YEAST Yeast 2000; 16: 219±229. Aminopeptidase yscCo-II: a new cobalt-dependent aminopeptidase from yeastÐpuri®cation and biochemical characterization IRMA HERRERA-CAMACHO*, ROSALVA MORALES-MONTERROSAS AND Â Z-ALVAREZ RUBEÂN QUIRO Area de BioquõÂmica, Centro de QuõÂmica del Instituto de Ciencias, Universidad AutoÂnoma de Puebla, 72000 Puebla, MeÂxico Saccharomyces cerevisiae aminopeptidase yscCo-II (APCo-II) was puri®ed to apparent homogeneity by gel ®ltration, af®nity chromatography and anion-exchange chromatography. APCo-II is an hexameric cobaltdependent metallo-enzyme with an estimated native molecular mass of 290 kDa. Enzyme activity is only detected in the presence of cobalt ions at pH 7.0. Substrate speci®city studies indicate that aminopeptidase yscCo-II cleaves only basic N-terminal residues. PMSF, Cu2+, 1,10-phenanthroline and bestatin were found to be very strong inhibitors of aminopeptidase yscCo-II activity. Kinetic studies indicated that the enzyme has a similar Km and KaCo (activation constant of cobalt) and, following extraction of cobalt from the enzyme, activity was recovered only after cobalt addition. Copyright # 2000 John Wiley & Sons, Ltd. KEY WORDS Ð Saccharomyces cerevisiae; protein degradation; cobalt-dependent aminopeptidase; protease INTRODUCTION The unicellular eukaryote, Saccharomyces cerevisiae, is especially suited for revealing the biological role of peptidases, since it is easily accessible to biochemical, genetic and molecular biological techniques. It is gradually becoming clear that peptide bond hydrolysis represents an essential mechanism in the regulation of cell metabolism at the post-translational level (SuaÂrez-Rendueles and Wolf, 1988; Fuller et al., 1988; Heinemeyer et al., 1991; Peters, 1994; Fujimura-Kamada et al., 1997; Hubbard, 1998). Much work has already been done on the two major proteolytic systems, the lysosomal and the cytosolic, that are involved in the degradation of proteins in S. cerevisiae (Hirsch et al., 1989; Fujiwara et al., 1990; Lee and Goldberg, 1996; Klionsky, 1998). The biochemical and genetic characteristics of the main enzymes of each system, e.g. the vacuolar exopeptidases and the multicatalytic enzyme complex or proteasome, have been investigated *Correspondence to: I. Herrera-Camacho, Area de BioquõÂmica del Centro de QuõÂmica, Instituto de Ciencias, Universidad AutoÂnoma de Puebla, Apartado Postal 1613, 72000 Puebla Pue, MeÂxico. E-mail: email@example.com CCC 0749-503X/2000/030219±11$17.50 Copyright # 2000 John Wiley & Sons, Ltd. very thoroughly (Achstetter and Wolf, 1985; Rivett, 1993; Richter-Rouff and Wolf, 1993; Lee and Goldberg, 1996; Groll et al., 1997; Gilon et al., 1998). Transcriptional regulation of the yeast vacuolar aminopeptidase yscI-encoding gene (APE1) by carbon sources has been shown in very recent years (Bordallo et al., 1995). Moreoever, there is the interesting example of a eukaryotic DNA-binding cysteine protease (Xu and Johnston, 1994), and the transcriptional regulation by nutrient limitation of the CPS1 gene by means of regulatory elements (Bordallo and SuarezRendueles, 1995). A number of proteases in yeast have been puri®ed and characterized, and their structural genes analysed (for reviews, see Achstetter and Wolf, 1985; Hirsch et al., 1989; Jones, 1991). Recent studies have identi®ed methionine aminopeptidase (Chang et al., 1990) and aminopeptidase yscXVI (Tisljar and Wolf, 1993); aminopeptidase Y (Yasuhara et al., 1994). In the past few years, a number of structural genes encoding yeast aminopeptidases have been cloned and sequenced, e.g. the BLH1 gene, encoding thiol-aminopeptidase (Enenkel and Wolf, 1993), and the aminopeptidase Y gene (Nishizawa et al., 1994). Received 21 January 1998 Accepted 10 October 1999 220 Â Z-ALVAREZ I. HERRERA-CAMACHO, R. MORALES-MONTERROSAS AND R. QUIRO Recently, cobalt-dependent enzymes have become an increasingly studied research ®eld and this has led to the description of new types of proteolytic enzymes from both prokaryotes and eukaryotes (Roderick and Matthews, 1993; Ar®n et al., 1995). Indeed, there is evidence that most of the aminopeptidase activities in yeast are stimulated, to a greater or lesser extent, by cobalt (Co2+) (Hirsch et al., 1989; Herrera-Camacho, 1984; Chang et al., 1990; Tisljar and Wolf, 1993). However, aminopeptidase yscCo is the only one described to date as detectable only in the presence of Co2+ (Achstetter et al., 1982). We will refer to this enzyme as AP yscCo-I, following the nomenclature proposed by Achstetter and Wolf (1985). In this paper, we report the puri®cation and characterization of another aminopeptidase that is only detectable in the presence of Co2+, which we name aminopeptidase yscCo-II (APCo-II). Its properties clearly differentiate this new aminopeptidase from all similar activities so far described in yeast. Enzyme assays All chromogenic peptide substrates were dissolved in water at a concentration of 10 mM, and stored at x20uC. Aminopeptidase yscCo-II activity was determined routinely at 37uC with the chromogenic substrate L-lysine-4-nitroanilide, and the enzyme was pre-incubated in the presence