Limited proteolysis of maize NADP-malic enzyme

The incubation of maize malic enzyme at 37 degrees C with trypsin at a ratio of 150:1 of malic enzyme to trypsin caused rapid and complete inactivation of enzyme activity. The inactivation was caused by fairly specific cleavage of the enzyme monomer (62 kDa) into 40 kDa and 20 kDa fragments. The intensity of 40 kDa band increased with the time of treatment of enzyme with trypsin from 2 to 30 min. Substrates, especially NADP (25 microM) provided almost total protection against trypsin inactivation of the enzyme activity. The studies carried out with various other endoproteases indicated that endoprotease Lys-C was most effective in inactivating malic enzyme activity. The kinetic properties of the truncated enzyme have been studied. The Km value for malate in case of native and modified enzyme was found to be identical. Km NADP for the modified enzyme was slightly higher indicating that after proteolysis the enzyme affinity for NADP had decreased. Limited proteolysis with trypsin did not show any appreciable change in fluorescence properties of the modified enzyme. Binding of NADPH to the enzyme was not affected after modification.     

Bacterial phosphotransferase system: regulation of mannitol enzyme II activity by sulfhydryl oxidation

The mechanism by which the oxidation-reduction potential regulates the bacterial phosphotransferase system in Escherichia coli has been investigated. Transphosphorylation experiments verified that the oxidizing agent, potassium ferricyanide, directly inhibits mannitol enzyme II activity. Phosphorylation of enzyme IImtl with enzyme I, heat-stable phosphocarrier protein of the phosphotransferase system, and phosphoenolpyruvate partially protects the enzyme from ferricyanide inhibition. The enzyme is even less sensitive to inhibition during catalytic turnover. Preincubation of unphosphorylated enzyme with ferricyanide, however, reversibly inactivates it even at high mannitol concentrations. The results are inconsistent with a regulatory mechanism in which sulfhydryl oxidation influences the affinity of the enzyme for the substrate. Instead, it is concluded that the oxidized enzyme is inactive.     

Immunological characterization of maize starch branching enzymes

Highly purified fractions of three starch branching enzymes from developing maize (Zea mays L.) endosperm were used to prepare antisera in rabbits. In double diffusion experiments, no immunoprecipitate was observed when branching enzyme IIa or IIb was tested against branching enzyme I antiserum. No immunoprecipitate was formed when branching enzyme I was tested against branching enzyme IIa or IIb antiserum. Increasing amounts of antisera in the above combinations also failed to inhibit enzyme activity. Branching enzyme IIa antiserum cross-reacted and formed spurs with branching enzyme IIb when compared with branching enzyme IIa antigen. Comparison of branching enzyme IIb antiserum with branching enzyme IIa also resulted in an immunoprecipitate. Increasing levels of branching enzyme IIa antiserum inhibited branching enzyme IIb as did the reciprocal combination. The data indicated that branching enzymes IIa and IIb are immunologically similar while branching enzyme I is distinct. The data supports the classification of starch branching enzymes based on genetic, kinetic, and chromatographic properties.     

宫川蜜柑根际土壤酶活性与土壤养分含量相关性的研究

研究了不同肥力水平的宫川蜜柑根际土壤酶的活性及其与土壤农化特性的关系。结果表明 :高产园的土壤酶活性显著高于低产园的土壤酶活性。经统计分析 ,土壤酶活性与养分含量均呈极显著相关。而且酶的活性在土壤中的分布有一定的规律性。其水平分布是在树冠内半径的 4 /5处至树冠滴水线范围内 ,酶的活性最高 ,由此处向内向外酶的活性逐渐降低 ;其垂直分布是 0～ 2 0 cm土层酶的活性最高 ,随土层的加深而逐渐降低     

