Characterization and autoprocessing of precursor and mature forms of human immunodeficiency virus type 1 (HIV 1) protease purified from Escherichia coli

A recombinant plasmid encompassing the human immunodeficiency virus type 1 (HIV 1) protease coding sequence and flanking regions (Ala-13 to Gly-185 of the pol open reading frame) has been expressed in two distinct strains of Escherichia coli, AR58 and AR68. In the first strain, AR58, the primary translation product, a 25 kilodalton (kDa) precursor protein, is short-lived and rapidly processes itself to the 11 kDa mature protease in vivo. In the second strain, AR68, the 25 kDa species is only partially processed, and it, a 13 kDa intermediate, and the mature 11 kDa enzyme accumulate at a ratio of 3:4.5:2.5, respectively. The 11 kDa mature protease from AR58 and the 25 kDa precursor from AR68 have been purified to homogeneity. The yield of 11 kDa enzyme from AR58 is approximately 0.02 mg/g wet weight of E. coli cell pellet. The protease has both the expected NH2- and COOH-terminal sequences. The yield of 25 kDa enzyme from AR68 is approximately 0.1 mg/g wet weight of E. coli cell pellet. In vitro, the 25 kDa precursor enzyme rapidly (t1/2 approximately equal to 9 min) processes itself into a species with a mass of approximately 13 kDa and a species with a mass of approximately 11 kDa. Both of these latter species can be separated by RP-HPLC, have the NH2-terminal sequence expected for the mature protease, and are active. The 11 kDa enzyme from AR58 comigrates with the 11 kDa enzyme from AR68 on RP-HPLC and SDS polyacrylamide gel electrophoresis. On extended incubation at 4 degrees C at either neutral or acidic pH all species of the protein exhibit further autodegradation at defined sequences. The availability of the mature, 11 kDa enzyme and the 25 kDa precursor will allow biochemical and physical studies on this critical viral enzyme.     

Investigation of the structure and localization of the urease of Helicobacter pylori using monoclonal antibodies

The urease of Helicobacter pylori (formerly Campylobacter pylori) has been partly purified by fast protein liquid chromatography. This material contained 10 nm doughnut-like structures when examined by electron microscopy and comprised three major polypeptides (61 kDa, 56 kDa and 28 kDa). Only two of these polypeptides (61 kDa and 28 kDa) were observed in urease-containing material isolated by preparative non-denatured PAGE. Monoclonal antibodies (mAbs) were produced which were directed against two of these polypeptides (56 kDa and 28 kDa). Only mAbs directed against the 28 kDa polypeptide inhibited or captured urease activity. These results suggest that the 56 kDa polypeptide is not essential for enzyme activity. Anti-urease mAbs were used in an indirect immunogold technique to localize the enzyme at the ultrastructural level. In both prefixed bacteria and ultrathin cryosectioned bacteria the enzyme was located on the cell surface and in material apparently shed from that surface.     

Posttranslational processing of polysaccharide lyase: maturation route for gellan lyase in Bacillus sp. GL1

Cells of Bacillus sp. GL1 extracellularly secrete a gellan lyase with a molecular mass of 130 kDa responsible for the depolymerization of a heteropolysaccharide (gellan), although the gene is capable of encoding a huge protein with a molecular mass of 263 kDa. A maturation route for gellan lyase in the bacterium was determined using anti-gellan lyase antibodies. The fluid of the bacterial exponentially growing cultures on gellan contained two proteins with molecular masses of 260 and 130 kDa, both of which reacted with the antibodies. The 260 kDa protein was purified from the cultured fluid and characterized. The protein exhibited gellan lyase activity and showed similar enzyme properties, such as optimal pH and temperature, thermal stability, and substrate specificity, to those of the 130 kDa gellan lyase. The N-terminal amino acid sequences of the 260 and 130 kDa enzymes were found to be identical. Determination of the C-terminal amino acid of the 130 kDa enzyme indicated that the 260 kDa enzyme is cleaved between the 1205Gly and 1206Leu residues to yield the mature form (130 kDa) of the gellan lyase. Therefore, the mature enzyme consists of 1170 amino acids (36Ala-1205Gly) with a molecular weight of 125,345, which is in good agreement with that calculated from SDS-PAGE analysis. Judging from these results, gellan lyase is first synthesized as a preproform (263 kDa) and then secreted as a precursor (260 kDa) into the medium through cleavage of the signal peptide. Finally, the precursor is post-translationally processed into the N-terminal half domain of 130 kDa as the mature form, the function of C-terminal half domain being unclear.     