of Co2+ and a buffer for 10 min at 37uC, after the start of the reaction with the substrate. An aliquot (0.5 ml) of the test solution contained an enzyme solution with 50 mM Tris±HCl buffer, pH=7.0, 0.5 mM CoCl2 and 1 mM of substrate solution. The test was stopped by adding 0.5 ml 40 mM EDTA (pH 8) and 20 mM chloroquine solution. The release of 4-nitroaniline was measured in the supernatant at 405 nm in a UV-160 Shimadzu photometer. One milliunit (mU) of enzyme activity is de®ned as the amount of enzyme that releases 1 nmol product/min under test conditions. The molar absorption coef®cient for 4-nitroaniline at 405 nm is e405nm=9900 Mx1 cmx1; this value was used for the calculation of enzyme activity. Protein determination MATERIALS AND METHODS Chemicals The chromogenic peptide substrates, as well as the proteinase inhibitors, were obtained from Sigma (USA) or Bachem (Switzerland). All other chemicals were of the highest purity available and were purchased from Merck (Germany), Sigma (USA) and Aldrich (USA). The yeast growth medium was obtained from DIFCO (USA). Matrices for gel chromatography were obtained from Pharmacia Biotech (Sweden). Protein was determined according to the method of Sedmak and Grossberg (1977), using crystalline bovine serum albumin as standard. Puri®cation of aminopeptidase yscCo-II All puri®cation steps were performed at 40uC. Step 1. Crude extract supernatant The preparation of crude extract supernatant was performed using the conditions of SuaÂrezRendueles et al. (1981). Yeast strain and growth conditions Step 2. Filtration chromatography and molecular mass Most experiments were performed with the diploid Saccharomyces cerevisiae wild-type strain, 1022. S cerevisiae strain II-21 (MATa lap1 lap2 lap3 lap4 leu2±3,112), which lacks aminopeptidase activity (Cuevas et al., 1989), was used as reference; Dr Paz SuaÂrez-Rendueles kindly supplied this strain. Yeast cells were grown in a minimal medium containing 0.7% Yeast Nitrogen Base without amino acids, and 2% glucose, in a rotary shaker at 32uC, and harvested towards the end of the exponential growth phase (Abs600nm= 1.0). After dialysis of crude extract supernatant against 20 mM MOPS/Tris, pH 7.0, 100 mM KCl, 5 mM N3Na (®ltration buffer), the extract was run on a column (2.6r60 cm) of 320 ml Sephadex G200 (diluted with ®ltration buffer) at a linear ¯ow rate of 10 ml/h. Fractions of 2.5 ml were collected and those containing the highest activity pooled. The following markers were used for molecular mass determinations of native APCo-II: egg albumin (43 kDa), bovine serum albumin (67 kDa), aldolase (158 kDa), catalase (232 kDa), and ferritin (450 kDa). Copyright # 2000 John Wiley & Sons, Ltd. Yeast 2000; 16: 219±229. 221 AMINOPEPTIDASE yscCo-II FROM YEAST Step 3. Af®nity chromatography For this step, the af®nity adsorbant was obtained by coupling the lys-phe dipeptide to AH-Sepharose-4B, using the carbodiimide method of Pharmacia-Biotech. Fractions (pooled according to step 2) were dialysed against 20 mM MOPSTris, pH 7.0 (af®nity buffer). They were then applied to a Lys-phe-AH-Sepharose-4B (column 1.5r10 cm), equilibrated with af®nity buffer and run with a buffer ¯ow rate of 12 ml/h. After release, the protein was washed with 20 mM acetate buffer, pH 5.5; APCo-II was eluted to 100 mM KCl in acetate buffer. Fractions of highest activity were pooled and again dialysed against af®nity buffer. Step 4. Ion exchange chromatography The APCo-II obtained in step 3 was applied to the DEAE-Cellulose (column 2r15 cm) and run in af®nity buffer with a ¯ow rate of 12 ml/h. After release of the protein, the APCo-II was eluted to 50 mM KCl in af®nity buffer. Fractions of highest activity were pooled, dialysed against af®nity buffer, and stored at 40uC until used for the characterization studies. Polyacrylamide gel electrophoresis (PAGE) After freeze-drying, a sample of puri®ed APCoII was loaded onto an SDS±PAGE gel (7% running gel; 3.5% stacking gel). Electrophoresis was performed under reducing conditions, as described by Laemmli (1970), at 60 V (stacking gel) and 100 V (running gel) in a Mini-Protean II apparatus (Bio Rad). Protein staining with silver was performed, as indicated by Morrisey (1981). The Mr of APCo-II was determined from the relative mobility of protein standards: phosphorylase b (97 kDa), bovine serum albumin (66 kDa), egg albumin (45 kDa) and carbonic anhydrase (29 kDa). Non-denaturing polyacrylamide gels were prepared in the same manner as for SDS±PAGE, but without the denaturing and reducing conditions. control was obtained with CoCl2 (1 mM) and phen (1 mM) in af®nity buffer and after incubation (18 h at 4uC), UV±VIS spectroscopy was performed. The enzyme, after dialysis with phen, was extensively dialysed against af®nity buffer with frequent changes of buffer. For regeneration of enzyme activity, the APCo-II solution (50 ml) was pre-incubated in 100 ml of af®nity buffer containing 2.5 mM of the various metals for 20 min at 37uC. The enzyme assay was initiated by adding the substrate, using the same conditions described in the section on enzyme assays. Kinetic studies All kinetic studies were