Transport and Activation of the Vacuolar Aspartic Proteinase Phytepsin in Barley (Hordeum vulgare L.)
1998; Elsevier BV; Volume: 273; Issue: 47 Linguagem: Inglês
10.1074/jbc.273.47.31230
ISSN1083-351X
AutoresStefanie Glathe, Jukka Kervinen, Manfred Nimtz, Grace H. Li, Gregory J. Tobin, Terry D. Copeland, David A. Ashford, Alexander Wlodawer, Júlia Costa,
Tópico(s)Biochemical and Structural Characterization
ResumoThe primary translation product of barley aspartic proteinase, phytepsin (EC 3.4.23.40), consists of a signal sequence, a propart, and mature enzyme forms. Here, we describe post-translational processing and activation of phytepsin during its transport to the vacuole in roots, as detected by using metabolic labeling and immunoprecipitation. After removal of the signal sequence, the glycosylated precursor of 53 kDa (P53) was produced and further processed to polypeptides of 31 and 15 kDa (P31 + P15) and, subsequently, to polypeptides of 26 and 9 kDa (P26 + P9), 45 min and 24 h after synthesis, respectively. The processing occurred in a late-Golgi compartment or post-Golgi compartment, because brefeldin A inhibited the processing, and P53 acquired partial endoglycosidase H resistance 30 min after synthesis, whereas P15 was completely resistant. The N-glycosylation inhibitor tunicamycin had no effect on transport, but the absence of glycans on P53 accelerated the proteolytic processing. Phytepsin was also expressed in baculovirus-infected insect cells. The recombinant prophytepsin underwent autoproteolytic activation in vitro and showed enzymatic properties similar to the enzyme purified from grains. However, a comparison of the in vitro/in vivoprocessing sites revealed slight differences, indicating that additional proteases are needed for the completion of the maturationin vivo. The primary translation product of barley aspartic proteinase, phytepsin (EC 3.4.23.40), consists of a signal sequence, a propart, and mature enzyme forms. Here, we describe post-translational processing and activation of phytepsin during its transport to the vacuole in roots, as detected by using metabolic labeling and immunoprecipitation. After removal of the signal sequence, the glycosylated precursor of 53 kDa (P53) was produced and further processed to polypeptides of 31 and 15 kDa (P31 + P15) and, subsequently, to polypeptides of 26 and 9 kDa (P26 + P9), 45 min and 24 h after synthesis, respectively. The processing occurred in a late-Golgi compartment or post-Golgi compartment, because brefeldin A inhibited the processing, and P53 acquired partial endoglycosidase H resistance 30 min after synthesis, whereas P15 was completely resistant. The N-glycosylation inhibitor tunicamycin had no effect on transport, but the absence of glycans on P53 accelerated the proteolytic processing. Phytepsin was also expressed in baculovirus-infected insect cells. The recombinant prophytepsin underwent autoproteolytic activation in vitro and showed enzymatic properties similar to the enzyme purified from grains. However, a comparison of the in vitro/in vivoprocessing sites revealed slight differences, indicating that additional proteases are needed for the completion of the maturationin vivo. aspartic proteinase endoglycosidase H polyacrylamide gel electrophoresis polyvinylidene difluoride matrix-assisted laser desorption ionization time-of-flight mass spectroscopy recombinant. Aspartic proteinases (APs)1 (EC 3.4.23) constitute one of the four superfamilies of proteolytic enzymes. They are present in a wide variety of organisms, such as viruses, fungi, plants, and animals. Common features of APs include an active site cleft that contains two catalytic aspartic acid residues (32 and 215 in pepsin), acidic pH optima for enzymatic activity, inhibition by pepstatin A, a conserved overall fold, and a preferential cleavage specificity for peptide bonds between amino acid residues with bulky hydrophobic side chains. Both intracellular and extracellular forms of APs are present in animal tissues (1Davies D. Annu. Rev. Biophys. Biophys. Chem. 1990; 19: 189-215Crossref PubMed Scopus (573) Google Scholar, 2Rawlings N.D. Barrett A.L. Methods Enzymol. 