- Email: [email protected]
Cooperative kinetics of human prostatic acid phosphatase
Biochimica et Biophysica Acta 1548 (2001) 257^264 www.bba-direct.com
Cooperative kinetics of human prostatic acid phosphatase Ewa Luchter-Wasylewska * Institute of Medical Biochemistry, Jagiellonian University, Collegium Medicum, Kopernika 7, 31-034 Krako¨w, Poland Received 27 February 2001; received in revised form 11 June 2001; accepted 13 June 2001
Abstract The steady-state kinetics of hydrolysis reaction catalysed by human prostatic acid phosphatase (PAP) by using 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as substrates has been studied at pH 5.5. The substrate binding curves were sigmoidal and Hill cooperation coefficient h was higher than 1 for each of the examined compounds. Thus, human prostatic acid phosphatase kinetics exhibits positive cooperativity towards the studied substrates. The extent of cooperativity was found to depend on the substrate used and on enzyme concentration. The highest cooperativity of PAP was observed for 1-naphthyl phosphate and the lowest for phosphotyrosine. When prostatic phosphatase concentration increased, Hill cooperation coefficient (h) and half saturation constant (K0:5 ) both grew, but the catalytic constant (kcat ) remained constant, for each of the substrates studied. Ligand-induced association^dissociation equilibrium of the active oligomeric species (monomer-dimer-tetramer-oligomers) is suggested. ß 2001 Published by Elsevier Science B.V. Keywords: Human prostate acid phosphatase; Positive cooperativity; Ligand-induced association^dissociation; Phosphotyrosine; Phenyl phosphate; 1-Naphthyl phosphate
1. Introduction Human prostatic acid phosphatase (orthophosphoric-monoester phosphohydrolase, acid optimum; EC 3.1.3.2) is a dimeric glycoprotein of molecular mass of about 100 kDa synthesized by the prostate gland and secreted into seminal £uid. An elevated level of prostatic phosphatase in serum has been used as marker of prostatic cancer for many years [1,2]. Human prostatic acid phosphatase catalyses the transfer of phosphate group from monoester into water (hydrolysis) or into alcohol (transphosphorylaAbbreviations: PAP, human prostatic acid phosphatase * Fax: +48-12-422-3272. E-mail address: [email protected] (E. LuchterWasylewska).
tion) in acidic solution [1]. Many aryl and alkyl phosphate esters, including phosphoproteins at phosphoserine, phosphothreonine [3] and phosphotyrosine residues [4,5] may be substrates in hydrolysis. In vitro PAP dephosphorylates a speci¢c tyrosine kinase, the EGF receptor from prostate carcinoma cells [6], the prostatic cytosolic 83 kDa protein [7], and also stimulates the synthesis of collagen and the production of alkaline phosphatase in bone-cultured osteoblasts [8]. In addition, PAP is a carrier protein for ABH blood group antigens of semen [9]. It has been suggested that human prostatic acid phosphatase regulates the activity of protein tyrosine kinases and plays a role in androgen-mediated cell proliferation signalling and di¡erentiation or metabolic regulation in prostate gland in vivo [5,10^12]. In normal prostate epithelial cells dephosphorylation of tyrosine residues in ErbB-2 (transmembrane glycoprotein
0167-4838 / 01 / $ ^ see front matter ß 2001 Published by Elsevier Science B.V. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 2 3 9 - 4
BBAPRO 36470 2-8-01
258
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
pp185 with sequence homology to EGF receptor and with an intrinsic tyrosine kinase activity) by cellular PAP is pivotal in maintaining slow growth rate and androgen responsiveness. In prostate carcinomas cellular PAP expression is suppressed and ErbB-2 is constitutively tyrosine hyper-phosphorylated which leads to the activation of MAP kinase in clinical archival prostate tumour specimens. Thus the ¢rst insight into molecular mechanism by which the androgen-refractory stage of prostate cancer occurs is proposed [10^12]. However, the physiological function and the natural substrate of PAP have not been elucidated so far. The molecule of human prostatic acid phosphatase consists of two identical subunits connected by noncovalent bonds [13]. From burst titration experiments two active centres per dimeric molecule of the enzyme were found [14]. The primary sequence of the human prostatic acid phosphatase polypeptide chain was determined by cloning and characterizing of the PAP cDNA [15^17]. The three-dimensional structure of PAP has recently been deduced from X-ray crystallographic structure of the wild-type human (1cvi.pdb; 2hpa.pdb) and recombinant rat enzyme (1rpt.pdb; 1rpa.pdb). Each monomeric subunit consists of two domains: the large and the small. The active centre of PAP is located in a large open cleft between domains, which enables the enzyme to accept a large variety of substrates [18,19]. The active site contains a histidine residue transiently phosphorylated during the reaction, several positively charged arginine residues forming a phosphate binding pocket, aspartic acid residue acting as a proton donor and also several hydrophobic residues as tryptophan, tyrosine, phenylalanine, isoleucine and leucine exerting regulatory and stabilizing e¡ects [12,14,16,20^24]. It has been so far assumed that the kinetics of PAP-catalysed reaction obeys the Michaelis^Menten equation and therefore Km as well as Vmax values for many substrates were calculated using this equation. Moreover Bardsley [25] found that PAP-catalysed kinetics of p-nitrophenyl phosphate hydrolysis did not obey the Michaelis^Menten equation. The positive cooperativity with p-nitrophenyl phosphate as substrate was suggested, however, by Ishibe et al. [8]. The aim of the present research was to study in detail the steady-state kinetics of PAP-catalysed hydrolysis of the arti¢cial substrates, 1-naphthyl phos-
phate, phenyl phosphate and phosphotyrosine, in order to recognize the type of the kinetics ^ hyperbolic or sigmoidal ^ and to ¢nd out whether the kinetics of human prostatic acid phosphatase displays cooperativity. The initial rate of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine hydrolysis was determined in 0.1 M acetate bu¡er (pH 5.5) using previously prepared continuous spectrophotometric assays elaborated by the author [26^28]. In this work the dependence of initial reaction rate on substrate concentration was analysed by the Hill equation, which is more general than the Michaelis^ Menten one. 2. Materials and methods 2.1. Materials O-Phospho-DL-tyrosine and disodium salts of phenyl phosphate, 1-naphthyl phosphate (Sigma) as well as sodium acetate, acetic acid and Tris (Fluka) were used. All other chemicals were of the highest purity available and were used without prior puri¢cation. 2.2. Enzyme preparation Homogeneous human prostatic acid phosphatase was puri¢ed from human seminal plasma using the a¤nity method [29]. The molar concentration of phosphatase was determined by measuring the absorbance at 280 nm (A1% = 14.4), assuming the molecular mass 100 kDa [1]. 2.3. Continuous estimation of phosphatase activity The phosphatase activity determinations with 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as substrates [26^28] were performed at pH 5.5 in 0.1 M acetate bu¡er. The reaction was started by adding 0.05 ml of enzyme solution (in 10 mM Tris^HCl bu¡er with 100 mM NaCl, pH 7.4) into 0.95 ml of substrate solution in 0.1 M acetate bu¡er, pH 5.5. In a blank sample, the enzyme was not added. The reaction was carried out for 1 min at 20³C. The enzyme stock solution and the substrate stock solution were kept on ice during the experiment; so spontaneous substrate hydrolysis
