Physiologically motivated strategies for control of the fed-batch cultivation of recombinant Escherichia coli for phenylalanine production

The efficiency of the fed-batch cultivation of recombinant Escherichia coli AT2471 for phenylalanine production is highly dependent on the distribution of the carbon flow between the main process products — biomass, phenylalanine, acetic acid and carbon dioxide. In order to enhance the process performance, the effects of several factors, namely glucose feeding, tyrosine feeding and oxygen supply, were investigated experimentally. As a result, a set of control strategies was developed, designed to tolerate phenylalanine synthesis at the expense of the remaining products. The DO was controlled to prevent acetic acid excretion due to oxygen limitation. The total amount of tyrosine fed was used to provide an optimal balance between biomass synthesis and that of phenylalanine. Special algorithms for control of the glucose feed rate were applied to eliminate the threat of acetic acid excretion due to overfeeding, and at the same time, to reduce excessive CO₂ evolution caused by unnecessarily severe glucose limitation. The joint application of these strategies resulted in greatly improved efficiency in the phenylalanine production process: the final phenylalanine concentration reached 46 g/l, the yield was above 17%, and the productivity-0.85 g/l·h. In combination, these data exceed the results reported by others, and are much higher than those obtained by use before the implementation of the proposed complex of techniques.     

Quantitative comparison of dynamic physiological feeding profiles for recombinant protein production with Pichia pastoris

Pichia pastoris is widely used for the production of recombinant proteins in industrial biotechnology. In general, industrial production processes describe fed-batch processes based on the specific growth rate. Recently, we introduced the specific substrate uptake rate (q _s) as a novel parameter to design fed-batch strategies for P. pastoris. We showed that a dynamic feeding strategy where the feed was adjusted in steps to the maximum specific substrate uptake rate was superior to more traditional strategies in terms of specific productivity. In the present study, we compare three different dynamic feeding strategies based on q _s for a recombinant P. pastoris strain with respect to cell physiology, methanol accumulation, productivity and product quality. By comparing (A) a feeding profile at constant high q _s, (B) a periodically adjusted feeding profile for a stepwise q _s ramp, and (C) a feeding profile at linear increasing q _s, we evaluated potential effects of the mode of feeding. Although a dynamic feeding strategy with stepwise increases of q _s to q _{s max} resulted in the highest specific productivity, a feeding profile where the feeding rate was stepwise increased to a constant high q _s value was superior in terms of the amount of active enzyme produced and in the amount of accumulated methanol. Furthermore, this feeding strategy could be run automatically by integrating an online calculator tool, thus rendering manual interventions by the operator unnecessary.     

Design of a microaerobically inducible replicon for high-yield plasmid DNA production

A pUC-derived replicon inducible by oxygen limitation was designed and tested in fed-batch cultures of Escherichia coli. It included the addition of a second inducible copy of rnaII, the positive replication control element. The rnaII gene was expressed from P_trc and cloned into pUC18 to test the hypothesis that the ratio of the positive control molecule RNAII to the negative control element, RNAI, was the determinant of plasmid copy number per chromosome (PCN). The construct was evaluated in several E. coli strains. Evaluations of the RNAII/RNAI ratio, PCN and plasmid yield normalized to biomass (Y_pDNA/X) were performed and the initial hypothesis was probed. Furthermore, in high cell-density cultures in shake flasks, an outstanding amount of 126 mg/L of plasmid was produced. The microaerobically inducible plasmid was obtained by cloning the rnaII gene under the control of the oxygen-responsive Vitreoscilla stercoraria hemoglobin promoter. For this plasmid, but not for pUC18, the RNAII/RNAI ratio, PCN and Y_pDNA/X efficiently increased after the shift to the microaerobic regime in fed-batch cultures in a 1 L bioreactor. The Y_pDNA/X of the inducible plasmid reached 12 mg/g at the end of the fed-batch but the original pUC18 only reached ca. 6 mg/g. The proposed plasmid is a valuable alternative for the operation and scale-up of plasmid DNA production processes in which mass transfer limitations will not represent an issue.     

