Synthesis and Hydrolysis by Cysteine and Serine Proteases of Short Internally Quenched Fluorogenic Peptides

Analytical Biochemistry 293, 71–77 (2001) doi:10.1006/abio.2001.5115, available online at http://www.idealibrary.com on

Synthesis and Hydrolysis by Cysteine and Serine Proteases of Short Internally Quenched Fluorogenic Peptides Robson L. Melo, Lira C. Alves, Elaine Del Nery, Luiz Juliano, 1 and Maria A. Juliano Department of Biophysics, Escola Paulista de Medicina, Universidade Federal de Sa˜o Paulo, Rua Treˆs de Maio 100, 04044-020 Sa˜o Paulo, Brazil

Received November 8, 2000; published online April 30, 2001

We developed sensitive substrates for cysteine proteases and specific substrates for serine proteases based on short internally quenched fluorescent peptides, AbzF-R-X-EDDnp, where Abz (ortho-aminobenzoic acid) is the fluorescent donor, EDDnp [N-(ethylenediamine)-2,4dinitrophenyl amide] is the fluorescent quencher, and X are natural amino acids. This series of peptides is compared to the commercially available Z-F-R-MCA, where Abz and X replace carbobenzoxy (Z) and methyl-7-aminocoumarin amide (MCA), respectively; and EDDnp can be considered a P 2ⴕ residue. Whereas MCA is the fluorescent probe and cannot be modified, in the series Abz-FR-X-EDDnp the amino acids X give the choice of matching the specificity of the S 1ⴕ enzyme subsite, increasing the substrate specificity for a particular protease. All Abz-F-R-X-EDDnp synthesized peptides (for X ⴝ Phe, Leu, Ile, Ala, Pro, Gln, Ser, Lys, and Arg) were assayed with papain, human cathepsin L and B, trypsin, human plasma, and tissue kallikrein. Abz-F-R-L-EDDnp was the best substrate for papain and Abz-F-R-R-EDDnp or AbzF-R-A-EDDnp was the more susceptible to cathepsin L. Abz-F-R-L-EDDnp was able to detect papain in the range of 1 to 15 pM. Human plasma kallikrein hydrolyzed AbzF-R-R-EDDnp with significant efficiency (k cat/K m ⴝ 1833 mM ⴚ1 s ⴚ1) and tissue kallikrein was very selective, hydrolyzing only the peptides Abz-F-R-A-EDDnp (k cat/K m ⴝ 2852 mM ⴚ1 s ⴚ1) and Abz-F-R-S-EDDnp (k cat/K m ⴝ 4643 mM ⴚ1 s ⴚ1). All Abz-F-R-X-EDDnp peptides were resistant to hydrolysis by thrombin and activated factor X. © 2001

Z-F-R-MCA is the most widely used substrate for cysteine proteases of the papain family, although it is also hydrolyzed by serine proteases of the trypsin family. MCA 2 is methyl-7-aminocoumarin amide, a fluorescent tag that changes its fluorescence properties if attached to a carboxyl group through an amide bond or if free as 7-aminomethylcoumarin (AMC) (1, 2). PeptidylMCA substrates, such as Z-F-R-MCA, interact with nonprime sites of proteases, except the MCA group that occupies the S⬘1 subsite [see Schechter and Berger’s nomenclature of protease subsites and substrate residues (3)]. This kind of short substrate is useful for evaluating the catalytic activity of proteases and has been used to map their specificity at nonprime sites. The use of internally quenched fluorescent peptides, on the other hand, gives information about both the nonprimed and the primed subsites, making use of the fluorescent quenched peptide which is ideal for the investigation of the substrate specificity of proteases (4). In these peptides the energy transfer occurs intramolecularly from the donor to the acceptor groups, which are separated by a peptide chain. This energy transference and the consequent quenching of fluorescence is highly dependent on the distance between the donor and the acceptor groups. As a consequence of separation of the two groups, the quenching effect disappears upon hydrolysis of the peptide. The sensitivity of this kind of substrates to detect protease activity depends on the increase of the fluorescence after their cleavage; therefore, the hydrolysis of short internally

Academic Press

Key Words: cathepsin B; cathepsin L; papain; trypsin; kallikrein.

1 To whom correspondence [email protected].

should

0003-2697/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

be

addressed.

