Progress in monitoring, modeling and control of bioprocesses during the last 20 years

The paper gives a review on the recent development of bioprocess engineering. It includes monitoring of product formation processes by flow injection analysis, various types of chromatographic and spectroscopic methods as well as by biosensors. The evaluation of mycelial morphology and physiology by digital image analysis is discussed also. It deals with advanced control of indirectly evaluated process variables by means of state estimation/observer, with the use of structured and hybrid models, expert systems and pattern recognition for process optimization and gives a short report on the state of the art of metabolic flux analysis and metabolic engineering.     

pH prediction and control in bioprocesses using mid-infrared spectroscopy

An on-line pH monitoring method based on mid-infrared spectroscopy relevant to bioprocesses is presented. This approach is non-invasive and does not require the addition of indicators or dyes, since it relies on the analysis of species of common buffers used in culture media, such as phosphate buffer. Starting with titrations of phosphoric and acetic acid solutions over almost the entire pH range (2-12), it was shown that the infrared spectra of all samples can be expressed as a linear combination of the molar absorbance of the acids and their deprotonated forms. In other words, pH had no direct influence on the molar infrared spectra themselves, but only on deprotonation equilibria. Accurate prediction (standard error of prediction for pH < 0.15 pH units) was achieved by taking into account the non-ideal behavior of the solutions, using the Debye-Hückel theory to estimate the activity coefficients. Batch cultures of E. coli were chosen as a case study to show how this approach can be applied to bioprocess monitoring. The discrepancy between the spectroscopic prediction and the conventional electrochemical probe never exceeded 0.12 pH units, and the technique was fast enough to implement a feedback controller to maintain the pH constant during cultivation.     

Data-based modeling and analysis of bioprocesses: some real experiences

Data-generated models find numerous applications in areas where the speed of collection and logging of data surpasses the ability to analyze it. This work is meant to addresses some of the challenges and difficulties encountered in the practical application of these methods in an industrial setting and, more specifically, in the bioprocess industry. Neural network and principal component models are the two topics that are covered in detail in this paper. A review of these modeling technologies as applied to bioprocessing is provided, and four original case studies using industrial fermentation data are presented that utilize these models in the context of prediction and monitoring of bioprocess performance.     

Coupled metabolic-hydrodynamic modeling enabling rational scale-up of industrial bioprocesses

Metabolomics aims to address what and how regulatory mechanisms are coordinated to achieve flux optimality, different metabolic objectives as well as appropriate adaptations to dynamic nutrient availability. Recent decades have witnessed that the integration of metabolomics and fluxomics within the goal of synthetic biology has arrived at generating the desired bioproducts with improved bioconversion efficiency. Absolute metabolite quantification by isotope dilution mass spectrometry represents a functional readout of cellular biochemistry and contributes to the establishment of metabolic (structured) models required in systems metabolic engineering. In industrial practices, population heterogeneity arising from fluctuating nutrient availability frequently leads to performance losses, that is reduced commercial metrics (titer, rate, and yield). Hence, the development of more stable producers and more predictable bioprocesses can benefit from a quantitative understanding of spatial and temporal cell-to-cell heterogeneity within industrial bioprocesses. Quantitative metabolomics analysis and metabolic modeling applied in computational fluid dynamics (CFD)-assisted scale-down simulators that mimic industrial heterogeneity such as fluctuations in nutrients, dissolved gases, and other stresses can procure informative clues for coping with issues during bioprocessing scale-up. In previous studies, only limited insights into the hydrodynamic conditions inside the industrial-scale bioreactor have been obtained, which makes case-by-case scale-up far from straightforward. Tracking the flow paths of cells circulating in large-scale bioreactors is a highly valuable tool for evaluating cellular performance in production tanks. The “lifelines” or “trajectories” of cells in industrial-scale bioreactors can be captured using Euler-Lagrange CFD simulation. This novel methodology can be further coupled with metabolic (structured) models to provide not only a statistical analysis of cell lifelines triggered by the environmental fluctuations but also a global assessment of the metabolic response to heterogeneity inside an industrial bioreactor. For the future, the industrial design should be dependent on the computational framework, and this integration work will allow bioprocess scale-up to the industrial scale with an end in mind.     