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
could be neglected. The concentration range of each of the studied substrates was between 0.2 and 5 Km . The acid phosphatase-catalysed hydrolysis of substrates was monitored continuously by measuring the increase of absorbance of the reaction products, namely of 1-naphthol at 320 nm (vO = 2080 M31 cm31 ) [28], of phenol at 274 nm (O = 1090 M31 cm31 ) [27] and of tyrosine at 286 nm (vO = 760 M31 cm31 ) [26], respectively. Measurements were performed using a single beam UV-VIS Gilford Response spectrophotometer. The plots of reaction product concentration versus time were linear for substrate concentrations much higher than K0:5 but non-linear (convex) for lower substrate concentrations. 2.4. Data analysis
259
In this work the results were presented using substrate saturation (v0 versus [S]0 ), double-reciprocal (1/v0 versus 1/[S]0 ) and Eadie^Hofstee (v0 versus v0 / [S]0 ) plots. In all plots the initial reaction rates (v0 ) were divided by the enzyme concentration ([E]0 ), to present several curves on one graph. 3. Results The results of 1-naphthyl phosphate hydrolysis catalysed by phosphatase at 17.04 nM concentration are shown in Fig. 1 as an example. The double-reciprocal (Lineweaver^Burk) plot is concave-up and the Eadie^Hofstee plot is curved instead of them both being linear, as it was so far regarded. They prove the striking departure from classical Michea-
2.4.1. Initial reaction rate calculation The linear least squares and non-linear quadratic least squares curve-¢tting program of the Gilford Spectrophotometer were used for the initial reaction rate calculation. This program calculated parameters for both linear and non-linear (second order polynomial) regression. It evaluated also the values of variance (s2 ) and selected which ¢t, linear or secondorder, showed lower variance value. 2.4.2. Determination of steady-state reaction constants The initial reaction rate, calculated using the Gilford Spectrophotometer program, was estimated for every concentration of the enzyme from triplicate measurement over a wide range of substrate concentrations. In order to determine steady-state reaction constants (kcat , K0:5 and h) the experimental results were ¢tted, using EZ-Fit (Perrella Scienti¢c; http:// www.JLC.net/Vfperrell; [30]) or SigmaPlot (Jandel Scienti¢c) computer programs, to Hill rate equation: v0
kcat E0 Sh0 K h0:5 Sh0
where v0 is the initial reaction rate, [E]0 and [S]0 the initial concentrations of enzyme and substrate, h the Hill cooperation coe¤cient, kcat the catalytic constant (turnover number) and K0:5 the half saturation constant.
Fig. 1. E¡ect of substrate concentration on initial rate for hydrolysis of 1-naphthyl phosphate catalysed by human prostatic acid phosphatase at concentration 17.04 nM. (A) Lineweaver^ Burk plot ; (B) Eadie^Hofstee plot. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 320 nm [28] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the EZ-Fit program [30]. Substrate concentration is expressed in mM and the initial reaction rate (M s31 ) divided by enzyme concentration (M), in s31 .
BBAPRO 36470 2-8-01
260
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
Fig. 2. Substrate saturation curves for hydrolysis of 1-naphthyl phosphate catalysed by human prostatic acid phosphatase. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 320 nm [28] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the Sigma-Plot program. PAP concentrations: 4.26 nM (b), 17.04 nM (S), 34.08 nM (F), 68.16 nM (8).
lis^Menten kinetics and suggest that human prostatic acid phosphatase kinetics exhibits positive cooperativity. The saturation curves presented on Figs. 2 and 3, obtained at several enzyme concentrations for 1-naphthyl phosphate and phenyl phosphate as substrates, respectively, were found to be sigmoidal. The
Fig. 3. Substrate binding curves for hydrolysis of phenyl phosphate catalysed by human prostatic acid phosphatase. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 274 nm [27] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the Sigma-Plot program. PAP concentrations: 8.72 nM (b), 34.88 nM (S), 69.76 nM (F), 139.52 nM (8).
hydrolysis of phosphotyrosine catalysed by PAP displayed also the sigmoidal dependence of activity on substrate concentration (data not shown). Thus the kinetics of hydrolysis catalysed by human prostatic acid phosphatase for each of the studied phosphoesters is sigmoidal appearing to be positively cooperative (Figs. 1^3). The values of half saturation constant (K0:5 ), turnover number (kcat ) and Hill cooperation coe¤cient (h) for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine catalysed by human prostatic acid phosphatase calculated from Hill equation are presented in Table 1. For each of substrates studied, the values of K0:5 and h both increase when enzyme concentration rises. The increase of Hill coe¤cient indicates that the cooperativity grows with phosphatase concentration. The cooperativity also strongly depends on the substrate used. The value of kcat , on the other hand, is constant for each of the studied substrates over the whole range of enzyme concentrations. The initial rate of the catalysed reaction at saturation substrate concentration for each of the studied substrates grows linearly with phosphatase concentration, suggesting that the spe-
Fig. 4. Initial reaction rate for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as a function of phosphatase concentration. Hydrolysis reaction of 8 mM 1-naphthyl phosphate (b), phenyl phosphate (S) and phosphotyrosine (F), catalysed by PAP at di¡erent concentrations, was conducted in 0.1 M acetate bu¡er, pH 5.5. Phosphatase activity was assayed by continuous spectrophotometric measurement of the absorbance at 320 nm for 1-naphthyl phosphate [28], 274 nm for phenyl phosphate [27] and 286 nm for phosphotyrosine [26]. Initial reaction rate (expressed as dA min31 ) was calculated by the Gilford Spectrophotometer program. The best linear ¢t of plots was performed using the Sigma-Plot program.
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
261
Table 1 Kinetic parameters for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine catalysed by human prostatic acid phosphatase Substrate
Enzyme concentration [E]0 (nM)
Half saturation constant K0:5 (mM)
Turnover number kcat (s31 )
Hill concentration coe¤cient h
1-Naphthyl phosphate
2.13 4.26 8.52 17.04 34.08 68.16 2.18 4.36 8.72 17.44 34.88 69.76 139.52 6.82 34.10 170.50
0.046 þ 0.003 0.060 þ 0.002 0.124 þ 0.006 0.182 þ 0.003 0.390 þ 0.016 0.631 þ 0.102 0.106 þ 0.010 0.139 þ 0.082 0.155 þ 0.013 0.247 þ 0.014 0.459 þ 0.040 0.788 þ 0.045 1.617 þ 0.057 0.773 þ 0.144 1.952 þ 0.143 5.771 þ 0.301
503 þ 12 519 þ 6 519 þ 11 524 þ 2 511 þ 11 513 þ 8 462 þ 19 499 þ 13 494 þ 12 419 þ 11 457 þ 22 461 þ 16 468 þ 10 152 þ 7 124 þ 6 132 þ 6
1.66 þ 0.16 1.70 þ 0.17 1.91 þ 0.14 2.65 þ 0.05 2.75 þ 0.28 3.59 þ 0.18 1.10 þ 0.08 1.32 þ 0.08 1.51 þ 0.11 1.66 þ 0.14 1.73 þ 0.24 2.26 þ 0.26 2.28 þ 0.15 1.08 þ 0.21 1.36 þ 0.25 1.73 þ 0.27
Phenyl phosphate
Phosphotyrosine
Phosphatase activity was assayed in 0.1 M acetate bu¡er (pH 5.5) at 20³C for 1 min in continuous manner [26^28] and initial rate was calculated by the Gilford Spectrophotometer program. The experimental results were ¢tted to Hill equation using non-linear regression software EZ-Fit.
