Protease II, a cytoplasmic endopeptidase from Escherichia coli , has been further characterized. In agreement with previous evidence of a trypsin‐like specificity, this enzyme cleaves essentially the carboxymethylated B chain of insulin at the Arg 22 ‐Gly 23 bond, but after prolonged periods of incubation it is also able to cleave the Tyr 16 ‐Leu 17 bond. Protease II and pancreatic trypsin are inhibited in a similar fashion by diisopropylphosphofluoridate and tosyl lysine chloromethyl ketone suggesting that seryl and histidyl groups are commonly involved in the active site of the two enzymes. The analysis of the pH dependence of the steady‐state kinetic parameters of two related substrates. N ‐benzoyl‐L‐arginine ethyl ester and N ‐benzoyl‐L‐arginine p ‐nitroanilide, indicates that two ionizing groups are present in the enzyme's active site, one has a p K (app) of 6.7–7.5, the other has a p K (app) of 8.7–9.6. These two enzyme groups control the binding of substrates, but the group with a p K around 9 does not appear to be essential for the catalysis. The assignment of these p K values is discussed in connection with the known features of the reaction mechanism of serine proteases. Attempts to obtain an insight into the catalytic mechanism from the kinetic analysis of the effects of added nucleophiles were unsuccessful. No ‘burst’ release of p ‐nitrophenol is observed from the hydrolysis of p ‐nitrophenyl‐ p′ ‐guanidobenzoate. Deuterium isotope effects for k cat / K m range from 1 for N ‐benzoyl‐L‐arginine ethyl ester to 1.5 for N ‐benzoyl‐L‐arginine amide. Furthermore, the kinetics of the protease‐II‐catalyzed hydrolysis of various synthetic substrates are characterized by relatively weak variations in the values of K m (app), but significant differences in the values of k cat are observed with closely related amide and ester substrates. This implies that K m may be related to K s , and that the step immediately following formation of the Michaelis enzyme‐substrate complex in the catalytic process, is rate‐determining. These findings do not agree with the acyl‐enzyme mechanism established for pancreatic serine hydrolases and α‐lytic protease.
Hydrolytic activities of isolated membrane fractions of Escherichia coli against chromogenic substrates, p-nitrophenyl ester and beta-naphthyl ester derivatives of N-substituted amino acids, were investigated by spectrophotometric and electrophoretic methods. Although detergents were absolutely necessary for the solubilization of enzymes, the amount of solubilized activities was increased by adding salt, such as NaCl or KCl. Two esterases were identified and separated by PAGE and by chromatography of the solubilized proteins in the presence of detergent. One hydrolyzed the alanine derivatives preferentially, whereas the other was mainly active on phenylalanine derivatives. Only the first was inactivated by diisopropyl fluorophosphate, a serine hydrolase inhibitor. Whereas the chymotrypsin-like enzyme was equally distributed between the inner and the outer membrane, the alanine activity was only detected in the inner membrane. They were both resistant to extraction with high salt concentrations, indicating their integral association with membranes. A study of the accessibility of these enzymes to their substrate in membrane vesicles with known polarity suggests that both alanine and phenylalanine activities are localized near the external surface of the cytoplasmic (inner) membrane. However, the phenylalanine activity (chymotrypsin-like enzyme) appears to be deeply buried inside the outer membrane. Because of its insensitivity to diisopropyl fluorophosphate, this last esterase seems to be distinct from the previously isolated periplasmic endopeptidase, protease I, which is also a chymotrypsin-like enzyme.
The recently isolated protease I of Escherichia coli leads to the partial degradation of RNA polymerase of E. coli and causes, in particular, the splitting of subunits ββ′ into fragments of approximately the same size as that of factor σ. This action is similar to that of papaïn or α‐chymotrypsin employed for short periods at much lower concentrations. The immunochemical study presented leads to the identification of the origin of the fragments obtained by limited proteolysis and shown the lack of cross reaction between subunits ββ′ and σ factor.
N-Acetyl-D-arginine linked to an agarose matrix has been used to purify protease II from Escherichia coli by affinity chromatography. The specific adsorption of protease II to this absorbent was achieved in 220 mM potassium phosphate buffer pH 7.6, and the enzyme was eluted with L-arginine. Enzyme preparations from cells harvested at late log phase have been resolved into two molecular species which differ in specific activity, kinetic constants and carbohydrate content. Both species appeared homogeneous by electrophoresis in conventional buffers and also in the presence of sodium dodecyl sulfate. Only one enzyme species was obtained by the same procedure using bacteria harvested at the middle of exponential growth.
