Characterization of the Recombinant Rat 175-kDa Hyaluronan Receptor for Endocytosis (HARE)
2003; Elsevier BV; Volume: 278; Issue: 44 Linguagem: Inglês
10.1074/jbc.m307201200
ISSN1083-351X
AutoresJanet A. Weigel, Paul H. Weigel,
Tópico(s)Cell Adhesion Molecules Research
ResumoHyaluronan (HA) and chondroitin sulfate (CS) clearance from lymph and blood in mammals is mediated by the HA receptor for endocytosis (HARE), which is present as two isoforms in rat and human (175/300 kDa and 190/315 kDa, respectively) in the sinusoidal endothelial cells of liver, spleen, and lymph nodes (Zhou, B., McGary, C. T., Weigel, J. A., Saxena, A., and Weigel, P. H. (2003) Glycobiology 13, 339–349). The small rat and human HARE proteins are not encoded directly by mRNA but are derived from larger precursors. Here we characterize the specificity and function of the 175-kDa HARE, expressed in the absence of the 300-kDa species, in stably transfected SK-Hep-1 cells. The HARE cDNA was fused with a leader sequence to allow correct orientation of the membrane protein. The recombinant rHARE contained ∼25 kDa of N-linked oligosaccharides and, like the native protein, was able to bind HA in a ligand blot assay, even after de-N-glycosylation. SK-HARE cell lines demonstrated specific 125I-HA endocytosis, receptor recycling, and delivery of HA to lysosomes for degradation. The Kd for the binding of HA (number-average molecular mass ∼ 133 kDa) to the 175-kDa HARE at 4 °C was 4.1 nm with 160,000 to 220,000 HA-binding sites per cell. The 175-kDa rHARE binds HA, dermatan sulfate, and chondroitin sulfates A, C, D, and E, but not chondroitin, heparin, heparan sulfate, or keratan sulfate. Surprisingly, recognition of glycosaminoglycans (GAGs) other than HA by native or recombinant HARE was temperature-dependent. Although competition was observed at 37 °C, none of the other GAGs competed for 125I-HA binding to SK-HARE cells at 4 °C. Anti-HARE monoclonal antibody-174 showed a similar temperature-dependence in its ability to block HA endocytosis. These data suggest that temperature-induced conformational changes may alter the GAG specificity of HARE. The results confirm that the 175-kDa rHARE does not require the larger HARE isoform to mediate endocytosis of multiple GAGs. Hyaluronan (HA) and chondroitin sulfate (CS) clearance from lymph and blood in mammals is mediated by the HA receptor for endocytosis (HARE), which is present as two isoforms in rat and human (175/300 kDa and 190/315 kDa, respectively) in the sinusoidal endothelial cells of liver, spleen, and lymph nodes (Zhou, B., McGary, C. T., Weigel, J. A., Saxena, A., and Weigel, P. H. (2003) Glycobiology 13, 339–349). The small rat and human HARE proteins are not encoded directly by mRNA but are derived from larger precursors. Here we characterize the specificity and function of the 175-kDa HARE, expressed in the absence of the 300-kDa species, in stably transfected SK-Hep-1 cells. The HARE cDNA was fused with a leader sequence to allow correct orientation of the membrane protein. The recombinant rHARE contained ∼25 kDa of N-linked oligosaccharides and, like the native protein, was able to bind HA in a ligand blot assay, even after de-N-glycosylation. SK-HARE cell lines demonstrated specific 125I-HA endocytosis, receptor recycling, and delivery of HA to lysosomes for degradation. The Kd for the binding of HA (number-average molecular mass ∼ 133 kDa) to the 175-kDa HARE at 4 °C was 4.1 nm with 160,000 to 220,000 HA-binding sites per cell. The 175-kDa rHARE binds HA, dermatan sulfate, and chondroitin sulfates A, C, D, and E, but not chondroitin, heparin, heparan sulfate, or keratan sulfate. Surprisingly, recognition of glycosaminoglycans (GAGs) other than HA by native or recombinant HARE was