1995; 248: 105-120Crossref PubMed Scopus (140) Google Scholar). Aspartic proteinases are synthesized as inactive precursors (zymogens) in which the N-terminal propeptide is bound to the active site cleft, thus preventing undesirable protein degradation and enabling spatial and temporal regulation of proteolytic activity. Pepsinogen, the inactive precursor of stomach pepsin, needs only a drop in pH for the autocatalytic cleavage of the propeptide to result in an active enzyme (1Davies D. Annu. Rev. Biophys. Biophys. Chem. 1990; 19: 189-215Crossref PubMed Scopus (573) Google Scholar). Procathepsin D, which is targeted to the lysosome largely via the mannose-6-phosphate receptor (3Metcalf P. Fusek M. EMBO J. 1993; 12: 1293-1302Crossref PubMed Scopus (133) Google Scholar), is activated after cleavage of its N-terminal 44 amino acids, most likely by lysosomal cysteine proteinases (4Samarel A.M. Ferguson A.G. Decker R.S. Lesch M. Am. J. Physiol. 1989; 257: C1069-C1079Crossref PubMed Google Scholar). Procathepsin D is also capable of acid-dependent autoactivation in vitro to yield a catalytically active (pseudo)cathepsin D (5Conner G.E. Richo G. Biochemistry. 1992; 31: 1142-1147Crossref PubMed Scopus (56) Google Scholar, 6Beyer B.M. Dunn B.M. J. Biol. Chem. 1996; 271: 15590-15596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). However, autocatalytic removal of the remaining 18-residue propeptide or the processing intermediate corresponding to pseudocathepsin D has not been observed in vivo (7Richo G.R. Conner G.E. J. Biol. Chem. 1994; 269: 14806-14812Abstract Full Text PDF PubMed Google Scholar). Barley AP (Hordeum vulgare AP), recently renamed phytepsin (EC 3.4.23.40) (8Barrett A.J. Eur. J. Biochem. 1997; 250: 1-6Crossref PubMed Scopus (38) Google Scholar), was originally isolated from grains in which it exists as two enzymatically active two-chain forms (9Sarkkinen P. Kalkkinen N. Tilgmann C. Siuro J. Kervinen J. Mikola L. Planta. 1992; 186: 317-323Crossref PubMed Scopus (104) Google Scholar). Sequence alignment of phytepsin with animal and microbial APs shows a high degree of similarity, with the exception of an inserted domain of approximately 100 residues that is plant-specific (10Runeberg-Roos P. Törmakangas K. Östman A. Eur. J. Biochem. 1991; 202: 1021-1027Crossref PubMed Scopus (104) Google Scholar, 11Cordeiro M.C. Xue Z.T. Pietrzak M. Pais M.S. Brodelius P.E. Plant Mol. Biol. 1994; 24: 733-741Crossref PubMed Scopus (74) Google Scholar, 12Asakura T. Watanabe H. Abe K. Arai S. Eur. J. Biochem. 1995; 232: 77-83Crossref PubMed Scopus (89) Google Scholar) and very similar to that of saposins (13Guruprasad K. Törmakangas K. Kervinen J. Blundell T.L. FEBS Lett. 1994; 352: 131-136Crossref PubMed Scopus (79) Google Scholar). The exact function of phytepsin is still controversial. Because phytepsin is an intracellular enzyme residing in leaf and root vacuoles (14Runeberg-Roos P. Kervinen J. Kovaleva V. Raikhel N.V. Gal S. Plant Physiol. 1994; 105: 321-329Crossref PubMed Scopus (96) Google Scholar) and in scutellar and aleuronal vacuole-like protein bodies in grains (15Marttila S. Jones B.L. Mikkonen A. Physiol. Plant. 1995; 93: 317-327Crossref Scopus (41) Google Scholar) and because it is able to cleave the C-terminal vacuolar targeting signal of barley prolectinin vitro (14Runeberg-Roos P. Kervinen J. Kovaleva V. Raikhel N.V. Gal S. Plant Physiol. 