Thermal stability of peroxidase from the african oil palm tree Elaeis guineensis Anabel Rodrı ´ guez 1, *, David G. Pina 1, *, Bele ´ nYe ´ lamos 2 , John J. Castillo Leo ´ n 3 , Galina G. Zhadan 1 , Enrique Villar 1 , Francisco Gavilanes 2 , Manuel G. Roig 4 , Ivan Yu. Sakharov 5 and Valery L. Shnyrov 1 1 Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Facultad de Biologı ´ a, Universidad de Salamanca, Salamanca, Spain; 2 Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Facultad de Quı ´ mica, Universidad Complutense, Madrid, Spain; 3 Escuela de Quı ´ mica, Universidad Industrial de Santander, Bucaramanga, Colombia; 4 Departamento de Quı ´ mica Fı ´ sica, Facultad de Quı ´ mica, Universidad de Salamanca, Salamanca, Spain; 5 Department of Chemical Enzymology, Faculty of Chemistry, Moscow State University, Moscow, Russia The thermal stability of peroxidase from leaves of the African oil palm tree Elaeis guineensis (AOPTP) at pH 3.0 was studied by differential scanning calorimetry (DSC), intrinsic fluorescence, CD and enzymatic assays. The spectral parameters as monitored by ellipticity changes in the far-UV CD spectrum of the enzyme as well as the increase in tryp- tophan intensity emission upon heating, together with changes in enzymatic activity with temperature were seen to be good complements to the highly sensitive but integral method of DSC. The data obtained in this investigation show that thermal denaturation of palm peroxidase is an irrevers- ible process, under kinetic control, that can be satisfactorily described by the two-state kinetic scheme, N À! k D, where k is a first-order kinetic constant that changes with tem- perature, as given by the Arrhenius equation; N is the native state, and D is the denatured state. On the basis of this model, the parameters of the Arrhenius equation were calculated. Keywords: peroxidase; differential scanning calorimetry; intrinsic fluorescence; circular dichroism; protein stability. Peroxidases (EC 1.11.1.7; donor:hydrogen-peroxide oxido- reductase) are enzymes that are widely distributed in the living world and that are involved in many physiological processes, including abiotic and biotic stress responses. Although the function of peroxidases is often seen primarily in terms of effecting the conversion of H 2 O 2 to H 2 O, this should not be allowed to obscure their wider participation in other reactions, such as cell wall formation, lignification, the protection of tissues from pathogenic microorganisms, etc. [1,2]. Several peroxidases have been isolated, sequenced and characterized. They have essentially been classified in three classes, supported in the first instance by comparison of aminoacid sequence data and confirmed by more recent crystal structure data (class I, intracellular prokaryotic peroxidases; class II, extracellular fungal peroxidases, and class III, secretory plant peroxidases [2]). Peroxidase has attracted industrial attention because of its usefulness as a catalyst in clinical biochemistry and enzyme immunoassays. Some modern applications of peroxidases include treatment of waste water containing phenolic compounds, the synthe- sis of several different aromatic chemicals and polymeric materials. The peroxidase most studied is the one obtained from horseradish roots (HRP), which is also the most commercially available one. However, other plant species may provide peroxidases with similar or even improved properties. Therefore, the availability of highly stable and active peroxidases from sources other than horseradish roots would go a long way toward the development of a catalytic enzyme with broad commercial and environmental possibilities [3]. Several publications have addressed the study of the conformational stability of peroxidases, but to date our understanding of their folding mechanism remains contradictory and unclear [4–11]. Factors affecting con- formational stability have been studied most intensively in proteins under reversible conditions [12,13]. However, after denaturation many proteins cannot refold in vitro due to modifications such as digestion, aggregation, loss of a prosthetic group, etc. [14,15]. Thus, the thermal denatura- tion of such proteins is often discussed in terms of the Lumry–Eyring model [16], in which a reversible unfolding step is followed by an irreversible denaturation step: N Ð U ! D, where N, U and D are the native, unfolded or partially unfolded, and denatured states of the protein, respectively [17]. However, use of the whole Lumry–Eyring kinetic model for the quantitative description of DSC traces is difficult because the corresponding system of differential equations does not have an analytical solution at varying temperatures. Although there are computer programs that allow the direct fitting of a system of differential