E-mail:

2 Abbreviations used: Z, benzyloxycarbonyl; MCA, methyl-7-aminocoumarin amide; Abz, ortho-amino benzoic acid; EDDnp, N-(2,4,dinitrophenyl) ethylenediamine; AMC, 7-aminomethylcoumarin; TFA, trifluoroacetic acid; BOP, benzotriazole-1-yl-oxy-tris(dimethylamino)phosphonium hexafluorophosphate; MALDI-TOF, matrix assisted laser desorption/ionization–time of flight; TPCK, N-tosyl-L-phenylalanine chloromethyl ketone.

71

72

MELO ET AL.

FIG. 1.

Schematic representation of the structure and subsite occupancy of the substrates Abz-F-R-X-EDDnp and Z-F-R-MCA.

quenched fluorescent peptides will result in higher fluorescent signals [for details see the reviews in (5, 6)]. In order to develop sensitive substrates for proteases with the properties of internally quenched fluorescent peptides we synthesized a series of peptides derived from Abz-F-R-X-EDDnp, where Abz (ortho-aminobenzoic acid) is the donor, EDDnp [N-(ethylenediamine)2,4-dinitrophenyl amide] is the quencher group, and X are natural amino acids. The comparison of this series of peptides with Z-F-R-MCA is shown in Fig. 1. Abz is substituted for the Z group, X is substituted for MCA, and EDDnp can be considered a P⬘2 residue. Whereas MCA is a fluorescent group and cannot be changed, X in Abz-F-R-X-EDDnp substrates are amino acids that give the choice of matching a better specificity for the S⬘1 enzyme subsite, improving the substrate susceptibility for a particular protease. All obtained peptides (for X ⫽ Phe, Leu, Ile, Ala, Pro, Gln, Ser, Lys, and Arg) were synthesized and assayed as substrates for papain, human cathepsin L and B, trypsin, human plasma, tissue kallikrein, thrombin, and activated factor X. MATERIALS AND METHODS

All solvents were appropriately distilled before use and DMF assayed for free amine residues (7). Fmocand Boc-amino acids and Z-F-R-MCA were purchased from Calbiochem–Nova Biochem. Fmoc-succinimide, di-tert-butyl dicarbonate (Boc 2O), trifluoroacetic acid (TFA), anisol, and 1,2-ethanedithiol were from Fluka. Benzotriazole-1-yl-oxy-tris(dimethylamino)phosphonium hexafluorophosphate (BOP) was from Watanabe Chemical Co. Synthesis of N ␣-Boc-Aminoacyl-EDDnp EDDnp was obtained in hydrochloride form as follows: 1.6 mol of ethylenediamine was dissolved in 360 ml of dry dioxane and cooled in an ice bath and under nitrogen atmosphere and 0.08 mol of 1-fluoro-2,4-dinitrobenzene, dissolved in 120 ml of dry dioxane, was

dropped under stirring during 1 h. The mixture was then kept under stirring during 3 h at room temperature. The solvent was then removed by evaporation under reduced pressure and water was added. The precipitate was collected by filtration and dissolved in EtOH and the pH was adjusted to 2 with concentrated HCl. The precipitate was collected and washed with chilled EtOH. Yield was 94%, and mp was 268 –270°C. N ␣-Boc-X-(CO ␣-EDDnp), where X stands for Phe, Leu, Ile, Ala, Pro, Ser(OBzl), Lys(Z), and Arg (Tos), was obtained by mixed anhydride procedure using isobutyl chlorocarbonate as previously described (8). All the amino acid derivatives were purified in silica gel chromatography using chloroform–methanol mixtures as solvent. The yield, melting point, and thin layer chromatography migration of the obtained amino acid derivatives are presented in Table 1. Peptide Synthesis Boc-Arg(Tos)-OH, Boc-Phe, and Boc-Abz were sequentially coupled to N ␣-deprotected aminoacylEDDnp using mixed anhydride or BOP as coupling procedures in solution. The final peptides were deprotected by anhydrous HF and purified by semi-preparative HPLC using an Econosil C-18 column (10 ␮m, 22.5 ⫻ 250 mm) with the solvent systems: (A) TFA/H 2O (1:1000) and (B) TFA/ MeCN /H 2O (1:900:100). The column was eluted at a flow rate of 5 ml/min with a 10 (or 30) to 50 (or 60)% gradient of solvent B over 30 or 45 min. Analytical HPLC was performed using a binary HPLC system from Shimadzu equipped with a SPD10AV Shimadzu UV-vis detector and a Shimadzu RF535 fluorescence detector, coupled to an Ultrasphere C-18 column (5 ␮m, 4.6 ⫻ 150 mm) which was eluted with solvent systems (A1) H 3PO 4/H 2O (1:1000) and (B1) MeCN/H 2O/H 3PO 4 (900:100:1) at a flow rate of 1.7 ml/min and a 10 – 80% gradient of B1 over 15 min. The HPLC column eluates were monitored by their absorbance at 220 nm and by fluorescence emission at 420 nm following excitation at 320 nm. The purity of ob-