1994; 105: 321-329Crossref PubMed Scopus (96) Google Scholar), phytepsin may represent a cathepsin D-like enzyme from plant cells. Accordingly, phytepsin may participate in protein processing and metabolic turnover (Ref. 16Kervinen J. The Handbook of Proteolytic Enzymes.in: Barrett A.J. Woessner F.F. Rawlings N. Academic Press, Inc. London, 1998Google Scholar and references therein). It has recently also been observed that phytepsin may play a role in the active autolysis in plant tissues undergoing developmentally regulated programmed cell death (17Runeberg-Roos P. Saarma M. Plant J. 1998; 15: 139-145Crossref PubMed Scopus (63) Google Scholar). Modification of phytepsin during its intracellular route to vacuoles involves several steps; however, the enzymology, sequence, and intracellular localization of these events is not known. The study presented here describes the mode of expression, processing, and activation of phytepsin during its transport. Furthermore, we demonstrate the autoactivation of phytepsin in vitro by using a recombinant enzyme expressed in insect cells. H. vulgare (cv. Sereia) grains were purchased from the Estação Nacional de Melhoramento de Plantas, Elvas, Portugal. A purified antiserum against phytepsin was prepared as described previously (14Runeberg-Roos P. Kervinen J. Kovaleva V. Raikhel N.V. Gal S. Plant Physiol. 1994; 105: 321-329Crossref PubMed Scopus (96) Google Scholar).l-[35S]methionine andl-[35S]cysteine (Pro-mix™ in vivocell labeling mix) and autoradiography films (Hyperfilm™-MP) were purchased from Amersham Pharmacia Biotech. Molecular weight standards for gel electrophoresis were purchased from Bio-Rad. Protein A coupled to Sepharose was purchased from Amersham Pharmacia Biotech. Endoglycosidase H (Endo H) was from Oxford GlycoSciences (Oxford, UK). Immobilon™-P polyvinylidene difluoride (PVDF) transfer membranes were purchased from Millipore Corp. (Bedford, MA). The remaining reagents were of analytical grade. Barley grains were surface-sterilized with 1% (w/v) sodium hypochlorite followed by 60% (v/v) ethanol and germinated on Petri dishes containing 15 ml of 0.8% (w/v) agar for 3 days at 28 °C in the dark. For the extraction of root proteins, nondenaturing solubilization buffer (30 mm Tris, pH 7.5, 1 mm EDTA, 0.25 m sucrose, 5% (w/v) polyvinylpolypyrrolidone, 0.15% (v/v) β-mercaptoethanol) was added to the roots frozen in liquid nitrogen. Five root tips were homogenized with 0.2 ml of the buffer using a Teflon homogenizer (Sigma). The extracts were cleared by centrifugation for 5 min at 10,000 rpm, and the supernatants were used for immunoprecipitation or Western blot analysis. Protein A-Sepharose (3 mg/sample) was resuspended in immunoprecipitation buffer (50 mm Tris, pH 7.5, 150 mm NaCl, 0.1% (v/v) Triton X-100, 0.05% (w/v) sodium deoxycholate, 10% (v/v) glycerol, 1% (w/v) bovine serum albumin, 1 mm EDTA) and washed twice. Rabbit antibodies were directly coupled to the Protein A-Sepharose by shaking for 20 min at room temperature. The proteins in the lysate supernatants were immunoprecipitated with aliquots of the protein A-Sepharose-antibody-protein complex for 1 h on ice. Immunoprecipitates were washed twice each with high salt (50 mm Hepes, pH 7.5, 0.5 m NaCl, 5 mmEDTA, 0.2% (v/v) Triton X-100, 0.1% (w/v) SDS), medium salt (50 mm Hepes, pH 7.5, 0.15 m NaCl, 5 mmEDTA, 0.2% (v/v) Triton X-100), and low salt (10 mm Tris, pH 7.5, 0.1% (v/v) Triton X-100) buffers (18Pinkas-Kramarski R. Shelly M. Glathe S. Ratzkin B. Yarden Y. J. Biol. Chem. 1996; 271: 19029-19032Abstract Full Text Full Text PDF PubMed Scopus (130) Google Scholar). The complex