equations to experimental data, there are as yet no publications in which DSC data have been interpreted through the use of the whole Lumry–Eyring kinetic model [18]. Therefore, to analyse the irreversible thermal denaturation of proteins, Correspondence to V. L. Shnyrov, Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Universidad de Salamanca, Plaza de los Doctores de la Reina, s/n, 37007 Salamanca, Spain. Fax: + 34 923 294579, Tel.: + 34 923 294465, E-mail: shnyrov@usal.es Abbreviations:ABTS,2,2¢-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid); DSC, differential scanning calorimetry; HRP, peroxidase from horseradish roots; AOPTP, peroxidase from the African Oil Palm Tree Elaeis guineensis. Enzyme: peroxidase (EC 1.11.1.7; donor:hydrogen-peroxide oxidoreductase). *Note: these authors contributed equally to this work. (Received 8 February 2002, accepted 12 April 2002) Eur. J. Biochem. 269, 2584–2590 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02930.x researchers generally look for simple models that are approximations to the Lumry–Eyring model [17,19–21]. Recently a novel peroxidase has been isolated from the leaves of the African oil palm tree Elaeis guineensis [5]. This peroxidase shows a characteristic spectrum for haem- containing proteins, with a Soret maximum at 403 nm. Its molecular mass as estimated by SDS/PAGE is 57 000, which is higher than the values published for other plant peroxidases [1], probably because of the higher degree of AOPTP glycosylation. It has also been found that AOPTP, similar to peroxidases earlier detected in the sweet potato, royal palm tree, tobacco, and tomato [22–24], is an anionic protein with a pI value of 3.8. Preliminary data [25] have suggested that AOPTP is stable over a broad pH-range, maximum stability being found at pH 7.0. Under acidic (pH 2.0) and alkaline (pH 12.0) conditions, AOPTP shows a lower stability but remains a highly stable enzyme, loosing not more than 20% of its initial activity for 30 min at 25 °C. In recent years there has been tremendous interest in the production of conducting polymers. Polyaniline is one such compound because it can be used in lightweight organic batteries, in microelectronics, in optical display, in anticor- rosive protection, in bioanalysis as a sensing element, etc. [26,27]. This is because it shows good electrical and optical properties as well as high environmental stability. It is well known that peroxidases can be used in the synthesis of polyaniline in the presence of hydrogen peroxide as a reduc- ting substrate and sulfonated polystyrene and poly(vinyl- phosphonic acid) as polymeric templates [28], which take place effectively at pH values below 4.0. Consequently, for the development of such biotechnological process, would be of interest to find and characterize peroxidases that are stable under acidic conditions, such as the enzyme considered here (peroxidase from African oil palm tree Elaeis guineensis). Here we describe a detailed investigation of the thermal denaturation of AOPTP at pH 3.0. This was studied by differential scanning calorimetry in the combination with structural probes, such as intrinsic fluorescence and circular dichroism, as well as enzymatic activity assays. The thermal unfolding of AOPTP was found to be irreversible and strongly scan-rate dependent, which led us to analyse this nonequilibrium process based on the simplest so-called two- state kinetic model: N À! k D ð1Þ which is a limiting case of the Lumry–Eyring model [17]. This model considers only two significantly populated macroscopic states, the initial or native state (N) and the final or denatured (D) state, transition between which is determined by a strongly temperature-dependent first-order rate constant (k). The data obtained demonstrate that AOPTP is a significantly more thermostable enzyme than other known peroxidases, that makes AOPTP an intriguing catalyst for scientific and commercial applications where stability at high temperatures is desirable. MATERIALS AND METHODS Materials 2,2¢-Azino-bis(3-ethylbenzthiazoline-6-sulfonicacid)(ABTS) was purchased from Amersham International plc (Buckinghamshire, UK). H 2 O 2 was obtained from Merck (Darmstadt, Germany) and quantified by UV spectropho- tometry at 230 nm (e ¼ 81 M )1 Æcm )1 ) [29]. Phenyl- Sepharose and Sephacryl S 200 were from Pharmacia Biotech (Uppsala, Sweden), DEAE cellulose was from Serva (Heidelberg, Germany), and other reagents were from Panreac (Barcelona, Spain). All reagents were of the highest purity available. Double-distilled water was used through- out. All measurements were carried out in 10 m M Na-phos- phate buffer, pH 3.0. Protein purification and determination AOPTP was purified from African oil palm tree leaves as described elsewhere [5]. Briefly, leaves