73

FLUOROGENIC SUBSTRATES FOR CYSTEINE AND SERINE PROTEASES TABLE 1

Characterization Data of Amino Acid Derivatives [N ␣-Boc-X-(CO ␣-EDDnp)] and Molecular Weights of the Peptides Abz-F-R-X-EDDnp, Obtained by MALDI-TOF Mass Spectrometry Abz-F-R-X-EDDnp molecular weight X

Rf (A)

Rf (B)

Rf (C)

mp (°C)

Yeld (%)

X

[H ⫹ 1] calculated

[H ⫹ 1] found

F L I A P S(OBz) K(2ClZ) R(Tos)

0.71 0.90 0.70 0.72 0.60 0.68 0.56 0.88

0.76 0.84 0.77 0.69 0.55 0.71 0.66 0.93

0.86 0.80 0.90 0.94 0.74 0.78 0.90 0.78

181–184 152–155 174–177 153–156 46–51 110–114 110–112 83–88

76 89 74 65 90 89 72 91

F L I A P S K R

795 761 761 719 745 735 776 804

796 762 762 720 746 736 777 805

Note. Thin layer chromatography conditions: solvent (A) ⫽ chloroform:methanol (93:7), solvent (B) ⫽ ethyl acetate:ethanol (4:1), and solvent (C) ⫽ chloroform:methanol:acetic acid (17:2:1). Molecular weight of the final products was obtained by MALDI-TOF mass spectrometry. The peptide Abz-F-R-Q-EDDnp was synthesized using the solid phase method by coupling the ␥-carboxyl group of Glu at the 2,4-dimethoxybenzylphenoxyacetic acid linker attached to the TG-resin from Novabiochem, which transforms Glu to Gln during the final deprotection step with TFA. Abz-F-R-Q-EDDnp molecular weight: calculated, 776; obtained, 777.

tained peptides was checked by amino acid sequencing, performed with a Shimadzu Sequencer Model No. PPSQ-23, and by MALDI-TOF in the reflectron mode (TofSpec-E from Micromass, Manchester, UK). The obtained molecular weights of the final peptides are shown in Table 1. The purity of the obtained peptides was higher than 95%, as determined by HPLC. Enzymes Human liver cathepsin B and human thrombin were purchased from Calbiochem/Novabiochem (La Jolla, CA). Human cathepsin L and papain were obtained according to Refs. (9, 10), respectively. The molar concentrations of the enzyme solutions were determined by active site titration with E-64 according to Ref. (11). Human plasma kallikrein was obtained and titrated as previously described (12–14). ␤-Trypsin was purified as described elsewhere (15) from a twice-crystallized bovine trypsin from Biobras Co. (Montes Claros, Minas Gerais, Brazil), previously treated with TPCK, and the operational molarities were determined by active site titration (16). Homogeneous preparations of human tissue kallikrein were obtained according to Shimamoto et al. (17) and kindly provided by Dr. J. Chao of the Medical University of South Carolina (Charleston, SC). Activated factor X was kindly provided by Dr. Bernard Le Bonniec (INSERM, U428, Universite´ Paris V, Faculte de Pharmacie, France). Fluorometric Enzyme Assay Hydrolysis of the internally quenched fluorogenic peptides was followed by fluorescence measurement at ␭ em ⫽ 420 nm and ␭ ex ⫽ 320 nm (emission and excitation wavelengths for Abz) in a Hitachi F-2000 spec-

