Packed-Bed Bioreactors Under the Influence of Chemical Modulators

This report considers the behaviour of packed-bed immobilized enzyme reactors operating in the presence of substrate and/or product sequestrators. In some cases, enzyme inhibition by the reaction product and presence of chemical modulators are also considered. For each case, an appropriate analytical model is developed. Using numerical simulations, it is shown that reactor performance is impaired by substrate sequestration. This effect can be partially reversed when competitive sequestration by product or modulator is operational.

In addition, a comparison is made between some of the predicted characteristics of the reactor and experimental data. It reveals the capabilities and limitations of the models employed.     

Mathematical analysis of two-phase mass transfer in a batch reactor for the chemical transformation of a steroid

A reactor is described for the conversion of the slightly water-soluble steroid testosterone (T) to 4-androstene-3, 17-dione (4-AD) by enzyme in the presence of excess cofactor. Since the enzyme is subject to substrate inhibition, reaction rates are strong functions of aqueous substrate concentration. High concentrations of the substrate, testosterone, per unit reactor volume are maintained within poly(dimethylsiloxane) beads that are suspended in the aqueous enzyme solution. Mass transfer (controlled by bead size, polymer to water volume ratio, enzyme loading) is used to control the degree and rate of conversion. The reactor dynamics are predicted over a wide range of reaction conditions. The product steroid is recovered in the polymeric beads from the enzyme solution.     

Application of ultrafiltration for enzyme retention during continuous enzymatic reaction

In most enzymatic reactions, batch or continuous, separation of the enzyme for reuse is difficult if not impossible. A process will be presented in which an Ultrafiltration membrane serves to separate the reaction products from the enzyme and the substrate. In this manner the enzyme may be retained and re-used. Furthermore, under these conditions, the enzyme need only be present in catalytic amounts regardless of the amount of product produced. Under proper operating conditions and proper ultrafiltration membrane selection, a pure solution of α-amylase from Bacillus subtilis may be retained with no loss in enzyme activity over a test period of 30 hr after steadystate has been achieved. In the presence of substrate, the membrane support and ultrafiltration cell serve as the reaction vessel for the hydrolysis of starch. The substrate is continuously pumped into the cell under constant ultrafiltration pressure. The di-, oligo-, and polysaccharides formed from the enzyme reaction then either pass through the membrane as products or are retained. The molecular weight distribution of the products is dependent on the nominal molecular weight cut-off of the membrane, absolute ultrafiltration pressure, enzyme-to-substrate ratio, temperature, and residence time of the substrate in the reactor. In addition to the partial hydrolysis of starch by α-amylase, some preliminary findings on the complete hydrolysis of starch by glucoamylase will also be presented. In these latter studies, the substrate may be completely hydrolyzed to glucose units.     

Three-dimensional numerical approach to investigate the substrate transport and conversion in an immobilized enzyme reactor

This numerical study evaluates the momentum and mass transfer in an immobilized enzyme reactor. The simulation is based on the solution of the three-dimensional Navier-Stokes equation and a scalar transport equation with a sink term for the transport and the conversion of substrate to product. The reactor consists of a container filled with 20 spherical enzyme carriers. Each of these carriers is covered with an active enzyme layer where the conversion takes place. To account for the biochemical activity, the sink term in the scalar transport equation is represented by a standard Michaelis-Menten approach. The simulation gives detailed information of the local substrate and product concentrations with respect to external and internal transport limitations. A major focus is set on the influence of the substrate transport velocity on the catalytic process. For reactor performance analysis the overall and the local transport processes are described by a complete set of dimensionless variables. The interaction between substrate concentration, velocity, and efficiency of the process can be studied with the help of these variables. The effect of different substrate inflow concentrations on the process can be seen in relation to velocity variations. The flow field characterization of the system makes it possible to understand fluid mechanical properties and its importance to transport processes. The distribution of fluid motion through the void volume has different properties in different parts of the reactor. This phenomenon has strong effects on the arrangement of significantly different mass transport areas as well as on process effectiveness. With the given data it is also possible to detect zones of high, low, and latent enzymatic activity and to determine whether the conversion is limited due to mass transfer or reaction resistances.     