ci¢c activity of PAP is constant over a broad enzyme concentration range (Fig. 4). 4. Discussion Detailed studies on the cooperative properties of human prostatic acid phosphatase (PAP) have been performed for the ¢rst time. The binding of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine to PAP has been investigated in 0.1 M acetate bu¡er at pH 5.5, the assumed optimum pH of this enzyme for low-molecular substrates [1]. The results of the experiments shown in Figs. 1^3 and Table 1 give strong evidence that kinetics of human prostatic acid phosphatase is sigmoidal and display the positive homotropic cooperativity towards studied substrates [31,32]. The degree of cooperativity depends on the substrate used, and for each of them on the enzyme concentration. The previously developed continuous spectrophotometric acid phosphatase assays [26^28] allowed determination of the initial reaction rate more precisely than the discontinuous ones used thus far and calculation of the kinetic constants more accurate. Quick
continuous initial rate estimation, as well as application of computer programs for calculations of both initial rate and kinetic parameters made it possible to perform a great number of experiments. The conclusions drawn on the basis of such amount of experiments were therefore more reliable. The sigmoidal saturation curves shown in this study (Figs. 1^3) demonstrate that the Hill equation can be ¢tted to experimental data. The calculated Hill cooperation coe¤cient h (Table 1) was always more than 1 (1.08^3.6). The Michaelis^Menten equation, where h is equal to 1, which has been used in practically all of the previous studies on PAP, is therefore not adequate. It should be noted, however, that some deviations from the Michaelis^Menten kinetics have already been reported. Bardsley et al. [25] have shown that the kinetics of p-nitrophenyl phosphate hydrolysis at pH 7.0 was described with a function of degree 4:4 and Ishibe et al. [8] have demonstrated the positive cooperativity with the same substrate at pH 6.25 when convex curvilinear Eadie^Hofstee relationship (v0 versus v0 /[S]0 ) was obtained, similar to that for 1-naphthyl phosphate obtained in this study and presented on Fig. 1. Moreover, Van Etten [20] has stated that the Line-
BBAPRO 36470 2-8-01
262
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
weaver^Burk plot is unsatisfactory for kinetic parameters calculations. The Hill coe¤cient (Table 1) for each of the studied substrate increases with enzyme concentration: for low phosphatase concentrations it is between 1 and 2, whereas for high PAP concentrations it attains values between 3 and 4. Since the Hill coe¤cient cannot exceed the number of enzyme subunits [31,32], it therefore seems that the active molecule of human prostatic acid phosphatase at pH 5.5, depending on protein concentration, is built of one (monomer), two (dimer) or four (tetramer) active subunits (and their mixture), respectively. These results suggest that PAP exhibits properties of associating^dissociating enzyme system [33,34] and the equilibrium between the oligomeric forms in solution, according to Ostwald's law, is shifted by the protein concentration itself. The dependence of the initial reaction rate at saturating substrate concentration on enzyme concentration is linear for each of the studied substrates (Fig. 4), indicating that speci¢c activity is constant over a studied phosphatase concentration range. According to the results obtained in this study, the catalytic activity, kcat , for each of the studied substrates remains constant over a whole enzyme concentration range (Table 1), supporting the suggestion that each of the existing oligomeric forms of human prostatic acid phosphatase (monomer, dimer, tetramer) has the same speci¢c activity. By extrapolating the kinetic results presented in this paper (Table 1) below the studied protein concentration range, a suggestion can be made that at very low enzyme concentrations PAP could possess a Hill coe¤cient equal to 1, obey the Michaelis^Menten equation and exist as an active monomer. This hypothesis is in contradiction to the conclusion of Kuciel et al. [35] drawn from the reactivation rate order studies of denatured PAP, which inferred that the single subunit of PAP was not active. On the other hand, extrapolation of the results presented in this paper above the studied range suggests that at physiological concentration (1035 M) in seminal £uid [2] human prostatic acid phosphatase could exist as much larger oligomers and therefore PAP-catalysed kinetics could exhibit in vivo very high cooperativity. This study showed, moreover, that the degree of
cooperativity of human prostatic acid phosphatase kinetics substantially depends on the substrate used (Table 1): the Hill coe¤cient h is di¡erent for various substrates at the same enzyme concentration. It can be best observed at the PAP concentration of about 34 nM. The cooperativity, expressed as a value of the Hill cooperativity coe¤cient h, was the highest for 1-naphthyl phosphate and the lowest for phosphotyrosine. On the other hand, the a¤nity of PAP to the substrates, expressed as the value of K0:5 , was the highest for 1-naphthyl phosphate (lowest K0:5 ) and the lowest for phosphotyrosine (highest K0:5 ). PAP thus shows the highest a¤nity and the highest cooperativity with 1-naphthyl phosphate (I in Scheme 1) in which the phosphate group is bonded to non-polar, £at, two-ring naphthyl moiety of high polarizability. A smaller e¡ect was observed for phenyl phosphate (II), containing only one-ring phenyl moiety of lower polarizability, and the smallest for phosphotyrosine (III), containing strongly polar and charged alanyl fragment attached to the phenyl ring. Therefore the cooperativity of kinetics and the a¤nity of PAP to substrates both increase with growing hydrophobicity, increasing polarizability but with decreasing charge of the substrate molecule. The crystallographic analyses of enzyme^substrate analogue complexes could provide a detailed insight into the structural basis of phosphatase^ligand interaction. Previously it was established that the native active PAP molecule exists as stable dimer at pH 5^7 [1,13,16,18,19,36]. The essential di¡erences between the suggestions arising from kinetic studies presented in this paper and the previous results on PAP subunit structure may arise from the distinction in protein concentration of the studied sample. The molecular mass determinations were performed at high protein concentration. On the other hand, the kinetic
Scheme 1. The molecules of 1-naphthyl phosphate (I), phenyl phosphate (II) and phosphotyrosine (III).
