Oxygen transfer and mycelial growth in a tubular loop fermentor

A 22 m long. 20 liter tubular loop fermentor (TLF) has been tested for oxygen transfer characteristics and as a reactor for mycelial growth. Model calculations show that the flow pressure drop has an important influence on the axial oxygen profiles. A design model that accounts for this influence is presented. Using the model, K_L a values are calculated from the results of sulfite oxidation experiments. These are correlated with power consumption and aeration rates. The K_L a dependence on aeration rate was found to be less than found with tank reactors. The growth kinetics of three metabolite-producing mycelial organisms in the TLF are presented: a Streptomyces, a Fusarium, and a Acrophialophora. In order to determine the influence of reactor type on the growth and product formation, these cultures have been grown in tanks and shake flasks. The antibiotic, product spectrum of Streptomyces is compared on the basis of inhibition tests and it is shown that the distribution of products is reactor dependent. The Fusarium culture produced a previously unknown metabolite, whose concentration in the loop fermentor was four times higher than in a shake flask. The Acrophialophora culture grew twice as fast in the loop fermentor, but produced essentially none of the specific product. Power Consumptions of up to 8 kW/m³ in the tubular fermentor did not appear to harm the mycelia.     

Comparison of two methods for designing calorimeters using stirred tank reactors

Calorimetry is a robust method for online monitoring and controlling bioprocesses in stirred tank reactors. Up to now, reactor calorimeters have not been optimally constructed for pilot scale applications. Thus, the objective of this paper is to compare two different ways for designing reactor calorimeters and validate them. The “heat capacity” method based on the mass flow of the cooling liquid in the jacket was compared with the “heat transfer” method based on the heat transfer coefficient continuously measured in the cultivation of Escherichia coli VH33 in a 50 L stirred tank reactor. It was found that the values of the “heat transfer” method agreed very well with the calculated values from the oxygen consumption. By contrast, the curve of the “heat capacity” method deviated from that of the oxygen consumption calculated with the oxycaloric equivalent. In conclusion, the “heat transfer” method has been proven to have a higher degree of validity than the “heat capacity” method. Thus, it is a better and more robust means to measure heat generation of fermentations in stirred tank bioreactors on a pilot scale. Biotechnol. Bioeng. 2013; 110: 180–190. © 2012 Wiley Periodicals, Inc.     

New technique for kla measurement between gas and water in aerated solvent-in-water dispersions

Summary A new technique is presented to determine gas-to-water overall volumetric mass transfer coefficients (k _l a) in a stirred-tank reactor containing solvent-in-water dispersions. The compound to be transferred from the gas to the water was toluene; the water-immiscible organic solvent was FC40, a perfluorocarbon. The k _l a was determined in steady-state conditions in the absence of biological consumption. Toluene removal was achieved by passing a continuous flow of toluene-free water through the reactor. When solvent was present it was separated from the water at the reactor outlet by means of a small settler and recycled back to the vessel. The k _l a was found to increase with the FC40 volume fraction. An enhancement of 1.9 times on an aqueous-phase-volume basis was found at 15 % (v/v) FC40.     

Ambruticin S production in air-lift and stirred-tank bioreactors

Summary Using the method of equi-inocular synchronized comparative fermentation (EISCF) the cultivation of Sorangium cellulosum So ce 10 and production of the polyketide antibiotic ambruticin S was compared in stirred-tank and air-lift reactors of different geometry. This method requires that inocula originate from the same pre-culture and cultivation parameters are synchronized to similar values. Similar ambruticin yields were obtained from both reactor systems provided that the concentration of dissolved oxygen (DO) was maintained above a certain value (ca. 40%). For cultivation of S. cellulosum it is the DO level rather than the oxygen transfer rate which presents the proper criterion for scale-up and comparative reactor studies. Offprint requests to: W.-D. Deckwer     

A Desktop Apparatus for Studying Interactions Between Microorganisms and Small-Scale Fluid Motion

Low levels of kinetic energy dissipation were successfully generated in a reactor using two submersible speakers. A software programme controlled the amplitude and frequency of the signal fed to each speaker and achieved good repeatability of flow conditions. The flow reactor had a near isotropic flow regime with a low mean flow, values were calculated from particle image velocimetry measurements. The flow characteristics compared well with grid turbulence reactors, though as no moving parts are present in this reactor design the strain rates were lower compared to oscillating grid set-ups. The low range of Reynolds numbers based on Taylor microscales (Re_λ~0.5–5.9) covered both turbulent and non-turbulent flow regimes. The small-scale fluid motion produced over the entire volume of this reactor makes it suitable for experiments examining the physiological responses of fluid motion on microorganisms.     