Mathematical modeling of diffusion and reaction in the hydrolysis of vegetable protein in an immobilized enzyme recycle reactor

Experimental investigation is by far the most effective approach for studying the behavior of physical systems. However, an enzymatic solubilization of vegetable protein is a complex combination of intrinsic problems, of which many are not easily adaptable to experimental investigation. Experimental designs to study enzyme vegetable protein reactions yield data which describe the extramembraneous activity of the immobilized enzyme. In a continuous recycle immobilized enzyme reactor, the microenvironment concentration of the substrate or product in the membrane phase, or the concentrations along the reactor axial length in the bulk phase are not discernible to the experimenter. However, the knowledge of such concentration profiles is important in weighing the significance of such factors as intermembrane diffusion, enzyme loading, wet membrane size, and the mode of operation of the reactor. The simulation of mathematical models, which describe the physical system within the constraints imposed, yields information which is vital to the understanding of the process occurring in the reactor. The kinetics and diffusion of an immobilized thermophilic Penicillium duponti enzyme at pH 3.4-3.7 and 50 degrees C was modeled mathematically. The kinetic parameters were evaluated by fitting a model to experimental data using nonlinear regression analysis. Simulation profiles of the effects of reactor geometry, substrate concentration, membrane thickness, and enzyme leading on the hydrolysis rate are presented. From the profiles generated by the mathematical model, the best operational reactor strategy is recommended.     

A novel reactor concept for the enzymatic reduction of poorly soluble ketones

Reductions of poorly soluble ketones often suffer from low total turnover numbers conferring to the coenzyme and large volumes which are needed for the conversion. The novel emulsion membrane reactor overcomes these limitations. From an emulsion consisting of an organic substrate and an aqueous buffer phase, the aqueous phase is separated selectively by using a hydrophilic ultrafiltration membrane and fed to a subsequent enzyme membrane reactor. The product outflow is recirculated to the emulsion stirred vessel and, due to the partition coefficients, the aqueous phase is recharged with substrate while the product is extracted. This new reactor concept will be compared to the classical enzyme membrane reactor. The latter was operated under the same conditions over a period of 4 months at a space-time yield of 21.2 g l⁻¹ day⁻¹. As a model system the enantioselective reduction of 2-octanone to (S)-2-octanol (ee > 99.5%) is used, carried out by a carbonyl reductase from Candida parapsilosis. NADH is regenerated by formate dehydrogenase from Candida boidinii. In comparison to the classical enzyme membrane reactor the total turnover number could be increased by a factor 9 using the novel emulsion membrane reactor.     

A silicone polymer as a Steroid Reservoir for enzyme-catalyzed Steroid reactions

Since steroids are only slightly soluble in the aqueous solutions in which enzymatic reactions take place, it is difficult to obtain high effective concentrations per unit reactor volume when enzymes are used to catalyze steroid reactions. In order to obtain high effective concentrations in the present work, we have used small particles of a hydrophobic polymer, poly (dimethyl siloxane), as a reservoir for the steroid substrate and product. The activity of a bacterial hydroxysteroid dehydrogenase in a buffer solution declines much more slowly in the presence of those polymer particles than in the presence of a comparable amount of butyl acetate or ethyl acetate, the organic solvents used as steroid reservoirs in previous work with steroid transforming enzymes. When another substrate of the hydroxysteroid dehydrogenase is loaded into the polymer particles and the particles are suspended in an aqueous solution containing the enzyme and its cofactor, more product is formed that when a similar solution is emulsified with butyl acetate.     

