Biokinetic parameters and behavior of Aeromonas hydrophila during anaerobic growth

Transcription

1 Biotechnol ett (28 3: DOI 1.17/s ORIGINA RESEARCH PAPER Biokinetic parameters and behavior of Aeromonas hydrophila during anaerobic growth Seungyeop ee Æ Jaai Kim Æ Seung Gu Shin Æ Seokhwan Hwang Received: 5 November 27 / Revised: 22 January 28 / Accepted: 25 January 28 / Published online: 12 February 28 Ó Springer Science+Business Media B.V. 28 Abstract The biokinetics of glucose metabolism were evaluated in Aeromonas hydrophila during growth in an anaerobic biosystem. After approx 34 h growth, A. hydrophila metabolized 5, mg glucose l into the end-products ethanol, acetate, succinate and formate. The maximum growth rate, l m, half saturation coefficients, K s, microbial yield coefficient, Y, cell mass decay rate coefficient, k d, and substrate inhibition coefficient, were.25 ±.3 h, 118 ± 31 mg glucose l,.12 lg DNA mg glucose,.1 h, and 3,18 ± 1,152 mg glucose l, respectively. These data were used to predict the performance of a continuous growth system with an influent glucose concentration of 5, mg l. Results of the analysis suggest that A. hydrophila will metabolize glucose at greater than 95% efficiency when hydraulic retention times (HRTs exceed 7 h, whereas the culture is at risk of washing out at an HRT of 6.7 h. Keywords Acidogenesis Aeromonas hydrophila Anaerobic digestion Biokinetics Inhibition S. ee J. Kim S. G. Shin S. Hwang (& School of Environmental Science and Engineering, Pohang University of Science and Technology, San 31, Hyoja-dong, Nam-gu, Pohang, Kyungbuk , Korea shwang@postech.ac.kr Introduction Acidogenesis is an anaerobic process in which microorganisms produce acid by metabolizing soluble sugars and amino acids in the absence of an electron acceptor or donor. The byproducts of this process include volatile fatty acids (VFA, which are major metabolic substrates for methanogens. Thus, the organisms involved in acidogenesis must be carefully monitored to enhance anaerobic digestion (Ruel et al. 22, Yang et al. 23. The kinetic model and data reported by the International Water Association (IWA task group have proven useful in manipulating this process (IWA Task Group for Mathematical Modeling of Anaerobic Digestion Processes 22; however, the IWA model uses the unit chemical oxygen demand (COD to refer to organic particulates and biomass, two variables that are difficult to distinguish using conventional experimental techniques. Furthermore, it is technically impossible to identify every microorganism responsible for acidogenesis in a particular system because many organisms aggregate as groups with similar parameters and are therefore unavoidably lumped together. The present study attempted to overcome these limitations by examining the growth kinetics and DNA concentration of a bacterial species known to produce large amounts of acid from carbohydrates. Aeromonas hydrophila is a facultative chemoorganotroph that is frequently isolated during the acidogenic phase of