trofluorometer. The concentrated stock solutions of the substrates (1 mg/ mL, in water–DMF 1:1) were diluted 10 to 50 times with the buffer for each specific enzyme and the final concentrations of the substrates were obtained by colorimetric determination of 2,4-dinitrophenyl group (extinction coefficient at 365 nm was 17,300 M ⫺1cm ⫺1). Fluorescence variations were converted into amount of hydrolyzed substrates by standard curves obtained by the fluorescence measurement of well-defined concentrations of each substrate after complete hydrolysis by papain or trypsin. Hydrolysis of Z-F-R-MCA was followed at ␭ em ⫽ 460 nm and ␭ ex ⫽ 380 nm (emission and excitation wavelengths for AMC), and the spectrofluorometer was calibrated with standard solutions of AMC. A 1-cm-pathlength cuvette containing 1 ml of the substrate solution was placed in the thermostatted cell compartment for 5 min. The enzyme solution was added and the increase of fluorescence with time was continuously recorded for 5 min. The slope was converted into moles of substrate hydrolyzed per minute based on the curves of fluorescence for standard solutions of complete hydrolyzed substrates. The enzyme concentrations for initial rate determinations were chosen to hydrolyze less than 5% of the initial substrate concentration. The kinetic parameters were calculated according to Wilkinson (18) as well as by the Eadie–Hofstee plots. The standard errors for K m and k cat determinations were less than 10%. The cleavage points of each substrate by each protease were determined by isolation of the fragments and their structure was determined by peptide sequencing. The kinetic parameters of hydrolysis were determined in concentrations of approximately ⫾1 K m for each substrate at 37°C, using the following buffer systems and enzyme concentrations: 0.1 M sodium phosphate, 5

74

MELO ET AL. TABLE 2

Fluorescence Increase after Total Hydrolysis of the Peptides Abz-F-R-X-EDDnp Peptides

F t/F i

Abz-F-R-F-EDDnp Abz-F-R-L-EDDnp Abz-F-R-I-EDDnp Abz-F-R-A-EDDnp Abz-F-R-Q-EDDnp Abz-F-R-S-EDDnp Abz-F-R-K-EDDnp Abz-F-R-R-EDDnp

37 19 57 119 17 64 41 26

Note. The fluorescence of the peptides (F i) was measured at ␭ em ⫽ 420 nm and ␭ ex ⫽ 320 nm in 0.1 M sodium phosphate, 5 mM EDTA, 0.25 mM NaCl, pH 6.8. Papain was added (5 nM) and the fluorescence was measured when it stabilized (F t).

mM EDTA, 0.25 M NaCl, pH 6.8 (papain) or 6.0 (cathepsin B and L), with enzyme concentration 0.1 to 15 nM; for trypsin, 0.1 M Tris, 10 mM CaCl 2, pH 8.0; for human plasma kallikrein, 50 mm Tris, 1 mM EDTA, pH 7.5, with enzyme concentration 0.1 to 5 nM; and for tissue kallikrein 50 mM Tris–HCl, pH 9.0, containing 1 mM EDTA. RESULTS AND DISCUSSION

The relations (F t/F i) of the fluorescence of each peptide solution after (F t) and before (F i) complete enzymatic hydrolysis are shown in Table 2. The fluores-

cence of each substrate after complete hydrolysis was significantly higher than that of the intact substrate. In the series Abz-F-R-X-EDDnp, the higher increases of fluorescence after complete hydrolysis were observed with peptides containing the amino acids Ala and Ser at the X position, which have smaller side chains. These peptides possibly are more flexible, which could result in a shorter average distance between Abz and EDDnp, consequently increasing the quenching effect. Table 3 shows the kinetic parameters for the hydrolysis of Z-F-R-MCA and of peptides from the series Abz-F-R-X-EDDnp by papain and human cathepsin L and B. The obtained values of k cat/K m for hydrolysis of Z-F-R-MCA were of the same order of magnitude as those previously reported for the three proteases (19, 20). All peptides generated from the series Abz-F-R-XEDDnp were hydrolyzed with higher k cat/K m values by papain and cathepsin L compared to hydrolysis of Z-FR-MCA. This occurs due to the lower K m values for the hydrolysis of Abz-F-R-X-EDDnp peptides, except with Abz-F-R-P-EDDnp which was resistant to hydrolysis by all assayed enzymes, due to the presence of Pro in the amino site of susceptible peptide bond. Only a very specific protease hydrolyzes at the Pro carboxyl group, for instance, the prolyl oligopeptidase (21). The k cat/K m values obtained with cathepsin L were higher than those for papain and cathepsin B, confirming that cathepsin L is the most potent cysteine protease of the papain family (11, 22). The peptide Abz-F-R-R-EDDnp was the best substrate for cathepsin L, indicating that

TABLE 3

Kinetic Parameters for Hydrolysis of Abz-F-R-X-EDDnp and Z-F-R-MCA by Papain, Human Cathepsin L, and Human Cathepsin B Papain

Substrates

K m (␮M)

k cat (s ⫺1)

Cathepsin L k cat/K m (mM ⫺1 s ⫺1)

K m (␮M)

k cat (s ⫺1)

Cathepsin B k cat/K m (mM ⫺1 s ⫺1)

K m (␮M)

k cat (s ⫺1)

k cat/K m (mM ⫺1 s ⫺1)

