By Angela Anastasi and Alan J. Barrett
Pitrilysin is a peptidase so far known only from Escherichia coli, which was originally discovered as an activity hydrolyzing polypeptide fragments of /3-galactosidase, and termed protease III.1 The enzyme was also encountered as an insulin-degrading activity from the periplasmic space of E. coli, and referred to as protease Pi.2 The name pitrilysin (EC 188.8.131.52, formerly EC 184.108.40.206) is now recommended3; the name is entirely trivial but may call to mind the fact that the enzyme is the product of the ptr gene in E. coli. The lysin ending is commonly used in the names of metalloendopeptidases. Much of the interest in pitrilysin has stemmed from the fact that it is a member of a group of metallopeptidases described as family M16 in  in this volume. Other members of this group of proteins are eukaryotic endopeptidases, insulysin (formerly known as insulinase; see ), the two subunits of the mitochondrial processing peptidase (see ), and a paired-basic processing enzyme (see  in this volume).
Cheng and Zipser assayed pitrilysin in a complementation assay, in which the catalytic activity of incomplete molecules of /3-galactosidase was restored by fragments derived from the N terminus of the intact form of the enzyme.1 The activating fragments (termed auto a) served as substrates for pitrilysin, so that the auto a activity was destroyed, and correspondingly little /3-galactosidase activity was regenerated, when pitrilysin activity was high.
In the majority of subsequent work with the enzyme, radiolabeled insulin has been used as substrate.2 4 However, the insulin assay has the disadvantage that it does not allow continuous monitoring of enzymatic activity, which is not ideal for kinetic work. This disadvantage has been overcome
1 Y.-S. E. Cheng and D. Zipser, J. Biol. Chem. 254, 4698 (1979).
2 A. L. Goldberg, K. H. S. Swamy, C. H. Chung, and F. S. Larimore, this series, Vol. 80, p. 680.
3 Nomenclature Committee of the International Union of Biochemistry and Molecular Biology, "Enzyme Nomenclature 1992." Academic Press, Orlando, Florida, 1992.
4 C. C. Dykstra and S. R. Kushner, J. Bacteriol. 163, 1055 (1985).
by the development of a fluorimetric assay.5 Both insulin-degrading and fluorimetric assays are described below.
To quantify pitrilysin activity by use of insulin as the substrate, reaction mixtures are set up containing 5-15 /xg of insulin, a trace amount of 125I-labeled insulin [about 12,000-15,000 counts/min (cpm)], and the enzyme, in a final volume of 0.5 ml of assay buffer (50 mM Tris-HCl, pH 7.5, containing 0.05% Brij 35). The radiolabeled insulin is obtainable from Lise Screen Ltd. (Watford, UK). After incubation for 1 hr at 37°, the reaction is stopped by the addition of 60 /x 1 of 100% (w/v) trichloroacetic acid solution and 40 ¡jA of bovine serum albumin solution (30 mg/ml) as carrier, with good mixing. The assay tubes are kept in ice for 30 min, and, after centrifugation, 0.4-ml samples of supernatant containing acid-soluble products are removed for gamma counting. The response of the assay is linear to about 35% degradation of the substrate. Another form of assay for insulin degradation, the plate assay, is described in ,
As is described in  in this volume, a quenched fluorescence substrate is one in which a short sequence of amino acids containing the scissile peptide bond links a potentially fluorescent group to a group that acts as an internal quencher of the fluorescence. Cleavage of the scissile bond results in the appearance of fluorescence, which is monitored.
The quenched fluorescence substrate for pitrilysin and insulysin, QF27, was designed on the basis of the structure of fragment 16-28 of the vasoactive intestinal peptide (VIP).5'6 The potential fluorophore is a derivative of 7-methoxycoumarin, and the quencher is a 2,4-dinitrophenyl group. Thus, the structure of QF27 is Mca-Nle-Ala-Val-Lys-Lys-Tyr-Leu-Asn-Ser-Lys(Dnp)-Leu-Asp-D-Lys. The enhancement of fluorescence resulting from hydrolysis is approximately 20-fold, and the sensitivity of the assay is simlar to that with insulin as substrate.
Reagents, The syntheses of QF27 and the reference standard, A^-acetyl-Ai6-Mca-L-Lys, have been described,5 and the compounds are commercially available from Calbiochem NovaBiochem (Nottingham, UK).
5 A. Anastasi, C. G. Knight, and A. J. Barrett, Biochem. J. 290, 601 (1993).
6 Abbreviations: Abz, o-aminobenzoyl; Dpa, dl-2-amino-3-(7-methoxy-4-coumaryl)propio-nic acid; Lys(Dnp), 7V6-(2',4'-dinitrophenyl)lysine; Mca, (7-methoxycoumarin-4-yl)acetyl; Nle, norleucine; VIP, vasoactive intestinal peptide. For amino acids, the standard three-letter or one-letter codes are used, except that k is d-lysine and in the oxidized B chain of insulin C is cysteic acid.