temperature-dependent. Although competition was observed at 37 °C, none of the other GAGs competed for 125I-HA binding to SK-HARE cells at 4 °C. Anti-HARE monoclonal antibody-174 showed a similar temperature-dependence in its ability to block HA endocytosis. These data suggest that temperature-induced conformational changes may alter the GAG specificity of HARE. The results confirm that the 175-kDa rHARE does not require the larger HARE isoform to mediate endocytosis of multiple GAGs. In a typical 70-kg adult, the total HA 1The abbreviations used are: HA, hyaluronic acid, hyaluronate, hyaluronan; CS, chondroitin sulfate; CS-A, chondroitin 4-sulfate; CS-C, chondroitin 6-sulfate; CS-D, chondroitin 2,6-sulfate; CS-E, chondroitin 4,6-sulfate; DS, dermatan sulfate; ECM, extracellular matrix; GAG, glycosaminoglycan; HARE, HA receptor for endocytosis; HBSS, Hanks' balanced salt solution; Hep, heparin; HS, heparan sulfate; KS, keratan sulfate; LECs, liver sinusoidal endothelial cells; mAb, monoclonal antibody; PBS, phosphate-buffered saline; PG, proteoglycan; SK-HARE; stable SK-Hep-1 cell lines expressing HARE; TBS, Tris-buffered saline; TBST, Tris-buffered saline containing 0.05% Tween 20. content is ∼15 g, and about one-third of this turns over every day (1Laurent T.C. Fraser J.R.E. Henriksen J.H. Degradation of Bioactive Substances: Physiology and Pathophysiology. CRC Press, Boca Raton, FL1991: 249-265Google Scholar). HA is a structural component in all vertebrate tissue matrices and plays key roles in cell proliferation and adhesion, morphogenesis, differentiation, inflammation, and wound healing (2Abatangelo G. Weigel P.H. (eds) New Frontiers in Medical Sciences: Redefining Hyaluronan. Elsevier Science B.V., Amsterdam2000Google Scholar, 3Toole B.P. J. Intern. Med. 1997; 242: 35-40Crossref PubMed Scopus (282) Google Scholar, 4Knudson C.B. Knudson W. FASEB J. 1993; 7: 1233-1241Crossref PubMed Scopus (601) Google Scholar, 5Laurent T.C. Fraser J.R.E. FASEB J. 1992; 6: 2397-2404Crossref PubMed Scopus (2085) Google Scholar, 6Evered D. Whelan J. CIBA Found. Symp. 1989; 143: 1-288Google Scholar). HA and other GAGs, in particular the various CSs, are synthesized and degraded continuously in tissues throughout the body. For example, the HA in skin has a metabolic half-life of only ∼1.5 days (7Tammi R. Saamanen A.M. Maibach H.I. Tammi M. J. Invest. Dermatol. 1991; 97: 126-130Abstract Full Text PDF PubMed Google Scholar). Some turnover of extracellular CS likely occurs in a coordinated way and by a similar mechanism with HA turnover, because some CS chains are covalently attached to PGs, which in turn are bound to HA chains released from the ECM. Coordinated turnover of HA and some of the body's large CS pool makes sense physiologically, because HA, CS, and other GAGs in the ECM could be released simultaneously after the cleavage of HA. CS could also be released from different ECMs by other mechanisms, such as regulated proteolysis of various aggregating or nonaggregating PGs. In this case, the final turnover (i.e. uptake and degradation) of such fragments that enter the lymph or blood could be mediated either by HARE or other CS-recognizing receptors or by receptors specific for the PG protein. The present model for the high turnover of HA and CS in ECMs throughout the mammalian body is that very large native HA molecules (approaching 107 Da) are partially digested to produce large HA fragments (∼106 Da) that are then released from the matrix (8Fraser J.R. Appelgren L.E. Laurent T.C. Cell Tissue Res. 1983; 233: 285-293Crossref PubMed Scopus (106) Google Scholar, 9Fraser J.R. Kimpton W.G. Laurent T.C. Cahill R.N. Vakakis N. Biochem. J. 1988; 256: 153-158Crossref PubMed Scopus (160) Google Scholar, 10Lebel L. Smith L. Risberg B. Gerdin B. Laurent T.C. J. Appl. Physiol. 