- Email: [email protected]
Enzyme activity determination on macromolecular substrates by isothermal titration calorimetry: application to mesophilic and psychrophilic chitinases
Biochimica et Biophysica Acta 1545 (2001) 349^356 www.elsevier.com/locate/bba
Enzyme activity determination on macromolecular substrates by isothermal titration calorimetry: application to mesophilic and psychrophilic chitinases Thierry Lonhienne
a;b
, Etienne Baise a , Georges Feller Charles Gerday a
a;
*, Vassilis Bouriotis b ,
a
b
Laboratory of Biochemistry, Institute of Chemistry B6, University of Liege, B-4000 Liege, Belgium Enzyme Technology Division, Institute of Molecular Biology and Biotechnology, 71110 Heraklion, Crete, Greece Received 17 October 2000; received in revised form 7 December 2000; accepted 12 December 2000
Abstract Isothermal titration calorimetry has been applied to the determination of the kinetic parameters of chitinases (EC 3.2.1.14) by monitoring the heat released during the hydrolysis of chitin glycosidic bonds. Experiments were carried out using two different macromolecular substrates: a soluble polymer of N-acetylglucosamine and the insoluble chitin from crab shells. Different experimental temperatures were used in order to compare the thermodependence of the activity of two chitinases from the psychrophile Arthrobacter sp. TAD20 and of chitinase A from the mesophile Serratia marcescens. The method allowed to determine unequivocally the catalytic rate constant kcat , the activation energy (Ea ) and the thermodynamic activation parameters (vG# , vH# , vS# ) of the chitinolytic reaction on the soluble substrate. The catalytic activity has also been determined on insoluble chitin, which displays an effect of substrate saturation by chitinases. On both substrates, the thermodependence of the activity of the psychrophilic chitinases was lower than that observed with the mesophilic counterpart. ß 2001 Elsevier Science B.V. All rights reserved. Keywords: Psychrophile; Extremophile; Microcalorimetry ; Isothermal titration calorimetry; Chitinase; Insoluble chitin
1. Introduction Chitin, a major component of the earth's biomass, is an unbranched polysaccharide composed of L-1,4linked N-acetyl-D-glucosamine (GlcNAc). The polysaccharide, designated as `animal cellulose', is a major component of the crustacean exoskeleton corre-
* Corresponding author: Fax: +32-4-366-33-64; E-mail: [email protected]
sponding to an annual production of billions of tons and resulting in a continuous `rain' of chitin on the ocean £oor. However, marine sediments contain relatively little chitin thanks to a wide distribution of marine bacteria degrading and catabolizing chitinous particles, allowing carbon and nitrogen to return to the ecosystem [1]. Arthrobacter sp. TAD20 is a Gram-positive bacterium of marine origin collected along the Antarctic ice shell that secretes mainly two chitinases in the culture medium in response to chitin induction. Psychrophilic bacteria living in permanently cold environments have to adapt to low temperatures by synthesizing cold-active enzymes. In-
0167-4838 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 0 ) 0 0 2 9 6 - X
BBAPRO 36362 24-1-01
350
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
deed, according to the Arrhenius equation kcat A e3E a =RT
1
relating the catalytic rate constant to the activation energy [2], any decrease in temperature causes an exponential decrease of the reaction rate, the magnitude of which depends on the activation energy, Ea . Accordingly, most enzymatic reactions display a reaction rate 2 or 3 times lower when the temperature is decreased by 10³C (Q10 = 2^3). As a consequence, the activity of a mesophilic enzyme is between 16 and 80 times lower when the reaction temperature is shifted from 37³C to 0³C. The main adaptation of psychrophilic enzymes resides in the decrease of the activation energy that renders the reaction rates less temperature-dependent, allowing to maintain sustainable activity at low temperatures [3,4]. The calculation of the activation thermodynamic parameters of the reaction (vG# , vH# , vS# ) is also of interest because it allows to determine the relative importance of the enthalpic and entropic contributions to the free energy of activation of compared enzymes [5,6]. Commonly used methods for the determination of glycosidase activity on polysaccharides are based on the release of reducing groups followed by their detection using 3,5-dinitrosalicylate [7], alkaline ferricyanide [8], alkaline copper [9] and in the particular case of chitin, the modi¢ed Schales procedure [10,11]. All these methods present the disadvantage of being end point methods, lacking precision and sensitivity, which in addition are a¡ected by various interfering compounds. In comparison, isothermal titration calorimetry (ITC) appears to be a convenient and powerful method that overcomes the complexity of the reaction mixtures and does not su¡er from interference problems. Its sensitivity only depends on the absolute value of the molar enthalpy of substrate hydrolysis. The use of microcalorimetry for enzymatic activity measurements has been considered previously [12^15] but essentially to record raw activities on soluble substrates without attention to the conditions required for steady-state kinetics. In this work, we have used ITC to measure chitinase activities on soluble and insoluble chitin over a wide range of temperatures; the method can be considered as general for direct and accurate determination of kinetic parameters derived from enzymatic hydro-
lysis of soluble and insoluble macromolecular substrates. 2. Materials and methods 2.1. Substrates N,NP-Diacetylchitobiose was from Sigma. Soluble chitin was prepared according to the following procedure. Chitin coarse £akes from crab shells (Sigma) were partially hydrolyzed by acid treatment as described for colloidal chitin preparation [16]. The acidic colloidal chitin solution was then centrifuged at 10 000Ug and the supernatant dialyzed against 20 mM Tris-HCl, pH 7.5. The sample was then concentrated on Amicon membrane (30 000 Da cuto¡) and loaded on a Sephacryl S-200 column (2.5U100 cm) equilibrated in 20 mM Tris-HCl, 500 mM NaCl, pH 7.5. The sugar concentration in the fractions was evaluated by the method of Dubois [17]. Fractions containing soluble chitin (40 000^100 000 Da) were pooled and the sample was dialyzed against 50 mM HEPES, pH 7.5 and then concentrated by ultra¢ltration (30 000 Da cuto¡) to a ¢nal concentration of 5 mg/ml. The sample was stored at 5³C. Powdered colloidal chitin from crabs (Sigma) was ¢ltered on a nylon net with a mesh of 300 Wm in order to eliminate the largest aggregates that would block the injection syringe. The ¢ltrate was centrifuged and washed twice with 50 mM HEPES, pH 7.5 in order to eliminate any soluble fraction. The sample was then dispersed in the above-mentioned bu¡er to achieve a ¢nal concentration of 10 mg/ml and frozen at 320³C. 2.2. Enzymes The two psychrophilic chitinases A and B were puri¢ed from cultures of the Gram-positive Antarctic bacterium Arthrobacter sp. TAD20. The strain was grown at 5³C for 5 days in 3 l of broth containing 5 g/l bactotryptone, 1 g/l yeast extract, 33 g/l sea salts, pH 7.3, and 1 g/l colloidal chitin to induce secretion of chitinases A and B. After centrifugation at 11 000Ug for 15 min, the supernatant was concentrated to 400 ml and dia¢ltrated against 20 mM TrisHCl, pH 6.5, using a Minitan tangential £ow ultra-