Characterization of a 46 kda insect chitinase from transgenic tobacco

A 46 kDa Manduca sexta (tobacco hornworm) chitinase was isolated from leaves of transgenic tobacco plants containing a recombinant insect chitinase cDNA, characterized, and tested for insecticidal activity. The enzyme was purified by ammonium sulfate fractionation, Q-Sepharose anion-exchange chromatography and mono-S cation-exchange chromatography. Although the gene for the chitinase encoded the 85 kDa full-length chitinase as previously reported by Kramer et al. [Insect Biochem. Molec. Biol. 23, 691–701 (1993)], the enzyme is produced in tobacco as a 46 kDa protein that is approximately four-fold less active than the 85 kDa chitinase. The N-terminal amino acid sequence of the 46 kDa chitinase is identical to that of the 85 kDa chitinase. The former enzyme is not glycosylated, whereas the latter contains approximately 25% carbohydrate. The pH and temperature optima of the 46 kDa chitinaseare similar to those of the 85 kDa chitinase. The former enzyme is more basic than the latter. The 46 kDa chitinase likely consists of the N-terminal catalytic domain of the 85 kDa chitinase and lacks the C-terminal domain that contains several potential sites for glycosylation. The 46 kDa chitinase is expressed in a number of plant organs, including leaves, flowers, stems and roots. Enzyme levels are higher in leaves and flowers than in stems and roots, and leaves from the middle portion of the plant have more chitinase than leaves from the top and bottom portions. Little or no enzyme is secreted outside of the plant cells because it remains in the intracellular space, even though its transit sequence is processed. When fed at a 2% dietary level, the 46 kDa chitinase caused 100% larval mortality of the merchant grain beetle, Oryzaephilis mercator. The results of this study support the hypothesis that insect chitinase is a biopesticidal protein for insect pests feeding on insect chitinase gene-containing transgenic plants.     

Purification and characterization of UDP-N-acetyl-D-glucosamine:dolichol phosphate N-acetyl-D-glucosamine-1-phosphate transferase involved in the biosynthesis of asparagine-linked glycoproteins in the mammary gland

The GlcNAc-1-P-transferase that initiates the dolichol cycle for the biosynthesis of asparagine-linked glycoproteins has been purified from the lactating bovine mammary gland. After solubilization from microsomes with 0.25% Nonidet P-40, the enzyme activity was stabilized with 20% glycerol, 20 micrograms/ml phosphatidylglycerol, 5 microM dolichol phosphate, and 2.5 microM UDP-GlcNAc. The purification protocol involved (NH4)2SO4 precipitation, gel filtration on Sephacryl S-300, DEAE-TSK, and hydroxylapatite chromatography. The purified enzyme was devoid of several readily detectable glycosyltransferases of the dolichol cycle. It showed two bands (A, 50 kDa and B, 46 kDa) on sodium dodecyl sulfate-polyacrylamide gel electrophoresis after either Coomassie Blue or silver staining. Antisera (anti-A and anti-B) raised against individual bands A and B inhibited the enzyme activity in solubilized microsomes. Each of the partially purified antibodies recognizes both bands A and B on Western blots of the enzyme; with the solubilized microsomes, the antibodies also recognize an additional polypeptide of approximately 70 kDa. When radioiodinated microsomes were immunoprecipitated with anti-B and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, again bands of 46, 50, and 70 kDa were observed. The peptide mapping of 50 and 46 kDa bands of the purified enzyme by chemical cleavage with N-chlorosuccinimide gave similar fragmentation patterns. The results indicate that either 70 kDa band is a precursor form of the enzyme or this polypeptide, representing the native enzyme or its subunit, is proteolyzed to smaller, enzymatically active peptide(s) of 50 and 46 kDa during purification despite the inclusion of several inhibitors against serine-proteases in all buffers used for tissue homogenization and enzyme purification. A number of properties of the purified enzyme, including its specific activation by Man-P-Dol were also characterized.     