1988; 64: 1327-1332Crossref PubMed Scopus (44) Google Scholar, 11Tzaicos C. Fraser J.R. Tsotsis E. Kimpton W.G. Biochem. J. 1989; 264: 823-828Crossref PubMed Scopus (18) Google Scholar, 12Laurent U.B. Dahl L.B. Reed R.K. Exp. Physiol. 1991; 76: 695-703Crossref PubMed Scopus (81) Google Scholar). These released HA fragments could still be bound to Link proteins and aggregating PGs (i.e. aggrecan, versican, neurocan, and brevican), which can contain covalently attached CS chains; all these components, therefore, would be released concurrently from ECM networks. These HA-PG fragments would also likely contain other proteins associated with particular PGs, such as growth factors. The released ECM fragments then enter lymphatic vessels and flow to regional lymph nodes, which are initial sites for the clearance and degradation of the HA and CS. Lymph nodes are the primary clearance sites, accounting for ∼85% of the HA turnover. The second clearance site is the liver, which accounts for ∼15% of the total body HA, and presumably CS, turnover. The HA/CS clearance and degradation in liver and lymph nodes is mediated by the same endocytic receptor (now designated HARE), which is expressed in the sinusoidal endothelial cells of these tissues (13Eriksson S. Fraser J.R. Laurent T.C. Pertoft H. Smedsrod B. Exp. Cell Res. 1983; 144: 223-228Crossref PubMed Scopus (261) Google Scholar, 14McGary C.T. Raja R.H. Weigel P.H. Biochem. J. 1989; 257: 875-884Crossref PubMed Scopus (125) Google Scholar). Although humans turn over ∼5 g of HA per day, the two HA/CS clearance systems utilizing HARE in lymph node and liver maintain a very low steady-state concentration of HA in blood (i.e. 10–100 ng/ml). Presumably, the removal of HA from lymph fluid and blood is very important for normal health. In particular, if the concentration of high molecular mass HA increased, then the viscosity of these latter fluids could increase to dangerous levels. For example, the passage of blood cells through narrow microcapillary beds would be impaired if blood viscosity increased. Coagulation homeostasis could also be perturbed, because HA specifically binds to human fibrinogen (15LeBoeuf R.D. Raja R.H. Fuller G.M. Weigel P.H. J. Biol. Chem. 1986; 261: 12586-12592Abstract Full Text PDF PubMed Google Scholar) and stimulates fibrin clot formation (16LeBoeuf R.D. Gregg R.R. Weigel P.H. Fuller G.M. Biochemistry. 1987; 26: 6052-6057Crossref PubMed Scopus (45) Google Scholar). Elevated levels of serum HA occur in several diseases, including some cancers (17Thylen A. Wallin J. Martensson G. Cancer. 1999; 86: 2000-2005Crossref PubMed Scopus (43) Google Scholar), psoriasis (18Lundin A. Engstrom-Laurent A. Hallgren R. Michaelsson G. Br. J. Dermatol. 1985; 112: 663-671Crossref PubMed Scopus (51) Google Scholar), scleroderma (19Freitas J.P. Filipe P. Emerit I. Meunier P. Manso C.F. Guerra Rodrigo F. Dermatology. 1996; 192: 46-49Crossref PubMed Scopus (27) Google Scholar), rheumatoid arthritis (20Manicourt D.H. Poilvache P. Nzeusseu A. van Egeren A. Devogelaer J.P. Lenz M.E. Thonar E.J. Arthritis Rheum. 1999; 42: 1861-1869Crossref PubMed Scopus (51) Google Scholar), and liver cirrhosis (21Yamada M. Fukuda Y. Nakano I. Katano Y. Takamatsu J. Hayakawa T. Acta Haematol. 1998; 99: 212-216Crossref PubMed Scopus (26) Google Scholar, 22Lai K.N. Szetp C.C. Lam C.W.K. Lai K.B. Wong T.Y.H. Leung J.C.K. J. Lab Clin. Med. 1998; 131: 354-359Abstract Full Text PDF PubMed Scopus (13) Google Scholar). Hepatic clearance of HA is such an important function that elevated serum HA is often used as a diagnostic tool to detect or monitor liver failure (23Bramley P.N. Rathbone B.J. Forbes M.A. Cooper E.H. Losowsky M.S. J. Hepatol. 