u. v! "17:” :m "(3.11,- , ..‘. :..u .u - J ~J~l m. .':{‘-14r."r;uvv-/ll .‘-.I -;. gr. ‘ir'nr'u awn"..- 4.2,.»- :1 .... Jun. ‘ .,. A ‘m "u, r u... ', . r: I "ll ' m .. .... ..’. .. ’ ". ...”)..q' a 1,, H's .}.. , .. .. . ..2'..;. ,.-. A ,. ...;... J... . .H . . aw}, mu :‘ n7 "1‘. ”.4va- "... . . . '4.,f.’..,:,.,.,_ , '1 1"!‘1' '7'..,..,.'..... ""'. 1-! 4 ”4-1., v: I r ‘-.'J',’;{ '..':"' ' é- ' ' ' u! ’ .9...” ,\ " ' ' '1' ...-.... ‘ " ”'w"? ' I’"'('I 11c ( " " " ‘\y“.-.. "RUM“ , ,2 ‘1‘ ) . 4 leL BIRAR IES ll \\ \\\\\\\\ \llllllllllllllllllll\\\\\\\\\\\\ \\\ This is to certify that the thesis entitled Kinetics of Growth and Catechol Production Bacillus stearothermophilus presented by Ramkumar Subramanian has been accepted towards fulfillment of the requirements for M.S. Chem. Engr. degree in WM WW4.» Major professor Date 2/24/92 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution _ tiamn‘r Michigan irate B a .‘ W _ {Tn ii, tg‘wflfli h‘i“ 3:41 a. :Q '16 up: .. '65:" 13% PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE Il MSU Is An Affirmative Action/Equal Opportunity Institution c:\cird\ddedm.pm3-o. 1 KINETICS OF GROWTH AND CATECHOL PRODUCTION BY BACILLUS STEAROTHERMOPHILUS By Ramkumar Subramanian A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE DEPARTMENT OF CHEMICAL ENGINEERING 1992 ABSTRACT KINETICS OF GROWTH AND CATECHOL PRODUCTION BY BACILLUS STEAROTHERMOPHILUS By Ramkumar Subramanian The purpose of this investigation was to study the growth and catechol production kinetics of Bacillus stearothermophilus. Bacillus stearothermophilus BR219, a phenol resistant thermophile, converts phenol to the specialty chemical catechol. In continuous culture studies on a dilute (DP) medium supplemented with 10 mM phenol, steady states were achieved between dilution rates of 0.25 and 1.3 hr'l. Phenol degradation was found to be uncoupled from growth. Bacillus stearothermophilus BR321, obtained by transposon mutagenesis of BR219, grows on DP medium supplemented with phenol and accumulates catechol. In batch culture, catechol concentrations up to 1 mM were obtained. A mathematical model was developed to describe batch growth for different initial phenol concentrations. In continuous culture experiments conducted at a dilution rate of 0.12 hr'l, steady states were obtained for feed phenol concentrations up to 0.25 mM, resulting in catechol concentrations, up to 0.1 mM. Above this phenol concentration, oscillatory dynamics were observed. .l .1 To my parents and sisters for their love, affection and support which make my accomplishments possible iv ACKNOWLEDGMENTS I wish to acknowledge the help of Dr. R. Mark Worden for all his support and guidance throughout the course of this research. I would like to thank Dr. Pat Oriel for all the useful suggestions he gave during the group meetings. I would also like to thank Dr. G. Gurujeyalakshmi and Dr. M. R. Natarajan for helping me learn microbiology laboratory techniques. Mike Bly helped me a lot, especially during the early phase of my research, and I would like to thank him for that. Finally, I would like to thank Bharath Rangarajan for the useful suggestions he gave during the development of the mathematical model. List of Tables List of Figures TABLE OF CONTENTS CHAPTER 1 Introduction CHAPTER 2 Background CHAPTER 3 Experimental Methods CHAPTER 4 Continuous Culture Experiments with Bacillus stearothermophilus BR219 CHAPTER 5 Batch Growth and Catechol Production Kinetics of Bacillus stearothermophilus BR321 CHAPTER 6 Continuous Culture Experiments with Bacillus stearothermophilus BR321 CHAPTER 7 Conclusions and Recommendations APPENDIX BIBLIOGRAPHY vi page vi 17 24 39 68 81 83 87 LIST OF TABLES Table page 4.1 Cell concentration and phenol degradation as a function of dilution rate 33 4.2 Specific growth and phenol degradation rates as a function of dilution rate 36 4.3 Outlet pH as a function of dilution rate 36 5.1 Phenol utilization and catechol production as a function of varying initial phenol concentration 44 6.1 Summary of steady state results from continuous culture with BR321 LIST OF FIGURES Figure 2.1 Meta pathway 2.2 Growth kinetics of BR219 2.3 Phenol degradation by BR219 2.4 Effect of tetracycline on catechol production in DP — 5 mM phenol 2.5 Effect of temperature on catechol production in DP - 5 mM phenol and 5 ug/mL tetracycline 2.6 Effect of phenol concentration on catechol production in DP - 5 mM phenol and 5 ug/mL tetracycline 3.1 Catechol assay 3.2 Phenol assay 4.1 Experimental setup 4.2 Unsteady state growth D = 0.25 4.3 Unsteady state growth D = 0.5 4.4 Unsteady state growth D = 1.00 4.5 Unsteady state growth D = 1.30 4.6 Cell and phenol concentrations as a function of dilution rate 4.7 Phenol degradation as a function of dilution rate 4.8 Specific growth rate and specific degradation rate as a function of dilution rate viii page 15 21 23 25 27 28 29 30 31 32 35 ix Figure 5.1 Cell concentration vs time in batch culture 5.2 Phenol concentration vs time in batch culture 5.3 Catechol concentration vs time in batch culture 5.4 Batch growth of BR321 in DP - 2.5 mM phenol with different concentrations of catechol present initially 5.5 Phenol concentration vs time when different concentrations of catechol are present initially 5.6 Catechol concentration vs time when different concentrations of catechol are present initially 5.7 Determination of u 5.8 Determination of kd 5.9 Cell death as a function of time for varying phenol concentrations 5.10 Determination of catechol inhibition constants 5.11 Cell concentration vs time in batch culture 5.12 Phenol concentration vs time in batch culture 5.13 Catechol concentration vs time in batch culture 5.14 Substrate concentration vs time in batch culture 5.15 Batch growth of BR321 in DP - 2.5 mM phenol with different concentrations of Catechol present initially 5.16 Phenol concentration vs time when different concentrations of catechol are present initially 5.17 Catechol concentration vs time when different concentrations of catechol are present initially 6.1 Unsteady state growth of BR321 with initial phenol concentration = 0.0 mM Page 41 42 43 46 47 48 53 54 55 56 59 6O 61 62 63 64 65 70 Figure 6.2 6.3 6.4 6.5 6.6 6.7 Unsteady state growth of BR321 with initial phenol concentration = 0.065 mM Unsteady state growth of BR321with initial phenol concentration = 0. 123 mM Unsteady state growth of BR321 with initial phenol concentration = 0. 239 mM Steady state cell concentration vs initial phenol concentration in continuous culture with BR321 Catechol concentration vs initial phenol concentration in continuous culture with BR321 Unsteady state growth of BR321 with initial phenol concentration = 0.36 mM Page 71 72 73 74 76 77 CHAPTER 1 INTRODUCTION The objective of this research was to study the growth and catechol (1,2-dihydroxybenzene) production kinetics of Bacillus stearothermophilus. This bioconversion could serve as an alternative method for producing catechol, i.e. bioconversion of phenol, as compared to chemical processes like o-chlorophenol oxidation and phenol hydroxylation using hydrogen peroxide, which are used in industry. Bacillus stearothermophilus BR219 degrades phenol via the meta pathway (Buswell, 1974; Adams and Ribbons, 1988) completely to carbon dioxide and water with catechol as the first intermediate in the process. Addition of tetracycline blocks this pathway by inhibiting catechol 2,3-dioxygenase, the subsequent enzyme in this process, resulting in the accumulation of catechol (Gurujeyalakshmi and Oriel, 1988a) Bacillus stearothermophilus—BR321 is a mutant developed from BR219 that contains transposon Tn916 inserted in the catechol 2,3- dioxygenase gene. This mutation prevents the formation of active catechol 2,3-dioxygenase and thus causes catechol accumulation without the addition of tetracycline (Natarajan and Oriel, 1992). When this research began, only BR219 had been isolated. The research plan was to study and optimize the growth rate of BR219 and subsequently 2 optimize catechol production after adding tetracycline. Batch experiments had been conducted with BR219 by Dr. Gurujeyalaksmi and Dr. Oriel (1988a-b,1989) and by Stefan Winter and Dr. R. M. Worden (Worden et al. 1991). Once isolated, strain BR321 was substituted for BR219 because BR321 did not require the addition of tetracycline for catechol accumulation. With