Use of glucose oxidase system in measuring aeration capacity of fermentors. Comparison of the dynamic and steady-state methods of kla measurement

A steady-state method for k_la determination has been presented using the Michaelis–Menten two-substrate kinetic equation for oxidation of glucose in the presence of the enzymes glucose oxidase and excess catalase. The conditions have been specified where spontaneous hydrolysis of lactone is sufficiently rapid, thus eliminating inhibitory action of lactone on the oxidation. In glucose oxidase-free batches, the k_la values were determined using various modification of the dynamic method. The dynamic methods in which gas interchange was effected without interrupting aeration and agitation of the batch yielded erroneously lower k_la values as compared to the results of steady-state methods if the measured k_la value was higher than 0.03 s^?1. The values yielded by the dynamic method in which the gas interchange was effected at the same time with turning on aeration and agitation of the batch agreed with values resulting from the steady-state method provided that the measured k_la values were lower than 0.08^?1 and the simultaneous interfacial transport of nitrogen and oxygen had been taken into account in the evaluation. At k_la values higher than 0.08 ^?1, this modification of the dynamic method also yielded lower k_la values as compared with the outcome of the steady-state method. The experiments performed do not, however, allow one to decide unambiguously on the whether these lower k_la values are due to failure of the adopted model to describe adequately the dynamic behavior of the system or whether they are true values differing from those yielded by the steady-state method on account of different physical properties of compared batches.     

Oxygen absorption in stirred tanks: A correlation for ionic strength effects

A new correlation is given for the prediction of the volumetric coefficient for mass transfer (K_La) in stirred tanks from dispersed gas bubbles to basal salt solutions of ionic strengths representative of fermentation media. The correlation includes the effects of both the operating parameters (agitation power per unit volume and gas superficial velocity) and the physicochemical properties of the system: interfacial tension, viscosity, density, diffusion, coefficient and, in particular, ionic strength. The effect of the latter was found to be most significant in the Newtonian systems of water-like viscosity investigated; no previous correlations have included the effect of ionic strength. K_La values were determined by using a dissolved oxygen probe to monitor the steady-state oxygen tension in continuous flow experiments, and/or the rate of change of oxygen tension in unsteady-state semibatch experiments. In the latter cases, results were computed by a nonlinear, least squares computer program which fitted the experimental data to a model of probe transient response characteristics. The general applicability of the model and the computational procedure was verified by comparing the results to those obtained with the same electrolyte solution in the steady-state mode. The experiments were run over a wide range of agitation power inputs, including those typical of both soluble- and insoluble-substrate fermentations. The correlation appears to be valid for both oxygen mass transfer with and without homogeneous chemical reaction in the liquid phase; in the former case, for example, sulfite oxidation, knowledge of the chemical reaction enhancement factor is required. In addition to predicting oxygen transfer capabilities, the correlation may be used for other sparingly soluble gases of interest in fermentation systems, such as methane, hydrogen, and carbon dioxide.     

Aerobic degradation of toluene in a closed chemostat with and without a head space outlet

The present paper describes the continuous aerobic cultivation of a Pseudomonas strain with toluene as the substrate in a closed chemostat with oxygen or air as the gas phase. Due to the constant supply of a nitrogen-saturated aqueous medium, nitrogen passes from the liquid phase of the chemostat into the gas phase (head space). This results in an increasing nitrogen content (asymptotic approach to 100%). The concomitant decrease in the partial pressure of the oxygen in the gas phase finally leads to an oxygen limitation for the bacteria in the medium and an incomplete toluene degradation. The critical nitrogen content of the gas phase at which oxygen limitation begins depends on the toluene concentration in the incoming medium. However, when the gas is continuously removed from the head space, the nitrogen content reaches a steady-state value of less than 100%, depending on the flow rate of the outgoing gas. The oxygen limitation and the associated incomplete toluene degradation can be prevented in this way. The method of gas removal from the head space to avoid oxygen limitation is also applicable when the reactor is supplied with air instead of oxygen. Waste waters contaminated with highly volatile pollutants can thus be biologically decontaminated under aerobic conditions, without shifting the pollution problem from the liquid to the gas phase.     

Dynamic measurement of the volumetric mass transfer coefficient in agitated vessels: Effect of the start-up period on the response of an oxygen electrode

The response of a polarographic oxygen electrode to a step change and to an exponential change in bulk oxygen concentration was studied theoretically and experimentally for the case where there is a significant liquid film resistance at the outerside of the membrane-covered electrode. The probe response has been described considering the start-up period of the concentration changes (the period of time that will elapse before the new concentration level is established and/or before the volumetric mass transfer coefficient k_La regains its steady-state value after the gas supply is opened to the fermentor). A linear change of the pertinent characteristics is assumed during this start-up period. It is shown that a substantial error could be introduced by neglecting the start-up period for cases frequently occurring in practice. In addition, the dependences of the probe response on the direct contact of bubbles with an electrode and on the fluid flow field around it were discussed.     