2 112 Biotechnol ett (28 3: anaerobic digestion in wastewater. (Joubert and Britz Recent studies suggest that this species is enriched during the initial phase of acidogenesis from any soluble carbohydrates (Kim 27. The potential of A. hydrophila to cause human disease has sparked further interest in characterizing the species, but little is known about the organism s growth kinetics and metabolic end products, two variables that are needed for the construction of biological models. Therefore, this study evaluated the species-specific growth parameters of A. hydrophila by analyzing bacterial DNA concentration and identifying the acidogenic end products of glucose metabolism. Materials and methods Biokinetic models The substrate inhibition model (i.e., Andrew s model was used to describe the growth of A. hydrophila (Grady et al The mathematic expressions are given as: X t ¼ X e lt l l ¼ m S K s þ S þ S2 dx dt ¼ ds dt ¼ 1 Y l m S ÞX K s þ S þ S2 k d l m S X K s þ S þ K S2 si ð1þ ð2þ ð3þ ð4þ where X t is the microbial concentration in the bioreactor at time t (microbial mass volume, X o is the initial microbial concentration (microbial mass volume, l is the specific microbial growth rate (time - 1, t is the incubation time (time, l m is the maximum specific growth rate if the substrate is not inhibitory (time, S is the residual substrate concentration (mass substrate volume, K s is the half-saturation coefficient equivalent to the substrate concentration at which the specific growth rate is half of its maximum (mass substrate volume, is the substrate inhibition coefficient (mass substrate volume, k d is the specific microorganism decay rate (time, and Y is the microbial yield coefficient (microbial mass (mass substrate used. Equations 1 and 2 were used to evaluate the kinetic parameters of l m, K s, and by fitting the specific growth rate (i.e., l to the initial substrate concentrations. Equations 3 and 4 were then used to approximate Y and k d, using the 4th-order Runge-Kutta method. Experimental conditions Aeromonas hydrophila was obtained from the Korean Collection for Type Cultures (KCTC and grown in an anaerobic chamber (Concept 4, Ruskinn Technology td. on media described in the KCTC instructions. Cell suspensions were collected at the onset of turbid growth and used as inocula. The following glucose-containing growth medium was prepared to give ratios of COD:nitrogen:phosphorous of 16:8:1 and a total COD of 5, mg l : glucose (4,868, yeast extract (5, NH 4 Cl (955.5, KH 2 PO4 (63.6, KH 2 PO 4 3H 2 O (, NaCl (6, KCl (185, MgSO 4 7H 2 O (.6, NTA (4., CaCl 2 2H 2 O (2, FeCl 3 6H 2 O (.1, MnCl 2 4H 2 O (.9, H 3 BO 3 (.2, CoCl 2 6H 2 O (1.5, CuCl 2 2H 2 O (2.2, NiCl2 6H 2 O (1.2, Na 2 SeO 3, 45% (.6, ZnCl 2 (.9, and citric acidh 2 O (15. An automated turbidity reader equipped with 256 simultaneous microcultivation units (Bioscreen C, absystems was used to generate microbial-specific growth rate data. The growth medium described above was used as the substrate. The base medium was autoclaved at 121 C for 15 min prior to the addition of filter-sterilized glucose. The initial ph of the culture medium was adjusted to 7. using 3 M NaOH. Each microcultivation unit was seeded with a small inoculum of microorganisms at 1% of the total culture volume, and bacteria were incubated at 35 C under anaerobic conditions. All experiments were replicated five times at each glucose concentration. A specific microbial growth rate, l, was estimated for each initial substrate concentration using Eq. 1. The calculated growth rates were then plotted against initial glucose concentrations to determine an overall relationship. The data were fitted to the plot using Eq. 2, with leastsquared errors at a 95% confidence interval (C.I.. The biokinetic parameters in Eq. 2 were estimated using the Marquardt evenberg algorithm (Sigma- Plot Ò 2, SPSS. Data used in the ordinary differential equation model were obtained using batch cultures grown in

3 Biotechnol ett (28 3: automated anaerobic bioreactors. The bioreactors contained working volumes of 4 l and were maintained at ph 7. by the addition of 3 M NaOH. Growth temperatures, culture media, sterilization steps, inocula, and seeding ratios were identical to those used in the microcultivation experiment. Samples were taken periodically to measure the concentrations of DNA, residual substrate, and metabolic products. The resulting data were compared with the solutions of Eqs. 3 and 4, thereby optimizing the parameters. Analytical methods Soluble total carbohydrate (STC concentrations were measured using the phenol/sulfuric acid method. Concentrations of VFAs were determined using a gas chromatograph equipped with an Innowax capillary column and a flame ionization detector. Helium was used as a carrier gas 2.5 ml min with a split ratio of 1:1. The concentrations of anions, including sulfate, were measured using an ion chromatograph equipped with a Metrosep A Supp 5 column (Metrohm. A mixture of 1.1 mm Na 2 CO 3 and 1.8 mm NaHCO 3 was used as an eluent at.7 ml min. The concentrations of formate, lactate, and succinate were determined using an identical ion chromatograph equipped with a PRP-X3 ion exclusion column (Hamilton, Reno using 1.5 mm HClO 4 as eluent at 1 ml min. DNA was extracted from bacteria grown in the bioreactor as follows. Appropriate dilutions were performed to ensure that the concentration of volatile suspended solids (VSSs was lower than 1.5 g l,in accordance with established guidelines (APHA- AWWA-WEF 25. Cells from each 1 ml sample were harvested by centrifugation at 1,g for 5 min and supernatants were discarded. The residual pellets were resuspended in 1 ml deionized and distilled water (DDW and centrifuged again to remove residual culture medium. The supernatants were carefully removed and pellets were resuspended in 1 ll of DDW. A fully automated nucleic acid extractor (Magtration System 6GC, PSS with magnetic bead technology was used to extract and purify genomic DNA from the bacterial samples. Genomic DNA concentrations were measured using a fluorometer (TD-7, Turner Designs with the PicoGreen dsdna quantification reagent (Molecular Probes, Inc.. Genomic DNA extracted from the samples was kept at -2 C until further use. All analyses were duplicated, and the results were presented as mean values. Results and discussion Production of VFAs as a result of glucose metabolism in batch culture Figure 1 shows temporal changes in concentrations of VFAs and glucose. Glucose was almost completely utilized by A. hydrophila after 34 h. Concentrations of the metabolic end-products ethanol, acetate, succinate, and formate were determined to be 1,432 ± 1, 1,252 ± 8, 682 ± 3, and 416 ± 5mgl as COD, respectively. The COD of major products, with an additional 71 ± 1mgl of residual COD from glucose, added up to 3,85 mg l, which represents about 93.3% of the soluble COD measured (4,125 ± 17 mg l, not shown. Thus, all of the known end products of acidogenesis were detected and measured in this study. Acetate and formate are major sources of methane: 7% of methane generated by anaerobic digestion is derived from acetate, while 15% is derived from formate, H 2, and CO 2 (Forday and Greenfield Ethanol and succinate, the key intermediates prior to COD as carbyhydrates Soluble Soluble carbohydrate Ethanol Acetate Formate Succinate Time (hours Fig. 1 Temporal changes in substrate concentration (soluble carbohydrates and formation of the metabolic end products of glucose fermentation (volatile fatty acids, ethanol, and succinate by Aeromonas hydrophila. Batch cultures were grown in 4 l media at 35 C, ph 7. Bars represent standard deviations COD as Succinate Ethanol, acids, fatty Volatile