Substrate stock solution: QF27 is dissolved in dimethyl sulfoxide at
2.5 mM, and stored at 4° Standard stock solution: Af2-Ac-Af6-Mca-L-Lys is dissolved in dimethyl sulfoxide at 5 mM, and stored at 4° Assay buffer: 50 mM Tris-HCl, pH 7.5, containing 0.05% Brij 35
Procedure. Fluorimetric assays with QF27 are conveniently made in a Perkin-Elmer (Norwalk, CT) spectrofluorimeter controlled by an IBM-compatible computer running the FLUSYS software.7 Other fluorimeters can be used, but software for the collection of kinetic data is highly desirable. The sample is contained in a quartz cuvette accurately thermostated at 37°, and is stirred continuously.
Excitation and emission wavelengths are set to 328 and 393 nm, respectively. The fluorimeter is zeroed against substrate in assay buffer and calibrated to read 1000 arbitrary units of fluorescence for a solution containing both reference standard at a concentration 10% that of substrate and working strength substrate. In routine assays, the substrate concentration is 10 /jlM (10 ¡A of the stock solution added), and the reference standard concentration is 1 (jlM (50 ¡A of 1/100 diluted stock solution). Because the substrate causes some quenching of the fluorescence of the product and standard, it is essential that the instrument is restandardized whenever the substrate concentration is changed, as in the determination of Km values (see , this volume).
Continuous assays are performed in a total volume of 2.5 ml of assay buffer with 10 ¡A of substrate stock solution (final 10 ¡jlM), and the reaction is started by the addition of about 0.05 milliunit (mU) of enzyme, 1 mU of activity being defined as that hydrolyzing 1 nmol of substrate per minute at 37°. The fluorescence is recorded, and the rate of increase in fluorescence over a suitable time interval (5-30 min) is calculated by linear regression analysis of the values against time.
Assays with QF27 have been found to show accurately measurable rates of hydrolysis, linear with time, for enzyme concentrations in the range 0.01-0.1 mU/ml (17-170 ng/ml). QF27 is not susceptible to hydrolysis only by pitrilysin, and it is also the first described synthetic substrate for insu-lysin.5
Pitrilysin has been purified from wild-type E. coli.2A However, since the preparation of plasmids containing the gene for pitrilysin,8'9 the preferred
7N. D. Rawlings and A. J. Barrett, Comput. Appl. Biosci. 6, 118 (1990).
8 C. C. Dykstra, D. Prasher, and S. R. Kushner, J. Bacteriol. 157, 21 (1984).
9 P. W. Finch, R. E. Wilson, K. Brown, I. D. Hickson, and P. T. Emmerson, Nucleic Acids Res. 14, 7695 (1986).
starting material has become one of the E. coli strains that overexpress the enzyme. Anastasi et al.5 compared two strains of E. coli: the strain PE004 that harbors the plasmid pPF307 (given by Professor P. T. Emmerson, Department of Biochemistry, University of Newcastle upon Tyne, Newcastle NE1 7RU, UK) and strain SK7814 containing plasmid pCDK35 (given by Dr. S. R. Kushner, Department of Genetics, Life Sciences Building, University of Georgia, Athens, GA 30602), both of which were grown in Luria-Bertani medium.10 For the PE004 strain, the medium was supplemented with 50 jUg/ml kanamycin, and the culture was grown initially at 30°; at Z)660 = 0.05-0.1, the temperature was shifted to 37° for 4-5 hr, for temperature induction.11 The SK7814 strain was grown in medium supplemented with 170 fig of chloramphenicol/ml for 5-6 hr to a D660 value of approximately 2.0.
Pitrilysin was identifiable in both cytoplasmic and periplasmic fractions of the overexpressing cells by assay and by immunoblot, two-thirds of the activity being present in the cytoplasmic fraction, but the specific activity of the periplasmic fraction was 6-fold greater; that fraction was therefore preferred as the source for purification. The periplasmic extract of strain PE004 had an initial specific activity one-third that of strain SK7814, and accordingly the latter strain was used for further work.
Approximately 12 g (wet weight) of SK7814 cells grown as above is harvested from culture (3 liters) by centrifugation at 5000 g for 30 min, then washed with 10 mM Tris-HCl, pH 7.5, containing 0.1 M NaCl, followed by 10 mM Tris-HCl, pH 7.5. All buffers used subsequently in the purification of pitrilysin contain 0.05% Brij 35. For the extraction of the periplasmic proteins, the pellet is mixed with 20 ml of chloroform, allowed to stand for 15 min at room temperature, and suspended in 180 ml of 10 mM Tris-HCl, pH 7.5.12 The suspension is centrifuged for 20 min at 6500g. The supernatant is removed, recentrifuged at 8000 g for 20 min, dialyzed against 20 mM Tris-HCl, pH 8.0, and run on a column (1.5 X 11.5 cm, 20 ml) of Q-Sepharose (Sigma, St. Louis, MO) equilibrated in the same buffer. The column is eluted with a gradient to 250 mM NaCl in the Tris buffer at a flow rate of 90 ml/hr. The fractions (2 ml) are assayed, activity being found at about 100 mM NaCl. Active fractions are combined and dialyzed against 20 mM potassium phosphate buffer, pH 7.0. The sample is run on a column (1x3 cm, 2.3 ml) of hydroxyapatite (Bio-Rad, Richmond, CA) and eluted
10 J. Sambrook, E. F. Fritsch, and T. Maniatis, "Molecular Cloning: A Laboratory Manual."
Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1989.