Purification and characterization of human liver beta-galactosidase from a patient with the adult form of GM1 gangliosidosis and a normal control

beta-Galactosidases were purified to homogeneity from livers of a normal control and a patient with the adult form of GM1 gangliosidosis. The purification was achieved by chromatography on DEAE-Sepharose fast flow, Con A-Sepharose, p-aminophenyl-1-thio-beta-D-galactopyranoside-Sepharose, and QAE-Mono Q. The normal and mutant enzymes were purified about 5000-fold with a yield of 10% and 1800-fold with a yield of 34%, respectively, and could hydrolyze 4-methylumbelliferyl-beta-D-galactoside, GM1 ganglioside, and asialofetuin. The purified normal enzyme was eluted from a TSK gel G-4000SW column as three symmetrical peaks of protein which were coincident with the three peaks of enzyme activity. The enzyme in these three peaks had apparent molecular weights of 800,000 (polymer), 140,000 (dimer), and 65,000 (monomer), whereas the mutant enzyme was eluted as two symmetrical peaks of protein and enzyme activity. The apparent molecular weight of a major monomeric form of the enzyme (beta-galactosidase A) was 60,000, and no dimeric form of the enzyme existed. Normal and mutant purified enzyme preparations migrated as a single major protein band with apparent molecular weights of 65,000 or 60,000, respectively, by SDS-polyacrylamide gel electrophoresis after treatment with mercaptoethanol. On isoelectric focussing, the mutant enzyme migrated more anodally than the normal enzyme. The mutant enzyme also had altered enzyme properties, such as pH optimum, Km values, substrate specificity and heat-stability. These data on the characteristics of the purified enzyme preparations provide the first direct evidence that patients with the adult form of GM1 gangliosidosis have a structurally altered beta-galactosidase.     

Enzymatic modification of vegetable protein: Immobilization of Penicillium duponti enzyme on reconstituted collagen and the use of the immobilized-enzyme complex for solubilizing vegetable protein in a recycle reactor

Penicillium duponti enzyme was immobilized on reconstituted collagen by macromolecular complication, impregnation, and covalent crosslinking techniques. The immobilization of the enzyme on collagen has a twofold purpose: (1) providing a protein microenvironment for the proteolytic enzyme; and (2) extending the useful life the enzyme once immobilized on the collagen matrix. Two types of collagen were used, one produced by the United States Department of Agriculture and the other produced by FMC. The USDA collagen contained unhydrolyzed telepeptide linkages and required pretreatment to reduce collagenaselike activity of the enzyme. Activity analysis of the immobilized enzyme complex showed that membranes with enzyme loading less than 10 mg enzyme/g of wet membrane in the reactor were dimensionally stable. The degree of crosslinking was an important parameter. Membranes with structural opening up to three times the initial dry thickness were found to be the maximum limit for controlled release of enzyme from the collagen membrane during enzymatic reaction. Higher activities and better stability of the enzyme in collagen membrane were found for covalent crosslinking of the enzyme to treated collagen films. The hydrolysis of soybean vegetable protein with the immobilized enzyme in a recycle reactor at enzyme loading of mg/g of wet membrane at 40°C, pH 3.4, produced 56.5% of soluble protein in 10h. The production is equivalent to 1.84 h total contact time between the substrate and the immobilized enzyme. The average productivity based on a stable enzyme activity and 20g of dry membrane was 329 mg of protein/g/mg of active enzyme immobilized. The productivity of the free enzyme in a batch reactor was 62.5 mg protein/h/mg enzyme.     

Substitution of glutamine for lysine at the pyridoxal phosphate binding site of bacterial D-amino acid transaminase. Effects of exogenous amines on the slow formation of intermediates

In bacterial D-amino acid transaminase, Lys-145, which binds the coenzyme pyridoxal 5'-phosphate in Schiff base linkage, was changed to Gln-145 by site-directed mutagenesis (K145Q). The mutant enzyme had 0.015% the activity of the wild-type enzyme and was capable of forming a Schiff base with D-alanine; this external aldimine was formed over a period of minutes depending upon the D-alanine concentration. The transformation of the pyridoxal-5'-phosphate form of the enzyme to the pyridoxamine-5'-phosphate form (i.e. the half-reaction of transamination) occurred over a period of hours with this mutant enzyme. Thus, information on these two steps in the reaction and on the factors that influence them can readily be obtained with this mutant enzyme. In contrast, these reactions with the wild-type enzyme occur at much faster rates and are not easily studied separately. The mutant enzyme shows distinct preference for D- over L-alanine as substrates but it does so about 50-fold less effectively than the wild-type enzyme. Thus, Lys-145 probably acts in concert with the coenzyme and other functional side chain(s) to lead to efficient and stereochemically precise transamination in the wild-type enzyme. The addition of exogenous amines, ethanolamine or methyl amine, increased the rate of external aldimine formation with D-alanine and the mutant enzyme but the subsequent transformation to the pyridoxamine-5'-phosphate form of the enzyme was unaffected by exogenous amines. The wild-type enzyme displayed a large negative trough in the circular dichroic spectrum at 420 nm, which was practically absent in the mutant enzyme. However, addition of D-alanine to the mutant enzyme generated this negative Cotton effect (due to formation of the external aldimine with D-alanine). This circular dichroism band gradually collapsed in parallel with the transformation to the pyridoxamine-5'-phosphate enzyme. Further studies on this mutant enzyme, which displays the characteristics of the wild-type enzyme but at attenuated rates, may yield information on the factors controlling the stereochemistry of the reaction as well as on the catalytic steps of the transaminase pathway.     

