Process limitations of a whole-cell P450 catalyzed reaction using a CYP153A-CPR fusion construct expressed in Escherichia coli.

Cytochrome P450s are interesting biocatalysts due to their ability to hydroxylate non-activated hydrocarbons in a selective manner. However, to date only a few P450-catalyzed processes have been implemented in industry due to the difficulty of developing economically feasible processes. In this study, we have used the CYP153A heme domain from Marinobacter aquaeolei fused to the reductase domain of CYP102A1 from Bacillus megaterium (BM3) expressed in Escherichia coli. This self-sufficient protein chimera CYP153A-CPRBM3 G307A mutant is able to selectively hydroxylate medium and long chain length fatty acids at the terminal position. ω-Hydroxylated fatty acids can be used in the field of high-end polymers and in the cosmetic and fragrance industry. Here, we have identified the limitations for implementation of a whole-cell P450-catalyzed reaction by characterizing the chosen biocatalyst as well as the reaction system. Despite a well-studied whole-cell P450 catalyst, low activity and poor stability of the artificial fusion construct are the main identified limitations to reach sufficient biocatalyst yield (mass of product/mass of biocatalyst) and space-time yield (volumetric productivity) essential for an economically feasible process. Substrate and product inhibition are also challenges that need to be addressed, and the application of solid substrate is shown to be a promising option to improve the process.

Quantification of kinetics for enzyme-catalysed reactions: implications for diffusional limitations at the 10 ml scale.

The effects of different reaction scales [100 microl reactions in 96-standard round well (SRW) plates and 10 ml reactions in 24-square well (SW) plates] have been investigated using, as a model, transketolase (TK)-catalysed reaction producing L-erythrulose. Reactions were carried out under non-shaking, shaking and at 10 ml scale stirring conditions to assess the effect of diffusional limitations. Statistical analysis confirmed the significance of the observed difference in reaction rates under given conditions. Only when the laboratory scale system (10 ml) was well mixed did the reaction rate become comparable to that in the microwells, where there is negligible diffusional limitation. These findings have important implications for the scale-up (or scale-down) of enzyme-catalysed reactions.

Choice of biocatalyst form for scalable processes.

The design of biocatalytic processes for industrial synthetic chemistry is determined in large part by the choice of isolated enzyme or whole-cell catalyst form. In the present paper, the considerations for choice are identified and some important classes of bioconversion are discussed in relation to the choice to be made. Recent developments in cell and protein engineering as well as reactor and process engineering are discussed in addition.

Measurement of strain-dependent toxicity in the indene bioconversion using multiparameter flow cytometry.

The bionconversion of indene to cis-(1S,2R)-indandiol, a potential key intermediate in the synthesis of Merck's HIV protease inhibitor, CRIXIVAN trade mark, can be achieved using Rhodococcus, Pseudomonas putida, and Escherichia coli strains. This study reports on the application of multiparameter flow cytometry for the measurement of cytoplasmic membrane integrity and membrane depolarization as indicators of toxic effects of the substrate, product, and by-products using each of these strains. Measurements of oxygen uptake rate (OUR) and optical density (OD) as indicators of metabolic activity and biomass growth, respectively, were also made. Measurements of the cytoplasmic membrane potential, cell viability, and respiratory activity provided a sensitive set of parameters to assess toxicity in the indene bioconversion and provided the basis for process improvements and strain selection. The toxic concentrations of the substrate, product, and by-products for each strain have been determined. The results show that it is possible to accumulate cis-(1S,2R)-indandiol and cis-1-amino-2-indanol up to 20 g/L without significant negative effects on cell physiology using any of the strains tested. The Gram-negative P. putida (421-5 and GM 730) and E. coli strains were more resistant to indene and the isolated chemicals of the biotransformation than the Gram-positive Rhodoccoccus I24 strain, possibly due to the presence of the outer membrane and efflux pump mechanisms. P. putida GM 730 and the E. coli TDO 123 strains responded similarly to toxic effects, and the E. coli TDO 123 strain was more resistant than the P. putida 421-5 strain. In addition to the recommendations for strain selection, the identified targets for bioprocess improvement include a combination of genetic as well as process engineering approaches.

