Effects of water-miscible solvents and polyhydroxy compounds on the structure and enzymatic activity of thermolysin

Journal of Biotechnology 127 (2006) 45–53

Effects of water-miscible solvents and polyhydroxy compounds on the structure and enzymatic activity of thermolysin Mohammad Pazhang, Khosro Khajeh ∗ , Bijan Ranjbar, Saman Hosseinkhani Department of Biochemistry and Biophysics, Faculty of Sciences, Tarbiat Modares University, P.O. Box 14115-175, Tehran, Iran Received 10 January 2006; received in revised form 16 May 2006; accepted 31 May 2006

Abstract The effect of organic solvents (n-propanol, isopropanol, dimethylformamide and dimethylsulfoxide) on the structure, activity and stability of thermolysin was the focus of this investigation. Results show the ability of the solvents to cause mixed inhibition of thermolysin, which was indicated by kinetic and structural studies (near-UV CD spectra and intrinsic fluorescence). Inhibitory effect of the solvents increased with increments in solvents log P. Thermoinactivation of thermolysin was studied at 80 ◦ C in 50% of solvents and showed that with the increase in solvent hydrophobicity, thermal stability of the enzyme decreased. For the stabilization of thermolysin at high temperature, additives such as glycerol, sorbitol and trehalose were employed. In the presence of DMF with a relatively low log P, trehalose was shown to be a good stabilizer, whereas glycerol had a marked stabilization effect in the presence of n-propanol and isopropanol with a relatively high log P. Consequently, it was concluded that the stabilizing effect of additives can be correlated with the log P of solvents. © 2006 Elsevier B.V. All rights reserved. Keywords: Thermolysin; Organic solvents; Mixed inhibition; log P; Stabilization; Additives

1. Introduction The remarkable potential of enzymes as practical catalysts is well recognized. In particular, they are being increasingly exploited for asymmetric synthetic transformation and fuelled by the growing demand for enantiopure pharmaceutical (Klibanov, 2001; Gupta Abbreviations: TLN, thermolysin; DMSO, dimethyl sulfoxide; DMF, dimethyl formamide ∗ Corresponding author. Tel.: +98 21 88009730; fax: +98 21 88009730. E-mail address: [email protected] (K. Khajeh). 0168-1656/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2006.05.017

and Roy, 2004). Organic solvents are advantageous in enzyme-catalyzed peptide synthesis, both to solubilize substrates and product and to manipulate reaction kinetics and equilibrium to increase product yield. The use of organic solvents, however, leads to rapid inactivation of enzyme upon denaturation, conformational rigidity or enzyme inhibition (Zaks and Klibanov, 1988; Rodakiewicz-Nowak et al., 2000; Simon et al., 2001). Protein molecules in aqueous solution are surrounded by a hydration layer composed of water molecules associated with the protein surface. It has been suggested that organic solvents tend to displace the water molecules both in the hydration layer and

46

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

in the interior of the protein, thereby distorting the interactions responsible for maintaining the native conformation of the enzymes (Khemelntisky et al., 1991). Thermolysin (EC 3.4.24.27) a 316 amino acid thermostable neutral metallo-endopeptidase produced by Bacillus thermoproteolyticus (Endo, 1962; Latt et al., 1969), is widely used in organic media for transesterification of sucrose, synthesis of peptides and utilized in industrial scale for the synthesis of aspartame precursor (Clapes et al., 1995; Khalaf et al., 1996; Ulijn et al., 2000; Pedersen et al., 2002). Thermolysin (TLN) catalyzes the hydrolysis of the peptide bond specifically on the amino acid of large hydrophobic residues, in particular leucine, isoleucine and phenylalanine (Morihara, 1967). In recent studies, English et al. (1999) applied multiple solvent crystal structure (MSCS) to map experimentally the surface of TLN to identify interaction sites complementary to small molecules such as isopropanol (English et al., 1999). Although native TLN has a Val-Lys dipeptide bound in the active site interacting with the S1 and S2 subsites (Holland et al., 1995), it is displaced at relatively modest concentrations of isopropanol (∼1–2 M). It was demonstrated that the interacting subsites could be experimentally ranked and only 2 of 12 interaction subsites identified using neat solvent were occupied at low concentrations. Although the interaction of small organic solvents with active site and surface of TLN has been reported (English et al., 2001), the inhibition effect of these molecules on the enzymatic activity of TLN on casein as substrate has largely been overlooked. The aim of the present investigation was to identify the influence of isopropanol, propanol, dimethyl sulfoxide (DMSO) and dimethyl formamide (DMF) on the structure and function of thermolysin. The inhibitory effects of each solvent was determined and discussed with regards to the mix inhibition pattern. In addition, the combined effects of polyols (glycerol, trehalose and sorbitol) and different organic media on the catalytic activity and stability of TLN at high temperatures were studied.