Purification and properties of the soluble hydrogenase from Desulfovibrio desulfuricans (strain Norway 4)

A soluble hydrogenase has been isolated from Desulfovibrio desulfuricans (strain Norway 4) grown on Postgate's medium. The enzyme differs significantly from a membrane-bound hydrogenase previously purified from the same organism grown on Starkey's medium. The enzyme consisted of two subunits of 56 kDa and 29 kDa compared with masses of 60 kDa and 27 kDa for the membrane-bound enzyme. Analysis of preparations of the soluble enzyme by various methods gave values of 5-10 iron atoms, 6 labile sulphur atoms and 0.45-0.8 nickel atom per molecule. The enzyme was unusual in that it contained selenium, in quantities equivalent to nickel. The highly purified active enzyme produced no electron-spin-resonance (ESR) signals in the oxidized state. ESR signals due to a [3Fe-xS] cluster and nickel were observed only in some of the less active fractions of the enzyme, demonstrating that neither of these ESR-detectable components is a prerequisite for hydrogenase activity. Treatment of D. desulfuricans (Norway) cells with EDTA released a minor fraction with hydrogenase activity, which might indicate the presence of a periplasmic enzyme.     

Structural organization of alpha-subunit from purified and microsomal toad kidney (Na+ + K+)-ATPase as assessed by controlled trypsinolysis

The membrane organization of the alpha-subunit of purified (Na+ + K+)-ATPase ((Na+ + K+)-dependent adenosine triphosphate phosphorylase, EC 3.6.1.3) and of the microsomal enzyme of the kidney of the toad Bufo marinus was compared by using controlled trypsinolysis. With both enzyme preparations, digestions performed in the presence of Na+ yielded a 73 kDa fragment and in the presence of K+ a 56 kDa, a 40 kDa and small amounts of a 83 kDa fragment from the 96 kDa alpha-subunit. In contrast to mammalian preparations (J?rgensen, P.L. (1975) Biochim. Biophys. Acta 401, 399-415), trypsinolysis of the purified amphibian enzyme led to a biphasic loss of (Na+ + K+)-ATPase activity in the presence of both Na+ and K+. These data could be correlated with an early rapid cleavage of 3 kDa from the alpha-subunit in both ionic conditions and a slower degradation of the remaining 93 kDa polypeptide. On the other hand, in the microsomal enzyme, a 3 kDa shift of the alpha-subunit could only be produced in the presence of Na+. Our data indicate that (1) purification of the amphibian enzyme with detergent does not influence the overall topology of the alpha-subunit but produces a distinct structural alteration of its N-terminus and (2) the amphibian kidney enzyme responds to cations with similar conformational transitions as the mammalian kidney enzyme. In addition, anti alpha-serum used on digested enzyme samples revealed on immunoblots that the 40 kDa fragment was better recognized than the 56 kDa fragment. It is concluded that the NH2-terminal of the alpha-subunit contains more antigenic sites than the COOH-terminal domain in agreement with the results of Farley et al. (Farley, R.A., Ochoa, G.T. and Kudrow, A. (1986) Am. J. Physiol. 250, C896-C906).     

Inactivation of maize NADP-malic enzyme by Cu2+-ascorbate

Maize malic enzyme was rapidly inactivated by micromolar concentrations of cupric nitrate in the presence of ascorbate at pH, 5.0. Ascorbate or Cu2+ alone had no effect on enzyme activity. The substrate L-malate or NADP individually provided almost total protection against Cu2+-ascorbate inactivation. The loss of enzyme activity was accompanied by cleavage of the enzyme. The cleaved peptides showed molecular mass of 55 kDa, 48 kDa, 38 kDa, and 14 kDa. Addition of EDTA, histidine and imidazole provided protection. The results of protection experiments with sodium azide, DABCO and catalase suggested that reactive oxygen species were generated resulting in loss of enzyme activity. This was further supported by experiments showing that the rate of enzyme inactivation was higher in D2O than in water. It is suggested that maize malic enzyme is modified by reactive oxygen species like singlet oxygen and H2O2 generated by Cu2+-ascorbate system and the modified amino acid residue(s) may be located at or near the substrate-binding site of the enzyme.     

