Biological Stoichiometry and Bioenergetics of Fusarium Oxysporum Ekt01/02 Proliferation Using Different Substrates in Cyanidation Wastewater

Cyanidation wastewater contains heavy metals, including high concentrations of ammonia and free cyanide (CN-). Aerobic growth of Fusarium oxysporum EKT01/02 in synthetic gold mine wastewater under different substrates was examined using biological stoichiometry and thermodynamic models in batch systems. The molecular weight of the dry biomass obtained was 23.03 g/C-mol, 33.14 g C-mol−1, and 27.06 g/C-mol in glucose with ammonia (GA), Beta vulgaris with ammonia (BA) and B. vulgaris with cyanide (BCN) cultures, respectively. The microbial growth model showed the highest biomass yield of 0.69 g dry cell/g substrate in BA cultures. The heat of reaction ΔHRXO and Gibbs energy dissipation per mole of biomass formed ΔGRXO were -652.55/-432.11 kJ/C-mol, -132.59/-471.19 kJ/C-mol, and -370.34/-225.35 kJ/C-mol-for GA, BA, and BCN cultures, respectively. The total Gibbs energy dissipated increased steadily over time and the metabolic rate of the F. oxysporum used was minimally adversely affected by the cyanidation wastewater as shown by the degree of reduction including the respiratory quotient quantified. The F. oxysporum proliferation was determined to be enthalpically driven in the cultures studied. This study revealed that the use of B. vulgaris agro-waste for the bioremediation of cyanidation wastewater is feasible and could engender sustainability of gold mining wastewater treatment processes. This article is protected by copyright. All rights reserved


INTRODUCTION
Cyanidation wastewater from gold mining operations contains high concentrations of heavy metals, ammonia and cyanide. Although the wastewater can be bioremediated, few studies report on the stoichiometric and thermodynamic analysis of such processes.
Thermodynamic analysis can predict the feasibility of a metabolic reaction and suitable conditions under which such a reaction can occur, thus addressing the feasibility of the process being studied. [1][2][3] Similarly, few studies report on the stoichiometric analysis of microbial proliferation and yield in bioremediation processes, although these factors determine the effectiveness of such bioprocesses because they are dependent on the microbial metabolic functions including cellular respiration of the isolates used. Furthermore, the stoichiometric coefficients define the efficiency of a specific microbial species in a defined process. Introducing bioenergetic analysis in such processes can further elucidate the feasibility of reactions under observation.
Gibbs energy dissipation per C-mol of biomass produced has been used to determine the balance between growth efficiency and metabolic rates using different thermodynamic models of microbial growth to justify the relationship between Gibbs energy dissipation and other parameters as the driving force for microbial growth and biomass yield in the presence of toxicants. [3][4][5] Available literature on bioenergetics, including stoichiometric analysis of microbial growth, has largely focused on the use of refined carbon sources such as glucose, sucrose, ethanol, and acetate, as substrates and/or electron donors, mostly in batch or fed-batch processes. [6][7][8][9][10][11] Although this approach can quantify the amount of Gibbs energy required to generate suitable quantities of biomass to support bioremediation reactions by varying the carbon sources used, it does not adequately describe systems in which a green chemistry approach is advocated for. [12] Thus, the challenge is to determine the energy requirements for a system in which a microorganism is grown on an agro-waste in the presence of a metabolic inhibitor such as cyanide, particularly for the bioremediation of cyanidation wastewater.
There are several reports on microbial remediation of industrial wastewater with numerous species of bacteria, fungi, algae and protozoa for the treatment of cyanidation wastewater. [13][14][15] Although the process is judged to be robust and environmentally benign, few mineral processing industries have adopted this treatment process due to the nutritional requirements essential for microbial growth on a large scale. The future of microbial remediation of wastewater depends on studies that identify renewable substrates such as agro-waste for microbial growth in bioremediation systems on a large scale. With the large quantity of agro-waste generated annually from processing of agricultural produce, this challenge can be mitigated. [16] For bioremediation processes, agro-waste can be used to provide sufficient macro-and micro-nutrient and/or carbon sources for microbial growth including biocatalytic functions to decontaminate wastewater. The presence of micro-and macro-nutrients such as proteins, soluble sugars and minerals in agro-waste can replace the use of refined carbon sources. [17] Furthermore, applying bioenergetics and biological stoichiometric analyses to systems in which agro-waste is used can further demonstrate the appropriateness of using suitable agro-waste in large scale wastewater treatment plants.
Therefore, this study seeks to promote the use of renewable feedstock in the bioremediation of cyanidation wastewater by applying biological stoichiometric and bioenergetic models to determine the functionality including requirements for a F. oxysporum species previously determined to be suitable for cyanide degradation. [12] Growth of the species on different substrates, namely glucose with ammonia as a nitrogen source (GA) for primary control experiments, Beta vulgaris (red beetroot) with ammonia as a nitrogen source (BA) for secondary control experiments and Beta vulgaris with cyanide as a nitrogen source (BCN), was undertaken for stoichiometric and bioenergetic experiments with free cyanide being the targeted contaminant for bioremediation.

