- Email: [email protected]
Cold-induced enzyme inactivation: how does cooling lead to pyridoxal phosphate–aldimine bond cleavage in tryptophanase?
Biochimica et Biophysica Acta 1594 (2002) 335^340 www.bba-direct.com
Cold-induced enzyme inactivation: how does cooling lead to pyridoxal phosphate^aldimine bond cleavage in tryptophanase? Tali Erez a , Garik Ya. Gdalevsky a , Chithra Hariharan a , Dina Pines a , Ehud Pines a , Robert S. Phillips b , Rivka Cohen-Luria a , Abraham H. Parola a; * a
Department of Chemistry, Faculty of Natural Sciences, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel b Departments of Chemistry and of Biochemistry and Molecular Biology, University of Georgia, Athens, GA 30602, USA Received 7 August 2001; received in revised form 9 November 2001; accepted 20 November 2001
Abstract The phenomenon of cold scission or cold lability, which entails a widespread variety of oligomeric enzymes, is still enigmatic. The effect of cooling on the activity and the quaternary structure of the pyridoxal 5P-phosphate (PLP)-dependent enzyme, tryptophanase (Tnase), was studied utilizing single photon counting time-resolved spectrofluorometry. Upon cooling of holo-wild-type (wt) Tnase and its W330F mutant from 25³C to 2³C, a reduction in PLP fluorescence lifetime and rotational correlation time as well as inactivation and dissociation from tetramers to dimers were observed for both enzymes. Fluorescence anisotropy invariably decreased as a consequence of cooling, whether it was accompanied by a slight decrease in activity without significant dissociation, or by a substantial decrease in activity that was associated with either a partial or major dissociation. These results support the suggested conformational change that precedes the PLP^aldimine bond scission. It is proposed that cold inactivation is initiated by the weakening of hydrophobic interactions, leading to conformational changes which are the driving force for the aldimine bond cleavage. ß 2002 Elsevier Science B.V. All rights reserved. Keywords: Tryptophanase; W330F mutant tryptophanase; Pyridoxal phosphate; Protein quaternary structure; Time resolved £uorescence anisotropy; Single photon spectro£uorometry
1. Introduction Tryptophanase (tryptophan indole-lyase, EC 4.1.99.1) is a widely distributed bacterial PLP-depen-
Abbreviations: Tnase, tryptophanase; wt-Tnase, wild-type Tnase; W330F Tnase, W330F mutant Tnase; PLP, pyridoxal 5P-phosphate; SOPC, S-(o-nitrophenyl)-L-cysteine; Tricine, N-[2hydroxy-1,1-bis(hydroxymethyl)-ethyl]-glycine ; CD, circular dichroism * Corresponding author: Fax: +972-8-647-2943. E-mail address: [email protected] (A.H. Parola).
dent enzyme that catalyzes K,L-elimination and L-replacement reactions of L-tryptophan and of a variety of other L-substituted L-amino acids [1,2]. Tryptophanase consists of four identical 52 kDa monomers. Each monomer binds one molecule of PLP, which form an aldimine bond with a lysine residue. The dissociation into monomers occurs only at denaturing conditions or at pH above 8.7 [1]. In a previous study on Tnase from Escherichia coli we have shown that wt-Tnase and its W330F mutant undergo a reversible cold inactivation at low temperature or even at 25³C at low protein concentrations [3]. It is generally thought that cooling weakens hydrophobic in-
0167-4838 / 02 / $ ^ see front matter ß 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 3 2 5 - 9
BBAPRO 36546 7-3-02
336
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
teractions in proteins [4,5]. Accordingly, we proposed that cooling of holo-Tnase favors its conversion into an inactive conformation that releases PLP [6]. Thus, the conformational change in holo-Tnase leads to breaking of the aldimine bond, release of PLP and formation of the apoenzyme, and to dissociation of the apoenzyme into dimers [3]. The mechanism of PLP binding to the wt and the W330F mutant apotryptophanase was investigated using stopped-£ow kinetics [6]. Based on the microscopic reversibility principle we suggested that during the cold inactivation process of wt Tnases a conformational change precedes the aldimine bond cleavage, as shown in Scheme 1, where E is the apoenzyme, EPLP is a non-covalent complex, E-PLP is the `open' inactive internal PLP^lysine aldimine species, and (E-PLP)1 is the `closed' active species that has an isomerized aldimine [3,6]. Another conformational change detectable only for the W330F mutant occurs subsequent to the aldimine bond cleavage. This relatively facile conformational change is expected particularly for the W330F mutant, since the replacement of tryptophan by phenylalanine leads to reduced hydrophobic surface [7], which presumably results in increased £exibility. In the present study, utilizing single photon counting time-resolved spectro£uorometry, we investigated the e¡ect of cold inactivation on the £uorescence lifetime and anisotropy decay of Tnase-bound PLP. Our previous circumstantial evidence for the proposed conformational change ((E-PLP)1 to E-PLP, in Scheme 1), which precedes the aldimine bond cleavage, gains hereby additional support. We suggest that the loss in Tnase activity is concomitant with the proposed conformational change, which is then followed by the cold dissociation into dimers.
Scheme 1.
