1. Fusarium solani pisii cutinase cloned and expressed in Escherichia coli was a gift from Corvas (Ghent, Belgium). Fungal cutinases are monomeric hydrolytic enzymes of low molecular weight (25 kDa). They display high hydrolytic activity toward both long- and short-chain fatty acid esters. The catalytic triad comprises Ser 120, His 188, and Asp 175. The optimal pH of Fusarium cutinase is slightly basic (8.5-9.5) and it loses half of its catalytic activity in 2.5 h at 80°C at pH 7.5 in phosphate buffer. A highly purified enzyme preparation (>98% as judged by sodium dodecyl sulfate electrophoresis [Pharmacia PhastSystem, France]) was used in this work. Other enzymes sources can also be used successfully e.g., commercial immobilized lipases, Novozyme 435 or Lipozyme IM20 (both Novo-Nordisk, Denmark), or Candida rugosa lipase (Sigma Chemical Co, France). However, for use in the solid-gas system, the latter enzyme must be purified and immobilized by adsorption as described in Note 2.

2. Chromosorb P Acid Washed (Prolabo, Rhone Poulenc, France), mesh 30/60 can be used for enzyme adsorption. Typically, an enzyme (50 mg) is dissolved in 20 mM phosphate buffer pH 7.5 (3 mL), and dry Chromosorb P (1.5 g) is added to the solution. After vigorous shaking for 5 min, the preparation is placed in a vacuum dessicator and dried overnight. The resulting enzyme-loaded Chromosorb is stored at 4°C over silica gel. Hydrolytic activity of the preparation is measured as described in Note 4.

3. The covalent attachment of the cutinase to a cation-exchanger resin (Ion Exchanger IV [mesh 20/50], Merck, France) is carried out using 1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluenesulfonate as the coupling agent. Previously dried resin (1 g) is added to a 5-mM carbodiimide solution prepared in 25 mM phosphate buffer, pH 8.0 (20 mL). The preparation is shaken at room temperature for 15 min and the resin is filtered and washed with MilliQ water (1 L). The resin is then packed into a chromatographic column (LKB, France, 20 cm length, 2.5 cm I.D) and percolated in a closed loop with 25 mM phosphate buffer pH 8.0 (25 mL), containing purified cutinase (10 mg). The flow rate of 5 mL/min is applied using a peristaltic pump (Minipulse 2 ,Gilson, France). The covalent binding of the protein to the resin is monitored as a decrease in absorbance at 280 nm by circulating the enzyme solution through a cuvet placed in a spectrophotometer (Uvikon 930, Kontron, France). When the absorbance is stabilized (typically a plateau is reached after passing the enzyme solution through the column for 1 h), the column is rinsed with water (500 mL) and the excess of carbodiimide is neutralized by passing a 200 mM glycine solution for 20 min. The resin is then rinsed with water (500 mL), collected and dried under vacuum overnight. The dry material (the protein loading is 5-6 mg of enzyme per gram of resin) is placed in a sealed vessel and stored at 4°C over silica gel. Hydrolytic activity of the preparation is measured as described in Note 4.

4. The hydrolytic activity of the different enzyme preparations is determined by a pH-stat method (Metrohm, Switzerland), using a 150-mM solution of triacetin in water as substrate. Experiments are carried out at 30°C and pH 8.0. A substrate solution (3.0 mL) is placed in the pH-stat vessel, and after setting the initial pH at 8.0, different amounts of solid immobilized cutinase or, in the case of free enzyme, 10 ^L of a 1-mg/mL enzyme solution are added. The acetic acid produced is titrated with a freshly prepared 25 mM NaOH solution. Typical hydrolytic activities of the cutinase preparations are as follows: Free enzyme: 880 U/mg of protein Adsorbed enzyme: 36 U/mg of protein Covalently bound enzyme: 78 U/mg of protein (1 U is defined as 1 ^mol of acetic acid liberated per min)

5. Nitrogen is preferred to air for two main reasons. First, substrates and products can be flammable, and the use of nitrogen minimizes the risks of explosion or fire. Second, solid-gas catalysis is always carried out at relatively high temperatures and the absence of oxygen greatly minimizes the inactivation of the biocatalyst through the combined deleterious effects of oxygen and temperature. Before use, all the tubing in the reactor must be passivated by flushing it with pure nitrogen.