23 ⫾ 2

76 ⫾ 7

3260

29,333 5273 20,769 59,167

3.5 ⫾ 0.2 3.1 ⫾ 0.2 11 ⫾ 1 4.3 ⫾ 0.2

828 806 118 109

23,000 14,615 56,363 63,000

14 ⫾ 1 6.1 ⫾ 0.5

2.9 ⫾ 0.1 2.5 ⫾ 0.1 1.3 ⫾ 0.1 0.47 ⫾ 0.03 NDH 0.28 ⫾ 0.01 0.45 ⫾ 0.02 NDH NDH

Z-F-R-MCA 10 ⫾ 1

23 ⫾ 2

2320

X F L I A P Q S K R

2.0 ⫾ 0.1

10.3 ⫾ 0.9

5150

Abz-F-R-X-EDDnp 1.7 ⫾ 0.3 0.54 ⫾ 0.05 1.06 ⫾ 0.08 1.03 ⫾ 0.06 1.7 ⫾ 0.1 1.3 ⫾ 0.1 2.8 ⫾ 0.1 4.0 ⫾ 0.2

10 ⫾ 0.6 29.7 ⫾ 0.9 7.3 ⫾ 0.1 28.9 ⫾ 2.3 NDH 28 ⫾ 2 31 ⫾ 2 35.7 ⫾ 0.8 30 ⫾ 1

5882 55,000 6887 28,058

0.15 ⫾ 0.02 1.1 ⫾ 0.1 0.26 ⫾ 0.02 0.12 ⫾ 0.01

16,471 23,847 12,750 7500

0.30 ⫾ 0.03 0.26 ⫾ 0.02 0.11 ⫾ 0.01 0.10 ⫾ 0.01

4.4 ⫾ 0.1 5.8 ⫾ 0.1 5.4 ⫾ 0.2 7.1 ⫾ 0.2 NDH 6.9 ⫾ 0.3 3.8 ⫾ 0.1 6.2 ⫾ 0.1 6.3 ⫾ 0.1

20 74

Note. Conditions: 0.1 M sodium phosphate, 5 mM EDTA, 0.25 mM NaCl, pH 6.8 (papain) or 6.0 (cathepsin B and L). Temperature: 37°C. Enzyme concentration: 0.1 to 15 nM. Substrate concentrations used were in the interval value of ⫾1 K m. NDH, no detected hydrolysis till [E] ⫽ 15 nM.

FLUOROGENIC SUBSTRATES FOR CYSTEINE AND SERINE PROTEASES

the S⬘1 subsite accepts basic residues quite well. According to this observation, a high k cat/K m value was also obtained with the hydrolysis of Abz-F-R-K-EDDnp by cathepsin L. On the other hand, the substrate Abz-FR-A-EDDnp was also hydrolyzed with a k cat/K m value near that of peptide Abz-F-R-R-EDDnp, indicating that the S⬘1 subsite has a broad specificity, accepting amino acids with large and positively charged side chains as well as small ones such as Ala. However, this subsite discriminates the isomeric forms of amino acid side chains such as in Ile and Leu, since Abz-F-R-L-EDDnp is hydrolyzed with a lower k cat/K m compared to Abz-FR-I-EDDnp in the series. The substrates with Ser and Gln have intermediate k cat/K m values. An analysis of the S⬘1 specificity of cathepsin L was previously reported using as substrates a series of peptides derived from Dns-F-R-X-W-A-OH (23), and the best substrates were in the order Ser ⬎ Ala ⬎ Lys. This partial disagreement indicates that the residues at other positions than that P⬘1 interfere with its fitting at the S⬘1 enzyme subsite. Papain hydrolyzed the substrate Abz-F-R-L-EDDnp with a k cat/K m (55,000 mM ⫺1 s ⫺1) value that is almost two times higher than the best substrate so far described for this enzyme, Abz-Q-V-V-A-G-A-EDDnp (k cat/K m ⫽ 31,000), which was based on the cystatin inhibitory site (19). The second best substrates were Abz-F-R-A-EDDnp and Abz-F-R-S-EDDnp, which were followed by the peptides containing Gln and Lys. The worst substrates of this series for papain were those containing Phe, Ile, and Arg. This order of hydrolytic preference with variations at P⬘1 is very parallel to that described for the substrates derived from Dns-F-R-XW-A-OH (23), except that Dns-F-R-F-W-A-OH is the second best of this series. Moreover, the preference of papain for Leu at the P⬘1 position was also observed in experiments based on nucleophile partitioning (24). Although papain and cathepsin L present high structural similarity, the high preference of the latter S⬘1 subsite for basic residues (Arg and Lys) and the former for Leu are two distinguishing hydrolytic activities observed with the substrates Abz-F-R-X-EDDnp that differentiate these enzymes. All of the internally quenched fluorescent peptides were poor substrates for cathepsin B compared to Z-FR-MCA, mainly due to low k cat values. Similar resistance to hydrolysis of Abz-F-R-X-EDDnp peptides was previously described for Abz-AFRSAAQ-EDDpn (25) and for peptides derived from the series AbzAFRSXAQ-EDDpn (22). Based on crystallographic and mutational analysis (22, 25, 26), the low endopeptidase activity of cathepsin B is attributed to the presence of an occluding loop in the S⬘ subsite of the enzyme. Besides this limitation for hydrolysis of Abz-F-R-XEDDnp peptides by cathepsin B, the data presented in Table 3 demonstrate the clear preference of S⬘1 for