A rapid purification procedure was developed for the isolation of caldesmon (CaD) from rabbit alveolar macrophage. The purified protein migrated with an apparent M(r) of 74,000 +/- 4000 on SDS-PAGE and cross-reacted with anti-gizzard CaD antibodies. A higher M(r) isoform was isolated from chicken gizzard. Their actin-binding parameters and effects on actomyosin-ATPase activity were investigated under identical experimental conditions. Electron microscope studies revealed that macrophage CaD was able to cross-link actin filaments into both networks and bundles. Compact F-actin bundles were predominantly or exclusively seen at cross-linker to actin molar ratios in the 1:20 to 1:10 range. Apparent K(a) at extrapolated saturation of the CaD-binding sites on F-actin was 1.2 x 10(6) M(-1) for macrophage CaD and 1.6 x 10(6) M(-1) for chicken gizzard CaD. CaD from either source was able to stimulate the actin-activated ATPase activity of macrophage myosin. Unexpectedly, chicken gizzard CaD also increased the ATPase activity of gizzard myosin. The degree of stimulation was approximately doubled in the presence of a large excess of Ca(2+)-calmodulin but was unaffected by the presence of macrophage tropomyosin. However, macrophage CaD did not behave as a Ca(2+)- and calmodulin-regulated actin-binding protein. These results, together with published data on other well-characterized actin bundling proteins, suggest that nonmuscle CaD could be essentially involved in the formation and organization of actin bundles at adhesion sites and cell surface projections. However, they afforded no evidence that the macrophage isoform might play a specific role in the Ca(2+)-dependent regulation of actin and myosin II interactions.
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTMacrophage .alpha.-actinin is not a calcium-modulated actin-binding proteinM. Pacaud and M. C. HarricaneCite this: Biochemistry 1993, 32, 1, 363–374Publication Date (Print):January 12, 1993Publication History Published online1 May 2002Published inissue 12 January 1993https://pubs.acs.org/doi/10.1021/bi00052a045https://doi.org/10.1021/bi00052a045research-articleACS PublicationsRequest reuse permissionsArticle Views65Altmetric-Citations14LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail Other access optionsGet e-Alertsclose Get e-Alerts
MI06Escherichia coli cells contain two membrane-bound esterases, one of which is localized in both inner and outer membranes and the other present only in the inner membrane (Pacaud, M. (1982) J. Bacteriol.149, in press).These two enzymes have been purified to homogeneity and shown to be capable of hydrolyzing a mixture of E. coli proteins.The inner membrane esterase has been called protease N ; the other esterase has been designated protease V. Protease N migrates as a single band on sodium dodecyl sulfate gels with an apparent Mr = 34,000.However, this enzyme displays an apparent M, = 660,000 on Ultrogel columns in the presence of Emulphogen BC-720.The large difference in size of the native enzyme appears to be due to its association with the detergent.The molecular weight of protease V could not be determined because of the interactions of this enzyme with detergents.Protease
ABSTRACT Under appropriate conditions macrophage cytosolic extracts can form a three-dimensional gel network of cross-linked actin filaments. These cytoplasmic gels are mainly composed of actin, filamin, alpha-actinin, and two new proteins of about 70000 and 55 000Mr (70 and 55 K). The behaviour of 70 K protein was found to be remarkably affected by Ca2+. Ca2+ treatment of isolated cytoplasmic gels led to the selective solubilization of the 70 K protein along with a 17 K polypeptide. Half-maximal recovery in the supernatant fraction was obtained from about 0·15μM free Ca2+. The cytoplasmic gel constituents solubilized in high ionic strength buffer were able to re-assemble into an insoluble actin network when returned to near physiological ionic conditions. However, the inclusion of micromolar Ca2+ prevented the re-association of 70 K protein with actin in these complexes. As compared to the 70 K protein, alpha-actinin was fully resistant to any variations in Ca2+ concentrations. On the other hand, purified 70 K protein displayed the property of assembling actin filaments into bundles at low Ca2+ concentrations (<0·15 μ M). However, the bundling activity decreased progressively at higher Ca2+, as detected by co-sedimentation and electron microscopy of the 70 K protein-actin mixtures. Half-maximal inhibition was observed at about 0·3 μM free Ca2+. Re-assembly of actin filaments into bundles occurred after chelation of Ca2+ by EGTA, indicating that the inhibitory effect of Ca2+ was reversible. Severing of actin filaments by 70 K protein was not observed in any of the solution conditions used. The Ca2+-dependent inhibition of the ability of 70 K protein to interact with actin networks resulted in a marked distortion of the overall organization of actin filaments, as revealed by thin-section electron microscopy of cytoplasmic gels formed in the presence and absence of Ca2+. Large zones of oriented bundles of filaments were replaced by a looser mesh. When the actin gel constituents were re-assembled in the presence of Ca2+ and exogenous gelsolin, it was also observed that the filament bundles (essentially generated by alpha-actinin) collapsed into dense aggregates. Furthermore, gelsolin did not significantly affect the ability of actin to re-combine with other proteins. The data presented here and previously led us to suspect that the Ca2+ control of the functional state of 70 K protein might be one of the prime factors in the triggering of rapid assembly and disassembly of microfilaments within macrophages.