were triturated and incubated with constant stirring in 10 m M phosphate buffer, pH 7.0, for 1 h at ambient temperature, and the homogen- ate obtained was filtered and centrifuged (7000 g,15min). For the extraction of coloured compounds, a two-phase system containing 14% (w/v) poly(ethylene glycol) and 20% (w/v) (NH 4 ) 2 SO 4 was used. Then, the aqueous phase containing peroxidase activity was applied to a phenyl- Sepharose column (1.5 · 30 cm) equilibrated with 100 m M phosphate buffer, pH 6.5, containing 1.7 M (NH 4 ) 2 SO 4 . The enzyme was eluted by decreasing the (NH 4 ) 2 SO 4 concen- tration, collected and concentrated using a YM-10 mem- brane (Amicon, cut-off 10 000) and applied to a Sephacryl S 200 column (2.5 · 41 cm) equilibrated with 5 m M Tris/HCl, pH 8.3. Elution was carried out in the same buffer. Fractions with enzymatic activity were collected and applied directly to a DEAE–cellulose column (0.9 · 9 cm) equili- brated with 5 m M Tris, pH 8.3. The peroxidase was eluted with a linear, 0–50 m M NaCl, gradient, dialyzed against distilled water, freeze-dried and stored at 4 °C. The purity of AOPTP were determined by SDS/PAGE. Electrophoresis was performed as described by Fairbranks et al. [30] on a Bio-Rad minigel apparatus, using a flat block with a polyacrylamide gradient of 5–25%. Gels were prefixed and stained using the method of Merril et al. [31]. Protein contents were determined by the Bradford assay [32]. The RZ (A 403 /A 280 ) for the AOPTP samples used in this work were 2.8–3.0. Differential scanning calorimetry DSC experiments were performed on a MicroCal MC-2D differential scanning microcalorimeter (MicroCal Inc., Northampton, MA) with cell volumes of 1.22 mL, inter- faced with a personal computer (IBM-compatible) as described previously [8]. Exhaustive cleaning of the cells was undertaken before each experiment. All protein solu- tions were dialyzed against the desired buffer, and the dialyzate was used as reference. All solutions were degassed by stirring under a vacuum prior to scanning. Different scan rates within the 0.5–1.5 KÆmin )1 rangewereemployedand an overpressure of 2 atm of dry nitrogen was always kept over the liquids in the cells throughout the scans. A background scan collected with a buffer in both cells was subtracted from each scan. The reversibility of the thermal transitions was checked by examining the reproducibility of the calorimetric trace in a second heating of the sample immediately after cooling from the first scan. The experi- mental calorimetric traces were corrected for the effect of Ó FEBS 2002 Stability of plant peroxidase (Eur. J. Biochem. 269) 2585 the instrument response time using the procedure described previously [33]. The molar excess heat capacity curves obtained by normalizing with the protein concentrations and the known volume of the calorimeter cell were smoothed and plotted using the Windows-based software package ( ORIGIN ) supplied by MicroCal. Data were ana- lyzed by the nonlinear least-squares fitting program, as reported elsewhere [19]. The correlation coefficient, r,used as a criterion for the accuracy of fitting, was calculated by the equation: r ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 À X n i ¼1 ðy i À y calc i Þ 2  X n i ¼1 ðy i À y m i Þ 2 s ð2Þ where y i and y calc i are, respectively, the experimental and calculated values of C ex p ; y m i is the mean of the experimental values of C ex p ,andn is the number of points. Typical protein concentrations for calorimetric experiments ranged between 1.0 and 2.5 mgÆmL )1 . Molar transition enthalpies, DH, refer to M ¼ 57 000 gÆmol )1 . Intrinsic fluorescence Fluorescence measurements were performed on a Hitachi F-4010 spectrofluorimeter. Exitation was carried out at 296 nm (with 5 nm excitation and emission slitwidths) in order to avoid the contribution of tyrosine to the intrinsic fluorescence spectrum of AOPTP. The temperature dependence of the emission fluorescence spectra was investigated using thermostatically controlled water circu- lating in a hollow brass cell-holder. The temperature of the sample cell was monitored with a thermocouple immersed in the cell under observation. Circular dichroism CD spectra in the far-ultraviolet range (190–250 nm) were recorded on a Jasco-715 spectropolarimeter, using a spectral band-pass of 2 nm and a cell path length of 1 mm with a protein concentration of 0.2 mgÆmL )1 . Spectra are averages of four scans at a scan rate of 50 nmÆmin )1 .Allspectra were background-corrected, smoothed, and converted to a mean residue ellipticity of [H] ¼ 10 M res ÆH obs Æl )1 Æp )1 ,where M res ¼ 115.5 is the mean residue molar mass, H obs is the ellipticity measured (degrees) at wavelength k, l is the optical path-length of the cell (dm), and p is the protein concen- tration (mgÆmL )1 ). Spectra were analyzed using the SELCON 3 software package [34]. To study the dependence of ellipticity on temperature, the samples were heated at a constant heating rate (% 1KÆmin )1 )usingaNeslabRT-11 programmable water bath. Activity assays AOPTP activity was assayed using ABTS as substrate [35]. Aliquots of enzyme solution were added to a spectral cuvette with 1-cm optical path length containing 0.4 m M ABTS and 5 m M H 2 O 2 in 50 m M acetate buffer, pH 5.0 in a final volume 2 mL. The rate of changes in absorbance at 405 nm due to ABTS radical formation was measured spectrophotometrically at 25 °C. Activities were calculated using a molar absorption coefficient of the ABTS oxidation product at 405 nm of 36.8 m M )1 Æcm )1 [36]. Kinetics of AOPTP thermal inactivation To study the kinetics of heat denaturation by intrinsic fluorescence, 0.02 mL samples of a 0.1-m M AOPTP solu- tion were added to 1.6 mL of buffer previously thermostat- ed at the desired temperature in the fluorimeter cuvette. The mixture was stirred constantly in the cuvette and the emmision intensity at a wavelength of 340 nm was recorded at a certain time interval. In all experiments, the time for temperture equilibrium to be reached in the cuvette after sample introduction did not exceed 5 s. An almost identical procedure was applied to study the kinetics of changes in peroxidase activity with temperature. Samples of AOPTP were incubated at the desired temperature under constant stirring. At certain times, aliquots were removed and immediately transferred to test tubes placed in a water–ice mixture to stop the inactivation process. Subsequently, enzyme activity was measured as described above. The measurements were made in triplicate and the data are presented as average values. RESULTS AND DISCUSSION Differential scanning calorimetry Figure 1 shows the calorimetric transitions of the thermal denaturation of AOPTP at pH 3.0, at three different scan rates. The heat absorption curve apparent T m (temperature at the maximum of the heat capacity profile) was found to be dependent on the scan rate and denaturation was always calorimetrically irreversible, as no thermal effect was observed in a second heating of the enzyme solution. Inspection of the DSC curves shown in Fig. 1 further reveals asymmetry in the shape of the peaks, which might arise from two overlapping transitions. This would be a reasonable possibility for AOPTP, which is a fairly large 50 60 70 80 90 0 10 20 30 40 C p ex (kcal K -1 mol -1 ) Temperature ( o C) Fig. 1. Temperature dependence of the excess molar heat capacity of AOPTP at scan rates of 0.5 (circles), 1.0 (squares) and 1.5 (triangles) KÆmin )1 at pH 3.0. Solid lines represent the best individual fit to each experimental curve using Eqn (3). Protein concentrations were % 2.5 mgÆmL )1 at a scan rate of 0.5 KÆmin )1 , % 2mgÆmL )1 at a scan rates of 1.0, and % 1.0 mgÆmL )1 at a scan rate of 1.5 KÆmin )1 . 2586 A. Rodrı ´ guez et al. (Eur. J. Biochem. 269) Ó FEBS 2002 protein and may, in principle, comprise several domains [37]. We analyzed this possibility by applying the successive annealing procedure [38]. Thus, AOPTP was first heated at ascanrateof60KÆh )1 in the microcalorimeter cell to a temperature of 69 °C, which would be close to the maximum for a putative first transition. The sample was cooledandthenheatedto90°C at the same scan rate. The reheating scan revealed that the only effect of the first scan was to decrease the peak intensity by a scale factor determined by the difference in the amounts of protein undergoing denaturation, and that there was no change in T m or any effect on the shape of the curve (not shown). These experiments rule out the possibility of overlapping independent transitions. The effect of the scan rate on the calorimetric profiles clearly indicated that they correspon- ded to irreversible, kinetically controlled transitions. For this reason the analysis of DSC transitions on the basis of equilibrium thermodynamics was ruled out [39] and was accomplished using the simple two-state irreversible model (Eqn 1), in which only the native (N) and final (irreversibly denatured) (D) states are significantly populated and in which the conversion from N to D is determined by a strongly temperature-dependent, first order rate constant (k) that changes with temperature, as given by the Arrhenius equation. In this case, the excess heat capacity C ex p is given by the following equation [19]: C ex p ¼ 1 v DH exp E A R 1 T à À 1 T   exp À 1 v Z T T 0 exp E A R 1 T à À 1 T  dT 8 < : 9 = ; ð3Þ where v ¼ dT/dt (KÆmin )1 ) is a scan rate value; DH is the enthalpy difference between the denatured and native states; E A is the activation energy of the denaturation process; R is a gas constant, and T* is temperature, where