Effect of glucose and oxygen on β-lactam biosynthesis by Cephalosporium acremonium

The effects of glucose consumption rate (q_s) and oxygen limitation on the control of cephalosporin C (Ceph C) biosynthesis and the activities of deacetoxycephalosporin C synthetase/hydroxylase (DAOC-SH) and acetyl coenzyme A: deacetylcephalosporin C o-acetyltransferase (DAC-AT) were investigated in cultivations of the highly productive Cephalosporium acremonium strain TR87 under conditions similar to those used in industrial production. A carefully optimised time course of q_s during the first part of fed batch cultivations was essential for maximal Ceph C production. The actual glucose concentration in the medium was of secondary importance. A decrease of q_s between 20 and 35 h of cultivation was found to induce the early onset of antibiotic synthesis. By subsequently maintaining q_s at a relatively low level using a controlled feed of glucose and a limiting amount of phosphate, maximal production rates were obtained. Oxygen starvation after the onset of Ceph C production led to a pronounced increase in penicillin N formation, a reduced Ceph C yield (−30%) and a strongly reduced activity of the two enzymes tested. In general, neither the time course nor the absolute levels of the two enzyme activities directly correlated with the actual production rates of Ceph C. This is the first time where an independent parameter (q_s) has been demonstrated to be responsible for triggering the synthesis of an antibiotic.     

Fed-batch operation in special microtiter plates: a new method for screening under production conditions

Batch and fed-batch operation result in completely different physiological conditions for cultivated microorganisms or cells. To close the gap between screening, which is hitherto exclusively performed in batch mode, and fed-batch production processes, a special microtiter plate was developed that allows screening in fed-batch mode. The fed-batch microtiter plate (FB-MTP) enables 44 parallel fed-batch experiments at small scale. A small channel filled with a hydrogel connects a reservoir well with a culture well. The nutrient compound diffuses from the reservoir well through the hydrogel into the culture well. Hence, the feed rate can easily be adjusted to the needs of the cultured microorganisms by changing the geometry of the hydrogel channel and the driving concentration gradient. Any desired compound including liquid nutrients like glycerol can be fed to the culture. In combination with an optical measuring device (BioLector), online monitoring of these 44 fed-batch cultures is possible. Two Escherichia coli strains and a Hansenula polymorpha strain were successfully cultivated in the new FB-MTP. As a positive impact of the fed-batch mode on the used strains, a fourfold increase in product formation was observed for E. coli. For H. polymorpha, the use of fed-batch mode resulted in a strong increase in product formation, whereas no measurable product formation was observed in batch mode. In conclusion, the newly developed fed-batch microtiter plate is a versatile, easy-to-use, disposable system to perform fed-batch cultivations at small scale. Screening cultures in high-throughput under online monitoring are possible similar to cultivations under production conditions.     

Fed-batch cultures of recombinant Escherichia coli with inhibitory substance concentration monitoring

Fed-batch cultures of recombinant Escherichia coli HB101 were investigated to obtain high cell density and large amounts of β-galactosidase (β-gal). E. coli HB1010 was transformed with a hybrid plasmid pTREZ1, which contained a β-gal gene controlled by the trp promoter. In fed-batch cultures of recombinant E. coli, when the cell concentration reached around 13 g/l, the cell growth stopped and large amounts of inhibitory substances have accumulated in the broth. These inhibitory substances were isolated and identified. Acetate produced by the cells was evidently the main inhibitor of cell growth and β-gal production. Since the cells proved to assimilate acetate, the feed rate was controlled with acetate concentration monitoring in the fed-batch culture. As a result, the acetate concentration was maintained at a low level and cells grew smoothly without acetate-induced inhibition. Cell concentration and β-gal quantity reached high values of 28 g/l and 64 U/ml, respectively.     

A new process for red pigment production by submerged culture of Monascus purpureus

Formation of red pigment by Monascus purpureus via diauxic growth on glucose and ethanol in submerged culture was optimized based on inoculum preparation and culture medium. A vegetative inoculum was prepared from spores grown on ethanol. The optimized culture medium was low in phosphates, and had an initial pH?of 5.5. The characteristics of Monascus purpureus grown on glucose and on ethanol were compared: the specific consumption rate of glucose (q_G) was higher than the specific consumption rate of ethanol (q_E), whereas the specific growth rate was greatest with ethanol. The specific production rate of red pigment (p_OD) and pigment yield (Y_OD/s) with glucose was twice that with ethanol. A novel fermentation process was developed with M. purpureus initially grown with controlled ethanol formation, and consumption of the latter during pigment formation.     