no antibiotic, it became possible to produce catechol while the cells were growing, thus making continuous reactor operation possible. In addition, the cost of tetracycline was eliminated. This thesis consists of experiments conducted with both BR219 and BR321. It has been divided into the following chapters: 1) Introduction 2) Background 3) Experimental Methods 4) Continuous Culture Experiments with BR219 5) Batch Experiments with BR321 6) Continuous Culture Experiments with BR321 7) Conclusions and Recommendations Chapter 2 summarizes the available literature pertinent to this research. Chapter 3 describes the experimental methods used in the course of this research, including fermentation protocols and techniques for measuring cell density and viability, and catechol and phenol concentrations. Chapter 4 deals with continuous culture experiments using the 3 parent strain (BR219). The experimental setup, procedure and data analysis are described in detail. Chapter 5 describes batch experiments conducted to analyze the catechol producing capabilities of BR321. The experimental setup and procedure are described in detail. A mathematical model was developed to describe changes in concentrations of viable cells, rate-limiting substrate, phenol, and catechol during batch growth. The details of this model are discussed in Chapter 5. Chapter 6 describes continuous culture experiments conducted using BR321. This experiment reflects the potential industrial feasibility of the bioprocess, as it demonstrates continuous production of catechol. Finally in Chapter 7, the conclusions drawn from this research are outlined. Ongoing research is also briefly described, and future plans and possibilities are given. CHAPTER 2 BACKGROUND In this chapter, the background information pertinent to this research is discussed. Topics include the current methods of production and uses of catechol, the advantages of using thermophilic bacteria, catechol metabolism, and growth studies of BR219 conducted by other researchers. Catechol Catechol, also known as pyrocatechol or 1,2-dihydroxybenzene or 1,2-benzenediol, is a specialty chemical used in industry in fur dyeing, leather tanning, photographic applications, and as a polymer intermediate (Kirk and Othmer, 1981). It is a colorless, crystalline compound with a phenolic odor and a sweet and bitter taste (ibid). Its main use is in the photography industry, where it is used to reduce the exposed silver halide grains in a photographic emulsion at a faster rate than the unexposed grains and provide a very fine grain and high contrast images (ibid). Catechol was first obtained in 1839 by dry distillation of catechin (ibid). In the early 20th century, catechol was manufactured by the hydrolysis of o-chlorophenol, as shown below. on ' CI 0“ 0 °" + OH‘ " (- + C“ o-chlorophen01 catechol Currently, there are no plants in the world producingcatechol by this method. Phenol hydroxylation, using hydrogen peroxide, as shown below, is used by most of the major producers of catechol. 63mg: 6335+ (.5... ' phenol Md ' hydmZumone About 78 % of the phenol is converted to dihydroxybenzenes, with every mole of phenol yielding 0.44 moles of catechol (Varagnat, 1976). Some plants use coal tar distillation to produce catechol. A detailed discussion of these production methods is given in the WW edited by Kirk and Othmer (1981). Annual production of catechol in the world is around 20,000 tonnes (Kirk and Othmer, 1981). The current price of catechol is $2.80/lb for purchases up to 22,000 pounds and $2.10/lb for more than 22,000 lbs (Chem. Marketing Rep. 1991a). Phenol currently costs $0.28/lb (Chem. Marketing Rep. 19910). Catechol’s utility in industry 6 is increasing, and its market is expanding. Any new technology for producing catechol will have to compete successfully against the existing technologies. The advantages of biochemical production of catechol are that phenol is a cheap substrate, the catalysts required for the reaction are inexpensive (bacteria), the operating temperature is less than the chemical process (55°C as compared to 80°C), and phenol is the only other aromatic compound in the product stream. In the chemical process, hydroquinone and small amounts of resorcinol are also formed, and successive distillations are required to get crude catechol. Hydrogen peroxide, which costs $0.25/1b (Chem. Marketing Rep., 1991b), has to be used on an equimolar basis in the chemical process for oxidation. These advantages for the biochemical process suggest that it may compete successfully with existing processes. Catechol is more easily oxidized than phenol (Kirk and Othmer, 1981). It can be converted by most oxidizing agents to its ortho and para benzoquinones, as shown below. OH 0- O I .03 I o H Kim ‘ .____,_ _._ QT' catechol 3,5-cyclohexadiene O. 4'24”” \ hydroxybenzoduinone o . no I We“ 0 dacisomueonie acid Waters (1964) discusses the the autoxidation of catechol when 7 exposed to atmospheric oxygen. Catechol first oxidizes to a mesomeric radical, which eventually oxidizes to the stable yellow quinone. The autoxidation process is shown below: cc:- :1 (1 oh: catechol 11150de . radical dim (Lg: ; (M): i dimer @3133- ;- cm: 'I (1.3:): yellow quinonc High pH values and high temperatures increase the autoxidation rate of catechol. Thermophilic bacteria Bailey and Ollis (1985b) classify thermophiles as those microorganisms that can grow in temperatures'ranging from 40 to 80°C, with the optimum temperature between 55 and 75°C. Sonnleitner and Fiechter (1983) describe thermophilic microorganisms as those that can grow and form products at temperatures above 65°C. The advantages of thermophilic bacteria in biotechnology have been discussed by Sonnleitner and Fiechter (1983) and Sbnnleitner (1983), and include increased productivity due to increased reaction rates due 8 to the higher temperature, reduced microbial contamination problems because of the higher temperature, ability to control the reaction rate by cooling alone, resistance to chemical denaturation and higher tolerance to products. It has been reported that meSOphiles (bacteria that can grow at lower temperatures than thermophiles i.e. around 37°C) can degrade phenol; however, phenol is an inhibitory substrate (Gurujeyalaksmi and Oriel, 1988a). The first thermophile to utilize phenol was reported in 1974 by J. A. Buswell (1974). Catechol Metabolism Gurujeyalaksmi and Oriel isolated a strain of Bacillus stearothermophilus (BR219) from river sediment that possesses the highest phenol tolerance reported to date (Gurujeyalaksmi and Oriel, 1988a). This strain can grow in phenol concentrations of 15 mM with optimal growth occurring at 10 mM (ibid). Phenol degradation was found to occur via the meta pathway (Buswell, 1974) (Fig. 2.1), as shown by the formation of catechol and 2-hydroxy muconic semialdehyde, the first two products in this pathway. Buswell and Twomey (1975) had also reported the utilization of phenols and cresols by Bacillus stearothermophilus, but their strain was inhibited by phenol at concentrations above 5 mM. The details of the metabolism were given in another paper by Buswell (1975) including the first few steps of the meta degradation pathway. Further steps in this pathway are discussed by Adams and Ribbons (1988). Dagley (1971) has discussed the catabolism of aromatic compounds by microorganisms in detail. Gurujeyalaksmi and Oriel studied the accumulation of catechol 1 2 4 5 I ) ) v phenol hydroxylase g. I catechol 2,3 - dioxygenase égxgw .otr .. .. (I): Cadmium. 