75

hydrophobic amino acids. This preference was also observed with the peptides derived from Dns-F-R-X-WA-OH (23), although they were hydrolyzed at two sites (R–X and X–W bonds) due to the carboxydipeptidase activity of cathepsin B. Table 4 shows the kinetic parameters for the hydrolysis of Z-F-R-MCA and Abz-F-R-X-EDDnp peptides by trypsin and human plasma kallikrein. These serine proteases are less efficient for the hydrolysis of these peptides than papain and cathepsin L. Trypsin has a preference at P⬘1 for Ala and Ser, followed by Arg. The peptides containing amino acids with large hydrophobic side chains at this position are quite resistant, particularly Abz-F-R-F-EDDnp, which was not hydrolyzed till the trypsin concentration of 5 nM. This observation justifies the poor hydrolysis of Z-F-R-MCA, since the large and hydrophobic MCA group is located at P⬘1. Human plasma kallikrein is more selective and AbzF-R-R-EDDnp was the best substrate we found for this protease. Very little information is available in the literature related to the subsite specificity of plasma kallikrein; however, the preference for Arg at P⬘1 is a significant observation due to the ability of human plasma kallikrein to hydrolyze several proenkephalinderived peptides (27). All identified cleavages in these peptides occurred either at the COOH-terminal or between pairs of basic amino acids. Plasma kallikrein has recognized Lys-Lys, Lys-Arg, and Arg-Arg as processing signals. Human urinary kallikrein also presented high selectivity but toward Abz-F-R-A-EDDnp (K m ⫽ 0.27 ␮M, k cat ⫽ 0.77 s ⫺1, k cat/K m ⫽ 2852 mM ⫺1 s ⫺1 ) and Abz-FR-S-EDDnp (K m ⫽ 0.28 ␮M, k cat ⫽ 1.3 s ⫺1, k cat/K m ⫽ 4643 mM ⫺1 s ⫺1 ). The preference for amino acids with small side chains is similar to that observed with trypsin. In addition, Ser is the amino acid present at the P⬘1 position in human kininogen, which is the natural substrate for the enzyme. We also examined the susceptibility of Abz-F-R-X-EDDnp peptides to thrombin and activated factor X, and all peptides were resistant to hydrolysis. Figure 2 shows that the sensitivity of this kind of substrate is very high, able to detect papain in the range of 1 to 15 pM using Abz-F-R-L-EDDnp, which is the best substrate for this enzyme. Similar sensitivity was observed with cathepsin L using Abz-F-R-REDDnp or Abz-F-R-A-EDDnp. In contrast, Z-F-R-MCA is less sensitive, detecting papain or cathepsin L at concentrations four times higher. In conclusion, the substitution of the MCA group by aminoacyl-EDDnp has the advantage of increasing the specificity of the substrates for the proteases, allowing the development of a new series of very sensitive and specific substrates for cysteine proteases based on short internally quenched fluorescent peptides. These

76

MELO ET AL. TABLE 4

Kinetic Parameters for Hydrolysis of Abz-F-R-X-EDDnp and Z-F-R-MCA by Trypsin, Human Plasma Kallikrein, and Human Tissue Kallikrein Trypsin

Substrates

K m (␮M)

k cat (s ⫺1)

k cat/K m (mM ⫺1 s ⫺1)

K m (␮M)

k cat (s ⫺1)

k cat/K m (mM ⫺1 s ⫺1)