The effect of applied sulphur on the growth, grain yield and control of powdery mildew in spring wheat

During the years 1998 and 1999, two field experiments were conducted at the University of Wales, Aberystwyth, UK, to test the effects of soil‐ and foliar‐applied sulphur (S) in spring wheat. S was applied at 0, 20, 40, 60 kg ha^?1 in 1998 and at 60 kg ha^?1 in 1999, using CaSO₄ as a source for the soil application and micronised S (Thiovit, 80%), with and without an organosilicone adjuvant, as a source for the foliar application. Senescence was retarded and grain yield was increased in 1998, following application of foliar S in conjunction with the organosilicone adjuvant. Application of foliar S was associated with a reduction in the level of mildew (Erysiphe graminis) recorded on the upper leaves and ears of the canopy. In 1999, grain yield was unaffected by treatments. A low level of mildew in the crop, particularly on the ears, is thought to be the reason for the lack of response in spite of the fact that senescence was retarded with foliar S application. A combined application of foliar S and commercial fungicide (cyproconazole) to the crop appeared to be more effective at controlling mildew than either S or fungicide applied alone. The study shows that there may be a role for S in a low‐input/organic wheat production system, where there is a need to reduce artificial inputs.     

茴香薄翅野螟虫口密度与油菜产量损失关系及其防治指标的研究

在网室内人工接茴香薄翅野螟Evergestis extimalis Scopoli卵,研究结果表明,随着幼虫数量的增加,油菜产量损失增加;与对照相比,当接卵数为216、399、858、1665和3279粒时,产量损失分别达3.13%,2.08%,12.5%,25.10%和43.33%。各网室实际接虫数(x)与其相应的小区油菜产量(y)的关系为:y=-0.0007x+4.8129,表明虫口密度与油菜产量为负相关。在青海省油菜种植区内每平米幼虫或卵数超过10头或粒时,应即时防治。     

On extremum seeking in bioprocesses with multivalued cost functions

Finding optimal operating modes for bioprocesses has been, for a long time, a relevant issue in bioengineering. The problem is of special interest when it implies the simultaneous optimization of competing objectives. In this paper, we address the problem of finding optimal steady states that achieve the best tradeoff between yield and productivity by using nonmodel ‐ based extremum‐seeking control with semiglobal practical stability and convergence properties. A special attention is paid to processes with multiple steady states and multivalued cost functions. © 2009 American Institute of Chemical Engineers Biotechnol. Prog., 2009     

水氮配合对绿洲沙地农田玉米产量、土壤硝态氮和氮平衡的影响

在黑河中游边缘绿洲沙地农田研究了不同的水氮配合对玉米产量、土壤硝态氮在剖面中的累积和氮平衡的影响.结果表明,施氮处理较不施氮处理产量增加48.22%～108.6%,施氮量超过225 kg hm-2,玉米产量不再显著增加.受土壤结构影响土壤硝态氮在土壤中呈"W"型分布,即土壤硝态氮含量在0～20 cm、140～160 cm和260～300 cm土层均出现峰值,并随施氮量增加,峰值增高.在常规高灌溉量处理硝态氮含量峰值最高值出现在260～300 cm土层,节水25%灌溉处理硝态氮含量峰值最高值出现在土壤表层0～20 cm土层.在常规高灌溉量处理0～300 cm土层中200～300土层硝态氮累积量所占比例最高,介于27.56%～51.86%之间;节水25%灌溉处理在0～300 cm土层中100～200土层硝态氮累积量所占比例最高,介于32.94%～38.07%之间;表明低灌溉处理下土壤硝态氮在土壤浅层累积较多,而高灌溉处理使更多的硝态氮淋溶至土壤深层.与2006年相比,2007年不施氮处理0～200 cm土层土壤硝态氮含量和积累量均明显减少;而施氮处理变化很小,在低灌溉处理甚至表现出硝态氮含量和积累量增加,表明施氮是土壤硝态氮累积的主要来源,而灌溉则使硝态氮向土壤深层淋溶.0～200 cm 土层土壤硝态氮累积量平均介于27.66～116.68 kg hm-2、氮素表观损失量平均介于77.35～260.96 kg hm-2,和施氮量均呈线性相关,即随施氮量增加,土壤硝态氮累积量和氮素表观损失量均增加,相关系数R2介于0.79～0.99之间,相关均显著.随施氮量增加,玉米总吸氮量和氮收获指数增加,氮的农学利用率降低,而灌溉的影响较小.施氮量超过225 kg hm-2时,地上部植株氮肥吸收利用率和籽粒氮肥吸收利用率开始有降低趋势.所以,在沙地农田,节水10%～25%的灌溉水平和225 kg hm-2的施氮水平可以在避免水肥过量投入的基础上减少土壤有机氮淋溶对地下水造成的污染威胁.     

