An insight into kinetics and thermodynamics of gamma-aminobutyric acid production by Enterococcus faecium CFR 3003 in batch fermentation

Abstract

In the present study, a kinetic and thermodynamic model for production of gamma-aminobutyric acid (GABA) using Enterococcus faecium CFR 3003 is proposed. Differential production of GABA in complex-minimal media indicated that the minimal medium (Modified TYG: Tryptone, Yeast extract and Glucose) supported the highest production of GABA. The influence of variables such as monosodium glutamate (MSG) concentration (1–5 %), temperature (25–60 °C) and pH (4.0–8.0) on GABA production was studied. The kinetics on the rate of MSG consumption (rs) as a function of MSG concentration [S] was studied by assuming Monod’s model for the MSG-uninhibited region and Briggs–Haldane model for the MSG-inhibited domain. The experimental kinetic parameters determined were found to be in good agreement with the predicted ones. Arrhenius model was used to calculate the thermodynamic parameters. The activation energy for the growth, activation enthalpy and entropy for GABA formation were found to be 80.13, 87.29 and 0.273 kJ mol⁻¹ K⁻¹, respectively. A process product yield (Y_P/S) of 0.64 g GABA g⁻¹ MSG could be reached with a volumetric rate of GABA formation, Q_P (0.159 g l⁻¹ h⁻¹) and MSG consumption Q_S (0.145 g l⁻¹ h⁻¹), respectively, under optimized conditions. This contributed to a GABA concentration of 8 g l⁻¹ at the end of 48 h of fermentation time, which is 2.9-fold higher than that under unoptimized conditions.

Introduction

Glutamate decarboxylase (GAD; 4.1.1.15) is widely distributed in nature from single cell organisms to mammalians (Matsukawa and Ueno 2007) and mediates decarboxylation of glutamate to gamma-aminobutyric acid (GABA). The mechanism involves the removal of α-carboxyl group from glutamate, adjacent to the α-amino group to produce GABA by releasing CO₂. GABA is a four-carbon amino acid and was discovered as an integral part of mammalian brain in 1950 (Roberts and Frankel 1950). It functions as a chief inhibitory neurotransmitter in brain and plays physiological roles in preventing diabetic conditions (Hagiwara et al. 2004), inducing insulin secretion from the pancreas (Adeghate and Ponery 2002), treating inhibitory motor disorders (Cho et al. 2007), and reducing blood pleasure (Inoue et al. 2003). Abnormalities in GAD function and reduced GABA levels are reported in people with many neurological disorders. A few lactic acid bacteria (LAB) are found to withstand acid stress encountered in food and in the gastrointestinal tract by expressing GAD in the presence of glutamate, by accepting proton. Thus, the extracellular glutamate and proton are taken up by the cells which further get decarboxylated to GABA by the action of GAD. The produced GABA is excreted out and the consumed intracellular proton results in the increase in intracellular pH (Cotter and Hill, 2003). Several LAB, Lactobacillus buchneri, (Cho et al. 2007), Lactobacillus brevis NCL912 (Li et al. 2008), Lactobacillus plantarum (Di Cagno et al. 2009), Lactococcus lactis subsp. lactis (Lu et al. 2009), Lactobacillus bulgaricus CFR 2028 (Gangaraju et al. 2014), Streptococcus salivarius subsp. thermophilus Y2 (Yang et al. 2008), and Enterococcus avium G-15 (Tamura et al. 2010) capable of producing GABA by expressing GAD have been reported. GABA-producing LAB have the potential for the development of GABA-enriched functional foods.

