min) before collecting extracts. Extracts were electrophoresed on an 18% gel with appropriate markers, and immunoblots were performed using anti-AT1R antibody (Fig. 6). We identified the presence of a 6- to 7-kDa fragment in pCMV/AT1R-transfected cells, which was not present in control extracts. This was consistent with the size of the cytoplasmic domain of the unmodified, native AT1R. Moreover, the intensity of this fragment was amplified (3.2-fold, P < 0.05, n = 3 experiments; Fig. 6A, lanes 7–9) with ANG II treatment. In cells treated with EDTA (Fig. 6B, lanes 11–12) or OPA (lanes 14–15) for 12 h before harvest, no detectable cleavage fragment accumulates..
Extracts were collected from CHO-K1 and A10 cells and fractionated into cytoplasmic, membrane, and nuclear components, and Western blot analyses were performed (Fig. 7). The carboxy-terminal tail was identified using anti-AT1R (carboxy terminal) antibodies. The 7-kDa cleavage fragment is detected in nuclear and cytoplasmic fractions predominantly and only in A10 cells.
Recovery of the AT1R Amino-Terminal Fragment from Media of ECFP/AT1R/EYFP-Transfected Cells
As mentioned earlier, no product conforming to the predicted size of ECFP/AT1R (less the carboxy-terminal cytoplasmic fragment) is detected in cell extracts prepared from pECFP/AT1R/EYFP-transfected cells, prompting us to question whether the extracellular domain of the AT1R might also be cleaved from the membrane. To test this hypothesis, CHO-K1 cells, which lack detectable AT1R (10, 21, 24, 39, 40), were transfected with pAT1R/EYFP or pECFP/AT1R/EYFP, and conditioned media were collected (15). Western blot analysis of protein concentrated from the conditioned media was conducted (Fig. 8). A band of ∼33 kDa is detected using both anti-GFP (cross-reacts with ECFP) and anti-AT1R (amino terminal specific) antibodies in proteins collected from pECFP/AT1R/EYFP-transfected cells. This size is consistent with that expected for CFP (∼30 kDa) plus amino terminus of AT1R (amino acids 1–28, ∼3 kDa), suggesting cleavage at or near the extracellular domain/TM1 juxtaposition. As expected, this band is not observed in proteins collected from pAT1R/EYFP-transfected cells. Without the CFP component, the 3-kDa amino-terminal cleavage fragment is not visible within the limitations of standard Western blot procedures.
The AT1 receptor is internalized from the plasma membrane and undergoes extensive recycling, accumulating in endosomes of the short-recycling pathway as well as the long-recycling perinuclear compartment (PNRC) (19). The function of the latter compartment is unknown, but materialization in the PNRC appears to slow the return of the receptor to the plasma membrane. At the start of the current study, we speculated that AT1R derived from IC endosomes (in particular, those of the PNRC) might function in an IC fashion to stimulate cell proliferation and hypertrophy. Indeed, functionally active IC holoreceptor might yet be derived from IC pools of AT1R, but our present studies indicate that at least some AT1R subunits undergo cleavage at the plasma membrane to produce a carboxy-terminal fragment population which traffics to cell nuclei and that these events are accompanied by measurable biological changes. Is this design precedented within the GPCR superfamily?
In addition to the many examples of single-pass transmembrane receptors, which as holoproteins or processed fragments translocate to the nucleus (5, 7, 11, 13, 22, 23, 25, 31, 34, 38, 40, 41, 44–46), a few multipass seven-transmembrane GPCRs, including the growth hormone-releasing hormone, endothelin-B, and GABAB receptors, have recently been found either to be processed or to be transported to the nucleus or both (6, 14, 16, 30).
Our previously published studies indicate that the expression of the single-fluorescent moiety fusion protein, AT1R/EYFP, stimulates proliferation of A10 VSMCs. The present study indicates that the expression of the related double-fluorescent protein, ECFP/AT1R/EYFP, similarly stimulates cell proliferation in ANG II-treated CCF and A10 cells. ECFP/AT1R/EYFP also undergoes cleavage with transport of the YFP domain to the nucleus and accumulation of a 36-kDa fragment, CT-CF. Interestingly, CT-CF accumulates regardless of ANG II treatment, but the quantity is amplified following ANG II treatment. However, the fragment only significantly translocates to the nucleus in ANG II-treated cells. Therefore, ANG II seems to have a role in both accumulation and transport of the CT-CF.