carried out at 37t0.1uC using a D-160 Shimadzu-Spectrometer, employing a Peltier system. Reaction rates were determined in the continuous assay, and the release of 4nitroaniline was monitored by the change in absorbance at 405 nm. The kinetic parameters were determined using Lineweaver-Burk plots and a program for regression and variance analysis. RESULTS Puri®cation and molecular properties APCoII was puri®ed to homogeneity of S. cerevisiae in three stages of chromatography. Determination of APCo-II activity was made after the ®rst stage of molecular-exclusion chromatography in Sephadex G-200, as shown in Figure 1. The peak of activity of APCo-II was not APCo-II metal(s) extraction and reactivation Divalent ion(s) present in the puri®ed enzyme were removed by dialysis for 18 h at 4uC of af®nity buffer (20 ml) containing 5 mM 1,10-phenanthroline (phen). The dialysis buffer was concentrated (20r) in a rotary evaporator and UV±VIS spectral analysis was performed. An internal Copyright # 2000 John Wiley & Sons, Ltd. Figure 1. Gel permeation chromatography of aminopeptidase yscCo-II on Sephadex G-200. Aminopeptidase activity without cobalt ($) and with cobalt (+), protein concentration (#). The crude extract supernatant was applied to the Sephadex G200 column. 100 ml aliquots of each fraction were employed for activity assays. For details, see Materials and Methods. Yeast 2000; 16: 219±229. 222 Â Z-ALVAREZ I. HERRERA-CAMACHO, R. MORALES-MONTERROSAS AND R. QUIRO observed until CoCl2 was added to the lys-NA substrate. When the elution position of APCo-II on gel ®ltration was correlated with molecular mass, a value of approximately 290 kDa was obtained. The other peak to the right corresponds primarily to AP-yscII, with an Mr of 85±100 kDa (Frey and RoÈhm, 1978; Hirsch et al., 1989). Given the recently described group of aminopeptidases, and as a check for the fact that this is a different activity, the strain II-21 (MATa, lap1, lap2, lap3, lap4, leu2±3, leu2±112), a mutant in four aminopeptidase activities, was used as reference. After chromatography in Sephadex G200, the Mr peak of 290 kDa of our APCo-II was observed (data not shown), as well as a second broad peak of less than 90 kDa, which most probably corresponds to the recently described aminopeptidases AP-yscXVI (Tisljar and Wolf, 1993) or AP-yscY (Yasuhara et al., 1994). The AP-yscII activity was not detected in this strain due to the lap2 mutation. Subsequent steps in the puri®cation involved subjecting the sample to a af®nity chromatography on a pre-equilibrated Lys-Phe-AH-Sepharose4B resin. Figure 2 shows that the APCo-II was eluted at high ionic strength (0.1 M KCl) and acidic pH (5.5). The enzyme binds very strongly to the ligand and could not be separated by means of the substrate lys-NA (data not shown). This method is very selective, and approximately 95% of the protein contaminants were eliminated (Figure 2 and Table 1). Figure 3 shows the last stage of puri®cation, ion exchange chromatogra- Figure 2. Af®nity chromatography of aminopeptidase yscCoII on lysine-phenyl-AH-Sepharose 4B. Aminopeptidase activity (2) and protein concentration (#). The aminopeptidase fraction obtained by Sephadex G-200 chromatography was applied to the lys-phe-AH-Sepharose 4B column. 100 ml aliquots of each fraction were employed for activity assays. For details, see Materials and Methods. Copyright # 2000 John Wiley & Sons, Ltd. phy, in which the APCo-II is bound to a cationic resin and is eluted at low ion strength (0.05 M KCl) whereas the rest of the protein contaminants remain bound to the column. Table 1 shows the stages of puri®cation, with a good yield of 27% and a puri®cation factor of 160. This factor could be even higher if it were based on the crude extract, which is impossible because of the interference of the other aminopeptidases. PAGE in native condition shows a single lane of proteins (Figure 4, line 1) which, when cleaved and incubated with the substrate and cobalt, reveals the activity of APCo-II (data not shown). Running the sample under denaturing conditions (PAGE±SDS), we observed a single band of apparent molecular mass 48 kDa (Figure 4, line 3). Given the Mr of 290 kDa obtained by gel ®ltration (Sephadex G-200) and this single lane in PAGE±SDS of 48 kDa, we ®nd that APCo-II is a homo-hexamer protein. The PAGE and PAGE± SDS experiments show that the enzyme has been puri®ed to homogeneity. APCo-II is rather unstable after storage of the puri®ed enzyme at 4uC; 50% and 20% of the initial activity was assayed after 4 and 8 days, respectively. The only way to keep it active for 2 months at 4uC was to concentrate it and keep it precipitated in a suspension with 4 M ammonium sulphate in 20 mM MOPS/Tris, pH=7. Properties of expression; dependence on cobalt and pH The activity of APCo-II was detected after the ®rst stage of size-exclusion chromatography. It is impossible to detect the activity of APCo-II in a crude soluble extract without interference because of the other aminopeptidase activities that cleave N-terminal lysine substrates, and because its activity is stimulated by the presence of Co2+ (Achstetter et al., 1982; Herrera-Camacho, 1984; Tisljar and Wolf, 1993; Yasuhara et al., 1994). In order to determine how the expression of APCo-II activity varies in terms of cellular growth, soluble extracts were carried out in different stages of growth, and after a molecularexclusion chromatography (Sephadex G-200), enzymatic activity was determined. APCo-II activity varies in the different stages of growth, with a minimal expression (5%) in early exponential phase, increasing to maximum (100%) in the late exponential phase, and decreasing (25%) during the early stationary phase (data not Yeast 2000; 16: 219±229. 