Application of multi-parameter flow cytometry using fluorescent probes to study substrate toxicity in the indene bioconversion.

The bioconversion of indene to cis-(1S,2R) indandiol, a potential key intermediate in the synthesis of Merck's HIV protease inhibitor, CRIXIVAN trade mark, can be achieved using a Rhodococcus strain. This study using Rhodococcus I24 reports on the application of multiparameter flow cytometry for the measurement of cell physiological properties based on cytoplasmic membrane (CM) integrity and membrane depolarization as indicators of toxic effects of the substrate, indene. Quantification of intact polarized CM, intact depolarized CM and permeabilized CM of a large population of bacterial cells has been conducted using specific intracellular and membrane-binding fluorescent stains. Measurements of oxygen uptake rate (OUR) and optical density (OD) as indicators of metabolic activity and biomass growth, respectively, were also made. Indene concentrations of up to 0.25 g/L (0.037 g indene/g dry cell weight) did not significantly (<5% compared to control) affect cell light-scattering properties, intact CM, membrane polarization, respiratory activity, or biomass growth. Between this value and 1.5 g/L (0.221 g indene/g dry cell weight), the changes in intact CM, respiratory activity and biomass growth were relatively insignificant (<5% compared to control), although dissipation of the membrane potential of a significant proportion of the cell population occurred at 0.50 g/L (0.074 g indene/g dry cell weight). At 2.5 g/L (0.368 g indene/g dry cell weight) there was a significant increase in the dead cell population, accompanied by changes in the extracellular cationic concentrations and substantial decrease in respiratory activity. The primary effect of indene toxicity was the disruption of the proton motive force across the cytoplasmic membrane which drives the formation of ATP. The disruption of the proton motive force may have been due to the measured changes in proton permeability across the membrane. In addition, indene may have directly inhibited the membrane-bound enzymes related to respiratory activity. The overall consequence of this was reduced respiratory activity and biomass growth. The cell physiological properties measured via flow cytometry are important for understanding the effects of toxicity at the cellular level which neither measurements of biomass growth or indandiol formation rates can provide since both are cell averaged measurements. The technique described here can also be used as a generic tool for measuring cell membrane properties in response to toxicity of other indene-resistant strains that may be possible to use as recombinant hosts to perform the biotransformation of indene. This study has demonstrated that flow cytometry is a powerful tool for the measurement of cell physiological properties to assess solvent toxicity on whole cell biocatalysts.

Use of isolated cyclohexanone monooxygenase from recombinant Escherichia coli as a biocatalyst for Baeyer-Villiger and sulfide oxidations.

The performance, in Baeyer-Villiger and heteroatom oxidations, of a partially purified preparation of cyclohexanone monooxygenase obtained from an Escherichia coli strain in which the gene of the enzyme was cloned and overexpressed was investigated. As model reactions, the oxidations of racemic bicyclo[3.2.0]hept-2-en-6-one into two regioisomeric lactones and of methyl phenyl sulphide into the corresponding (R)-sulphoxide were used. Enzyme stability and reuse, substrate and product inhibition, product removal, and cofactor recycling were evaluated. Of the various NADPH regeneration systems tested, 2-propanol/alcohol dehydrogenase from Thermoanerobium brockii appeared the most suitable because of the low cost of the second substrate and the high regeneration rate. Concerning enzyme stability, kosmotropic salts were the only additives able to improve it (e.g., half-life from 1 day in diluted buffer to 1 week in 1 M sodium sulphate) but only under storage conditions. Instead, significant stabilization under working conditions was obtained by immobilization on Eupergit C (half-life approximately 2.5 days), a procedure that made it possible to reuse the catalyst up to 16 times with complete substrate (5 g x L(-1)) conversion at each cycle. Reuse of free enzyme was also achieved in a membrane reactor but with lower efficiency. Water-organic solvent biphasic systems, which would overcome substrate inhibition and remove from the aqueous phase, where reaction takes place, the formed product, were unsuccessful because of their destabilizing effect on cyclohexanone monooxygenase. More satisfactory was continuous substrate feeding, which shortened reaction times and, very importantly, yielded in the case of bicyclo[3.2.0]hept-2-en-6-one (10 g x L(-1)) both lactone products with high optical purity (enantiomeric excess > or = 96%), which was not the case when all of the substrate was added in a single batch.

