A Cometabolic Kinetics Model Incorporating Enzyme Inhibition, Inactivation, and Recovery: II. Trichloroethylene Degradation Experiments Roger L. Ely,' Michael R. Hyman? Daniel J. Arp? Ronald 6. G~enther,~and Kenneth J. Williamson'* Departments of 'Civil Engineering, 'Botany & Plant Pathology, and 3Mathematics Oregon State University, Corvallis, Oregon 9733 1 Received June 6, 7994lAccepted December 13, 1994 A cometabolism enzyme kinetics model has been pre- fects included assumptions requiring experimental verifica- sented which takes into account changes in bacterial ac- ti~n.~Among them was the assumption that, although in- tivity associated with enzyme inhibition, inactivation of hibition, inactivation, and recovery create an inherently enzyme resulting from product toxicity, and respondent synthesis of new enzyme. Although this process is inher- nonequilibriurn, unsteady-state condition, changes in en- ently unsteady-state, the model assumes that cometa- zyme levels, specifically changes in concentrations of en- bolic degradation of a compound exhibiting product tox- zyme/substrate complexes (ES, El, and ESZ), occur slowly icity can be modeled as pseudo-steady-state under cer- enough that the process may be modeled as pseudo-steady- tain conditions. In its simplified form, the model also state provided that enzyrnehbstrate associatiorddissocia- assumes that enzyme inactivation is directly proportional to nongrowth substrate oxidation, and that recovery is tion reactions are fast compared to the net rate of change of directly proportional to growth substrate oxidation. In enzyme quantities. Also, the amount of enzyme inactivated, part 1, model derivation, simplification, and analyses E', was assumed to equal some function, h, of the non- were described. In this article, model assumptions are growth substrate oxidized, P,; and the amount of new en- tested by analyzing data from experiments examining E,,,, trichloroethylene (TCE) degradation by the ammonia- zyme synthesized, in response to inactivation, was oxidizing bacterium Nitrosomonas europaea in a quasi- assumed to equal some function, g, of the growth substrate steady-state bioreactor. Model solution results showed oxidized, P,. In simplifying the model, it was assumed that TCE to be a competitive inhibitor of ammonia oxidation, h was equal to a constant fraction, f, representing the with TCE affinity for ammonia monooxygenase (AMO) amount of enzyme inactivated per nongrowth substrate ox- being about four times greater than that of ammonia for idized, that is, E' = jP2. It was assumed also that g was the enzyme. Inhibition was independent of TCE oxidation and occurred essentially instantly upon exposure to TCE. equal to a constant fraction, fs, representing the amount of In contrast, inactivation of AM0 occurred more gradually new enzyme synthesized, in response to inactivation, per and was proportional to the rate and amount of TCE ox- growth substrate oxidized, that is, En,, = fpl.Using these idized. Evaluation of other 0,-dependent enzymes and assumptions, equations were derived to describe cometa- electron transport proteins suggested that TCE-related damage was predominantly confined to AMO. In re- bolic transformation rates in general and rates of ammonia sponse to inhibition and/or inactivation, bacterial recov- and TCE oxidation by N. europaea in particular, in the ery was initiated, even in the presence of TCE, implying absence of significant cell growth or death. that membranes and protein synthesis systems were N. europaea are autotrophic, ammonia-oxidizing bacte- functioning. Analysis of data and comparison of model ria. Ammonia monooxygenase (AMO) catalyzes the reduc- results showed the inhibition/inactivation/recovery con- tion and insertion of an oxygen atom from molecular 0, cept to provide a reasonable basis for understanding the effects of TCE on AM0 function and bacterial response. into ammonia, oxidizing the ammonia to hydroxylamine. The model assumptions were verified except that ques- As indicated in Figure 1, oxidation of NH, to NH20H is a tions remain regarding the factors controlling recovery reductant-consuming step, requiring two electrons that must and its role in the long term. 0 1995 John Wiley & Sons, Inc. be supplied by the subsequent oxidation of hydroxylamine Key words: Nitrosomonas europaea ammonia oxida- to nitrite. l1 Two other electrons obtained in oxidizing hy- tion kinetics model trichloroethylene cometabolism TCE droxylamine to nitrite enter the electron transport chain, providing energy for cell growth and maintenan~e.