Efficient feeding profile optimization for recombinant protein production using physiological information

A multivariate study was performed aiming at the optimization of a recombinant rhamnose inducible E. coli induction system with alkaline phosphatase as target product. The effects of typical factors with impact on post- as well as pre-induction feeding rates were investigated with respect to the space–time yield of the target product. The goal was increased understanding as well as quantitative characterization of these factors with respect to their physiological impact on the model system. The optical density (OD) at which the culture was induced had a strong positive effect on the space–time yield. Pre-induction growth rate (k) had a second-order effect, while induction feed rate drop (J), a factor defining the linear post-induction feed rate, was interacting with (k). However, explanation of the observed effects to acquire more understanding regarding their effect on cell metabolism was not straight forward. Hence, the original process parameters were transformed into physiological more meaningful parameters and served as the basis for a multivariate data analysis. The observed variance with respect to observed volumetric activity was fully explained by the specific substrate uptake rate (q _s) and induction OD, merging the process parameters pre-induction growth rate (k) and feed rate drop (J) into the physiological parameter specific substrate uptake rate (q _s). After transformation of the response volumetric activity (U/ml) into the biomass specific activity (U/g_biomass), the observed variance was fully explained solely by the specific substrate uptake rate (q _s). Due to physiological multivariate data analysis, the interpretation of the results was facilitated and factors were reduced. On the basis of the obtained results, it was concluded that the physiological parameter q _s rather than process parameters (k, J, induction OD) should be used for process optimization with respect to the feeding profile.     

Batch xylitol production by<Emphasis Type="Italic"> Candida guilliermondii</Emphasis> FTI 20037 from sugarcane bagasse hemicellulosic hydrolyzate at controlled pH values

Batch fermentation of sugarcane bagasse hemicellulosic hydrolyzate by the yeast Candida guilliermondii FTI 20037 was performed using controlled pH values (3.5, 5.5, 7.5). The maximum values of xylitol volumetric productivity (Q _p=0.76 g/l h) and xylose volumetric consumption (Q _s=1.19 g/l h) were attained at pH 5.5. At pH 3.5 and 7.5 the Q _p value decreased by 66 and 72%, respectively. Independently of the pH value, Y _x/s decreased with the increase in Y _p/s suggesting that the xylitol bioconversion improves when the cellular growth is limited. At the highest pH value (7.5), the maximum specific xylitol production value was the lowest (q _pmax=0.085 g/l h.), indicating that the xylose metabolism of the yeast was diverted from xylitol formation to cell growth.List of symbols P _max xylitol concentration (g/l) - Q _x volumetric cell production rate (g/l h) - Q _s volumetric xylose uptake rate (g/l h) - Q _p volumetric xylitol production rate (g/l h) - q _pmax specific xylitol production (g/g h) - q _smax specific xylose uptake rate (g/g h) - _max specific cell growth rate (h^–1) - Y _p/s xylitol yield coefficient, g xylitol per g xylose consumed (g/g) - Y _p/x xylitol yield coefficient, g xylitol per g dry cell mass produced (g/g) - Y _x/s cell yield coefficient, g dry cell mass per g xylose consumed (g/g) - _cell percentage of the cell yield from the theoretical value (%) - _xylitol percentage of xylitol yield from the theoretical value (%)     

Multi-omic based production strain improvement (MOBpsi) for bio-manufacturing of toxic chemicals

Robust systematic approaches for the metabolic engineering of cell factories remain elusive. The available models for predicting phenotypical responses and mechanisms are incomplete, particularly within the context of compound toxicity that can be a significant impediment to achieving high yields of a target product. This study describes a Multi-Omic Based Production Strain Improvement (MOBpsi) strategy that is distinguished by integrated time-resolved systems analyses of fed-batch fermentations. As a case study, MOBpsi was applied to improve the performance of an Escherichia coli cell factory producing the commodity chemical styrene. Styrene can be bio-manufactured from phenylalanine via an engineered pathway comprised of the enzymes phenylalanine ammonia lyase and ferulic acid decarboxylase. The toxicity, hydrophobicity, and volatility of styrene combine to make bio-production challenging. Previous attempts to create styrene tolerant E. coli strains by targeted genetic interventions have met with modest success. Application of MOBpsi identified new potential targets for improving performance, resulting in two host strains (E. coli NST74ΔaaeA and NST74ΔaaeA cpxP_o) with increased styrene production. The best performing re-engineered chassis, NST74ΔaaeA cpxP_o, produced ∼3 × more styrene and exhibited increased viability in fed-batch fermentations. Thus, this case study demonstrates the utility of MOBpsi as a systematic tool for improving the bio-manufacturing of toxic chemicals.     

