Oligomeric structure and substrate induced inhibition of human cathepsin C.

Cathepsin C has been purified from human kidney by a modified procedure. Human cathepsin C was isolated as pure protein with a pI close to 6.0. The enzyme was shown to have a molecular mass of 200 kDa and to consist of four identical subunits, each composed of three different polypeptide chains, two of them disulfidebound. Their NH-terminal amino acid sequences were determined. Two chains showed pronounced similarity with the heavy and light chains of other papain-like cysteine proteinases, whereas the third one corresponded to the prosequence of the enzyme, thus showing that a substantial part of the proregion remains bound in the mature enzyme. The kinetics of substrate hydrolysis deviated substantially from standard Michaelis-Menten kinetics, demonstrating substrate inhibition at higher substrate concentrations. These data are explained by a sequential cooperative interaction model, where an enzyme molecule can bind up to four substrate molecules but where only the binary enzyme-substrate complex is catalytically active. Substrate inhibition was observed over the whole range of pH activity. From the pH activity profile it can be concluded that at least three ionizable groups with pK values 4.2, 6.8, and 7.7 are involved in substrate hydrolysis. Human cathepsin C thus appears to differ qualitatively from other cysteine proteinases of different origin.

Cathepsin C has been purified from human kidney by a modified procedure. Human cathepsin C was isolated as pure protein with a pI close to 6.0. The enzyme was shown to have a molecular mass of 200 kDa and to consist of four identical subunits, each composed of three different polypeptide chains, two of them disulfidebound. Their NH 2 -terminal amino acid sequences were determined. Two chains showed pronounced similarity with the heavy and light chains of other papain-like cysteine proteinases, whereas the third one corresponded to the prosequence of the enzyme, thus showing that a substantial part of the proregion remains bound in the mature enzyme. The kinetics of substrate hydrolysis deviated substantially from standard Michaelis-Menten kinetics, demonstrating substrate inhibition at higher substrate concentrations. These data are explained by a sequential cooperative interaction model, where an enzyme molecule can bind up to four substrate molecules but where only the binary enzymesubstrate complex is catalytically active. Substrate inhibition was observed over the whole range of pH activity. From the pH activity profile it can be concluded that at least three ionizable groups with pK a values 4.2, 6.8, and 7.7 are involved in substrate hydrolysis. Human cathepsin C thus appears to differ qualitatively from other cysteine proteinases of different origin.
Cathepsin C, known also as dipeptidyl aminopeptidase I (DPPI), cathepsin J, or dipeptidyl transferase (EC 3.4.14.1), is a lysosomal cysteine peptidase (1)(2)(3) belonging to the papain family (4). It is present in a variety of tissues from rat and human sources (5). A survey of human tissues showed that spleen and kidney are the richest sources of cathepsin C. In addition, serum levels were highest in hepatic diseases, followed by peripheral arterial disease, thromboembolism, myocardial infarction, diabetes mellitus, and prostatic hypertrophy. Like other lysosomal cysteine proteinases cathepsin C is involved in intracellular protein degradation (1), and it has been reported that cathepsin C activity is present at higher levels in cytotoxic lymphocytes and myeloid cells, indicating more specific but as yet unknown roles in these immune effector cells (6).
Cathepsin C is active against different substrates in the pH range between 3.5 and 8.0 (7). It cleaves primarily peptide and protein substrates having an unsubstituted amino terminus (8 -15), although it can also degrade substrates with a blocked amino terminus (16). Early studies showed that the enzyme requires halide ions and sulfhydryl reagents to achieve maximal hydrolytic activity (11,17), indicating that it is a cysteine proteinase. It is specifically inactivated by Gly-Phe-diazomethyl ketone, a typical thiol-blocking reagent, with a K i value of about 10 nM (18). Recently, it has been shown to be inhibited by E-64, another typical thiol-blocking reagent, and leupeptin, both at high concentrations (2). Cathepsin C is also inhibited by rat stefin A and chicken cystatin, two protein inhibitors of cysteine peptidases from the cystatin superfamily (2,19). There are differing reports concerning the molecular mass and subunit composition of cathepsin C. Thus bovine cathepsin C was found to be an oligomeric enzyme of about 200 kDa, composed of eight subunits in the form of two tetramers with two different types of subunits (14). The rat liver enzyme was also estimated to have a molecular mass of 200 kDa (12), but more recently 160 kDa, consisting of two different subunits (2). The enzyme from porcine spleen was isolated as a mercurial derivative of 56 kDa, which formed after reducing an active dimer of 110 kDa (20).
