Abstract

The major aims of biology to understanding life at a systems level. Escherichia coli is a metabolically versatile bacterium able to respond to changes in environmental factors availability. The effect of pH downshift on fermentation characteristics was investigated in a continuous culture of Escherichia coli at aerobic and micro-aerobic conditions. Regardless of oxygen availability, higher levels of acetate were associated with lower biomass yields and lower glucose consumption rates at pH 5.5 as compared to the observations made at pH 7.0. Observed gene expressions indicated that the down- regulation of the glucose uptake rate corresponded to the down-regulation of ptsG gene expression which in turn was caused by the up-regulation of mlc gene under the positive control of Crp. In accordance with up-regulation of arcA gene expression at acidic conditions, the expressions of TCA cycle-related genes such as icdA and gltA, and the respiratory chain gene cyoA were down-regulated, whereas cydB gene expression was up-regulated. Decreased activity of the TCA cycle caused more acetate formation at lower pH levels. Under micro-aerobic condition, higher levels of formate and lactate were produced at lower pH due to up-regulation of pflA, yfiD and ldhA genes. Meanwhile, lower levels of ethanol were produced due to the down-regulation of adhE gene at lower pH, as compared to the observation at neutral pH. The combined effect of pH and temperature on gene expression was also investigated and observed that decreases in the specific glucose consumption rate were associated with increases in the specific acetate production rate. This type of information is useful for the production of recombinant proteins, bio-molecules, simultaneous saccharification and fermentation (SSF) and strain improvement.

1. Introduction

The ability of bacterium to respond rapidly and effectively to environmental perturbation is a distinguishing and vital aspect of their physiology. The molecular toolbox, enabling genetic engineering and studying of regulation and gene expression, is presumably the largest that exists for one particular organism ¹. Hence, Escherichia coli have been widely used by applied microbiologists to try to steer the metabolism of this organism toward the production of molecules with biotechnological value. E. coli, a facultative anaerobe and other related enteric bacteria show a number of genetic responses to pH changes in their growth environment by regulating gene expression ^{2, 3, 4} and protein profiles ⁵. Although most acid tolerance systems are activated at the late growth phase and/or the stationary phase, acid tolerance is also observed in the exponential growth phase of E. coli under aerobic conditions and this is advantageous from the productivity point of view. Some of these acid resistance systems depend on the available extracellular amino acids such as glutamate, arginine and lysine. In this system, the intracellular proton is consumed by the reductive decarboxylation of the amino acid followed by the excretion of the product, such as γ-amino butyric acid (GABA) from the cytoplasm to the periplasm. This mechanism has been shown to increase gadAB, which encodes glutamate decarboxylase and gadC, which encodes the glutamate: γ-amino butyric acid (GABA) antiporter in response to acid stress, heat shock and stationary phase signals in E coli ². Organic acids such as acetic acid, lactic acid etc. accumulate at the late growth phase or the stationary phase in typical batch cultures, and are known to increase GadA and GadB proteins at low pH ⁶. The sigma factor σ^S or RpoS, which increase at the stationary phase of growth, as well as Crp (catabolite receptor protein) are involved in acid resistance ⁷. It has also been shown that acid pH lowers cAMP levels in exponentially growing cells in the minimal glucose medium, potentially resulting in the elevation of RpoS. It has also been shown that the two component system of EnvZ (sensor) and OmpR (regulator) regulate protein expression. Given the role of OmpR as a possible key regulator for acid adaptation, the ompR mutant is thus sensitive to acid exposure ⁸. The effects of pH on gene expression are also known to be influenced by other environmental factors, most notably the availability of oxygen. The effect of the availability of oxygen on gene expression at low pH has been investigated in a number of studies that focus on protein responses under different environmental conditions using two-dimensional gel electrophoresis (2-D gels) ⁹. Under anaerobic and or micro-aerobic conditions, additional genes, such as ackA, lpdA, and ompT were induced ¹⁰. It has also been shown that, at anaerobic conditions, amino acid decarboxylase is increased ^{11, 12} whereas the oxygen-induced cytochrome o is repressed ¹³. To avoid deleterious concentrations in the cell caused by the production at low pH, ldhA is induced by acid, thus producing lactate instead of the more harmful acetate plus formate ¹⁴. Although the underlying mechanism for this acid tolerance is not yet clear, it could be activated by adapting cells at pH values ranging from 4.5 to 5.8. Despite the wealth of literatures on protein responses under different environmental conditions, only relatively few genomic studies has paid attention on the effect of pH downshift on E. coli metabolism. The effect of pH on metabolism is of practical interest, since culture pH is often not controlled in the industry. Meanwhile, the production of recombinant proteins and organic acids such as ethanol, lactate etc. by microbial fermentation is a widely used process in pharmaceutical, food and chemical industries. This necessitates the need to find favorable conditions (eg. temperature, pH etc.) for both the enzymatic hydrolysis and fermentation process in simultaneous saccharification and fermentation (SSF). In the present study, therefore, we investigated metabolism changes in E. coli at low pH with respect to gene expressions under aerobic and micro-aerobic conditions. The study also investigated the combined effects of both pH and temperature on E. coli metabolism.