11 S. Yasuda and T. Takagi, J. Bacteriol. 154, 1153 (1983).
12 G. F.-L. Ames, C. Prody, and S. Kustu, J. Bacteriol. 160, 1181 (1984).
with a gradient (60 ml) of 20-250 mM potassium phosphate buffer, pH 7.0. Active fractions are combined, dialyzed against 25 mili Bis-Tris-HCl, pH 6.3, and run on the Pharmacia (Piscataway, NJ) FPLC (fast protein liquid chromatography) Mono P (HR 5/20) column at a flow rate of 0.5 ml/min. The column is preequilibrated with the Bis-Tris buffer and is developed with Polybuffer 74 diluted 10-fold (pH 4.0). Active fractions are combined, dialyzed into 50 mili Tris-HCl, pH 7.5, and supplemented with glycerol to 40% (v/v) before the preparation is stored at -20°.
Typically, the final preparation of pitrilysin is 145-fold purified, in 4% yield.5 The protein appears as a single band in sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis, with mobility slightly higher than that of /8-galactosidase (116 kDa), consist with the molecular mass of 107 kDa calculated from the amino acid sequence.
The ptr gene encoding pitrilysin was cloned as a part of a 19-kb BamHl fragment and was found to be physically located between the recB and recC genes, whose products are subunits of exonuclease V. When the genes were amplified for the overexpression of exonuclease V, an increase in pitrilysin activity was obtained incidentally.8 The complete nucleotide sequence of the ptr gene9 shows that pitrilysin is a member of family M16 of metallopeptidases (, this volume).
Pitrilysin contains zinc,13 and site-directed mutagenesis has implicated the two histidine residues in the sequence -His-Tyr-Leu-Glu-His- in the binding of zinc. Moreover, both of the His residues and the Glu in the same sequence are required for the catalytic activity of the enzyme.14
The pH optimum for hydrolysis of QF27 by pitrilysin is about pH 7.5. The enzyme shows no activation by thiol compounds (unlike the homologous insulysin; see ), merely being inhibited at high concentrations, as is normal for metallopeptidases. The metal dependence of pitrilysin is described below, but no addition of divalent cations to the solution is normally required for full activity.
Little has been known about the specificity of pitrilysin for cleavage of peptide bonds, although Cheng and Zipser showed cleavage of insulin B
13 L. Ding, A. B. Becker, A. Suzuki, and R. A. Roth, J. Biol. Chem. 267, 2414 (1992).
14 A. B. Becker and R. A. Roth, Proc. Natl. Acad. Sci. U.S.A. 89, 3835 (1992).
chain at the -Tyr16+Leu17-bond and more slowly at -Phe24+Tyr25-.' The points of cleavage of a number of oligopeptides by pitrilysin were determined by use of high-performance liquid chromatography analysis, by Anastasi et al.,5 and the results are summarized in Fig. 1. With each of the substrates shown, all susceptible bonds were cleaved at similar rates so that the relative heights of HPLC peaks for the products did not change appreciably with time of incubation. There was no further degradation of the products. In that small set of scissile bonds, no clear preference for amino acids around the bond hydrolyzed by pitrilysin is apparent, although the residue in the PI position is commonly Leu or Tyr.
Kinetics. Kinetic parameters for the hydrolysis of the synthetic substrate QF27 by pitrilysin are Km = 7.7 ± 0.75 ¡¿M and kCM = 2.9 sec"1 (taking an Mr of 107,000, and assuming that a preparation of 580 mU/mg was fully active). The value of kcJKm was thus 3.8 X 105 M"1 sec"1. It should be noted that two bonds are cleaved in QF27 at approximately equal rates, so the rate of cleavage of each would be about one-half the total rate of hydrolysis.
QF27 has also been found to be a substrate for Drosophila insulysin, which forms products identical to those resulting from the action of pitrilysin. The Km of Drosophila insulysin for QF27 is 18 ± 3.3 ¡¿M.
Rates of hydrolysis of peptides were determined by HPLC analysis after timed incubations and expressed relative to the rate for insulin B chain (Table I). The rate of disappearance of intact insulin B chain was the greatest in the series, despite the fact that most of the peptides were cleaved at two or more bonds. Substrates of pitrilysin so far discovered range in
Insulin B chain (oxidized)
Was this article helpful?