2. Materials and methods 2.1. Materials S-(o-Nitrophenyl)-L-cysteine (SOPC) was synthesized as described in Phillips et al. [8]. PLP, Tricine and 2-mercaptoethanol were purchased from Sigma. 2.2. Experimental methods 2.2.1. Preparation of enzymes Tnase was isolated from SVS 370 E. coli containing the wt and the site directed W330F mutant tnaA gene on plasmids, as described by Phillips and Gollnick [7,9]. Additional puri¢cation was achieved by chromatography on DEAE-Sephadex A-50, as described by Honda and Tokushige [10]. Protein concentration was determined following the absorbance at 278 nm, taking A1% values of 9.19 and 7.64 for holo-wt and W330F-Tnases, respectively [7,9]. 2.2.2. Enzymatic activity Activity was measured by the spectrophotometric method using the chromogenic substrate analog, S-(o-nitrophenyl)-L-cysteine (SOPC), as described by Suelter et al. [11]. Cold inactivation studies were carried out by incubating the wt and the mutated enzymes for 20 h at 2³C at a concentration of 1 mg/ml, either in 100 mM potassium phosphate or 50 mM Tricine^100 mM KCl bu¡er. Control samples incubated at 25³C fully preserved their activity. For activity measurements that were carried out at 25³C, 10^20-Wl enzyme aliquots were immediately stirred into 1 ml ¢nal volume of 100 mM potassium phosphate bu¡er (pH 8.0), 2 mM EDTA, 2 mM 2-mercaptoethanol or 1 mM dithiothreitol, preincubated at 25³C. Initial activity was measured for 1 min by following the decline in OD370 nm . This time period is short relative to the time for reactivation (forming reaction intermediates with PLP) of cold-inactivated apotryptophanase (t1=2 = 23 min) [3,6]. One unit of tryptophanase is de¢ned as the amount of enzyme required for the decomposition of 1 Wmol SOPC at 1 min at 25³C. 2.2.3. Single photon-counting spectro£uorometry Time-dependent £uorescence and £uorescence ani-
BBAPRO 36546 7-3-02
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
sotropy measurements were performed by using the time correlated single photon counting system [12,13]. The £uorescence was sampled during ¢xed accumulation times for the parallel, the perpendicular, and the magic angle relative orientation between the excitation and emission polarization. During £uorescence measurements, samples were gently mixed; temperature was maintained either at 25³C or 2³C, at the respective sample preincubation temperature. A thermostated cell holder that was stabilized to better than 0.05³C, by a Neslab (Portsmouth, NH) chiller, was used. 3. Results and discussion The e¡ect of cooling on the quaternary structure and the associated conformational changes of Tnase were the purpose of the present study. The phenomenon of cold scission or cold lability entails a widespread variety of oligomeric enzymes [14^30] including PLP-dependent enzymes [31]. At low temperatures and low concentrations these enzymes undergo a reversible loss of activity concomitant with
337
dissociation into their corresponding subunits. It is generally suggested that cooling evoke major changes in hydrophobic interactions that lead to destabilization of enzyme quaternary structure [4,5]. Fluorescence lifetime and rotational correlation time of holo-wt Tnase and its W330F mutant from E. coli, under conditions known to induce cold inactivation (e.g., at 2³C and low protein concentration, 1 mg/ ml), were measured. Excitation of the bound PLP at 350 nm, following emission at 400 nm (under these conditions free PLP is undetectable), resulted in the single photon correlation measurements data shown in Table 1. In addition, the associated percentage of cold inactivation and dissociation obtained from enzyme kinetics measurements and HPLC analyses [3] are shown too. Upon cooling from 25³C to 2³C, a reduction in £uorescence lifetime of holo-wt Tnase either in 100 mM potassium phosphate or in 50 mM Tricine^KCl bu¡er, and of its W330F mutant in 50 mM Tricine^ KCl bu¡er, was observed. This reduction could not be ascribed to a trivial explanation of a decline in di¡usional quenching, since the latter, which is common at low temperatures, usually leads to increased
Table 1 E¡ect of cooling on the £uorescence characteristics of PLPa bound to wt and W330F Tnaseb Bu¡er (pH 7.5)
T (³C)
dc (ns)
P1 (f1 ) (ns)
P2 (f2 ) (ns)
M2
Activity (%) Dissociation (%)d
I wt 100 mM potassium phosphate
II
III
IV
V
VI
VII
VIII
25 2 25 2
3.6 1.7 4.3 2.3
7.6 0.5 6.0 1.0
(0.35) (0.35) (0.34) (0.33)
0.021 0.025 0.062 0.063
(0.05) (0.05) (0.06) (0.07)
1.2 1.3 1.5 1.2
100e 75 þ 5f 100e 60 þ 3e
^ 5 5 ^ 20 þ 5
1 2 3 4
25 2 25 2
4.4 4.4 3.5 1.6
6.1 1.5 6.5 2.6
(0.38) (0.23) (0.37) (0.28)
0.121 0.124 0.125 0.125
(0.02) (0.17) (0.03) (0.12)
1.3 2.3 1.3 1.1
100e 10 þ 5f 100e 10 þ 2e
^ 20 þ 5 ^ 70 þ 5
5 6 7 8
50 mM Tricine^100 mM KCl
W330F 100 mM potassium phosphate 50 mM Tricine^100 mM KCl a
Vex = 350 nm; Vem = 400 nm. Protein concentration 1 mg/ml. c Time-resolved emission spectral data, obtained at the magic angle, were ¢tted to a discrete lifetime model. Data processing was done by the ISS Global analysis program (ISS, Champagne, IL). d Followed by HPLC at 5³C, taken from Erez et al. [3]. Dash indicates no signi¢cant dissociation. e Followed by enzyme kinetics, taken from Erez et al. [3]. Activity of 100% represents 42^50 and 40^46 Wmol min31 mg31 in wt and W330F Tnase, respectively, measured at 25³C, pH 8. See Section 2.2. f Followed by enzyme kinetics, obtained in the present study. b
BBAPRO 36546 7-3-02
338
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