~ INTRODUCTION AND BACKGROUND Many metabolic and phylogenetic similarities exist be- Derivation of a cometabolism enzyme kinetics model incor- tween ammonia-oxidizing and methane-oxidizing ba~teria,~ porating bacterial inhibition, inactivation, and recovery ef- but much more is known about the soluble and particulate methane monooxygenases (sMMO and pMMO) found in * To whom all correspondence should be addressed. methane-oxidizing bacteria than about AMO. AM0 is Biotechnology and Bioengineering, Vol. 46, Pp. 232-245 (1995) 0 1995 John Wiley & Sons, Inc. CCC 0006-3592195lO30232-14 Quasi-Stead y- Sta te Reactor Experiments N. europaea cells were grown axenically in batch cultures (1.5 L) as described previously. lS The growth medium con- sisted of 25 mM (NH4),SO4, 3 mM KH2P0,, 735 @I MgSO,, 200 pA4 CaCl,, 10 @I FeSO,, 17 @I EDTA, 0.7 To Electron pA4 CuSO,, and 0.04% (wt/vol) Na2C03, and was buffered 2 e- Transport Chain with a phosphate solution (pH 8.0) to final concentrations of 43 mM potassium phosphate and 4 mM sodium phosphate. Figure l. Reactions catalyzed by N, europaea in converting ammonia to Cells were harvested by centrifugation at or near the end of nitrite. AM0 is ammonia monooxygenase. HA0 is hydroxylamine oxi- doreductase. the log-growth phase 3 days after inoculation; washed with 500 mL of solution containing 80 mM phosphate and 100 mM carbonate buffers (pH 7.8); centrifuged again; and re- membrane bound as is pMMO, and its substrate is thought suspended in 500 mL of the same phosphateicarbonate so- to be NH, rather than NH: .24 AM0 catalyzes the insertion lution, except that 80 mM phosphate and 50 mM HEPES of from 1802into NH, to produce NH,'80H." In buffer (pH 7.8) were used in experiments 6 through 9. Cell addition, it catalyzes the oxidation of many nongrowth suspensions were sealed within the reactor vessel and used substrates such as methane and other n-alkanes (to c,), immediately after harvesting. n-alkenes (to C,), aromatics, and halogenated hydrocar- The reactor vessel (Wheaton Double-Sidearm Cellstir) bons, including TCE.4,6,'4,'6"'7.'9.22,27Whereas evidence had a total volume of 1.67 L comprised of 0.5 L of liquid suggests considerable similarity between AM0 and (the cell suspension) and 1.17 L of headspace (see Fig. 1). p~~~,12,13,18,21,2S,26 . its ability to catalyze the oxidation The top of the reactor was sealed with a Teflon-lined, gas- of aromatic hydrocarbons and straight-chain alkanes above keted cap. The two upper reactor sidearms were sealed with C, indicates that, catalytically, AM0 may be more similar screw-cap, septum vial caps fitted with Teflon-gasketed to sMMO than to the less versatile pMMO system. Mininert valves, and the lower sidearm was sealed with a N. europaea were used in this study for four major rea- hypo-vial Mininert valve (Supelco). Silicone septa in the sons: (1) Nitrifying bacteria are particularly attractive for in Mininert valves were replaced immediately prior to each situ bioremediation processes because they are ubiquitously experiment. Turbulent mixing of the reactor liquid, with present in natural soil environments and their activity can be extreme vortexing and air entrainment, was provided by a stimulated by adding ammonia and oxygen; (2) substrate Teflon-coated, 3-in. magnetic stirring bar operated at high utilization and other laboratory experiments may be con- speed by a magnetic stirrer. ducted with them under a variety of conditions without sig- This mixing method, selected after conducting gas-liquid nificant changes in cell concentrations because they grow mass transfer evaluations of several mixer and baffle con- slowly (generation time of about 8 h); (3) inherent compli- figurations, accomplished complete gas-liquid equilibrium cations of interpreting results from mixed culture experi- in less than 2 min after a pulse injection of carbon tetra- ments can be avoided because methods and techniques have chloride (CT) into the reactor (data not shown). This indi- been developed for maintaining and employing pure cul- cates that, while a brief time period was necessary to tures of this species under laboratory conditions; and (4) achieve equilibrium after a pulse addition of volatile com- they may function as a model for other monooxygenase- pound, gas-liquid mass transfer retardation effects were not and dioxygenase-utilizing bacteria. The goals of this re- significant during the much slower process of TCE degra- search were: (1) to distinguish and quantify the processes dation. A syringe pump arrangement delivered, at a con- affecting bacterial activity during TCE cometabolism in the stant rate, a solution containing ammonium sulfate in 80 absence of significant cell growth or death; (2) to determine mM phosphate/100 mM carbonate buffer (pH 7.8). The sy- the kinetics of TCE degradation by ammonia-oxidizing bac- ringe pump contained a lO-mL, gas-tight, her-lock, glass teria; (3) to investigate conditions potentially conducive to syringe