In the present communication, attempts were made to study the whole cell kinetics and thermodynamic aspects of GABA production using Enterococcus faecium CFR 3003. The advantage and the significance of carrying out a kinetic study for the whole cells is that working with purified enzyme adds to the downstream process because the enzyme, GAD, is localized intracellularlly. Thus, this nesseciates cell lysis using a suitable cell disruptor, and further purification steps should be performed to obtain a pure enzyme prior to the kinetic study. In spite of the fact that there is a need to consider cell growth and substrate transformation rate, the whole cell process works out as economical when compared with that of the process using purified enzyme. This is mainly due to the costs involved in purification. Also, there is a lack of scientific information on the kinetic and thermodynamic parameters on the GABA production process. The thermodynamic parameters estimated at the laboratory level hold good for the industrial scale, as these parameters are independent of reactor scale. The kinetic parameters estimated might vary a bit in the industrial scale due to mass transfer effects. If the industrial process is operated in a reaction-controlled region, the mass transfer effects can be neglected. The data generated in the study gave an insight into the kinetic and thermodynamic phenomena involved in GABA production and, thus, this could be an essential step in bridging the existing gap in correlating the basic scientific data towards translating the same into industrial process engineering reality.

Materials and methods

Chemicals

Monosodium glutamate (MSG) and GABA were procured from Sigma-Aldrich, Saint Louis, MO, USA. All other chemicals used were of analytical grade unless otherwise stated.

Microorganism

Initial screening: A number of lactic cultures available with the researcher’s group were screened for production of GABA. The strain Lactobacillus bulgaricus CFR 2028 was obtained from a culture collection of the Central Food Technological Research Institute (CFTRI), and could produce 22.7 mM GABA (2.34 g l⁻¹) in MRS medium, while after single factor optimization (minimal TYG medium supplemented with 2 % substrate concentration at 37 °C with an initial medium pH of 7.0), we could reach a yield of 46 mM (4.80 g l⁻¹) as reported in our earlier communication (Gangaraju et al. 2014). The screening trails were continued further for selecting high-yielding GABA strains. We were successful in isolating an organism from fermented traditional indigenous ayurvedic preparations producing much higher yields than L. bulgaricus CFR 2028. The morphological analysis of the isolated organism was found to be coccus and was identified to be E. faecium by Internal transcribed spacer (ITS) amplification using oliginucleotides 5′-GTCGTAACAAGGTAGCCGTA-3′ (forward primer ITSF) and 5′-GCCAAGGCATCCACC-3′ (reverse primer ITSReub). The genomic DNA was extracted using a MDI Genomic DNA Miniprep Kit (Advanced Microdevices, Ambala Cantt, India). The PCR cycling program for amplification was 95 °C for 5 min, 30 cycles of 94 °C for 1 min, 53.4 °C for 1 min 45 s and 72 °C for 1 min 20 s, with a final extension at 72 °C for 15 min. A PCR product was sequenced by Xceleris Genomics Lab, Ahmedabad, India. The strain identified as E. faecium has now been deposited at the in-house culture collection (Central Food Repository; CFR) and given the number 3003.

Comprehensive evaluation and selection of the potential medium for industrial production of GABA

Evaluation of five different media was carried out and the composition of the same is as follows.

• de Man Rogosa Sharpe medium [MRS₁ (g l⁻¹)]: Protease peptone, 10; Beef extract, 10; Yeast extract, 5; Dextrose, 20; Polysorbate 80, 1; Ammonium citrate, 2; Sodium acetate, 5; Magnesium sulfate, 0.1; Manganese sulfate, 0.05; Dipotassium phosphate, 2.0; MSG, 10; initial pH, 6 (Ratanaburee et al. 2011).

• Modified MRS (g l⁻¹): Yeast extract, 6; Dextrose, 25; Polysorbate 80, 1; Tryptone, 6; MgSO₄ ^.7H₂O, 0.2; MnSO₄ ^.4H₂O, 0.05; MSG, 10; initial pH, 5 (Li et al. 2008).

• MRS₅ (g l⁻¹): Protease peptone, 10; Beef extract, 10; Yeast extract, 5; Dextrose, 20; Polysorbate 80, 1; Ammonium citrate, 2; Sodium acetate, 5; MSG, 50; initial pH, 6.5 (Cho et al. 2007).