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
¢ltration unit (Millipore) ¢tted with PTGC membranes (10 000 Da cuto¡). The sample was then loaded on a QFF-Sepharose column (2U20 cm) equilibrated in the above-mentioned bu¡er. The £ow through containing the chitinolytic activity was dialyzed against 20 mM Tris-HCl, pH 8 and loaded on another QFF-Sepharose column (2U20 cm) equilibrated in the same bu¡er and eluted with a NaCl linear gradient (0^200 mM, 250^250 ml). Chitinase A ( þ 25 mg) was obtained as a pure component whereas chitinase B eluted in the £ow through was concentrated to 10 ml and applied to a Sephacryl S-200 column (2.5U100 cm) eluted with 20 mM Tris-HCl, 100 mM NaCl, pH 7.5. The chitinolytic fractions corresponded to pure chitinase B ( þ 90 mg). The puri¢ed chitinases A and B were conditioned in HEPES 50 mM, pH 7.5, concentrated to 5 mg/ml and kept at 5³C. In these conditions the enzymes were stable for at least 3 months. The molecular masses calculated from the DNA sequence of chitinases A (EMBL accession No. asp250585) and B (EMBL accession No. asp250586) are 89 415 and 57 123 Da respectively. Chitinase A and chitobiase from Serratia marcescens were kindly provided by Prof. C. Vorgias. 2.3. Microcalorimetry Thermograms were recorded on a MCS Isothermal Titration Calorimeter Unit (MicroCal, Northampton, MA). The molar enthalpy of chitin L-1,4 linkage was determined by ¢lling the ITC sample cell (1.33 ml) with a 1 mg/ml solution of N,NP-diacetylchitobiose (Mr = 424.4), corresponding to 3.13 Wmoles N,NP-diacetylchitobiose, in 50 mM HEPES, pH 7.5. The enthalpy of the hydrolysis reaction corresponded to the heat generated after total substrate hydrolysis by chitobiase from S. marcescens. This was measured by integrating the area between the calorimetric trace and the baseline [13]. The molar enthalpy of bond hydrolysis (583 cal/mol) was obtained by dividing the integrated area (in Wcal) by 3.13 Wmoles of N,NP-diacetylchitobiose. In the case of enzyme activity determination, reactions were initiated by injecting 5^10 Wl of enzyme solution into the sample cell ¢lled with the substrate solution: 5 mg/ml and 10 mg/ml of soluble and insoluble chitin in 50 mM HEPES, pH 7.5 were used respectively. The stirring rate provided
351
by the syringe paddle rotation was set to 400 rpm in order to maintain a uniform substrate suspension. The catalytic rate constant kcat (min31 ) expressed as the number of L-1,4 linkages hydrolyzed per minute and per active site was calculated by dividing the heat generated in the reaction (in Wcal released per minute per mole of enzyme) by the molar enthalpy of chitin L-1,4 linkage (583 cal/mol). The activation energy was calculated from the slope (3Ea /R) of Arrhenius plots and the thermodynamic activation parameters of the chitinolytic reactions were calculated according to the following equations [6,18]:
v G # RTU 23:76 ln T3ln kcat
2
in kJ/mol, with kcat in s31 to conform to Bolzmann and Planck constant units
v H # E a 3RT
3
v S # v H # 3v G# =T
4
Errors on vG# were calculated using the relation v G# Err RTU kcat Err =kcat
5
3. Results 3.1. Enthalpy of chitin L-1,4 linkage hydrolysis The molar enthalpy corresponding to the hydrolysis of chitin was measured by monitoring the heat released during the total hydrolysis of N,NP-diacetylchitobiose, the dimer of N-acetylglucosamine, in the presence of chitobiase (Fig. 1). This was performed at 7³C, 15³C, 25³C, and 35³C in order to check the temperature dependence of the molar enthalpy. The data (average of three measurements) showed no signi¢cant variation with temperature. This is in agreement with the results obtained on maltohexaose for K-1,6 linkage [19]. Consequently, the mean of all measurements (583 þ 13 cal/mol linkage) was taken as the molar enthalpy of chitin L-1,4 linkage hydrolysis. 3.2. Calorimetric determination of enzyme activity on soluble chitin The molecular mass dispersion of the soluble chi-
BBAPRO 36362 24-1-01
352
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
Fig. 1. Typical calorimetric traces (heat £ow versus time) obtained after addition of chitobiase (5 Wg) to the ITC sample cell containing N,NP-diacetylchitobiose (3.13 Wmoles). The arrow corresponds to the enzyme injection. The integrated area under the curve divided by the amount of N,NP-diacetylchitobiose hydrolyzed in the cell is equal to the enthalpy change (vH) of the reaction. Note that in these conditions a steady state is not reached as indicated by the progressive decrease of activity resulting from substrate depletion. This experiment was performed in 50 mM HEPES, pH 7.5 at 15³C.
tin polymer has been estimated by gel ¢ltration on a Sephacryl S-200 column calibrated with polyethylene glycol of various masses. The soluble chitin mass ranges from 40 000 to 100 000 Da, corresponding to a maximum polymer length of about 400 units. The optimal bu¡er for activity of the three chitinases was 50 mM HEPES, pH 7.5. The kcat value was determined by recording the reaction rate under steadystate conditions. The substrate (5 mg/ml) was at saturating concentration for the three enzymes assayed, at all temperatures investigated. Fig. 2 shows a typical calorimetric trace (heat versus time) obtained after three successive injections of equal amounts of chitinase. The signal output is negative because it represents the variation of the current feedback (vCFB) required to compensate for the heat released by the reaction in the sample cell. Each injection gives rise to the same heat £ow production, demonstrating the linearity of activity as a function of enzyme concentration. The catalytic rate constant kcat (min31 ) was calculated by converting the heat £ow experimental values (Wcal/s) to moles of L-1,4 glycosidic linkages hydrolyzed per minute. In order to check the validity of this method, two reactions
Fig. 2. Typical calorimetric trace of chitinase assay on soluble chitin. Arrows correspond to the injection of equal amounts of enzyme (12.3 Wg of chitinase B) into the reaction cell. vCFB is the variation of the current feedback and is proportional to the heat released in the sample cell following hydrolysis of L-1,4 linkages. Under these conditions, a steady state is reached as demonstrated by the constant enzyme activity (heat £ow) after injection. The experiment was performed at 15³C in 50 mM HEPES, pH 7.5 containing 5 mg/ml soluble chitin.
were run at 25³C in rigorously identical conditions using chitinase B from Arthrobacter sp. TAD20. One reaction was recorded by microcalorimetry as described. In the second reaction, the release of reducing groups during soluble chitin hydrolysis was fol-
Fig. 3. Temperature dependence of kcat for soluble chitin hydrolysis using chitinases A (a) and B (b) from the Antarctic strain Arthrobacter sp. TAD20 and chitinase A (c) from S. marcescens. (Inset) Arrhenius plots (ln kcat versus reciprocal absolute temperature) for the three chitinases. Solid lines correspond to linear regressions on the experimental data.