223 AMINOPEPTIDASE yscCo-II FROM YEAST Table 1. Puri®cation of aminopeptidase yscCo-II. Puri®cation step Activity (mU) Protein (mg) Speci®c activity (mU/mg) Yield (%) Puri®cation (-fold) Soluble extract Sephadex G-200 Lys-phe-AH-sepharose-4B DEAE±cellulose ± 134.1 99.6 36.5 89.7 5.32 0.22 0.009 ± 25.2 452.9 4051.7 ± 100.0 74.3 27.2 ± 1.0 17.9 160.7 shown). The optimum pH of APCo-II is 7 (Figure 5A), and its activity is strictly dependent on Co2+, with a maximum of activity at concentrations of 0.5 mM CoCl2 (Figure 5B). Due to these characteristics, it is different from the other known aminopeptidases; APCo-yscI (Achstetter et al., 1982) expresses itself primarily in the stationary phase, has a basic optimum pH (8.5), and presents its maximum activity at 0.05±0.1 mM CoCl2. In the case of AP-yscII and AP-yscY, enzymatic activities are not dependent on Co2+; AP-yscII stimulates its activity with Co2+ (Herrera-Camacho, 1984) and, in the case of AP-Y, Co2+ has different effects (activating or inhibiting), depending on the substrate used (Yasuhara et al., 1994). Substrate speci®city breaking substrates with N-terminal basic amino acids, such as lys-NA and arg-NA. With neutral acidic terminal amino acids, as well as substrates of carboxypeptidase yscY (BTNA) and substrates of dipeptidyl-aminopeptidases (x-pro-NA), the enzyme is unable to effect hydrolysis. The Michaelis±Menten constant determined with lys-NA and arg-NA as substrates and calculated from Lineweaver±Burk was found to be 0.16 mM for both. APCo-II hydrolyses lys-NA and arg-NA at the same rate. In the case of APCo-I, lys-NA is hydrolysed at a rate three times higher than for arg-NA (Achstetter et al., 1982). In the case of AP-Y, it hydrolyses a greater number of substrates, with a preference for N-terminal peptides and dipeptides, rather than N-terminal amino acids (Yasuhara et al., 1994). The effect of divalent metals on enzyme activity The substrate speci®city of APCo-II, was investigated using a broad spectrum of different synthetic chromogenic aminoacyl derivatives. Table 2 shows that APCo-II has a preference for Given the fact that APCo-II activity can only be detected in the presence of Co2+, and in order to ®nd out whether this activity were detectable when Figure 3. Ion exchange chromatography of aminopeptidase yscCo-II on DEAE±cellulose. Aminopeptidase activity (2) and protein concentration (#). The aminopeptidase fraction obtained by lys-phe-AH-Sepharose 4B chromatography was applied to DEAE-cellulose column. 100 ml aliquots of each fraction were employed for activity assays. For details, see Materials and Methods. Figure 4. Polyacrylamide gel electrophoresis (PAGE) of puri®ed aminopeptidase yscCo-II. Line 1: APCo-II puri®ed under native condition (PAGE). Line 2: molecular weight markers in SDS-PAGE (97 kDa, phosphorylase b; 66 kDa, bovine serum albumin; 45 kDa, egg albumin; 29 kDa, carbonic anhydrase). Line 3: APCo-II puri®ed under denaturalized condition (SDS±PAGE). The gel was stained with silver. For details, see Materials and Methods. Copyright # 2000 John Wiley & Sons, Ltd. Yeast 2000; 16: 219±229. 224 Â Z-ALVAREZ I. HERRERA-CAMACHO, R. MORALES-MONTERROSAS AND R. QUIRO Figure 5. Optimal conditions of aminopeptidase yscCo-II activity. (A) Effect of pH on APCo-II activity. The following buffers were used: Mes±Tris (5.5±6.5) (+), MOPS±Tris (6.5±7.5) (&), Tris±HCl (7.5±9.0) ($). (B) Effect of Co2+on the activity from APCo-II at pH=7.0. 0.5 mg protein, corresponding to homogeneity of APCo-II, were employed as the enzyme. For details of activity assays, see Materials and Methods. the Co2+ is replaced by another divalent metal, a study was carried out using different metal centres in the presence and absence of Co2+ (Table 3). Taking the activity shown with Co2+ as 100, we found that no other divalent cation of those assayed shows any important APCo-II activity (Table 3, ®rst column). Ca2+ and Mg2+ recover a minimal percentage (12%) of the activity, which suggests that APCo-II is strictly dependent on Co2+. On the other hand, when determining the activity of APCo-II in the presence of Co2+, the effect of the metal centres on enzyme activity can be observed. Table 3, last column, shows that the presence of Cu2+ totally inhibited APCo-II activity. The divalent metal ions Cr2+, Zn2+ and Ni2+ showed levels of inhibition of 83%, 70% and Table 2. Substrate yscCo-II. speci®city of aminopeptidase Substrate Relative hydrolysis (%) Lys-4-NA Arg-4-NA Pro-4-NA Leu-4-NA Glu-4-NA Ala-pro-4-NA Arg-pro-4-NA N-benzoyl-L-tyrosyl-4-NA 