Proteome analysis of the Escherichia coli heat shock response under steady-state conditions

In this study a proteomic approach was used to investigate the steady-state response of Escherichia coli to temperature up-shifts in a cascade of two continuously operated bioreactors. The first reactor served as cell source with optimal settings for microbial growth, while in the second chemostat the cells were exposed to elevated temperatures. By using this reactor configuration, which has not been reported to be used for the study of bacterial stress responses so far, it is possible to study temperature stress under well-defined, steady-state conditions. Specifically the effect on the cellular adaption to temperature stress using two-dimensional gel electrophoresis was examined and compared at the cultivation temperatures of 37°C and 47.5°C. As expected, the steady-state study with the double bioreactor configuration delivered a different protein spectrum compared to that obtained with standard batch experiments in shaking flasks and bioreactors. Setting a high cut-out spot-to-spot size ratio of 5, proteins involved in defence against oxygen stress, functional cell envelope proteins, chaperones and proteins involved in protein biosynthesis, the energy metabolism and the amino acid biosynthesis were found to be differently expressed at high cultivation temperatures. The results demonstrate the complexity of the stress response in a steady-state culture not reported elsewhere to date.     

Control of the degree of hydrolysis of a protein modification with immobilized protease by the pH-drop method

A mathematical model, DH = 1/k(10^?2ΔpH ? 1), between the pH-drop (ΔpH) and degree of hydrolysis (DH) of an enzymatic modification of casein was developed to assess the DH in a packed-bed column reactor by directly monitoring the pH value of the modified protein system. It was demonstrated that the linear DH range and the k value of the equation were dependent on the reactor type and the specificity of the proteolytic enzymes immobilized on chitin used in the present study, but no effect of the substrate casein concentration on the linear DH range was observed. Since DH and ΔpH values of the modified casein correlated with the flow rate in a packed-bed column reactor, it was suggested that the DH value, in a considerably wide range of casein modification with a certain immobilized protease in a column reactor, could be controlled by adjusting the flow rate of the substrate and monitored by a pH-meter. This relationship might be used as a basis for scale-up and long-term operation of enzymatic modification of proteins by immobilized protease in a column reactor.     

Kinetic behavior of heterogeneous populations in completely mixed reactors

The kinetic behavior of heterogeneous microbial populations was studied in a continuous flow completely mixed reactor operated at various dilution rates. Glucose was used as the growth-limiting nutrient. The physiological growth parameters for cells harvested from continuous flow reactors were determined using batch experiments. It, was found that the growth parameters, maximum growth rate (μ_m), saturation constant (k_s), and cell yield (Y) vary for each dilution rate, and cannot be considered as precise constants in depicting the kinetic behavior of heterogeneous populations. In addition, it was found that the yield coefficients obtained from batch experiments were always lower than those obtained from continuous flow experiments. Levels of substrate and biological solids calculated for different dilution rates using growth constants from batch experiments did not agree with the experimental values observed in steady-state experiments. However, when the yield values from, the continuous flow experiments were used in conjunction with batch values for μ_m and k_s the theoretical and experimental dilute-out curves agreed fairly closely (within the range needed for engineering prediction) until the culture began to wash out of the unit. In general, the data substantiated the use of the single phase relationship between growth rate and substrate concentration described by the Monod equation, μ = μ_mS/(k_s + s).     