• GYP (g l⁻¹): Dextrose, 10; Yeast extract, 10; Protease peptone, 5; Sodium acetate, 2; MgSO₄ ^.7H₂O, 0.2; MnSO₄ ^.4H₂O, 0.01; FeSO₄ ^.7H₂O, 0.01; NaCl, 0.01; MSG, 10; initial pH, 6.8 (Choi et al. 2006).

• TYG (g l⁻¹): Dextrose, 10; Yeast extract, 5; Tryptone, 5, MSG, 10; initial pH, 7 (Nomura et al. 1999).

The fermentation was carried by inoculating different media (100 ml in 500-ml Erlenmeyer flasks) with 24-h-old culture and the flasks were incubated at 37 °C for 48 h.

Effect of substrate concentration on GABA production

The effect of substrate concentration on GABA production as a function of fermentation time was studied. TYG medium (100 ml in 500-ml Erlenmeyer flasks) supplemented with 1, 2, 3, 4 and 5 % MSG was inoculated with 24-h-old culture of E. faecium CFR 3003. The initial media pH was adjusted to 7.0. The flasks were incubated at 37 °C and samples were drawn at regular intervals.

Effect of temperature on growth of E. faecium CFR 3003 and GABA production

To study the effect of temperature on growth and GABA production by E. faecium CFR 3003, TYG medium supplemented with 2 % MSG was inoculated with 24-h-old culture and the flasks were incubated at different temperatures (25, 30, 35, 37, 40, 45, 50, 55, 60 °C) for 48 h. The initial media pH was adjusted to 7.0. The samples were drawn at regular time intervals.

Effect of pH on GABA production

TYG medium containing 2 % MSG at the initial pH values of 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, and 8 were used for cultivation of E. faecium CFR 3003 to study the effect of initial medium pH on GABA production. The flasks were incubated at 37 °C.

Analytical methods

Quantitative analysis GABA and residual MSG using HPLC

Supernatant (10 μl) was derivatized with ortho-phthaladehyde-mercaptoethanol [OPA-MCE (20 μl), prepared by dissolving 10 mg OPA and 10 μl MCE in 2.5 ml acetonitrile] reagent and boric acid buffer (0.4 M, pH 10.4; 100 μl). Derivatization was carried out at 30 °C ± 2 for 5 min and the reaction mixture was filtered using microfilter (Millipore, 0.45 μm), and analyzed by HPLC Schimadzu LC-20 AD (Schimadzu , Kyoto, Japan) equipped with a UV detector (SPA-20A). The samples (20 μl) were injected using a HPLC injector syringe (Hamilton) and the analysis was carried out at room temperature (30 °C ± 2) using aan nalytical column (Hypersil ODS). The flow-rate of mobile phase [sodium acetate (1.64 g) and triethylamine (TEA, 200 μl) in 1000 ml 20 % (v/v) acetonitrile] was 0.8 ml min⁻¹ and the wavelength for detection was 338 nm (Li et al. 2008).

Biomass estimation

Growth of E. faecium CFR 3003, in terms of optical density (OD) was determined by measuring the culture turbidity at 600 nm during the course of fermentation (Robbe-Saule et al. 2006). Biomass (g l⁻¹) was obtained from the standard graph of OD versus biomass.

pH measurement

The pH of the supernatant was measured using a digital pH meter equipped with a glass electrode over a range 0.0–14.0 (Analab Scientific Instruments, Gujarat, India).

Theoretical and modelling aspects

Effect of substrate concentration on GABA production

Monod’s Model: MSG uninhibited domain

To determine the relationship between MSG consumption rate (rs) and MSG concentration [S], the differential method of analysis was employed. The experimental rate values were determined from MSG concentration versus time. Monod’s model in the form of below equation was assumed for its simplicity for the region of the MSG-uninhibited domain.

$$ -\mathrm{rs}=\frac{\mathrm{rs},\ \max *\left[\mathrm{S}\right]}{\mathrm{Ks}+\left[\mathrm{S}\right]} $$

(1)

$$ -\frac{1}{\mathrm{rs}}=\frac{\mathrm{Ks}}{\mathrm{rs}, \max *\left[\mathrm{S}\right]}+\frac{1}{\mathrm{rs}, \max } $$

(2)

The kinetic parameters, Ks (saturation constant) and rs,max (maximum specific rate of MSG consumption) were obtained by plotting experimentally measured values of $-\frac{1}{\mathrm{rs}}$ versus $\frac{1}{\left[\mathrm{S}\right]}$.