100.0 105.2 11.7 12.2 5.4 1.4 2.8 0.0 Activities against all substrates was tested with puri®ed enzymes and are expressed relative to the rate of hydrolysis of L-lysine-4-nitroanilide (100% was 4035 mU/mg). Activities were measured at pH 7.0 with 1 mM substrate and 0.5 mM CoCl2 at 37uC. Incubation for 30 min and enzyme activity was determined as described in Materials and Methods. Copyright # 2000 John Wiley & Sons, Ltd. 40%, respectively. Hg2+, Mg2+ and Mn2+ have only a very weak effect as inhibitors. In the case of Hg2+, the concentration must be increased ®ve times (5 mM) in order to produce a total inhibition of APCo-II activity. Different concentrations were used for each metal (chlorides and sulphates), and it could be observed that these have no signi®cant effect on the APCo-II response to the different metals. It is interesting to observe how Zn2+, which is a structural part of several enzymes, e.g. AP-yscI (Metz and RoÈhm, 1976), AP-Y (Yasuhara, 1994) and very probably also of AP-yscII (GarcõÂaAlvarez et al., 1991), behaves in others as an inhibitor, e.g. APCo-I (Achstetter et al., 1982), thiol-AP (Enenkel and Wolf, 1993), as well as this enzyme, APCo-II. Cu2+ proves to be a strong inhibitor of APCo-II as well as AP-II (HerreraCamacho, 1984) and AP-Y (Yasuhara et al., 1994), whereas in the case of APCo-I, it has no effect (Achstetter et al., 1982). Table 3. Divalent metal ion dependency of aminopeptidase yscCo-II activity and their effect. Cation added (conc. 1 mM) CoCl2 CoSO4 CaCl2 CaSO4 CuCl2 CuSO4 CrCl2 MgCl2 MgSO4 MnCl2 MnSO4 NiSO4 HgCl2 HgCl2 (5 mM) HgSO4 ZnCl2 ZnSO4 Activity (%) Without CoCl2 0 100* 99 14 12 3 2 5 12 11 4 3 1 0 0 0 3 2 With 1 mM CoCl2 100* 105 103 96 94 4 2 17 80 76 79 74 60 72 2 73 30 33 *Enzyme activity (control), 100% was 4019 mU/mg. For the ion dependency of APCo-II, the enzyme puri®ed was preincubated 10 min at 37uC in the presence of the respective metal ion(s) and, for the effect of divalent cation, the enzyme was pre-incubated with 1 mM CoCl2, after addition of respective metal ion(s). Yeast 2000; 16: 219±229. 225 AMINOPEPTIDASE yscCo-II FROM YEAST Effect of protease inhibitors The results of the study of APCo-II with different inhibitors are shown in Table 4. On assaying the effect of the chelating agents on APCo-II activity, we observed that 1,10-phenanthroline showed an inhibition of 90% at a concentration of 1 mM. At the same concentration, the other chelating agents (ANT, EDTA and chloroquine) show 50% inhibition. Inhibition with 1,10-phenanthroline is re-established when Co2+ is added (data not shown). Of the inhibitors of microbial origin, bestatin at concentrations of 4 mg/ml has a strong inhibiting effect (close to 100%) on APCo-II. At the same concentration, antipain and leupeptin produced a much weaker inhibition of 15% and 25%, respectively. In order to elicit an inhibition of 100% in the case of antipain, a concentration of 2.5 times higher is required. Bestatin proved to be a potent inhibitor of APCo-II and, in contrast, does not have or produce any effects on the activities of APCo-I, Table 4. thiol-AP, and AP-Y (Achstetter et al., 1982; Enenkel and Wolf, 1993; Yasuhara et al., 1994). Other inhibitors that interact directly with the Zn2+ of metallo-aminopeptidases are hydroxamates (Holmes and Mattheus, 1981), in our study leu-hydroxamate and lys-hydroxamate, which provoked an inhibition of 80% at 1 mM. With respect to inhibitors of sulphhydryl groups, p-HMB at concentrations of 1 mM did not have any effect on APCo-II activity and requires higher concentrations in order to produce an inhibition of 90%. This result is analogous to the effect produced with metal Hg2+, which has great af®nity for the sulphhydryl groups. PMSF and TLCK, inhibitors of serine proteases, inhibit APCo-II at concentrations of 1 mM with 100% and 50%, respectively. Given the strong inhibition with PMSF, it would be plausible to believe that residues of serine, tyrosine or threonine were involved in the enzyme catalysis. PMSF, a potent inhibitor of APCo-II, produces little or no effect on the activities of APCo-I, thiol-AP and AP-Y (Achstetter et al., Effect of protease inhibitors on aminopeptidase yscCo-II activity. Inhibitors Final concentration Relative speci®c activity (%) ± 1,10-Phenanthroline ± 1 mM 5 mM 1 mM 5 mM 1 mM 5 mM 1 mM 5 mM 4 mg/ml 10 mg/ml 4 mg/ml 10 mg/ml 4 mg/ml 10 mg/ml 1 mM 5 mM 1 mM 5 mM 1 mM 5 mM 1 mM 5 mM 1 mM 5 mM 100 9 0 51 21 53 10 52 5 86 7 6 0 73 49 27 14 19 13 99 10 5 0 43 6 Chloroquine EDTA Nitrilotriacetic acid Antipain Bestatin Leupeptin Leu-hydroxamate Lys-hydroxamate p-Hydroxymercuribenzoate (pHMB) Phenylmethylsulphonyl ¯uoride (PMSF) Tosyl-lysine chloromethyl ketone (TLCK) Puri®ed aminopeptidase yscCo-II (4087 mU/mg) was incubated in the presence of the indicated inhibitors for 30 min and enzyme activity was determined as described in Materials and Methods. Copyright # 2000 John Wiley & Sons, Ltd. Yeast 2000; 16: 219±229. 