2. Materials and methods Thermolysin, trehalose and sorbitol were purchased from Sigma (St. Louis, MO, USA). All other chemi-

cals were from Merck (Darmstadt, Germany) and were reagent grade. 2.1. Protease activity and kinetic measurements The protease activity of thermolysin was performed at 25 ◦ C in 50 mM Tris buffer, pH 7.2 containing 10 mM CaCl2 in 0.5 ml reaction volume. The reaction was carried out with 0.5% casein as substrate and 20 ␮g/ml enzyme concentration. After 5 min, the enzymatic activity was halted with 500 ␮l of 10% trichloroacetic acid (TCA) and subsequently, absorbance of the supernatant measured at 280 nm. One unit was considered the amount of enzyme, which releases soluble fragments equivalent to 0.001 A280 nm per minute at standard condition. In all experiments, water/organic solvent mixtures were prepared by mixing required amounts of the components (organic solvents, aqueous buffer and substrate solution) and the pH of the reaction mixture was adjusted to 7.2. In order to determine kinetic parameter, measurements were carried out using different substrate concentrations and the final concentration of the enzyme was 20 ␮g/ml. Blanks for each casein concentration were also set up and the differential absorbance used to determine the activity. Steady state kinetic parameters were determined at constant solvent content and experimental data was analyzed graphically and numerically using the Lineweaver–Burk equation (Price and Stevens, 1999; Copeland, 2000). One of the ways in which inhibition of enzymecatalyzed reactions can be discussed is in terms of a general scheme shown below:

It is assumed that the enzyme-containing complexes are in equilibrium with each other, i.e. that the breakdown of ES to generate product does not significantly disturb the equilibrium (Price and Stevens, 1999; Copeland, 2000). In the present work, all data obtained for a particular system (consisting of 100 experimental points, on average) were analyzed using mix model of enzyme inhibition as described in the following equa-

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

tion:







1 + [IK0I] Km 1 + [IK0i] 1 1 + = v0 Vmax [S0 ] Vmax

 (1)

where vo is the initial reaction rate, Vmax the maximum reaction rate, Km the binding constant for the casein to the enzyme and Ki and KI are the inhibition constants for binding of inhibitor to enzyme or enzyme–substrate complex, respectively. In this model a mix inhibitor displays finite but unequal affinity for both the free enzyme and the ES complex; hence the dissociation constants from each of these enzyme forms must be considered in the kinetic analysis of these inhibitors. Ki and KI were determined using secondary plots as described by following equations:   [I0 ] 1 + 1 KI = (2)  Vmax Vmax slope for inhibited reaction



[I0 ] = slope for uninhibited reaction × 1 + Ki

 (3)

 Hence a secondary plot of 1/Vmax against [I0 ] will be linear, the intercept on the [I0 ] axis giving −KI ; a graph of slope of primary plot against [I0 ] will also be linear, the intercept on the [I0 ] axis giving −Ki (Price and Stevens, 1999; Copeland, 2000).