4 114 Biotechnol ett (28 3: propionate, can also serve as substrates for methonogenesis via conversion to acetate and hydrogen. (a.2 Monod model (R 2 =.86 Growth kinetics of A. hydrophila in media containing glucose growth rate (hr.15.1 Inhibition model (R 2 =.97 Figure 2a shows the overall growth kinetics (i.e., l of A. hydrophila as a function of initial glucose concentration. Among the five replicates of each trial at a given glucose concentration, the highest and lowest values were excluded and the specific growth rate was estimated by averaging the three remaining l values. The use of different models (i.e., the Monod model versus the inhibition model changed the appearance of the curve. A Monod equation that excluded the inhibition term in Eq. 2 was initially used to assess the data; however, the specific growth rate did not follow Monod kinetics. In accordance with the substrate inhibition model, the specific growth rate increased with glucose concentration to a maximum before decreasing with a further increase in glucose concentration. The data were subsequently aligned to the inhibition model using a non-linear, least-squared method, which generated the following values at a 95% C.I.: l max =.25 ±.3 h ; K s = 118 ± 31 mg glucose l ; = 3,18 ± 1,152 mg glucose l. Figure 2b shows temporal changes in microbial DNA and substrate concentrations in batch kinetic experiments. These data were used to estimate the other biokinetic coefficients, Y and k d in Eqs. 3 and 4, by numerical optimization. The results of optimization using the Marquardt-evenberg algorithm were Y =.12 lg DNA mg glucose and k d =.1 h. The model prediction with parameters optimized is shown in the Fig. 2b with solid lines. A t-test did not detect a difference between the experimental values and the model predictions at an a level of.5. The biokinetic values identified in this study are within the range of the parameters previously reviewed by the IWA task group (IWA Task Group for Mathematical Modeling of Anaerobic Digestion Processes 22. The IWA group found the kinetic parameters for glucose acidogenesis to be between.41 and d for l max, and between.22 and.63 g COD l for K s ; however, the parameter k d is not addressed in the reference and data for Y are expressed as COD COD, which cannot be compared Specific (b concentration Glucose Initial glucose concentration Data used in kinetic analysis Time (hr Fig. 2 (a Observed and predicted specific growth rates of Aeromonas hydrophila as a function of initial glucose concentration. Batch cultures were grown at 35 C, ph 7. Bars represent the standard error of triplications: ( experimental specific growth rate, (- - - Monod model, (- inhibition model. (b Temporal changes in concentrations of glucose concentration and A. hydrophila DNA. Batch cultures were grown at 35 C, ph 7, in an anaerobic bioreactor with a working volume of 4 l: ( observed cell mass concentration, (s observed glucose concentration, ( numerical model prediction with data generated in the present study. It may still be possible to use DNA concentration as biomass to determine information about the variables l max and K s. Predicting the performance of a continuous culture system The kinetic parameters evaluated in this study can be used to predict the performance of a continuous culture of A. hydrophila. The dynamic state biokinetic expressions with substrate inhibition were used to predict the changes in microbial and substrate concentrations in a continuously stirred tank reactor (CSTR at various hydraulic retention times (HRTs, which are given as: D NA concentration ( µ g