1991; 13: 8-13Abstract Full Text PDF PubMed Scopus (43) Google Scholar). In previous studies, we used a specific mAb and a ligand blot assay to purify two membrane-bound HA-binding proteins from rat LECs (24Zhou B. Oka J.A. Weigel P.H. J. Biol. Chem. 1999; 274: 33831-33834Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar) and human spleen (25Zhou B. McGary C.T. Weigel J.A. Saxena A. Weigel P.H. Glycobiology. 2003; 13: 339-349Crossref PubMed Scopus (48) Google Scholar). The rHARE isoforms are 175 and 300 kDa and the hHARE isoforms are 190 and ∼315 kDa. These HARE isoforms are expressed in the liver sinusoids, the venous sinuses of the red pulp in spleen, and the medullary sinuses of lymph nodes (25Zhou B. McGary C.T. Weigel J.A. Saxena A. Weigel P.H. Glycobiology. 2003; 13: 339-349Crossref PubMed Scopus (48) Google Scholar, 26Zhou B. Weigel J.A. Fauss L. Weigel P.H. J. Biol. Chem. 2000; 275: 37733-37741Abstract Full Text Full Text PDF PubMed Scopus (231) Google Scholar). The small HARE isoforms contain a single subunit, whereas the large isoforms contain multiple disulfide-bonded subunits, e.g. the ∼315-kDa hHARE has two subunits (250 and 220 kDa) in a ratio of ∼3:1 (25Zhou B. McGary C.T. Weigel J.A. Saxena A. Weigel P.H. Glycobiology. 2003; 13: 339-349Crossref PubMed Scopus (48) Google Scholar). Although they are different sizes, all of the subunits in both HARE isoforms appear to be derived from the same large (2551-amino acid) precursor protein, which has also been called Stabilin 2 (27Politz O. Gratchev A. McCourt P.A. Schledzewski K. Guillot P. Johansson S. Svineng G. Franke P. Kannicht C. Kzhyshkowska J. Longati P. Velten F.W. Johansson S. Goerdt S. Biochem. J. 2002; 362: 155-164Crossref PubMed Scopus (247) Google Scholar). When expressed in SK-Hep-1 cells in the absence of the large isoform, the small rHARE isoform co-localized with clathrin as expected for a coated pit-coupled endocytic HA receptor (28Zhou B. Weigel J.A. Saxena A. Weigel P.H. Mol. Biol. Cell. 2002; 13: 2853-2868Crossref PubMed Scopus (66) Google Scholar). The two HARE species, therefore, appear to be functionally independent isoreceptors for HA. In this study, we functionally characterized the 175-kDa rHARE in stable cell lines. Previous cellular studies have used rat LECs, which express both the 175- and ∼300-kDa HARE isoforms. This study is the first examination of the ligand specificity and endocytic activity of a single HARE isoform in the absence of the other. Materials and Buffers—Na125I was from Amersham Pharmacia Corp. 125I-HA was prepared as described previously (29Raja R.H. LeBoeuf R. Stone G. Weigel P.H. Anal. Biochem. 1984; 139: 168-177Crossref PubMed Scopus (66) Google Scholar) using a hexylamine derivative of HA oligosaccharides (number-average molecular mass = 133,000 based on gel permeation chromatography coupled to multiangle light scattering analysis), modified only at the reducing ends. Male Sprague-Dawley rats (200 g) were from Charles River Laboratories. BSA Fraction V was from Intergen Co. Collagenase was from Roche Applied Science. The preparation and characterization of mouse mAbs against the rat HARE were described previously (26Zhou B. Weigel J.A. Fauss L. Weigel P.H. J. Biol. Chem. 2000; 275: 37733-37741Abstract Full Text Full Text PDF PubMed Scopus (231) Google Scholar). Tris, SDS, ammonium persulfate, N,N′-methylenebisacrylamide, and SDS-PAGE standards were from Bio-Rad. Digitonin was from ACROS Organics. Unless noted otherwise, other chemicals and reagents were from Sigma Chemical Co. All GAGs (with the exception of heparin, which came from Sigma) were obtained from Seikagaku Corp. Nitrocellulose membranes were from Schleicher & Schuell. Goat anti-mouse (polyvalent) IgG-alkaline phosphatase conjugate was from Sigma. HBSS and