5.2 ⫾ 0.3

2.6 ⫾ 0.2

Z-F-R-MCA 500

2.7 ⫾ 0.2

0.43 ⫾ 0.02

159

X F L I A P Q S K R

Plasma kallikrein

Abz-F-R-X-EDDnp a

10.6 ⫾ 0.7 17 ⫾ 1 9.4 ⫾ 0.9 6.9 ⫾ 0.6 8.7 ⫾ 0.7 7.8 ⫾ 0.9

NDH 2.4 ⫾ 0.1 15 ⫾ 1 24 ⫾ 2 NDH 3.5 ⫾ 0.3 23 ⫾ 1 k cat/K m ⫽ 349 (mM ⫺1 s ⫺1) b 8.2 ⫾ 0.9

226 882 2553 507 2643 1051

1.1 ⫾ 0.1 2.7 ⫾ 0.2 1.2 ⫾ 0.1 0.73 ⫾ 0.05 0.12 ⫾ 0.01

NDH NDH 0.10 ⫾ 0.01 0.25 ⫾ 0.02 NDH NDH 0.41 ⫾ 0.01 0.099 ⫾ 0.008 0.22 ⫾ 0.01

90 92

342 135 1833

Note. Conditions: for trypsin, 0.1 M Tris, 10 mM CaCl 2, pH 8.0. For human plasma kallikrein, 50 mM Tris, 1 mM EDTA, pH 7.5. Temperature: 37°C. Enzyme concentration: 0.1 to 5 nM. Substrate concentrations used were in the interval value of ⫾1 K m. a NDH, no detected hydrolysis till [E] ⫽ 5 nM. b k cat/K m value was determined in pseudo-first-order condition because the K m value was too high, precluding a precise evaluation of the kinetic parameters.

are useful alternatives for detection and specificity studies of this family of proteases. Although the serine proteases did not hydrolyze the Abz-F-R-X-EDDnp series of peptides as efficiently as cysteine proteases, we found selective short substrates for human plasma kallikrein and human urinary kallikrein.

ACKNOWLEDGMENTS This work was supported by the Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo, Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico, and Financiadora de Estudos e Projetos through the PADCT- Biotecnologia III program. We also acknowledge the excellent technical assistance of Mrs. Eglelisa G. Andrade.

REFERENCES

FIG. 2. Rate of hydrolysis of Abz-F-R-L-EDDnp (䊐) and Z-F-RMCA (F) by different concentrations of papain. Conditions: 50 mM NaHPO 4, 5 mM EDTA, 200 mM NaCl, 5 mM DTT, 0.62 ␮M substrate, and 10 min for preactivation of the enzyme. Slops for each curve are 8.1 (‚ fluorescence/min) for Abz-F-R-L-EDDnp and 2.1 (‚ fluorescence/min) for Z-F-R-MCA.

1. Zimmerman, M., Yurewicz, E., and Patel, G. (1976) A new fluorogenic substrate for chymotrypsin. Anal. Biochem. 70, 258 – 262. 2. Zimmerman, M., Ashe, B., Yurewicz, E. C., and Patel, G. (1977) Sensitive assays for trypsin, elastase and chymotrypsin using new fluorogenic substrates. Anal. Biochem. 78, 47–51. 3. Schechter, I., and Berger, A. (1967) On the size of the active site in proteases. I. Papain. Biochem. Biophys. Res. Commun. 27, 157–162. 4. Yaron, A., Carmel, A., and Katchalski-Katzir, E. (1979) Intramolecularly quenched fluorogenic substrates for hydrolytic enzymes. Anal. Biochem. 95, 228 –235. 5. Gershkovich, A. A., and Kholodovych, V. V. (1996) Fluorogenic substrates for proteases based on intramolecular fluorescence energy transfer (IFETS). J. Biochem. Biophys. Methods 33, 135– 162. 6. Knight, C. G. (1995) Fluorimetric assays of proteolytic enzymes. Methods Enzymol. 248, 18 –34. 7. Stewart, J. M., and Young, J. D. (1984) in Solid Phase Peptide Synthesis, 2nd ed., Pierce Chemical Company, Rockford, IL. 8. Chagas, J. R., Juliano, L., and Prado, E. S. (1991) Intramolecularly quenched fluorogenic tetrapeptide substrates for tissue and plasma kallikrein. Anal. Biochem. 192, 419 – 425.