(up to 10%) lifetime [32,33]. In our case we see a decrease in £uorescence lifetime, which could perhaps be accounted for by the higher exposure of PLP in the active site to bulk water. Cooling resulted in a marked decline in rotational correlation time values (Table 1: P1 (column IV), rows 2, 4, 6, 8 vs. 1, 3, 5, 7). The change in £uorescence rotational correlation time does not depend on the lifetime but rather on the freedom of PLP rotation. In all samples analyzed, there is a signi¢cant decline in the rotational correlation time upon cooling, regardless of whether the plugged magic-angle lifetime values during analysis decreased (Table 1, column III, rows 2, 4, 8 vs. 1, 3, 7) or remained unaltered (column III, row 6 vs. 5). The observed time resolved decay of anisotropy pro¢les were analyzed assuming two components of relaxation motion: a major rotational correlation time at the nanosecond time range, and a minor one at the picosecond time range. These rotational correlation time values are too short to re£ect changes associated even with the relatively low molecular mass monomeric protein of the 52 kDa. We assumed, as a rule of thumb, that 3 kDa of molecular mass corresponds to 1 ns in rotational correlation time [33]. Thus, the longest one, 7.6 ns, would be expected from a protein of 7.6U3 kDa = 22.8 kDa. This value is less than half of the molecular mass of the monomer, let alone a dimer or a tetramer with molecular masses of 104 and 208 kDa, respectively. Alternatively, the two components may be attributed to PLP rotational motion around the aldimine bond and to PLP segmental oscillation (wobbling in a cone) within the free volume available in the active site, in the nanosecond and picosecond time domain, respectively. When bound to the active site through the aldimine bond (i.e., in (E-PLP)1 ), PLP freedom of rotation and wobbling are further restricted by hydrogen bonding and salt bridges. In contrast, these restrictions on PLP rotational motion are alleviated in the relatively more `open' active site (E-PLP) following the conformational change suggested to precede the aldimine bond cleavage (E-PLP to EPLP). A contribution from free-dissociated PLP as a potential rotating body is excluded, since under the experimental conditions employed, free PLP £uorescence was undetectable. Moreover, the anisotropy invariably decreased as a consequence of cooling, whether it
was accompanied by a slight decrease in enzyme activity without signi¢cant dissociation (Table 1, column VII, row 2 vs. 1) or by a substantial decrease in enzyme activity that was associated with either a partial or major dissociation (Table 1, column VIII, rows 4, 6, 8 vs. 3, 5, 7). Therefore, one may conclude that PLP is fully sensing the environment at the active site, with its aldimine bond being either completely or at least partially intact, i.e., in either E-PLP or EPLP, respectively. Qualitatively, similar reductions in the rotational correlation times were observed regardless of whether the holo-Tnase is the wt or the W330F mutant, in spite of the di¡erences in the tendencies of these two proteins to undergo dissociation from a tetrameric into a dimeric structure (Table 1, column VIII). Thus, the data observed here presumably re£ect a pre-aldimine dissociation phenomenon, e.g., a conformational change ((E-PLP)1 to E-PLP) which is concomitant with the hydrogen and salt bonding relaxation. CD experiments did not reveal changes in the aromatic or peptidic region of either wt or W330FTnase during cold inactivation [3]. The present study, however, suggests that when one examines the active site directly through its £uorescent PLP cofactor, the rotational dynamics and lifetimes of PLP do sense changes which are cumulatively expressed primarily at the active site, yet are undetectable by CD which senses rather the Tnase backbone. To summarize, both changes in lifetime and rotational correlation time could be ascribed to the change induced by cooling from a `closed' (25³C, (E-PLP)1 ) to an `opened' (2³C, (E-PLP)) conformation of the active site. That this may take place was already proposed before [6]. Thus, the proposed conformational change that occurs prior to the aldimine bond cleavage alluded to in our stopped £ow kinetic study of PLP binding to apo-Tnase, gains further support [6]. In conclusion, it is suggested that cold inactivation is initiated by the weakening of hydrophobic interactions, leading to conformational changes which are the driving force for the aldimine bond cleavage. Acknowledgements We wish to express our gratitude and appreciation
BBAPRO 36546 7-3-02
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
to Professor Yuri M. Torchinsky for introducing us to the study of tryptophanase. This work was partially supported by the James Franck Center for Laser-Matter Interaction and by the O¤ce of Naval Research of the USA (Grant no. NOOO14-96-1-80) to A.H.P.
References
[15]
[16]
[17]
[1] E.E. Snell, Tryptophanase: structure, catalytic activities, and mechanism of action, Adv. Enzymol. 42 (1975) 287^333. [2] Yu.M. Torchinsky, Ya. Kawata, Tryptophanase: structural, spectral and catalytic properties, in: T. Fukui, K. Soda (Eds.), Molecular Aspects of Enzyme Catalysis, Kodansha, Tokyo/VCH, New York, 1994, pp. 165^190. [3] T. Erez, G.Ya. Gdalevsky, R.S. Phillips, Yu.M. Torchinsky, A.H. Parola, Cold inactivation and dissociation into dimers of Escherichia coli Tryptophanase and its W330F mutant form, Biochim. Biophys. Acta 1384 (1998) 365^372. [4] W. Kauzmann, Some factors in the interpretation of protein denaturation, Adv. Protein Chem. 14 (1959) 1^63. [5] P.E. Bock, C. Frieden, Another look at the cold lability of enzymes, Trends Biochem. Sci. 3 (1978) 100^103. [6] T. Erez, R.S. Phillips, A.H. Parola, Pyridoxal phosphate binding to wild type, W330F, and C298S mutants of Escherichia coli apotryptophanase: unraveling the cold inactivation, FEBS Lett. 433 (1998) 279^282. [7] R.S. Phillips, P.D. Gollnick, The environments of Trp-248 and Trp-330 in tryptophan indole-lyase from Escherichia coli, FEBS Lett. 268 (1990) 213^216. [8] R.S. Phillips, K. Ravichandran, R.L. Von Tersch, Synthesis of L-tyrosine from phenol and S-(o-nitrophenyl)-L-cysteine catalyzed by tyrosine phenol-lyase, Enzyme Microbiol. Technol. 11 (1989) 80^83. [9] R.S. Phillips, P.D. Gollnick, Evidence that cysteine 298 is in the active site of tryptophan indole-lyase, J. Biol. Chem. 264 (1989) 10627^10632. [10] T. Honda, M. Tokushige, E¡ects of the temperature and monovalent cations on the activity and quaternary structure of tryptophanase, J. Biochem. 100 (1986) 679^685. [11] C.H. Suelter, J. Wang, E.E. Snell, Direct spectrophotometric assay of tryptophanase, FEBS Lett. 66 (1976) 230^232. [12] E. Pines, D. Pines, T. Barak, B.-Z. Magnes, L.M. Tolbert, J.E. Haubrich, Isotope and temperature e¡ects in ultrafast proton-transfer from a strong excited-state acid, Ber. Bunsengesellsch. Phys. Chem. 102 (1998) 511^517. [13] M. Zamai, V.R. Caiolfa, D. Pines, E. Pines, A.H. Parola, Nature of interaction between basic ¢broblast growth factor and the antiangiogenic drug 7,7-(carbonyl-bis[imino-Nmethyl-4,2-pyrrolecarbonylimino[N-methyl-4,2-pyrrole]-carbonylimino])-bis-(1,3-naphthalene disulfonate), Biophys. J. 75 (1998) 672^682. [14] R. Shukuya, G.W. Schwert, Glutamic acid decarboxylase. 3.