D-alpha-Hydroxyglutarate dehydrogenase of Rhodospirillum rubrum.

D-alpha-Hydroxyglutarate dehydrogenase of R. rubrum grown anaerobically in the light was partially purified and some properties were investigated. 1. The enzyme catalyze stoichiometrically the dehydrogenation reaction of D-alpha-hydroxyglutarate into alpha-oxoglutarate, coupled with the reduction of 2, 6-dichlorophenolindophenol. 2. Cytochrome c2, cytochrome c, and ferricyanide are effective as electron acceptors with the crude enzyme but not with the purified one, whereas NAD+ and NADP+ are completely ineffective. The enzyme is thought to play a role in the electron transport system of the organism. 3. D-alpha-Hydroxyglutarate is virtually the sole substrate for the enzyme. The apparent activity against L-alpha-hydroxyglutarate is presumed to be due to contamination of the L-isomer sample with the D-isomer. The enzyme shows barely detectable activity against both isomers of malate and virtually no activity against DL-lactate and glycolate. 4. Both isomers of malate and oxalate, which are presumably substrate analogues, inhibit the enzyme activity. 5. The enzyme is not an inducible enzyme but rather is a constitutive one for R. rubrum, unlike from the enzyme of Pseudomonas putida which is an inducible enzyme for the catabolism of lysine.     

Immobilization of Aspergillus oryzae tannase and properties of the immobilized enzyme

Tannase enzyme from Aspergillus oryzae was immobilized on various carriers by different methods. The immobilized enzyme on chitosan with a bifunctional agent (glutaraldehyde) had the highest activity. The catalytic properties and stability of the immobilized tannase were compared with the corresponding free enzyme. The bound enzyme retained 20·3% of the original specific activity exhibited by the free enzyme. The optimum pH of the immobilized enzyme was shifted to a more acidic range compared with the free enzyme. The optimum temperature of the reaction was determined to be 40 °C for the free enzyme and 55 °C for the immobilized form. The stability at low pH, as well as thermal stability, were significantly improved by the immobilization process. The immobilized enzyme exhibited mass transfer limitation as reflected by a higher apparent K_m value and a lower energy of activation. The immobilized enzyme retained about 85% of the initial catalytic activity, even after being used 17 times.     

Characterization of rat muscle fructose 1,6-bisphosphatase

Fructose 1,6-bisphosphatase has been purified from rat muscle. Although the specific activity of the enzyme in the crude extract of rat muscle was extremely low, purification by the present procedure is highly reproducible. The purified enzyme showed a single band in SDS-polyacrylamide gel electrophoresis. The subunit molecular weight of the muscle enzyme was 37,500 in contrast to 43,000 in the case of the liver enzyme. Immunoreactivity of the muscle enzyme to anti-muscle and anti-liver fructose 1,6-bisphosphatase sera was clearly distinct from that of the liver enzyme. All one-dimensional peptide mappings of the muscle enzyme with staphylococcal V8 protease, chymotrypsin, and papain showed different patterns from those of the liver enzyme. When incubated with subtilisin, the extent of activation of muscle fructose 1,6-bisphosphatase at pH 9.1 was smaller than that of the liver enzyme. The subtilisin digestion pattern of the muscle enzyme on SDS-polyacrylamide gel electrophoresis was distinct from that of the liver enzyme. The AMP-concentration giving 50% inhibition of the muscle enzyme was 0.54 microM, whereas that of the liver enzyme was 85 microM. The concentrations of fructose 2,6-bisphosphate that gave 50% inhibition of rat muscle and liver enzymes were 6.3 and 1.5 microM, respectively. Fructose 1,6-bisphosphatase protein was not detected in soleus muscle by immunoelectroblotting with anti-muscle fructose 1,6-bisphosphatase serum.     