Briggs–Haldane Model: MSG inhibited domain

The dynamics of MSG inhibitory effects on GABA production was evaluated by the Briggs–Haldane model. Haldane’s inhibitory kinetics is as follows:

$$ -\mathrm{rs}=\frac{\mathrm{rs}, \max *\left[\mathrm{S}\right]}{\left(\mathrm{Ks}+\left[\mathrm{S}\right]+\frac{{\left[\mathrm{S}\right]}^2}{\mathrm{Ki}}\right)} $$

(3)

$$ -\frac{\left[\mathrm{S}\right]}{\mathrm{rs}}=\frac{\mathrm{Ks}}{\mathrm{rs}, \max }+\frac{\left[\mathrm{S}\right]}{\mathrm{rs}, \max }+\frac{{\left[\mathrm{S}\right]}^2}{\mathrm{rs}, \max *\mathrm{Ki}} $$

(4)

By fitting the experimental data into the above simplified equation and by plotting $-\frac{\left[\mathrm{S}\right]}{\mathrm{rs}}$ versus $\left[\mathrm{S}\right]$, kinetic parameters were estimated (Sukumar 2009).

Thermodynamics

Effect of temperature on the growth of E. faecium CFR 3003 and deactivation kinetics

To describe the relationship of growth of E. faecium CFR 3003 and its thermal deactivation, the empirical Arrhenius equation given below was employed:

$$ \upmu =\mathrm{A}.{\mathrm{e}}^{-\frac{\mathrm{Ea}}{\mathrm{RT}}} $$

(5)

$$ \ln \left(\upmu \right)= \ln \left(\mathrm{A}\right)-\frac{\mathrm{Ea}}{\mathrm{RT}} $$

(6)

By plotting $\ln\;\left(\mu \right)$ against $\frac{1}{\mathrm{T}}$, and substituting Arrhenius constant (A), activation energy (Ea) was determined. Universal gas constant (R) = 8.314 J mol⁻¹ K⁻¹. μ is the specific growth rate.

Effect of temperature on GABA production and deactivation kinetics

To describe the relationship of GABA production and thermal deactivation, the empirical Arrhenius equation in terms of specific rate of product formation, $\left(\mathrm{qp}\right)$ was used as below (Rajoka et al. 2005).

$$ \mathrm{qp}=\mathrm{T}.\frac{{\mathrm{k}}_{\mathrm{B}}}{h}.{\mathrm{e}}^{\frac{\varDelta {\mathrm{S}}^{*}}{\mathrm{R}}}.{\mathrm{e}}^{\frac{-\varDelta {\mathrm{H}}^{*}}{\mathrm{R}\mathrm{T}}} $$

(7)

$$ \ln \left(\frac{\mathrm{qp}}{\mathrm{T}}\right)= \ln \left(\frac{{\mathrm{k}}_{\mathrm{B}}}{h}\right)+\frac{\varDelta {\mathrm{S}}^{*}}{\mathrm{R}}-\frac{\varDelta {\mathrm{H}}^{*}}{\mathrm{R}\mathrm{T}} $$

(8)

By plotting $\ln \left(\frac{\mathrm{qp}}{\mathrm{T}}\right)$ against $\frac{1}{\mathrm{T}}$, ${\varDelta \mathrm{S}}^{*}$and ${\varDelta \mathrm{H}}^{*}$can be determined. Planck’s constant (h) = 6.63 × 10⁻³⁴ Js; Boltzman constant ${\Big(\mathrm{k}}_{\mathrm{B}\;}\Big)$ = 1.38 × 10⁻²³ J K⁻¹; Avogadro’s no. (N) = 6.023 × 10²³ mol⁻¹; Universal gas constant (R) = 8.314 J mol⁻¹ K⁻¹. Specific rate of product formation, $\left(\mathrm{qp}\right)$ was calculated by multiplying specific growth rate $\Big(\upmu$) with Y_p/x.