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
353
Table 1 Thermodynamic activation parameters for chitinases A and B from the Antarctic Arthrobacter sp. TAD20 and for chitinase A from the mesophile S. marcescens at 15³C Parameter 31
kcat (min ) vG# (kJ mol31 ) Ea (kJ mol31 ) vH# (kJ mol31 ) TvS# (kJ mol31 )
Chitinase A Arthrobacter
Chitinase B Arthrobacter
Chitinase A S. marcescens
103 þ 8 69.2 þ 0.2 62.6 þ 1.1 60.2 þ 1.1 39.0 þ 1.3
81 þ 7 69.7 þ 0.2 62.1 þ 1.1 59.7 þ 1.1 310.0 þ 1.3
235 þ 8 67.2 þ 0.1 76.7 þ 1.4 74.3 þ 1.4 7.1 þ 1.5
lowed by the modi¢ed Schales procedure [11] and activity was calculated using N-acetyl-D-glucosamine as standard. The ITC method yielded a kcat value of 227 þ 12 min31 , in good agreement with the kcat value of 207 þ 15 min31 obtained by the colorimetric method. Fig. 3 shows the thermodependence of the chitinolytic activity for the three chitinases studied in the present work. The activation energies of the chitinolytic reactions were calculated by determining the slope (3Ea /R) of Arrhenius plots (Fig. 3, inset) and the thermodynamic activation parameters were calculated at 15³C for the three chitinases (Table 1). 3.3. Evidence for insoluble chitin saturation by chitinases As shown in Fig. 4, the successive injections of chitinase aliquots into the substrate suspension show a substrate saturation phenomenon. Indeed, as the number of identical enzyme injections increases, the corresponding heat release (vCFB) decreases to reach zero when chitin is totally saturated by the enzyme. Any further injection of enzyme does not increase the heat release; the activity is at its maximum level and is not limited by substrate depletion. This was checked by undertaking two reactions of hydrolysis under identical conditions. The ¢rst one was followed using the microcalorimeter whereas the second was incubated at 15³C. After an elapsed time corresponding to the measurement of the saturation phenomenon in the ¢rst reaction, the second batch reaction was stopped by heating the solution for 10 min at 90³C in order to inactivate the enzyme. The solution was then used as substrate for a new calorimetric experiment in the conditions of the ¢rst reaction. A similar signal output was again observed showing that the phenomenon is not due to substrate
Fig. 4. Insoluble chitin hydrolysis. (Upper panel) Calorimetric traces of a chitinase assay showing the saturation e¡ect of insoluble chitin by the enzyme. Arrows indicate injections of equal amounts of enzyme (70 Wg of chitinase B from Arthrobacter sp. TAD20 per injection) into the reaction cell. The negative signal following injections corresponds to the heat of dilution of the enzyme in the substrate. (Lower panel) At su¤ciently low enzyme concentration, a linear relationship between the heat released and the amount of enzyme injected into the cell is reached. Arrow indicates injections of 10 Wg and 20 Wg of chitinase B from Arthrobacter sp. TAD20 corresponding to curve a and b, respectively. Experiments were performed at 15³C in 50 mM HEPES, pH 7.5 containing 10 mg/ml insoluble chitin.
BBAPRO 36362 24-1-01
354
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
Fig. 5. E¡ect of temperature on the catalytic activity of chitinases A (a) and B (b) from Arthrobacter sp. TAD20 and of chitinase A (c) from S. marcescens towards insoluble chitin. Specific activity is given in Wmole of L-1,4 linkage hydrolyzed per Wmole of enzyme per minute. The averages are the results of three measurements.
depletion nor to inhibition by the released products but to substrate saturation by the enzyme. In order to calculate kcat and to avoid the substrate saturation e¡ect, the amount of enzyme injected into the cell was reduced to a level at which a linear relationship was observed between the heat £ow released and the chitinase concentration (Fig. 4). All further activity measurements were performed under these conditions. The catalytic activity of the three chitinases plotted as a function of temperature is shown in Fig. 5: the weaker thermodependence of the psychrophilic enzymes is also observed. 4. Discussion Psychrophilic enzymes perform catalysis at a temperature close to the freezing point of water [20]. The comparative study of psychrophilic and mesophilic chitinases by isothermal titration calorimetry (Table 1) reveals that both psychrophilic chitinases A and B display a typical `cold adapted' behavior on soluble chitin, i.e. a lower activation energy, Ea [18]. However, this lower activation energy is not translated into a higher speci¢c activity at low and moderate temperatures when compared to the mesophilic enzyme. Indeed, the mesophilic chitinase A from
S. marcescens is at least 2-fold more active than the psychrophilic enzymes at 15³C (Table 1). Analysis of the thermodynamic parameters clearly shows that the di¡erence in the turnover number is due to an unfavorable entropic contribution to the free energy of activation (vG# ) for the psychrophilic chitinases whereas this contribution is favorable for the mesophilic chitinase (Table 1). Two reasons may be suggested to explain this phenomenon. (i) According to the current hypothesis, the generally observed increased £exibility of psychrophilic enzymes is the main structural feature that provides an enhanced ability to undergo conformational changes during catalysis at low temperature [20]. However, this optimization of the catalytic rate constant is partly countered by an unfavorable entropic factor arising from the higher £exibility of the enzyme that contributes to increase the conformational entropy of the preactivated state [21]. (ii) It is probable that the soluble chitin used for catalytic activity measurements is not the optimal substrate for the enzymes studied here since chitinases have a widespread substrate speci¢city, i.e. various types of chitin characterized by di¡erent arrangement of strands, degrees of acetylation and cross-linking to other structural components [22]. For this reason, it seems more valuable to only pay attention to the variation of kcat of the psychrophilic and mesophilic enzymes with temperature, this characteristic being represented by the activation energy, Ea . The di¡erence in substrate speci¢city of the three chitinases is well demonstrated by the distinct results obtained on insoluble chitin. On this substrate, the psychrophilic chitinase B is clearly more active than the mesophilic chitinase A at low and moderate temperatures (Fig. 5) whereas the mesophilic enzyme maintains a large advantage when compared to the psychrophilic chitinase A, which displays a weak activity on insoluble chitin. These results point out the usefulness of microcalorimetry to perform accurate enzyme activity determination on natural macromolecular substrates. Moreover, the precision of the method allowed us to calculate the thermodynamic activation parameters on soluble chitin, which would be di¤cult to obtain accurately with end point methods. As far as insoluble chitin hydrolysis is concerned, two aspects are worth mentioning, i.e. the relatively long lag period (5^15 min) preceding the steady state