was resuspended in 25 μl of SDS-polyacrylamide gel electrophoresis (PAGE) sample buffer, boiled for 5 min, and analyzed by SDS-PAGE (19Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207218) Google Scholar). The proteins were transferred from the gel to a PVDF membrane in a Bio-Rad semi-dry transfer cell for 1 h at 15 V. The PVDF membrane was blocked in Tris-buffered saline containing 0.05% (v/v) Tween 20 and 3% (w/v) dried lowfat milk for 45 min. The antiserum and the second antibody (alkaline phosphatase-coupled goat-anti-rabbit IgG) were used at 1/500 and 1/8000 dilutions, respectively, in blocking solution containing 1% dried lowfat milk. Detection was performed by the enhanced chemiluminescence (ECL) method (Amersham Pharmacia Biotech) or with 4-chloro-1-naphthol (20Costa J. Grabenhorst E. Nimtz M. Conradt H.S. J. Biol. Chem. 1997; 272: 11613-11621Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). For each time point, 5 roots from 3-day-germinated grains were cut ∼1 cm from the tip, and the cut ends were dipped in 20 μl of essential B5 plant growth medium containing 4% (w/v) sucrose and 100 μCi of Pro-mix™ (l-[35S]methionine/cysteine) for the indicated pulse periods. The chase was performed in 20 μl of essential B5 plant growth medium containing 4% (w/v) sucrose, 1 mml-methionine, and 0.5 mml-cysteine for the indicated time periods. Protein extracts were prepared from roots frozen in liquid nitrogen and were immunoprecipitated with anti-phytepsin serum. The proteins (∼40 μg) were separated by SDS-PAGE in a gradient gel (10–20%). The gel was dried, and 35S-labeled proteins were detected by fluorography. Metabolic labeling and pulse-chase experiments were performed as described above, except that for each time point, 10 root tips were incubated with 200 μCi of Pro-mix™. After immunoprecipitation, the protein was dissociated from the antibodies with 40 μl of 0.1 m glycine-HCl buffer, pH 3.0, then incubated in 50 mm citrate buffer, pH 5.5, containing 1 mg/ml SDS and 0.2 m β-mercaptoethanol at 95 °C for 10 min. Phenylmethylsulfonyl fluoride, E-64, and pepstatin A were added to final concentrations of 1 mm, 10 μg/ml, and 10 μg/ml, respectively. Six milliunits of Endo H were added to a final volume of 100 μl, and the mixture was incubated at 37 °C for 18 h. The protein was then precipitated with ethanol and analyzed by SDS-PAGE in 12 or 10–20% polyacrylamide gradient gels. Root tips (1 cm) were dipped for 1 h in 14 μl of essential B5 plant growth medium containing 4% (w/v) sucrose and inhibitors at the following concentrations: 7 μm brefeldin A1 in methanol and 10 μg/ml tunicamycin A1 in methanol. The control samples were incubated with the same volume of methanol added. The roots were radiolabeled as described above. Duplicate samples of 1 μl of metabolically labeled extract were spotted on small pieces of Whatman 3MM filter paper. The papers were dried, and one of the duplicates was washed 5 times with 5 ml of 5% (v/v) trichloroacetic acid and dried. The papers were placed at the bottom of plastic vials, 3 ml of liquid scintillation mixture (Beckman) was added, and the radioactivity was counted in the carbon channel of a Beckman scintillation counter. Sequences encoding the complete preprophytepsin (1863 base pairs (10Runeberg-Roos P. Törmakangas K. Östman A. Eur. J. Biochem. 