Comparison of kinetic and thermodynamic parameters of GABA production using E. Faecium CFR 3003 in 2-L shake flask experiments

The kinetics of GABA production, cell mass formation, substrate consumption parameters and their thermodynamics under optimized fermentation conditions in 2-L shake flask experiments were evaluated. The procedure followed by Rajoka et al. (2005) and Ahmed et al. (2013) was adopted for determining kinetic parameters and thermodynamic parameters. Specific growth rate (μ) was determined as a slope of straight line between ln (X/X₀) and fermentation time (h). Specific GABA yield (Y_P/X; g GABA g⁻¹ cells) was determined using the relationship dP/dX; Process GABA yield (Y_P/S; g GABA g⁻¹ MSG) was determined from dP/dS. Specific rate of GABA formation (q_p) and specific rate of MSG consumption (q_s) were determined using (Y_P/X)^.μ and (dS/dX)^.μ_, respectively. Growth yield coefficient (Y_X/S) was g cells formed g⁻¹ MSG utilized and cell mass productivity (Q_x) expressed as g cells l⁻¹ h⁻¹. The volumetric rate of GABA formation and MSG consumption (Q_P and Q_S) were determined from a plot of GABA (g l⁻¹) and MSG (g l⁻¹) as a function of fermentation time, respectively.

Nucleotide sequence

Based on the blast search result of ITS amplified region, the organism was identified as Enterococcus faecium. The organism showed similarity 100 % with E. faecium T110 genome, 99 % with 16S–23S rRNA spacer DNA of E. faecium, 93 % similarity with E. hirae, 92 % similarity with E. mundtti, and 90 % similarity with E. faecalis. The 16–23S rRNA spacer amplified region is as follows:

CTACACAATTTGTTTTTACTTTGTTCAGTTTTGAGAGGTTTACTCTCAAACACTTTTGTTCATTGAAAACTGGATATTTGAAGTAAATGTAAGTAATACAAACCGAGAACACCGCGTTGAATGAGTTTTTTAATAAGTTCAATTGCTTATTTTCTTGATCTAACTTCTATCGCTAGAAGAAGGATCAAAACCCAACCGCAAGGTTGATAA

Results and discussion

Differential production of GABA in complex-minimal media

Maximum GABA production by E. faecium CFR 3003 was observed in TYG medium (Fig. 1). Under the conditions of nutrient stress in minimal TYG medium, cells might have switched to survival mode, catabolising the available MSG and other nutrients for GABA production. The results are also in line with our previous findings for GABA production by L. bulgaricus CFR 2028 (Gangaraju et al. 2014). The possible reason for lower GABA production in other media could be that the regulatory factors which govern and favour the growth might not favour the efficient expression of GAD (Castanie-Cornet and Foster 2001).

Fig. 1

GABA production in complex-minimal media

Effect of substrate concentration on GABA production

An increase in GABA production was observed with increase in MSG concentration up to 2 %, steep decrease was observed over 2 % (Fig. 2). The decrease in GABA production over 2 % MSG is attributed to its inhibition and same is evaluated by assuming Briggs–Haldane substrate inhibition model. E. faecium CFR 3003 produced highest GABA of 5.5 g l⁻¹ at an initial concentration of 2 % MSG, and this was considered as optimum MSG level for further experiments.