'rCA Cycle 2 hydroxymuconic aemialdehyde A) 2 hydroxymuconic aemialdehyde dehydrogenase 4 oxaloerotonate (enol) B) 4 oxalocrotonate tautomeraae 4 oxalocrotonate (keto) C) 4 oxalocrotonate decarboxylase 2 oxopent 4 enoate D) 2 hydroxy 6 oxohepta 2,4 dienoate hydrolase 4 hydroxy 2 oxalovalerate - E) 2 oxopent 4 enoate hydrataae F) 4 hydroxy 2 oxovalarate aldolase Fig 2.1 Meta pathway (Egon Buswell (1975) and Adams-and Ribbons (1988)) 10 by Bacillus stearothermophilus BR219. They reported that addition of tetracycline inhibited the enzyme catechol 2,3-dioxygenase, which oxidizes catechol, resulting in the accumulation of catechol (1989). They also reported that the optimal concentration of tetracycline for inhibition to occur was 5 ug/mL at 55°C. They further said that the highest conversion of phenol to catechol occurred at 5 mM phenol (16% phenol conversion), while the highest concentration of catechol was obtained when 10 mM phenol was used as substrate (Figs. 2.2-2.6). For low phenol concentrations, the cells began to die immediately after a key nutrient was depleted, i.e. there was no stationary phase (Fig. 2.2). However, when higher phenol concentrations were used, a small stationary phase occurred before death of cells began. Typical maximum optical densities (650 nm) obtained with the dilute media were around 0.35. Gurujeyalaksmi and Oriel (1988a) measured doubling times of around 100 minutes with the first isolates of the bacterium. Kinetic studies of this bacteria were conducted by Stefan Winter under the guidance of Dr. R. M. Worden. His research focused on determining the optimal concentration of phenol required for the bioconversion process, the effect of pH on growth and process control of this fermentation. They showed that pH does not affect growth in the range of 6-8, and minimal growth was observed at pH 5.5 (Worden et al., 1991). Typical maximum optical densities (650 nm) obtained in rich media were around 2.8. Gurujeyalakshmi and Oriel (1988b) ‘reported that when tetracycline inhibited the phenol degradation pathway, the accumulation of catechol was higher for strains of BR219 carrying a conjugating transposon Tn916. Subsequently Drs. Natarajan and Oriel (1992), om Emma: E5 was _Ecmxo_o>m_2:o Eoi 2.2m Co 8:8: £280 3. .5 Ev 2:: om .9 0* on . _ . 11 EEONH 2E9" 2E9" .zEmu EEO" .88 .82: 62c. 80 .28 3:2”. .025 are .88 6:2: egg mm .88 3:2: 625 0.0 .88 saga egg I om (Silun new) HlMOJO 12 Emma; Bro vco _Ecmxo_o>o?.50 80¢ mpmmm E cozovoumov Becca ON .9... C5 mEP on om on 0+. on om 2 o _ _> _ . _ . _ . _ _ _ _ _ _ O .4 n. v 1N— m w ... ... ... ... ... . ON :5 om u .28 3:2: 655 «Le - 28.9 n 6:00 Becca BEE qu ...¢N :8 OF u .280 .055. BEE Ono . 26 m u 6:00 6:25 BEE : H (ww) uogioJiueouoo loueqd 13 883 8:0 uco _meo_o>o_E:Q Sci 65% 2E m I do 5 cozoanoa 6:038 co oczoxoobmo. .6 «8.5 YN .mE. EV 2:: em on Na /\ méoxoobfi ..E\oio— DID looF oczoxoobfi ._E\o.:.m i 1 05350.33 4E\mi¢ GIG LONF oc=o>oob3 4E\min Elm. 056.8953 DEBS; 0:0 056.88%: oz I 10: (mu/671) °ou03 quoeioo l4 88; go 95 Esmxgoasso Eeu 0E6>00b8 ._E\mim nco 6:96 28 m I do 5 6302006 6,6060 :0 838388 6 60.5 ON .9... ON AEV 08$ @— _ N— _ 0000 $18 900 GIG 9.00 I (mu/617’) 'ouog |oqoeioQ 15 Ammo—v 6:0 0:0 _E;mv_o6>0_830 Eoi 0:630:00. ..E\o.:.m use 6.66 I an E 6502006 650060 co cozobcoocoo 6:26 6 62m. QN .m_.._ O3 0:5 mm . .VN ON m: N, m 4 o _ .m- _ m. . _ . _ . ...F I P H , .nio 10¢ low IONF 38,: :5 om ale .0? .82: :6 2 mm . .22: 22 S 0.0 - .825 22 m xix . CON (qty/671) ‘ouoo loqoaiog 16 successfully mutagenized BR219 with transposon Tn916 to yield a mutant (BR321) that accumulated catechol throughout the growth phase. Thus, with BR321 the tetracycline addition step could be eliminated in the catechol production process. CHAPTER 3 EXPERIMENTAL METHODS Introduction In this chapter the techniques used in the experiments are described in detail. These techniques include measurements of cell density and viability, catechol concentration, and phenol concentration. The methods for making LB-Broth and DP media, and LB and DP plates are also described. Media used 1) LB-Broth (Maniatis et al., 1982) This contains per liter 10 g tryptone (Difco, Detroit, Michigan) 5g yeast extract (Difco, Detroit, Michigan) 5g sodium chloride 2) DP medium (Buswell, 1974) This contains per liter 1 g ammonium chloride 0.5 g potassium phosphate, dibasic 0.2 g yeast extract (Difco, Detroit, Michigan) 17 18 0.1 g casamino acids (Difco, Detroit, Michigan) 0.02 g magnesium sulphate 1 mL trace element solution, which contains per liter 500 ug boric acid 40 ug copper sulphate 100 ug potassium iodide 200 ug ferric chloride 400 ug manganous sulphate aliquot of phenol (for desired concentration) For LB or DP media, the different components were added to reverse osmosis (RO) water to a final volume of 1 L and dissolved by constant stirring. The pH was adjusted to 7.2 using small amounts of hydrochloric acid or sodium hydroxide. The resulting medium was then autoclaved as described below and then stored at room temperature. Sterilization The media were sterilized at 121°C using 15 psig steam in a Castle sterilizer (Wilmot Castle company, New York) for 15 minutes for small quantities of media (250 mL) and 45 minutes for large quantities of media (12 L). LB and DP plates The media (LB or DP) were made up as described earlier. Before adjusting the final volume to 1 L, 10 g of agar or gelrite and 1 g of magnesium chloride were added. The solution was then sterilized as 19 described earlier. After cooling to 40 - 50°C, the solution was poured into the plates. The plates were left at room temperature for a day 'to check for contamination and then stored inverted at 4°C. Cell Density The cell density was measured by turbidity at 650 nm using a Perkin Elmer (Norwalk, Connecticut) Lambda 3A spectrophotometer with R0 water as the blank. When the optical density was above 0.8, the sample was quantitatively diluted with DP or LB medium (as needed) for better accuracy. Colony Counts The cell viability was measured by plating a known volume of cell broth on LB plates, after asuitable dilution to give 30 - 300 colonies per plate, and then incubating the plates in an oven at 55°C (Boyd, 1988). Cell Dry Weight To measure the cell concentration based on dry weight, 25 mL of the broth was withdrawn from the reactor with a volumetric pipette, and centrifuged for 10 minutes at 5900g. The supernatant was carefully removed using a pipette. Twenty-five mL of sterile water was added, and the cells were resuspended using a vortex machine. The cells were centrifuged again, and once again, the supernatant removed. Finally, the cells were transferred quantitatively to a pre-dried, pre-weighed, aluminum pan and dried for 20 hours at 100°C. The pan was cooled in a desiccator and weighed. The difference in the pan weight with and 20 without the cells gave the dry weight of cells. Catechol Assay Catechol was measured as described by Irie et al. (1987) with some minor modifications. A reagent solution was made that contained 0.1 % by weight 4-amino antipyrene 2 % by weight sodium carbonate 0.02 N sodium hydroxide An appropriate amount of sample (to obtain the desired final concentration) was mixed with 3 mL of the reagent solution. After sitting for 20 minutes, the absorbance was measured at 515 nm with the reagent solution as the blank. The reagent solution is colorless at 515 nm and can be replaced by RC water as the blank. Fig. 3.1 shows a typical calibration curve for this assay. This assay was repeated many times to verify its reproducibility and accuracy. The catechol assay is highly linear in the range of 0.0001 to 0.012 g/L (coefficient of linearity - 0.9997). Phenol assay Phenol was measured using a technique described by R. W. Martin (1949). Eight mL of R0 water, 240 uL of a 2 % by weight solution of 4oamino antipyrene, 800 uL of a 2 N ammonium hydroxide solution, 800 uL of a 2% by weight potassium ferricyanide solution, and a small aliquot of sample containing phenol were added to a test tube. The solution was well mixed and allowed to react for 30 minutes, 21 .892 .2038 3 .90 A._\0v 0:00 65000.00 N 0.0 0 90.0. 000.0 000.0 .000 «00.0 000.0 . _ . _ . _ . _ . _ . IO.O M. \ \ \\ \ ..Nd \\ \ .. \\ \ 14.0 \ \\ \ \\ . 10.0 \ \ . m\ 0.0 WU 919 'Cl'O 22 before the absorbance was measured at 510 nm, using the same mixture without the sample as the blank. Figure 3.2 shows a typical calibration curve for this assay. This assay is highly linear up to phenol concentrations of 0.005 g/L. These assays were calibrated before each set of experiments to verify the existing calibration. The difference in the calibration curves obtained was always less than 2%. 23 000.0 82 .80.: a...” .9... A._\0v 0:00 6:050 00.0.0 . 00.0.0 . No.00 . «0.0.0 . 000.m.0 \» - \ \ -md \ x\ . \ \ [0.0 \ \ - \\ 10.0 \ . \ \\ 10.0 .\ . 