The cDNA sequence of rat cathepsin C (21) is highly homologous to those of cathepsins B, H, L, and S and papain (22)(23)(24)(25)(26). However, the propeptide region of cathepsin C is substantially longer than those of other cysteine peptidases but without pronounced amino acid sequence similarity to any of them (21). Recently, partial amino acid sequences of rat (2,21) and human (27) cathepsin C have been reported. The biosynthesis and processing of cathepsin C were investigated by pulse-chase experiments in cultured rat macrophages (28), showing that the enzyme is first synthesized as procathepsin C with a molecular mass of 55 kDa. Within 1 h procathepsin C is cleaved and modified into mature cathepsin C having two chains of 25 and 7.8 kDa. Cathepsin C then oligomerizes just before entering the lysosomes. This behavior is unique among papain-like cysteine peptidases.
The above results apply mainly to cathepsin C of non-human origin, and there is little information concerning the properties and structure of human cathepsin C (27,29). It has been reported that human cathepsin C is a glycoprotein with a molecular mass of 200 kDa consisting of eight subunits of 24 kDa (27), while preliminary studies in this laboratory have indicated the existence of four subunits of 50 kDa. While differences in substrate specificity among human, bovine and porcine cathepsin C (27) have been reported, the detailed kinetics have not so far been studied for human cathepsin C.
We present new more detailed studies on the properties of human cathepsin C, its amino acid sequence analysis, characterization of its oligomeric structure, and kinetics of its interaction with synthetic substrates.
Purification Procedure-Cathepsin C was purified from human kidney as follows. All the procedures were carried out at 4°C unless otherwise indicated. Frozen kidneys (1 kg) were partially thawed, cut into small pieces, and homogenized in 1.5 liter of 100 mM sodium acetate buffer, pH 5.0, containing 300 mM NaCl and 1 mM EDTA (buffer A). Nonsoluble material was removed by centrifugation at 10,000 ϫ g for 15 min. The supernatant was adjusted to pH 4.2 and incubated at 37°C for 2 h. Following additional centrifugation at 10,000 ϫ g for 15 min, the supernatant was precipitated with solid ammonium sulfate to 70% saturation on ice. The suspension was stirred for 30 min, and then the precipitated proteins were collected by centrifugation at 10,000 ϫ g for 15 min. The precipitate was resuspended in a small volume of 20 mM sodium acetate buffer, pH 5.2, containing 1 mM EDTA (buffer B). Following dialysis against the same buffer, insoluble material was removed by centrifugation, and the supernatant was applied to a CM-Sephadex C-50 column (4 ϫ 25 cm), equilibrated with buffer B. After thorough washing of the column, bound proteins were eluted with a linear salt gradient (0 -500 mM NaCl in buffer B) at a flow rate of 42 ml/h. Fractions (14 ml) were collected and assayed for catalytic activity using Gly-Phe-4M␤NA 1 as described below. The peak containing cathepsin C activity, which eluted at 250 mM NaCl, was pooled and concentrated by ultrafiltration (Amicon YM-10). The concentrate was applied to a Sephacryl S-200 column (3 ϫ 140 cm) equilibrated with 100 mM sodium acetate buffer, pH 5.5, containing 300 mM NaCl and 1 mM EDTA (buffer C) at a flow rate of 33 ml/h. Catalytically active fractions with molecular mass of 200 kDa were pooled and concentrated (Amicon YM-10). An appropriate volume of 50 mM dithioerithritol was then added to the sample to give a final concentration of 5 mM. After 15 min of stirring at room temperature, the dithioerithritol was removed by dialysis against 100 mM phosphate buffer, pH 6.0, containing 300 mM NaCl and 1 mM EDTA (buffer D). The sample was stirred overnight with 50 ml of activated thiol-Sepharose 4B equilibrated with the same buffer. The suspension was then poured into a column, and the unbound material was washed off with buffer D. Proteins, including cathepsin C, were eluted with 20 mM Cys in buffer D. 5-ml fractions were collected and assayed for catalytic activity. Active fractions were pooled, concentrated (Amicon YM-10), and dialyzed against 20 mM bis-Tris buffer pH 6.0 containing 1 mM EDTA (buffer E). The dialyzed sample was then applied to a Q-Sepharose FF column (1 ϫ 15 cm) equilibrated with buffer E. The column was washed extensively, and bound proteins were eluted with a linear salt gradient (0 -250 mM NaCl in buffer E) at a flow rate of 1 ml/min. Cathepsin C eluted at 150 mM NaCl. From 1 kg of human kidneys, 2 mg of pure cathepsin C were obtained.