Fig. 2

Effect of MSG concentration on GABA production

Monod’s model: MSG uninhibited domain

GABA production at an initial MSG concentration of 2 % indicating that the system follows the classical Monod model (Samanta et al. 2008). The suitability of the Monod equation to fit the data was evaluated using the experimental data (rs and [S]) within the MSG uninhibited region. Simulated work led to a reasonable straight line with the following equations for the plot $-\frac{1}{\mathrm{rs}}$ versus $\frac{1}{\left[\mathrm{S}\right]}$ with a slope and intercept of $\frac{\mathrm{Ks}}{\mathrm{rs}, \max }$ and $\frac{1}{\mathrm{rs}, \max }$ , respectively.

$$ \begin{array}{cc}\hfill -\frac{1}{\mathrm{rs}}=17.48\frac{1}{\left[\mathrm{S}\right]}+0.77;{\mathrm{R}}^2=0.980\hfill & \hfill \left(1\%\ \mathrm{MSG}\right)\hfill \end{array} $$

(9)

$$ \begin{array}{cc}\hfill -\frac{1}{\mathrm{rs}}=17.308\frac{1}{\left[\mathrm{S}\right]}+0.50;{\mathrm{R}}^2=0.988\hfill & \hfill \left(2\%\ \mathrm{MSG}\right)\hfill \end{array} $$

(10)

Kinetic parameters of Monod’s model were determined using the experimental data used for deriving the above equations. The predicted profiles of $\hbox{--} \mathrm{rs}$ versus $\left[\mathrm{S}\right]$ are compared with experimental data (Fig. 3). Monod’s model agreed well with the experimental results at 1 and 2 % MSG with standard deviation of 0.10 and 0.06, respectively.

Fig. 3

Rate of MSG consumption versus MSG concentration: a MSG = 1 %, b MSG = 2 %

Thus, it can be concluded that the reaction engineering performance of the present system on the MSG-uninhibited domain can effectively be described by the Monod model. The average numerical values for rs,max and Ks were found to be 1.64 mM h⁻¹ and 28.53 mM, respectively. The rate equation for MSG-uninhibited domain may therefore be written as

$$ -\mathrm{rs}=\frac{1.64*\left[\mathrm{S}\right]}{28.53+\left[\mathrm{S}\right]} $$

(11)

Brigg’s–Haldane model: MSG inhibited domain

The following equations were obtained for 3, 4 and 5 % MSG concentration, respectively, by assuming Briggs–Haldane model and by fitting a 2nd-order polynomial equation.

$$ \begin{array}{cc}\hfill -\frac{\left[\mathrm{S}\right]}{\mathrm{rs}}=0.0028{\left[\mathrm{S}\right]}^2+0.533\left[\mathrm{S}\right]+19.73;{\mathrm{R}}^2=0.998\hfill & \hfill \left(3\%\ \mathrm{MSG}\right)\hfill \end{array} $$

(12)

$$ -\frac{\left[\mathrm{S}\right]}{\mathrm{rs}}=-0.0017{\left[\mathrm{S}\right]}^2+0.681\left[\mathrm{S}\right]+45.23;{\mathrm{R}}^2=0.993\left(4\%\ \mathrm{MSG}\right) $$

(13)

$$ \begin{array}{cc}\hfill -\frac{\left[\mathrm{S}\right]}{\mathrm{rs}}=0.0028{\left[\mathrm{S}\right]}^2+0.916\left[\mathrm{S}\right]+64.59;{\mathrm{R}}^2=0.991\hfill & \hfill \left(5\%\ \mathrm{MSG}\right)\hfill \end{array} $$

(14)

Kinetic parameters of Briggs–Haldane model were determined using the experimental data used for deriving the above equations. The predicted profiles of $\hbox{--} \mathrm{rs}$ versus $\left[\mathrm{S}\right]$ are compared with experimental data (Fig. 4). Briggs–Haldane model fitted the data with standard deviation of 0.04, 0.34 and 0.19 for 3, 4 and 5 % MSG respectively.