PBS were formulated according to the Invitrogen catalog formulations. Medium 1 is Eagle's Basal Medium (Invitrogen) supplemented with 100 mg/liter succinic acid sodium salt, 75 mg/liter succinic acid, 2.4 g/liter HEPES, and 0.22 g/liter NaHCO3. Medium 1/BSA is Medium 1 supplemented with 0.1% BSA (w/v). TBS contains 20 mm Tris-HCl, pH 7.0, and 150 mm NaCl. Preparation of LECs—Rat LECs were prepared by a collagenase liver perfusion procedure and purified by differential and Percoll gradient centrifugation as previously described (13Eriksson S. Fraser J.R. Laurent T.C. Pertoft H. Smedsrod B. Exp. Cell Res. 1983; 144: 223-228Crossref PubMed Scopus (261) Google Scholar, 24Zhou B. Oka J.A. Weigel P.H. J. Biol. Chem. 1999; 274: 33831-33834Abstract Full Text Full Text PDF PubMed Scopus (59) Google Scholar). Contaminating Kupffer cells were allowed to adhere to a glass Petri dish, and nonadherent LECs were then collected, cultured on fibronectin-coated 24-well plates and used the same day. Selection of Stable Transfectants Expressing the 175-kDa HARE— SK-Hep-1 cells (from American Type Culture Collection, Manassas, VA) were transfected with purified p175HARE-κ DNA using FuGENE 6 and subjected to selection using G418 as described by Zhou et al. (28Zhou B. Weigel J.A. Saxena A. Weigel P.H. Mol. Biol. Cell. 2002; 13: 2853-2868Crossref PubMed Scopus (66) Google Scholar). Because the protein is not directly encoded by an mRNA species, the 4708-bp cDNA sequence in this construct contains, at the 5′-end, the membrane insertion (leader) sequence of the mouse immunoglobulin κ chain, so that the resulting protein is inserted in the plasma membrane in the correct orientation. The 46-amino acid κ chain sequence was derived from pSecTag2, and the final construct was assembled in pcDNA3.1 (both from Invitrogen). Cloning rings were used to isolate individual colonies of antibiotic-resistant transfected cells after 2–3 weeks. Cells were detached by treatment with 0.05% trypsin and 0.53 mm EDTA for 5 min at room temperature, collected, and grown in 12-well plates to assess HARE protein expression and function by enzyme-linked immunosorbent assay, Western blot, and 125I-HA binding assays. Positive cultures were further purified by dilution cloning, and the final cloned cell lines were designated by numbers, e.g. SK-HARE-36. Western and Ligand Blot Assays—Cell lysates were mixed with equal volumes of a 2× SDS sample buffer (30Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207478) Google Scholar), without reducing agent, to give final concentrations of 16 mm Tris-HCl, pH 6.8, 2% (w/v) SDS, 5% glycerol (v/v), and 0.01% bromphenol blue. After SDS-PAGE, the contents of the gel were electrotransferred to a nitrocellulose membrane overnight at 10 V at 4 °C using 25 mm Tris, pH 8.3, 192 mm glycine, 20% methanol, and 0.01% SDS in a Genie blotter apparatus from Idea Scientific. For the ligand blot assay, the nitrocellulose membrane was treated first with TBS containing 0.1% Tween 20 at 4 °C for 2 h, or TBST (Tris-buffered saline containing 0.05% Tween 20) overnight, and then incubated with 1–2 μg/ml 125I-HA in 150 mm NaCl, 10 mm HEPES, pH 7.4, and 5 mm EDTA without, or with, a 100- to 150-fold excess of nonlabeled HA (as competitor) to assess total and nonspecific binding, respectively (31Yannariello-Brown J. Zhou B. Ritchie D. Oka J.A. Weigel P.H. Biochem. Biophys. Res. Commun. 