6. For satisfactory saturation of the carrier gas, it is recommended to use flasks such as the one depicted in Fig. 2. This vessel allows good saturation at high flow rates of the carrier gas to be achieved. Typically, these flasks contain about 150 mL of pure substrate (50 mL in the first chamber and 100 mL in the second one) and can be operated at flow rates of up to 20 mL/min. The other advantage of this design is that most of the saturation occurs at the first stage. Because each compound has a different enthalpy of vaporization, it is easier to obtain the same temperature in the first oven when the energy consuming vaporization is minimal in the second stage.

7. As the equilibrium binding of water molecules to proteins is determined by water activity, it is one of the key parameters in the system. Thus, one has to be aware of the effect of alterations to the hydration level. Numerous studies in liquid systems showed that the amount of water associated with the enzyme correlates well with the retention of catalytic activity. Furthermore, enzymes in dehydrated (or partially hydrated) form are more resistant to thermal inactivation. Thus, water plays a dual (and antagonist) role. On the one hand, it helps to maintain the cata-lytically competent conformation of enzymes by giving some flexibility to the protein molecules (i.e., enhances the catalytic rate). On the other hand, water is involved in the reactions leading to thermal inactivation of the biocatalyst. By controlling the total amount of water available in the microenvironment of the enzyme, an optimal degree of hydration can be established and applied with the view of achieving good catalytic rate and thermal stability (11-14). Further information on the effects of aw on the properties of enzyme in nonaqueous systems can be found in a recent review (16).

8. One of the key parameters in the solid-gas biocatalysis is the "availability" of different chemical species in the gas phase to the enzyme. The thermodynamic parameter corresponding to each "availability" is the activity of the compound. Because of the lack of sensors for on-line measurements, a calculation method is needed in order to obtain numerical values. For each compound, determination of its thermodynamic activity requires the knowledge of its partial pressure in the

Fig. 2. Saturation flask.

1st Stage 2nd Stage

Fig. 2. Saturation flask.

reaction gas phase and access to the vapor pressure curve as a function of temperature. One can assume that a carrier gas, after bubbling through a substrate solution, is in equilibrium with the liquid phase and, therefore, the partial pressure of the substrate in the gas leaving the saturation apparatus is equal to the vapor pressure corresponding to the saturation pressure above the pure compound. In order to calculate the composition of the gas, different molar flows for each compound (carrier + substrates + water) have to be known. The carrier flow in each line is calculated using:

QV N2 normalized RT

Then, with the knowledge of the molar flow of the carrier gas, it is possible to calculate the different molar flows in the outlet of the saturation flasks using the equation that includes saturation pressure:

n Pvsat

From these, the partial pressure of each compound in the gas entering the bioreactor is determined by:

and the activity of each compound in the reactor stage is calculated as follows:

PPnsat at the temperature of the bioreactor

For the foregoing calculations, the following parameters are used: R Ideal gases constant (8.314 J/mol.K)

T Temperature (K)

Pa Absolute pressure in the system (atm)

QN2 normalized Normalized volumetric flow of carrier gas in line n (L/h at 273 K and 1 atm) QN2 Molar flow of carrier gas in line n (mol/h)

QX Molar flow of substrate X in line n (mol/h)

PpXsat Saturation pressure of compound X in line n (atm)

PpX Partial pressure of compound X in the gas entering the bioreactor (atm)

aX Activity of compound X in the bioreactor (dimensionless)

QVtotal Total volumetric flow in the bioreactor (L/h)

QVtotal Total molar flow in the bioreactor (mol/h)

To determine PpX sat, see Note 12.

Assuming that the applied conditions are far from the critical temperature and the critical pressure for each compound, and using the ideal gas law PV = nRT, one can have a good estimate of the total volumetric flow and the residence time in the bioreactor:

QVtotal = ——- (L/h at T = temperature of the bioreactor in K) (5)