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
experiments were conducted at low enzyme concentration. There is a need for detailed investigation on the concentration-dependent dissociation^association process of PAP in native conditions both in the absence and in the presence of ligand. Phosphate ester formation and hydrolysis, catalysed by kinases and phosphatases, are important processes linked to energy metabolism, to metabolic regulation and to a variety of cellular signal transduction pathways [37^39]. The phosphorylation level can be modulated by changes in the activities of either protein kinases or protein phosphatases. Hormones, growth factors and metabolites regulate their activities. Protein phosphatases do not simply constitutively reverse the e¡ects of protein kinases, but rather themselves play central and speci¢c roles in cellular physiology. The malfunctioning of protein phosphatases can result in severe defects such as the development of cancer [39]. Kinetics of the reaction catalysed by acid, alkaline and phosphoprotein phosphatases from various plant and animal tissues has been extensively studied. The regulation of activity of phosphatases may be performed in many di¡erent ways. Generally, alkaline phosphatases do not obey Michaelis^Menten kinetics, but acid phosphatases mostly do [40]. The examples of non-Michealian behaviour are derepressible and repressible acid phosphatases from the yeast Yarrowia lipolytica: the ¢rst exhibits kinetics described with rate equation of 2:2 minimum degree [41] and the second manifests negative cooperativity [42]. In some phosphoprotein phosphatases positive cooperativity is observed. The exocellular enzyme from the yeast Y. lipolytica is one example [43]. The human prostatic acid phosphatase, representing a distinct subgroup of protein tyrosine phosphatases, may play a key role in regulating the growth and androgen responsiveness of human prostate cancer cells by dephosphorylating ErbB-2, an in vivo substrate, on tyrosine residues [5,10^12]. The research presented above contributes new data to the kinetic behaviour of human prostatic acid phosphatase in vitro, increasing the possibility of learning about its in vivo biological role and its contribution in cellular as well as in fertilization processes. Further studies are required to clarify the mechanism of the PAP association^dissociation process and the activity regulation at in vivo conditions.
263
Acknowledgements The author is indebted to Dr Piotr M. Laidler for many helpful discussions. This work was supported by research grant BBN-501/ P/134/L from the Polish Committee for Scienti¢c Research (Komitet Badan Naukowych).
References [1] W. Ostrowski, Human prostatic acid phosphatase: physicochemical and catalytic properties, in: E. Spring-Mill, E.S.E. Hafez (Eds.), Male Accessory Sex Glands, Elsevier/North Holland, 1980, pp. 197^213. [2] W.S. Ostrowski, R. Kuciel, Clin. Chim. Acta 226 (1994) 121^129. [3] E. Wasylewska, J. Czubak, W.S. Ostrowski, Acta Biochim. Polon. 30 (1983) 175^184. [4] H.C. Li, J. Cherno¡, L.B. Chen, A. Kirschonbaum, Eur. J. Biochem. 138 (1984) 45^51. [5] M.F. Lin, G.M. Clinton, Adv. Prot. Phosphatases 4 (1987) 199^228. [6] M.F. Lin, G.M. Clinton, Mol. Cell. Biol. 8 (1988) 5477^ 5485. [7] H. Lee, T.M. Chu, C.L. Lee, Prostate 19 (1991) 251^ 263. [8] M. Ishibe, R.N. Rosier, J.E. Puzas, J. Clin. Endocrinol. Metab. 73 (1991) 785^792. [9] M. Nakazato, Tohoku J. Exp. Med. 174 (1994) 155^165. [10] T.C. Meng, M.F. Lin, J. Biol. Chem. 273 (1998) 22096^ 22104. [11] T.C. Meng, M.S. Lee, M.F. Lin, Oncogene 19 (2000) 2664^ 2677. [12] X.Q. Zhang, M.S. Lee, S. Zelivianski, M.F. Lin, J. Biol. Chem. 276 (2001) 2544^2550. [13] E. Luchter-Wasyl, W. Ostrowski, Biochim. Biophys. Acta 365 (1974) 349^359. [14] W. Ostrowski, E.A. Barnard, Biochemistry 12 (1973) 3893^ 3898. [15] P. Vihko, P. Virkkunen, P. Henttu, K. Roiko, T. Solin, M.L. Huhtala, FEBS Lett. 236 (1988) 275^281. [16] R.L. Van Etten, R. Davidson, P.E. Stevis, H. MacArthur, D.L. Moore, J. Biol. Chem. 266 (1991) 2313^2319. [17] F.S. Sharief, S.S.L. Li, Biochem. Biophys. Res. Commun. 184 (1992) 1468^1476. [18] G. Schneider, Y. Lindquist, P. Vihko, EMBO J. 12 (1993) 2609^2615. [19] C.G. Jacob, K. Lewinski, R. Kuciel, W. Ostrowski, L. Lebioda, Prostate 42 (2000) 211^218. [20] R.L. Van Etten, Ann. N. Y. Acad. Sci. 390 (1982) 27^50. [21] K. Ostanin, A. Saeed, R.L. Van Etten, J. Biol. Chem. 269 (1994) 8971^8978. [22] K.S. Porvari, A.M. Herralla, R.M. Kurkela, P.A. Taavitsai-
BBAPRO 36470 2-8-01
264
[23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264 nen, Y. Lindquist, G. Schneider, P.T. Vihko, J. Biol. Chem. 269 (1994) 22642^22646. Z. Zhang, K. Ostanin, R.L. Van Etten, Acta Biochim. Polon. 44 (1997) 659^672. K. Moller, R.L. Van Etten, Biospectroscopy 3 (1997) 85^96. W.G. Bardsley, P. Le¡, J. Kavanagh, R.D. Waight, Biochem. J. 187 (1980) 739^765. I. Apostol, R. Kuciel, E. Wasylewska, W.S. Ostrowski, Acta Biochim. Polon. 32 (1985) 187^197. E. Luchter-Wasylewska, Anal. Biochem. 241 (1996) 167^172. E. Luchter-Wasylewska, Acta Biochim. Polon. 44 (1997) 853^859. R.L. Van Etten, M. Saini, Clin. Chem. 24 (1978) 1525^1530. F.W. Perrella, Anal. Biochem. 174 (1988) 437^447. K.E. Neet, Methods Enzymol. 249 (1995) 519^567. L. Acerenza, E. Mizraji, Biochim. Biophys. Acta 1339 (1997) 155^166. B.I. Kurganov, Allostericheskije Fermenty (in Russian), Nauka, Moscow, 1978, pp. 117^174.
[34] T.W. Traut, Crit. Rev. Biochem. Mol. Biol. 29 (1994) 125^ 163. [35] R. Kuciel, A. Bakalova, A. Mazurkiewicz, A. Bilska, W. Ostrowski, Biochem. Int. 22 (1990) 329^334. [36] M. Derechin, W. Ostrowski, M. Galka, E.A. Barnard, Biochim. Biophys. Acta 250 (1971) 143^154. [37] J.M. Denu, J.A. Stuckey, M.A. Saper, J.E. Dixon, Cell 87 (1996) 361^364. [38] D. Barford, A.K. Das, M.P. Eglo¡, Annu. Rev. Biophys. Biomol. Struct. 27 (1998) 133^164. [39] S. Zolnierowicz, M. Bollen, EMBO J. 19 (2000) 483^488. [40] M.F. Holyaerts, T. Manes, J.L. Millan, J. Biol. Chem. 272 (1997) 22781^22787. [41] A. Moran, F.J. Burguillo, M.C. Lopez, A. Dominguez, Biochim. Biophys. Acta 990 (1989) 288^296. [42] F.J. Gonzalez, C. Fauste, F.J. Burguillo, A. Dominguez, Biochim. Biophys. Acta 1162 (1993) 17^27. [43] P. Jolivet, C. Queiroz-Claret, E. Bergeron, J.C. Meunier, Int. J. Biochem. Cell Biol. 30 (1998) 783^796.