0.0 WU 019 'CI'O CHAPTER 4 CONTINUOUS CULTURE EXPERIMENTS WITH BACILLUS STEAROTHERMOPHILUS BR219 Introduction The main objective of these experiments was to study the growth of Bacillus stearothermophilus BR219 in continuous culture. Batch growth studies, conducted in both rich and dilute media (Gurujeyalakshmi and Oriel, 1988a-b,1989; Worden et al, 1991), have been discussed in Chapter 2. Continuous culture experiments were performed to better understand the growth and phenol degradation kinetics of this organism . Materials and Methods The bacterium Bacillus stearothermophilus BR219 was obtained from the laboratory of Dr. P. Oriel. The experiments were conducted in a 1.3 liter Bioflo II - New Brunswick (Edison, New Jersey) fermentor . The experimental setup is shown in Fig. 4.1. The detailed procedure is given below. The inoculum was prepared by transferring a single colony from a LB plate of BR219 to a sterile flask containing 100 mL of sterile LB- Broth and incubating for 13 -17 hours at 55°C with shaking at 200 rpm. 24 25 “upon Hmucoaduomxu H.v mam .03? .53? . 50>; 523. , . mama . gnaw i 2.52; _..A 55: , m a m L _ I. n T. _ Illll . m g 9.8 5:55 «:33 5:33 55... m 0.32.2: .2250 .92. 6.5: AVI 3.2.. .94.! zmhmgofi 26 The feed carboy, connecting tubes, overflow vessel, and the fermentor vessel were sterilized and connected aseptically. The reactor was filled using a pump and then heated to 55°C. Air sparging at 3 L/min and agitation at 100 rpm were started. A 5 % by volume inoculum was injected into the reactor, using a sterile syringe and needle. Once growth began, continuous feed was started. The reactor was monitored occasionally to see if steady state had occurred. This was done by measuring the turbidity at 650 nm as described in Chapter 3. If the cell density remained the same for 3 residence times, steady state was assumed. At steady state, the cell turbidity, the cell dry weight, and the inlet and outlet phenol concentrations were measured. The pump was then adjusted to increase the liquid flow rate, thus increasing the dilution rate. The dilution rate was varied from 0.25 to 1.5 hr]. Phenol concentration in the feed was kept steady at 10 mM. Sterile conditions were maintained for the most part; however, due to the high concentration of phenol and the high temperature involved, aseptic operation may not have been necessary. Results Cell-concentration profiles leading to each of the steady states are plotted in Figs. 4.2-4.5. The steady state cell (X) and phenol (P) concentrations are plotted as a function of the dilution rate (D) in Fig. 4.6. The corresponding steady state data are also tabulated in Table 4.1. The cell concentration was constant between dilution rates of 0.25 to 1.3 hr’l, while the phenol degradation varied from a high of 20 % at a dilution rate of 0.25 hr‘1 to 5 % at a dilution rate of 1.3 hr'1(Fig. 4.7). 27 2.0. u o 555 33.... >832: we a: $532.3 “:2: 0000 00_0¢ 00.0w 00.0w. 00.0 P 0 . 00.0 - 100.0 ..0—0 [9.0 1:.0 3.0 (wu 099 0'0) woo “so 28 00¢F 00W— 3 u 0 555 33m €038: m... .0: 0&0— Ammsziv m2: 000 0mm p 0m¢ 00m - 0 Noav I¢0A0 100A0 Imogv 10—A0 NPAV (wu 099 0'0) °ouoo Inez) 29 005 P 00; 00.0 F - u 0 £265 3on .8082: we .0: $052.20 02: 000 _ 0mm 0m? 0mm _ (wu 099 00) mos H80 30 00m om; u g £55 2% €032: m... .90 80523 mi: 00m 00F 00F 00 0 n — — - _ p _ n p p - — p — p n — p P — — @000 (UULJ 099 0'0) °0u00 net) 31 (wu 099 0'0) woo Ileo 8.0.. 5:25 .6 cozoca 0 mo mcozobcoocoo 65%. use :00 0+ .0_.._ oi 0 BE 20:35 . f .92 mm on who who Jo A0 0.0 . oo o 0200 520:“. 92. I 88 o . 028 5sz 950 i 0200 ....GO GIG 8.0., _ 128.0 A 2.9. 0 0/0 . 188.0 9.? . .885 8.0.. . $ .| . mmi 111x 9.8 o and 88.0 (1/5) NouvalNHONoo 'lONZ-lHd 32 28 c0026 00 8:83 0 mo 50008000 .655. we .90 Ami: BE 20:35 .9; N... 0. — 0.0 0.0 9.0 N0 . — p _ _ _ n _ . .V peponep loueqd % 33 ' Because of the imprecision of the dry weight biomass assay at the low cell concentrations, the standard deviations of the dry weight measurements were very large (7.4 x 10'4 g/25 mL). Table 4.1: Cell concentration and phenol degradation as a function of dilution rate. Dilution rate Cell Concentration % Phenol degraded (hr’l) OD. 650 nm dry weight (g/25 mL) 0.25 0.113 0.009 20.1 0.5 0.104 0.007 8.5 1.0 0.105 0.007 5.8 1.3 0.110 - 5.2 1.5 wash out At low dilution rates, the reactor was fairly stable. Minor changes in the feed concentrations, i.e. when the carboys were changed, did not affect the cell concentration significantly. However as the dilution rate was increased, the reactor stability decreased. When the pump was switched from a dilution rate of 0.5 to 1 hr'1 and from 1 to 1.3 hr'l, sudden lag phases resulted which lasted for several hours. The onset of a lag phase caused the majority of the cells to be washed out within three residence times. When such a lag phase was observed, 34 the pump was switched off until growth became apparent again. The theory of chemostats, as discussed by Bailey and Ollis (1985a) and by Marison (1988) show the relation between growth rates and dilution rates. Mass balances on mass and phenol in the reactor assuming sterile feed and negligible cell death rate give dX/dt = -DX + “X (4.1) d? /dt=1)(1>in - Pom) -q x (4.2) out p At steady state, dX/dt = 0 and dPom/dt = 0 (4.3) Hence, p. = D (4.4) qp = D(Pin - Pong/X (4.5) where Pin and P out are inlet and outlet phenol concentrations. The specific growth rate (u), which is the rate of increase in cell concentration per unit of cell concentration, and the specific phenol degradation rate (qp), which is the rate of phenol consumed per unit of cell concentration, are shown in Fig. 4.8 as a function of the dilution rate. They are also shown in Table 4.2. 35 (NA) 3013 Hmoao 0010303 30.. 000020 00 000002 0 mo .30.. 000008000 208% 0:0 38 £380 0:6on 0.0 .90 0:3 020 20:30 m: 3 N; 0. _ 0.0 0.0 .90 0.0 0.0 0.0 — _ — . — . P w — p _ . _ . 000.0 L n «.0- L.\\\\\\\\¢r Jo -800 1 $4 < l to- 48.0 0.04 000.0 0.0.. 1000.0 0.7 105.0 NT -050 ....T 1.8.0 . 05000.00 90 05 105000.00 I - 0; . 20.0 (N 000 'l/b) Elva NOILVGVEIOEO OIdIOEdS 36 Table 4.2: Specific growth and phenol degradation rates as a function of dilution rate Dilution rate Specific growth rate Specific phenol deg. rate (hr-1) (hr-1) (g/(L ODU hr» 0.25 0.25 0.00183 0.5 0.5 0.00167 1.0 1.0 0.00237 1.3 1.3 0.00258 Table 4.3 shows the outlet pH and dilution rate values. The inlet pH was maintained constant at 7.2 . Table 4.3: Outlet pH as a function of dilution rate Dilution rate Outlet pH (hr-1) 0.25 7.0 0.5 7.08 1.0 7.12 1.3 7.13 37 When the dilution rate was increased from 1.3 hr'1 to 1.5 hr'1 the cells began to be washed out. The pump was switched off until growth became apparent and then turned on again. When three such attempts failed, it was concluded that at this dilution rate a stable, nontrivial steady state could not be reached. An experiment was conducted to study the potential loss of phenol due to evaporation. In this experiment, DP medium was continuously fed at a dilution rate of 1.2 hr‘l. Phenol concentration was measured in the inlet and outlet streams. There were no bacteria present, but all the other conditions were as before (55°C, air sparging at 3 L/min, and agitation at 100 rpm). There was no measurable loss of phenol for more than 3 residence times, and hence, it was concluded that evaporative losses of phenol were negligible. Discussion As shown in Fig. 4.8, over a five fold (500%) increase in speCific growth rates there was only a 45% increase in the specific-phenol degradation rate. This suggests that phenol degradation is uncoupled from growth. It may also suggest that the enzymes needed for this purpose are under tight genetic control (Worden 'et al., 1991). The small change in the phenol degradation rates may be a result of the adaptation of the culture during the experiment. The steady states were achieved gradually from the lowest to the highest dilution rate, and this may have resulted in strain improvement through natural selection (ibid). Note that even though pH was not controlled in these experiments, the pH did not vary much with dilution rate. It has already 38 been shown that pH within the range of 6-8 does not significantly affect growth (see Chapter 2). This result suggests that tight pH control may not be needed in a commercial application. These results show that Bacillus stearothermophilus BR219 can grow on 10 mM phenol at 55°C at high specific growth rates (up to 1.3 hr'l). It has already been shown that the organism can grow on 15 mM phenol (Gurujeyalakshmi and Oriel, 1988a), which is the highest known microbial tolerance for phenol. Also, the high growth rates were achieved on a dilute medium such as DP supplemented with phenol. This combination of high growth rates, extreme phenol