The role of enzyme sequestration in the regulation of the adenylate cyclase of Dictyostelium discoideum

Although the adenylate cyclase of Dictyostelium discoideum cannot be activated by its cAMP agonist in vitro, its in vivo activation can be demonstrated by rapidly breaking and assaying the cells, over 10-fold higher activity being observed for stimulated cells than for basal cells. We report here that when basal cells are broken in the presence of labeled ATP and then rapidly assayed, they display 8-fold more adenylate cyclase activity than cells broken in the presence of unlabeled ATP. This suggests that a significant amount of the enzyme in extracts of basal cells is sequestered within vesicles that can be loaded with substrate at the time of cell lysis, but then rapidly seal. In contrast to the results obtained with basal cells, when cells activated in vivo are broken in the presence of labeled ATP, there is less than 2-fold increase in adenylate cyclase activity. Thus, a much smaller percentage of the observed adenylate cyclase activity of stimulated cells appears to be due to sequestered enzyme than of basal cells. Two models are discussed that account for these observations. One model envisions that roughly equal populations of sequestered and nonsequestered enzyme are produced upon breakage of both basal and activated cells, but that sequestered enzyme in basal extracts becomes uniquely activated in vitro. The other model proposes that the differences in observed activity are due directly to differences in sequestration. According to this latter model, nearly all of the -fold activation previously observed for the D. discoideum adenylate cyclase can be accounted for by a change in sequestration of the enzyme rather than by an intrinsic alteration in the enzyme per se. It therefore suggests a novel mode of regulation whereby an enzyme may be packaged within vesicles and its activity controlled by modulating the permeability of the vesicles to its substrate or effectors.     

Optimum design of a series of CSTRs performing reversible Michaelis-Menten kinetics: a rigorous mathematical study

The optimum design of a given number of CSTRs in series performing reversible Michaelis-Menten kinetics in the liquid phase assuming constant activity of the enzyme is studied. In this study, the presence of product in the feed stream to the first reactor, as well as the effect of the product intermediate concentrations in the downstream reactors on the reaction rate are investigated. For a given number of N CSTRs required to perform a certain degree of substrate conversion and under steady state operation and constant volumetric flow rate, the reactor optimization problem is posed as a constrained nonlinear programming problem (NLP). The reactor optimization is based on the minimum overall residence time (volume) of N reactors in series. When all the reactors in series operate isothermally, the constrained NLP is solved as an unconstrained NLP. And an analytical expression for the optimum overall residence time is obtained. Also, the necessary and sufficient conditions for the minimum overall residence time of N CSTRs are derived analytically. In the presence of product in the feed stream, the reversible Michaelis-Menten kinetics shows competitive product inhibition. And this is, because of the increase in the apparent rate constant K^' _m that results in a reduction of the overall reaction rate. The optimum total residence time is found to increase as the ratio (‚₀) of product to substrate concentrations in the feed stream increases. The isomerization of glucose to fructose, which follows a reversible Michaelis-Menten kinetics, is chosen as a model for the numerical examples.     

Chemo-enzymatic epoxidation-process options for improving biocatalytic productivity

The reactor choice is crucial when designing a process where inactivation of the biocatalyst is a problem. The main bottleneck for the chemo-enzymatic epoxidation has been found to be enzyme inactivation by the hydrogen peroxide, H(2) O(2) , substrate. In the work reported here, the effect of reaction parameters on the reaction performance have been investigated and used to establish suitable operating strategies to minimize the inactivation of the enzyme, using rapeseed methyl ester (RME) as a substrate in a solvent-free system. The use of a controlled fed-batch reactor for maintaining H(2) O(2) concentration at 1.5 M resulted in increased productivity, up to 76 grams of product per gram of biocatalyst with higher retention of enzyme activity. Further investigation included a multistage design that separated the enzymatic reaction and the saturation of the RME substrate with H(2) O(2) into different vessels. This setup showed that the reaction rate as well as enzyme inactivation is strongly dependent on the H(2) O(2) concentration. A 20-fold improvement in enzymatic efficiency is required for reaching an economically feasible process. This will require a combination of enzyme modification and careful process design.     