Fig. 4

Rate of MSG consumption versus MSG concentration (a) [MSG] = 3 % (b) [MSG] = 4 % (c) [MSG] = 5 %

The variation of kinetic parameters with initial MSG concentration is presented in Table 1. The rs,max decreased and Ks increased with initial MSG concentration. This increase in Ks values can be correlated to the decrease in the affinity for MSG by E. faecium CFR 3003 in the MSG-inhibited domain (Agarry and Solomon 2008). It is thus apparent that the inhibition becomes prominent as the initial MSG concentration increased above 2 %. Thus, rs,max varies with initial MSG concentration and is found to be a strong function of initial MSG concentration. The overall rate equation under MSG inhibited domain may therefore be written as

Table 1 Kinetic parameters obtained from Monod and Briggs–Haldane models

$$ -\mathrm{rs}=\frac{1.47*\left[\mathrm{S}\right]}{57.60+\left[\mathrm{S}\right]+\frac{{\left[\mathrm{S}\right]}^2}{306.30}} $$

(15)

The overall magnitude of Ki in the present case is greater than the Ks value indicating that the overall system dynamics cannot be explained by the Briggs–Haldane model alone. This necessitates the addition of two biological switches, which will be active in a given present system. An integrated equation capable of describing the dynamics of overall MSG consumption rate may therefore be compounded as follows (Samanta et al. 2008):

$$ -\mathrm{rs}=\upbeta \frac{1.64*\left[\mathrm{S}\right]}{28.53+\left[\mathrm{S}\right]}+{\textsf{C}\hspace{-1.7ex}{=}} \frac{1.47*\left[\mathrm{S}\right]}{57.60+\left[\mathrm{S}\right]+\frac{{\left[\mathrm{S}\right]}^2}{306.30}} $$

(16)

where, β and € are dynamic biological switches. β = 1 and € = 0 when MSG ≤ 2 %; β = 0 and € = 1 when MSG ≥ 2 %. The critical value MSG concentration for the present study is identified as 2 %.

Thermodynamics

Effect of temperature on growth of E. faecium CFR 3003 and GABA production

Incubation temperature plays a significant role on the growth of organism and GABA production. E. faecium CFR 3003 produced 5.3 g l⁻¹ of GABA at 37 °C at the end of 48 h. The results are in line with the studies by Komatsuzaki et al. (2005) and Huang (2007) in which they have reported that L. paracasei NFRI 7415 and L. brevis CGMCC 1306, respectively produced highest GABA at 37 °C.

Lower activation energy (E_a) and lower enthalpy and entropy of activation for product formation are considered as a prerequisite to check the stability of the culture at incubation temperature (Declerck et al. 2003). The activation energy (E_a) for any cell growth are reported to be inbetween 34 and 80 kJ mol⁻¹ (Aiba et al. 1973), and in the present case for E. faecium CFR 3003 it was estimated to be 80.13 kJ mol⁻¹ (Fig. 5). The activation enthalpy for GABA formation (${\varDelta \mathrm{H}}^{*}$ _(f)) was graphically determined from Fig. 6. The values of thermodynamic parameters determined are listed in Table 2. The activation enthalpy for GABA formation (${\varDelta \mathrm{H}}^{*}$ _(f)) and that for thermal deactivation (${\varDelta \mathrm{H}}^{*}$ _(d)) were found to be 87.29 and 100.57 kJ mol⁻¹, respectively. The value of (${\varDelta \mathrm{H}}^{*}$ _(d)) was found higher than that for (${\varDelta \mathrm{H}}^{*}$ _(f)). This infers that its deactivation rate increases much faster with temperature than the product formation rate; as a result, the decline in GABA productivity was observed above a threshold value (Rajoka et al. 2005).

Fig. 5

Arrhenius plot for the graphical estimation of thermodynamic parameters for growth of E. faecium CFR 3003

Fig. 6

Arrhenius plot for the graphical estimation of thermodynamic parameters of both GABA formation and thermal deactivation

Table 2 Thermodynamic parameters for GABA production and deactivation of GABA production by E. faecium CFR 3003

The activation entropy for GABA formation (${\varDelta \mathrm{S}}^{*}$ _(f)) was found to be −0.273 kJ mol⁻¹ K⁻¹. This negative sign reflects that deactivation phenomenon implies a little disorderness during GABA formation at temperature up to 37 °C. The activation entropy for thermal deactivation was found to be positive (${\varDelta \mathrm{S}}^{*}$ _(d) = 0.579 kJ mol⁻¹ K⁻¹), reflecting the disorder happening during GABA formation above an optimum temperature.