[18]
[19]
[20]
[21] [22]
[23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
339
The inactivation of the enzyme at low temperatures, J. Biol. Chem. 235 (1960) 1658^1661. I.V. Velichiko, S.E. Volk, V.Yu. Dudarenkov, N.N. Magretova, V.Ya. Chernyak, A. Goldman, B.S. Cooperman, R. Lahti, A.A. Baykov, Cold lability of the mutant forms of Escherichia coli inorganic pyrophosphatase, FEBS Lett. 359 (1995) 20^22. J.J. Irias, M.R. Olmsted, M.F. Utter, Pyruvate carboxylase. Reversible inactivation by cold, Biochemistry 8 (1969) 5136^ 5148. R.S. Feldberg, P. Datta, Cold inactivation of L-threonine deaminase from Rhodospirillum rubrum; involvement of hydrophobic interaction, Eur. J. Biochem. 21 (1971) 447^454. R.A. Heller, R.G. Gould, Reversible cold inactivation of microsomal 3-hydroxy-3-methylglutaral coenzyme A reductase from rat liver, J. Biol. Chem. 249 (1974) 5254^5260. Y. Nakanishi, K. Okamoto, F. Isohashi, Subcellular distribution of ATP-stimulated and ADP-inhibited acetyl Co-A hydrase in livers from control and clo¢brate treated rats: comparison of the cytosolic and peroxisomal enzyme, J. Biochem. 115 (1994) 328^332. K. Nakashima, F.B. Rudolph, T. Wakabayashi, H.A. Lardy, Rat liver pyruvate carboxylase, V. Reversible dissociation by chloride salts of monovalent cations, J. Biol. Chem. 250 (1975) 331^336. L. King, G. Weber, Conformational drift of dissociated lactate dehydrogenase, Biochemistry 25 (1986) 3637^3640. R.V. Parry, J.C. Turner, P.A. Rea, High purity preparations of higher plant vacuolar H -ATPase reveal additional subunits. Revised subunit composition, J. Biol. Chem. 264 (1989) 20025^20032. Y. Moriyama, N. Nelson, Cold inactivation of vacuolar proton-ATPases, J. Biol. Chem. 264 (1989) 3577^3582. F. Brigani, W.P. Fong, D.S. Auld, B.L. Valle, In vitro dissociation and reassociation of human alcohol dehydrogenase class 1 isozymes, Biochemistry 28 (1989) 5374^5379. T.M. Huang, G.G. Chang, Characterization of the tetramerdimer-monomer equilibrium of the enzymatically active subunits of the pigeon liver malic enzyme, Biochemistry 31 (1992) 12658^12664. A.N. Fanjul, R.N. Farias, Cold-sensitive cytosolic 3,5,3Ptriiodo-L-thyronine-binding protein and pyruvate kinase from human erythrocytes share similar regulatory properties of hormone binding by glycolytic intermediates, J. Biol. Chem. 268 (1993) 175^179. J.P. Krall, G.E. Edwards, PEP carboxylases from two C4 species of Panicum with markedly di¡erent susceptibilities to cold inactivation, Plant Cell. Physiol. 34 (1993) 1^11. K.E. Hightower, R.E. McCarty, In£uence of nucleotides on the cold stability of chloroplast coupling factor, Biochemistry 35 (1996) 10051^10057. B. Tomkinson, Association and dissociation of the tripeptidyl-peptidase II complex as a way of regulating the enzyme activity, Arch. Biochem. Biophys. 376 (2000) 275^280. M. Isupov, A.A. Antson, E.J. Dodson, G.G. Dodson, I.S. Dementieva, L.N. Zakomirdina, K.S. Wilson, Z. Dauter,
BBAPRO 36546 7-3-02
340
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
A.A. Lebedev, E.H. Harutyunyan, Crystal structure of tryptophanase, J. Mol. Biol. 276 (1998) 603^623. [31] E.E. Snell, S.J. Di Mari, Schi¡ base intermediates in enzyme catalysis, in: P.D. Boyer (Ed.), The Enzymes, Vol. 2, Academic Press, New York, 1970, pp. 335^370. [32] P. Fuchs, A. Parola, P.W. Robbins, E.R. Blout, Microvis-
cosity of the membrane lipids of normal, transformed and revertant 3T3 cells, Proc. Natl. Acad. Sci. USA 72 (1975) 3351^3354. [33] A.H. Parola, Membranes lipid^protein interactions, in: M. Shinitzky (Ed.), Biomembranes, Physical Aspects, Vol. 2, Balaban Publishers VCH, Brooklyn, NY, 1993, pp. 159^277.