An immobilized cell reactor with simultaneous product separation. I. Reactor design and analysis

A four-phase reactor-separator (gas, liquid, solid, and immobilized catalyst) is proposed for fermentations characterized by a volatile product and nonvolatile substrate.In this reactor, the biological catalyst is immobilized onto a solid column packing and contacted by the liquid containing the substrate.A gas phase is also moved through the column to strip the volatile product into the gas phase. The Immobilized Cell Reactor-Separator (ICRS) consists of two basic gas-liquid flow sections: a cocurrent "enricher" followed by a countercurrent-"stripper".In this article, an equilibrium stage model of the reactor is developed to determine the feasibility and important operational variables of such a reactor-separator. The ICRS concept is applied to the ethanol from whey lactose fermentation using some preliminary immobilized cell reactor performance data. A mathematical model for a steady-state population based on an adsorbed monolayer of cells is also developed for the reactor. The ICRS model demonstrated that the ICRS should give a significant increase in reactor productivity as compared to an identically sized Immobilized Cell Reactor (ICR) with no separation. The gas-phase separation of the product also allows fermentation of high inlet substrate concentrations. The model is used to determine the effects of reactor parameters on ICRS performance including temperature, pressure, gas flow rates, inlet substrate concentration, and degree of microbial product inhibition.     

Evaluation of D-amino acid oxidase from Rhodotorula gracilis for the production of alpha-keto acids: a reactor system

The study reports on the development of a bioreactor for the production of alpha-keto acids from D,L- or D-amino acids using Rhodotorula gracilis D-amino acid oxidase. D-Amino acid oxidase was co-immobilized with catalase on Affi-Gel 10 matrix, and the reactor was operated as a continuous-stirred tank reactor (CSTR) or stirred tank with medium recycling conditions. The optimum substrate concentration and quantity of biocatalyst were determined (5 mM and 1.2 mg/L, respectively). Under optimum operating conditions, product formation was linearly related to both substrate and enzyme concentration, showing the system to be highly flexible. Under these conditions, in a stirred tank, over 90% conversion was achieved in 30 min with a maximum production of 0.23 g of pyruvic acid/day/enzyme units. Product was recovered by ion exchange chromatography. The operational stability of the reactor was high (up to 9.5 h of operation without loss of activity) and the inactivation half-life was not reached even after 18 h or 36 bioconversion cycles. This represents the first case of a reactor developed successfully with a D-amino acid oxidase. (c) 1994 John Wiley & Sons, Inc.     

Influence of substrate and product diffusion on the heterogeneous kinetics of enzymic reversible reactions

When a reversible reaction is catalyzed by a surface bound enzyme, the diffusion of both substrate and product can considerably modify the kinetic properties of the reaction. According to this theoretical analysis, the enzyme activity is decreased due to the presence of substrate and product concentration gradients in the enzyme microenvironment, and the relative kinetic importance of the two diffusion steps mainly depend on the value of a dimensionless criterion inversely proportional to the equilibrium constant. Moreover diffusional effects increase with increasing bound enzyme activity, but decrease with increasing substrate and product concentration. Analytical expressions are established for the limiting values of substrate and product concentrations in the enzyme microenvironment, as well as for the increase in half-maximal-activity substrate concentration in the presence of substrate and product diffusional limitations.     

Performance of a β-d-galactosidase hollow fibre reactor

β-d-Galactosidase was immobilized in a hollow fibre ultrafiltration module. The hydrolysis of 2-nitrophenyl β-d-galactopyranoside (ONPG) was significantly affected by enzyme loading, flow rate and substrate concentration. Pretreatment of hollow fibres with a protein was necessary to minimize enzyme inactivation. Residence time distribution studies indicated that the product of the reaction (ONP) was significantly adsorbed by the fibres, which resulted in the reactor taking 10–30 h to achieve steady-state. An equation based on Michaelis-Menten kinetics and a plug-flow model adequately described the performance of the reactor with regard to operating variables, even though some diffusion effects were observed.     

Optimal temperature policy for immobilized enzyme packed bed reactor performing reversible Michaelis-Menten kinetics using the disjoint policy.