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
and the substrate saturation e¡ect (Fig. 4). Such a long lag phase in the time course of chitin hydrolysis has been previously reported [23]. According to the proposed model, this lag corresponds to the opening of crystalline regions of the colloid surface therefore exposing new ¢bril ends. The steady state is reached when ¢bril ends are formed continuously. The saturation phenomenon can be understood if one considers that most of the ¢bril population forms a largely interconnected network buried in the colloidal particle and therefore not accessible to enzymatic attack because of di¡usion constraints and crystallinity regions [23]. The hydrolysis of powdered colloidal chitin can be represented as the hydrolysis of bonds located at the surface of the particle. When all the surface is covered by chitinases, the possible excess of enzyme is therefore not in contact with the substrate and can be considered as inactive. The ITC method clearly detects such substrate saturation phenomenon and, subsequently, the conditions of the reaction can be easily adapted to accurately record the speci¢c activity at steady state. The essential feature of the ITC method lies in the universal heat exchange involved in any biochemical reaction. It implies the possibility of using any type of substrate and especially natural macromolecular substrates. Acknowledgements We thank Prof. Constantinos E. Vorgias (University of Athens) for kindly providing chitinase A and chitobiase from S. marcescens. This work was supported by the European Union under the form of a TMR contract CT970131 and by the Biotech contract COLDZYME BIO4-CT96-0051. References [1] C.E. Zobell, S.C. Rittenberg, The occurrence and characteristics of chitinoclastic bacteria in the sea, J. Bacteriol. 35 (1937) 275^287. [2] A. Fersht, Enzyme Structure and Mechanism, W.H. Freeman, New York, 1985. [3] S. Davail, G. Feller, E. Narinx, C. Gerday, Cold adaptation of proteins. Puri¢cation, characterization, and sequence of the heat-labile subtilisin from the Antarctic psychrophile Bacillus TA41, J. Biol. Chem. 269 (1994) 17448^17453.
355
[4] G. Feller, T. Lonhienne, C. Deroanne, C. Libioulle, J. Van Beeumen, C. Gerday, Puri¢cation, characterization, and nucleotide sequence of the thermolabile alpha-amylase from the Antarctic psychrotroph Alteromonas haloplanctis A23, J. Biol. Chem. 267 (1992) 5217^5221. [5] P.S. Low, J.L. Bada, G.N. Somero, Temperature adaptation of enzymes: roles of the free energy, the enthalpy, and the entropy of activation, Proc. Natl. Acad. Sci. USA 70 (1973) 430^432. [6] T. Lonhienne, C. Gerday, G. Feller G, Psychrophilic enzymes: revisiting the thermodynamic parameters of activation may explain local £exibility, Biochim. Biophys. Acta 1543 (2000) 1^10. [7] G.H. Miller, Use of dinitrosalicylic acid reagent for determination of reducing sugar, Anal. Chem. 31 (1959) 426^ 428. [8] J.F. Robyt, S. Bemis, Use of the autoanalyzer for determining the blue value of the amylose-iodine complex and total carbohydrate by phenol-sulfuric acid, Anal. Biochem. 19 (1967) 56^60. [9] S. Dygert, L.H. Li, D. Florida, J.A. Thoma, Determination of reducing sugar with improved precision, Anal. Biochem. 13 (1965) 367^374. [10] T. Imoto, K. Yagishita, A simple activity measurement of lysozyme, Agric. Biol. Chem. 35 (1971) 1154^1156. [11] J.L. Reissig, J.L. Strominger, L.F. Leloir, A modi¢ed colorimetric method for the estimation of N-acetylamino sugars, J. Biol. Chem. 217 (1955) 959^966. [12] B.A. Williams, E.J. Toone, Calorimetric evaluation of enzyme kinetic parameters, J. Org. Chem. 58 (1993) 3507^ 3510. [13] P.E. Morin, E. Freire, Direct calorimetric analysis of the enzymatic activity of yeast cytochrome c oxidase, Biochemistry 30 (1991) 8494^8500. [14] L. Sica, R. Gilli, C. Briand, J.C. Sari, A £ow microcalorimetric method for enzyme activity measurements: application to dihydrofolate reductase, Anal. Biochem. 165 (1987) 341^348. [15] S. Rosenstein, H.D. Brown, Microcalorimetric assay of acetylcholinesterase, Biochim. Biophys. Acta 629 (1980) 195^ 198. [16] S.C. Hsu, J.L. Lockwood, Powdered chitin agar as a selective medium for enumeration of actinomycetes in water and soil, Appl. Microbiol. 29 (1975) 422^426. [17] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Rebers, F. Smith, Colorimetric method for determination of sugars and related substances, Anal. Chem. 28 (1956) 350^356. [18] G. Feller, C. Gerday, Psychrophilic enzymes: molecular basis of cold adaptation, Cell. Mol. Life Sci. 53 (1997) 830^ 841. [19] R.N. Goldberg, D. Bell, Y.B. Tewari, M.A. McLaughlin, Thermodynamics of hydrolysis of oligosaccharides, Biophys. Chem. 40 (1991) 69^76. [20] G. Feller, J.L. Arpigny, E. Narinx, C. Gerday, Molecular adaptations of enzymes from psychrophilic organisms, Comp. Biochem. Physiol. 118A (1997) 495^499.
BBAPRO 36362 24-1-01
356
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
[21] P.A. Fields, G.N. Somero, Hot spots in cold adaptation: localized increases in conformational £exibility in lactate dehydrogenase A4 orthologs of Antarctic notothenioid ¢shes, Proc. Natl. Acad. Sci. USA 95 (1998) 11476^11481. [22] A.L. Svitil, D.L. Kirchman, A chitin-binding domain in a marine bacterial chitinase and other microbial chitinases:
implications for the ecology and evolution of 1,4-beta-glycanases, Microbiology 144 (1998) 1299^1308. [23] R.A. Smucker, Interpretative model for enzymatic chitin hydrolysis, in: R. Muzzarelli, C. Jeuniaux, G.W. Gooday (Eds.), Chitin in Nature and Technology, Plenum Press, New York, 1986, pp. 254^258.