Inoculum Preparation
An isolated Fusarium oxysporum EKT01/02 (Accession no: KU985430/KU985431) was cultivated in a synthetic gold mine wastewater containing metal concentrations similar to those reported in a previous study. [18] The wastewater had the following constituents (per litre): 47 mg to be the highest pH in cyanidation wastewater. Uninoculated bioreactors were used in control experiments. Samples were taken periodically for biomass concentration measurements in a Jenway 6715 UV/Visible spectrophotometer at wavelength of 300 nm in triplicate using a calibration curve relating optical density (OD) to dry biomass weight. [12] The results showed the limiting substrate concentration for growth of F. oxysporum EKT01/02 was 300 mg/L on both glucose and B. vulgaris. [19] All reagents were of analytical grade from Merck (Germany).

Agro-waste Preparation
The B. vulgaris agro-waste was obtained from an agro-processing facility in close proximity to Cape Peninsula University of Technology, Cape Town, dried at 80 °C for seven days and pulverised to less than 100 µm in a grinder (Bosch MKM 7000, Germany).

Experimental Culture Conditions
The cultivation was carried out in a 1 litre stirred tank reactor at ambient temperature i.e. 25 ± 2 °C. A 10 % (v/v) F. oxysporum culture (48 h old), was inoculated on synthetic wastewater containing 300 mg glucose as refined carbon source, followed by experiments on 300 mg pulverized B. vulgaris agro-waste and subsequently on 300 mg pulverized B. vulgaris with 100 mg CN -/L in the form of KCN added to the synthetic wastewater. An overhead stirrer fitted with a four blade propeller at 250 rpm provided mixing and aeration was at 0.4 L/min. Biomass was harvested once the carbon source was exhausted and/or when the stationary microbial growth phase was reached. Harvested biomass was centrifuged at 10 000 rpm for 10 min at 4 °C in an Avanti ® J-E centrifuge (Beckman Coulter, Inc. USA), washed thrice in sterile distilled water, dried for at least 12 h in a Duran ® vacuum desiccator (DURAN Group GmbH, Germany) until the sample weight was constant, and stored at -20 °C for further analyses. All procedures were repeated until a suitable quantity of dry biomass was obtained.

Analytical Procedures
Biomass concentration was determined daily and expressed in grams dried biomass per litre culture medium (g/L). The dry samples from the desiccator were further dried at 100 °C for 24 h in an oven to remove residual moisture, before milling with a mortar and pestle prior to elemental analysis for C, H, and N by a Thermo Flash EA 1112 series analyser in a Helium carrier gas (Thermo Fisher Scientific Inc. Waltham, USA). The analyser combusts the sample with oxygen to produce CO2, H2O and N2 which are separated in a gas chromatograph and analysed by a thermal conductivity detector. The peaks were integrated and percentages calculated for C, H, and N. All The percentage of ash in dry biomass was determined by drying at 100 °C in an oven to constant weight as previously explained. The dried biomass was ashed in an EMF 260 furnace (Kiln Contracts Pty Ltd, Cape Town, South Africa) at 550 °C for 2 h done in triplicate. The fraction of oxygen was computed by difference from the total dry weight as follows: where , , , , ℎ are fractions of -O-, -C-, -H-, -N-and ash respectively, on a dry biomass basis.