5 Biotechnol ett (28 3: ! dx dt ¼ l m S k d 1 X ð5þ K s þ S þ S2 s ds dt ¼ S i S 1 l m S X ð6þ s Y K s þ S þ S2 In these equations, s is the hydraulic retention time (time and S i is the influent substrate concentration (mass substrate volume. The initial glucose and microbial concentrations are assumed to be 5, mg l and 5 lg DNA l, respectively, and the influent glucose concentration is assumed to be 5, mg l. Figure 3 shows temporal changes in the (a residual glucose concentration and (b microbial DNA concentration of A. hydrophila, and depicts steady states at different HRTs. Conditions of steady state were assumed when DNA and residual glucose concentrations varied by less than.5%. The residual carbohydrate concentrations decreased and thereafter remained steady, except when the HRT was less than 6.7 h. Based on these data, the estimated efficiency of glucose metabolism would be greater than 95% at HRTs longer than 7 h. The DNA concentration of A. hydrophila rapidly increased to different maximum values and gradually decreased to steady values at 38 and 43 h after start-up for HRT of 7 and 1 h, respectively. Aeromonas hydrophila cultures are predicted to wash out at an HRT of 6.7 h, resulting in a continuous increase in the residual glucose concentration to the initial concentration of 5, mg l. This is close to the washout point of 5.9 h HRT based on calculations using the apparent maximum growth rate. The unstable behavior of the culture at 7 h HRT indicates that a very small change in flow rate near the washout point could have a drastic impact on the system s performance. These are typical patterns of substrate utilization and microbial growth in CSTR operations, as reported in previous studies. Conclusion The growth kinetics of carbohydrate metabolism by A. hydrophila were determined by assessing specific growth rates in regard to initial substrate concentrations, and by solving a series of differential equations More than 9% of the metabolic end products of (a concentration glucose Residual (b D NA concentration ( µ g Time (hours Time (hours Fig. 3 Residual concentrations of (a glucose and (b Aeromonas hydrophila DNA as a function of time. Arrows indicate different HRTs acidogenesis from glucose were identified in this study. Finally, data from the study were used to simulate the behavior of a continuous growth system. The kinetic parameters estimated by monitoring DNA concentrations are consistent with those described in previous reports. DNA concentration is a useful way to distinguish actual biomass concentration from other particulates found at sites of anaerobic digestion. Moreover, analysis of DNA samples by real-time quantitative PCR could identify specific species or groups of microorganisms within the biomass (Yu et al. 25. The biokinetic variables measured in this study can be used in the future to construct models of anaerobic digestion, especially based on specific species such as A. hydrophila. Acknowledgements This research was supported in part by the Brain Korea 21 program, the Advanced Environmental Biotechnology Research Center (AEBRC; Grant No: R11-

6 116 Biotechnol ett (28 3: , and the Ministry of Environment as The Eco-Technopia 21 Project. References APHA-AWWA-WEF (25 Standard methods for the examination of water and wastewater, 21th edn. American Public Health Association, Washington, DC Forday W, Greenfield PF (1983 Anaerobic-digestion. Effluent Water Treat J 23: Grady CP Jr, Daigger GT, im HC (1999 Biological wastewater treatment, 2nd edn. Marcel Dekker, New York, NY IWA Task Group for Mathematical Modeling of Anaerobic Digestion Processes (22 Anaerobic digestion model no. 1 (ADM1. IWA, ondon Joubert WA, Britz TJ (1987 Characterization of aerobic, facultative anaerobic, and anaerobic bacteria in an acidogenic phase reactor and their metabolite formation. Microb Ecol 13: Kim J (27 Characterizing population dynamics in acidogenesis of various substrates by use of 16s rrna gene quantification: diversity and biokinetics. Ph.D. Thesis, School of Environmental Science and Engineering, Pohang, Korea, Pohang University of Science and Technology, 17 pp Ruel SM, Comeau Y, Ginestet P, Heduit A (22 Modeling acidogenic and sulfate-reducing processes for the determination of fermentable fractions in wastewater. Biotechnol Bioeng 8: Yang K, Yu Y, Hwang S (23 Selective optimization in thermophilic acidogenesis of cheese-whey wastewater to acetic and butyric acids: partial acidification and methanation. Water Res 37: Yu Y, ee C, Kim J, Hwang S (25 Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction. Biotechnol Bioeng 89:67 679