k is equal to 1min )1 . The excess heat capacity functions obtained for AOPTP were analysed by fitting the data to the two-state irreversible model (Eqn 3), either individually or by fitting this theor- etical expression simoultaneously to all the experimental curves, using the scan rate as an additional variable. The highest likelihood values for E A and T* obtained with the nonlinear least squares minimization procedure are shown in Table 1. It may be seen that the calculated and experimental curves are in good agreement. Also, the parameters obtained from individual fits were in reasonable agreement with those obtained from the global fit, indica- ting that the two-state irreversible model offers a good explanation of the AOPTP denaturation process. Addition- ally it should be noted that no dependence of the shape of the DSC contour on the AOPTP concentration was found at a scan rate of 60 KÆh )1 in the 0.7–3.8 mgÆmL )1 range. No pronounced dependence of the denaturation enthalpy on scan rate was observed (see Table 1). These data argue against an effect of intermolecular aggregation on the DSC traces obtained. Fluorescence and enzymatic activity Conformational changes in the surroundings of AOPTP aromatic side chains were detected by intrinsic fluorescence spectroscopy. The emission spectra from 300 to 400 nm of intact and thermally denatured AOPTP are represented in Fig. 2. Intact AOPTP displayed a low emission intensity due to energy transfer to haem, which, as can be seen in Fig. 3, significantly increased in the denatured enzyme owing to a change in the relative orientation or distance between the haem and tryptophan residue(s) [40]. Therefore, the intrinsic fluorescence of AOPTP was monitored at 340 nm for thermal denaturation. Figure 3A shows the kinetic data on AOPTP denaturation as observed by changes in the fluorescence intensity obtained at five different temperatures. This figure shows that although the denaturation rate does increase with temperature, the Table 1. Arrhenius equation parameter estimates for the two-state irreversible model of the thermal denaturation of AOPTP at pH 3.0. Parameter Temperature scan rate (KÆmin )1 ) 0.5 1.0 1.5 Global fitting DH, kcalÆmol )1 251 ± 9 257 ± 7 256 ± 7 T*, K 347.6 ± 0.2 347.6 ± 0.2 347.3 ± 0.3 347.5 ± 0.3 E A , kcalÆmol )1 99.7 ± 1.2 98.8 ± 1.4 101.1 ± 0.9 102.1 ± 1.4 r 0.9990 0.9987 0.9989 0.9959 300 320 340 360 380 400 0 5 10 15 20 25 30 Wavelength (nm) Fluorescence intensity (relative units) Fig. 2. Fluorescence spectra of intact at 25 °C (solid line) and thermally denatured at 80 °C (dashed line) 1 l M AOPTP at pH 3.0. Excitation wavelength, 296 nm. Ó FEBS 2002 Stability of plant peroxidase (Eur. J. Biochem. 269) 2587 final level of intrinsic fluorescence is independent of the denaturation temperature. This supports the idea that the thermal denaturation of AOPTP is not a reversible equilib- rium process between the native and denatured enzyme because if this was the case the relative amounts of native and denatured states would be expected to show a definite temperature dependence. Therefore, this appears to be a kinetic phenomenon involving an irreversible process. The same experimental approach was applied to the enzymatic activity assays, as the denaturation of any enzyme is expected to abolish its biological activity, allowing us to monitor thermally induced conformational changes in the catalytic surroundings by measuring the loss of enzy- matic activity vs. time at different temperatures (Fig. 3B). The best fit of the experimental data, represented as continuous lines in Fig. 3, was achieved with an exponential function: F ¼ F 1 þðF 0 À F 1 Þ exp ðÀktÞð4Þ where F is the function value at a given time (t)andF 0 and F 1 are normalization parameters (at t ¼ 0, F ¼ F 0 ,andat t ¼1, F ¼ F 1 ), indicating a first-order kinetic process. The temperature dependence of the rate constants obtained from the data shown in Fig. 3 was expressed by the Arrhenius equation: k ¼ exp E A R 1 T à À 1 T  ð5Þ and is represented in Fig. 4. Thus, the activation energy and T* can be calculated from the linear fit of both the fluorescence and enzymatic assay data. The value thus obtained (E A ¼ 110.8 ± 3.2 kcalÆmol )1 )and(T* ¼ 345.9 ± 1.8 K), were in satisfactory agreement with the values obtained from the DSC experiments (Table 1). Circular dichroism CD is one of the most sensitive physical technique for determining structures and monitoring the structural 15 20 25 30 Fluorescence intensity at 340 nm a 020406080 0.0 0.2 0.4 0.6 0.8 1.0 Relative activity Time (min) b 020406080100 0.01 100 0.1 1 Log (activity) Time (min) Fig. 3. Temperature dependence of the thermal denaturation kinetics of AOPTP at pH 3.0 as monitored by intrinsic fluorescence (a) and per- oxidase activity shown at normal (b) and semilog scale (b, insert). Symbols refer to the experimental data at different temperatures: 73.6 °C(s), 70.9 °C(d), 69.2 °C(n), 68.7 °C(m), and 65.9 °C(,)in (a); 71.0 °C(s), 68.0 °C(d), 66.5 °C(n), and 65.2 °C(m)in(b). 