Statistical Analysis
Since all experiments were performed in triplicate, reproducibility was expressed as a standard deviation obtained from the data set (n = 3). Normality of sample distribution was assessed using Shapiro-Wilk's test (p>0.05) [20,21] with inspection of skewness and kurtosis measures and standard errors, [22,23] including visual inspection of box plots, histograms and normal Q-Q plots. Test of equality of variances in samples (homogeneity of variance) (p>0.05) was done using parametric and non-parametric Levene's test for approximately normally and non-normally distributed sample data respectively. [24,25] The statistical analyses were performed in an IBM Statistical Package for the Social Sciences (SPSS) software v24.0.

Stoichiometric Microbial Analysis
Microbial growth models represent a material balance of the system in compliance with the law of conservation of mass. The overall stoichiometry of a biological reaction can be estimated using either the method of half reactions or regularities. [10; 26] The general form of such a biological stoichiometric reaction can be described by Equation 2. [3] 1 ⁄ By sequentially decoupling the overall reaction into catabolic and anabolic reactions, assuming the electron donor is first completely catabolised and a fraction of catabolism products is used to synthesize new biomass, [3] we have Equations 3 and 4: Anabolism: Where S, A, NS, X, and P represent the energy source, electron acceptor, nitrogen source, dry biomass and reduced electron acceptori.e. the by-products. For an aerobic culture, S can be any reducible carbon source such as glucose, fructose, methanol, methane, etc., which acts as an electron donor being primarily oxidized to CO2 during the catabolic reaction. In this study, P was annotated to represent water while A was denoted to represent oxygen.
Furthermore, each empirical model was converted to a unit carbon elemental formula and the molecular weights were estimated from data in  (6) and (7) were thus stoichiometrically balanced using an elemental analysis approach.

Energy Balances for Biological Systems
The biological stoichiometry of a defined process is incomplete without the exploratory analysis of an energy balance. The standard enthalpy of formation (∆ ) and Gibbs energy (∆ ) values (at pH = 7, 101.325 kPa, and 298 K) available in literature (Table 1) were used for bioenergetic models taking into account the stoichiometric coefficients from the microbial models to determine the heat of reaction(∆ ). Furthermore, to determine experimental values for biomass enthalpy of formation (∆ ), including heat of combustion (∆ ) were obtained as described above, from which a model representing the combustion of a unit mass of biomass can be derived. The biomass enthalpy of formation was calculated for an ion containing carbon mole (ICC/mole) by multiplying the heat of combustion of the dry biomass with the mass of 1 C-mole biomass as shown in Equation (8): where is the mass of 1 C-mole of the dry biomass. The heat of reaction evolved in the synthesis of 1 C-mole of biomass was calculated using Hess's Law-Equation (9): where n are the appropriate stoichiometric coefficients.
The Gibbs energy is the major driving force of microbial growth. [1,3,5,9,29] The energy exchange that accompanies a biological growth process can be well defined from the initial state to completion under both isothermal and isobaric conditions: where ∆ , ∆ , ∆ are the Gibbs energy, enthalpy and entropy changes respectively, accompanying microbial growth. Once the bioenergetic properties of the inputs and outputs are known, values of ∆ , ∆ , and ∆ can be estimated for microbial growth models. The quantity of Gibbs energy needed to synthesise 1 C-mole of microbial biomass has been previously modelled by Heijnen and van Dijken [30,31] using an empirical correlation. Their findings indicated Gibbs energy of a reaction (∆ ) for synthesizing 1 C-mole of biomass depends mostly on the degree of reduction ( ) of the carbon donor and the number of carbon atoms as expressed in the model-Equation (11). [30,31] This model was used to determine the Gibbs energy needed to synthesise 1 C-mole of biomass and the degree of freedom can be estimated using Equation (12): This article is protected by copyright. All rights reserved

Quantifying Microbial Growth and Bioenergetic Kinetic Parameters
The Gibbs energy dissipation for biomass growth and maintenance (1⁄ ) was estimated: [11] 1 = 1 + where 1⁄ was the Gibbs energy requirement for synthesising a unit C-mole of biomass as defined in Equation (12), with being the maintenance Gibbs energy, approximated to 4.5 kJ C-mol -1 h -1 at 298K. The specific microbial growth rate ( ) was estimated using Equation (14): where and −1 were biomass concentrations (g dry biomass weight/L) at times −1 (h), respectively.