BBAPRO 36470 2-8-01
Cooperative kinetics of human prostatic acid phosphatase Ewa Luchter-Wasylewska * Institute of Medical Biochemistry, Jagiellonian University, Collegium Medicum, Kopernika 7, 31-034 Krako¨w, Poland Received 27 February 2001; received in revised form 11 June 2001; accepted 13 June 2001
Abstract The steady-state kinetics of hydrolysis reaction catalysed by human prostatic acid phosphatase (PAP) by using 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as substrates has been studied at pH 5.5. The substrate binding curves were sigmoidal and Hill cooperation coefficient h was higher than 1 for each of the examined compounds. Thus, human prostatic acid phosphatase kinetics exhibits positive cooperativity towards the studied substrates. The extent of cooperativity was found to depend on the substrate used and on enzyme concentration. The highest cooperativity of PAP was observed for 1-naphthyl phosphate and the lowest for phosphotyrosine. When prostatic phosphatase concentration increased, Hill cooperation coefficient (h) and half saturation constant (K0:5 ) both grew, but the catalytic constant (kcat ) remained constant, for each of the substrates studied. Ligand-induced association^dissociation equilibrium of the active oligomeric species (monomer-dimer-tetramer-oligomers) is suggested. ß 2001 Published by Elsevier Science B.V. Keywords: Human prostate acid phosphatase; Positive cooperativity; Ligand-induced association^dissociation; Phosphotyrosine; Phenyl phosphate; 1-Naphthyl phosphate
1. Introduction Human prostatic acid phosphatase (orthophosphoric-monoester phosphohydrolase, acid optimum; EC 3.1.3.2) is a dimeric glycoprotein of molecular mass of about 100 kDa synthesized by the prostate gland and secreted into seminal £uid. An elevated level of prostatic phosphatase in serum has been used as marker of prostatic cancer for many years [1,2]. Human prostatic acid phosphatase catalyses the transfer of phosphate group from monoester into water (hydrolysis) or into alcohol (transphosphorylaAbbreviations: PAP, human prostatic acid phosphatase * Fax: +48-12-422-3272. E-mail address: [email protected] (E. LuchterWasylewska).
tion) in acidic solution [1]. Many aryl and alkyl phosphate esters, including phosphoproteins at phosphoserine, phosphothreonine [3] and phosphotyrosine residues [4,5] may be substrates in hydrolysis. In vitro PAP dephosphorylates a speci¢c tyrosine kinase, the EGF receptor from prostate carcinoma cells [6], the prostatic cytosolic 83 kDa protein [7], and also stimulates the synthesis of collagen and the production of alkaline phosphatase in bone-cultured osteoblasts [8]. In addition, PAP is a carrier protein for ABH blood group antigens of semen [9]. It has been suggested that human prostatic acid phosphatase regulates the activity of protein tyrosine kinases and plays a role in androgen-mediated cell proliferation signalling and di¡erentiation or metabolic regulation in prostate gland in vivo [5,10^12]. In normal prostate epithelial cells dephosphorylation of tyrosine residues in ErbB-2 (transmembrane glycoprotein
0167-4838 / 01 / $ ^ see front matter ß 2001 Published by Elsevier Science B.V. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 2 3 9 - 4
BBAPRO 36470 2-8-01
258
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
pp185 with sequence homology to EGF receptor and with an intrinsic tyrosine kinase activity) by cellular PAP is pivotal in maintaining slow growth rate and androgen responsiveness. In prostate carcinomas cellular PAP expression is suppressed and ErbB-2 is constitutively tyrosine hyper-phosphorylated which leads to the activation of MAP kinase in clinical archival prostate tumour specimens. Thus the ¢rst insight into molecular mechanism by which the androgen-refractory stage of prostate cancer occurs is proposed [10^12]. However, the physiological function and the natural substrate of PAP have not been elucidated so far. The molecule of human prostatic acid phosphatase consists of two identical subunits connected by noncovalent bonds [13]. From burst titration experiments two active centres per dimeric molecule of the enzyme were found [14]. The primary sequence of the human prostatic acid phosphatase polypeptide chain was determined by cloning and characterizing of the PAP cDNA [15^17]. The three-dimensional structure of PAP has recently been deduced from X-ray crystallographic structure of the wild-type human (1cvi.pdb; 2hpa.pdb) and recombinant rat enzyme (1rpt.pdb; 1rpa.pdb). Each monomeric subunit consists of two domains: the large and the small. The active centre of PAP is located in a large open cleft between domains, which enables the enzyme to accept a large variety of substrates [18,19]. The active site contains a histidine residue transiently phosphorylated during the reaction, several positively charged arginine residues forming a phosphate binding pocket, aspartic acid residue acting as a proton donor and also several hydrophobic residues as tryptophan, tyrosine, phenylalanine, isoleucine and leucine exerting regulatory and stabilizing e¡ects [12,14,16,20^24]. It has been so far assumed that the kinetics of PAP-catalysed reaction obeys the Michaelis^Menten equation and therefore Km as well as Vmax values for many substrates were calculated using this equation. Moreover Bardsley [25] found that PAP-catalysed kinetics of p-nitrophenyl phosphate hydrolysis did not obey the Michaelis^Menten equation. The positive cooperativity with p-nitrophenyl phosphate as substrate was suggested, however, by Ishibe et al. [8]. The aim of the present research was to study in detail the steady-state kinetics of PAP-catalysed hydrolysis of the arti¢cial substrates, 1-naphthyl phos-
phate, phenyl phosphate and phosphotyrosine, in order to recognize the type of the kinetics ^ hyperbolic or sigmoidal ^ and to ¢nd out whether the kinetics of human prostatic acid phosphatase displays cooperativity. The initial rate of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine hydrolysis was determined in 0.1 M acetate bu¡er (pH 5.5) using previously prepared continuous spectrophotometric assays elaborated by the author [26^28]. In this work the dependence of initial reaction rate on substrate concentration was analysed by the Hill equation, which is more general than the Michaelis^ Menten one. 2. Materials and methods 2.1. Materials O-Phospho-DL-tyrosine and disodium salts of phenyl phosphate, 1-naphthyl phosphate (Sigma) as well as sodium acetate, acetic acid and Tris (Fluka) were used. All other chemicals were of the highest purity available and were used without prior puri¢cation. 2.2. Enzyme preparation Homogeneous human prostatic acid phosphatase was puri¢ed from human seminal plasma using the a¤nity method [29]. The molar concentration of phosphatase was determined by measuring the absorbance at 280 nm (A1% = 14.4), assuming the molecular mass 100 kDa [1]. 2.3. Continuous estimation of phosphatase activity The phosphatase activity determinations with 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as substrates [26^28] were performed at pH 5.5 in 0.1 M acetate bu¡er. The reaction was started by adding 0.05 ml of enzyme solution (in 10 mM Tris^HCl bu¡er with 100 mM NaCl, pH 7.4) into 0.95 ml of substrate solution in 0.1 M acetate bu¡er, pH 5.5. In a blank sample, the enzyme was not added. The reaction was carried out for 1 min at 20³C. The enzyme stock solution and the substrate stock solution were kept on ice during the experiment; so spontaneous substrate hydrolysis