A number of different regulatory domains and functions map to the carboxy terminus of the AT1R. Specific residues in the carboxy-terminal tail play roles in G protein coupling and receptor uptake, whereas phosphorylation of serine and threonine residues by PKA and PKC may result in uncoupling from G proteins and receptor desensitization. Using a carboxy-terminal deletion mutant (Δ309–359), which is very similar to the mutant that we generated and used in this study, Inagami and associates (33) demonstrated that the mutant AT1 receptor shares a similar ANG II-binding affinity and maximum binding value as wild-type but markedly reduced G protein interaction. This suggests that the carboxy-terminal cytoplasmic domain is involved in G protein coupling but not in cell-surface presentation or ANG II binding. Our imaging studies suggest that ECFP/AT1RΔCT/EYFP is properly transported to the plasma membrane upon ANG II treatment but that the fluors remain coincidental, and there is no accumulation of nuclear fluorescence indicating no cleavage and no nuclear transport. Our immunoblot studies corroborate this conclusion. Importantly, therefore, the cytoplasmic carboxy terminal domain is obligatory for processing.
Although there have not been any previous reports directly demonstrating that the AT1R undergoes biologically functional proteolytic cleavage, there exists some indirect supporting evidence. Modrall and colleagues (29) postulated that receptor downregulation might occur independently of receptor endocytosis. Using endocytosis-deficient mutants (carboxy terminus deleted), they showed that receptors were downregulated both by 125I-labeled ANG II endocytosis and by radioligand assays for AT1 receptor binding sites. They further demonstrated that the endocytosis-deficient mutant receptors (Δ309–359 and Δ311–359) were fully capable of rapid downregulation comparable with that of the wild-type receptor. They suggest, therefore, that an alternative pathway, that of receptor degradation, might be responsible for loss of cell-surface receptor. Our studies, showing nuclear accumulation of the carboxy-terminal cytoplasmic fragment of the AT1R, accompanied by alterations in signal transduction and cell proliferation, suggest that the degradation of plasma membrane receptor actually represents a biologically functional-regulated proteolysis.
The ANG II type 2 (AT2) receptor, in addition to its traditional plasma membrane function, appears to have a unique and important IC role. The AT2 receptor, which shares only 32% amino acid sequence homology with the AT1 receptor, has recently been shown to bind to a transcription factor and to drive its localization to the nucleus. The rat AT2 receptor carboxy terminus binds to the transcription factor promyelocytic zinc finger protein (PLZF) on yeast two-hybrid screens of a human heart library (37). Confocal microscopy showed that ANG II induces cytosolic PLZF to colocalize with AT2R at the plasma membrane and then drives the receptor and PLZF to internalize. PLZF slowly appears in the nucleus, whereas AT2R accumulates in the perinuclear region. Nuclear PLZF binds to a number of genes that contribute to protein synthesis, and the authors suggest that these AT2 receptor-mediated changes in gene regulation could, in effect, contribute to cardiac hypertrophy. The AT2 receptor, in whole or in part, does not appear to translocate into the nucleus, unlike the AT1R, yet this is an example of a receptor that, through the cytoplasmic carboxy terminus, serves a unique IC chaperone function and, thus, contributes to alterations in gene expression.
The existence of prototypical plasma membrane receptors within cell nuclei and accompanying evidence that at least some of these are involved in transcriptional regulation of gene expression prompt us to ask why some receptors have evolved to occupy multiple cellular locations and presumably perform several functions. The most reasonable answers are that nuclear accumulation of a conventional “plasma membrane” receptor or receptor product may 1) contribute to the amplification of a downstream response to an external stimulus, 2) prolong a response to a stimulus, and/or 3) reduce the degeneracy of ligand-receptor signaling pathways and nuclear responses that are shared by multiple cell-surface receptors (i.e., increase specificity).
The discovery of traditional plasma membrane receptors in the nucleus is a conundrum that has plagued scientists working in related disciplines for several years. Although clear evidence exists for the presence of uncleaved holoreceptors, including those for FGF-1, ErbB-3, and epidermal growth factor receptor in the nucleus, the mechanism by which protein possessing hydrophobic domains, and associated with a lipid bilayer, may escape the membrane and acquire soluble properties that permit nuclear transport and nuclear function remains a mystery. Receptors, such as that for ErbB-4, Alzheimer precursor protein, and now AT1R, that undergo cleavage to release soluble cytoplasmic domains, which are then transported to the cell nucleus, pose less of a conceptual challenge.
This work was supported by the Ochsner Clinic Foundation and National Heart, Lung, and Blood Institute Grant HL-072795.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society