Inhibitory effect of dihydroxyacetone onGluconobacter oxydans: Kinetic aspects and expression by mathematical equations

Summary Microbial conversion of glycerol into dihydroxyacetone (DHA) byGluconobacter oxydans was subjected to inhibition by excess substrate. Comparison of cultures containing increasing initial DHA contents (0 to 100 g l^–1) demonstrated that DHA also inhibited this fermentation process. The first effect was on bacterial growth (cellular development stopped when DHA concentration reached 67 gl^–1), and then on oxidation of glycerol (DHA synthesis only occurred when the DHA concentration in the culture medium was lower than 85 g l^–1). Productivity, specific rates and, to a lesser extent, conversion yields decreased as initial concentrations of DHA increased. The changes in the specific parameters according to increasing initial DHA contents were described by general equations. These formulae satisfactorily express the concave aspect of the curves and the reduction in biological activity when the cells were in contact with DHA concentrations of up to 96 g l^–1.Abbreviations X, S, P biomass, substrate, product concentrations - r _x,r _s,r _p rates of growth, consumption and production - ,q _s,q _p specific rates of growth, glycerol consumption and DHA production - Y _x/s, Y_p/s conversion yields of substrate into biomass and product - K _s constant of affinity of cells to the substrate - K _ip product inhibition constant - P _m threshold concentration of DHA in substrate     

The oxygen requirements of yeasts for the fermentation of d-xylose and d-glucose to ethanol

Summary The effect of oxygen availability on d-xylose and D-glucose metabolism by Pichia stipitis, Candida shehatae and Pachysolen tannophilus was investigated. Oxygen was not required for fermentation of d-xylose or d-glucose, but stimulated the ethanol production rate from both sugars. Under oxygen-limited conditions, the highest ethanol yield coefficient (Y_e/s) of 0.47 was obtained on d-xylose with. P. stipitis, while under similar conditions C. shehatae fermented d-xylose most rapidly with a specific productivity (q_pmax) of 0.32 h^-1. Both of these yeasts fermented d-xylose better and produced less xylitol than. P. tannophilus. Synthesis of polyols such as xylitol, arabitol, glycerol and ribitol reduced the ethanol yield in some instances and was related to the yeast strain, carbon source and oxygen availability. In general, these yeasts fermented d-glucose more rapidly than d-xylose. By contrast Saccharomyces cerevisiae fermented d-glucose at least three-fold faster under similar conditions.Nomenclature q_pmax maximum specific rate of ethanol production (g ethanol per g dry biomass per hour) - Y_e/s ethanol yield (g ethanol per g substrate utilized) - Y_p/s polyol yield (g polyol per g substrate utilized) - Y_x/s biomass yield (g dry biomass per g substrate utilized) - _max maximum specific growth rate (per hour)     

Effects of cultural conditions on the production of phenylalanine from a plasmid-harboring E. coli strain

Summary Phenylalanine production from E. coli KA 197/pJN6 (plasmid harboring genes for aro F, phe A^FBR, Amp^R and Tc^R) was studied under varying nutritional conditions in batch and continuous cultures. In batch culture experiments where growth was deliberately interrupted by limiting concentrations of sulphate and phosphate the phenylalanine production continued from the non-growing cells. However, the depletion of phosphate resulted in an immediate cessation of phenylalanine production but thereafter a low specific rate of phenylalanine formation resumed, while the decrease in specific rate of product formation was less after sulphate depletion. In the chemostat experiments, however, phosphate limitation was the only case where the specific rate of phenylalanine formation remained constant, while at the corresponding time in sulphate and glucose limited chemostats it was declining respectively had ceased.     

Oxygen transfer correlation in high cell density culture of recombinant E. coli

Summary A recombinant E. coli BL21[pET3a-T2M2] was cultivated in fed-batch cultures and cell mass increased to more than 70g/L. The volumetric oxygen transfer coefficient was estimated in a range of various fermentation parameters (agitation speed, oxygen flow rate and cell mass concentration) and finally the oxygen transfer correlation in bioreactor containing the recombinant E. coli cultures was determined as: k_spla = 0.0195 (P_g/V)^0.55 (V_s)^0.64 (1+2.12X+0.20X²)^–0.25.     