226 Â Z-ALVAREZ I. HERRERA-CAMACHO, R. MORALES-MONTERROSAS AND R. QUIRO 1982; Enenkel and Wolf, 1993; Yasuhara et al., 1994). APCo-II metal(s) extraction and regeneration of enzyme activity In order to examine the possible existence of a metallic centre in APCo-II, the enzyme was studied in presence of the chelating agent 1,10phenanthroline (phen). The endogenous metal ion(s) was removed by dialysis against phen and the resulting phen±Co coordination compound was followed spectrophotometrically. The results (shown in the Figure 6A) demonstrate that we have detected a band at 510 nm, corresponding to Co2+ d±d transitions (see control Co2+, Figure 6B), and also a charge-transfer band associated with UV±phen transitions (see control phen, Figure 6D). The same pattern is obtained in a phen±Co system which was used as an internal control (Figure 6C). These results indicate that APCo-II is a metallo-enzyme containing Co2+ in its structure. On the other hand, the APCo-II following interaction with phen (and after extensive dialysis against buffer) was inactive and showed a 100% recovery of activity after Co2+ addition (Table 5). Other metal ions like Ca2+, Cu2+, Mg2+, Mn2+, Ni2+ and Zn2+ could not restore the enzyme activity. These results indicate that the aminopeptidase activity is dependent only on the divalent cation Co2+. Kinetic studies of APCo-II Since the enzyme under study is noticeably cobalt-dependent, we studied its kinetic behaviour by measuring the velocity of product formation as a function of the substrate and of Co2+. The APCo-II Lineweaver-Burk plot, in the presence of different ®xed Co2+ concentrations (Figure 7), shows a family of lines with differents slopes that converge on the axis, 1/V0. The Vmax is independent of Co2+ concentrations; however, the Km decreases as Co2+ increases. The maximum effect of Co2+ on APCo-II activity was found at 0.5 mM, with no difference in Km differences at higher concentrations. The measured Km and Vmax values for APCo-II (at 0.5 mM of CoCl2) are 0.16 mM and 3.12 nmol/min, respectively. The data, shown in the form of a Lineweaver±Burk plot in Figure 7, are entirely consistent with a `competitive activation' where the Co2+ is an `essential activator' (Dixon and Webb, 1979). By plotting [Co] vs. 1/Kmap, a Co-apparent activation constant of Ka=0.28 mM was obtained (insert, Figure 7), a value that is very similar to the Km (0.16 mM), and supports the strict cobalt dependence of APCo-II. Table 5. cations. Figure 6. Aminopeptidase yscCo-II metal(s) extraction. (A) Spectrum of the concentrated dialisate after dialysis of APCoII with 1,10-phenanthroline. (B) CoCl2 spectrum. (C) (phen± CoCl2) complex spectrum. (D) 1,10-phenanthroline spectrum. For details, see Materials and Methods. Copyright # 2000 John Wiley & Sons, Ltd. Regeneration of enzyme activity by divalent Cation added (0.5 mM) Recovery activity (%) Control (x phen) Control (+ phen) CoCl2 CaCl2 CuCl2 MgCl2 MnCl2 NiCl2 ZnCl2 100* 2 109 2 0 3 5 0 0 *100% corresponding to activity of the nontreated enzyme, and is 4006 mU/mg. The enzyme, after dialysis with 1,10-phenanthroline (phen), was extensively dialysed against af®nity buffer. The APCo-II solution (50 ml) was pre-incubated in the presence of 2.5 mM of various divalent cations (100 ml) at 37uC for 20 min; the substrate was then added and the enzyme activity determined as described in Materials and Methods. Yeast 2000; 16: 219±229. AMINOPEPTIDASE yscCo-II FROM YEAST 1/Vo (nmol/min) Km(Co 0.5mM) = 0.161 mM Vmax = 3.12nmol/min Figure 7. Kinetic study of APCo-II by Cobalt. The initial data shown in Lineweaver±Burk transformation. CoCl2 concentration shown in the ®gure are (+) 0.1 mM, (+) 0.3 mM, (&) 0.5 mM and (2) 1 mM. A secondary plot showing the variation of (1/Kmap) with cobalt concentration is given in the inset ®gure. For details of kinetic analyses, see Materials and Methods. DISCUSSION APCo-II activity is strictly dependent on Co2+, which cannot replaced by any other divalent cation tested. The APCo-II activity is speci®c for N-terminal basic amino acid substrates. APCo-II was puri®ed to homogeneity from the cell extract by a three-step procedure. The puri®cation, as estimated from the speci®c activity, was approximately 160-fold, with a yield of 27%. This value was based on the original activity following ®ltration chromatography, since it is impossible to obtain an accurate measure of the enzyme's activity from the crude extract due to interference from the other aminopeptidases. The apparent molecular mass of the native enzyme, calculated after gel ®ltration data, was found to be 290 kDa. In contrast, under denaturing SDS±PAGE conditions, a single band of about 48 kDa was found. These data indicate that aminopeptidase yscCo-II is a hexameric enzyme made up of six identical subunits. Sequencing of the amino-terminal region of the puri®ed enzyme, using an Applied Biosystems 477A Protein Sequencer, showed that the N-terminus of the protein was blocked. In the budding yeast S. cerevisiae, with the exception of AP ysc-I (which has an Mr of 640 kDa; Frey and RoÈhm, 1978), APCo-II has an Mr which, at 290 kDa, is considerably higher than that of the other aminopeptidases described in the literature (Hirsch et al., 1989; Chang et al., Copyright # 2000 John Wiley & Sons, Ltd. 227 1990; Tisljar and Wolf, 1993; Yasuhara