Purification and characterization of an 8-hydroxy-5-deazaflavin-reducing hydrogenase from the archaebacterium Methanococcus voltae

A methylviologen and 8-hydroxy-5-deazaflavin(F420)-reducing hydrogenase was purified over 800-fold to near homogeneity from the archaebacterium Methanococcus voltae with 10 U mg-1 F420-reducing activity. It is the only hydrogenase in this organism. The enzyme showed Km values of 16 microM for F420 and 1.2 mM for methylviologen. A turnover number of 1050 min-1 was calculated for the minimal active unit. The protein tends to aggregate. The molecular mass of the minimal active unit is 105 kDa. Larger molecules of 745 kDa were regularly observed. The enzyme was resolved into subunits with molecular masses of 55 kDa, 45 kDa, 37 kDa and 27 kDa by SDS/polyacrylamide gel electrophoresis. Reversible conversion of an anionic into an uncharged form was observed by DEAE-cellulose chromatography with concomitant changes in substrate specificities. The methylviologen-reducing activity was heat-resistant up to 65 degrees C and was not affected by antiserum raised against the native enzyme, while F420 reduction was inactivated by both treatments. Nickel and selenium contents were determined as 0.6-0.7 mol each, FAD content as 1 mol and iron as 4.5 mol/mol protein (105 kDa), respectively. Electron micrographs taken from the purified enzyme show ring-shaped molecules of 18 nm diameter, which represent the high-molecular-mass species of the enzyme.     

Occurrence of immunoreactive 80 kDa and non-immunoreactive diacylglycerol kinases in different pig tissues.

We surveyed diacylglycerol kinase in different pig tissues by using rabbit antibody immunospecific to the brain 80 kDa enzyme [Kanoh, Iwata, Ono & Suzuki (1986) J. Biol. Chem. 261, 5597-5602]. Among the other tissues examined, the immunoreactive 80 kDa enzyme was found only in the thymus and, to a much lesser extent, in the spleen, although this enzyme species was widely distributed in a variety of brain regions. Other tissues such as platelets, kidney, heart and liver contained little, if any, immunoreactive enzymes. Gel filtration of cytosolic enzymes from several tissues revealed the presence of three major activity peaks, apparently corresponding to 280, 120 and 80 kDa. Thymus and spleen contained the immunoreactive 80 kDa species together with non-immunoreactive 280 kDa enzyme. In the case of platelets, the kinase consisted almost exclusively of non-immunoreactive 120 kDa species with some 280 kDa enzyme. In an attempt to characterize the different kinase forms, the thymus enzyme was chosen for further studies because of its high activity. No immunoreactive proteins were detected in Western-blot analysis when the 280 kDa enzyme was solvent-extracted, proteinase-treated or preincubated in the presence of Ca2+. In comparison with the 80 kDa species, the 280 kDa enzyme was much more heat-stable and less dependent on deoxycholate in the assay mixture. Although the purification of different forms of the kinase is required to confirm the presence of isoenzymes, the results show that there exist several immunologically distinct diacylglycerol kinase species.     

Two-stage autolysis of the catalytic subunit initiates activation of calpain I

Calcium-induced autolysis of bovine erythrocyte calpain I occurs in multiple stages. Initially, a 14 amino acid segment is cleaved from the N-terminus of the native 80 kDa catalytic subunit, yielding a 78 kDa form of the subunit. Then, an additional 12 amino acid segment is cleaved from the N-terminus, forming a 76 kDa subunit. The 76 kDa enzyme is the active form of the catalytic subunit that is able to proteolyze the 30 kDa regulatory subunit as well as exogenous substrates. While the initial autolytic step requires high calcium, the 76 kDa enzyme form is active in microM calcium and can cleave the amino termini of native 80 kDa and intermediate 78 kDa enzyme forms at low calcium. Both intramolecular and intermolecular proteolysis of the catalytic subunit appear to yield the same products.     