Enzyme Assay-Cathepsin C was routinely assayed by the hydrolysis of Gly-Phe-4M␤NA at 37°C. 50 l of enzyme solution was activated for 5 min with 200 l of 4 mM cysteine in the assay buffer (100 mM phosphate buffer, pH 6.0, 20 mM NaCl, 1 mM EDTA), followed by the addition of 200 l of 125 M substrate in assay buffer. After 10 min of incubation, the reaction was stopped by adding 500 l of a mixed solution of commercial Fast Garnet GBC salt with chloromercuribenzoate-EDTA reagent (30).
Determination of Protein Concentration-Protein concentration of cathepsin C was determined by the method of Bradford (31), using papain as a standard. The concentration of papain was determined spectrophotometrically using the absorption coefficient 2.39 liters⅐g Ϫ1 cm Ϫ1 (32) and M r 23,500 (33).
Determination of Molecular Weight by Gel Filtration-A column of Superdex 200 HR (300 ϫ 10 mm) attached to fast protein liquid chromatography (Pharmacia Biotech Inc.) was used for determination of the molecular mass of human cathepsin C. The buffer was 100 mM potassium phosphate, pH 7.1, containing 150 mM NaCl. Determination was performed at 24°C and at a flow rate of 0.5 ml/min. The column was calibrated with aldolase (160 kDa), bovine serum albumin (67 kDa), ovalbumin (43 kDa), chymotripsinogen A (25 kDa), and cytochrome c (12.3 kDa).
SDS-PAGE-Electrophoresis was performed on a PhastSystem apparatus (Pharmacia) using 8 -25% gradient gels, following the manufacturer's instructions. A sample was prepared in three different ways. Nonreduced and nondenaturated sample was mixed with SDS buffer and applied to the gel. Denaturated but nonreduced sample was mixed with the same buffer and boiled at 100°C for 5 min and than applied to the gel. Reduced sample under denaturating conditions was boiled for 5 min at 100°C in SDS buffer containing 2-mercaptoethanol.
Electrophoretic Titration Curve-Electrophoretic titration curves were obtained according to Ref. 34 on a PhastSystem apparatus (Pharmacia), following the manufacturer's instructions.
Reduction and Alkylation of Cathepsin C-The sample was dissolved in 500 mM Tris buffer, pH 8.0, containing 6.0 M guanidinium chloride and 2 mM EDTA and reduced with 10 mM 2-mercaptoethanol at 40°C in an atmosphere of argon. 15 h later, the sample was alkylated for 20 min with 20 mM iodoacetic acid.
Peptide Purification and Separation-Cathepsin C was analyzed by HPLC (Milton Roy LCD, United Kingdom) on a ChromSpher C8 column (3 ϫ 100 mm) equilibrated with 0.1% trifluoroacetic acid. Cathepsin C was eluted with a linear gradient of acetonitrile (0 -70%) in the starting solution. Optical density was monitored continuously at 215 nm; the flow rate was 1 ml/min. NH 2 -terminal Sequence Analysis-Amino acid sequence analyses were performed on an Applied Biosystems liquid phase Sequenator model 475 A, connected on-line to a 120A phenylthiohydantoin-derivative analyzer from the same manufacturer.