1991; 202: 1021-1027Crossref PubMed Scopus (104) Google Scholar)) were cloned into the KpnI restriction site of the baculovirus transfer vector pBlueBac4.5 (Invitrogen Corp., Carlsbad, CA). The construct was cotransfected with Autographa californica multiple nuclear polyhedrosis viral DNA into Spodoptera frugiperda(Sf9) cells, and recombinant baculoviruses were derived using standard methodologies (21O'Reilly D.R. Miller L.K. Luckow V.A. Baculovirus Expression Vectors: A Laboratory Manual. Oxford University Press, New York1994Google Scholar). For protein production, Sf9 suspension cultures were infected with A. californicamultiple nuclear polyhedrosis virus-phytepsin in complete Grace's medium (about 600 ml) supplemented with 7% fetal bovine serum. Four days after infection, the conditioned culture supernatant was clarified for 20 min at 6,000 × g and concentrated to approximately 2% of the initial volume at 4 °C by ultrafiltration (Amicon YM30 membrane, Amicon, Danvers, MA). The concentrated supernatant was diluted 1.3-fold with cold 0.5 m sodium acetate, pH 4.0, and phytepsin was purified according to Sarkkinenet al. (9Sarkkinen P. Kalkkinen N. Tilgmann C. Siuro J. Kervinen J. Mikola L. Planta. 1992; 186: 317-323Crossref PubMed Scopus (104) Google Scholar) by affinity chromatography on a pepstatin-agarose column, with the exception that no washing with pH 7.5 buffer was carried out and the elution was performed with 0.1 mTris-HCl, pH 8.8, 0.2 mm dithiothreitol, 0.1 mNaCl. Further purification was performed by ion exchange chromatography on a Mono Q column (Amersham Pharmacia Biotech). Recombinant phytepsin purified on the Mono Q column (2.2 mg/ml in 20 mm Tris-HCl, pH 8.0, 0.1 m NaCl) was mixed with an equal volume of 0.2m incubation buffer and kept at 37 °C for up to 90 min. Incubation buffers were as follows: sodium lactate, pH 3.7, sodium acetate, pH 4.5 and 5.5, and sodium phosphate, pH 6.5. Samples were removed after various times and frozen at –70 °C. Processing products were separated by SDS-PAGE (PhastSystem, Amersham Pharmacia Biotech) and stained with Coomassie Brilliant Blue R-250 or electroblotted onto a PVDF membrane for N-terminal sequencing. The Coomassie Brilliant Blue-stained protein bands in the membrane were excised and individually sequenced on an Applied Biosystems 477A gas-phase sequencer, and the phenylthiohydantoin amino acids were identified on-line with a Model 120 analyzer (Applied Biosystems, Inc., Foster City, CA). Phytepsin samples were diluted with 20 mm sodium lactate, pH 3.7, and preincubated for 1 h at 37 °C, and the activity was measured according to Sarkkinen et al. (9Sarkkinen P. Kalkkinen N. Tilgmann C. Siuro J. Kervinen J. Mikola L. Planta. 1992; 186: 317-323Crossref PubMed Scopus (104) Google Scholar) using bovine hemoglobin as the substrate at pH 3.7. The assay temperature was 37 °C. One unit of activity corresponded to the enzymatic activity that liberated trichloroacetic acid-soluble reaction products equivalent to 1 mg of bovine serum albumin in 1 h at 37 °C. Glycoproteins were analyzed using a matrix of 22.4 mg of 3,5-dimethoxy-4-hydroxycinnamic acid in 400 μl of acetonitrile and 600 μl of 0.1% (v/v) trifluoroacetic acid in H2O as the UV-absorbing material. The solubilized samples were mixed with the same volume of matrix, and 1 μl of the mixture was spotted onto the stainless steel tip and dried at room temperature. The concentration of the analyte was ∼5–25 pmol/μl. Measurements were performed on a Bruker REFLEXTM MALDI/TOF mass spectrometer using a N2 laser (337 nm) with a 3-ns pulse width and 107–108 watt/cm2 irradiance at the surface (0.2 mm2 spot). Spectra were recorded at an acceleration voltage of 28.5 kV in the linear mode, using the delayed extraction facility. Affinity-purified phytepsin preparation from barley grains typically contains two enzyme forms of approximately 32 + 16 kDa and 29 + 11 kDa and occasionally some higher molecular mass precursors. In earlier studies (9Sarkkinen P. Kalkkinen N. Tilgmann C. Siuro J. Kervinen J. Mikola L. Planta. 1992; 186: 317-323Crossref PubMed Scopus (104) Google Scholar, 10Runeberg-Roos P. Törmakangas K. Östman A. Eur. J. Biochem. 