et al., 1994). The thiol aminopeptidase yeast (Enenkel and Wolf, 1993) is the one that comes closest in Mr (220 kDa); however, it differs clearly from APCo-II, since the latter has a wider substrate pro®le. Hydrolysing substrates with acidic, basic and neutral N-terminal amino acids, APCo-II is speci®c for basic N-terminal amino acids. Its inhibitor pattern, on the other hand, is very different, especially with bestatin and PMSF, which are potent inhibitors of APCo-II, and have no effect on thiol±AP (Enenkel and Wolf, 1993). The same inhibitors also have no signi®cant effect on APCo-I (Achstetter et al., 1982), AP± yscII (Herrera-Camacho, 1984) or on AP±Y (Yasuhara et al., 1994). Given the strong inhibition with PMSF, it would be plausible to believe that residues of serine, tyrosine or threonine were involved in the enzyme catalysis. Bestatin, an inhibitor of microbial origin, has been described as an inhibitor of metallo-enzymes (Salvesen and Nagase, 1989), interacting with the metal centres of these enzymes. In studies with aminopeptidases of Aeromonas, the metallic centre that has previously been replaced by the enzyme's Zn2+, and the direct interaction of bestatin with Co2+, has been demonstrated (Wilkes and Prescott, 1985). On the other hand, a study of the absorption spectrum of cobalt(II)-substituted Aeromonas aminopeptidase showed that it is markedly perturbed by the presence of amino acid-hydroxamates (Wilkes and Prescott, 1987). Later studies of these metallo-enzyme inhibitors claim that bestatin and the hydroxamates are necessary to know the reaction mechanism of APCo-II and its active centre. Due to inhibition by chelating agents and reactivation by ion metals, and the effect the other inhibitors, e.g. bestatin and hydroxamate, APCo-II is classi®ed as a metallo-enzyme. This seems to be the most common class for yeast aminopeptidases. A phen±Co coordination compound is formed after incubation of APCo-II with 1,10-phenanthroline; this result is entirely consistent with the presence of the cobalt in the molecular structure of APCo-II. The APCo-II activity is strict cobaltdependent, having a Co-apparent activation constant very similar to the Km. The results of the kinetic studies (Figure 7) clearly show a `competitive activation' in which Co2+ is an `essential activation' (Dixon and Webb, 1979). The characteristic mechanism of this type of activation established that the Co2+ binds to the free enzyme Yeast 2000; 16: 219±229. 228 Â Z-ALVAREZ I. HERRERA-CAMACHO, R. MORALES-MONTERROSAS AND R. QUIRO (E), and afterwards this complex (ECo) binds to substrate (S), in a obligatory order of reaction, written as: Ka E , ECo , ECoS These results lend support to the cobalt±metalloenzyme character for APCo-II, and the involvement of the Co2+ metal in the catalytic function. The recently described family of cobaltdependent enzymes, such as the methionine aminopeptidases, is of particular interest. Ar®n et al. (1995) described the structure of cobaltdependent methionine aminopeptidase from E. coli and identi®ed the motif that binds the active site cobalt ions (Roderick and Matthews, 1993). APCo-II possibly belongs to this family, since cobalt has been extracted from the structure of the protein. However, to establish unambigously the true coordination of cobalt in APCo-II and to identify the amino acids of the active site, metal analysis, X-ray crystallographic analysis, and sequencing of the amino-terminal region of the APCo-II is required. Such studies are under way in our laboratory. The studies described here of the new cobalt-dependent aminopeptidase, yscCo-II, provide an important starting point for these further investigations. ACKNOWLEDGEMENTS This work was supported in part by SEPDGICSA and CONACYT of MeÂxico. The authors are especially grateful to Steve Oliver for his help with English usage. REFERENCES Achstetter T, Ehmann C, Wolf DH. 1982. Aminopeptidase Co, a new yeast peptidase. Biochem Biophys Res Commun 109: 341±347. Achstetter T, Wolf DH. 1985. Proteinases, proteolysis and biological control in the yeast Saccharomyces cerevisiae. Yeast 1: 139±157. Ar®n SM, Kendall RL, Hall L, Weaver LH, Stewart AE, Matthews BW, Bradshaw RA. 1995. Eukaryotic methionyl aminopeptidases: two classes of cobaltdependent enzymes. Proc Natl Acad Sci U S A 92: 7714±7718. Bordallo J, SuaÂrez-Rendueles MP. 1995. Cis and transacting regulatory elements required for regulation of the CPS1 gene in Saccharomyces cerevisiae. Mol Gen Genet 246: 580±589. Bordallo J, Cueva R, SuaÂrez-Rendueles MP. 1995. Copyright # 2000 John Wiley & Sons, Ltd. Transcriptional regulation of the yeast vacuolar aminopeptidase yscI encoding gene (APE1) by carbon sources. FEBS Lett 364: 13±16. Chang YH, Teichert U, Smith JA. 1990. Puri®cation and characterization of a methionine aminopeptidase from Saccharomyces cerevisiae. J Biol Chem 265: 19892±19897. Cuevas R, GarcõÂa-Alvarez N, SuaÂrez-Rendueles MP. 1989. Yeast vacuolar aminopeptidase yscI. Isolation and regulation of the APE1 (LAP4) structural gene. FEBS Lett 259: 125±129. Dixon M, Webb EC. 1979. Enzymes, 3rd edn. Academic Press: New York. Enenkel C, Wolf DH. 