A novel split-cylinder airlift reactor for fed-batch cultures

Conventional airlift reactors are not adequate to carry out variable volume processes since it is not possible to achieve a proper liquid circulation in these reactors until the liquid height is higher than that of the downcomer. To carry out processes of variable volume, the use of a split-cylinder airlift reactor is proposed, in the interior of which two multi-perforated vertical baffles are installed in order to provide several points of communication between the reactor riser and downcomer. This improves the liquid circulation and mixing at any liquid volume. In fed-batch cultures, it is important to know how liquid height affects the hydrodynamic characteristics and the volumetric oxygen transfer coefficient since this impacts on the kinetic behavior of any fermentation. Thus, in the present work, the effect of the liquid height on the mixing time, the overall gas hold-up, and the volumetric oxygen transfer coefficient of the proposed airlift reactor were determined. The mixing time was increased and the volumetric oxygen transfer coefficient decreased due to the increase of the liquid height in the reactor in all the superficial gas velocities tested, which corresponded to a pseudohomogeneous flow regime. The experimental values of the mixing time and the mass-transfer coefficient were properly described through correlations in which the U_GR/H_L ratio was used as the independent variable. Thus, this variable might be used to describe the hydrodynamic behavior and the oxygen transfer coefficient of airlift reactors when such reactors are used in processes where the liquid volume changes with time. However, these correlations are useful for the particular device and for the particular operating conditions at which they were obtained. These empirical correlations are useful to understand some factors that influence the mixing time and volumetric oxygen transfer coefficient, but such correlations do not have a sufficient predictive potential for a satisfactory reactor design. The overall gas hold-up values were not significantly affected when the liquid height was lower than the downcomer height. However, the values decreased abruptly when the reactor was operated with liquid heights over the downcomer height, especially at high superficial gas velocities.     

Venturi aeration and thermophilic aerobic sewage sludge digestion in small-scale reactors

Summary The aeration performance of two venturi aeration reactors (operating volumes 381 and 251) was studied for an air-water system. It was found that the mass transfer coefficient (k _la) could be described in terms of the superficial gas velocity (V _s) alone by the simple expressionk _La=aV _infb ^supS with constantsa=0.313,b=0.579 for the 38-1 reactor anda=0.214,b=0.534 for the 25-1 reactor. A similar relationship was obeyed when the 38-1 reactor was aerated with a diffuser tile (a=17.0,b=1.52). A linear relationship betweenk _La and gas hold-up was observed for the 38-1 reactor with both venturi and diffuser aeration. The 25-1 reactor was used successfully for the thermophilic aerobic digestion of sewage sludge. A mean sludge temperature rise of 30°C was observed. Chemical oxygen demand, pH, and total solids content of the digested sludge differed significantly from the feed sludge and were similar to values obtained for full-scale thermophilic aerobic digestion. No significant differences between inorganic solids content, dissolved oxygen concentration, or redox potential were observed between feed and digested sludge.     

Oxygen transfer in an airlift reactor with multiple net draft tubes

A pilot scale airlift reactor with multiple net draft tubes was developed to improve oxygen transfer in the reactor. The reactor was 0.29 m in diameter and 2 m height. A steadystate sulfite oxidation method was applied to determine an overall volumetric mass transfer coefficient. Oxygen transfer of the proposed airlift reactor can be 60–100% higher than that of bubble columns under the same operating conditions.List of Symbols C ^* mol·dm^–3 saturated concentration of dissolved oxygen - C _L mol·dm^–3 bulk concentration of dissolved oxygen - G mol/min nitrogen flow rate - k _L a hr^–1 the volumetric gas-liquid mass transfer coefficient - Mo ₂ g/mol molecular weight of oxygen - OTR g/min the oxygen transfer rate - U _g cm/s superficial air velocity - V _L dm³ volume of the liquid phase - _in oxygen mole ratio in the inlet gas - _out oxygen mole ratio in the outlet gas     

Two‐compartment processing as a tool to boost recombinant protein production

Pichia pastoris is used extensively as a production platform for many recombinant proteins. The dissolved oxygen (DO) is one of the most important factors influencing protein production. The influence of the DO on productivity has not been studied independent from the feed rate. In this work, various DO levels were investigated independent from the feed rate. The model system was recombinant P. pastoris under the control of methanol‐induced alcohol oxidase promoter, which expressed HRP as the target protein. No significant effect was observed in terms of titer and specific productivity, which is a confirmation of the fact that the DO in a one‐compartment system cannot boost productivity for the model system under study. Hence, a two‐compartment system (a single reactor coupled with a plug flow reactor) was designed and implemented in order to apply oxygen‐related stress in the plug flow reactor and allow the cells to be recovered in the main reactor. Doing so, more than two‐fold increase in the titer and productivity and three‐fold increase in protein‐specific activity were achieved. Hence, partial application of oxygen‐related stress in the two‐compartment system was proposed as a process technology to enhance protein production.     