BBAPRO 36362 24-1-01
Enzyme activity determination on macromolecular substrates by isothermal titration calorimetry: application to mesophilic and psychrophilic chitinases Thierry Lonhienne
a;b
, Etienne Baise a , Georges Feller Charles Gerday a
a;
*, Vassilis Bouriotis b ,
a
b
Laboratory of Biochemistry, Institute of Chemistry B6, University of Liege, B-4000 Liege, Belgium Enzyme Technology Division, Institute of Molecular Biology and Biotechnology, 71110 Heraklion, Crete, Greece Received 17 October 2000; received in revised form 7 December 2000; accepted 12 December 2000
Abstract Isothermal titration calorimetry has been applied to the determination of the kinetic parameters of chitinases (EC 3.2.1.14) by monitoring the heat released during the hydrolysis of chitin glycosidic bonds. Experiments were carried out using two different macromolecular substrates: a soluble polymer of N-acetylglucosamine and the insoluble chitin from crab shells. Different experimental temperatures were used in order to compare the thermodependence of the activity of two chitinases from the psychrophile Arthrobacter sp. TAD20 and of chitinase A from the mesophile Serratia marcescens. The method allowed to determine unequivocally the catalytic rate constant kcat , the activation energy (Ea ) and the thermodynamic activation parameters (vG# , vH# , vS# ) of the chitinolytic reaction on the soluble substrate. The catalytic activity has also been determined on insoluble chitin, which displays an effect of substrate saturation by chitinases. On both substrates, the thermodependence of the activity of the psychrophilic chitinases was lower than that observed with the mesophilic counterpart. ß 2001 Elsevier Science B.V. All rights reserved. Keywords: Psychrophile; Extremophile; Microcalorimetry ; Isothermal titration calorimetry; Chitinase; Insoluble chitin
1. Introduction Chitin, a major component of the earth's biomass, is an unbranched polysaccharide composed of L-1,4linked N-acetyl-D-glucosamine (GlcNAc). The polysaccharide, designated as `animal cellulose', is a major component of the crustacean exoskeleton corre-
* Corresponding author: Fax: +32-4-366-33-64; E-mail: [email protected]
sponding to an annual production of billions of tons and resulting in a continuous `rain' of chitin on the ocean £oor. However, marine sediments contain relatively little chitin thanks to a wide distribution of marine bacteria degrading and catabolizing chitinous particles, allowing carbon and nitrogen to return to the ecosystem [1]. Arthrobacter sp. TAD20 is a Gram-positive bacterium of marine origin collected along the Antarctic ice shell that secretes mainly two chitinases in the culture medium in response to chitin induction. Psychrophilic bacteria living in permanently cold environments have to adapt to low temperatures by synthesizing cold-active enzymes. In-
0167-4838 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 0 ) 0 0 2 9 6 - X
BBAPRO 36362 24-1-01
350
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
deed, according to the Arrhenius equation kcat A e3E a =RT
1
relating the catalytic rate constant to the activation energy [2], any decrease in temperature causes an exponential decrease of the reaction rate, the magnitude of which depends on the activation energy, Ea . Accordingly, most enzymatic reactions display a reaction rate 2 or 3 times lower when the temperature is decreased by 10³C (Q10 = 2^3). As a consequence, the activity of a mesophilic enzyme is between 16 and 80 times lower when the reaction temperature is shifted from 37³C to 0³C. The main adaptation of psychrophilic enzymes resides in the decrease of the activation energy that renders the reaction rates less temperature-dependent, allowing to maintain sustainable activity at low temperatures [3,4]. The calculation of the activation thermodynamic parameters of the reaction (vG# , vH# , vS# ) is also of interest because it allows to determine the relative importance of the enthalpic and entropic contributions to the free energy of activation of compared enzymes [5,6]. Commonly used methods for the determination of glycosidase activity on polysaccharides are based on the release of reducing groups followed by their detection using 3,5-dinitrosalicylate [7], alkaline ferricyanide [8], alkaline copper [9] and in the particular case of chitin, the modi¢ed Schales procedure [10,11]. All these methods present the disadvantage of being end point methods, lacking precision and sensitivity, which in addition are a¡ected by various interfering compounds. In comparison, isothermal titration calorimetry (ITC) appears to be a convenient and powerful method that overcomes the complexity of the reaction mixtures and does not su¡er from interference problems. Its sensitivity only depends on the absolute value of the molar enthalpy of substrate hydrolysis. The use of microcalorimetry for enzymatic activity measurements has been considered previously [12^15] but essentially to record raw activities on soluble substrates without attention to the conditions required for steady-state kinetics. In this work, we have used ITC to measure chitinase activities on soluble and insoluble chitin over a wide range of temperatures; the method can be considered as general for direct and accurate determination of kinetic parameters derived from enzymatic hydro-
lysis of soluble and insoluble macromolecular substrates. 2. Materials and methods 2.1. Substrates N,NP-Diacetylchitobiose was from Sigma. Soluble chitin was prepared according to the following procedure. Chitin coarse £akes from crab shells (Sigma) were partially hydrolyzed by acid treatment as described for colloidal chitin preparation [16]. The acidic colloidal chitin solution was then centrifuged at 10 000Ug and the supernatant dialyzed against 20 mM Tris-HCl, pH 7.5. The sample was then concentrated on Amicon membrane (30 000 Da cuto¡) and loaded on a Sephacryl S-200 column (2.5U100 cm) equilibrated in 20 mM Tris-HCl, 500 mM NaCl, pH 7.5. The sugar concentration in the fractions was evaluated by the method of Dubois [17]. Fractions containing soluble chitin (40 000^100 000 Da) were pooled and the sample was dialyzed against 50 mM HEPES, pH 7.5 and then concentrated by ultra¢ltration (30 000 Da cuto¡) to a ¢nal concentration of 5 mg/ml. The sample was stored at 5³C. Powdered colloidal chitin from crabs (Sigma) was ¢ltered on a nylon net with a mesh of 300 Wm in order to eliminate the largest aggregates that would block the injection syringe. The ¢ltrate was centrifuged and washed twice with 50 mM HEPES, pH 7.5 in order to eliminate any soluble fraction. The sample was then dispersed in the above-mentioned bu¡er to achieve a ¢nal concentration of 10 mg/ml and frozen at 320³C. 2.2. Enzymes The two psychrophilic chitinases A and B were puri¢ed from cultures of the Gram-positive Antarctic bacterium Arthrobacter sp. TAD20. The strain was grown at 5³C for 5 days in 3 l of broth containing 5 g/l bactotryptone, 1 g/l yeast extract, 33 g/l sea salts, pH 7.3, and 1 g/l colloidal chitin to induce secretion of chitinases A and B. After centrifugation at 11 000Ug for 15 min, the supernatant was concentrated to 400 ml and dia¢ltrated against 20 mM TrisHCl, pH 6.5, using a Minitan tangential £ow ultra-