Quantum yield and energy dependence for photooxidation of chlorophyllide in greening wheat

Dark grown leaves of wheat were irradiated with red light of different intensities, at a temperature close to 0°C. The rate of photoreduction of the protochlorophyllide 650-form into chlorophyllide 684-form was measured. On continued irradiation the chlorophyllide 684-form was photodecomposed. By comparing the rates of the two processes the quantum yield for photooxidation of the chlorophyllide 684-form was calculated. The quantum yield was 2°10^-5 at an intensity of 2200 W m^-2, and increased with decreasing light intensity to 3.2°10^-5 at an intensity of 170 W m^-2.     

Stem growth characteristics of high yielding Miscanthus correlate with yield,development and intraspecific competition within plots

High yielding perennial grasses are utilized as biomass for the bioeconomy and to displace fossil fuels. Many such grasses, including Miscanthus, are largely undomesticated. The main Miscanthus crop is a naturally occurring hydrid M. × giganteus (Mxg). All above ground biomass from Miscanthus is harvested. Stem traits correlate strongly with yield and therefore understanding the seasonal progression of stem growth should identify routes for improved yield. If such studies utilized high yielding commercial genotypes growing in plots the conclusions are likely to be more commercially relevant. Stem elongation was measured from five high yielding genotypes, 10 plants per plot from 20 plots in a replicated field trial over 4 years. Richards growth function produced an accurate fit to stem elongation. Differentials, double differentials and integrals of the parameterized function produced six growth characteristics, describing growth rate, timing and duration of the logarithmic growth phase and area under the growth curve. Maximum growth rate was correlated with yield and compensatory interactions were identified, for example plants with higher maximal growth rates had shorter durations of logarithmic growth. Plant position within plots of lower yielding genotypes did not affect growth characteristics but had a significant effect on late season growth characteristics in higher yielding genotypes. Two high yielding genotypes were compared over 3 years and growth parameterized using four different factors. The inverse correlation between maximum growth rate and duration of logarithmic growth was consistent across years and factors in both genotypes except when parameterized using temperature and only in Mxg. This suggested that different limitations to growth were exerted on the two genotypes which may help explain the exceptional performance of the Mxg genotype. We discuss the implications of the identified complex interactions in growth characteristics for approaches to maximize seasonal yield in perennial biomass crops.     

Development of a large-scale biocalorimeter to monitor and control bioprocesses

Calorimetry has shown real potential at bench-scale for chemical and biochemical processes. The aim of this work was therefore to scale-up the system by adaptation of a standard commercially available 300-L pilot-scale bioreactor. To achieve this, all heat flows entering or leaving the bioreactor were identified and the necessary instrumentation implemented to enable on-line monitoring and dynamic heat balance estimation. Providing that the signals are sufficiently precise, such a heat balance would enable calculation of the heat released or taken up during an operational (bio)process. Two electrical Wattmeters were developed, the first for determination of the power consumption by the stirrer motor and the second for determination of the power released by an internal calibration heater. Experiments were designed to optimize the temperature controller of the bioreactor such that it was sufficiently rapid so as to enable the heat accumulation terms to be neglected. Further calibration experiments were designed to correlate the measured stirring power to frictional heat losses of the stirrer into the reaction mass. This allows the quantitative measurement of all background heat flows and the on-line quantitative calculation of the (bio)process power. Three test fermentations were then performed with B. sphaericus 1593M, a spore-forming bacterium pathogenic to mosquitoes. A first batch culture was performed on a complex medium, to enable optimization of the calorimeter system. A second batch culture, on defined medium containing three carbon sources, was used to show the fast, accurate response of the heat signal and the ability to perfectly monitor the different growth phases associated with growth on mixed substrates, in particular when carbon sources became depleted. A maximum heat output of 1100 W was measured at the end of the log-phase. A fed-batch culture on the same defined medium was then carried out with the feed rate controlled as a function of the calorimeter signal. A maximum heat output of 2250 W was measured at the end of the first log-phase. This work demonstrates that real-time quantitative calorimetry is not only possible at pilot-scale, but could be readily applied at even larger scales. The technique requires simple, readily available devices for determination of the few necessary heat flows, making it a robust, cost-effective technique for process development and routine monitoring and control of production processes.     