There is a lack of scientific information on the kinetic and thermodynamic parameters on GABA production process. The thermodynamic parameters estimated at the laboratory level hold good for the industrial scale, as these parameters are independent of reactor scale. The kinetic parameters estimated might vary a bit in industrial scale due to mass transfer effects. If the industrial process is operated in a reaction-controlled region, the mass transfer effects can be neglected. Thus, the data generated are an essential step in bridging the existing gap in correlating the basic scientific data towards translating the same into industrial process engineering reality.

Effect of pH on GABA production

The optimum pH for GABA production by E. faecium CFR 3003 was found to be 6.5 (Fig. 7). Optimum pH for GABA production is reported to be strain-specific (Li et al. 2010; Yang et al. 2008). Lactococcus lactis produced highest amount of GABA (7.2 g l⁻¹) at an optimum pH ranged from 7.5 to 8.0 (Lu et al. 2009), and L. plantarum DSM19463 synthesized the maximum GABA at pH 6.0 (Di Cagno et al. 2009).

Fig. 7

Effect of initial medium pH on GABA production

Comparison of kinetic and thermodynamic parameters of GABA production using E. Faecium CFR 3003 in 2 L shake flask experiments

The kinetic parameters determined for GABA production by E. faecium CFR 3003 under optimized conditions (pH 6.0; MSG concentration 2 %, temperature of incubation 37 °C) are reported in Table 3. A process product yield (Y_P/S) of 0.64 g GABA g⁻¹ MSG could be reached with a volumetric rate of GABA formation, Q_P (0.159 g l⁻¹ h⁻¹) and MSG consumption Q_S (0.145 g l⁻¹ h⁻¹), respectively under optimized conditions.

Table 3 Growth and GABA production kinetics of E. faecium CFR 3003

The value of fermentation parameter obtained under optimized conditions is higher than those reported previously (Park and Oh 2007; Cho et al. 2011) and this could be attributed to the presence of high metabolic activities of the organism (Ahmed et al. 2013).

Conclusion

In conclusion, E. faecium CFR 3003 was efficient in GABA production and evaluation of GABA production in different complex-minimal media, suggesting the maximum GABA in minimal medium. The influence of variables such MSG concentration, temperature and pH on GABA production revealed its optimum points (2 % MSG, 37 °C and pH 6.5). The use of the desired MSG concentration was critical for better GABA levels. This indicated that MSG concentration in the medium greatly affects the GABA production rate. The correlation between rs, max and Ks was studied by fitting Monod’s and Briggs–Haldane models. The rs,max decreased and Ks increased with initial MSG concentration. This increase in Ks values was correlated to the decrease in the affinity for MSG by E. faecium CFR 3003 in the MSG-inhibited domain. The thermodynamic parameters for GABA production using E. faecium CFR 3003 have also been estimated by assuming suitable models. The activation energy for the growth of E. faecium CFR 3003, activation enthalpy and entropy for GABA formation were found to be 81.13, 87.29 and 0.273 kJ mol⁻¹ K⁻¹, respectively. A process product yield (Y_P/S) of 0.64 g GABA g⁻¹ MSG could be reached with a volumetric rate of GABA formation, Q_P (0.159 g l⁻¹ h⁻¹) and MSG consumption Q_S (0.145 g l⁻¹ h⁻¹), respectively, under optimized conditions.

The kinetic and thermodynamic evaluation of GABA production process using E. faecium CFR 3003 not only adds to the existing scientific knowledge but could also throw light on the possibilities of using this interesting bacterium for industrial applications.