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
¢ltration unit (Millipore) ¢tted with PTGC membranes (10 000 Da cuto¡). The sample was then loaded on a QFF-Sepharose column (2U20 cm) equilibrated in the above-mentioned bu¡er. The £ow through containing the chitinolytic activity was dialyzed against 20 mM Tris-HCl, pH 8 and loaded on another QFF-Sepharose column (2U20 cm) equilibrated in the same bu¡er and eluted with a NaCl linear gradient (0^200 mM, 250^250 ml). Chitinase A ( þ 25 mg) was obtained as a pure component whereas chitinase B eluted in the £ow through was concentrated to 10 ml and applied to a Sephacryl S-200 column (2.5U100 cm) eluted with 20 mM Tris-HCl, 100 mM NaCl, pH 7.5. The chitinolytic fractions corresponded to pure chitinase B ( þ 90 mg). The puri¢ed chitinases A and B were conditioned in HEPES 50 mM, pH 7.5, concentrated to 5 mg/ml and kept at 5³C. In these conditions the enzymes were stable for at least 3 months. The molecular masses calculated from the DNA sequence of chitinases A (EMBL accession No. asp250585) and B (EMBL accession No. asp250586) are 89 415 and 57 123 Da respectively. Chitinase A and chitobiase from Serratia marcescens were kindly provided by Prof. C. Vorgias. 2.3. Microcalorimetry Thermograms were recorded on a MCS Isothermal Titration Calorimeter Unit (MicroCal, Northampton, MA). The molar enthalpy of chitin L-1,4 linkage was determined by ¢lling the ITC sample cell (1.33 ml) with a 1 mg/ml solution of N,NP-diacetylchitobiose (Mr = 424.4), corresponding to 3.13 Wmoles N,NP-diacetylchitobiose, in 50 mM HEPES, pH 7.5. The enthalpy of the hydrolysis reaction corresponded to the heat generated after total substrate hydrolysis by chitobiase from S. marcescens. This was measured by integrating the area between the calorimetric trace and the baseline [13]. The molar enthalpy of bond hydrolysis (583 cal/mol) was obtained by dividing the integrated area (in Wcal) by 3.13 Wmoles of N,NP-diacetylchitobiose. In the case of enzyme activity determination, reactions were initiated by injecting 5^10 Wl of enzyme solution into the sample cell ¢lled with the substrate solution: 5 mg/ml and 10 mg/ml of soluble and insoluble chitin in 50 mM HEPES, pH 7.5 were used respectively. The stirring rate provided
351
by the syringe paddle rotation was set to 400 rpm in order to maintain a uniform substrate suspension. The catalytic rate constant kcat (min31 ) expressed as the number of L-1,4 linkages hydrolyzed per minute and per active site was calculated by dividing the heat generated in the reaction (in Wcal released per minute per mole of enzyme) by the molar enthalpy of chitin L-1,4 linkage (583 cal/mol). The activation energy was calculated from the slope (3Ea /R) of Arrhenius plots and the thermodynamic activation parameters of the chitinolytic reactions were calculated according to the following equations [6,18]:
v G # RTU 23:76 ln T3ln kcat
2
in kJ/mol, with kcat in s31 to conform to Bolzmann and Planck constant units
v H # E a 3RT
3
v S # v H # 3v G# =T
4
Errors on vG# were calculated using the relation v G# Err RTU kcat Err =kcat
5
3. Results 3.1. Enthalpy of chitin L-1,4 linkage hydrolysis The molar enthalpy corresponding to the hydrolysis of chitin was measured by monitoring the heat released during the total hydrolysis of N,NP-diacetylchitobiose, the dimer of N-acetylglucosamine, in the presence of chitobiase (Fig. 1). This was performed at 7³C, 15³C, 25³C, and 35³C in order to check the temperature dependence of the molar enthalpy. The data (average of three measurements) showed no signi¢cant variation with temperature. This is in agreement with the results obtained on maltohexaose for K-1,6 linkage [19]. Consequently, the mean of all measurements (583 þ 13 cal/mol linkage) was taken as the molar enthalpy of chitin L-1,4 linkage hydrolysis. 3.2. Calorimetric determination of enzyme activity on soluble chitin The molecular mass dispersion of the soluble chi-
BBAPRO 36362 24-1-01
352
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
Fig. 1. Typical calorimetric traces (heat £ow versus time) obtained after addition of chitobiase (5 Wg) to the ITC sample cell containing N,NP-diacetylchitobiose (3.13 Wmoles). The arrow corresponds to the enzyme injection. The integrated area under the curve divided by the amount of N,NP-diacetylchitobiose hydrolyzed in the cell is equal to the enthalpy change (vH) of the reaction. Note that in these conditions a steady state is not reached as indicated by the progressive decrease of activity resulting from substrate depletion. This experiment was performed in 50 mM HEPES, pH 7.5 at 15³C.
tin polymer has been estimated by gel ¢ltration on a Sephacryl S-200 column calibrated with polyethylene glycol of various masses. The soluble chitin mass ranges from 40 000 to 100 000 Da, corresponding to a maximum polymer length of about 400 units. The optimal bu¡er for activity of the three chitinases was 50 mM HEPES, pH 7.5. The kcat value was determined by recording the reaction rate under steadystate conditions. The substrate (5 mg/ml) was at saturating concentration for the three enzymes assayed, at all temperatures investigated. Fig. 2 shows a typical calorimetric trace (heat versus time) obtained after three successive injections of equal amounts of chitinase. The signal output is negative because it represents the variation of the current feedback (vCFB) required to compensate for the heat released by the reaction in the sample cell. Each injection gives rise to the same heat £ow production, demonstrating the linearity of activity as a function of enzyme concentration. The catalytic rate constant kcat (min31 ) was calculated by converting the heat £ow experimental values (Wcal/s) to moles of L-1,4 glycosidic linkages hydrolyzed per minute. In order to check the validity of this method, two reactions
Fig. 2. Typical calorimetric trace of chitinase assay on soluble chitin. Arrows correspond to the injection of equal amounts of enzyme (12.3 Wg of chitinase B) into the reaction cell. vCFB is the variation of the current feedback and is proportional to the heat released in the sample cell following hydrolysis of L-1,4 linkages. Under these conditions, a steady state is reached as demonstrated by the constant enzyme activity (heat £ow) after injection. The experiment was performed at 15³C in 50 mM HEPES, pH 7.5 containing 5 mg/ml soluble chitin.
were run at 25³C in rigorously identical conditions using chitinase B from Arthrobacter sp. TAD20. One reaction was recorded by microcalorimetry as described. In the second reaction, the release of reducing groups during soluble chitin hydrolysis was fol-
Fig. 3. Temperature dependence of kcat for soluble chitin hydrolysis using chitinases A (a) and B (b) from the Antarctic strain Arthrobacter sp. TAD20 and chitinase A (c) from S. marcescens. (Inset) Arrhenius plots (ln kcat versus reciprocal absolute temperature) for the three chitinases. Solid lines correspond to linear regressions on the experimental data.