tolerance, and potential unimportance of pH control and sterile conditions, makes this organism well suited for phenol biodegradation. In addition, catechol is an intermediate in the phenol-degradation pathway. Thus, it may be possible to produce catechol commercially from phenol using BR219. However, at present, tetracycline addition is needed to block unwanted microbial catechol oxidation. CHAPTER 5 BATCH GROWTH AND CATECHOL PRODUCTION KINETICS OF BACILLUS STEAROTHERMOPHILUS BR321 Introduction The successful transposon mutagenesis of Bacillus stearothermophilus BR219 to create a new strain BR321 in the laboratory of Dr. Oriel led to this study. This new strain accumulates catechol in the absence of tetracycline, thus allowing catechol production during the growth phase. Details of the transposon mutagenesis procedure are given by Natarajan and Oriel (1992). In this research, the kinetics of growth, phenol degradation and catechol production of strain BR321 were examined in batch culture. A simple mathematical model was then developed to explain the batch kinetics of this organism. Materials and Methods The experiments were conducted in 250 mL Bellco flasks with stainless steel caps that allow gas exchange while maintaining sterility. An Aquatherm water bath shaker (New Brunswick, Edison, New Jersey) was used to maintain temperature and provide shaking. A Fisher-Scientific (Springfield, New Jersey) Model 235C micro- centrifuge was used to concentrate the cells in preparing the inoculum. The various assays were done as described in Chapter 3. 39 40 The inoculum was prepared as described in Chapter 4. The cells, grown in LB-Broth for 13 - 17 hours, were microfuged in Reagiergefabe (Sarstedt, Germany) test tubes for 2 minutes and resuspended in DP medium before being used as an inoculum. This procedure reduced the lag phase in DP medium. One hundred mL of DP medium was used in all experiments; the inoculum was 5 mL (5 %). Samples were taken every hour for the first four hours and afterwards every four hours. The samples were assayed for cell turbidity during the first few hours, and the samples taken every four hours were used to measure cell viability, and concentrations of catechol and phenol. Experimental Results Natarajan and Oriel (1992) showed that catechol is produced throughout the growth phase by Bacillus stearothermophilus BR321. To understand the kinetics of this organism and to determine the effect of phenol concentration on catechol production, batch growth experiments were conducted at phenol concentrations of 0, 0.8, 1.6 and 2.4 mM. Figures 5.1, 5.2, and 5.3 show the time profiles of growth, phenol concentration and catechol concentration for the different cases. As seen by Fig. 5.1, phenol concentration did not seem to affect growth at the concentrations tested. Table 5.1 shows the percentage of phenol utilized and catechol produced in these different cases. 41 0.0 :5 0.0.1.080 .22: I :5 0.0188 .225 0.0 25 3n .88 .82: «£0 :5 ....Nuocoo .22: 0.0 20:00 0300 5 0E: m> 0000550000 :00 ...m .9... 0.0 as: on on or 0 . _ I p _ O._. 0.9 10.09 10.009 u¢0+mo._. homo; 100+m0.— 111 00 L/sneo 42 :5 0.0".88 .820 I :5 0._u.ocoo .82: 0.0 :2 ....N u .88 .82: E 0.v 8330 0300 E 08: 00 m> 000.000.0808 .0000; NA... .9... 0.0 9:: om O 0 F') (ww) 'ouog louaqd 43 .88 .88.. xlx .88 .88.. 0.0 .88 .88.. are 0+. 0.330 00000 E 08: 0> .0000 600300 0.0 .9... 0.0 8.: on 0.0. 3 ouog quoetoo (Wm) ' 44 Table 5.1 Phenol utilization and catechol production as a function of varying initial phenol concentration Phenol Concentration % Phenol degraded Catechol Produced (mM) (mM) Initial Final 0.802 0.14 82.5 0.42 1.52 0.54 64.47 0.73 2.39 1.07 55.23 0.859 Assuming that phenol degradation occurs only by the meta pathway, and that catechol 2,3-dioxygenase production is entirely disabled by the transposon insertion, the molar catechol yield should equal the molar phenol consumption. However, Table 5.1 indicates that this equality did not hold. This discrepancy may be due to the autoxidation of catechol as evidenced by the increasing brown color of the flasks with time. The amount of catechol unaccounted for increased with increasing phenol concentrations. The growth rates in the different cases seems to be similar, and the cells died rapidly after reaching a maximum at about 8 hours for all cases. This observation suggests that a lack of substrate may have caused the death of the cells. However, the slopes of the curves in the 'cell death region are steeper for higher initial phenol concentration, and hence higher catechol concentrations, suggesting that accumulation of catechol may have accelerated cell death. 45 To investigate this phenomenon further, an experiment was conducted in which catechol was added in different amounts to flasks containing DP medium supplemented with 2.5 mM phenol. Catechol was added to the flasks before inoculating them. The concentration of catechol was varied from 0 to 0.46 mM. Otherwise, the experiment was conducted as before. Figures 5.4, 5.5 and 5.6 show the concentrations of viable cells, catechol and phenol respectively, as a function of time, for these experiments. The results indicate that the addition of 0.135 mM catechol did not affect the growth of the cell; however, higher concentrations of catechol did inhibit cell growth. Note that when there was no external catechol added, or when 0.15 mM or 0.26 mM was added, the maximum catechol concentration exceeded 0.46 mM. However, when 0.46 mM catechol was added initially, the cells did not grow, and no additional catechol was produced. This observation suggested that, through some unknown mechanism, externally added catechol is more inhibitory than catechol produced by the cells. Model Development The batch growth experiments showed that the growth rate was relatively unaffected by the phenol concentration in the range of 0 to 2.4 mM during the first 8 hours. After 8 hours cell death occurred. From these results, it was assumed as a first approximation that the specific growth rate was independent of the substrate concentration during the growth phase, and that substrate ran out in 8 hours, initiating cell death. The carbon substrate consisted of complex mixtures of carbon sources (yeast extract and casamino acids) whose concentrations could not be individually meaSured. The cells were likely to have grown on 46 0.00:; 00000.0 300300 .0 00000008000 000.200 00; .88.. z... 3 u .0 ... 52m. .0 .050 0.8 ....m 0... 0.0 8.: .88 ..8 0...... I 2... 9.0 n .88 08 0...... 0.0 2... 8.0 u .88 ..8 0...... m0 :9 ...... 0.9.0 u .88 ..8 0...... are loop [000— Ivo+mp umo+uF I00+mF vi OOL/SII93 47 2:. 9.0 :8 mad 2:. 0.90 6:00 6:00 6:00 6:00 00 0.0.0.0. 00000.0 80 .000300 ..0 00200000008 “00.0.0.0 000; 00.0 m> 0000:0808 .0500 0.0 .9... 05 88F .000 .200. I .000 .200. 0:0 .«00 .200. Elm. .000 BEE E 'ouoo loueqd (ww) 3.05:: «comma 20 650300 00 mcosobcoocoo 28820 00;; «E: m> cozobcoocoo 65200 9m .9... AEV 08:. 48 6:00 .«00 .025 I :2 95 u .88 .8 .28.. 0.0 :2 03 u .88 .8 .28.. mm :5 8.0 u .88 .8 .28.. «:4 (ww) 'ouoo loqoamo 49 more preferable substrates first and others later. However, as a first approximation, all substrates were lumped together, for model development purposes. As seen in Fig. 5.4, the cell death rates seemed to increase with increasing catechol concentration. Hence, catechol was assumed to be toxic to the cells. The model consists of two sets of equations: one for the growth phase and another for the death phase. Uptake of both phenol and the critical carbon substrate, as well as catechol production were assumed to occur only during the growth phase. The specific growth and death rates were related by a Dirac Delta type term, whereby, when the substrate concentration reaches zero, specific growth rate is zero, and as long as substrate concentration is greater than zero the specific death rate is zero. The model assumptions are stated below. Model Assumptions 1) The specific growth rate and specific phenol degradation rate are constant, when substrate is present in the reactor. 2) When the substrate is completely consumed, phenol uptake and cell growth cease, and cell death begins. 3) Catechol is toxic to the cells, but phenol is not toxic in the concentration range used. 4) Substrate is utilized for both cell growth and maintenance. 5) There is a molar equivalence between phenol utilized and catechol produced. 