47

holder. Results are expressed as molar ellipticity [θ] (◦ cm2 dmol−1 ), based on a mean amino acid residue weight (MWR) assuming average weights of 110. The concentration of enzyme solution was adjusted to 1.5 mg/ml. The molar ellipticity [θ] was calculated from the formula [θ]λ = (θ × 100MWR)/(cl), where c is the protein concentration in mg/ml, l the light path length in centimeters, and θ is the measured ellipticity in degrees wavelength λ. 2.4. Fluorescence measurements The intrinsic fluorescence intensity is an excellent parameter to monitor the polarity of aromatic side chains environment in a protein. Thermolysin has 3 tryptophan and 28 tyrosine residues. Fluorescence emission spectra of thermolysin at various concentrations of organic solvents were recorded on a PerkinElmer luminescence spectrometer LS 50B. Twenty micrograms per millilitre of TLN was prepared in different concentrations of organic solvents in 50 mM Tris buffer, pH 7.2 containing 10 mM CaCl2 and the system was mixed and allowed to equilibrate for 5 min. Emission spectra from 300 to 400 nm were recorded using an excitation wavelength of 280 nm.

2.2. Thermal stability studies in aqueous organic media The time course of irreversible thermoinactivation was measured by incubation of the enzyme (0.2 mg/ml) at 80 ◦ C in the absence and presence of 50% (v/v) of each solvent. At regular intervals, samples were removed, cooled on ice and the residual activity was determined. In each experiment, activity of the enzyme solution or enzyme/organic solvent mixtures kept on ice was considered as control (100%). 2.3. Circular dichroism (CD) measurements CD spectra were recorded in the near-UV range on a JASCO (Tokyo, Japan) J-715 spectropolarimeter equipped with a thermostatically-controlled cell

Fig. 1. Activity of thermolysin at different concentrations of npropanol (), isopropanol (), DMF (), and DMSO (). Different concentrations of solvents were prepared in 50 mM Tris buffer containing 10 mM CaCl2, and subsequently pH of solutions was adjusted to 7.2. Standard deviations were within 5% of the experimental values. For more details see Section 2.

48

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

Results presented in this paper are the mean from at least three repeated experiments in a typical run to confirm reproducibility

3. Results and discussion 3.1. Enzyme activity and structural characterization The effects of organic media on enzymatic reaction rates have been suggested to be of two types: (a) distortion of enzymes or (b) probable inhibitors through specific interactions with enzymes, which

could lead to changes in the reaction kinetics and substrate specificity (Rodakiewicz-Nowak et al., 2000; Miroliaei and Nemat-Gorgani, 2002). In this study, the activity of TLN has been estimated in the presence of different concentration of isopropanol, propanol, DMF and DMSO (Fig. 1). Catalytic potential of TLN decreased as the concentrations of organic solvents were increased. Results show that the enzyme was strongly inhibited by low concentration of isopropanol and n-propanol as compared to DMF or DMSO. The solvent perturbations may readily disrupt the tertiary structures of the protein, which results in the reduction of its enzymatic activity. To follow these structural changes of TLN, near UV CD and intrinsic fluores-

Fig. 2. Near-UV CD spectra of thermolysin at 50 mM of borate buffer, pH 7.2 (control) and different concentrations of n-propanol (a) and DMF (b). Enzyme concentration of 1.5 mg/ml was used in all experiments. (c and d) Fluorescence intensity spectra of thermolysin at 50 mM Tris buffer (pH 7.2) in the presence of n-propanol (c) and DMF (d). The excitation wavelength was 280 nm and the enzyme concentration was 20 ␮g/ml.

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

cence measurements were made in 0–20% (v/v) aqueous propanol, isopropanol, DMF and DMSO. These measurements showed slight changes of the enzyme conformation in the presence of low concentrations of propanol (Fig. 2). Also, the tertiary structure of the enzyme was not affected in the presence of isopropanol and DMSO (data not shown). With respect to these findings and catalytic activity studies which showed a decrease in activity of TLN without any structural changes, it is suggested that the enzyme was inhibited in the low concentrations of these solvents. With

49

increasing DMF concentration, the tryptophan fluorescence of TLN gradually decreased (Fig. 2d), suggesting a change in the tertiary structure of the protein molecule and consequently the exposure of the buried tryptophan residues to the polar solvents. Studies on the catalytic activities of thermolysin at various aqueous organic solvents are of special importance with regards to synthetic reaction, and for an insight into enzymatic catalysis in aqueous organic media in general. Reaction rates were determined for different substrate concentrations using fresh enzyme

Fig. 3. Lineweaver–Burk plots related to thermolysin inhibition at different concentrations of n-propanol (a), isopropanol (b), DMF (c), and DMSO (d). Concentration of the enzyme in all experiments is 20 ␮g/ml. 1/v is the reciprocal initial velocity in unit−1 . One unit was described as the amount of enzyme, which releases soluble fragments equivalent to 0.001 A280 nm per minute at standard condition. Standard deviations were within 4% of the experimental values. For more details see Section 2.