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
353
Table 1 Thermodynamic activation parameters for chitinases A and B from the Antarctic Arthrobacter sp. TAD20 and for chitinase A from the mesophile S. marcescens at 15³C Parameter 31
kcat (min ) vG# (kJ mol31 ) Ea (kJ mol31 ) vH# (kJ mol31 ) TvS# (kJ mol31 )
Chitinase A Arthrobacter
Chitinase B Arthrobacter
Chitinase A S. marcescens
103 þ 8 69.2 þ 0.2 62.6 þ 1.1 60.2 þ 1.1 39.0 þ 1.3
81 þ 7 69.7 þ 0.2 62.1 þ 1.1 59.7 þ 1.1 310.0 þ 1.3
235 þ 8 67.2 þ 0.1 76.7 þ 1.4 74.3 þ 1.4 7.1 þ 1.5
lowed by the modi¢ed Schales procedure [11] and activity was calculated using N-acetyl-D-glucosamine as standard. The ITC method yielded a kcat value of 227 þ 12 min31 , in good agreement with the kcat value of 207 þ 15 min31 obtained by the colorimetric method. Fig. 3 shows the thermodependence of the chitinolytic activity for the three chitinases studied in the present work. The activation energies of the chitinolytic reactions were calculated by determining the slope (3Ea /R) of Arrhenius plots (Fig. 3, inset) and the thermodynamic activation parameters were calculated at 15³C for the three chitinases (Table 1). 3.3. Evidence for insoluble chitin saturation by chitinases As shown in Fig. 4, the successive injections of chitinase aliquots into the substrate suspension show a substrate saturation phenomenon. Indeed, as the number of identical enzyme injections increases, the corresponding heat release (vCFB) decreases to reach zero when chitin is totally saturated by the enzyme. Any further injection of enzyme does not increase the heat release; the activity is at its maximum level and is not limited by substrate depletion. This was checked by undertaking two reactions of hydrolysis under identical conditions. The ¢rst one was followed using the microcalorimeter whereas the second was incubated at 15³C. After an elapsed time corresponding to the measurement of the saturation phenomenon in the ¢rst reaction, the second batch reaction was stopped by heating the solution for 10 min at 90³C in order to inactivate the enzyme. The solution was then used as substrate for a new calorimetric experiment in the conditions of the ¢rst reaction. A similar signal output was again observed showing that the phenomenon is not due to substrate
Fig. 4. Insoluble chitin hydrolysis. (Upper panel) Calorimetric traces of a chitinase assay showing the saturation e¡ect of insoluble chitin by the enzyme. Arrows indicate injections of equal amounts of enzyme (70 Wg of chitinase B from Arthrobacter sp. TAD20 per injection) into the reaction cell. The negative signal following injections corresponds to the heat of dilution of the enzyme in the substrate. (Lower panel) At su¤ciently low enzyme concentration, a linear relationship between the heat released and the amount of enzyme injected into the cell is reached. Arrow indicates injections of 10 Wg and 20 Wg of chitinase B from Arthrobacter sp. TAD20 corresponding to curve a and b, respectively. Experiments were performed at 15³C in 50 mM HEPES, pH 7.5 containing 10 mg/ml insoluble chitin.
BBAPRO 36362 24-1-01
354
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
Fig. 5. E¡ect of temperature on the catalytic activity of chitinases A (a) and B (b) from Arthrobacter sp. TAD20 and of chitinase A (c) from S. marcescens towards insoluble chitin. Specific activity is given in Wmole of L-1,4 linkage hydrolyzed per Wmole of enzyme per minute. The averages are the results of three measurements.
depletion nor to inhibition by the released products but to substrate saturation by the enzyme. In order to calculate kcat and to avoid the substrate saturation e¡ect, the amount of enzyme injected into the cell was reduced to a level at which a linear relationship was observed between the heat £ow released and the chitinase concentration (Fig. 4). All further activity measurements were performed under these conditions. The catalytic activity of the three chitinases plotted as a function of temperature is shown in Fig. 5: the weaker thermodependence of the psychrophilic enzymes is also observed. 4. Discussion Psychrophilic enzymes perform catalysis at a temperature close to the freezing point of water [20]. The comparative study of psychrophilic and mesophilic chitinases by isothermal titration calorimetry (Table 1) reveals that both psychrophilic chitinases A and B display a typical `cold adapted' behavior on soluble chitin, i.e. a lower activation energy, Ea [18]. However, this lower activation energy is not translated into a higher speci¢c activity at low and moderate temperatures when compared to the mesophilic enzyme. Indeed, the mesophilic chitinase A from
S. marcescens is at least 2-fold more active than the psychrophilic enzymes at 15³C (Table 1). Analysis of the thermodynamic parameters clearly shows that the di¡erence in the turnover number is due to an unfavorable entropic contribution to the free energy of activation (vG# ) for the psychrophilic chitinases whereas this contribution is favorable for the mesophilic chitinase (Table 1). Two reasons may be suggested to explain this phenomenon. (i) According to the current hypothesis, the generally observed increased £exibility of psychrophilic enzymes is the main structural feature that provides an enhanced ability to undergo conformational changes during catalysis at low temperature [20]. However, this optimization of the catalytic rate constant is partly countered by an unfavorable entropic factor arising from the higher £exibility of the enzyme that contributes to increase the conformational entropy of the preactivated state [21]. (ii) It is probable that the soluble chitin used for catalytic activity measurements is not the optimal substrate for the enzymes studied here since chitinases have a widespread substrate speci¢city, i.e. various types of chitin characterized by di¡erent arrangement of strands, degrees of acetylation and cross-linking to other structural components [22]. For this reason, it seems more valuable to only pay attention to the variation of kcat of the psychrophilic and mesophilic enzymes with temperature, this characteristic being represented by the activation energy, Ea . The di¡erence in substrate speci¢city of the three chitinases is well demonstrated by the distinct results obtained on insoluble chitin. On this substrate, the psychrophilic chitinase B is clearly more active than the mesophilic chitinase A at low and moderate temperatures (Fig. 5) whereas the mesophilic enzyme maintains a large advantage when compared to the psychrophilic chitinase A, which displays a weak activity on insoluble chitin. These results point out the usefulness of microcalorimetry to perform accurate enzyme activity determination on natural macromolecular substrates. Moreover, the precision of the method allowed us to calculate the thermodynamic activation parameters on soluble chitin, which would be di¤cult to obtain accurately with end point methods. As far as insoluble chitin hydrolysis is concerned, two aspects are worth mentioning, i.e. the relatively long lag period (5^15 min) preceding the steady state