50 Model Based on the above assumptions the following model was developed: Growth phase : dX 717 = uX—kCC"X (5.1) (15 (T! = ‘ys/XHX -mX (5,2) dP — = —k X d! P (5.3) E _ _dP d: _ E (5.4) Where; )1 is the specific growth rate (hr'l) X is the cell concentration (cells/100 LIL) C is the catechol concentration (mM) kc (mM‘2-3 hr'l) and n are the catechol inhibition constants kp is the specific phenol degradation rate (mM phenol 100 uL/hr cells) S is the substrate concentration (g/L) yS/x is the substrate yield coefficient (g substrate 100 uL/ L cells) m is the maintenance term (g substrate 100 uL/ L hr cells) Death Phase 51 E = -k.X -k.C"X (5.5) (31—? = 0 (5.6) {if = 0 (5.7) g = 0 (5.8) where, kd is the specific death rate (hr'l) Evaluation of model constants 1) The specific growth rate (u) The cell turbidity data for the case where there was no catechol inhibition were used to tabulate u. Integrating the cell. mass balance equation for batch growth, E‘H-X ' ! J—=wt X0 0 X 1’1 (z) - l” (5.9) (5.10) 52 The best fit of Eqn. 5.10 to Fig. 5.7 yielded t1 = 0.32 hr'l, with the coefficient of linearity = 0.958. When S is zero, it is set to zero. 2) The specific death rate (kd) The rate of cell death in the absence of substrate was calculated by fitting a straight line to the cell concentration data shown in Fig. 5.1, using that part of the curve where cell concentration is decreasing with time. Integrating the cell death equation for batch growth, dX _ = _ 5.11 d: kdx ( ) X dX' t J T = _kdIdt' Xmax 0 X In (X ) = —kdl (5.12) max The best fit of Eqn. 5.12 to Fig. 5.8 yielded kd = -0.30 hr‘1 with a correlation of linearity of 0.995. When S is greater than zero, kd is set to zero. 3) The catechol inhibition constants (kc and n) The difference in the death rates, shown by the negative slopes in Fig. 5.9, is attributed to the different levels of catechol. A log plot of the death rates vs catechol concentration in the reactor, as shown in Fig. 5.10 yielded values for the constants kc (slope) and n (antilog of the intercept) from Eqn. 5.1 as follows kc =0.63 (catechol concentration is in mM) 53 1 mo 5:050:83 5m .90 EV 2:: M. m w w o . 3.83": x _.mmt \x- \ IV.NI \x - \ \\.. . \ ..o.m| \ \ r \ \ - \ A 0.7 \ . \ l Nél (uuu 099 'G'O -— ‘ouoo Heo)ul 54 ..x .6 885558 3.. .8 EV 8c: on mm ON 9 OF m ._...__... :hmqolflvx Xi— :0— INF : (Tr/ COL/sues) Ul 55 mcozobcoocoo 6:020 mEb0> Lo. we: 8 8:83 o 8 £26 :8 Wm .90 95 2:: so 25 2.0 :2: 22 3 688m :5 ON .888 :5 am .88; 25 o In n5 n5 n5 do 0030:0808 :00 :6 .9; EV 8.: on ON 0.. o . _ . I — . — _ O.F 10.9 10.00, iodoop ivo+moé tmo+mo.— _ imo+mo.F Woo L/sneo o.v :5 m.ou.88 .825 x :5 Snacoo .88“. o .0000 3:25 4 05 08:. on cm 0? _ . _ . _ x u n x I o o o o :10, a c llllllllllll l /< 2330 5300 E 06: m> cozobcoocoo 6:020 Nfim a: us 'ouog |oueqd (WW) 61 2E md SE 0.. 2E TN :5 wd 2E m.— EE TN 23.30 5300 5 0E.“ m> cozobcoocoo .2038 2d 91.. E. 8.: 0.“. on ON 0 — . _ n 9.0 6:00 .002... mm .280 .0005. O . ..ocoo .ocozn. 4 iNd .0000 .0025 II . .0000 .0025 ---- u u . .0000 .0095 I l n n 1v o O I o 10.0 < o o 1 a $6 a no.— \. . -3 (ww) ‘ouoo loqoaiog 23.30 5800 c. 2.0: m> .ocoo ouobmnam 3.0 .0: c5 8:: 0..~ on om . _ p _ . SE in u .800 .0520 l l 62 (1/5) '0003 910.113an 63 3.0.00. .0880 .000200 .0 9000:0808 00280.0 5.; .00000 200 mdtmo 0. Fwnmm .0 5300.0 0300 9.0 .0... e... 8.0 on 8 ON 2 o. m o - _ — p — _ - _ - — F - n _ — - r - - — . _ - n — - F F .020. 2E 9.0 n .0000 .000 .200. 0 2E mud n .0000 ..00 .200. .u low :E 3.0 n .0000 ..8 .020. < EE 0 n .0000 ..00 .020. .l. ....E 9.0 n .0000 ..00 .020. ll 0 too" 20.. mud n ..0000 ..00 .020. I l w 20.. 00.0 n .0000 ..00 .200. l I H . [82 fl 0 Lam: .m .l [3+0— . i00+m.. I .ocoo .ocou .0050 .0500 .ocoo .OCOO .ocoo .030. .200. 60.5 60.... .205 .208. i i 6%.. i 1 Wm O 63.0. n. 4 3.0.00. 000020 80 .0538 n.0 2000:0808 0.080....0 000; 90.0 m> 0000:0808 .00000 3.0 a... E. 80 mm on 0N ON mp 0.. n I I I I — I I I - — I I I I — I .I I - — I - I I — I I I I — I I I 0 o o I l . ..l .0l. lllll l. a q (ww) °ou03 Iouaqd 65 .«oo 3.03.... «583 So .0538 .0 2052.588 E88...” cock on... m> cozobcmocoo .2088 :6 a... AEV mE... on mm om mm (ww) 'ouoo loqoemg 66 coloration of the' flasks with time. Hence, the model overpredicts the catechol concentration. Also, some of the catechol could be present within the cell and may come out as cell death occurs. This catechol would not be measured, and the data shown would not account for it. The model, however, assumes that all the catechol diffuses out as soon as it is produced. This could be another reason, although not as significant, why the model overpredicts the catechol concentration. The disagreement between the model and the data in the catechol-challenge experiments suggests that the catechol added externally exerts a stronger inhibitory effect than catechol produced by the cells. The catechol concentrations added were much lower (maximum added concentration = 0.46 mM) than the maximum concentrations achieved (1 mM) when no catechol was added. A likely explanation for this trend is that added catechol has a longer residence time in the reactor, and hence oxidizes to a greater extent than the catechol produced by the cells. The oxidation products may be more toxic than the catechol itself. Also, added catechol is present during the growth phase, and this catechol may have a toxic effect on the enzyme induction; it may repress either enzyme production or enzyme activity. Catechol added externally may also induce a lag phase from which the cells might not recover. Catechol produced by the cells also oxidizes. However, the majority of the catechol is produced late in the growth cycle, even as substrate runs out. By the time this catechol oxidizes, cell death due to lack of substrate would have already begun. The experimental data, as seen in Figs. 5.4 and 5.11, show that when 0.135 mM of catechol was added, the growth kinetics’ were not 67 affected. This is significant because it shows that catechol concentrations in this range do not affect growth. Continuous catechol removal methods could conceivably be used to keep the concentration below this level during the fermentation. One of the major difficulties of this research was that the substrate concentration could not be measured directly during the fermentation. A better understanding of this process would be possible if the substrate concentration could be measured and if studies on the enzymatic reactions were carried out. This model, however, seems to . explain the kinetics of batch growth fairly well, and the fact that it is simple to use is a clear advantage. Bacillus stearothermophilus BR321 grows at a high rate on a dilute medium such as DP and converts phenol to the specialty chemical catechol. Catechol concentrations as high as 1 mM have been achieved in this process; however, both externally added catechol and catechol produced by the cells affect the kinetics of this organism. A simple model has been suggested and used to understand the kinetics of this organism, and it fits the growth and phenol-uptake data well. The ability to measure the substrate concentrations will further help in understanding the system. The high growth rates and ability to grow on dilute, inexpensive media seem to indicate that this process, in due course, may become a suitable alternative for catechol production. CHAPTER 6 CONTINUOUS CULTURE EXPERIMENTS WITH BACILLUS STEAROTHERMOPHILUS BR321 Introduction The main objective of the continuous culture experiments was to evaluate the feasibility of this process for making catechol continuously and to understand the growth kinetics of BR321 in continuous culture. As seen in the previous chapters, BR321 grows in DP medium supplemented with phenol and accumulates catechol throughout the growth phase. Continuous production of catechol could make this process more industrially viable, so it was decided to investigate the growth kinetics and catechol production in continuous culture. Materials and Methods The experimental setup and procedure were similar to the one in Chapter 4, except catechol concentration was monitored. The turbidity assay, which was used as the indicator for steady state in the experiments with BR219, could not be used in this experiment since oxidized catechol gives the broth a brownish color, thus masking the turbidity. Also, sterility was more important in this set of