50

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

Table 1 Effect of different concentrations of the organic solvents on the kinetic parameters and thermostability of thermolysin Solvent

Concentration (mM)

Vm (unit)a

 (% casein) Km

 Vm /Km

1104 988 870 747 630

Ki (mM)

KI (mM)

log P

Half lifeb (min)

0.18 ± 0.02

0.25 ± 0.01

0.34

3

n-Propanol

0 26.6 65.6 106.6 133

95 87 81 68 63

± ± ± ± ±

5 4 5 3 4

0.086 0.088 0.093 0.091 0.100

± ± ± ± ±

0.003 0.004 0.005 0.004 0.005

Isopropanol

0 26 65 104 130

95 84 81 78 71

± ± ± ± ±

5 4 3 4 3

0.086 0.082 0.091 0.100 0.097

± ± ± ± ±

0.003 0.004 0.003 0.005 0.004

1104 1024 890 780 732

0.25 ± 0.03

0.45 ± 0.03

0.14

4

DMF

0 130 260 390 650

95 84 66 65 56

± ± ± ± ±

5 4 3 4 2

0.086 0.082 0.064 0.065 0.056

± ± ± ± ±

0.003 0.005 0.003 0.004 0.002

1104 1024 1031 1015 1000

7.63 ± 0.31

1.05 ± 0.07

−1.01

6

DMSO

0 140 280 420 700

95 80 75 72 67

± ± ± ± ±

5 4 3 4 2

0.086 0.077 0.070 0.071 0.067

± ± ± ± ±

0.003 0.004 0.003 0.004 0.002

1104 1038 1057 1014 1003

7.70 ± 0.25

1.84 ± 0.06

−1.34

More than 20 min

a b

One unit is the amount of enzyme which releases soluble peptide fragments equivalent to 0.001 A280 nm per minute at standard condition. Half-life for inactivation of thermolysin at 80 ◦ C in the presence of each solvent (50%, v/v).

samples each time and were carried out at different casein concentrations in the presence of different concentrations of organic solvents. It is worthy to note that casein was soluble in low concentrations of solvents tested, with subtle effects on its intrinsic fluorescence spectrum (data not shown). The kinetic parameter values of thermolysin were determined from Lineweaver–Burk plots (Fig. 3). The dependences of the Km , Vmax and Vmax /Km values of thermolysin with different concentrations of n-propanol, isopropanol, DMF, DMSO are presented in Table 1. It is observed that increasing the concentration of DMF and DMSO caused a noticeable decrease of the Km and Vmax values. On the other hand, for isopropanol and propanol, the Km values of TLN in the presence of low solvent concentrations slightly increased with significant decrease in Vmax value of the enzyme. From Table 1, it can be deduced that the decrease in Vmax /Km for different concentrations of propanol and isopropanol is due to Vmax rather than Km . Results clearly indicated differences in the sensitivity of TLN to these solvents (Table 1). The reciprocal plot related to the reversible inhibitory effect of DMF, DMSO, propanol and isopropanol showed

mixed inhibition patterns (Fig. 3), from which Ki and KI values were determined from the secondary plots as described in Section 2 (Table 1). From the relatively large values of Ki and KI for DMF and DMSO, it is apparent that the inhibitory effects of these solvents are much less significant. Thermolysin could hydrolyze the peptide bond specifically on the hydrophobic amino acids. Running through the middle of the protein, it has a large and rigid active site cleft consisting of at least four subsites (S2 , S1 , S1 , S2 ) with the main specificity pocket being the S1 subsite, which is known to prefer hydrophobic group (Morihara and Tsuzuki, 1970). It has been observed that isopropanol can occupy all four subsites on thermolysin active site (English et al., 1999). Therefore, solvents with higher hydrophobicity have better binding tendencies to the active site of TLN and hence, higher inhibitory effects. In these experiments, Ki and KI values had an inverse relationship with log P (partition coefficient between water and n-octanol) values of the examined solvents (Fig. 4). On the other hand, results showed that inhibition strength of the solvents increased with increments in hydrophobicity.