BBAPRO 36362 24-1-01
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
and the substrate saturation e¡ect (Fig. 4). Such a long lag phase in the time course of chitin hydrolysis has been previously reported [23]. According to the proposed model, this lag corresponds to the opening of crystalline regions of the colloid surface therefore exposing new ¢bril ends. The steady state is reached when ¢bril ends are formed continuously. The saturation phenomenon can be understood if one considers that most of the ¢bril population forms a largely interconnected network buried in the colloidal particle and therefore not accessible to enzymatic attack because of di¡usion constraints and crystallinity regions [23]. The hydrolysis of powdered colloidal chitin can be represented as the hydrolysis of bonds located at the surface of the particle. When all the surface is covered by chitinases, the possible excess of enzyme is therefore not in contact with the substrate and can be considered as inactive. The ITC method clearly detects such substrate saturation phenomenon and, subsequently, the conditions of the reaction can be easily adapted to accurately record the speci¢c activity at steady state. The essential feature of the ITC method lies in the universal heat exchange involved in any biochemical reaction. It implies the possibility of using any type of substrate and especially natural macromolecular substrates. Acknowledgements We thank Prof. Constantinos E. Vorgias (University of Athens) for kindly providing chitinase A and chitobiase from S. marcescens. This work was supported by the European Union under the form of a TMR contract CT970131 and by the Biotech contract COLDZYME BIO4-CT96-0051. References [1] C.E. Zobell, S.C. Rittenberg, The occurrence and characteristics of chitinoclastic bacteria in the sea, J. Bacteriol. 35 (1937) 275^287. [2] A. Fersht, Enzyme Structure and Mechanism, W.H. Freeman, New York, 1985. [3] S. Davail, G. Feller, E. Narinx, C. Gerday, Cold adaptation of proteins. Puri¢cation, characterization, and sequence of the heat-labile subtilisin from the Antarctic psychrophile Bacillus TA41, J. Biol. Chem. 269 (1994) 17448^17453.
355
[4] G. Feller, T. Lonhienne, C. Deroanne, C. Libioulle, J. Van Beeumen, C. Gerday, Puri¢cation, characterization, and nucleotide sequence of the thermolabile alpha-amylase from the Antarctic psychrotroph Alteromonas haloplanctis A23, J. Biol. Chem. 267 (1992) 5217^5221. [5] P.S. Low, J.L. Bada, G.N. Somero, Temperature adaptation of enzymes: roles of the free energy, the enthalpy, and the entropy of activation, Proc. Natl. Acad. Sci. USA 70 (1973) 430^432. [6] T. Lonhienne, C. Gerday, G. Feller G, Psychrophilic enzymes: revisiting the thermodynamic parameters of activation may explain local £exibility, Biochim. Biophys. Acta 1543 (2000) 1^10. [7] G.H. Miller, Use of dinitrosalicylic acid reagent for determination of reducing sugar, Anal. Chem. 31 (1959) 426^ 428. [8] J.F. Robyt, S. Bemis, Use of the autoanalyzer for determining the blue value of the amylose-iodine complex and total carbohydrate by phenol-sulfuric acid, Anal. Biochem. 19 (1967) 56^60. [9] S. Dygert, L.H. Li, D. Florida, J.A. Thoma, Determination of reducing sugar with improved precision, Anal. Biochem. 13 (1965) 367^374. [10] T. Imoto, K. Yagishita, A simple activity measurement of lysozyme, Agric. Biol. Chem. 35 (1971) 1154^1156. [11] J.L. Reissig, J.L. Strominger, L.F. Leloir, A modi¢ed colorimetric method for the estimation of N-acetylamino sugars, J. Biol. Chem. 217 (1955) 959^966. [12] B.A. Williams, E.J. Toone, Calorimetric evaluation of enzyme kinetic parameters, J. Org. Chem. 58 (1993) 3507^ 3510. [13] P.E. Morin, E. Freire, Direct calorimetric analysis of the enzymatic activity of yeast cytochrome c oxidase, Biochemistry 30 (1991) 8494^8500. [14] L. Sica, R. Gilli, C. Briand, J.C. Sari, A £ow microcalorimetric method for enzyme activity measurements: application to dihydrofolate reductase, Anal. Biochem. 165 (1987) 341^348. [15] S. Rosenstein, H.D. Brown, Microcalorimetric assay of acetylcholinesterase, Biochim. Biophys. Acta 629 (1980) 195^ 198. [16] S.C. Hsu, J.L. Lockwood, Powdered chitin agar as a selective medium for enumeration of actinomycetes in water and soil, Appl. Microbiol. 29 (1975) 422^426. [17] M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Rebers, F. Smith, Colorimetric method for determination of sugars and related substances, Anal. Chem. 28 (1956) 350^356. [18] G. Feller, C. Gerday, Psychrophilic enzymes: molecular basis of cold adaptation, Cell. Mol. Life Sci. 53 (1997) 830^ 841. [19] R.N. Goldberg, D. Bell, Y.B. Tewari, M.A. McLaughlin, Thermodynamics of hydrolysis of oligosaccharides, Biophys. Chem. 40 (1991) 69^76. [20] G. Feller, J.L. Arpigny, E. Narinx, C. Gerday, Molecular adaptations of enzymes from psychrophilic organisms, Comp. Biochem. Physiol. 118A (1997) 495^499.
BBAPRO 36362 24-1-01
356
T. Lonhienne et al. / Biochimica et Biophysica Acta 1545 (2001) 349^356
[21] P.A. Fields, G.N. Somero, Hot spots in cold adaptation: localized increases in conformational £exibility in lactate dehydrogenase A4 orthologs of Antarctic notothenioid ¢shes, Proc. Natl. Acad. Sci. USA 95 (1998) 11476^11481. [22] A.L. Svitil, D.L. Kirchman, A chitin-binding domain in a marine bacterial chitinase and other microbial chitinases:
implications for the ecology and evolution of 1,4-beta-glycanases, Microbiology 144 (1998) 1299^1308. [23] R.A. Smucker, Interpretative model for enzymatic chitin hydrolysis, in: R. Muzzarelli, C. Jeuniaux, G.W. Gooday (Eds.), Chitin in Nature and Technology, Plenum Press, New York, 1986, pp. 254^258.
BBAPRO 36362 24-1-01