Substrate-dependent dissociation of malate thiokinase.

Malate thiokinase has been purified to apparent homogeneity by employing conventional purification techniques along with affinity chromatography. The enzyme is composed of two nonidentical subunits (alpha subunit Mr=34,000, beta subunit Mr=42,500) to yield an alpha 4 beta 4 structure for the native enzyme. Phosphorylation of the enzyme by ATP occurs exclusively on the alpha subunit. The phosphorylated enzyme is acid labile and base stable consistent with phosphorylation of a histidine residue. Dephosphorylation of the enzyme is promoted by ADP, succinate, malate, and coenzyme A plus inorganic phosphate. Phosphorylation of the enzyme leads to a reversible change in the sedimentation properties of the enzyme; the native enzyme exhibits an S20,w of approximately 10, whereas the phosphoenzyme exhibits an S20,w of approximately 7. Formation of the 7 S form of the enzyme is also observed when coenzyme A and succinyl-CoA interact with the enzyme. The ratio of alpha to beta subunits in both the 10 S and 7 S forms of the enzyme is approximately 1.0, suggesting that the 7 S form of the enzyme has an alpha 2 beta 2 structure.     

Enzyme synthesis in the regulation of hepatic `malic'' enzyme activity

A homogeneous preparation of ;malic' enzyme (EC 1.1.1.40) from livers of thyroxine-treated rats was used to prepare in rabbits an antiserum to the enzyme that reacts monospecifically with the ;malic' enzyme in livers of rats in several physiological states. Changes in enzyme activity resulting from modification of the state of the animal are hence due to an altered amount of enzyme protein. The antiserum has been used to precipitate out ;malic' enzyme from heat-treated supernatant preparations of livers from both adult and neonatal rats, in a number of physiological conditions, that had been injected 30min earlier with l-[4,5-(3)H]leucine. The low incorporations of radioactivity into the immunoprecipitable enzyme have permitted the qualitative conclusion that changed enzyme activity in adult rats arises mainly from alterations in the rate of enzyme synthesis. The marked increase in ;malic' enzyme activity that occurs naturally or as a result of thyroxine treatment of the weanling rat is likewise due to a marked increase in the rate of enzyme synthesis possibly associated with a concurrent diminished rate of enzyme degradation.     

Characterization and partial purification of two enzymes transferring N-acetylglucosamine to dolichyl monophosphate and ribonuclease A

Two N-acetylglucosamine (GlcNAc) transferases which catalyze the incorporation of GlcNAc into GlcNAc-P-P-dolichol (dolichol enzyme) and into bovine pancreatic ribonuclease A (RNAseA enzyme) were solubilized from the rat liver microsomes in a non-ionic detergent, Triton X-100. Both enzyme activities were adsorbed on activated CH-Sepharose 4B, and could be eluted with a linear KCl gradient. Two enzyme activities were separated by this column with the dolichol enzyme eluting before the RNAseA enzyme. A 49-fold and 136-fold purification was achieved for the dolichol and the RNAseA enzyme, respectively. The addition of exogeneous dolichyl phosphate resulted in a 3-5-fold stimulation of the purified dolichol enzyme, but did not affect the purified RNAseA enzyme. The addition of RNAseA stimulated only the RNAseA enzyme. Whereas, tunicamycin could inhibit only the dolichol enzyme. The purified dolichol enzyme had a Km of 14 X 10(-6) M for UDP-GlcNAc and the reaction was saturated with about 0.25 M dolichyl phosphate. The purified RNAseA enzyme had a Km of 4.55 X 10(-6) M for UDP-GlcNAc and was saturated with about 0.36 mM RNAseA. The pH optima and the metal ion requirement for the two enzymes were different. These results suggest that because of the different properties of these two enzymes they may have distinct functions regarding the core glycosylation of N-linked glycoproteins. It is well established that the dolichol enzyme catalyzes the formation of the first dolichol-linked intermediate GlcNAc-P-P-dolichol, whereas according to the present finding, the RNAseA enzyme may catalyze the transfer of GlcNAc directly from UDP-GlcNAc into acceptor protein.     