A new technique for the performance evaluation of clean-in-place disinfection of biofilms

A concentric cylinder reactor (CCR) is described that enables the steady-state kinetics of microbial biofilms to be evaluated under conditions of constant nutrient flow and variable shear-stress. The reactor has been used to evaluate the influence of fluid dynamic shear on the extent and mode of detachment of bacteria from biofilms. Using a food factory isolate of Pseudomonas aeruginosa, a general increase in the overall growth rate and detachment of the biofilms (cfu cm⁻² min⁻¹) with time was shown for each biofilm, regardless of the prevailing shear. As the shear rate was increased beyond 0.123 ms⁻¹, populations tended toward a pseudo steady-state. Sudden changes in shear force, however, caused dramatic changes in the productivity of steady-state populations. The CCR provides an effective means of testing disinfectant activity, particularly for clean-in-place situations, and allows for an examination of the residual effects of a cleansing programme on a treated surface for three different chemical classes of disinfectant. Utilisation of the CCR would, therefore, provide enhanced ability to determine the efficacy and efficiency of chemical products for use in sanitation protocols. Journal of Industrial Microbiology & Biotechnology (2000) 25, 235–241. Received 09 February 2000/ Accepted in revised form 27 June 2000     

Branched respiratory chain in aerobically grown Staphylococcus aureus —oxidation of ethanol by cells and protoplasts

Addition of ethanol and some other primary alcohols, except methanol, to cells and protoplasts (but not membrane particles) considerably stimulated the rate of oxygen consumption. This additional respiration was strongly inhibited by 0.1 mM KCN. The cyanide inhibition curve of endogenous substrate oxidation was slightly biphasic while in the presence of ethanol it became clearly biphasic having K _i values of approx. 0.1 and 0.5 mM. Based on the steady-state cytochrome spectra in the presence of 0.1 mM KCN, we attributed the lower K _i to cytochrome a ₆₀₂. Proteolysis of protoplasts external membrane proteins did not change the rate of endogeneous substrate oxidation but prevented the inhibition of this respiration by low concentrations of KCN and stimulation of oxygen consumption by ethanol. The activity of NAD⁺-dependent ethanol dehydrogenase in the cytoplasm was found to be 520 nmol NADH-x min^–1 x mg^–1 protein. Proteolysis of external membrane proteins apparently inhibits the operation of the cytochrome a ₆₀₂-containing electron transport branch inducing the suppression of electron flow from NADH to oxygen.     

Control of mitochondrial respiration in the heart in vivo

The role of the hydrolysis products of adenosine triphosphate (ATP), adenosine diphosphate (ADP) and inorganic phosphate (Pi), in the control of myocardial respiration was evaluated in vivo using ³¹P NMR. These studies were conducted to evaluate whether increases in the ATP hydrolysis products can be detected through the cardiac cycle or during increases in cardiac work. ³¹P NMR data acquisitions gated to various portions of the cardiac cycle (50 msec time resolution) revealed that cytosolic ATP, ADP and Pi did not change over the course of the cardiac cycle. These metabolites were also monitored during steady-state increases in cardiac work in conjunction with measurements of coronary blood flow and oxygen consumption. No changes were observed during 2 to 3 fold increases in myocardial oxygen consumption induced by various methods. These results demonstrate that the cytosolic ATP, ADP, and Pi concentrations remain relatively constant throughout the cardiac cycle and during physiological increases in cardiac work and oxygen consumption. Furthermore, it is shown that ADP and Pi cannot be solely responsible for the regulation of cardiac respiration in vivo based on the in vitro Km values of these compounds for oxidative phosphorylation. It is concluded that other mechanisms, working in concert with the simple kinetic feedback of ATP hydrolysis products, must be present in the cytosol to provide control of myocardial respiration in vivo.     

Rapid Method for Assessing Oxygen Consumption Rate of Cells from Transient-state Measurements of Pericellular Dissolved Oxygen Concentration

Recently we described a method for estimating the oxygen consumption rate (OCR) of cells in static culture from equilibrium measurements of dissolved oxygen concentration (dO₂), using an oxygen-sensing microplate and the steady-state solution to Fick's Law (Guarino et al. 2004). Here we describe a complementary method for estimating OCR from the transient-state rate of change of measured dO₂. Although the system is open to the atmosphere and subject to a significant lag in sensor response, the rate of change of the measured dO₂ immediately after seeding correlated directly with both cell number and steady-state OCR. This transient-state method is linear with cell number to a much higher density than is possible with the steady-state method because it derives from measurements made before diffusion limitations can be established. For a given sensor thickness, the same correlation line between the transient and non-diffusion-limited steady-state estimates of OCR was found to apply for various preparations of rat hepatocytes. The correlation slope varied predictably with sensor thickness. Thus, despite the non-idealities of this system, the initial rate measurement offers a rapid method to obtain an estimate of absolute OCR. To demonstrate the utility of this method, we purposefully treated rat hepatocytes in ways expected to change OCR. Cells deprived of oxygen by storage under several centimeters of medium showed decreases in both OCR and viability with time. Likewise, the OCR of hepatocytes exposed to the oxidative phosphorylation inhibitor rotenone decreased, whereas those exposed to the uncoupler dinoseb increased.