Expanded thermodynamic model for microbial true yield prediction

Thermodynamic methods to predict true yield and stoichiometry of bacterial reactions have been widely used in biotechnology and environmental engineering. However, yield predictions are often inaccurate for certain simple organic compounds. This work evaluates an existing method and identifies the cause of prediction errors for compounds with low degree of reductance of carbon. For these compounds, carbon, not energy or reducing equivalents, constrains growth. Existing thermodynamically-based models do not account for the potential of carbon-limited growth. The improved method described here consists of four balances: carbon balance, nitrogen balance, electron balance, and energy balance. Two efficiency terms, K1 and K2 are defined and estimated from a priori analysis. The results show that K1 and K2 are nearly the same in value so that only one coefficient, K = 0.41 is used in the modified model. Comparisons with observed yields show that use of the new model and parameters results in significantly improved yield estimation based on inclusion of the carbon balance. The average estimation error is less than 6% for the data set presented.     

Stemming the obesity epidemic: a tantalizing prospect

Objective: Obesity is a growing problem worldwide, but there are no good methods to assess the future course of the epidemic and the potential influence of interventions. We explore the behavior change needed to stop the obesity epidemic in the U.S. Research Methods and Procedures: We modeled the population distribution of BMI as a log‐normal curve of which the mean shifts upward with time due to a positive population energy balance. Interventions that decrease food intake or increase physical activity result in more favorable trends in BMI. Results: The recently observed trend in average BMI implies that the average U.S. adult over‐consumes by ～10 kcal/d. If this trend continues unaltered, obesity prevalence will exceed 40% for men and 45% for women in 2015. To stop the epidemic, it suffices to decrease caloric consumption by ～10 kcal or walk an extra 2 to 3 minutes per day, on average. Discussion: This leads to a paradox: little behavior change seems sufficient to halt the epidemic, but in practice this proves hard to achieve. The obesogenic environment is the likely culprit. Individuals trying to maintain a healthy weight need to be supported by environments that stimulate physical activity and do not encourage over‐consumption. Research should show what measures are effective.     

Optimal cofactor swapping can increase the theoretical yield for chemical production in Escherichia coli and Saccharomyces cerevisiae

Maintaining cofactor balance is a critical function in microorganisms, but often the native cofactor balance does not match the needs of an engineered metabolic flux state. Here, an optimization procedure is utilized to identify optimal cofactor-specificity “swaps” for oxidoreductase enzymes utilizing NAD(H) or NADP(H) in the genome-scale metabolic models of Escherichia coli and Saccharomyces cerevisiae. The theoretical yields of all native carbon-containing molecules are considered, as well as theoretical yields of twelve heterologous production pathways in E. coli. Swapping the cofactor specificity of central metabolic enzymes (especially GAPD and ALCD2x) is shown to increase NADPH production and increase theoretical yields for native products in E. coli and yeast—including l-aspartate, l-lysine, l-isoleucine, l-proline, l-serine, and putrescine—and non-native products in E. coli—including 1,3-propanediol, 3-hydroxybutyrate, 3-hydroxypropanoate, 3-hydroxyvalerate, and styrene.     

Thermodynamic yield predictions for biodegradation through oxygenase activation reactions

Activation reactions involve modification of recalcitrant substrates to forms that are more readily degradable. These reactions require specialized enzymes and cosubstrates, including molecular oxygen and reduced electron carriers. In these reactions, microorganisms invest electrons and cannot capture energy or carbon for synthesis. The subsequent degradation of the intermediates formed in activation reactions releases electrons, energy, and carbon that the organisms use for growth. The overall yield is reduced due to the required activation investments. A mathematical method to predict cell yields of oxygenase activation reactions is developed using electron and energy balances. Predicted yields are compared with experimental yields for methane, organic chelating agents, and aromatic hydrocarbons.     