A new iron oxidase from a moderately thermophilic iron oxidizing bacterium strain TI-1.

Iron oxidase was purified from plasma membranes of a moderately thermophilic iron oxidizing bacterium strain TI-1 in an electrophoretically homogeneous state. Spectrum analyses of purified enzyme showed the existence of cytochrome a, but not cytochrome b and c types. Iron oxidase was composed of five subunits with apparent molecular masses of 46 kDa (alpha), 28 kDa (beta), 24 kDa (gamma), 20 kDa (delta), and 17 kDa (epsilon). As the molecular mass of a native enzyme was estimated to be 263 kDa in the presence of 0.1% n-dodecyl-beta-D-maltopyranoside (DM), a native iron oxidase purified from strain TI-1 seems to be a homodimeric enzyme (alpha beta gamma delta epsilon)(2). Optimum pH and temperature for iron oxidation were pH 3.0 and 45 degrees C, respectively. The K(m) of iron oxidase for Fe(2+) was 1.06 mM and V(max) for O(2) uptake was 13.8 micromol x mg(-1) x min(-1). The activity was strongly inhibited by cyanide and azide. Purified enzyme from strain TI-1 is a new iron oxidase in which electrons of Fe(2+) were transferred to haem a and then to the molecular oxygen.     

Identification and characterization of proteases involved in specific proteolysis of vitellogenin and yolk proteins in salmonids.

A pepstatin A-sensitive enzyme involved in yolk formation was purified from the masu salmon (Oncorhynchus masou) ovary using in vitro generation of yolk proteins from purified vitellogenin to assay enzymatic activity. Purification of the enzyme involved precipitation of ovarian extracts by water and ammonium sulfate followed by five steps of column chromatography. After SDS-PAGE and Western blotting, the purified enzyme appeared as a single approximately 42 kDa band that was immunoreactive to anti-human cathepsin D. The course of proteolytic cleavage of the three major yolk proteins (lipovitellin, beta'-component, and phosvitin) in fertilized masu salmon and Sakhalin taimen (Hucho perryi) eggs and embryos was visualized by SDS-PAGE and Western blotting using specific antisera. Major yolk protein bands appeared in positions corresponding to 92 kDa, 68 kDa, and 22 kDa (lipovitellin-derived peptides), as well as 17 kDa (beta'-component). During embryo development, the 92 kDa and 22 kDa bands gradually decreased in intensity, becoming undetectable in alevins. The 68 kDa band and a minor 24 kDa band became more intense after the eyed stage. Two additional peptides, corresponding to 40 and 28 kDa, newly appeared in alevins. During embryonic growth, the beta'-component band (17 kDa) persisted and phosvitin appeared to be progressively dephosphorylated. In vitro analysis of lipovitellin proteolysis indicated that the enzyme involved is a Pefabloc SC-sensitive serine protease. These results demonstrate, for the first time, that a cathepsin D-like protease and serine proteases play key roles in yolk formation and degradation, respectively, in salmonid fishes.     

Proteolytic release of keratinocyte transglutaminase.

Human keratinocytes express a particulate transglutaminase that can be released from the membrane by limited proteolysis with trypsin or plasmin to yield a form that is congruent to 80 kDa. The enzyme from cultured cells was also releasable by endogenous proteolysis to yield a catalytically active fragment of congruent to 80 kDa. Endogenous release was strongly dependent upon temperature and Ca2+ concentration and was inhibited by iodoacetate, but not by leupeptin, antipain or phenylmethanesulphonyl fluoride. These phenomena raise the possibility of partial translocation of transglutaminase activity to the cytoplasm by proteolysis to which the enzyme is subject during terminal differentiation. In addition, hydrodynamic measurements showed that the endogenously released enzyme was monomeric in solution (79 kDa), whereas that solubilized by hydroxylamine without proteolysis appeared dimeric (190 kDa). The latter dimeric state may reflect either an altered conformation of the enzyme or post-translational modification beyond fatty acid esterification.     