Isolation and properties of rat plasma lecithin-cholesterol acyltransferase

Lecithin-cholesterol acyltransferase was purified from rat plasma and the properties of this enzyme during the purification procedures and those of the purified enzyme were investigated in comparison with the human enzyme. The rat enzyme was not adsorbed on hydroxyapatite, which was employed for the purification of the human enzyme. When purified human enzyme was incubated at 37 degrees C in 0.1 mM phosphate buffer (pH 7.4; ionic strength, 0.00025), no alteration of enzyme activity was observed for up to 6 h. In the case of the rat enzyme, however, approximately 40% of the enzyme activity was lost under the same conditions. The human enzyme and rat enzyme were both retained on a Sepharose 4B column to which HDL3 was covalently linked, in 39 mM phosphate buffer, pH 7.4. Although the human enzyme was eluted from the column in 1 mM phosphate buffer, the rat enzyme was dissociated from the column at a lower buffer concentration (0.1 mM phosphate buffer). These findings indicate that the rat enzyme effectively associated with HDL3 in 39 mM phosphate buffer, pH 7.4, but the association was more sensitive to increase of ionic strength compared with that of the human enzyme.     

Immobilized isocrite dehydrogenase: some aspects of its catalytic, chemical and immunological properties

Isocitrate dehydrogenase from Azotobacter vinelandii has been immobilized on Sepharose 4B with an efficiency of between 60 and 75%. The immobilized enzyme is assayed by a flow technique which monitors a final steady state level of product formation. By the assay system described it is estimated that the immobilized enzyme retains between 30 and 40% of the catalytic activity of the free enzyme. Studies have been carried out on the substrate dependence of the enzyme. The enzyme requires magnesium ions with optimal concentrations of 10⁻³m and above. The dependence on isocitrate and TPN⁺ concentrations was determined and analyzed by double-reciprocal plots. The immobilized enzyme is inactivated by DTNB [5,5′-dithiobis(2-nitrobenzoic acid)] and reactivated by DTT (dithiothreitol). The DTNB-modified enzyme can be reactivated by potassium cyanide. Comparison of these reactions with those of the free enzyme suggest that the steric environment of the active site was not grossly altered by immobilization. Some supporting evidence is derived from the identity of the energies of activation, 16,600 cal/mole, of free and immobilized enzyme catalyzed oxidation of isocitrate. Furthermore, the immobilized enzyme is inactivated by antibody prepared against the free enzyme. The covalently attached enzyme is resistant to tryptic digestion except in the presence of 2 m urea. This suggests that exposed lysyl residues which may be the primary site of attack by trypsin are utilized in immobilization. Treatment of the enzyme with 2 m urea unfolds the enzyme to a conformation which has very little activity but which recovers full activity upon removal of the urea. Interaction of the enzyme with antibody suggest that the antibody reacts univalently. The second valence can be satisfied by addition of free enzyme. The free enzyme bound to the immobilized enzyme-antibody complex is active. Preliminary attempts to dissociate the enzyme-antibody complexes have been unsuccessful.     

Evidence that two covalent intermediates, phosphoryl and malonyl enzymes, are formed during malonyl-coenzyme A synthetase catalysis

The isolation of malonyl-coenzyme A synthetase from Pseudomonas fluorescens grown on malonate has been reported recently (Kim, Y.S., and Bang, S.K. (1985) J. Biol.Chem. 260, 5098-5104). This enzyme is phosphorylated in the presence of ATP and Mg2+. The phosphoryl group appears on one subunit of the enzyme composed of two different subunits, and the phosphoryl enzyme is acid labile and base stable. The phosphoryl group on the enzyme is released by the incubation of the phosphoryl enzyme with malonate and malonyl enzyme is formed. The malonyl enzyme is acid labile and also relatively unstable under basic conditions. The malonyl group is found on the subunit of the enzyme which is phosphorylated. Malonyl-CoA is formed when malonyl enzyme reacts with coenzyme A. These results suggest that two convalent intermediates, phosphoryl and malonyl enzyme, are sequentially formed in the synthesis of malonyl-coenzyme A by malonyl-coenzyme A synthetase catalysis.     