Kinetics of Substrate Hydrolysis-Substrate hydrolysis was investigated at 25°C in the pH range 3.0 -8.0. The following buffers were used: at pH 3.0, 100 mM citrate buffer; at pH 4.0 and 5.0, 100 mM acetate buffer; at pH 6.0 and 7.0, 100 mM phosphate buffer; and at pH 8.0, 100 mM borate buffer. All of the buffers contained 1 mM EDTA, 100 mM NaCl, and 3% dimethyl sulfoxide. Substrate, Gly-Phe-4M␤NA (10 -600 M final concentration), was dissolved in the desired buffer. Prior to experiments, cathepsin C was activated with 100 mM phosphate buffer, pH 6.0, containing 1 mM EDTA, 100 mM NaCl, and 2 mM dithioerythritol. The reaction was started by adding 50 l of activated enzyme (0.14 -1.4 nM final concentration, depending on pH) to 2.95 ml of the substrate solution. The product release was then continuously monitored at excitation and emission wavelengths of 335 and 415 nm, respectively, with a Perkin Elmer LS-3 spectrofluorimeter, controlled by Flusys software (35). At least five experiments were done at each substrate concentration, each with less than 5% of substrate consumption.
pH Activity Profile-The pH activity profile of cathepsin C was determined by measuring k cat /K m at 0.2 pH interval over the range of pH The concentration of substrate in the reagent mixture was 0.25 M. The reaction buffers were prepared from 20 mM citric acid and 40 mM sodium phosphate, mixed in appropriate aliquots to get the desired pH value. All buffers also contained 0.1 M NaCl and 1 mM EDTA. Samples were monitored to ensure that pH did not fluctuate during the enzyme reaction.

RESULTS
Purification of Cathepsin C-Cathepsin C was purified from human kidney as described under "Experimental Procedures." Initially, human kidney supernatant was exposed to slightly acidic pH during incubation at 37°C. Following ammonium sulfate precipitation and ion exchange chromatography on a CM-Sephadex column, gel filtration on Sephacryl S-200 was carried out. In this step, the high molecular weight cathepsin C was separated from the low M r papain-like cysteine peptidases. In the next step, the majority of proteins not having a free thiol group were separated by covalent chromatography on thiol-Sepharose. Finally, cathepsin C was additionally purified on DEAE-Sepharose, from which it eluted as a single peak at 100 mM NaCl. 2 mg of cathepsin C were obtained from 1 kg of starting material. The enzyme was stored at 4°C, since freezing resulted in the loss of the oligomeric structure (described below).
Determination of Molecular Weight by Gel Filtration-The cathepsin C isolated from human kidney was eluted as a symmetrical peak of molecular mass 200 Ϯ 10 kDa.
SDS-PAGE-Samples of cathepsin C, not frozen and repetitively frozen several times, were dissolved in SDS buffer at room temperature. They both migrate as single bands with molecular masses of ϳ200 and ϳ50 kDa, respectively, on nonreducing SDS-PAGE (Fig. 1). This indicates that repetitive freezing irreversibly destroys the oligomeric structure of cathepsin C. It suggests also that the cathepsin C molecule consists of four equal subunits. After incubation of cathepsin C in SDS buffer at 100°C, it migrated as two bands with molecular masses of ϳ23 kDa and less than 10 kDa, while under reductive conditions it ran as a triple band with molecular masses of ϳ23, ϳ16, and Ͻ10 kDa (Fig. 1B).
Electrophoretic Titration Curve-The electrophoretic titration curve shows that cathepsin C migrates as a single band over the pH range between 3.5 and 9.0 (Fig. 2), indicating that cathepsin C is homogeneous in the whole pH range investigated. A pI value of 6.0 was determined from the same curve. NH 2 -terminal Sequence Analysis-After separation on reverse-phase HPLC, cathepsin C was subjected to NH 2 -terminal sequence analysis. Three NH 2 -terminal sequences were found: DTPAXCTYLD, LPQSWDWRN, and DPFNPFELTN. Alignment of these three sequences with those of rat cathepsin C NH 2 -terminal sequences (2) and rat procathepsin C sequence deduced from cDNA (21) indicates that these NH 2 -termini correspond to those of the proregion (residue Ϫ205 of rat procathepsin C), to the heavy chain (residue 1 of rat procathepsin C), and to the light chain (residue 165 of rat procathepsin C) (Fig. 3).