As a consequence of these calculations, special care should be taken to ensure accurate control of the physical parameters, such as temperatures in all stages of the reactor, absolute pressure, and the molar flows of the carrier gas. 9. When no carrier gas is passed through the lines, the saturation flasks will be equilibrated at the temperature of the saturation oven. However after applying the flow, the temperature in the saturation flasks may decrease because of insufficient heat transfer to counteract the loss of energy resulting from vaporization of substrates. If so, the oven temperature must be corrected and monitored until the desired value in the final saturation chamber is obtained. In some cases, when the three flasks give three different stable temperatures, flow rates of the carrier gas must be recalculated. 10. For analyses, the vapor phase leaving the reactor is sampled using a 0.25-mL loop on a six-way valve (Valco) maintained at 150°C. Samples are automatically injected in the split injector of a gas chromatograph (Hewlett Packard, model 5890 A) equipped with a thermal conductivity detector (TCD) for the detection of water and a flame ionization detector (FID) for the analysis of all other products. An CP-Sil 19-CB fused-silica capillary column (25 m x 0.32 mm I.D. x 1.2 ^m film thickness, Chrompack, France) is used. The split ratio is 25:1. The injector is kept at 200°C and detectors are kept at 250°C. The column temperature is held at 50°C for 1 min, then programmed to increase to 110°C at 10°C/min and kept for 2 min at this temperature. The carrier gas is nitrogen and the flow rate is 2 mL/min. Hydrogen and air are supplied to the FID at 30 and 300 mL/min, respectively. Quantitative data are obtained after integration on a HP 3396A integrator.

11. Note that thermodynamic activity is an equilibrium parameter. Thus, for meaningful measurements, a certain equilibration time prior to the analysis is necessary for the solid phase to equilibrate with the gas phase. This time depends on the quantity of the solid phase packed in the reactor and the molar flows of different compounds. For example, in the case of enzyme hydration with water, the equilibration time can be assessed from the isotherm sorption curve for the enzyme preparation (see Note 13) and the molar flow of water through the reactor. For a transesterification, where water does not participate in the reaction, the amount of water in the gas phase entering and leaving the reactor at equilibrium should be the same. This can be easily monitored by GC and, once this is achieved, it is assumed that the system is equilibrated. When all the compounds present in the gas phase participate in the reaction, calibration curves for all of them have to be constructed as a function of partial pressure (see Note 14). Then, a correct mass balance between the inlet and the outlet gas is a good indicator that the solid and the gas phases are in equilibrium.

12. If saturation pressure curves for pure compounds are not available in the literature, determination of vapor pressure curves as a function of temperature can be carried out using a MiniVap VP apparatus (Grabner, Austria). In a typical experiment, vacuum is applied over a closed cell with an accurate temperature regulation. Once the zero absolute pressure is reached in the vessel and the temperature is at the low starting point value, a sample of pure compound (1.0 mL) is injected into the cell, using a three-way valve. Then, the temperature gradient is programmed to reach the end point selected and continuous measurements of temperature and pressure in the cell are performed. The data (Psat and T) are recorded and the curve is fitted using the exponential formula for determining saturation pressure at a given temperature in the range tested.

13. Isotherm sorption curves can be determined using an Autosorb system (Biosystemes, France). A dried preparation of supported (100 mg) or free (20 mg) enzyme is deposited into a glass container that is placed in a temperature/water activity controlled chamber. The temperature can be varied in the range from 10° to 80°C. The enzyme sample should be previously incubated at aw = 0 for at least 48 h under dried nitrogen. The water activity is programmed to reach different plateaus using increments of 0.1, and once a new plateau is reached, the corresponding water activity is maintained stable for 48 h. Every hour, the water content is determined by weighing on an electronic balance in the chamber until the weight is stabilized for all the samples loaded in the carousel. The water content for each aw is then calculated. When working with dehydrated enzymes, the isotherm sorption curve is only constructed by increasing the water activity. Isotherm sorption curves can also be determined manually by gravimetric methods and equilibration of the sample over saturated salt solutions (17).

14. The calibration procedure requires special attention. As a fixed volume is injected in the gas chromatograph and because gaseous samples are subject to variations of temperature or pressure, the calibration must be performed under conditions close to those used in the experiment. The total flow, the resulting back-pressure, and the temperature of the sampling loop must be the same under all the conditions tested. Then, different gas compositions can be created by mixing the flows from the three saturated nitrogen lines in various proportions. Calculations presented in Note 8 allow the determination of the respective partial pressures in the gas phase. If the back-pressure is significantly different in different experiments, a correction and normalization on the absolute pressure must be applied. Knowing that,

response factors can be determined from the curve AreaX = f [PpX]. The response factors for products of the reactions can be obtained by injecting liquid standards made to contain known amounts of nontransformed substrates and products.

0 0

Post a comment