Differences in redox and kinetic properties between NAD-dependent and O2-dependent types of rat liver xanthine dehydrogenase

Reductive titrations of a NAD-dependent type (type-D) and an O2-dependent type (type-O) of rat liver xanthine dehydrogenase showed that only the type-D enzyme formed a pronounced stable FAD semiquinone (FADH*). The FAD semiquinone was less stabilized in the presence of NAD. The Vmax value for xanthine-NAD activity of type-D enzyme was close to that for xanthine-O2 activity of type-O enzyme, while the Vmax value for xanthine-O2 activity of type-D enzyme was about one-fourth of that of type-O enzyme. The Km value for O2 of type-D enzyme was about five times as large as that of type-O enzyme. The absorbance spectrum of type-D enzyme during turnover with xanthine and O2 as substrates showed a considerable amount of FADH* formation, but that with xanthine and NAD as substrates showed only a negligible one. Low xanthine-O2 activity of type-D enzyme, as compared with that of type-O enzyme, seems to be explained by the conformational change occurring in conversion from type-O to type-D enzyme, which results in different reactivity of FAD to molecular oxygen and a higher fraction of FADH* during turnover. The binding of NAD may possibly increase the fraction of FADH2, resulting in a Vmax value of xanthine-NAD activity almost as high as that of xanthine-O2 activity of type-O enzyme.     

多聚半乳糖醛酸内切酶的固定化及其性质研究

棉花枯萎病菌多聚半乳糖醛酸内切酶在pH大于7时不稳定,故对它进行多种化学修饰而又不影响其活性,必须在pHd小于7的体系中进行。本文报道将PGAUase在还原剂存在下,与稀酸处理的Sepharose 4B交联,获得较高活力的固定化酶。固定化酶催化动力学表明,最适pH为4,4,最适温度为55℃,在pH1至8.0范围内稳定。和溶液酶比较,对热稳定性提高,但对碱稳定性下降。以多聚半乳糖醛酸为底物,Km为0.27mmol/L,Vmax为66.67nmol/L·min,均大于溶液酶(Km=0.07mmol/L,Vmax=28.00nmol/L·min)。在pH4.8,30℃,聚半乳糖醛酸在固相酶的柱中循环水解不同的时间降解产物经圆盘电泳和等电聚焦测定,得到不同大小的寡糖片段混合物,证明固相酶和溶液酶的作用方式相同,同时使以酶解法制备一定大小的有生物活性的寡糖分子成为可能。     

Effects of phospholipids on L-lactate dehydrogenase from membranes of Escherichia coli. Activation and stabilization of the enzyme with phospholipids.

Membrane-bound L-lactate dehydrogenase was freed from the detergent used during purification. The detergent-free enzyme had about one-half the specific activity of the enzyme in 1.0% Tween 80, and was only partially sensitive to the specific antibody. This enzyme was activated about 3-fold with phosphatidylglycerol, cardiolipin, or a mixture of phospholipids. The phospholipid-activated enzyme had a similar Km value for L-lactate to that of the membrane enzyme and was completely inhibited by the specific antibody. On heat treatment, the phospholipid-activated enzyme was more stable than detergent-free enzyme and was as stable as membrane-bound enzyme. The alpha helical content of the enzyme increased 1.7-fold during preincubation with these lipids and the alpha helix became more stable during heat treatment than that of the detergent-free enzyme. These results suggest that the enzyme showed monomolecular dispersion in the lipid bilayer and that its conformation, including its active site and secondary structure, was different from that of the detergent-free enzyme. Phosphatidylethanolamine, dilauroyl lecithin and lecithin from egg yolk had none of the above effects on the activity or the secondary structure of the enzyme. On the other hand, mixtures of each of these lipids and cholate had essentially similar effects to phosphatidylglycerol.     

Purification and characterization of dipeptidyl peptidase IV from Streptococcus salivarius HHT.

Dipeptidyl peptidase IV (DAP IV) was purified from Streptococcus salivarius HHT by anion-exchange chromatography, gel filtration and affinity chromatography after lysis of cell walls with N-acetylmuramidase. DAP IV was purified 114-fold with a yield of 16.6% from total activity of the crude extract. The purified enzyme was shown to be homogeneous by disc gel electrophoresis. The molecular weight of the enzyme was estimated to be about 109,000 by gel filtration and 47,000 by sodium dodecylsulphate SDS-polyacrylamide gel electrophoresis, suggesting that the native enzyme is a dimeric form. The optimum pH for the reaction was 8.7 in Gly-NaOH buffer, and the isoelectric point of the enzyme was pH 4.2. The enzyme hydrolysed specifically N-terminal X-Pro from X-Pro-p-nitroanilides. The enzyme activity was hardly affected by various cations, sulphydryl-blocking reagents and metal chelators. The enzyme activity was markedly inhibited by 1 mM diisopropylfluoride, and the desialysed enzyme was attacked by proteinases.