1991; 202: 1021-1027Crossref PubMed Scopus (104) Google Scholar) molecular weight estimations for different chains were based on SDS-PAGE analyses (Fig. 1) and on calculations from the cDNA-derived protein sequence when the N-terminal sequence was known. However, details about the processing of the different polypeptides at the C terminus were not known. To more accurately describe the sizes of the polypeptides resulting from the processing events, we used MALDI-TOF-MS for the analysis of reduced processed products that corresponded to the polypeptides constituting the two isoforms of phytepsin from barley grains. The sizes determined for two or three independent phytepsin preparations were 9.2–9.5, 15.3–15.8, 26.4–26.7, 30.5–30.8, and 46–47 kDa. Accordingly, the polypeptides were named P9, P15, P27, P31, and P47 (indicated on the left side of Fig. 1). When we analyzed barley root extracts by Western blotting with the same anti-phytepsin antibody, we detected polypeptides P9, P15, and P31. We also detected a polypeptide of ∼26 kDa in roots instead of 27 kDa as previously observed in grains (14Runeberg-Roos P. Kervinen J. Kovaleva V. Raikhel N.V. Gal S. Plant Physiol. 1994; 105: 321-329Crossref PubMed Scopus (96) Google Scholar), and additional polypeptides of approximately 42, 46, and 53 kDa (P42, P46, and P53, respectively) (Fig. 1, lane 2). The similarity among the molecular weights of polypeptides below 31,000 indicates that the processing of phytepsin follows a similar pathway in both grains and roots. P53 and P46 correspond to glycosylated prophytepsin and one-chain phytepsin, respectively. P42 probably corresponds to an additional independent isoenzyme characteristic of the root, as we discuss later in the text. Thus, the roots constitute a good model for studying the expression and processing of phytepsin during the intracellular transport to its final cellular location, the vacuole. The expression and processing of phytepsin were followed by pulse-chase labeling of the roots with [35S]methionine/cysteine. After labeling, the roots were homogenized under nondenaturing conditions, and the processing products were immunoprecipitated with anti-phytepsin antiserum and analyzed by SDS-PAGE (Fig. 2). The precursor P53 appeared during the first 30 min of the pulse, and its half-life was estimated to be 3 h because it was no longer detected 6 h after the beginning of the chase (Fig. 2, A and B). The primary processing products, P31 and P15, appeared 45 min after the pulse (Fig. 2 A), and the further processed polypeptides, P26 and P9, were only observed after 24 h of chase (Fig. 2 B). The processing products of phytepsin were still detected 3 days after the chase (data not shown), indicating a slow turnover rate. The pulse-chase analysis also showed the appearance of an additional protein, P42, 30 min after the pulse. The intensity of this protein remained constant for the entire 24-h chase period, which suggests that it is not related to the proteolytic processing described above. To study whether the processing of P53 occurred before prophytepsin reached the Golgi complex, we treated the roots with the fungal antibiotic brefeldin A, which is known to inhibit Golgi-mediated vesicular traffic by disrupting the Golgi apparatus (22Satiat-Jeunemaitre B. Cole L. Bourett T. Howard R. Hawes C. J. Micros. (Oxf.). 