1996; 218: 314-319Crossref PubMed Scopus (20) Google Scholar). The nitrocellulose membrane was washed five times for 5 min each with TBST and dried at room temperature. The bound 125I-HA was detected by autoradiography using Kodak BioMax MS film exposed at –85 °C for 6–48 h. Nonspecific binding in this assay is typically <5%. For Western blot assays, the nitrocellulose membranes were blocked with 1% BSA in TBS at 4 °C overnight either after the ligand blot assay (the membranes were rewet with TBST first) or directly after SDS-PAGE and electrotransfer. The membrane was then incubated with anti-rat HARE mAbs (e.g. 1 μg/ml IgG) at 22 °C for 1 h, washed three times for 5 min each with TBST, and incubated with goat anti-mouse IgG-alkaline phosphatase conjugate (1:1500 dilution) for 1 h at room temperature. The nitrocellulose was washed five times for 5 min each with TBST and incubated with p-nitro blue tetrazolium and sodium 5-bromo-4-chloro-3-indolyl phosphate p-toluidine for color development (Bio-Rad), which was stopped by washing the membrane with distilled water. 125I-HA Binding or Endocytosis—Stably transfected SK-HARE cell lines were grown to confluence in Dulbecco's modified Eagle's medium with 10% fetal calf serum containing 0.4 mg/ml G418 in tissue culture multi-well dishes (usually 24-well plates). The cells were washed, incubated at 37 °C in fresh medium without serum for 1 h, the plates were then placed on ice, and the cells washed two times with HBSS prior to all experiments. Medium containing 1–2 μg/ml 125I-HA with or without the noted concentration of IgG or other GAG (as competitor) was added to each well, and the cells were incubated either on ice for 60 min to assess cell surface binding or at 37 °C to allow internalization of ligand. At the noted times the medium was aspirated, the cells were washed three times with HBSS and lysed in 0.3 n NaOH, and radioactivity and protein content were determined. All values were normalized for cell protein per well and are presented as cpm/μg of protein. Degradation of 125I-HA—Degradation of 125I-HA was measured by a cetylpyridinium chloride precipitation assay as described by McGary et al. (32McGary C.T. Yannariello-Brown J. Kim D.W. Stinson T.C. Weigel P.H. Hepatology. 1993; 18: 1465-1476Crossref PubMed Scopus (18) Google Scholar). 50-μl samples of medium were mixed with 250 μl of 1 mg/ml HA in 1.5-ml microcentrifuge tubes. Alternatively, 100-μl samples of cell lysate (in 0.3 n NaOH) were mixed with 47 μl of 0.6 n HCl, 28 μl of distilled water, and 125 μl of 2 mg/ml HA. After mixing, 300 μl of 6% (w/v) cetylpyridinium chloride in distilled water was added, and the tubes were mixed by vortexing. After 10 min at room temperature, the samples were centrifuged at 9000 rpm in an Eppendorf model 5417 microcentrifuge, using a swinging bucket rotor, at 22 °C for 5 min. A sample (300 μl) of the supernatant was taken for determination of radioactivity, and the remainder was removed by aspiration. The tip of the tube containing the precipitated pellet was cut off and put into a gamma counter tube, and radioactivity was determined. Degradation was measured as the time-dependent increase of nonprecipitable radioactivity. About 80% of the total radioactivity was precipitable at the beginning of each experiment. HA fragments that are smaller than ∼50 monomers do not precipitate with cetylpyridinium chloride (32McGary C.T. Yannariello-Brown J. Kim D.W. Stinson T.C. Weigel P.H. Hepatology. 1993; 18: 1465-1476Crossref PubMed Scopus (18) Google Scholar), and because radioautolysis continuously generates smaller fragments, the background in this cetylpyridinium chloride precipitation assay increases with age of the radioiodinated HA. General—Protein content was determined by the method of Bradford (33Bradford M.M. Anal. Biochem. 1976; 72: 248-254Crossref PubMed Scopus (217389) Google Scholar) using BSA as a standard. SDS-PAGE was performed according to the method of Laemmli (30Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207478) Google Scholar). Western blotting was performed as described by Burnette (34Burnette W.N. Anal. Biochem. 