BBAPRO 36546 7-3-02
Cold-induced enzyme inactivation: how does cooling lead to pyridoxal phosphate^aldimine bond cleavage in tryptophanase? Tali Erez a , Garik Ya. Gdalevsky a , Chithra Hariharan a , Dina Pines a , Ehud Pines a , Robert S. Phillips b , Rivka Cohen-Luria a , Abraham H. Parola a; * a
Department of Chemistry, Faculty of Natural Sciences, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84105, Israel b Departments of Chemistry and of Biochemistry and Molecular Biology, University of Georgia, Athens, GA 30602, USA Received 7 August 2001; received in revised form 9 November 2001; accepted 20 November 2001
Abstract The phenomenon of cold scission or cold lability, which entails a widespread variety of oligomeric enzymes, is still enigmatic. The effect of cooling on the activity and the quaternary structure of the pyridoxal 5P-phosphate (PLP)-dependent enzyme, tryptophanase (Tnase), was studied utilizing single photon counting time-resolved spectrofluorometry. Upon cooling of holo-wild-type (wt) Tnase and its W330F mutant from 25³C to 2³C, a reduction in PLP fluorescence lifetime and rotational correlation time as well as inactivation and dissociation from tetramers to dimers were observed for both enzymes. Fluorescence anisotropy invariably decreased as a consequence of cooling, whether it was accompanied by a slight decrease in activity without significant dissociation, or by a substantial decrease in activity that was associated with either a partial or major dissociation. These results support the suggested conformational change that precedes the PLP^aldimine bond scission. It is proposed that cold inactivation is initiated by the weakening of hydrophobic interactions, leading to conformational changes which are the driving force for the aldimine bond cleavage. ß 2002 Elsevier Science B.V. All rights reserved. Keywords: Tryptophanase; W330F mutant tryptophanase; Pyridoxal phosphate; Protein quaternary structure; Time resolved £uorescence anisotropy; Single photon spectro£uorometry
1. Introduction Tryptophanase (tryptophan indole-lyase, EC 4.1.99.1) is a widely distributed bacterial PLP-depen-
Abbreviations: Tnase, tryptophanase; wt-Tnase, wild-type Tnase; W330F Tnase, W330F mutant Tnase; PLP, pyridoxal 5P-phosphate; SOPC, S-(o-nitrophenyl)-L-cysteine; Tricine, N-[2hydroxy-1,1-bis(hydroxymethyl)-ethyl]-glycine ; CD, circular dichroism * Corresponding author: Fax: +972-8-647-2943. E-mail address: [email protected] (A.H. Parola).
dent enzyme that catalyzes K,L-elimination and L-replacement reactions of L-tryptophan and of a variety of other L-substituted L-amino acids [1,2]. Tryptophanase consists of four identical 52 kDa monomers. Each monomer binds one molecule of PLP, which form an aldimine bond with a lysine residue. The dissociation into monomers occurs only at denaturing conditions or at pH above 8.7 [1]. In a previous study on Tnase from Escherichia coli we have shown that wt-Tnase and its W330F mutant undergo a reversible cold inactivation at low temperature or even at 25³C at low protein concentrations [3]. It is generally thought that cooling weakens hydrophobic in-
0167-4838 / 02 / $ ^ see front matter ß 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 3 2 5 - 9
BBAPRO 36546 7-3-02
336
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
teractions in proteins [4,5]. Accordingly, we proposed that cooling of holo-Tnase favors its conversion into an inactive conformation that releases PLP [6]. Thus, the conformational change in holo-Tnase leads to breaking of the aldimine bond, release of PLP and formation of the apoenzyme, and to dissociation of the apoenzyme into dimers [3]. The mechanism of PLP binding to the wt and the W330F mutant apotryptophanase was investigated using stopped-£ow kinetics [6]. Based on the microscopic reversibility principle we suggested that during the cold inactivation process of wt Tnases a conformational change precedes the aldimine bond cleavage, as shown in Scheme 1, where E is the apoenzyme, EPLP is a non-covalent complex, E-PLP is the `open' inactive internal PLP^lysine aldimine species, and (E-PLP)1 is the `closed' active species that has an isomerized aldimine [3,6]. Another conformational change detectable only for the W330F mutant occurs subsequent to the aldimine bond cleavage. This relatively facile conformational change is expected particularly for the W330F mutant, since the replacement of tryptophan by phenylalanine leads to reduced hydrophobic surface [7], which presumably results in increased £exibility. In the present study, utilizing single photon counting time-resolved spectro£uorometry, we investigated the e¡ect of cold inactivation on the £uorescence lifetime and anisotropy decay of Tnase-bound PLP. Our previous circumstantial evidence for the proposed conformational change ((E-PLP)1 to E-PLP, in Scheme 1), which precedes the aldimine bond cleavage, gains hereby additional support. We suggest that the loss in Tnase activity is concomitant with the proposed conformational change, which is then followed by the cold dissociation into dimers.
Scheme 1.