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
could be neglected. The concentration range of each of the studied substrates was between 0.2 and 5 Km . The acid phosphatase-catalysed hydrolysis of substrates was monitored continuously by measuring the increase of absorbance of the reaction products, namely of 1-naphthol at 320 nm (vO = 2080 M31 cm31 ) [28], of phenol at 274 nm (O = 1090 M31 cm31 ) [27] and of tyrosine at 286 nm (vO = 760 M31 cm31 ) [26], respectively. Measurements were performed using a single beam UV-VIS Gilford Response spectrophotometer. The plots of reaction product concentration versus time were linear for substrate concentrations much higher than K0:5 but non-linear (convex) for lower substrate concentrations. 2.4. Data analysis
259
In this work the results were presented using substrate saturation (v0 versus [S]0 ), double-reciprocal (1/v0 versus 1/[S]0 ) and Eadie^Hofstee (v0 versus v0 / [S]0 ) plots. In all plots the initial reaction rates (v0 ) were divided by the enzyme concentration ([E]0 ), to present several curves on one graph. 3. Results The results of 1-naphthyl phosphate hydrolysis catalysed by phosphatase at 17.04 nM concentration are shown in Fig. 1 as an example. The double-reciprocal (Lineweaver^Burk) plot is concave-up and the Eadie^Hofstee plot is curved instead of them both being linear, as it was so far regarded. They prove the striking departure from classical Michea-
2.4.1. Initial reaction rate calculation The linear least squares and non-linear quadratic least squares curve-¢tting program of the Gilford Spectrophotometer were used for the initial reaction rate calculation. This program calculated parameters for both linear and non-linear (second order polynomial) regression. It evaluated also the values of variance (s2 ) and selected which ¢t, linear or secondorder, showed lower variance value. 2.4.2. Determination of steady-state reaction constants The initial reaction rate, calculated using the Gilford Spectrophotometer program, was estimated for every concentration of the enzyme from triplicate measurement over a wide range of substrate concentrations. In order to determine steady-state reaction constants (kcat , K0:5 and h) the experimental results were ¢tted, using EZ-Fit (Perrella Scienti¢c; http:// www.JLC.net/Vfperrell; [30]) or SigmaPlot (Jandel Scienti¢c) computer programs, to Hill rate equation: v0
kcat E0 Sh0 K h0:5 Sh0
where v0 is the initial reaction rate, [E]0 and [S]0 the initial concentrations of enzyme and substrate, h the Hill cooperation coe¤cient, kcat the catalytic constant (turnover number) and K0:5 the half saturation constant.
Fig. 1. E¡ect of substrate concentration on initial rate for hydrolysis of 1-naphthyl phosphate catalysed by human prostatic acid phosphatase at concentration 17.04 nM. (A) Lineweaver^ Burk plot ; (B) Eadie^Hofstee plot. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 320 nm [28] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the EZ-Fit program [30]. Substrate concentration is expressed in mM and the initial reaction rate (M s31 ) divided by enzyme concentration (M), in s31 .
BBAPRO 36470 2-8-01
260
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
Fig. 2. Substrate saturation curves for hydrolysis of 1-naphthyl phosphate catalysed by human prostatic acid phosphatase. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 320 nm [28] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the Sigma-Plot program. PAP concentrations: 4.26 nM (b), 17.04 nM (S), 34.08 nM (F), 68.16 nM (8).
lis^Menten kinetics and suggest that human prostatic acid phosphatase kinetics exhibits positive cooperativity. The saturation curves presented on Figs. 2 and 3, obtained at several enzyme concentrations for 1-naphthyl phosphate and phenyl phosphate as substrates, respectively, were found to be sigmoidal. The
Fig. 3. Substrate binding curves for hydrolysis of phenyl phosphate catalysed by human prostatic acid phosphatase. Phosphatase activity was assayed in continuous manner by measuring the absorbance at 274 nm [27] in 0.1 M acetate bu¡er (pH 5.5) at 20³C. The initial rate of the reaction was calculated by the Gilford Spectrophotometer program. The curves drawn are the best-¢t curves calculated by Hill equation using the Sigma-Plot program. PAP concentrations: 8.72 nM (b), 34.88 nM (S), 69.76 nM (F), 139.52 nM (8).
hydrolysis of phosphotyrosine catalysed by PAP displayed also the sigmoidal dependence of activity on substrate concentration (data not shown). Thus the kinetics of hydrolysis catalysed by human prostatic acid phosphatase for each of the studied phosphoesters is sigmoidal appearing to be positively cooperative (Figs. 1^3). The values of half saturation constant (K0:5 ), turnover number (kcat ) and Hill cooperation coe¤cient (h) for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine catalysed by human prostatic acid phosphatase calculated from Hill equation are presented in Table 1. For each of substrates studied, the values of K0:5 and h both increase when enzyme concentration rises. The increase of Hill coe¤cient indicates that the cooperativity grows with phosphatase concentration. The cooperativity also strongly depends on the substrate used. The value of kcat , on the other hand, is constant for each of the studied substrates over the whole range of enzyme concentrations. The initial rate of the catalysed reaction at saturation substrate concentration for each of the studied substrates grows linearly with phosphatase concentration, suggesting that the spe-
Fig. 4. Initial reaction rate for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine as a function of phosphatase concentration. Hydrolysis reaction of 8 mM 1-naphthyl phosphate (b), phenyl phosphate (S) and phosphotyrosine (F), catalysed by PAP at di¡erent concentrations, was conducted in 0.1 M acetate bu¡er, pH 5.5. Phosphatase activity was assayed by continuous spectrophotometric measurement of the absorbance at 320 nm for 1-naphthyl phosphate [28], 274 nm for phenyl phosphate [27] and 286 nm for phosphotyrosine [26]. Initial reaction rate (expressed as dA min31 ) was calculated by the Gilford Spectrophotometer program. The best linear ¢t of plots was performed using the Sigma-Plot program.
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
261
Table 1 Kinetic parameters for hydrolysis of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine catalysed by human prostatic acid phosphatase Substrate
Enzyme concentration [E]0 (nM)
Half saturation constant K0:5 (mM)
Turnover number kcat (s31 )
Hill concentration coe¤cient h
1-Naphthyl phosphate
2.13 4.26 8.52 17.04 34.08 68.16 2.18 4.36 8.72 17.44 34.88 69.76 139.52 6.82 34.10 170.50
0.046 þ 0.003 0.060 þ 0.002 0.124 þ 0.006 0.182 þ 0.003 0.390 þ 0.016 0.631 þ 0.102 0.106 þ 0.010 0.139 þ 0.082 0.155 þ 0.013 0.247 þ 0.014 0.459 þ 0.040 0.788 þ 0.045 1.617 þ 0.057 0.773 þ 0.144 1.952 þ 0.143 5.771 þ 0.301
503 þ 12 519 þ 6 519 þ 11 524 þ 2 511 þ 11 513 þ 8 462 þ 19 499 þ 13 494 þ 12 419 þ 11 457 þ 22 461 þ 16 468 þ 10 152 þ 7 124 þ 6 132 þ 6
1.66 þ 0.16 1.70 þ 0.17 1.91 þ 0.14 2.65 þ 0.05 2.75 þ 0.28 3.59 þ 0.18 1.10 þ 0.08 1.32 þ 0.08 1.51 þ 0.11 1.66 þ 0.14 1.73 þ 0.24 2.26 þ 0.26 2.28 þ 0.15 1.08 þ 0.21 1.36 þ 0.25 1.73 þ 0.27
Phenyl phosphate
Phosphotyrosine
Phosphatase activity was assayed in 0.1 M acetate bu¡er (pH 5.5) at 20³C for 1 min in continuous manner [26^28] and initial rate was calculated by the Gilford Spectrophotometer program. The experimental results were ¢tted to Hill equation using non-linear regression software EZ-Fit.