Elemental Analysis
Elemental analysis of the biomass ( The mass of 1 C-mole of biomass and the elemental formula were quantified to be within the range of previous research - Table 3. The higher C-molar mass (33.14 g C-mole) observed when cultures were grown on agro-waste can be attributed to the excess macro-and micro nutrients available within B. vulgaris which were not present in the refined carbon source used and/or the rigidification of the fungal cell membranes including accumulations of extracellular polymeric substances, as the biomass strived to protect itself from cyanide toxicity. The degree of reduction indicates there are more available electrons during cyanide biodegradation which may be linked to the constituents available in B. vulgaris. [33] The degree of reduction on agro-waste (BA) was also similar to cultures in which glucose was used, an indication that use of agro-waste as a carbon source has minimal impact on the performance of the cultures. In comparison with similar filamentous fungi reported by Duboc et al, [26] the degree of reduction and dry biomass weight of GA agrees with their report.

Microbial Growth Model
The microbial growth models used to represent aerobic growth of F. oxysporum on GA, BA, and BCN are shown in Table 4, organized into catabolic, anabolic and overall metabolic stoichiometric reactions. The catabolic equations represent the oxidation of the carbon source (glucose or B. vulgaris waste). The nitrogen source (ammonia or cyanide) reacts with the catabolic products to produce biomass as shown in the anabolic equations. In reality, catabolic and anabolic processes are interdependent during growth, although they are theoretically constructed independently to elucidate the metabolism process. The overall metabolic description of a process is what is required to describe the actual biomass generated for bioremediation studies.
During catabolism, oxidation of the carbon source provides the ATP required to catalyse the anabolic mechanisms. In turn, anabolism conserves the chemical form of the non-thermal energy contained within the carbon source. Therefore, metabolism can be said to be an energy conservation process i.e. all the non-thermal energy remaining within the carbon source for microbial growth processes. [29] Meanwhile, for complete aerobic oxidation of the substrate, a nonconservative process is followed with minimal non-thermal energy being required for the conservation of energy within biomass. The growth efficiency can be estimated as a quantifiable ratio between available electrons (AE) in conserved biomass to those that are available in the nonconservative reactions. The AE can be classified as a degree of reduction for a unit carbon atom.
This article is protected by copyright. All rights reserved For growth on BCN, cyanide can be converted to cyanate by cyanide monoxygenase, followed by conversion of cyanate to ammonia and carbon dioxide with cyanate as catalyst. Alternatively, cyanide can be oxidised directly using cyanide dioxygenase to produce ammonia and carbon dioxide as shown in Equation (15). [34,35] ( ) + 2( ) + 2 ( ) + + ( ) → 2( ) + 3( ) + ( ) + (15) The ammonia by-product can be consumed with other by-products to generate biomass including the carbon dioxide from cyanide biocatalytic decomposition which accounts for the higher molar production of carbon dioxide observed in BCN cultures compared with those grown in BA. This contributes to the higher AE observed in the BCN cultures.