The optimal temperature policy that maximizes the time-averaged productivity of a continuous immobilized enzyme packed bed reactor is determined. This optimization study takes into consideration the enzyme thermal deactivation with substrate protection during the reactor operation. The general case of reversible Michaelis-Menten kinetics under constant reactor feed flow rate is assumed. The corresponding nonlinear optimization problem is solved using the calculus of variations by applying the disjoint policy. This policy reduces the optimization problem into a differential-algebraic system, DAE. This DAE system defines completely the optimal temperature-time profiles. These profiles depend on the kinetic parameters, feed substrate concentration, operating period, and the residence time and are characterized by increasing form with time. Also, general analytical expressions for the slopes of the temperature and residual enzyme activity profiles are derived. An efficient solution algorithm is developed to solve the DAE system, which results into a one-dimensional optimization problem with simple bounds on the initial feed temperature. The enzymatic isomerization of glucose into fructose is selected as a case study. The computed productivities are very close to that obtained by numerical nonlinear optimization with simpler problem to solve. Moreover, the computed conversion profiles are almost constant over 90% of the operating periods, thus producing a homogeneous product.     

Kinetics studies in a continuous steady state hollow fiber membrane enzyme reactor

Porous hollow cellulose fibers have been used to separate a nonflowing enzyme solution of alkaline phosphatase from a continuous flow of substrate. The porosity of the hollow fiber membrane allows the substrate and product to diffuse freely through the membrane while restricting the permeation of the enzyme. The resulting “immobilized” enzyme system has been shown to behave as a continuous reactor—converting p-nitrophenylphosphate to p-nitrophenol. By varying the concentrations, flow rate, etc., either diffusion or enzyme kinetics can be studied. The continual influx of product and removal of substrate at steady state allows the study of kinetics of relatively short half-life enzymes and unstable systems.     

Kinetic mechanism of phosphoenolpyruvate carboxykinase (GTP) from rat liver cytosol. Product inhibition, isotope exchange at equilibrium, and partial reactions

Initial velocity studies of rat liver cytosolic P-enolpyruvate carboxykinase in the direction of P-enolpyruvate formation gave intersecting double reciprocal plots indicating that the reaction conforms to a sequential reaction pathway. A complete product inhibition study with MnGDP-, P-enolpyruvate, and HCO3- as product inhibitors indicated that all patterns were noncompetitive. Isotope exchange at equilibrium with exchange between the substrate/product pairs GTP/GDP oxalacetate/HCO3-, and oxalacetate/P-enolpyruvate while varying the concentration of substrate/product pairs in fixed constant ratio gave no complete inhibitory patterns as the concentration of the constant ratio pairs approached saturation. The exchange rates between the substrate/product pairs differed by a factor of 40 when compared under the same assay conditions. These results were interpreted in terms of a random reaction mechanism in which true dead-end complexes do not form and in which the rate-limiting step is not the interconversion of the ternary quarternary central complexes. In addition to the formation of P-enolpyruvate from oxalacetate and MnGTP2-, the enzyme catalyzes the decarboxylation of oxalacetate to pyruvate in the absence of MnGTP2-. This reaction occurs only slowly in the absence of GDP and most rapidly in the presence of MnGDP-. When only MnGTP2- and oxalacetate are present, no pyruvate is formed, and oxalacetate is converted stoichiometrically to P-enolpyruvate. The enzyme also catalyzes the exchange of [14C]GDP into GTP in the absence of P-enolpyruvate. This exchange is stimulated by the presence of HCO3-. When enzyme is incubated with MnGTP2- in the presence or absence of HCO3-, there is no hydrolysis to form GDP and P1. The two partial reactions, namely the exchange of [14C]GDP with the E.HCO3.MnGTP or E.MnGTP complex and the formation of pyruvate from the E.oxalacetate.MnGDP complex provide pathways by which the expected dead-end complexes can be converted to enzyme forms which can return to the catalytic or exchange sequence.     