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

Fig. 4. Effect of solvent hydrophobicity on the inhibition constants; Ki () and KI (). (1) DMSO, (2) DMF, (3) isopropanol and (4) npropanol. Ki and KI values decrease with increasing in solvent log P (hydrophobicity). For more details see Section 3.

51

Fig. 5. The time course of inactivation of thermolysin at 80 ◦ C suspended in 50 mM Tris buffer, pH 7.2 (*), and 50% of each solvent; n-propanol (), isopropanol (), DMF () and DMSO (). The enzyme (0.2 mg/ml) was incubated at 80 ◦ C and at regular intervals, samples were removed and cooled on ice and their residual activities were determined. Standard deviations were within 6% of the experimental values. For more details see Section 2.

3.2. Enzyme stability and stabilization Conventional biochemical wisdom holds that enzymes become unstable and/or inactive when exposed to organic solvents. This inactivation indeed occurs (via protein denaturation) when water-miscible organic solvent are added to aqueous protein solutions. Irreversible thermoinactivation of thermolysin in the absence and presence of 50% (v/v) of each solvent (DMF, DMSO, isopropanol, n-propanol) at 80 ◦ C were investigated in this study. Results indicate that in the absence of organic solvents and at 80 ◦ C, thermolysin was not thermally inactivated. However, in the presence of propanol, isopropanol and DMF at 50% concentration, and increasing incubation periods at 80 ◦ C, the residual activity of the enzyme diminished, indicating a decrease in the thermal stability of the enzyme (Fig. 5). In the presence of DMSO, practically no inactivation occurred but n-propanol induced the most prominent change on thermolysin stability. Comparison between the stability of thermolysin and solvents log P (Table 1) showed that the thermoinactivation rate of the enzyme was enhanced concomitantly with increasing solvent log P. Different strategies have been developed to enhance the stability of enzymes (Jaenicke, 2000; F´ag´ain,

2003): protein engineering, immobilization, chemical modifications, and the use of salts or polyols (Arakawa and Timasheff, 1982; Chen and Arnold, 1991; Timasheff, 1993; Scouten et al., 1995; Rishi et al., 1998; Khajeh et al., 2001; Davis, 2003). Sorbitol, glycerol, and different mono- or disaccharides are commonly used as protein stabilizers (Lin and Timasheff, 1996; Simon et al., 2002). Obtaining stable biocatalysts in organic media is one of the main targets of biotechnology. In our experiments, the stabilizing effects of trehalose, sorbitol and glycerol on the thermal stability of thermolysin were studied in the presence of organic solvents (Fig. 6). All compounds exhibited a good stabilizing effect and were capable of retarding the thermoinactivation of thermolysin at 80 ◦ C in the presence of each solvent. n-Propanol and isopropanol are solvents with a relatively high log P compared to DMF which has a higher destabilization effect on thermolysin (Fig. 6a and b). These solvents can bind to hydrophobic pockets on thermolysin and distort tertiary structure of the enzyme. Glycerol and sorbitol may compete with n-propanol or isopropanol for binding to the hydrophobic sites of proteins. A study of intra- and intermolecular bonds within the protein conformation by Raman methods revealed

52

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53

Fig. 6. Combination effects of 50% of solvents; n-propanol (a) isopropanol (b) and DMF (c) with 20% of sorbitol (), trehalose () and glycerol () on thermostability of thermolysin at 80 ◦ C. Thermostability of the enzyme in the absence and presence of each solvent are shown by (*) and (), respectively. Standard deviations were within 6% of the experimental values. For more details see Section 2.