experiments than with BR219, since lower phenol concentrations were involved. All experiments were done at a dilution rate of 0.12 hr'l. Catechol was assayed periodically, and it served as the indicator for reaching steady 68 69 state. At steady state the cell viability and dry weight were measured. Results Cell concentration profiles leading to steady states are plotted in Figs. 6.1-6.4. The steady state cell dry weight and viability are plotted as a function of the initial phenol concentration in Fig. 6.5. The steady state results are summarized in Table 6.1. The cell dry weight seemed to remain constant with initial phenol concentration; however, the results have high standard deviations as seen in Fig. 6.5. The steady state catechol concentration is plotted as a function of the feed phenol concentration in Fig. 6.6. When the feed phenol concentration was changed to 0.36 mM from 0.24 mM, the catechol concentration began to oscillate (Fig. 6.7). This experiment was run for more than 50 residence times, and in this period the catechol concentration went through three cycles. Some interesting characteristics were observed during this experiment. The period of the oscillations increased from one cycle to the next. Between cycles there was a lag period which seemed to increase from one cycle to the next, i.e. 10 hours between the first and second cycles to about 100 hours between the second and third cycles. The feed carboy had to be changed more than ten times during the course of this experiment, and these changes might have caused small perturbations in the feed concentration. The maximum steady state catechol concentration achieved was 0.08 mM. The reactor liquid turned increasingly brown as the feed phenol concentration was increased, suggesting increasing concentrations of catechol oxidation products. Autoxidation of catechol is also suggested by the results given in Table 6.1, where the moles of phenol consumed are greater 2c.mvo H cozcbcwocoo_ocmca_c::: 5.; .Nnmm .0 £266 83m €8.95 ..m .9... E. 2:: om ow o_n o_N . o.— 70 it )E J ) (wu 099 0'0) woo Heo 71 (mm) °ou03 loqoeiog 2E 000.0 n 0050500950 .0093 .05.... 5.; 52m .6 5.55 8% 38.25 .3 .9. AEV mEfi om on 0.. on ON or o 00.0 . _ _ p _ _ _ . _ . .u OP.O 5.? ..NS 8.0.. . 1.1.0 mod: .. 12.0 3.01 . u . 6 . . 19.0 no or .0000 650300 0.0 . As: one do. .88 :8 I omd mod (uuu 099 0'0) 'ouoo “IIGO 72 00 2:. mm. .0 H 00.00.052.00 .9.ch .03.... £3, 52m .0 5.55 8% 0.8.95 m... .9. 0... we: 00 0.V 0m 0m 0 F 0 p _ . _ . _ p _ . _ . .V0.0 100.0 no.0 (ww) °ouog |oqoeiog 73 .2... mmmd H ..o..0.....oo..oo .0023 .03.... ...... .Nmmm. .0 £2.80 80% 3.089.: 0.0 .0... .... a... on 0.0 0.0 0% 0_n 0_N 0__. mod .18... 150.0 100.0 00.0 (ww) 'ouog |oqoeioo 74 (11» 92/6) 81190 10 wfileM Mo wNmmm ...—£2. 0.2.530 m303c$c00 E COSOLHCUUCOU .90.... .00.... m> 00.00....00000 ..00 0.0.0 300% 0.0 .0... 0...... .0000 .0..0..n. .05.... 0N0 0N..0 9.0 0....0 9W0 00.0 , no.0! 0000 0 ”2.0.02. .00 O . 0:00 0.00.) x mo+mo n . .000... I 0:00 0.005 I r 88.0- . X 1 - 1m0+mo.m 0000.01 .. 0000.01 . . ... ll 1mo+moR 0000.01 ... . . e . 050.0..H tmo+mo.m 0.8.01 II MW . 00+m F. "171 oo L/sneo 9lq0!/\ 75 2.0 00 0:2 00.03:. 0 .0 00.000000 0.03 30081030 .3. ...................... 0.0; 0000.0 0000. .0: 0.0-------:-- V090 8.5.0 20.0 no. a. m... 8.00... 02.0 .80... £00 a... x a: 88... 2.... $0... 0.0... n... x .0... $08... $0... 0.0 0.0 so. 0. 5.. 2.00... 0.0 02.5 .28. 50228.50 3... 02.3.8.8. Se 2.0. ...... .28. .250 0000005 3:00.00 .25.... .2... 2.00 2......» to :00 .22... 3...... :25 5.3 0.3.3 0:005:50 80.0 3.000. 880 .000: 00 .308800 ...0 030... 76 memm 5:; 03:30 mzoaczcoo E scrotcmocoo 6:050 BEE m> 8:828:00 6:00.00 ed .9... 3.5 .88 68.: 6...; mud om..o 9.0 omd . mod — b — p - b _ - P b p - 620200098 X 600.: ll (Wm) p83npOJd loqaemg 77 2E mmd u co_.0._.c00cow. 6:050 BEE £3, .mnmm *0 £305 0.3m €00.95 No .9... AEV 08:. com ow¢ own. 0mm co. » n - No.0 ax. 109.0 Imod + . mod :36 N—d (ww) ‘ouog loqdemo 78 than the moles of catechol produced. hdodel The model, described in Chapter 5, was extended to the continuous process by including a flow term in each differential equation. Growth Phase %— -_- uX-ch"X+DXf-DX (6.1) ‘15 — _ pX—mX-l-DS -DS (6.2) ‘5; - ys/x f 5% = —kpX+DPf—DP (6.3) %‘ti = —DC+kpX (6.4) D is the dilution rate (hr‘l), the subscript f refers to the feed, and all other symbols have the same meaning as in Chapter 5. Death Phase dX .37 = — kdX - ch"X + DXf— DX (6.5) dS —-- = -D 606 dt DSf S ( ) 79 dP E = DPf-DP (6.7) dC _ 2? ' DC (6.8) Discussion The model predictions for the cell and catechol concentrations are shown in Figs. 6.5 and 6.6 respectively. The model predicts that the substrate concentration approaches zero at steady state. The cell concentration predicted by the model is lower than the experimental value. This error may be because the modeling constants were evaluated from batch-growth experiments in Erlenmeyer flasks, while the continuous experiments were conducted in a New Brunswick Bioflo II fermentor, which has better control, mixing properties and aeration. Thus, the constants used in the model might not be appropriate for the continuous culture studies. The catechol concentrations predicted by the model are higher than the experimental values, due in part to autoxidation of catechol, which is not taken into account in the model. Also, the catechol concentration remains constant at the steady state value rather than peaking late in the experiment, as in the batch experiments. Long term exposure to catechol was found to strongly inhibit the catechol production kinetics, as discussed in Chapter 5. The catechol concentrations obtained in these experiments were low compared [to those obtained in the batch experiments. The model predicts that such low concentrations should not affect the viability of the cells. However, the catechol oxidation by-products may be more toxic to the cells than the catechol itself. This hypothesis is consistent 80 with the decreasing cell viability with increasing feed phenol concentrations i.e. increasing product catechol concentrations. Catechol is colorless, but, as the experiment progressed, the reactor became slightly brown owing to catechol oxidation. This brown color became more evident as the product concentration increased in the reactor. The biochemistry of this fermentation is complicated. The difficulty in measuring substrate consumption makes the modeling of this system harder. The mathematical model can predict the kinetics of this organism in batch culture when catechol is not present initially in the reactor. However, it fails to accurately predict continuous culture data. It underpredicts the cell concentration at' low feed phenol concentrations and does not predict decreasing cell viability with increasing feed phenol concentration. The model also predicts steady catechol production when the feed phenol concentration was raised to 0.36 mM, while the actual data show oscillatory dynamics, as seen in Fig. 6.3. This instability is thought to be due to the inhibitory effects of catechol and its oxidation products. Understanding the effects of catechol oxidation on the cells will help in modeling this system better. Figs. 6.1 and 6.2 show that continuous catechol production can be achieved at concentrations up to 0.08 mM using 0.24 mM phenol in the feed. Thus, steady catechol production in a continuous fermentor has been demonstrated experimentally. Continuous productionof catechol would be advantageous for large scale industrial production. CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS Kinetic studies on two strains of Bacillus stearothermophilus i.e. BR219 and BR321 were carried out. Bacillus stearothermophilus BR219 can grow on high phenol concentrations (10 mM) at 55°C at high growth rates (1.3 hr'l). The results of continuous culture experiments show that because of its high specific growth rate, extreme phenol tolerance, and potential unimportance of pH control and sterile conditions, this strain may be well suited for use in bioreactors to degrade phenol. Bacillus stearothermophilus BR321, a mutant of BR219, can convert phenol to the specialty chemical catechol while growing on a dilute medium like DP. Thus, it is well suited for continuous catechol production. A summary of important results obtained during the course of this research is given below. Summary of Results In