FLUOROGENIC SUBSTRATES FOR CYSTEINE AND SERINE PROTEASES ´ ., Plouffe, C., Takebe, S., Manson, P., 9. Carmona, E., Dufour, E Mort, S. M., and Me´nard, R. (1996) Potency and selectivity of the cathepsin L propeptide as an inhibitor of cysteine proteases. Biochemistry 35, 8149 – 8157. 10. Me´nard, R., Khouri, H.E., Plouffe, C., Dupras, R., Rippoli, D., Vernet, T. Tessier, D. C., Lalberte´, F., Thomas, D. Y., and Storer, A. C. (1990) A protein engineering study of the role of aspartate 158 in the catalytic mechanism of papain. Biochemistry 29, 6706 – 6713. 11. Barrett, A. J., and Kirshke, H. (1981) Cathepsin B, cathepsin H, and cathepsin L. Methods Enzymol. 80, 535–561. 12. Sampaio, C. A. M., Sampaio, M. U., and Prado, E. S. (1984) Active-site titration of horse urinary kallikrein. Hoppe Seyler’s Z. Physiol. Chem. 365, 297–302. 13. Oliva, M. L., Grisolia, D., Sampaio, M. U., and Sampaio, C. A. M. (1982) Properties of highly purified human plasma kallikrein. Agents Actions Suppl. 9, 52–57. 14. Kouyoumdjian, M., Borges, D. R, Michelacci, Y. M., Guimara˜es, J. A., Sampaio, C. A. M., and Prado, J. L. (1987) Purification and characterization of the alpha form of rat plasma kallikrein. Braz. J. Med. Biol. Res. 20, 549 –552. 15. Dias, C. L. F., and Rogana, E. (1986) Autolysis of ␤-trypsin at pH 3.0. Braz. J. Med. Biol. Res. 19, 11–18. 16. Chase, J. T., and Shaw, E. (1970) Titration of trypsin, plasmin and thrombin with p-nitrophenyl p⬘-guanidinobenzoate HCl. Methods Enzymol. 19, 20 –22. 17. Shimamoto, K., Chao, J., and Margolius, H. S. (1980) The radioimmunoassay of human urinary kallikrein and comparisons with kallikrein activity measurements J. Endocrinol. Metab. 51, 840 – 848. 18. Wilkinson, G. N. (1961) Statistical estimations in enzyme kinetics. Biochem. J. 80, 324 –332. 19. Gauthier, F., Moreau, T., Lalmanach, G., Brillard-Bourdet, M., Di Martino, M. F., and Juliano, L. (1993) A new, sensitive

20.

21.

22.

23.

24. 25.

26.

27.

77

fluorogenic substrate for papain based on the sequence of cystatin inhibitory site. Arch. Biochem. Biophys. 306, 304 – 308. Na¨gler, D. K., Tam. W., Storer, A. C., Krupa, J. C., Mort, J. S., and Me´nard, R. (1999) Interdependency of sequence and positional specificities for cysteine proteases of papain family. Biochemistry 38, 4868 – 4874. Rawlings, N. D., Polgar, L., and Barrett, A. J. (1991) A new family of serine-type peptidases related to prolyl oligopeptidase. Biochem. J. 279, 907–908. Portaro, F. C., Santos, A. B., Cezari, M. H. S., Juliano, M. A., Juliano, L., and Carmona, E. (2000) Probing the specificity of cysteine proteinases at subsites remote from the active site: Analysis of P 4, P 3, P⬘2 and P 3⬘ variations in extended substrates. Biochem. J. 347, 123–129. Me´nard, R., Carmona, E., Pouffe, C., Bro¨mme D., Konishi, Y., Lefevre, J., and Storer, A. C. (1993) The specificity of S⬘1 subsite of cysteine proteases. FEBS Lett. 328, 107–110. Alecio, M. R., Dann, M. L., and Lowe G. (1974) The specificity of the S1⬘ subsite of papain. Biochem. J. 141, 495–501. Nagler, D. K., Storer, A. C., Portaro, F. C., Carmona, E., Juliano, L., and Me´nard, R. (1997) Major increase in endopeptidase activity of human cathepsin B upon removal of occluding loop contacts. Biochemistry 36, 12609 –12615. Musil, D., Zucic, D., Turk, D., Engh, R. A., Mayr, I., Huber, R., Popovic, T., Turk, V., Towatari, T., Katunuma N., and Bode, W. (1991) The refined 2.15 A X-ray crystal structure of human liver cathepsin B: The structural basis for its specificity. EMBO J. 10, 2321–2330. Metters, K. M., Rossier, J., Paquin, J., Chretien, M., and Seidah, N. G. (1988) Selective cleavage of proenkephalin-derived peptides (less than 23,300 daltons) by plasma kallikrein. J. Biol. Chem. 263, 12543–12553.