(560l) Temperature Dependence of pH Dynamic Behavior in the Glucose Oxidase-Ferricyanide-Glucose-NaOH Reaction System in a CSTR

Our work is focused on detailed experimental observations of bistability in the glucose oxidase-ferricyanide-glucose-NaOH system in a continuous stirred tank reactor (CSTR) closed to atmosphere under systematically varied temperature and input NaOH concentrations.

The continuous stirred tank reactor has four inlet streams delivering solution of glucose oxidase, glucose, ferricyanide and NaOH and one output stream. The pH electrode screwed into the reactor keeps the system closed to atmosphere. Temperature is measured by a thermometer placed in the output stream from the reactor. The temperature of stock solutions and reaction mixture is controlled and maintained by a dry thermostat and by a magnetic stirrer and heater. Input concentrations of the enzyme (14.7U/ml), glucose (12mM) and ferricyanide (234mM) are maintained constant, while temperature, which significantly affects the activity of the enzyme [1], and NaOH input concentration are systematically varied. From previous studies it is well known that the system displays multiple steady states and a bistable hysteretic switch [2], but no oscillations were found, unless an external computer controlled loop is added. Here we report pH hysteresis curves for a fixed NaOH input concentration (5mM) and for temperatures in the range from 25°C to 40°C. We found the pH of the two coexisting steady states to range from pH~3 to pH~12. For an optimal temperature ~30°C [1] we also measured pH hysteresis curves with varying NaOH input concentration from 1mM to 10mM. These measurements provide information about the extent of the hysteretic behavior in the system. Interestingly, we also report pH oscillatory behavior with an amplitude of 0.4 pH units and periods between 8 to 15 minutes. Such dynamics has not yet been observed in the system without any external negative feedback loop.

As the simple Michaelis-Menten type of models including pH sensitivity do not describe the observed dynamics (especially the oscillatory dynamics), we formulate a detailed kinetic mechanism of the reaction to account for both the observed bistable and oscillatory behaviors as follows: glucose oxidase (GOX) is a two-substrate enzyme reacting with glucose and ferricyanide ; β-D-glucose binds to the active site of the unprotonated oxidized enzyme and its coenzyme flavin adenine dinucleotide (FAD). The active site of the enzyme is protonated due to glucose creating δ-gluconolactone simultaneously with reduction of the enzyme. The δ-gluconolactone reacts with water forming gluconic acid, which decreases pH. Solution of D-glucose contains both anomers α (36%) and β (64%)[3]. The enzyme consumes β-D-glucose immediately after it mutarotates from α-D-glucose to β-D-glucose. In the next reaction step, the coenzyme FAD forms FADH^- and further reacts with one ferricyanide ion taking first hydrogen from FADH^- to form ferrocyanide, H⁺ and FAD. The second molecule of ferricyanide deprotonates FAD forming second ferrocyanide, H⁺ and deprotonated FAD [4]. Deprotonated FAD is reprotonated from active site of the enzyme forming unprotonated oxidized form of the enzyme, which can again react with another molecule of β-D-glucose[5]. The unprotonated oxidized form of the enzyme can further be protonated, which reversibly decreases affinity of the enzyme towards the substrate. According to this mechanism, we formulated a model, which accounts for both bistability and oscillatory behavior. To analyze this model we used stoichiometric network analysis [6] and numerical bifurcation theory [7].

[1] J. Singh and N. Verma, Advances in Applied Science Research 4 (3), 250-257, (2013).

[2] V. K. Vanag, D. G. Míguez and I. R. Epstein, The Journal of Chemical Physics 125, 194515:1-12, (2006).

[3] H. S. Lee and J. Hong, Journal of Biotechnology 84, 145-153, (2000).

[4] A. A. Shul’ga, M. Koudelka-Hep, N. F. de Rooij, Sensors and Actuators B 26-27, 432-435, (1995).

[5] M. Meyer, G. Wohlfahrt, J. Knäblein and D. Schoburg, Journal of computer-Aided Molecular Design 12, 425-440, (1998).

[6] V. Radojković and I. Schreiber, Phys. Chem. Chem. Phys. 20, 9910-9921, (2018).

[7] M. Kubíček and M. Marek: Computational methods in bifurcation theory and dissipative structures. Springer Verlag, New York 1983; M. Kohout, I. Schreiber and M. Marek, Compt. and Chem. Eng. 26, 517-527, (2002).