ci¢c activity of PAP is constant over a broad enzyme concentration range (Fig. 4). 4. Discussion Detailed studies on the cooperative properties of human prostatic acid phosphatase (PAP) have been performed for the ¢rst time. The binding of 1-naphthyl phosphate, phenyl phosphate and phosphotyrosine to PAP has been investigated in 0.1 M acetate bu¡er at pH 5.5, the assumed optimum pH of this enzyme for low-molecular substrates [1]. The results of the experiments shown in Figs. 1^3 and Table 1 give strong evidence that kinetics of human prostatic acid phosphatase is sigmoidal and display the positive homotropic cooperativity towards studied substrates [31,32]. The degree of cooperativity depends on the substrate used, and for each of them on the enzyme concentration. The previously developed continuous spectrophotometric acid phosphatase assays [26^28] allowed determination of the initial reaction rate more precisely than the discontinuous ones used thus far and calculation of the kinetic constants more accurate. Quick
continuous initial rate estimation, as well as application of computer programs for calculations of both initial rate and kinetic parameters made it possible to perform a great number of experiments. The conclusions drawn on the basis of such amount of experiments were therefore more reliable. The sigmoidal saturation curves shown in this study (Figs. 1^3) demonstrate that the Hill equation can be ¢tted to experimental data. The calculated Hill cooperation coe¤cient h (Table 1) was always more than 1 (1.08^3.6). The Michaelis^Menten equation, where h is equal to 1, which has been used in practically all of the previous studies on PAP, is therefore not adequate. It should be noted, however, that some deviations from the Michaelis^Menten kinetics have already been reported. Bardsley et al. [25] have shown that the kinetics of p-nitrophenyl phosphate hydrolysis at pH 7.0 was described with a function of degree 4:4 and Ishibe et al. [8] have demonstrated the positive cooperativity with the same substrate at pH 6.25 when convex curvilinear Eadie^Hofstee relationship (v0 versus v0 /[S]0 ) was obtained, similar to that for 1-naphthyl phosphate obtained in this study and presented on Fig. 1. Moreover, Van Etten [20] has stated that the Line-
BBAPRO 36470 2-8-01
262
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
weaver^Burk plot is unsatisfactory for kinetic parameters calculations. The Hill coe¤cient (Table 1) for each of the studied substrate increases with enzyme concentration: for low phosphatase concentrations it is between 1 and 2, whereas for high PAP concentrations it attains values between 3 and 4. Since the Hill coe¤cient cannot exceed the number of enzyme subunits [31,32], it therefore seems that the active molecule of human prostatic acid phosphatase at pH 5.5, depending on protein concentration, is built of one (monomer), two (dimer) or four (tetramer) active subunits (and their mixture), respectively. These results suggest that PAP exhibits properties of associating^dissociating enzyme system [33,34] and the equilibrium between the oligomeric forms in solution, according to Ostwald's law, is shifted by the protein concentration itself. The dependence of the initial reaction rate at saturating substrate concentration on enzyme concentration is linear for each of the studied substrates (Fig. 4), indicating that speci¢c activity is constant over a studied phosphatase concentration range. According to the results obtained in this study, the catalytic activity, kcat , for each of the studied substrates remains constant over a whole enzyme concentration range (Table 1), supporting the suggestion that each of the existing oligomeric forms of human prostatic acid phosphatase (monomer, dimer, tetramer) has the same speci¢c activity. By extrapolating the kinetic results presented in this paper (Table 1) below the studied protein concentration range, a suggestion can be made that at very low enzyme concentrations PAP could possess a Hill coe¤cient equal to 1, obey the Michaelis^Menten equation and exist as an active monomer. This hypothesis is in contradiction to the conclusion of Kuciel et al. [35] drawn from the reactivation rate order studies of denatured PAP, which inferred that the single subunit of PAP was not active. On the other hand, extrapolation of the results presented in this paper above the studied range suggests that at physiological concentration (1035 M) in seminal £uid [2] human prostatic acid phosphatase could exist as much larger oligomers and therefore PAP-catalysed kinetics could exhibit in vivo very high cooperativity. This study showed, moreover, that the degree of
cooperativity of human prostatic acid phosphatase kinetics substantially depends on the substrate used (Table 1): the Hill coe¤cient h is di¡erent for various substrates at the same enzyme concentration. It can be best observed at the PAP concentration of about 34 nM. The cooperativity, expressed as a value of the Hill cooperativity coe¤cient h, was the highest for 1-naphthyl phosphate and the lowest for phosphotyrosine. On the other hand, the a¤nity of PAP to the substrates, expressed as the value of K0:5 , was the highest for 1-naphthyl phosphate (lowest K0:5 ) and the lowest for phosphotyrosine (highest K0:5 ). PAP thus shows the highest a¤nity and the highest cooperativity with 1-naphthyl phosphate (I in Scheme 1) in which the phosphate group is bonded to non-polar, £at, two-ring naphthyl moiety of high polarizability. A smaller e¡ect was observed for phenyl phosphate (II), containing only one-ring phenyl moiety of lower polarizability, and the smallest for phosphotyrosine (III), containing strongly polar and charged alanyl fragment attached to the phenyl ring. Therefore the cooperativity of kinetics and the a¤nity of PAP to substrates both increase with growing hydrophobicity, increasing polarizability but with decreasing charge of the substrate molecule. The crystallographic analyses of enzyme^substrate analogue complexes could provide a detailed insight into the structural basis of phosphatase^ligand interaction. Previously it was established that the native active PAP molecule exists as stable dimer at pH 5^7 [1,13,16,18,19,36]. The essential di¡erences between the suggestions arising from kinetic studies presented in this paper and the previous results on PAP subunit structure may arise from the distinction in protein concentration of the studied sample. The molecular mass determinations were performed at high protein concentration. On the other hand, the kinetic
Scheme 1. The molecules of 1-naphthyl phosphate (I), phenyl phosphate (II) and phosphotyrosine (III).