Purification and characterization of protein methylase II from Helicobacter pylori

Protein methylase II (AdoMet:protein-carboxyl O-methyltransferase, EC 2.1.1.24) was identified and purified 115-fold from Helicobacter pylori through Q-Sepharose ion exchange column, AdoHcy-Sepharose 4B column, and Superdex 200 HR column chromatography using FPLC. The purified preparation showed two protein bands of about 78 kDa and 29 kDa molecular mass on SDS-PAGE. On non-denaturing gel electrophoresis, the enzyme migrated as a single band with a molecular mass of 410 kDa. In addition, MALDI-TOF-MS analysis and Superdex 200 HR column chromatography of the purified enzyme showed a major mass signal with molecular mass values of 425 kDa and 430 kDa, respectively. Therefore, the above results led us to suggest that protein methylase II purified from H. pylori is composed of four heterodimers with 425 kDa (4x(78+29)=428 kDa). This magnitude of molecular mass is unusual for protein methylases II so far reported. The enzyme has an optimal pH of 6.0, a K(m) value of 5.0x10(-6) M for S-adenosyl-L-methionine and a V(max) of 205 pmol methyl-(14)C transferred min(-1) mg(-1) protein.     

Matrix metalloproteinase-9 in a unique proteoglycan form in avian embryonic growth plate cartilage

The appearance of a high molecular weight gelatinolytic enzyme (230 kDa) correlated with cartilage collagen loss in chick embryonic tibias cultured with lipopolysaccharide. This 230 kDa enzyme was purified and its activity was measured on synthetic and natural substrates. The enzyme was activated by aminophenylmercuric acetate and inhibited by ethylenediaminetetraacetic acid, phenanthroline, marimastat or tissue inhibitors of metalloproteinases. Amino acid sequences of peptides derived from the purified enzyme showed identity with avian MMP-9. Digestion of the intact enzyme with chondroitinase decreased the size of the molecule to 80 kDa on SDS-PAGE. When chick embryonic tibia cultures were radiolabeled with (35)S-sulfate, the radiolabel co-purified with the 230 kDa gelatinase. Chondroitinase treated 230 kDa gelatinase also reacted with specific anti-chondroitin sulfate antibodies and FACE analysis revealed a predominance of chondroitin-4-sulfate. These results demonstrate this avian matrix metalloproteinase contained glycosaminoglycan chains. To our knowledge, this is the first report of a matrix metalloproteinase in a proteoglycan form.     

Purification of phospholipid methyltransferase from rat liver microsomal fraction.

Phospholipid methyltransferase, the enzyme that converts phosphatidylethanolamine into phosphatidylcholine with S-adenosyl-L-methionine as the methyl donor, was purified to apparent homogeneity from rat liver microsomal fraction. When analysed by SDS/polyacrylamide-gel electrophoresis only one protein, with molecular mass about 50 kDa, is detected. This protein could be phosphorylated at a single site by incubation with [alpha-32P]ATP and the catalytic subunit of cyclic AMP-dependent protein kinase. A less-purified preparation of the enzyme is mainly composed of two proteins, with molecular masses about 50 kDa and 25 kDa, the 50 kDa form being phosphorylated at the same site as the homogeneous enzyme. After purification of both proteins by electro-elution, the 25 kDa protein forms a dimer and migrates on SDS/polyacrylamide-gel electrophoresis with molecular mass about 50 kDa. Peptide maps of purified 25 kDa and 50 kDa proteins are identical, indicating that both proteins are formed by the same polypeptide chain(s). It is concluded that rat liver phospholipid methyltransferase can exist in two forms, as a monomer of 25 kDa and as a dimer of 50 kDa. The dimer can be phosphorylated by cyclic AMP-dependent protein kinase.     

Regulation of squid visual phospholipase C by activated G-protein alpha.