Bioenergetic Parameters
In addition, changes in thermodynamic properties can be calculated, although not precisely, by using the microbial growth models and known thermodynamic properties of reactants and products except for biomass for which a true standard state is unknown. The ∆ determinations as described earlier in a bomb calorimeter were -12.23 ± 0.02, -13.15 ± 0.03, -15.54 ± 0.06 kJ/g for biomass obtained from GA, BA, and BCN cultures, respectively. All measurements were performed in triplicate. The experimental enthalpy of combustion for B. vulgaris waste was -431.1 ± 0.3 kJ C-mol -1 (n = 6). Generally, Thornton's rule [36] can be used for estimating heat of combustion of organic substances, as for many organic substances, their heat of combustion is directly proportional to the number of atoms of oxygen consumed during combustion, as described by Equation (16):  Table   1 were used to determine the changes in bioenergetic parameters accompanying the aerobic growth of the F. oxysporum isolate used as shown in Table 5.
The more exothermic ∆ calculated using Equation (8) indicated higher values for growth on B. vulgaris, than on glucose as previously observed. The accuracy of these values is a function of the validity of the molecular formula of the carbohydrate used to represent the B. vulgaris agrowaste which has a direct influence on the accuracy of the bioenergetic parameter determinations. ∆ and ∆ shown in Table 5 is an indication of spontaneous metabolic processes in both refined and agro-waste carbon source. Furthermore, from bioenergetic analysis, the growth on BA was hypothetically spontaneous at varying temperature due to negative enthalpy and positive entropy changes for such a system. This may be directly linked to other added nutritional value components such as proteins, vitamins, and other minerals besides the available carbohydrates which are available in the agro-waste used thus can dissociate at different rates depending on the culture temperature. The estimated change in entropy values in all cases was determined to be weak, therefore, the growth processes were observed to be enthalpically driven which is similar to most previous reports. [2,26,29] The results in Figure 1, show a gradual increase in the total Gibbs energy dissipated over time.
Previous reports indicated that the Gibbs energy dissipation for biomass growth including maintenance (1⁄ ) increases gradually in batch cultures, [6,37] achieving increasing metabolic rates although resulting in low biomass yield. The microbial growth model showed highest biomass yield based on substrate and oxygen in BA cultures - Table 6. The results in Table 6 and Figure 1 concur with observations in previous studies that showed an increase in energy requirements is largely due to constraints in synthesising biomass from a carbon and/or an energy source which causes reduction in specific growth rate as the process approaches the stationary phase. [1,6] By comparison, the growth on BA showed the lowest energy requirements for microbial growth with the highest dry biomass yield and maximum specific growth rate as shown in Table 6, while the Gibbs energy dissipated on GA was quantifiably large resulting in a lower maximum specific growth rate and dry biomass yield. The increase in energy requirements occurred at a specific growth rate of 0.0008 h -1 after 6 days on GA, meanwhile, prior to that, i.e. after 4 days, there was a decrease in energy requirements on cultures grown in BCN due to an increase in the specific growth rate from 0.0032 to 0.004 h -1 prior to cultures reaching the stationary phase. The relatively high biomass yield in BCN compared to GA may be due to the combined effect of an elongated catabolic pathway, presence of stored mucilage in the cells and the requirement to assimilate micro-and macro-nutrients available in B. vulgaris. [4,17] The biomass yield based on oxygen consumption varied but the respiratory quotient (R.Q) was similar for the isolate in all cultures studied, an indication that the metabolic performance of the isolate was largely identical irrespective of the substrate used.    Table 1 Thermodynamic properties of compounds used at 298.15 K and 101.325 kPa [10] Substance Formula ∆ (kJ/mol) Glucose 140.3 a The data was adapted from Finch et al. [28] Table 2 Elemental analysis of dry biomass as a mass fraction (g/100 g dry biomass) measured in triplicate. The standard deviation is indicated in the parenthesis (n=3)  Table 3 Elemental formula of filamentous fungi and mass of 1 C-mole for dry biomass ( ) and the degree of reduction(  Table 4 Microbial growth equations for aerobic growth of F. oxysporum on GA, BA, and BCN based on suggested model Equations (3) and (4) [3] Growth on GA Catabolism: 6 12 6( ) + 6 2( ) → 6 2 ( ) + 6 2( ) Anabolism: 1 Table 6 Kinetic parameters of F. oxysporum on glucose with ammonia (GA), Beta vulgaris with ammonia (BA) and Beta vulgaris with cyanide (BCN). The standard deviation is indicated in parenthesis (n=3)