1996; 181: 162-177Crossref PubMed Scopus (142) Google Scholar). Incubation of root cells with brefeldin A before metabolic labeling followed by pulse-chase experiments showed the accumulation of P53, whereas for the nontreated cells at the same chase times, P53 was processed to the two-chain form P31 + P15 (Fig. 3). Brefeldin A affected the processing for several hours after treatment, and only partial processing of P53 was seen at the 2- and 5-h time points. The experiment with brefeldin A clearly shows that processing of P53, which leads to the formation of the two-chain form of phytepsin, occurs only after the precursor has reached the Golgi complex or has migrated beyond it. Brefeldin A did not have any effect on the P42 polypeptide, further corroborating our assumption that it is independent from the described proteolytic pathway. Endo H removes oligomannose but not complex-type N-linked glycans from glycoproteins. Processing of oligomannose to complex-type glycans occurs in the Golgi complex, and therefore, resistance to Endo H indicates localization of a glycoprotein at or beyond the Golgi complex. The prophytepsin sequence contains a singleN-glycosylation site located in P15, and the attached glycans in the mature P31 + P15 form are known to be of the plant complex type (23Costa J. Ashford D.A. Nimtz M. Bento I. Frazão C. Esteves C.L. Faro C.J. Kervinen J. Pires E. Verı́ssimo P. Wlodawer A. Carrondo M.A. Eur. J. Biochem. 1997; 243: 695-700Crossref PubMed Scopus (46) Google Scholar). During the time course of a 30-min chase, the 53-kDa precursor was partially sensitive to Endo H (detected as a wider band), with a shift of about 2 kDa (Fig. 4). This result shows that the glycans linked to the P53 proform are of the oligomannose type for about 30 min after the beginning of the chase, which corresponds to the time taken for the enzyme to reach the Golgi complex, where the glycans are modified. In contrast, the P15 chain was not sensitive to Endo H digestion at any time point (Fig. 4 B), indicating that P15 contains only complex-type glycans. Therefore, we conclude that P15 is produced only when prophytepsin has passed the Golgi complex and, most likely, in transit to or within the vacuole. Plant APs contain a conserved utilized glycosylation site in their plant-specific insert (10Runeberg-Roos P. Törmakangas K. Östman A. Eur. J. Biochem. 1991; 202: 1021-1027Crossref PubMed Scopus (104) Google Scholar). Since it has been suggested that the plant-specific insert might be important for transport to the vacuole (13Guruprasad K. Törmakangas K. Kervinen J. Blundell T.L. FEBS Lett. 1994; 352: 131-136Crossref PubMed Scopus (79) Google Scholar), we investigated the importance of glycans for the intracellular transport. We found that incubation of root cells with theN-glycosylation inhibitor tunicamycin did not inhibit processing of P53 (Fig. 5), but that the processing of P53 to P31 + P15 and P26 + P9 was actually accelerated when glycosylation was inhibited. These results suggest that the glycan moiety of phytepsin protects the enzyme from premature proteolytic cleavage in the Golgi apparatus. To enable a closer study on the processing pattern of phytepsin, we developed a recombinant expression method for this enzyme. Although several APs, including pepsinogen (24Lin X. Wong R.N.S. Tang J. J. Biol. Chem. 1989; 264: 4482-4489Abstract Full Text PDF PubMed Google Scholar), procathepsin D (5Conner G.E. Richo G. Biochemistry. 1992; 31: 1142-1147Crossref PubMed Scopus (56) Google Scholar, 6Beyer B.M. Dunn B.M. J. Biol. Chem. 1996; 271: 15590-15596Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar) and phytepsin, 2J. Kervinen, unpublished information. have been produced in bacterial expression systems, a general problem with these methods has been the very low yield of correctly folded product. Therefore, we chose a baculovirus-infected insect cell expression method for this study. Sf9 cells were infected by a recombinant baculovirus genome containing the complete coding region of preprophytepsin. After