continuous culture, Bacillus stearothermophilus BR219 achieved high specific growth rates (up to 1.3 hr'l) on a dilute medium (DP) supplemented with 10 mM phenol. Steady states were achieved at dilution rates from 0.25 hr’1 to 1.3 hr'l. Automatic pH control was not required, and microbial contamination did not become a problem, even after loss of aseptic conditions. The specific phenol degradation rate was relatively unaffected by the growth rate, varying only forty 81 82 five percent over a five fold increase in the growth rate. In batch culture, Bacillus stearothermophilus BR321 grew rapidly (0.3 hr'l) on a dilute medium (DP) while converting phenol to catechol. Catechol concentrations as high as 1 mM were achieved using 2.4 mM phenol feed. Cell death began immediately following the growth phase, presumably due to both a lack of substrate and catechol toxicity. Catechol autoxidation products may also affect the growth kinetics of this organism. A mathematical model was developed to explain the batch kinetics of this organism, and it included cell death due to lack of substrate and catechol toxicity. The model fit the growth and phenol- uptake data well. In continuous culture experiments at a dilution rate of 0.12 hr‘l, BR321 produced catechol for phenol concentrations in the feed up to 0.24 mM. Above this concentration oscillatory dynamics Were observed. Further Studies This research demonstrates the ability to biologically produce catechol from phenol. However, because catechol is more toxic than phenol, research is needed on the separation of catechol as it is formed in batch and continuous processes. The economic feasibility of this process also needs to be evaluated in detail. Scale-up studies would help in developingthis process further. Kinetic studies on the autoxidation of catechol and its effect on growth are needed to improve the existing model and possibly increase reactor performance. This research has given important insights in the potential use of Bacillus stearothermophilus for catechol production on a large scale. Future work may lead to a full scale commercial process. APPENDIX C ------ Ramkumar Subramanian C ------ ChE 899 Masters Thesis Research C ------ This program solves the two differential equations using IMSL ---------- C ------ subroutine IVPRK (by the 5th order Runge — Kutta Method) and ---------- C ------ evaluates the values of various parameters C INTEGER XPARMINEQ PARAMETER (MXPARM-50,NEQ-4) COMMON KP,YSX,S INTEGER IDO,ISTEP REAL FCN,PARAM(MXPARM),T,TEND,TOL,Y(NEQ),KP,YSX,S EXTERNAL FCNIIVPRK'SSET C CHARACTER RFILE*80 TYPE'(X,A,$)'.'ENTER THE FILE NAME FOR RESULTS :' ACCEPT'(A)’.RFILE OPEN(UNIT-2,NAME-RFILE,TYPE-’NEW') C4 ----- Nomenclature —- C Y(l) - Cell Concentration (cells/100 micro L) C Y(2) - substrate Concentration (g/L) C Y(3) - Phenol Concentration (mM) C Y(4) - Catechol Concentration (mM) C KP - 3e-7 - Specific Phenol Degradation Rate C YSX - 2.4e-7 - Substrate Yield Coefficient C S - 0.05*YSX - Maintenance Term C Note : Only in the program I use S as the maintenance term C MD - 0.3 i Specific Growth Rate C KD - 0.3 - Specific Death Rate C N and KC are the Catechol Inhibition constants - 2.3 and 0.64 resp. C ------ Input initial conditions Y(2)-0.3 Write (*,*) ’ T,X0,PHE,CAT,KP,YSX,S' READ (*:*) T,Y(1),Y(3)oY(4):KPoYSX:S C ------- Input data TOL-0.1 C ------ Setting param to default CALL SSET(MXPARM,0,PARAM,1) C ------ Selecting values for PARAM(1,10) PARAM(1) -1.E-04 PARAM(4)-1.E¢ PARAM(10)-1. C IDO-l WRITE(2,*)' Time [X] [S] [P] [C]' WRITE(2,*) T,Y DO 50 ISTEP-l,32 TEND-FLOAT(ISTEP) CALL IVPRK(IDO,NEQ,PCN,T,TBND,TOL,PARAM,Y) WRITE(2,*) T,Y SO CONTINUE C IDO-3 CALL IVPRK(IDO,NEQ,FCN,T,TEND,TOL,PARAM.Y) END C ------ Subroutine FCN begins now SUBROUTINE FCN(NEQ,T,Y,YPRIME) COMMON KP,YSX,S INTEGER NEQ REAL Y(NEQ).YPRIME(NEQ),MU,KC,YSX,KP,N,KD,S 83 84 MU-0.3 XC-0.63 N-2.3 KD-0.3 IF(Y(2).GT.0.0S) THEN IP(Y(4).LE.0.0) Y(4)-0 YPRIME(l)-MU*Y(1)-(KC)*(Y(4)**N)*Y(1) ELSE YPRIME(1)-(-KC)*(Y(4)**N)*Y(1)-KD*Y(1) END IF . IF(T.LE.2.5) YPRIME(1)-0.0 YPRIME(2)-(-YSX)*(MU*Y(1)+S*Y(1)) IF(T.LE.2.5 .Or. Y(2).LT.0.01) YPRIME(2)-0.0 YPRIME(3)'(-KP)*Y(1) YPRIME(4)-KP*Y(1) IP(T.LE.2.5.0R.Y(3).LE.0.02) YPRIME(3)-0.0 IF(T.LE.2.5.0R.Y(3).LE.0.02) YPRIME(4)'0 IF(Y(4).LE.0.0) Y(4)-0.0 RETURN END 85 C ------- Ramkumar Subramanian C ------- ChE 899 Masters thesis C ------- Solution to Continuous model C? ----- This program solves the two differential equations using IMSL ---------- C ------ subroutine IVPRK (by the 5th order Runge - Kutta Method) and ---------- C ------ evaluates the values of various parameters ‘ C INTEGER XPARMINBQ PARAMETER (MXPARM-50,NEQ-4) COMMON P,KP,YSX,S INTEGER IDO,ISTEP REAL FCN,PARAM(MXPARM),T,TEND,TOL,Y(NEQ):P,YSX,KP,S EXTERNAL FCN.IVPRK,SSET CHARACTER RFILE*80 TYPE'(X,A:$)'¢’ENTER THE FILE NAME FOR RESULTS :' ACCEPT'(A)’,RFILE OPEN(UNIT-2,NAME-RFILE,TYPE-'NEW') C ------ Nomenclature -—-- - --- C D - 0.12 - Dilution Rate (l/hr) . C All other parameters are defined in the batch model C ------ Input initial conditions Y(2)-0.3 Write (*,*) ' T,X0,PHE,CAT,TIME LIMIT.STEP,RP.YSX,S’ READ (*,*) T,Y(1),P,Y(4),LIMIT,LIMIT$.KP,YSX,S Y(3)-P C ------- Input data ’ ----------------- TOL-0.1 C ------ Setting param to default -- CALL SSET(MXPARM.0,PARAM,1) C ------ Selecting values for PARAM(1,10) PARAM(1) -1.E-04 PARAM(4)-1.E4 PARAM(10)-1. IDO-l WRITE(2.*)' Time [X] [S] [P] [Cl' warren,” T,¥ DO 50 ISTEP-1.LIMIT.LIMITS TEND-FLOAT (ISTEP) CALL IVPRK(IDO,NEQ,PCN,T,TEND,TOL,PARAM,Y) WRITE(2,*) T,Y 50 CONTINUE IDO-3 ' ’ CALL IVPRK(IDO,NEQ,FCN,T,TEND,TOL,PARAM,Y) END C ------ Subroutine FCN begins now SUBROUTINE FCN(NEO,T.Y,YPRIME) COMMON P,KP,YSX,S INTEGER NEQ ' REAL Y(NEQ) IYPRIME (NEQ)IMU(KC'KP'KD' D,P,YSX,S MU-0.3 KC-0.63 N-2.3 D-0.12 86 KD-0.3 - IF(Y(2).GT.0.05) THEN YPRIME(1)-MU*Y(1)-D*Y(1)-(KC)*(Y(4)**N)*Y(1) ELSE YPRIME(1)-(-KC)*(Y(4)**N)*Y(1)-KD*Y(1)-D*Y(1) END IF YPRIME(2)-(-YSX)*(MU*Y(1)+S*Y(1)-D*(Y1))-D*Y(2)+D*0.3 IF(Y(2).LE.0.0001) YPRIME(2)-0.3*D-D*Y(2) IF(Y(2).GT.0.05.AND.Y(3).GT.0.02) THEN YPRIME(3)-(-KP)*Y(1)-D*Y(3)+P*D YPRIME(4)-KP*Y(1)-D*Y(4) ELSE YPRIME(3)-P*D-D*Y(3) YPRIME(4)-D*Y(4)*(-1) END IF IF(Y(3).LT.0.0) Y(3)-0.0 IP(Y(4).LT.0.0) Y(4)-0.0 RETURN END BIBLIOGRAPHY BIBLIOGRAPHY Adams D. and Ribbons D.W.(1988), J. Gen. Microbiol., 134, 3179. Bailey J.E and Ollis D.F (1985a) Biochemical Engineering Fundamentals, 2nd ed., McGraw Hill, p. 385. Bailey J.E and Ollis D.F (1985b) Biochemical Engineering Fundamentals, 2nd ed., McGraw Hill, p. 392. Boyd R.E. (1988), General Microbiology, 2nd ed., Times Mirror/ College Publishing, p. 402. Buswell J.A. (1974), Biochem. Biophys. Res. Commun., 60, 934. Buswell J.A. (1975), J. Bacteriol., 124, 1077. Buswell J.A. and Twomey D.G. (1975), J. Gen. Microbiol., 87, 377. Chemical marketing Reporter (Sept 23, 1991a), 37. Chemical marketing Reporter (Sept 23, 1991b), 40. Chemical marketing Reporter (Sept 23, 1991c), 41. Dagley S. (1971), Advances in Microbial Physiology, Academic Press, 6, 1. Gurujeyalaksmi G. and Oriel P. (1988a), App. Environ. Microbiol. 55,500. ' Gurujeyalaksmi G. and Oriel P. (1988b), Proceedings of the Federation of European Microbiological Societies Symposium ed. Da Costa M.S., J.C. Duarte, and Williams R.A. D, Elsevier London,p.113. Gurujeyalaksmi G. and Oriel P. (1989), Biotechnol. Lett. 11, 689. Irie S., K. Shirai, S. Doi, and Yorifuji T. (1987), Agr. Biol. Chem, 51, 1489. Kirk-Othmer Encyclopedia of Chemical Technology (1981), 3rd ed., 13, 39 ' Maniatis T., E.F. Fritsch, and Sambrook J., (1982), Molecular Cloning - a laboratory manual, Cold Spring Harbor Laboratory, p. 68. Marison I.W. (1988) in Biotechnology For Engineers, ed. Alan Scragg, Ellis Horwood ~West Sussex, England p. 200. 87 88 Martin R.W. (l949), Anal. Biochem., 21, 1419. Natarajan M.R. and Oriel P. (1992), Biotechnology Progress (in press). Sonnleitner B. (1983), Microbial Activities, Springer-Verlag, Berlin, p.69. Sonnleitner B. and Fiechter A. (1983) Trends Biotechnol. 1, 74. Varagnat J. (1976), Ind. Eng. Chem. Product Res. Dev., 15, 212. Waters W.A. (1964), Mechanisms of Oxidation of Organic Compounds, John Wiley & Sons, p. 146 Worden R.M., R. Subramanian, M.J. Bly, S. Winter and Aronson C.L. (1991), Applied Biochem. and Biotechnol. 28/29, 267. H. mm...“- . my 'v(‘:.-tlu~r'll:l'~pu .n l -.r\ l [‘vji "n.1,. HICHI l GQN STATE UNIV. LIBRARIES WHIHI'llWWWWII“llHlllHWWlW 1293010553158