Xylitol formation by Candida boidinii in oxygen limited chemostat culture

Summary Production of xylitol by Candida boidinii NRRL Y-17213 occurs under conditions of an oxygen limitation. The extent to which substrate is converted to xylitol and its coproducts (ethanol, other polyols, acetic acid), and the relative flow rates of substrate to energetic and biosynthetic pathways is controlled by the degree of oxygen limitation.With decrease in oxygen concentration in the inlet gas, for a constant dilution rate of 0.05 1/h. the specific oxygen uptake rate decreased from 1.30 to 0.36 mmol/gh Xylitol was not produced at specific oxygen uptake rates above 0.91 mmol/gh. Upon shift to lower oxygen rates, specific xylitol production rate increased more rapidly than specific ethanol production rate:Nomenclature D dilution rate (1/h) - DOT dissolved oxygen tension (%) - m_o2 maintenance coefficient (mmol O₂/g cell mass h) - q_o2 specific oxygen uptake rate (mmol O₂/g cell mass h) - q_s specific xylose uptake rate (g xylose/g cell mass h) or (mmol xylose/g cell mass h) - q_x specific xylitol production rate (g xylitol/ g cell mass h) or (mmol xylitol/ g cell mass h) - q_e specific ethanol production rate (g ethanol/ g cell mass h) or (mmol ethanol/ g cell mass h) - q_CO2 specific carbon dioxide production rate (mmol CO₂/g cell mass h) - S xylose concentration (g/1) - Y_cm/s cell mass yield coefficient, (g cell mass/mmol xylose) or (g cell mass/ g xylose consumed) - Y_cm/O2 cell mass yield coefficient, (g cell mass/mmol O₂) - Y_X/S xylitol yield coefficient (g xylitol/g xylose consumed) - Y_x/O2 xylitol yield coefficient (g xylitol/mmol O₂) - Y_e/s ethanol yield coefficient (g ethanol/g xylose consumed) - OUR oxygen uptake rate (mmol O₂/1h) - specific growth rate (1/h)     

Dependency of growth yield,maintenance and K s-values on the dissolved oxygen concentration in continuous cultures of Azotobacter vinelandii

Azotobacter vinelandii was grown diazotrophically in sucrose-limited chemostat cultures at either 12, 48, 108, 144 or 192 M dissolved oxygen. Steady state protein levels and growth yield coefficients (Y) on sucrose increased with increasing dilution rate (D). Specific rate of sucrose consumption (q) increased in direct proportion to D. Maintenance coefficients (m) extrapolated from plots of q versus D, as well as from plots of 1/Y versus 1/D exhibited a nonlinear relationship to the dissolved oxygen concentration. Constant maximal theoretical growth yield coefficients (Y ^G) of 77.7 g cells per mol of sucrose consumed were extrapolated irrespective of differences in ambient oxygen concentration. For comparison, glucose-, as well as acetate-limited cultures were grown at 108 M oxygen. Fairly identical m- and Y ^G-values, when based on mol of substrate-carbon with glucose and sucrose grown cells, indicated that both substrates were used with the same efficiency. However, acetate-limited cultures showed significantly lower m- and, at comparable, D, higher Y-values than cultures limited by either sucrose or glucose. Substrate concentrations (K _s) required for half-maximal growth rates on sucrose were not constant, they increased when the ambient oxygen concentration was raised and, at a given oxygen concentration, when D was decreased. Since biomass levels varied in linear proportion to K _s these results are interpreted in terms of variable substrate uptake activity of the culture.Abbreviations D dilution rate - K _s substrate concentration required for half maximal growth rate - m maintenance coefficient - q specific rate of substrate consumption - Y growth yield coefficient - Y ^G maximum theoretical growth yield coefficient     

Technological properties of indigenous wine yeast strains isolated from wine production regions of Turkey

The purpose of this study was to evaluate the important technological and fermentative properties of wine yeast strains previously isolated from different wine producing regions of Turkey. The determination of the following important properties was made: growth at high temperatures; fermentative capability in the presence of high sugar concentration; fermentation rate; hydrogen sulfide production; killer activity; resistance to high ethanol and sulfur dioxide; foam production; and enzymatic profiles. Ten local wine yeast strains belonging to Saccharomyces, and one commercial active dry yeast as a reference strain were evaluated. Fermentation characteristics were evaluated in terms of kinetic parameters, including ethanol yield (Y_P/S), biomass yield (Y_X/S), theoretical ethanol yield (%), specific ethanol production rate (q_p; g/gh), specific glucose uptake rate (q_s; g/gh), and the substrate conversion (%). All tested strains were able to grow at 37 °C and to start fermentation at 30° Brix, and were resistant to high concentrations of sulfur dioxide. 60 % of the strains were weak H₂S producers, while the others produced high levels. Foam production was high, and no strains had killer activity. Six of the tested strains had the ability to grow and ferment at concentrations of 14 % ethanol. Except for one strain, all fermented most of the media sugars at a high rate, producing 11.0–12.4 % (v/v) ethanol. Although all but one strain had suitable characteristics for wine production, they possessed poor activities of glycosidase, esterase and proteinase enzymes of oenological interest. Nine of the ten local yeast strains were selected for their good oenological properties and their suitability as a wine starter culture.     