We have previously demonstrated the existence of two types of endopeptidase in Escherichia coli. A purification procedure is described for one of these, designated protease II. It has been purified about 13,500-fold with a recovery of 24%. The isolated enzyme appears homogeneous by electrophoresis and gel filtration. Its molecular weight is estimated by three different methods to be about 58,000. Its optimal pH is around 8. Protease II activity is unaffected by chelating agents and sulfhydryl reagents. Amidase and proteolytic activities are stimulated by calcium ion, which decreases the enzyme stability. Like pancreatic trypsin, this endopeptidase catalyses the hydrolysis of alpha-amino-substituted lysine and arginine esters. It appears distinct from the previously isolated protease I, which is a chymotrypsin-like enzyme. The apparent Michaelis constant for hydrolysis of N-benzoyl-L-arginine ethyl ester is 4.7 X 10(-4) M. The esterase activity is inhibited by diisopryopylphosphorofluoridate (Ki(app) equals 2.7 X 10(-3) M) and tosyl lysine chloromethyl ketone (Ki(app) equals 1.8 X 10(-5) M), indicating that serine and histidine residues may be present in the active site. However, protease II is insensitive to phenylmethanesulfonyl fluoride and several natural trypsin inhibitors. Its amidase and esterase activities are competitively inhibited by free arginine and aromatic amidines. The proteolytic activity measured on axocasein is very low. In contrast to trypsin, protease II is without effect on native beta-galactosidase. It easily degrades aspartokinase I and III. Nevertheless both enzymes are resistant to proteolysis in the presence of their respective allosteric effectors. These results provide further evidence that such differences in protease susceptibility can be related to the conformational state of the substrate. The possible implication of structural changes in the mechanism of preferential proteolysis in vivo, is discussed.
Protease I, a periplasmic endopeptidase from Escherichia coli has been further purified by a modified procedure. While the purified protein consists of a single polypeptide chain of about 21000 daltons, its molecular weight in dilute salt solution was estimated to be near 43000. suggesting that the enzyme has a marked tendency to dimerize. It has only one disulphide bond and is very sensitive to urea. In agreement with previous evidence of a chymotrypsin‐like specificity, hydrolytic assays of various p‐nitrophenyl esters of N‐substituted amino acids showed that phenylalanine and tyrosine derivatives are the best substrates for the enzyme. The K m (app) for N‐benzoyloxycarbonyl‐L‐tyrosine‐p‐nitrophenyl ester at pH 7.5 in 100 mM sodium phosphate buffer at 25 oC was found to be 0.2 mM. In contrast to chymotrypsin, protease I is unable to hydrolyse N‐acetyl‐L‐phenylalanine ethyl ester and its tyrosine analogue. Moreover, the enzyme appears devoid of amidase activity and exhibits a low activity upon polypeptides. At 37 oC, it cleaves the carboxymethylated B‐chain of bovine insulin at four points: Phe 25 ‐Tyr 26 , Phe 24 ‐Phe 25 , Leu 15 ‐Tyr 16 and Ser 9 ‐His 10 . From a detailed study of peptides bonds hydrolyzed, it was concluded that protease I has a stringent requirement for both residues forming the scissile bond, and appears to possess an extended hydrophobic binding site.
'/%3e%3cpath d='m450 224.315-44.467 136.87 116.401-84.559h-143.87l116.403 84.559z' style='fill:none;stroke:%23006233;stroke-width:14.62993431;stroke-miterlimit:4;stroke-dasharray:none'/%3e%3c/svg%3e)