that the polyols (glycerol and sorbitol) have little effect on the structural organization of water, suggesting that their protective effects arise from direct interactions (specific or non-specific) with the enzyme polypeptide (Simon et al., 2002). The interaction between lysozyme and sorbitol was studied by NMR spectroscopy, and anomalous relaxation properties of Ala and Thr methyl groups were observed, indicating the modifications of local motions. Water displacement was also found in the protein structure, which reveals a complex interplay of different interactions (Wimmer et al., 1997). DMF, with relatively low log P, can distort protein structure by striping the protein molecule of essential water. In this situation, trehalose has a good stabilizing effect compared to sorbitol and glycerol (Fig. 6c), which may be correlated to the ability to increase the surface tension of water. It has been suggested that interactions between water molecules around thermolysin and between water molecules and the enzyme increase with increasing surface tension and it becomes difficult for DMF to strip essential water from protein molecules. Also, sugars (e.g. trehalose) appear to enter the lattice structure of water surrounding the protein molecules and in this way strengthen it, thereby stabilizing the protein structure (Simon et al., 2002). 4. Conclusion Our systematic studies on the catalytic measurements, stability and stabilization of thermolysin permit the following conclusions. Fluorescence and near-UV circular dichroism studies showed that no changes in the tertiary structure

of the enzyme can be seen at low concentration of npropanol, isopropanol, DMF and DMSO. Kinetic measurements showed that n-propanol and DMSO have highest and lowest inhibitory potential for thermolysin, respectively. Kinetic parameters of thermolysin such as Ki , KI and catalytic efficiency,  /K  , are correlated to log P of solvents. Vmax m In buffered 50% of each solvent, irreversible thermoinactivation results indicated that n-propanol has high destabilization effect and practically no inactivation can be seen in the presence of DMSO. Therefore, DMSO, with low inhibitory potential and denaturation capacity for thermolysin, is a suitable solvent for biotechnological applications of thermolysin in nonaqueous media. Stabilization results showed that trehalose has a significant stabilization effect on the enzyme in the presence of DMF with a relatively low log P compared to sorbitol and glycerol, whereas glycerol is a good stabilizer in the presence of n-propanol and isopropanol with high log P compared to DMF.

Acknowledgment The authors express their gratitude to the research council of Tarbiat Modares University for the financial support during the course of this project.

References Arakawa, T., Timasheff, S.N., 1982. Stabilization of protein structure by sugars. Biochemistry 21, 6536–6544.

M. Pazhang et al. / Journal of Biotechnology 127 (2006) 45–53 Chen, K., Arnold, F.H., 1991. Enzyme engineering for nonaqueous solvents: random mutagenesis to enhance activity of subtilisin E in polar organic media. Biotechnology 9, 1073–1077. Clapes, P., Torres-Luis, J., Adlercreutz, P., 1995. Enzyme peptide synthesis in low water content systems: preparative enzymatic synthesis of [Leu]-and [Met]-enkephalin derivatives. Bioinorg. Ned. Chem. 3, 244–255. Copeland, R.A., 2000. Enzymes: A Practical Introduction to Structure, Mechanism, and Data Analysis. John Wiley & Sons, Inc., New York. Davis, B.G., 2003. Chemical modification of biocatalysts. Curr. Opin. Biotech. 14, 379–386. Endo, E., 1962. Studies on protease produced by thermophilic bacteria. J. Ferment. Technol. 40, 346–353. English, A.C., Dones, S.H., Caves, L.C.D., Groom, R.C., Hubbard, R.E., 1999. Locating interaction sites on proteins: the crystal structure of thermolysin soaked in 2–100% isopropanol. Proteins 37, 628–640. English, A.C., Groom, R.C., Hubbard, R.E., 2001. Experimental and computational mapping of the binding surface of a crystalline protein. Prot. Eng. 14, 47–59. ´ F´ag´ain, O.C., 2003. Enzyme stabilization–recent experimental progress. Enzyme. Microb. Tech. 33, 137–149. Gupta, M.N., Roy, I., 2004. Enzymes in organic media. Forms, functions and applications. Eur. J. Biochem. 271, 2575–2583. Holland, D.R., Hausrath, A.C., Juers, D., Matthews, B.W., 1995. Structural analysis of zinc substitutions the active site of thermolysin. Prot. Sci. 4, 1955–1965. Jaenicke, R., 2000. Stability and stabilization of globular proteins in solution. J. Biotechnol. 79, 193–203. Khajeh, K., Ranjbar, B., Naderi-Manesh, H., Habibi, A.E., NematGorgani, M., 2001. Chemical modification of bacterial ␣amylase: changes in tertiary structure and the effect of additional calcium. Biochim. Biophys. Acta 1548, 229–237. Khalaf, N., Govardham, C.P., Lalonde, J.J., Persichetti, R.A., Wang, Y.-F., Margolin, A.L., 1996. Cross-linked enzyme crystals in organic solvents. J. Am. Chem. Soc. 118, 5494–5495. Khemelntisky, Y.L., Mozhaev, V.V., Belova, Sergeeva, M.V., Martinek, K., 1991. Denaturation capacity: a new quantitative criterion for selection of organic solvents as reaction media in biocatalysis. Eur. J. Biochem. 198, 31–41. Klibanov, A.M., 2001. Improving of enzymes by using them in organic solvents. Nature 409, 241–246. Latt, S., Holmquist, B., Valle, B.L., 1969. Thermolysin: a zinc metalloenzyme. Biochem. Biophys. Res. Commun. 37, 333–339.