BBAPRO 36470 2-8-01
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264
experiments were conducted at low enzyme concentration. There is a need for detailed investigation on the concentration-dependent dissociation^association process of PAP in native conditions both in the absence and in the presence of ligand. Phosphate ester formation and hydrolysis, catalysed by kinases and phosphatases, are important processes linked to energy metabolism, to metabolic regulation and to a variety of cellular signal transduction pathways [37^39]. The phosphorylation level can be modulated by changes in the activities of either protein kinases or protein phosphatases. Hormones, growth factors and metabolites regulate their activities. Protein phosphatases do not simply constitutively reverse the e¡ects of protein kinases, but rather themselves play central and speci¢c roles in cellular physiology. The malfunctioning of protein phosphatases can result in severe defects such as the development of cancer [39]. Kinetics of the reaction catalysed by acid, alkaline and phosphoprotein phosphatases from various plant and animal tissues has been extensively studied. The regulation of activity of phosphatases may be performed in many di¡erent ways. Generally, alkaline phosphatases do not obey Michaelis^Menten kinetics, but acid phosphatases mostly do [40]. The examples of non-Michealian behaviour are derepressible and repressible acid phosphatases from the yeast Yarrowia lipolytica: the ¢rst exhibits kinetics described with rate equation of 2:2 minimum degree [41] and the second manifests negative cooperativity [42]. In some phosphoprotein phosphatases positive cooperativity is observed. The exocellular enzyme from the yeast Y. lipolytica is one example [43]. The human prostatic acid phosphatase, representing a distinct subgroup of protein tyrosine phosphatases, may play a key role in regulating the growth and androgen responsiveness of human prostate cancer cells by dephosphorylating ErbB-2, an in vivo substrate, on tyrosine residues [5,10^12]. The research presented above contributes new data to the kinetic behaviour of human prostatic acid phosphatase in vitro, increasing the possibility of learning about its in vivo biological role and its contribution in cellular as well as in fertilization processes. Further studies are required to clarify the mechanism of the PAP association^dissociation process and the activity regulation at in vivo conditions.
263
Acknowledgements The author is indebted to Dr Piotr M. Laidler for many helpful discussions. This work was supported by research grant BBN-501/ P/134/L from the Polish Committee for Scienti¢c Research (Komitet Badan Naukowych).
References [1] W. Ostrowski, Human prostatic acid phosphatase: physicochemical and catalytic properties, in: E. Spring-Mill, E.S.E. Hafez (Eds.), Male Accessory Sex Glands, Elsevier/North Holland, 1980, pp. 197^213. [2] W.S. Ostrowski, R. Kuciel, Clin. Chim. Acta 226 (1994) 121^129. [3] E. Wasylewska, J. Czubak, W.S. Ostrowski, Acta Biochim. Polon. 30 (1983) 175^184. [4] H.C. Li, J. Cherno¡, L.B. Chen, A. Kirschonbaum, Eur. J. Biochem. 138 (1984) 45^51. [5] M.F. Lin, G.M. Clinton, Adv. Prot. Phosphatases 4 (1987) 199^228. [6] M.F. Lin, G.M. Clinton, Mol. Cell. Biol. 8 (1988) 5477^ 5485. [7] H. Lee, T.M. Chu, C.L. Lee, Prostate 19 (1991) 251^ 263. [8] M. Ishibe, R.N. Rosier, J.E. Puzas, J. Clin. Endocrinol. Metab. 73 (1991) 785^792. [9] M. Nakazato, Tohoku J. Exp. Med. 174 (1994) 155^165. [10] T.C. Meng, M.F. Lin, J. Biol. Chem. 273 (1998) 22096^ 22104. [11] T.C. Meng, M.S. Lee, M.F. Lin, Oncogene 19 (2000) 2664^ 2677. [12] X.Q. Zhang, M.S. Lee, S. Zelivianski, M.F. Lin, J. Biol. Chem. 276 (2001) 2544^2550. [13] E. Luchter-Wasyl, W. Ostrowski, Biochim. Biophys. Acta 365 (1974) 349^359. [14] W. Ostrowski, E.A. Barnard, Biochemistry 12 (1973) 3893^ 3898. [15] P. Vihko, P. Virkkunen, P. Henttu, K. Roiko, T. Solin, M.L. Huhtala, FEBS Lett. 236 (1988) 275^281. [16] R.L. Van Etten, R. Davidson, P.E. Stevis, H. MacArthur, D.L. Moore, J. Biol. Chem. 266 (1991) 2313^2319. [17] F.S. Sharief, S.S.L. Li, Biochem. Biophys. Res. Commun. 184 (1992) 1468^1476. [18] G. Schneider, Y. Lindquist, P. Vihko, EMBO J. 12 (1993) 2609^2615. [19] C.G. Jacob, K. Lewinski, R. Kuciel, W. Ostrowski, L. Lebioda, Prostate 42 (2000) 211^218. [20] R.L. Van Etten, Ann. N. Y. Acad. Sci. 390 (1982) 27^50. [21] K. Ostanin, A. Saeed, R.L. Van Etten, J. Biol. Chem. 269 (1994) 8971^8978. [22] K.S. Porvari, A.M. Herralla, R.M. Kurkela, P.A. Taavitsai-
BBAPRO 36470 2-8-01
264
[23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
E. Luchter-Wasylewska / Biochimica et Biophysica Acta 1548 (2001) 257^264 nen, Y. Lindquist, G. Schneider, P.T. Vihko, J. Biol. Chem. 269 (1994) 22642^22646. Z. Zhang, K. Ostanin, R.L. Van Etten, Acta Biochim. Polon. 44 (1997) 659^672. K. Moller, R.L. Van Etten, Biospectroscopy 3 (1997) 85^96. W.G. Bardsley, P. Le¡, J. Kavanagh, R.D. Waight, Biochem. J. 187 (1980) 739^765. I. Apostol, R. Kuciel, E. Wasylewska, W.S. Ostrowski, Acta Biochim. Polon. 32 (1985) 187^197. E. Luchter-Wasylewska, Anal. Biochem. 241 (1996) 167^172. E. Luchter-Wasylewska, Acta Biochim. Polon. 44 (1997) 853^859. R.L. Van Etten, M. Saini, Clin. Chem. 24 (1978) 1525^1530. F.W. Perrella, Anal. Biochem. 174 (1988) 437^447. K.E. Neet, Methods Enzymol. 249 (1995) 519^567. L. Acerenza, E. Mizraji, Biochim. Biophys. Acta 1339 (1997) 155^166. B.I. Kurganov, Allostericheskije Fermenty (in Russian), Nauka, Moscow, 1978, pp. 117^174.
[34] T.W. Traut, Crit. Rev. Biochem. Mol. Biol. 29 (1994) 125^ 163. [35] R. Kuciel, A. Bakalova, A. Mazurkiewicz, A. Bilska, W. Ostrowski, Biochem. Int. 22 (1990) 329^334. [36] M. Derechin, W. Ostrowski, M. Galka, E.A. Barnard, Biochim. Biophys. Acta 250 (1971) 143^154. [37] J.M. Denu, J.A. Stuckey, M.A. Saper, J.E. Dixon, Cell 87 (1996) 361^364. [38] D. Barford, A.K. Das, M.P. Eglo¡, Annu. Rev. Biophys. Biomol. Struct. 27 (1998) 133^164. [39] S. Zolnierowicz, M. Bollen, EMBO J. 19 (2000) 483^488. [40] M.F. Holyaerts, T. Manes, J.L. Millan, J. Biol. Chem. 272 (1997) 22781^22787. [41] A. Moran, F.J. Burguillo, M.C. Lopez, A. Dominguez, Biochim. Biophys. Acta 990 (1989) 288^296. [42] F.J. Gonzalez, C. Fauste, F.J. Burguillo, A. Dominguez, Biochim. Biophys. Acta 1162 (1993) 17^27. [43] P. Jolivet, C. Queiroz-Claret, E. Bergeron, J.C. Meunier, Int. J. Biochem. Cell Biol. 30 (1998) 783^796.
BBAPRO 36470 2-8-01