2. Materials and methods 2.1. Materials S-(o-Nitrophenyl)-L-cysteine (SOPC) was synthesized as described in Phillips et al. [8]. PLP, Tricine and 2-mercaptoethanol were purchased from Sigma. 2.2. Experimental methods 2.2.1. Preparation of enzymes Tnase was isolated from SVS 370 E. coli containing the wt and the site directed W330F mutant tnaA gene on plasmids, as described by Phillips and Gollnick [7,9]. Additional puri¢cation was achieved by chromatography on DEAE-Sephadex A-50, as described by Honda and Tokushige [10]. Protein concentration was determined following the absorbance at 278 nm, taking A1% values of 9.19 and 7.64 for holo-wt and W330F-Tnases, respectively [7,9]. 2.2.2. Enzymatic activity Activity was measured by the spectrophotometric method using the chromogenic substrate analog, S-(o-nitrophenyl)-L-cysteine (SOPC), as described by Suelter et al. [11]. Cold inactivation studies were carried out by incubating the wt and the mutated enzymes for 20 h at 2³C at a concentration of 1 mg/ml, either in 100 mM potassium phosphate or 50 mM Tricine^100 mM KCl bu¡er. Control samples incubated at 25³C fully preserved their activity. For activity measurements that were carried out at 25³C, 10^20-Wl enzyme aliquots were immediately stirred into 1 ml ¢nal volume of 100 mM potassium phosphate bu¡er (pH 8.0), 2 mM EDTA, 2 mM 2-mercaptoethanol or 1 mM dithiothreitol, preincubated at 25³C. Initial activity was measured for 1 min by following the decline in OD370 nm . This time period is short relative to the time for reactivation (forming reaction intermediates with PLP) of cold-inactivated apotryptophanase (t1=2 = 23 min) [3,6]. One unit of tryptophanase is de¢ned as the amount of enzyme required for the decomposition of 1 Wmol SOPC at 1 min at 25³C. 2.2.3. Single photon-counting spectro£uorometry Time-dependent £uorescence and £uorescence ani-
BBAPRO 36546 7-3-02
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
sotropy measurements were performed by using the time correlated single photon counting system [12,13]. The £uorescence was sampled during ¢xed accumulation times for the parallel, the perpendicular, and the magic angle relative orientation between the excitation and emission polarization. During £uorescence measurements, samples were gently mixed; temperature was maintained either at 25³C or 2³C, at the respective sample preincubation temperature. A thermostated cell holder that was stabilized to better than 0.05³C, by a Neslab (Portsmouth, NH) chiller, was used. 3. Results and discussion The e¡ect of cooling on the quaternary structure and the associated conformational changes of Tnase were the purpose of the present study. The phenomenon of cold scission or cold lability entails a widespread variety of oligomeric enzymes [14^30] including PLP-dependent enzymes [31]. At low temperatures and low concentrations these enzymes undergo a reversible loss of activity concomitant with
337
dissociation into their corresponding subunits. It is generally suggested that cooling evoke major changes in hydrophobic interactions that lead to destabilization of enzyme quaternary structure [4,5]. Fluorescence lifetime and rotational correlation time of holo-wt Tnase and its W330F mutant from E. coli, under conditions known to induce cold inactivation (e.g., at 2³C and low protein concentration, 1 mg/ ml), were measured. Excitation of the bound PLP at 350 nm, following emission at 400 nm (under these conditions free PLP is undetectable), resulted in the single photon correlation measurements data shown in Table 1. In addition, the associated percentage of cold inactivation and dissociation obtained from enzyme kinetics measurements and HPLC analyses [3] are shown too. Upon cooling from 25³C to 2³C, a reduction in £uorescence lifetime of holo-wt Tnase either in 100 mM potassium phosphate or in 50 mM Tricine^KCl bu¡er, and of its W330F mutant in 50 mM Tricine^ KCl bu¡er, was observed. This reduction could not be ascribed to a trivial explanation of a decline in di¡usional quenching, since the latter, which is common at low temperatures, usually leads to increased
Table 1 E¡ect of cooling on the £uorescence characteristics of PLPa bound to wt and W330F Tnaseb Bu¡er (pH 7.5)
T (³C)
dc (ns)
P1 (f1 ) (ns)
P2 (f2 ) (ns)
M2
Activity (%) Dissociation (%)d
I wt 100 mM potassium phosphate
II
III
IV
V
VI
VII
VIII
25 2 25 2
3.6 1.7 4.3 2.3
7.6 0.5 6.0 1.0
(0.35) (0.35) (0.34) (0.33)
0.021 0.025 0.062 0.063
(0.05) (0.05) (0.06) (0.07)
1.2 1.3 1.5 1.2
100e 75 þ 5f 100e 60 þ 3e
^ 5 5 ^ 20 þ 5
1 2 3 4
25 2 25 2
4.4 4.4 3.5 1.6
6.1 1.5 6.5 2.6
(0.38) (0.23) (0.37) (0.28)
0.121 0.124 0.125 0.125
(0.02) (0.17) (0.03) (0.12)
1.3 2.3 1.3 1.1
100e 10 þ 5f 100e 10 þ 2e
^ 20 þ 5 ^ 70 þ 5
5 6 7 8
50 mM Tricine^100 mM KCl
W330F 100 mM potassium phosphate 50 mM Tricine^100 mM KCl a
Vex = 350 nm; Vem = 400 nm. Protein concentration 1 mg/ml. c Time-resolved emission spectral data, obtained at the magic angle, were ¢tted to a discrete lifetime model. Data processing was done by the ISS Global analysis program (ISS, Champagne, IL). d Followed by HPLC at 5³C, taken from Erez et al. [3]. Dash indicates no signi¢cant dissociation. e Followed by enzyme kinetics, taken from Erez et al. [3]. Activity of 100% represents 42^50 and 40^46 Wmol min31 mg31 in wt and W330F Tnase, respectively, measured at 25³C, pH 8. See Section 2.2. f Followed by enzyme kinetics, obtained in the present study. b
BBAPRO 36546 7-3-02
338
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