Kinetics of Substrate Inhibition of Cathepsin C-Preliminary experiments showed that the rate of hydrolysis of Gly-Phe-4M␤NA was substantially decreased at high substrate concentrations in the range 100 -600 M. In order to check whether product inhibition is occurring, the rate of product formation was followed in the presence and absence of 500 M dipeptide Gly-Phe using 20 mM Gly-Phe-4M␤NA as substrate. No change in the initial rate was detected in the presence of dipeptide, indicating that the rate decrease is due to the high substrate concentration. Since the enzyme is a tetramer, composed of four presumably identical subunits, it was proposed that the decrease in reaction rate resulted from cooperative interactions between subunits of the enzyme in complexes ES 2 , ES 3 , ES 4 . The arrangement of subunits in a tetrameric enzyme may be tetrahedral, linear, or square, depending on the number of interactions between subunits (36,37). A number of different models exist to explain such cooperative behavior (Adair-Pauling, Monod-Wyman-Changeux, Frieden, Koshland-Nemethy-Filmer, Wong-Endrenyi; reviewed in Refs. 38 and 39)). However, the binding equations described did not take into account possible substrate inhibition. Various models were fitted to the experimental data, assuming that only the binary complex is active. Such a simplified model is shown in Scheme I, where E represents tetrameric enzyme, K s is the dissociation constant and a, b, and c are correction factors by which the intrinsic binding constant has been changed with binding of first, second, and third substrate molecule, respectively. The best fit with the lowest 2 error was obtained with the Wong-Endrenyi equation for tetrahedral arrangement of subunits (Fig. 4). The rate of product hydrolysis, corresponding to this mechanism, is described by where v is the velocity of product formation, V M is maximal velocity, and [S] is substrate concentration. The results support the assumption that only the binary enzyme-substrate complex is fully active, whereas all higher complexes are inactive or almost inactive. As can be seen in Table I, the binding of the second molecule is only slightly less favorable than the binding of the first substrate molecule, whereas a drastic drop in affinity is observed for the binding of the third substrate molecule (Fig. 5), indicating a large negative cooperativity effect. The binding of the fourth substrate molecule is highly favorable,  although the value of c is somewhat uncertain due to the large error in calculations of the last parameter.
In a further attempt to check whether the ternary and higher enzyme-substrate complexes are active, the data were fitted by the following equation, modified from that of , to take into account reduced activity of higher complexes.
The parameters d, e, and f represent coefficients by which the activities of the complexes ES 2 , ES 3 and ES 4 are reduced due to substrate inhibition. Fitting Equation 2 to the experimental data, for the case where the ternary complex is active (i.e. d Ͼ 0, e ϭ f ϭ 0), did not reduce the 2 error. The best fit was obtained using Equation 2, with d very close to zero (Ͻ0.02). This indicates that the ternary complex has low or zero activity, as proposed in Scheme I. Similar results were obtained for the systems where higher complexes were assumed to be active, supporting the above conclusion that only ES has significant activity. The effect of pH on the rate of substrate hydrolysis was studied in the pH range from 3.0 to 8.0. Substrate inhibition was observed at all pH values, and the experimental data at each pH value could be best described using Equation 1. As can be seen from Table I, all of the kinetic constants are strongly pH-dependent. The pH dependence of the second order rate constant for substrate hydrolysis, k cat /K m , exhibits a bellshaped profile (Fig. 6), indicating that at least two ionizable groups of cathepsin C are involved in the interaction. In a detailed analysis, the reaction rate was measured every 0.2 pH unit as described under "Experimental Procedures." The pH activity data profile obtained in this way was best fitted by the equation corresponding to a model with three different disso-ciable groups of the enzyme involved in substrate hydrolysis (40).
where (k cat /K m ) lim represents the limiting value of k cat /K m . In Fig. 6 the best fit, corresponding to the pK a values 4.2 Ϯ 0.05, 6.8 Ϯ 0.1, and 7.7 Ϯ 0.1 is shown.