2.90 2.92 2.94 2.96 -3 -2 -1 0 ln k 10 3 / T in K Fig. 4. Dependence of the logarithm of the inactivation rate constant (min )1 ) on the reciprocal value of the absolute temperature as monitored by intrinsic fluorescence (solid symbols) and enzymatic activity assays (open symbols) for AOPTP at pH 3.0. Thelinewasfittedbylinear regression. 200 220 240 -10000 -5000 0 5000 10000 15000 [Θ] (deg cm 2 dmol -1 ) Wavelength (nm) 40 50 60 70 80 90 -7000 -6000 -5000 -4000 Temperature ( o C) [Θ] (deg cm 2 dmol -1 ) Fig. 5. CD spectra in the far-ultraviolet spectral region of intact (solid line) and irreversible thermally denatured (dashed line) 2 l M AOPTP at pH 3.0 and 25 °C. (Inset) Temperature dependences of ellipticity at 222 nm for AOPTP at pH 3.0 obtained upon heating with a constant scan rate of % 1KÆmin )1 . Solid line is best fit obtained using Eqn (7). 2588 A. Rodrı ´ guez et al. (Eur. J. Biochem. 269) Ó FEBS 2002 changes occurring in biomacromolecules [41], affording a direct interpretation of the changes in protein secondary structure. Figure 5 shows the far-UV CD spectra of intact (solid line) and thermally denatured (dashed line) AOPTP at pH 3.0. The fractions of a helix, a strand, turns, and unordered secondary structures obtained following the SELCON3 self-consistent method [34] are given in Table 2. It is clear that AOPTP is significantly different from other haem peroxidases from plants for which, despite the low level of sequence homology (often less than 20%), the overall folding and the organization of the secondary structure is conserved [42]. The structure of haem peroxidases from plants is formed by 10–11 a helices (c. 40%), linked by loops and turns, while a structures are essentially absent or are only a minor component [43]. By contrast, intact AOPTP contains a considerable amount of a-structure (% 38%) and only 15% of a helices, at pH 3.0. This probably makes this enzyme more stable in comparison with horseradish peroxidase which under the same experimental conditions has 42% of a helices and only 11% of a structure [8]. Upon heating AOPTP to the denaturation temperature, the shape of the CD spectrum changes, showing an increase in unordered structure from % 30%, for the intact enzyme, up to % 50% for the denatured one (see Table 2). The process of thermal denaturation of AOPTP was monitored directly by following the changes in molar ellipticity at 222 nm as at this wavelength the changes in ellipticity are significant upon heating. On increasing temperature (Fig. 5, insert), irreversible cooperative transitions to the denatured state occurred, which were analyzed using a nonlinear least squares fitting (see lines through the data points in Fig. 5, insert). In this case, the fraction of denatured AOPTP, F U was calculated from the spectral parameter used to follow denaturation (y) prior to the minimization procedure, according to the expression: F U ¼ðy À y N Þ=ðy U À y N Þð6Þ where y N ¼ a 1 + a 2 T and y U ¼ b 1 + b 2 T represents the mean values of the y characteristic of the native and denatured conformations, respectively, obtained by linear regressions of pre- and post-transitional baselines; T is the temperature. In this case, the parameter used to follow denaturation, y, can be expressed as a function of the kinetic parameters by equation [19]: y ¼ y U À½y U À y N  exp 1 v ð T T 0 exp E A R 1 T à À 1 T  dT 8 < : 9 = ; ð7Þ Fitting of the experimental data to this equation afforded the T* parameter and the activation energy for AOPTP. These results were 347.2 ± 1.6 K and 106.0 ± 1.4 kcalÆ mol )1 , respectively, which are similar to the values for the same parameters obtained by the other methods used in this work. Thus, all these independent experimental approaches support the conclusion that AOPTP thermal denaturation can be interpreted in terms of the irreversible two-state kinetic model, and that only two states, native and denatured, are populated in its denaturation process. Finally, it is interesting to compare the thermal stability of AOPTP with that of other peroxidases. In our previous publication [8] we reported the results of a detailed investigation of the thermal denaturation of horseradish peroxidase isoenzyme c under the same experimental conditions as those used here. It is clear that AOPTP is substantially more thermostable than HRPc. Thus, the T m for AOPTP at a scan rate of 60 KÆh )1 is 72.3 ± 0.2 °C while for HRPc this value is only 60.2 ± 0.2 °C.The Arrhenius denaturation energy of AOPTP obtained by different methods, 103 ± 6 kcalÆmol )1 , is a high value in comparison