The use of oxygen uptake rate measurements to control the supply of toxic substrate: toluene hydroxylation by Pseudomonas putida UV4.

During toluene hydroxylation, catalyzed by Pseudomonas putida UV4 one molecule of oxygen is added to the aromatic ring to produce the dihydroxylated (non-aromatic) ring structure, toluene cis-glycol. Toluene, which is toxic to the cells at aqueous phase concentration above ( approximately 2.4 mmol), is fed to the reactor. A feed-back control system based on oxygen uptake rate measurements was used to control the feed rate, and thus maintain the aqueous phase toluene concentration in the desired range for zero order kinetics.

Large scale production of cyclohexanone monooxygenase from Escherichia coli TOP10 pQR239.

The cyclohexanone monooxygenase (CHMO) from Acinetobacter calcoaceticus NCIMB 9871 has been cloned into Escherichia coli in an L-arabinose inducible vector. The recombinant E. coli containing the L-arabinose inducible CHMO was grown at 1.5 litres under controlled conditions to determine the parameters for growth and induction. It was found that induction with 0.1% (w/v) L-arabinose at late logarithmic phase of growth and growth for a further 2.5 to 3 h gave the optimal CHMO titre ( approximately 3500 U.l(-1,) 630 U. g dry cell weight(-1)). High dissolved oxygen concentrations were shown to be deleterious to the CHMO titre. This influenced the strategy for growth and induction, and was optimal when the oxygen uptake rate was maximized but the dissolved oxygen concentration was zero. Finally, a 300 litre scale fermentation was carried out giving a total CHMO titre of >8 x 10(5) U.

Advances in enzyme technology--UK contributions.

Enzyme technology has been a recognised part of bioprocess engineering since its inception in the 1950s and 1960s. In this article the early history of enzyme technology is discussed and the subsequent developments in enzyme isolation, enzyme modification and process technology are described. These creative developments have put enzyme technology in a position of huge potential to contribute to environmentally compatible and cost effective means of industrial chemical synthesis. Recent developments in protein modification to produce designer enzymes are leading a new wave of enzyme application.

Alkaline biocatalysis for the direct synthesis of N-acetyl-D-neuraminic acid (Neu5Ac) from N-acetyl-D-glucosamine (GlcNAc).

Integration between the alkaline epimerization of N-acetyl-D-glucosamine (GlcNAc) to N-Acetyl-D-mannosamine (ManNAc) and the N-acetyl-D-neuraminic acid (Neu5Ac) aldolase-catalyzed biotransformation has been assessed experimentally. GlcNAc epimerization took place above pH 9.0, and the initial rate of ManNAc formation increased exponentially to 10.37 mmol/L per hour at pH 12. However, above this pH, severe degradation of pyruvate occurred. A value of 31.3% molar conversion on Pyr was achieved in an integrated biotransformation. The "pseudo"-steady state at the end of the reaction was comparable to the equilibrium achieved with a combination of an epimerase and aldolase enzymes. The integrated reaction proved feasible, but at the expense of pyruvate and Neu5Ac aldolase degradation.

Application of in situ product-removal techniques to biocatalytic processes.

Biocatalytic processes for the manufacture of small, highly functionalized molecules frequently have limited productivity. A common reason for this is the presence of the reaction products that can cause inhibitory or toxic effects (making poor use of the enzyme) or promote unfavourable equilibria (giving low conversions). In each case, the product needs to be removed as soon as it is formed in order to overcome these constraints and hence increase the productivity of the biocatalytic process. Here, we review the need for in situ product removal and the process research required for its implementation.

Boron based separations for in situ recovery of L-erythrulose from transketolase-catalyzed condensation.