DISCUSSION
A number of isolation procedures have been described for the purification of cathepsin C from different mammalian tissues (12,17,20), but only two used human tissues as a starting material (27,29). We isolated cathepsin C from human kidney, known to be a rich source of lysosomal cysteine peptidases. In our purification scheme the successive affinity chromatography steps used by McGuire and co-workers (27) are replaced by cation exchange and covalent chromatographies, which resulted in a higher yield of pure enzyme as shown by sequence homogeneity and electrophoresis (Fig. 2).
Several attempts have been made to determine the oligomeric structure of cathepsin C, with results differing in the number and size of the enzyme subunits. When dissociation of bovine cathepsin C was studied in the presence of guanidinium chloride and urea (14), an intermediate with a molecular mass of 53 kDa was observed, which could be further dissociated into smaller subunits with molecular mass of 24.5 kDa, suggesting that cathepsin C consists of eight subunits. Furthermore, two different NH 2 -terminal amino acids were found, suggesting that two different types of subunits are present in the enzyme. Similar conclusions were reported for the human spleen enzyme (27) and for the rat enzyme (28), since these researchers found that the enzyme is a 200-kDa protein composed of 24-kDa subunits, as determined by SDS-PAGE. In addition, another band was detected on SDS-PAGE, corresponding to the light chain of the enzyme (28). The molecular mass of 200 kDa, as we determined for the native cathepsin C by gel filtration, is in good agreement with these results, as well as the molecular mass of 23 kDa and Ͻ10 kDa of the subunits, determined by SDS-PAGE under nonreducing conditions (Fig. 1B). However, three different components with molecular masses 23, 16, and Ͻ10 kDa were obtained by SDS-PAGE under reducing conditions, indicating that at least two components are connected FIG. 3. Position of the NH 2 -terminal amino acid sequences of the human cathepsin C polypeptide chains in comparison with the rat preprocathepsin C. A, rat preprocathepsin C amino acid sequence deduced from cDNA (21). B, NH 2 -terminal amino acid sequences of rat cathepsin C (2). C, NH 2 -terminal amino acid sequences of human cathepsin C. Identical amino acid residues of human and rat cathepsin C are boxed. with disulfide bonds. These results are in agreement with those reported for rat liver cathepsin C (2). They reported a molecular weight of 160 kDa with two different kinds of subunits, present in a 1:1 molar ratio. One of the subunits was found to be a glycoprotein, which could under denaturing conditions dissociate into two components with molecular masses of 19 -24 kDa and 6 kDa. The NH 2 -terminal sequence of the larger component was found to be very similar to those of rat cathepsins B, H, and L, whereas that of the smaller component exhibited considerable similarity to those of the light chains of these cathepsins. The other subunit was also found to be gly-cosylated and exhibited a molecular mass of 17 kDa on SDS-PAGE, but its NH 2 -terminal amino acid sequence did not show any similarity to the partial sequences of other cathepsins. Moreover, the alignment of the NH 2 -terminal sequences of rat and human cathepsin C with that of rat procathepsin C (Fig. 3) revealed that the sequences are highly similar. Furthermore, in the mature enzyme, the 16-or 17-kDa fragment (2), which corresponded to the proregion of the enzyme (Fig. 3), was present, indicating that a substantial part of the proregion still remains bound in the mature enzyme. This is further supported by the finding that monomeric procathepsin C (55 kDa) is only slightly larger than the monomeric form of the mature enzyme ( Fig. 1A) or of the intermediate of 53 kDa (14), which presumably corresponded to the monomeric form. Although this finding is unusual for lysosomal cysteine proteinases, where only cathepsin H is a slight exception (41,42), it is well known that large parts of proregions can remain bound to the active form of an enzyme. Examples are the various serine proteinases from the blood coagulation cascade (43,44). From the results presented here we conclude that cathepsin C is an oligomeric enzyme, consisting of four identical subunits, each composed of three different polypeptide chains.