Product catalyzes the deamidation of D145N dehalogenase to produce the wild-type enzyme

Aspartate 145 plays an essential role in the active site of 4-chlorobenzoyl-CoA dehalogenase, forming a transient covalent link at the 4-position of the benzoate during the conversion of the substrate to 4-hydroxybenzoyl-CoA. Replacement of Asp 145 by residues such as alanine or serine results in total inactivation, and stable complexes can be formed with either substrate or product. The Raman spectroscopic characterization of some of the latter is described in the preceding publication (Dong et al.). The present work investigates complexes formed by D145N dehalogenase and substrate or product. Time-resolved absorption and Raman difference spectroscopic data show that these systems evolve rapidly with time. For the substrate complex, initially the absorption and Raman spectra show the signatures of the substrate bound in the active site of the asparagine 145 form of the enzyme but these signatures are accompanied by those for the ionized product. After several minutes these signatures disappear to be replaced with those closely resembling the un-ionized product in the active site of wild-type dehalogenase. Similarly, for the product complex, the absorption and Raman spectra initially show evidence for ionized product in the active site of D145N, but these are rapidly replaced by signatures closely resembling the un-ionized product bound to wild-type enzyme. It is proposed that product bound to the active site of asparagine 145 dehalogenase catalyzes the deamidation of the asparagine side chain to produce the wild-type aspartate 145. For the complexes involving substrate, the asparagine 145 enzyme population contains a small amount of the WT enzyme, formed by spontaneous deamidation, that produces product. In turn, these product molecules catalyze the deamidation of Asn 145 in the major enzyme population. Thus, conversions of substrate to product and of D145N to D145D dehalogenase go on simultaneously. The spontaneous deamidation of asparagine 145 has been characterized by allowing the enzyme to stand at RT in Hepes buffer at pH 7.5. Under these conditions deamidation occurs with a rate constant of 0.0024 h-1. The rate of product-catalyzed deamidation in Hepes buffer at 22 degrees C was measured by stopped-flow kinetics to be 0.024 s-1, 36000 times faster than the spontaneous process. A feature near 1570 cm-1 could be observed in the early Raman spectra of both substrate and product-enzyme complexes. This band is not associated with either substrate or product and is tentatively assigned to an ester-like species formed by the attack of the product's 4-O- group on the carbonyl of asparagine's side chain and the subsequent release of ammonia. A reaction scheme is proposed, incorporating these observations.     

Stereoselective synthesis of (S)-(+)-Naproxen catalyzed by carboxyl esterase in a multicompartment electrolyzer

The stereoselective hydrolysis of racemic 2-substituted propionates, catalyzed by carboxyl esterase, provides a cost-competitive route to produce the optically pure, anti-inflammatory drug Naproxen. In the present work, we describe the application of the multicompartment electrolyzer reactor (ME) for the stereoselective hydrolysis of a racemic Naproxen ester, (R,S)-ethoxyethyl-[2-(6-methoxy-2-naphtyl)]propionate, catalyzed by a carboxyl esterase.The enzyme was trapped in a reactor chamber, delimited by two isoelectric membranes encompassing the pI value of the enzyme, together with the neutral substrate. After 90 min, a conversion of 45% was obtained with an enantiomeric excess of 84%. The reaction product, (S)-(+)-Naproxen, was electrophoretically removed in continuous from the reaction chamber and collected in a contiguous, more acidic chamber, separated from the enzyme and from the unreacted substrate. Moreover, at the end of the reaction, it was possible to recover the enzyme from the reactor and use it again.     

Modeling and simulation of a pressure-swing reactor for the conversion of poorly soluble substrate by immobilized enzyme: the case of d-hydantoinase reaction

An immobilized enzyme reactor system for converting poorly soluble substrate is proposed. In this stirred batch reactor, the solid substrate and immobilized enzyme suspensions are separated by a microporous filter. The advantage of separating the solid substrate from immobilized enzyme is that the fouling and breakage of the immobilized enzyme usually encountered in the stirred tank reactor can be prevented. Pressure swing can be applied to enhance the mass transfer between the two compartments. The hydrolytic reaction converting the poorly soluble substrate p-hydroxyphenylhydantoin (pHPH) into soluble N-carbamoyl-p-d-hydroxyphenylglycine (CpHPG) by immobilized d-hydantoinase is carried out in this reactor. The performance of this pressure-swing reactor is studied by simulation using a simple kinetic model. The pressure-swing operation increases the overall production rate significantly. The pressure swing also makes the reactor perform better for converting the solid substrate at higher concentration.