1981; 112: 195-203Crossref PubMed Scopus (5926) Google Scholar) with minor modifications (26Zhou B. Weigel J.A. Fauss L. Weigel P.H. J. Biol. Chem. 2000; 275: 37733-37741Abstract Full Text Full Text PDF PubMed Scopus (231) Google Scholar). 125I radioactivity was measured using a Packard Auto-Gamma Counting system. Digital images were captured using an Alpha Innotech Fluorochem 8000 or a Molecular Dynamics Personal Model densitometer. Images were taken in Corel Photo Paint (version 9.0) as JPG files, cropped and processed identically and then transferred to Corel Draw (version 9.0) for annotation. N-terminal amino acid sequence analysis was performed by the University of Oklahoma Health Sciences Center Molecular Biology Resource facility. Expression of the 175-kDa rHARE Protein in Stable Cell Lines—The 175-kDa rHARE protein is not directly encoded by a distinct mRNA, but rather is generated by proteolysis from a larger precursor (28Zhou B. Weigel J.A. Saxena A. Weigel P.H. Mol. Biol. Cell. 2002; 13: 2853-2868Crossref PubMed Scopus (66) Google Scholar). Therefore, to create stable cell lines expressing only this HARE isoform, we used a synthetic cDNA coding for the 1431-amino acid protein fused at the N terminus to the leader sequence of the mouse immunoglobulin κ light chain. This leader sequence serves as a membrane insertion signal that allows correct orientation of the protein and trafficking to the cell surface. This vector was used to transfect SK-Hep-1 cells, after which multiple stable cell lines expressing HARE were cloned using antibiotic selection. The SK-Hep-1 cell line has been used by us and by others for similar studies, and it does not display specific 125I-HA binding or endocytosis activity. Additionally, it has no endogenous surface HA receptors and no cross-reactivity with the anti-HARE mAbs. Seven independent SK-HARE clones were obtained, all of which had similar characteristics with respect to 175-kDa rHARE expression and function. Based on Western analyses, all cell lines expressed comparable levels of HARE protein and showed similar HA-binding activity in ligand blots (Fig. 1). Each of the previously described anti-175-kDa rHARE mAbs that recognize the native nonreduced protein (26Zhou B. Weigel J.A. Fauss L. Weigel P.H. J. Biol. Chem. 2000; 275: 37733-37741Abstract Full Text Full Text PDF PubMed Scopus (231) Google Scholar) also individually recognizes the recombinant rHARE in Western blots (not shown). Untransfected cells and SK-Hep-1 cells, transfected with the same vector containing an unrelated cDNA insert, displayed only a low level of nonspecific 125I-HA uptake at 37 °C and showed no bands in similar Western or ligand blots (not shown). In experiments examining multiple cell lines, the molecular mass of the recombinant 175-kDa HARE was 182 ± 3 kDa (n = 10), compared with 180 ± 4 kDa (n = 6) for the native rat LEC protein. Two bands were apparent in each SK-HARE clone (Fig. 1), a major larger protein and a minor smaller protein. Amino acid sequence analysis revealed multiple amino acids released at each cycle, indicating that neither protein contains a unique N terminus, but rather is a mixture of proteins with many different N termini. The sequence data for the smaller minor HARE band were consistent with at least two proteins, one starting at Asp80 (DL) and another at Gly121 (GVI). The latter HARE variant corresponds well to a minor sequence found in the native rat 175-kDa HARE that begins at Val122 (28Zhou B. Weigel J.A. Saxena A. Weigel P.H. Mol. Biol. Cell. 2002; 13: 2853-2868Crossref PubMed Scopus (66) Google Scholar). The sequence data for the major recombinant 175-kDa HARE band were consistent with cleavage of the signal peptide, followed by variable proteolysis within a region