Phospholipase C (PLC) is the key enzyme in the phototransduction cascade of invertebrate rhabdomeric photoreceptors. In addition to 130 kDa PLC, a 95 kDa protein recognized by antibody against the catalytic site of PLC was found in the squid retina. The PLC-like 95 kDa protein (95 kDa PLC) was produced from 130 kDa PLC by an intrinsic protease in the presence of calcium. The 130 kDa PLC was stimulated by the active form of Gq-class G-protein alpha (Gq alpha), but the 95 kDa PLC was not, although their PLC activity was similar. A 35 kDa fragment, the counterpart of 95 kDa PLC, was not recognized by antibodies against catalytic site or N-terminal site of the 130 kDa PLC, indicating that the cleavage site is on the C-terminal side beyond the catalytic site. In the presence of a large excess of the 35 kDa fragment, 95 kDa PLC was stimulated by Gq alpha to a similar extent as intact 130 kDa PLC. These results indicate that the C-terminal polypeptide of PLC is necessary for regulation of its enzyme activity by Gq alpha. The uncoupling of PLC from Gq alpha, caused by limited proteolysis, is therefore a candidate regulatory mechanism of the phototransduction cascade in rhabdomeric photoreceptors.     

Characterization of trimming Man9-mannosidase from pig liver. Purification of a catalytically active fragment and evidence for the transmembrane nature of the intact 65 kDa enzyme.

An alpha 1,2-mannosidase (Man9-mannosidase) involved in N-linked oligosaccharide processing has been purified about 16,000-fold from pig liver crude microsomes (microsomal fractions) by CM-Sepharose and DEAE-Sephacel chromatography, concanavalin A (Con A)-Sepharose chromatography and, as the key step of the procedure, affinity chromatography on immobilized N-5-carboxypentyl-l-deoxymannojirimycin (CP-dMM). On SDS/polyacrylamide-gel electrophoresis under reducing conditions, the isolated enzyme migrated as a single protein band with a molecular mass of 49 kDa. The enzyme does not bind Con A and is not susceptible to glycopeptidase F, indicating that it lacks N-linked oligosaccharides of the high-mannose or complex type. Purified Man9-mannosidase has a pH optimum close to 6.0 and requires bivalent cations for activity, with Ca2+ being most effective. The enzyme is inhibited strongly by basic sugar analogues of mannose such as 1-deoxymannojirimycin (dMM, Ki approximately 5 microM), N-methyl-dMM (Ki approximately 55 microM) and CP-dMM (Ki approximately 150 microM), whereas NN-dimethyl-dMM and the mannosidase II inhibitor swainsonine were hardly or not at all inhibitory. A homogeneous preparation of the 49 kDa enzyme cleaves specifically three of the four alpha 1,2-mannosidic linkages in the natural Man9-GlcNAc2 (M9) substrate. The relative rates by which the parent and intermediate structures are hydrolysed were found to be about 3:2:5 for M9, M8 and M7 respectively. The enzyme displays only marginal activity toward the remaining alpha 1,2-mannosidic linkages in the Man9-GlcNAc2 oligosaccharide (relative rate of M6 hydrolysis approximately 0.02) and is not active against nitrophenyl and methylumbelliferyl alpha-mannosides. This unique substrate specificity suggests that Man9-mannosidase processing differs from that catalysed by other trimming alpha 1,2-mannosidases hitherto reported. A polyclonal antibody raised against the denatured 49 kDa polypeptide not only recognizes a protein band of similar size in Western blots of crude microsomes, but also reacts strongly with a 65 kDa protein species. On trypsin treatment of detergent-solubilized microsomes, the 65 kDa protein is converted specifically into a stable 49 kDa fragment, indicating a precursor-product relationship between the two proteins. We conclude from this observation that the 65 kDa protein represents the intact form of Man9-mannosidase from which the 49 kDa enzyme which we have isolated has been generated, with retention of catalytic activity, by proteolysis during purification. Proteolytic studies with sealed microsomes suggest that the intact 65 kDa enzyme is a protein with a membrane-spanning domain, as well as a cytosolic polypeptide domain of size at least 3 kDa.