In this article we report on the application of in situ product removal (ISPR) (the concurrent recovery of a product during the product formation process) as a means of improving the productivity of bioconversions. The Escherichia coli transketolase-catalyzed condensation of glycolaldehyde with beta-hydroxypyruvate to yield L-erythrulose (and carbon dioxide) was chosen as a model system. Those ISPR methods based on phenylboronate-diol interactions showed greatest potential for use as a selective means of removing L-erythrulose from the reaction medium. Soluble, insoluble, and immobilized boronates were investigated. Concentrations of free phenylboric acid of 100 mM and above were toxic to transketolase, thus rendering the use of these methods unsuitable for ISPR. However, one of the immobilized phenylboronate resins (Affi-Gel 601) was not toxic to the enzyme, although significant levels of nonspecific binding of both substrates were observed. When ISPR was performed on the model reaction using this resin with substrate feeding, it proceeded to completion.

Choice of microbial host for the naphthalene dioxygenase bioconversion.

The use of whole cell biotransformations for single and multistep enzyme conversions is gaining widespread application. In this study the naphthalene dioxygenase nah A gene was transferred into Pseudomonas aeruginosa PAC 1R, Escherichia coli JM107 and Pseudomonas putida PpG 277. The effect of ethanol on these genetically engineered Gram-negative bacteria was studied by measurement of enzyme activity, stability and cell integrity. Ethanol has been used in biotransformations as a co-substrate carbon source for co-factor recycling and as a co-solvent increasing dissolved substrate and product levels. Ethanol increased the dissolved substrate (naphthalene) concentration slightly and dissolved product ((+)-cis-(1R,2S)-dihydroxy-1,2-dihydronaphthalene) by approximately 30% at 4% (w/v) ethanol. Both P. aeruginosa PAC 1R and P. putida PpG 277 showed decreased activity with increasing ethanol concentration whilst E. coli enzyme activity increased with increasing ethanol concentration being comparable to that when glucose was used as a carbon source. This project highlighted the many factors involved in the selection of microbial hosts for whole cell biotransformation processes.

Enzyme-catalysed carbon-carbon bond formation: large-scale production of Escherichia coli transketolase.

Escherichia coli strain JM107/pQR700 possesses the vector pBGS18, a high copy number plasmid carrying kanamycin resistance, into which a 4.4 kb fragment containing the transketolase gene had been cloned. The bacterium was grown at 20 and 1000 1 scale for the production of transketolase. The specific growth rate was maintained at 0.15 h-1 until the bacterial concentration reached 20 g dry wt per litre at which point the culture was harvested. The clarified cell extract obtained after disruption of the bacteria in a high-pressure homogeniser contained about 230 U ml-1 of the enzyme, which represented about 40% of the total protein released. No further purification was done at large scale as the clarified cell extract could be used satisfactorily for biotransformations.

In situ product removal as a tool for bioprocessing.

In situ product removal (ISPR) is the fast removal of product from a producing cell thereby preventing its subsequent interference with cellular or medium components. Over the past 10 years ISPR techniques have developed substantially and its feasibility (with improvements in yield or productivity) for several processes demonstrated. Assessment of progress, however, compared to the potential benefits inherent in the ISPR approach to bioprocessing reveals that these are far from being exploited fully. Here we indicate future directions including application of the ISPR approach to a wider range of product groups and the development of novel, more specific ISPR methodologies, applicable under sterile conditions in the immediate vicinity of the producing cells. General guidelines for adaptation of an appropriate ISPR approach for a given product are also provided.

Lewis cell studies to determine reactor design data for two-liquid-phase bacterial and enzymic reactions.

Substrate transfer rates from organic to aqueous phases were measured in the presence and absence of biocatalyst in the reaction medium, using modified Lewis cells. These measurements, in combination with intrinsic aqueous phase biocatalytic reaction kinetics, were used to confirm that benzyl acetate hydrolysis by pig liver esterase and toluene oxidation by a strain of Pseudomonas putida occur uniformly throughout the bulk of the aqueous phase. Such data may be used to provide a basis for two-liquid-phase biocatalytic reactor design.