Effect of changes in diet energy density on feed intake, milk yield and metabolic parameters in dairy cows in early lactation

The purpose of this experiment was to investigate how early lactating cows adjust their metabolism and production to acute, but moderate changes in the energy density of the diet. Sixty dairy cows were randomly assigned to one of four treatments: two change-over groups (HNH and NHN) and two control groups (HHH and NNN), where H and N refer to a high and normal energy density in the total mixed ration (TMR), respectively. The experimental period covered the first 9 weeks post calving, which was split up in three 3-week periods. Thus, cows assigned to HNH or NHN shifted TMR in weeks 4 and 7 after calving while cows assigned to HHH or NNN were fed the same TMR for all 9 weeks. Results from cows on treatment HNH were compared with group HHH while cows on treatment NHN were compared with group NNN. When the diet changed from N to H and H to N, cows increased and decreased their dry-matter intake (DMI), respectively compared with control groups. Cows adjusted milk yield accordingly to changes in DMI, although not always significantly. Energy-corrected milk yield was not significantly affected by any of the changes in the energy density of the diet but generally showed same tendencies as milk yield. Non-esterified fatty acids (NEFA), beta-hydroxybutyrate in blood and milk and triacylglycerol and glycogen content in the liver were not significantly affected by changes in the energy density of the diet, except from NEFA at one change. Glucose increased more when the diet changed from N to H and increased less when the diet changed from H to N, compared with control groups, although not always significantly. Collectively, these results suggest that cows adjust their DMI and partly milk yield according to the energy density of the diet and therefore only limited effects were observed in physiological parameters.     

Basic concepts and principles of stoichiometric modeling of metabolic networks

Metabolic networks supply the energy and building blocks for cell growth and maintenance. Cells continuously rewire their metabolic networks in response to changes in environmental conditions to sustain fitness. Studies of the systemic properties of metabolic networks give insight into metabolic plasticity and robustness, and the ability of organisms to cope with different environments. Constraint-based stoichiometric modeling of metabolic networks has become an indispensable tool for such studies. Herein, we review the basic theoretical underpinnings of constraint-based stoichiometric modeling of metabolic networks. Basic concepts, such as stoichiometry, chemical moiety conservation, flux modes, flux balance analysis, and flux solution spaces, are explained with simple, illustrative examples. We emphasize the mathematical definitions and their network topological interpretations.     

Object-oriented fuzzy expert system for on-line diagnosing and control of bioprocesses

Summary An object-oriented fuzzy expert system to support on-line control of an automated fermentation plant is described. The major elements of the system consist of a fuzzy inference engine, a database, a knowledge base, and an expression evaluater. The expression evaluater calculates specific rates for growth, and substrate and product formation at different physiological states during the cultivation from the measured data. The specific rates are then compared with the standard target rates stored in the database. If differences outside the set tolerances were observed, the inference engine analyses the reasons for the faults on the basis of the knowledge represented in the form of a knowledge network and fuzzy membership functions of the process variables. The fuzzy expert system was developed on the basis of a shell constructed by using the object oriented Smalltalk/V Mac programming environment, with Lactobacillus casei lactic acid fermentation as the example of process application.Visiting scientist from Helsinki University of Technology at RIKEN Correspondence to: P. Linko or I. Endo     

Modeling and simulation of lactic acid fermentation with inhibition effects of lactic acid and glucose

An unstructured mathematical model for lactic acid fermentation was developed. This model was able to predict the inhibition effects of lactic acid and glucose and was confirmed to be valid with various initial concentrations of lactic acid and glucose. Simulation of energy production was made using this mathematical model, and the relationship between the kinetics of energy metabolism and lactic acid production was also analyzed.     

Disease assessment methods and their use in simulating growth and yield of peanut crops affected by leafspot disease

Peanut (Arachis hypogaea) crops in Benin often experience late leafspot (Cercosporidium personatum), which causes severe yield losses associated with leaf defoliation and necrosis. The objective of this research was to determine the best method of disease assessment and to test its utility in the CROPGRO‐peanut model to simulate growth and yield as affected by late leafspot in early and late maturing peanut cultivars grown at different sowing dates under rain‐fed conditions (without irrigation) in northern Benin. Two peanut cultivars TS 32‐1 and 69–101 were sown on three dates between May and August during 1998 and 1999. In both years there was severe occurrence of late leafspot and the progression of disease was earlier and faster with later sowing dates. Overall, the long duration cultivar 69–101 produced greater yield than the short duration cultivar TS 32‐1. The CROPGRO‐peanut model was able to predict and simulate the observed crop and pod dry matter over time when input on percent diseased leaf area and percent defoliation were provided. Of several disease assessments, the best approach was to input measured percent main‐stem defoliation above the fourth node and percent diseased leaf area estimated from visual leafspot score.