of ∼40 amino acids starting from this cleavage site at Ala–25 (AAQPA) to Ser16 (SIFR). The recombinant rHARE, therefore, is processed by SK-Hep-1 cells and LECs in a similar manner to yield a variety of proteolytic cleavage products. The native and recombinant rHARE proteins contained 20–25 kDa of N-linked oligosaccharides that could be released by treatment with endoglycosidase-F (Fig. 2). The size of the apparent HARE core protein from SK-HARE cells was ∼155 kDa; this is consistent with the predicted mass of the protein derived from the synthetic cDNA (minus the 21-residue signal peptide) and is essentially identical in size to the deglycosylated native rat liver protein. As with the native HARE, the deglycosylated recombinant 175-kDa HARE retained its HA-binding activity. Glycosylation of the protein is not required for its ability to bind ligand (Fig. 2, left panel). The binding of 125I-HA to SK-HARE cells at 4 °C showed a typical hyperbolic increase that plateaued after 1 h (Fig. 3A). 125I-HA binding curves in the presence of increasing concentrations of unlabeled HA (Fig. 3B) were essentially identical among the SK-HARE cell lines. Based on the midpoint of these binding curves from multiple experiments with several SK-HARE clones, the apparent Kd for HA binding was 4.3 ± 1.1 nm (n = 6). Two independent equilibrium binding experiments (as in Fig. 3C), each using SK-HARE cell lines #26 and #36, were performed and analyzed by the method of Scatchard (35Scatchard G. Ann. N. Y. Acad. Sci. 1949; 61: 660-672Crossref Scopus (17809) Google Scholar). In each experiment, we observed a single class of binding sites with identical Kd values of 4.1 nm, as calculated by first order linear regression analyses (cc ≤ –0.95). B max values ranged from 160,000 to 220,000 HA-binding sites/cell, which is similar to the number of binding sites observed previously on rat LECs (36Raja R.H. McGary C.T. Weigel P.H. J. Biol. Chem. 1988; 263: 16661-16668Abstract Full Text PDF PubMed Google Scholar). Endocytosis and Degradation of HA Mediated by the 175-kDa rHARE—All stable cell lines mediated the specific and continuous endocytosis of 125I-HA for many hours at 37 °C (Fig. 4). Specificity of HA endocytosis ranged from 67% to 79%, as assessed in the presence of excess unlabeled HA, and the cells internalized an average of 4.2-times their number of surface HA receptors within 3 h (Table I). Western analysis confirmed that HARE protein content was not decreased during this period, indicating that HARE was not delivered to lysosomes for degradation following internalization of HA-HARE complexes (not shown). The constant cellular HARE content despite the high ratio of HA molecules taken up per HARE indicates that the recombinant 175-kDa HARE is a recycling endocytic receptor, as expected based on earlier studies with rat LECs (14McGary C.T. Raja R.H. Weigel P.H. Biochem. J. 1989; 257: 875-884Crossref PubMed Scopus (125) Google Scholar). Endocytosis of 125I-HA mediated by HARE in these stable cell lines led to degradation of the internalized ligand (Fig. 4). Degradation products first accumulated intracellularly (presumably in lysosomes) and then, after a long lag time, appeared in the medium. We previously found in rat LECs (32McGary C.T. Yannariello-Brown J. Kim D.W. Stinson T.C. Weigel P.H. Hepatology. 1993; 18: 1465-1476Crossref PubMed Scopus (18) Google Scholar) that the HA-hexylamine derivative, which is modified at the reducing end to contain a hydroxyphenol group, behaves as a residualizing label in that degradation products first reach high levels in lysosomes before appearing in the medium (37Strobel J.L. Baynes J.W. Thorpe S.R. Arch. Biochem. Biophys. 1985; 240: 635-645Crossref PubMed Scopus (46) Google Scholar). A
Referência(s)