(up to 10%) lifetime [32,33]. In our case we see a decrease in £uorescence lifetime, which could perhaps be accounted for by the higher exposure of PLP in the active site to bulk water. Cooling resulted in a marked decline in rotational correlation time values (Table 1: P1 (column IV), rows 2, 4, 6, 8 vs. 1, 3, 5, 7). The change in £uorescence rotational correlation time does not depend on the lifetime but rather on the freedom of PLP rotation. In all samples analyzed, there is a signi¢cant decline in the rotational correlation time upon cooling, regardless of whether the plugged magic-angle lifetime values during analysis decreased (Table 1, column III, rows 2, 4, 8 vs. 1, 3, 7) or remained unaltered (column III, row 6 vs. 5). The observed time resolved decay of anisotropy pro¢les were analyzed assuming two components of relaxation motion: a major rotational correlation time at the nanosecond time range, and a minor one at the picosecond time range. These rotational correlation time values are too short to re£ect changes associated even with the relatively low molecular mass monomeric protein of the 52 kDa. We assumed, as a rule of thumb, that 3 kDa of molecular mass corresponds to 1 ns in rotational correlation time [33]. Thus, the longest one, 7.6 ns, would be expected from a protein of 7.6U3 kDa = 22.8 kDa. This value is less than half of the molecular mass of the monomer, let alone a dimer or a tetramer with molecular masses of 104 and 208 kDa, respectively. Alternatively, the two components may be attributed to PLP rotational motion around the aldimine bond and to PLP segmental oscillation (wobbling in a cone) within the free volume available in the active site, in the nanosecond and picosecond time domain, respectively. When bound to the active site through the aldimine bond (i.e., in (E-PLP)1 ), PLP freedom of rotation and wobbling are further restricted by hydrogen bonding and salt bridges. In contrast, these restrictions on PLP rotational motion are alleviated in the relatively more `open' active site (E-PLP) following the conformational change suggested to precede the aldimine bond cleavage (E-PLP to EPLP). A contribution from free-dissociated PLP as a potential rotating body is excluded, since under the experimental conditions employed, free PLP £uorescence was undetectable. Moreover, the anisotropy invariably decreased as a consequence of cooling, whether it
was accompanied by a slight decrease in enzyme activity without signi¢cant dissociation (Table 1, column VII, row 2 vs. 1) or by a substantial decrease in enzyme activity that was associated with either a partial or major dissociation (Table 1, column VIII, rows 4, 6, 8 vs. 3, 5, 7). Therefore, one may conclude that PLP is fully sensing the environment at the active site, with its aldimine bond being either completely or at least partially intact, i.e., in either E-PLP or EPLP, respectively. Qualitatively, similar reductions in the rotational correlation times were observed regardless of whether the holo-Tnase is the wt or the W330F mutant, in spite of the di¡erences in the tendencies of these two proteins to undergo dissociation from a tetrameric into a dimeric structure (Table 1, column VIII). Thus, the data observed here presumably re£ect a pre-aldimine dissociation phenomenon, e.g., a conformational change ((E-PLP)1 to E-PLP) which is concomitant with the hydrogen and salt bonding relaxation. CD experiments did not reveal changes in the aromatic or peptidic region of either wt or W330FTnase during cold inactivation [3]. The present study, however, suggests that when one examines the active site directly through its £uorescent PLP cofactor, the rotational dynamics and lifetimes of PLP do sense changes which are cumulatively expressed primarily at the active site, yet are undetectable by CD which senses rather the Tnase backbone. To summarize, both changes in lifetime and rotational correlation time could be ascribed to the change induced by cooling from a `closed' (25³C, (E-PLP)1 ) to an `opened' (2³C, (E-PLP)) conformation of the active site. That this may take place was already proposed before [6]. Thus, the proposed conformational change that occurs prior to the aldimine bond cleavage alluded to in our stopped £ow kinetic study of PLP binding to apo-Tnase, gains further support [6]. In conclusion, it is suggested that cold inactivation is initiated by the weakening of hydrophobic interactions, leading to conformational changes which are the driving force for the aldimine bond cleavage. Acknowledgements We wish to express our gratitude and appreciation
BBAPRO 36546 7-3-02
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
to Professor Yuri M. Torchinsky for introducing us to the study of tryptophanase. This work was partially supported by the James Franck Center for Laser-Matter Interaction and by the O¤ce of Naval Research of the USA (Grant no. NOOO14-96-1-80) to A.H.P.
References
[15]
[16]
[17]
[1] E.E. Snell, Tryptophanase: structure, catalytic activities, and mechanism of action, Adv. Enzymol. 42 (1975) 287^333. [2] Yu.M. Torchinsky, Ya. Kawata, Tryptophanase: structural, spectral and catalytic properties, in: T. Fukui, K. Soda (Eds.), Molecular Aspects of Enzyme Catalysis, Kodansha, Tokyo/VCH, New York, 1994, pp. 165^190. [3] T. Erez, G.Ya. Gdalevsky, R.S. Phillips, Yu.M. Torchinsky, A.H. Parola, Cold inactivation and dissociation into dimers of Escherichia coli Tryptophanase and its W330F mutant form, Biochim. Biophys. Acta 1384 (1998) 365^372. [4] W. Kauzmann, Some factors in the interpretation of protein denaturation, Adv. Protein Chem. 14 (1959) 1^63. [5] P.E. Bock, C. Frieden, Another look at the cold lability of enzymes, Trends Biochem. Sci. 3 (1978) 100^103. [6] T. Erez, R.S. Phillips, A.H. Parola, Pyridoxal phosphate binding to wild type, W330F, and C298S mutants of Escherichia coli apotryptophanase: unraveling the cold inactivation, FEBS Lett. 433 (1998) 279^282. [7] R.S. Phillips, P.D. Gollnick, The environments of Trp-248 and Trp-330 in tryptophan indole-lyase from Escherichia coli, FEBS Lett. 268 (1990) 213^216. [8] R.S. Phillips, K. Ravichandran, R.L. Von Tersch, Synthesis of L-tyrosine from phenol and S-(o-nitrophenyl)-L-cysteine catalyzed by tyrosine phenol-lyase, Enzyme Microbiol. Technol. 11 (1989) 80^83. [9] R.S. Phillips, P.D. Gollnick, Evidence that cysteine 298 is in the active site of tryptophan indole-lyase, J. Biol. Chem. 264 (1989) 10627^10632. [10] T. Honda, M. Tokushige, E¡ects of the temperature and monovalent cations on the activity and quaternary structure of tryptophanase, J. Biochem. 100 (1986) 679^685. [11] C.H. Suelter, J. Wang, E.E. Snell, Direct spectrophotometric assay of tryptophanase, FEBS Lett. 66 (1976) 230^232. [12] E. Pines, D. Pines, T. Barak, B.-Z. Magnes, L.M. Tolbert, J.E. Haubrich, Isotope and temperature e¡ects in ultrafast proton-transfer from a strong excited-state acid, Ber. Bunsengesellsch. Phys. Chem. 102 (1998) 511^517. [13] M. Zamai, V.R. Caiolfa, D. Pines, E. Pines, A.H. Parola, Nature of interaction between basic ¢broblast growth factor and the antiangiogenic drug 7,7-(carbonyl-bis[imino-Nmethyl-4,2-pyrrolecarbonylimino[N-methyl-4,2-pyrrole]-carbonylimino])-bis-(1,3-naphthalene disulfonate), Biophys. J. 75 (1998) 672^682. [14] R. Shukuya, G.W. Schwert, Glutamic acid decarboxylase. 3.