An elevation of the molar growth yield of Zymomonas mobilis during aerobic exponential growth

Elevated values of molar growth yield (Y_x/s = 14–26 g mol^–1) were obtained during exponential growth (μ > 0.4 h^–1) of Zymomonas mobilis ATCC 29191 by using reduced concentrations of glucose (6.25–100 mM) and increased oxygen supply (E _h > 300 mV) in the growth medium, as compared to the Y_x/s of anaerobic exponential growth (8–10 g mol^–1). Aerobically grown cells showed an increased maximum growth rate (μ_max), and a reduced specific glucose consumption rate (q_s), and specific ethanol formation rate (q_p), thus demonstrating a more pronounced energy-coupling growth under oxic conditions. These results can be neither explained by the concept of a solely operating Entner-Doudoroff pathway as an ATP source in aerobically growing cultures of Z. mobilis nor considered to be consistent with existing data on the lack of the Pasteur effect in this bacterium. Therefore, the results rather give evidence for the essential contribution of aerobic ATP generation under the reported conditions. Received: 24 September 1996 / Accepted: 9 December 1996     

Growth and energy production in Bacteroides amylophilus

The molar growth yield (Y _m) of Bacteroides amylophilus strain WP91 on maltose was 68±2 g/mol when determined from batch cultures at the peaks of maximal growth. Continued incubation led to considerable cell lysis. When calculated from batch cultures in exponential phase (specific growth rate, =0.57 h^-1) Y _m was 101 g/mol. The maximum value of Y _m in maltose-limited chemostat cultures at the maximum dilution rate (D) attainable (D==0.39 h^-1) was about 79 g/mol. Ammonia-Fmited chemostat cultures metabolized maltose with a much reduced efficiency and this was associated with a difference in morphology and chemical composition of the cells. The theoretical maximum molar growth yields (Y _m ^max ) were 55 and 114 g/mol for ammonia- and maltose-limited growth respectively. However, if account was taken of extracellular nitrogen-containing material in ammonia-limited cultures, Y _m ^max became 60. The maintenance coefficient (m _s), estimated from the lines relating the specific rate of maltose consumption (q _m) and D (where m _s=q _m at D=0), was 7.4±0.6×10^-4 mol maltose/g x h for both nutrient limitations. A difference in maintenance energy demand, independent of growth-rate, could not account, therefore, for the observed differences in Y _m between ammonia- and maltose-limited growth.     

Specific light uptake rates can enhance astaxanthin productivity in <Emphasis Type="Italic">Haematococcus lacustris</Emphasis>

Lumostatic operation was applied for efficient astaxanthin production in autotrophic Haematococcus lacustris cultures using 0.4-L bubble column photobioreactors. The lumostatic operation in this study was performed with three different specific light uptake rates (q _e) based on cell concentration, cell projection area, and fresh weight as one-, two- and three-dimensional characteristics values, respectively. The q _e value from the cell concentration (q _e1D) obtained was 13.5 × 10^?8 μE cell^?1 s^?1, and the maximum astaxanthin concentration was increased to 150 % compared to that of a control with constant light intensity. The other optimum q _e values by cell projection area (q _e2D) and fresh weight (q _e3D) were determined to be 195 μE m^?2 s^?1 and 10.5 μE g^?1 s^?1 for astaxanthin production, respectively. The maximum astaxanthin production from the lumostatic cultures using the parameters controlled by cell projection area (2D) and fresh weight (3D) also increased by 36 and 22 % over that of the controls, respectively. When comparing the optimal q _e values among the three different types, the lumostatic cultures using q _e based on fresh weight showed the highest astaxanthin productivity (22.8 mg L^?1 day^?1), which was a higher level than previously reported. The lumostatic operations reported here demonstrated that more efficient and effective astaxanthin production was obtained by H. lacustris than providing a constant light intensity, regardless of which parameter is used to calculate the specific light uptake rate.