The kinetics of substrate hydrolysis showed a significant decrease in reaction rate at high substrate concentration, thus deviating from the simple Michaelis-Menten kinetics. As further shown, this can be explained in terms of substrate inhibition. A similar observation was also made for beef spleen cathepsin C (45), although Heinrich and Fruton used different substrates. This phenomenon can be explained by cooperativity between the subunits upon substrate binding. Kinetically, the experimental data could be best interpreted by the model in which an enzyme molecule can bind up to four substrate molecules and where only the binary enzyme-substrate complex is active (Scheme I). This model for substrate inhibition is also consistent with molecular weight and sequence data, such that each of the four identical subunits has an active site similar to those of other cysteine proteinases. Although the arrangement of subunits in the cathepsin C molecule is unknown, the model suggests that all four substrate binding sites have equal initial affinities for substrate. A theoretical model, corresponding to the tetrahedral arrangement of subunits (37) gave a slightly better result than the corresponding models for the linear and square geometry of the subunits. All attempts to fit other models, including the simple Adair-Pauling (39) and Koshland (36) models, resulted in substantially larger 2 error and significant deviations from the experimental data. As shown in Table I, binding of the first substrate molecule to any of the four binding sites has little effect on the affinity for the second substrate molecule. Increasing substrate concentration, however, enables an enzyme molecule to bind additional substrate molecules. Binding of the second substrate molecule seems not to be productive and, in addition, it reduces the catalytic activ- ity at the first site (Fig. 4, Table I). In addition, binding of the second substrate molecule drastically decreases the binding affinity for the third substrate molecule, exhibiting a large negative cooperativity effect. In Fig. 5 it is seen that, over the range of experimental substrate concentrations, the enzyme exists predominantly as ES and ES 2 . Calculations for the population of enzyme-substrate complexes have been made on the basis of the model in Scheme I. In contrast, binding of the fourth substrate molecule is highly favorable, indicating positive cooperativity (Table I). As a consequence, the population of ES 3 is negligible. However, binding of these additional substrate molecules has no effect on overall enzyme activity, since it is clear from Fig. 5 that the inhibition of activity is largely, if not entirely, due to formation of ES 2 .
McGuire et al. (27) observed normal Michaelis-Menten behavior for the hydrolysis of a similar substrate in a similar concentration range, which could be explained by the fact that they were using frozen enzyme, which was shown to irreversibly lose its quaternary structure (Fig. 1A).
The kinetics of substrate hydrolysis, investigated between pH 3.0 and 8.0, showed substrate inhibition over the whole pH range (Table I), indicating that this phenomenon is not an artifact but a general feature of this enzyme. From a detailed investigation of the pH activity profile three pK a values of 4.2, 6.6, and 7.7 were determined (Fig. 6). The three-dimensional structure of cathepsin C is not known, and therefore any assignment of these values to particular groups of the enzyme would be speculative. Nevertheless, cathepsin C exhibits considerable sequence homology with other papain-like cysteine proteinases, including the conserved Cys-25 and His-159 (papain numbering), which are essential for the catalytic activity (46). However, the pK a values for the formation and decomposition of the thiolate-imidazolium ion pair, which is the reactive component, are 3.4 and 8.3 in papain or 3.5 and 8.6 in cathepsin B (40,47). Therefore it is very likely that the pK a values we observed belong to neighboring charged groups, which are involved in the catalytic activity of cathepsin C. As suggested by Brocklehurst (48), use of substrates in the extreme pH regions, where cysteine proteinases are only slightly active, may well mask the ionization of the reactive ion pair. Cathepsin C exhibits some additional similarities with lysosomal cysteine proteinases. It has a pH optimum around 6 ( Fig. 6, Table I), typical for the lysosomal cysteine proteinases (49,50), and is extremely unstable at neutral pH (2), a property characteristic of cathepsins B and L (51,52). Substrate-induced inhibition was demonstrated also at pH values where the enzyme is stable (Table I), confirming that the inhibition is not associated with protein instability.
In conclusion, cathepsin C, isolated from human kidneys, exhibits a number of structural similarities with other lysosomal papain-like cysteine proteinases. However, some important differences from other members of the papain family have been shown: (i) cathepsin C is an oligomeric enzyme, consisting of four identical subunits; (ii) each of these subunits is composed of three different polypeptide chains, two of them disulfide-bound; (iii) the kinetics of substrate hydrolysis deviate from Michaelis-Menten kinetics, reflecting both negative and positive cooperative effects of substrate binding between the subunits; (iv) substrate binding leads to inhibition of hydrolysis, observed over the whole pH range of enzyme activity.