[18]
[19]
[20]
[21] [22]
[23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
339
The inactivation of the enzyme at low temperatures, J. Biol. Chem. 235 (1960) 1658^1661. I.V. Velichiko, S.E. Volk, V.Yu. Dudarenkov, N.N. Magretova, V.Ya. Chernyak, A. Goldman, B.S. Cooperman, R. Lahti, A.A. Baykov, Cold lability of the mutant forms of Escherichia coli inorganic pyrophosphatase, FEBS Lett. 359 (1995) 20^22. J.J. Irias, M.R. Olmsted, M.F. Utter, Pyruvate carboxylase. Reversible inactivation by cold, Biochemistry 8 (1969) 5136^ 5148. R.S. Feldberg, P. Datta, Cold inactivation of L-threonine deaminase from Rhodospirillum rubrum; involvement of hydrophobic interaction, Eur. J. Biochem. 21 (1971) 447^454. R.A. Heller, R.G. Gould, Reversible cold inactivation of microsomal 3-hydroxy-3-methylglutaral coenzyme A reductase from rat liver, J. Biol. Chem. 249 (1974) 5254^5260. Y. Nakanishi, K. Okamoto, F. Isohashi, Subcellular distribution of ATP-stimulated and ADP-inhibited acetyl Co-A hydrase in livers from control and clo¢brate treated rats: comparison of the cytosolic and peroxisomal enzyme, J. Biochem. 115 (1994) 328^332. K. Nakashima, F.B. Rudolph, T. Wakabayashi, H.A. Lardy, Rat liver pyruvate carboxylase, V. Reversible dissociation by chloride salts of monovalent cations, J. Biol. Chem. 250 (1975) 331^336. L. King, G. Weber, Conformational drift of dissociated lactate dehydrogenase, Biochemistry 25 (1986) 3637^3640. R.V. Parry, J.C. Turner, P.A. Rea, High purity preparations of higher plant vacuolar H -ATPase reveal additional subunits. Revised subunit composition, J. Biol. Chem. 264 (1989) 20025^20032. Y. Moriyama, N. Nelson, Cold inactivation of vacuolar proton-ATPases, J. Biol. Chem. 264 (1989) 3577^3582. F. Brigani, W.P. Fong, D.S. Auld, B.L. Valle, In vitro dissociation and reassociation of human alcohol dehydrogenase class 1 isozymes, Biochemistry 28 (1989) 5374^5379. T.M. Huang, G.G. Chang, Characterization of the tetramerdimer-monomer equilibrium of the enzymatically active subunits of the pigeon liver malic enzyme, Biochemistry 31 (1992) 12658^12664. A.N. Fanjul, R.N. Farias, Cold-sensitive cytosolic 3,5,3Ptriiodo-L-thyronine-binding protein and pyruvate kinase from human erythrocytes share similar regulatory properties of hormone binding by glycolytic intermediates, J. Biol. Chem. 268 (1993) 175^179. J.P. Krall, G.E. Edwards, PEP carboxylases from two C4 species of Panicum with markedly di¡erent susceptibilities to cold inactivation, Plant Cell. Physiol. 34 (1993) 1^11. K.E. Hightower, R.E. McCarty, In£uence of nucleotides on the cold stability of chloroplast coupling factor, Biochemistry 35 (1996) 10051^10057. B. Tomkinson, Association and dissociation of the tripeptidyl-peptidase II complex as a way of regulating the enzyme activity, Arch. Biochem. Biophys. 376 (2000) 275^280. M. Isupov, A.A. Antson, E.J. Dodson, G.G. Dodson, I.S. Dementieva, L.N. Zakomirdina, K.S. Wilson, Z. Dauter,
BBAPRO 36546 7-3-02
340
T. Erez et al. / Biochimica et Biophysica Acta 1594 (2002) 335^340
A.A. Lebedev, E.H. Harutyunyan, Crystal structure of tryptophanase, J. Mol. Biol. 276 (1998) 603^623. [31] E.E. Snell, S.J. Di Mari, Schi¡ base intermediates in enzyme catalysis, in: P.D. Boyer (Ed.), The Enzymes, Vol. 2, Academic Press, New York, 1970, pp. 335^370. [32] P. Fuchs, A. Parola, P.W. Robbins, E.R. Blout, Microvis-
cosity of the membrane lipids of normal, transformed and revertant 3T3 cells, Proc. Natl. Acad. Sci. USA 72 (1975) 3351^3354. [33] A.H. Parola, Membranes lipid^protein interactions, in: M. Shinitzky (Ed.), Biomembranes, Physical Aspects, Vol. 2, Balaban Publishers VCH, Brooklyn, NY, 1993, pp. 159^277.
BBAPRO 36546 7-3-02