not only with value for HRPc (38±1kcalÆ mol )1 ) but also in comparison with those found for other plant peroxidases [4]. Coupled with its high catalytic potential [44], the unique high thermostability of AOPTP promises good perspectives for this peroxidase in biotech- nological applications. ACKNOWLEDGEMENTS This work was supported by NATO Linkage Grant LST.CLG 975189 (to M. G. R., I. Y. S. and V. L. S.). D.G.P. is a fellowship holder from Fundac¸ a ˜ oparaaCieˆ ncia e a Tecnologia, Portugal (Ref. SFRH/BD/ 1067/2000). We thank N. S. D. Skinner for proof-reading the manu- script. REFERENCES 1. Dunford, H.B. (1991) Horseradish peroxidase: structure and kinetic properties. In Peroxidases in Chemistry and Biology,Vol.II (Everse, J., Everse, K.E. & Grisham, M.B., eds), pp. 1–24. CRC Press, Boca Raton, FL, USA. 2. Welinder, K.G. (1992) Superfamily of plant, fungal and bacterial peroxidases. Curr. Opin. Struct. Biol. 2, 388–393. 3. Krell, H W. (1991) Peroxidase. An important enzyme for diag- nostic test kits. In Biological, Molecular and Physiological Aspects of Plant Peroxidases (Lobarsewski,J.,Greppin,H.,Penel,C.& Gaspar, T., eds), pp. 469–478. University M. Curie-Sklodowska and University Geneva, Lublin and Geneva. 4. McEldoon, J.P. & Dordick, J.S. (1996) Unusual thermal stability of soybean peroxidase. Biotechnol. Progr. 12, 555–558. 5. Sakharov, I.Yu., Castillo, L.J., Areza, J.C. & Galaev, I.Y. (2000) Purification and stability of peroxidase of African oil palm Elaies guineensis. Bioseparation 9, 125–132. 6. Tams, J.W. & Welinder, K.G. (1996) Unfolding and refolding of Coprinus cinereus peroxidase at high pH, in urea, and at high temperature. Effect of organic and ionic additives on these pro- cesses. Biochemistry 35, 7573–7579. Table 2. Secondary structure (%) determined by CD spectroscopy for intact and thermally denatured AOPTP at pH 3.0. Protein state a Helices b Strands b Turns Unordered Regular Distorted Total Regular Distorted Total Intact 5.6 9.3 14.9 27.2 10.6 37.8 20.2 29.7 Denatured 4.8 6.9 11.7 12.5 10.6 23.1 14.3 49.6 Ó FEBS 2002 Stability of plant peroxidase (Eur. J. Biochem. 269) 2589 7. Chattopadhyay, K. & Mazumdar, S. (2000) Structural and con- formational stability of horseradish peroxidase: effect of tem- perature and pH. Biochemistry 39, 263–270. 8. Pina, D.G., Shnyrova, A.V., Gavilanes, F., Rodrı ´ guez, A., Leal, F., Roig, M.G., Sakharov, I.Yu., Zhadan, G.G., Villar, E. & Shnyrov, V.L. (2001) Thermally induced conformational changes in horseradish peroxidase. Eur. J. Biochem. 268, 120–126. 9. Pappa, H.S. & Cass, A.E.G. (1993) A step towards understanding the folding mechanism of horseradish peroxidase. Tryptophan fluorescence and circular dichroism equilibrium studies. Eur. J. Biochem. 212, 227–235. 10. Das, T.K. & Mazumdar, S. (1995) pH-induced conformational perturbation in horseradish peroxidase. Picosecond tryptophan fluorescence studies on native and cyanide-modified enzymes. Eur. J. Biochem. 227, 823–828. 11. Tsaprailis, G., Wing Sze Chan, D. & English, A.M. (1998) Con- formational states in denaturants of cytochrome c and horseradish peroxidases examined by fluorescence and circular dichroism. Biochemistry 37, 2004–2016. 12. Privalov, P.L. (1989) Thermodynamic problems of protein struc- ture. Annu. Rev. Biophys. Biophys. Chem. 18, 47–69. 13. Freire, E. (1995) Thermal denaturation methods in the study of protein folding. Methods Enzymol. 259, 144–168. 14. Schmid,F.X.,Mayr,L.M.,Mucke,M.&Scho ¨ nbrunner, E.R. (1993) Prolyl isomerases: role in protein folding. Adv. Protein Chem. 44, 25–66. 15. Klibanov, A.M. & Akhern, T.J. (1987) Thermal stability of pro- teins. In Protein Engineering (Oxender, D.L. & Fox, C.F., eds), pp. 213–218. Alan R. Liss, New York. 16. Lumry, R. & Eyring, E. (1954) Conformation changes of proteins. J. Phys. Chem. 58, 110–120. 17. Sanchez-Ruiz, J.M. (1992) Theoretical analysis of Lumry–Eyring models in differential scanning calorimetry. Biophys. J. 61, 921–935. 18. Lyubarev, A.E. & Kurganov, B.I. (2000) Analysis of DSC data relating to proteins undergoing irreversible thermal denaturation. J. Therm. Anal. Cal. 62, 51–62. 19. Kurganov, B.I., Lyubarev, A.E., Sanchez-Ruiz, J.M. & Shnyrov, V.L. (1997) Analysis of differential scanning calorimetry data for proteins. Criteria of validity of one-step mechanism of irreversible protein denaturation. Biophys. Chem. 69, 125–135. 20. Lyubarev, A.E. & Kurganov, B.I. (1998) Modeling of irreversible thermal protein denaturation at varying temperature. I. The model involving two consecutive irreversible steps. Biochemistry (Moscow) 63, 434–440. 21. Lyubarev, A.E. & Kurganov, B.I. (1999) Modeling of irreversible thermal protein denaturation at varying temperature. II. The complete kinetic model of Lumry and Eyring. Biochemistry (Moscow) 64, 832–838. 