incubation for 4 days, a prominent polypeptide of 53 kDa was found in the medium, as analyzed by Western blotting using anti-phytepsin antiserum (not shown). After pepstatin-agarose column chromatography followed by ion exchange chromatography, a typical yield of about 0.5 mg of purified protein was obtained from 1 liter of the cell medium. MALDI-TOF-MS analysis revealed two proteins of 52,847 and 53,062 Da (Fig. 6), which migrated in SDS-PAGE as one broad band at 53 kDa (Fig. 7 A, lane 1). N-terminal sequencing gave one unambiguous sequence of EAEGLVRIAL (Fig. 8). This N-terminal sequence is similar to that previously obtained from in vitro expression in the presence of canine pancreatic microsomes (14Runeberg-Roos P. Kervinen J. Kovaleva V. Raikhel N.V. Gal S. Plant Physiol. 1994; 105: 321-329Crossref PubMed Scopus (96) Google Scholar), which indicates that Sf9 cells are able to cleave the signal sequence from this plant protein to produce a secreted recombinant prophytepsin (rP53). In an isoelectric focusing gel, rP53 migrated to a pI of ∼5.3, which was identical to that observed for phytepsin purified from barley grains (data not shown).Figure 7Autoproteolytic processing of recombinant prophytepsin in vitro. A, purified prophytepsin from insect cell medium (lane 1) was mixed with an equal volume of 0.2 m sodium lactate, pH 3.7, and incubated at 37 °C. Samples were removed after 7, 15, and 60 min (lanes 2–4, respectively); lane 5, incubation for 60 min with 50 μm pepstatin; lane 6, phytepsin purified from barley grains (3 μg). B, incubation of prophytepsin for 90 min at pH 3.7, 4.5, 5.5, and 6.5 (lanes 1–4, respectively). All samples were analyzed by electrophoresis in 20% SDS-polyacrylamide gels and stained with Coomassie Brilliant Blue.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 8Proteolytic processing scheme of phytepsin. ↓, in vivo cleavage site; ▾, in vitro autocatalytic cleavage site. Bold Double underlining denotes the N-terminal polypeptides from in vivo processing (10Runeberg-Roos P. Törmakangas K. Östman A. Eur. J. Biochem. 1991; 202: 1021-1027Crossref PubMed Scopus (104) Google Scholar) and overlining N-terminal polypeptides fromin vitro autoprocessing. Bold cross, glycosylation site. Approximate locations of catalytic residues in the active site are marked by DTG/DSG. The plant-specific domain is shaded.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The molecular weight for rP53 predicted from the cDNA is 51,779. The higher molecular weight observed by MALDI-TOF-MS is due to the presence of oligosaccharides, confirmed by positive staining observed using the periodic acid Schiff method to detect the protein in the gel and indicates that the single potential glycosylation site is occupied. The observed main peak at 52,847 is consistent with the presence on pro-phytepsin of an N-linked oligosaccharide with the structure Man3GlcNAcFucGlcNAc characteristic of proteins expressed in insect cells (predicted molecular weight of 52, 839 (25Grabenhorst E. Hofer B. Nimtz M. Jäger V. Conradt H.S. Eur. J. Biochem. 1993; 215: 189-197Crossref PubMed Scopus (60) Google Scholar), assuming the presence of one sodium atom). The additional signals at 53,062 and 53,273 are probably due to the addition of one or two matrix molecules (sinapinic acid) with the concomitant loss of water (M + nx206) or to the presence of larger oligosaccharide chains. To study the capability of prophytepsin for autoproteolytic processing, we incubated rP53 in buffers over the pH range 3.7–6.5 at 37 °C and removed the samples at various time points. At pH 3.7, the processed polypeptides of 36 and 17 kDa (rP36 and rP17, respectively) were detected after 7 min of incubation, ind
Referência(s)