1993. BLH1 codes for a yeast aminopeptidase, the equivalent of mammalian bleomycin hydrolase. J Biol Chem 268: 7036±7043. Frey J, RoÈhm KH. 1978. Subcellular localization and levels of aminopeptidases and dipeptidases in Saccharomyces cerevisiae. Biochim Biophys Acta 527: 31±41. Fujimura-Kamada K, Nouvet FJ, Michaelis S. 1997. A novel membrane-associated metalloprotease, Ste24p, is required for the ®rst step of NH2-terminal processing of the yeast a-factor precursor. J Cell Biol 136: 271±285. Fujiwara T, Tanaka K, Orino E, Yoshimura T, Kumatori A, Yamaguchi K, Shin K, Kakikazu A, Nakanishi S, Ichihara A. 1990. Proteosomes are essential for yeast proliferation. J Biol Chem 263: 16604±16613. Fuller RS, Sterne RE, Thorner J. 1988. Enzymes required for yeast prohormone processing. Ann Rev Physiol 50: 345±368. GarcõÂa-Alvarez N, Cueva R, SuaÂrez-Rendueles MP. 1991. Molecular cloning of soluble aminopeptidases from Saccharomyces cerevisiae. Sequence analysis of aminopeptidase yscII, a putative zinc-metallopeptidase. Eur J Biochem 202: 993±1002. Gilon T, Chomsky O, Kulka RG. 1998. Degradation signals for ubiquitin system proteolysis in Saccharomyces cerevisiae. EMBO J 17: 2759±2766. Groll M, Ditzel L, LoÈwe J, Stock D, Brochtler M, Bartunik HD, Huber R. 1997. Structure of 20S Ê resolution. Nature proteasome from yeast at 2.4 A 386: 463±471. Heinemeyer W, Kleinschmidt JA, Saidowsky J, Escher C, Wolf DH. 1991. Proteinase yscE, the yeast proteosome/multicatalytic proteinase: mutants unravel its necessity for cell survival. EMBO J 10: 555±562. Herrera-Camacho I. 1984. Mise en evidence de la modulation directe des activities proteolytiques de la levure Saccharomyces cerevisiae par des metabolites. TheÁse de Docteur Biologie Cellulaire et Moleculaire, UniversiteÂ Pierre et Marie Curie, France. Hirsch HH, SuaÂrez-Rendueles P, Wolf DH. 1989. Yeast Saccharomyces cerevisiae proteinases: structure, characteristics and function. In Molecular and Cell Yeast 2000; 16: 219±229. AMINOPEPTIDASE yscCo-II FROM YEAST Biology of Yeasts, Walton EF, Yarranton GT (eds). Blackie: London; 134±200. Holmes MA, Mattheus BW. 1981. Binding of hydroxamic acid inhibitors to crystalline thermolysin suggest a pentacoordinate zinc intermediate in catalysis. Biochemistry 20: 6912±6921. Hubbard SJ. 1998. The structural aspects of limited proteolysis of native proteins. Biochem Biophys Acta 1382: 191±206. Jones EE. 1991. Three proteolytic systems in the yeast Saccharomyces cerevisiae. J Biol Chem 266: 7963±7966. Klionsky DJ. 1998. Non-classical protein sorting to the yeast vacuole. J Biol Chem 273: 10807±10810. Laemmli UK. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227: 680±685. Lee DH, Goldberg AL. 1996. Selective inhibitors of the proteasome-dependent and vacuolar pathways of protein degradation in Saccharomyces cerevisiae. J Biol Chem 271: 27280±27284. Metz G, RoÈhm KH. 1976. Yeast aminopeptidase I. Chemical composition and catalytic properties. Biochem Biophys Acta 429: 933±949. Morrisey JH. 1981. Silver stain for proteins in polyacrylamide gels: a modi®ed procedure with enhanced uniform sensitivity. Anal Biochem 17: 307±310. Nishizawa M, Yasuhara T, Nakai T, Fujiki Y, Ohashi A. 1994. Molecular cloning of the aminopeptidase Y gene of Saccharomyces cerevisiae. J Biol Chem 269: 13651±13655. Peters JM. 1994. Proteasomes: protein degradation machines of the cell. Trends Biochem Soc 19: 377±382. Richter-Rouff B, Wolf DH. 1993. Proteosome and cell cycle. FEBS Lett 336: 34±36. Copyright # 2000 John Wiley & Sons, Ltd. 229 Rivett AJ. 1993. Proteosomes: multicatalytic proteinase complexes. Biochem J 291: 1±10. Roderick SL, Matthews BW. 1993. Structure of the cobalt-dependent methionine aminopeptidase from Escherichia coli: a new type of proteolytic enzyme. Biochemistry 32: 3907±3912. Salvesen G, Nagase H. 1989. Inhibition of proteolytic enzymes. In Proteolytic Enzymes, Beynon RJ, Bond JS (eds). IRL Press: Oxford; 83±104. Sedmak JJ, Grossberg SE. 1977. A rapid sensitive and versatile for protein using Coomassie Brilliant Blue G-250. Anal Biochem 79: 544±552. SuaÂrez-Rendueles MP, Schwencke J, Garcia-Alvarez N, Gascon S. 1981. A new X-prolyl-dipeptidyl aminopeptidase from yeast associated with a particulate fraction. FEBS Lett 131: 296±300. SuaÂrez-Rendueles MP, Wolf DH. 1988. Proteinase function in yeast: biochemical and genetic approaches to a central mechanism of post-translational control in the eukaryotic cell. FEMS Microbiol Rev 54: 17±46. Tisljar U, Wolf DH. 1993. Puri®cation and characterization of the cystinyl bond cleaving yeast aminopeptidase yscXVI. FEBS Lett 322: 191±196. Wilkes HS, Prescott JM. 1985. The slow tight binding of bestatin and amastatin to aminopeptidases. J Biol Chem 260: 13154±13162. Wilkes HS, Prescott JM. 1987. Hydroxamate-induced spectral perturbations of cobalt Aeromonas aminopeptidase. J Biol Chem 262: 8621±8625. Xu HE, Johnston SA. 1994. Yeast bleomycin hydrolase is a DNA-binding cysteine protease. J Biol Chem 269: 21177±21183. Yasuhara T, Nakai T, Ohashi A. 1994. Aminopeptidase Y, a new aminopeptidase from Saccharomyces cerevisiae. J Biol Chem 269: 13644±13650. Yeast 2000; 16: 219±229.