22. Marangoni,A.G.,Brown,E.D.,Stanley,D.W.&Yada,R.Y. (1989) Tomato peroxidase: rapid isolation and partial character- ization. J. Food. Sci. 54, 1269–1271. 23. Gazaryan, I.G. & Lagrimini, L.M. (1996) Purification and unusual kinetic properties of a tobacco anionic peroxidase. Phytochemistry 41, 1029–1034. 24. Lindgren,A.,Ruzgas,T.,Gorton,L.,Cso ¨ regi, E., Bautista Ardila, G., Sakharov, I.Yu. & Gazaryan, I.G. (2000) Biosensors based on novel peroxidases with improved properties in direct and mediated electron transfer. Biosensors Bioelectronics 15, 491–497. 25. Sakharov, I.Yu. (2001) Unusual stability of the heme-peroxidase from palm tree leaves Elaeis guineensis. J. Inorg. Biochem. 86, 415. 26. MacDiarmid, A.G. (1997) Polyaniline and polypyrrole: Where are we headed? Synthetic Metals 84, 27–34. 27. Shoji, E. & Freund, M.S. (2001) Potentiometric sensors based on the inductive effect on the pK(a) of poly (aniline): a nonenzymatic glucose sensor. J. Am. Chem. Soc. 123, 3383–3384. 28. Liu, W., Cholli, A.L., Nagarajan, R., Kumar, J., Tripathy, S., Bruno, F.F. & Samuelson, L. (1999) The role of template in the enzymatic synthesis of conducting polyaniline. J. Am. Chem. Soc. 121, 11345–11355. 29. Pick, E. & Keisari, Y. (1980) A simple colorimetric method for the measurement of hydrogen peroxide produced by cells in culture. J. Immunol. Methods 38, 161–170. 30. Fairbanks, G., Steck, T. & Wallach, D.F.N. (1971) Electro- phoretic analysis of the major polypeptides of the human ery- throcyte membrane. Biochemistry 10, 2606–2617. 31. Merril, C.R., Goldman, D., Sedman, S.A. & Ebert, M.H. (1981) Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211, 1437–1438. 32. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 33. Lopez Mayorga, O. & Freyre, E. (1987) Dynamic analysis of differential scanning calorimetry data. Byophys. Chem. 87, 87–96. 34. Sreerama, N., Venyaminov, S.Yu. & Woody, R.W. (1999) Esti- mation of the number of alpha-helical and beta-strand segments in proteins using circular dichroism spectroscopy. Prot. Sci. 8, 370–380. 35. Childs, R.E. & Bardsley, W.G. (1975) The steady-state kinetics of peroxidase with 2,2¢-azino-di-(3-ethyl-benzthiazoline-6-sulfonic acid) as chromogen. Biochem. J. 145, 93–103. 36. Smith, A.T., Santama, N., Dacey, S., Edwards, M., Bray, R.C., Thorneley, R.N.F. & Burke, J.F. (1990) Expression of a synthetic gene for horseradish peroxidase C in Escherichia coli and folding and activation of the recombinant enzyme with Ca 2+ and heme. J. Biol. Chem. 265, 13335–13343. 37. Garel, J.R. (1992) Folding of large proteins: multidomain and multysubunite proteins. In Protein Folding (Creighton,T.E.,ed.), pp. 405–454. W.H. Freeman, New York. 38. Shnyrov, V.L. & Zhadan, G.G. (2000) Irreversible thermal denaturation of complex biological structures. In Recent Res. Devel. Physical. Chem. (Pandalai, S.G., ed.), pp. 351–367. Trans- world Research Network, Trivandrum, India. 39. Freire, E., van Osdol, W.W., Mayorga, O.L. & Sanchez-Ruiz, J.M. (1990) Calorimetrically determined dynamics of complex unfolding transitions in proteins. Annu. Rev. Biophys. Biophys. Chem. 19, 159–188. 40. Hill, B.C., Horowitz, P.M. & Robinson, N.C. (1986) Detection, characterization, and quenching of the intrinsic fluorescence of bovine heart cytochrome c oxidase. Biochemistry 25, 2287–2292. 41. Venyaminov, S.Yu. & Yang, J.T. (1996) Determination of protein secondary structure. In Circular Dichroism and the Conformational Analysis of Biomacromolecules (Fasman, G.D., ed.), pp. 69–107. Plenum Press, New York. 42. Welinder, K.G. & Gajhede, M. (1993) Structure and evolution of peroxidases. In Plant Peroxidases Biochemistry and Physiology (Greppin, H., Rasmussen, S.K., Welinder, K.G. & Penel, C., eds), pp. 35–42. University of Copenhagen and University of Geneva, Geneva, Switzerland. 43. Banci, L. (1997) Structural properties of peroxidases. J. Biotech- nol. 53, 253–263. 44. Sakharov, I.Yu. (2001) Long-term chemiluminescent signal is produced in the course of luminol peroxidation catalyzed by peroxidase isolated from leaves of african oil palm tree. Bio- chemistry (Moscow) 66, 515–519. 2590 A. Rodrı ´ guez et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . Spain; 5 Department of Chemical Enzymology, Faculty of Chemistry, Moscow State University, Moscow, Russia The thermal stability of peroxidase from leaves of the African oil palm tree Elaeis guineensis. generally look for simple models that are approximations to the Lumry–Eyring model [17,19–21]. Recently a novel peroxidase has been isolated from the leaves of the African oil palm tree Elaeis guineensis. process. Finally, it is interesting to compare the thermal stability of AOPTP with that of other peroxidases. In our previous publication [8] we reported the results of a detailed investigation of the thermal