53

Lin, T.Y., Timasheff, S.N., 1996. On the role of surface tension in the stabilization of globular proteins. Prot. Sci. 57, 372–381. Miroliaei, M., Nemat-Gorgani, M., 2002. Effect of organic solvent on stability and activity of two related alchol dehydrogenases: a comparative study. Int. J. Biochem. Cell B 34, 169–175. Morihara, K., 1967. The specificity of various neutral and alkaline proteins from microorganisms. Biochem. Biophys. Res. Commun. 26, 657–661. Morihara, K., Tsuzuki, H., 1970. Thermolysin: kinetic study with oligopeptides. Eur. J. Biochem. 15, 374–380. Pedersen, N.R., Halling, P.J., Pedersen, L.H., Wimmer, R., Matthiesen, R., Veitman, O.R., 2002. Efficient transesterification of sucrose catalysed by the metalloprotease thermolysin in dimethylsulfoxide. FEBS Lett. 519, 181–184. Price, N.C., Stevens, L., 1999. Fundamentals of Enzymology: The Cell and Molecular Biology of Catalytic Proteins. Oxford University Press, Inc., New York. Rishi, V., Anjum, F., Ahmad, F., Pfeil, W., 1998. Role of noncompatible osmolytes in the stabilization of proteins during heat stress. Biochem. J. l 329, 137–143. Rodakiewicz-Nowak, J., Kasture, S.M., Duek, B., Haber, J., 2000. Effect of various water-misible solvents on enzymatic activity of fungal laccase. J. Mol. Catal. B Enzym. 11, 1–11. Simon, L.M., Kotorman, M., Garab, G., Laczko, I., 2001. Structure and activity of ␣-chymotrypsin and trypsin in aqueous organic media. Biochem. Biophys. Res. Commun. 280, 1367–1371. Simon, L.M., Kotorman, M., Garab, G., Laczko, I., 2002. Effects of polyhydroxy compounds on the structure and activity of ␣chymotrypsin. Biochem. Biophys. Res. Commun. 293, 416–420. Scouten, W.H., Luong, J.H.-T., Brown, R.S., 1995. Enzyme or protein immobilization techniques for applications in biosensor design. TIBTECH 13, 178–185. Timasheff, S.N., 1993. The control of protein stability and association by weak interactions with water: How do solvents affect these processes? Annu. Rev. Biophys. Biomol. Struct. 22, 67– 97. Ulijn, R.V., Erbeldinger, M., Halling, P.J., 2000. Comparision of methods for thermolysin catalyzed peptide synthesis including a novel more active site. Biotechnol. Bioeng. 69, 633–638. Wimmer, R., Olsson, M., Petersen, M.T.N., Hatti-Kaul, R., Petersen, S.B., M¨uller, N., 1997. Towards a molecular level understanding of protein stabilization: the interaction between lysozyme and sorbitol. J. Biotechnol. 55, 85–100. Zaks, A., Klibanov, A.M., 1988. Enzymatic catalysis in nonaqueous solvents. J. Biol. Chem. 263, 3194–3201.