2. Materials and Methods

The strains used were Escherichia coli BW25113 (lacI^q rrnB_T14 ΔlacZ_WJ16 hsdR514 ΔaraBAD_AH33 ΔrhaBAD_LD78) ¹⁵. The inoculum was prepared by transferring cells from a glycerol stock (0.1 ml) to a 50 ml L-shaped test tube containing 10 ml of Lysogeny broth (LB) medium and incubating overnight at 37° ubsequently, 1 ml of the culture broth was transferred into a 500 ml T- shaped flask containing 100 ml LB medium. Synthetic M9 minimal media ontaining 10 g/L glucose, 48 mM Na₂HPO₄, 22 mM KH₂PO₄, 10 mM NaCl, and 30 mM (NH₄)₂SO₄ was used as the main culture medium. Additionally, the following components were filter sterilized and then added (per liter of final medium): 1 ml of 1 M MgSO₄, 1 ml of 0.1 mM CaCl₂, 1 ml of 1 mg of Vitamin B₁ per liter, and 10 ml of trace element solution containing (per liter) 0.55 g of CaCl₂, 1 g of FeCl₃, 0.1 g of MnCl₂.4H₂O, 0.17 g of ZnCl₂, 0.043 g of CuCl₂.2H₂O, 0.06 g of CoCl₂.6H₂O and 0.06 g Na₂MoO₄.2H₂O. Using a 1 L working volume, continuous cultivations were conducted at a dilution rate of 0.2 h^-1 in a 2 L jar fermentor (M-100, Rikakikai Co. Tokyo, Japan). The culture pH was kept constant at 5.5± 0.05, 6.0± 0.5 or 7.0± 0.05 by adding either 2N NaOH or 2N HCl. Culture temperature was maintained at 37°C ± 0.5°C or 42 ± 0.5°C. Air flow rate was maintained at 1 L min^-1 and an agitation speed of 350 rpm was selected to ensure that the dissolved oxygen level remains about 30 - 40% of air saturation in the aerobic cultivation. Micro-aerobic cultivation was initiated by aerobic cultivation for 2 h followed by a stoppage in the supply of air and reduction in the agitation speed to around 100 rpm, such that the cultivation was nearly anaerobic. CO₂ and O₂ concentrations were measured using the off-gas analyzer (DEX-2562, ABLE Co., Japan).

Cell concentration was determined using a spectrophotometer (Ubet-30, Jasco Co., Tokyo, Japan) by converting the optical density (OD) of the culture broth at a wave length of 600 nm to dry cell weight (DCW) per liter. This approach is based on the previously reported relationship between OD and DCW. Glucose concentration was measured using enzymatic kit (Wako Co., Osaka, Japan). Acetate, formate, lactate, succinate, and ethanol concentrations were also determined using enzymatic kits (Boehringer Co., Mannheim, Germany).

Exactly 2.5 µl of culture broth was suspended into 5 µl RNAprotect bacteria reagent. The samples were kept on ice for 5 minutes. After centrifugation at 10,000 rpm (4°C, 10 min), the supernatant was discarded, and the pellet was stored in RNA isolation at -80°C until used. Total RNA was isolated from E. coli cells using Qiagen RNeasy Mini Kit (QIAGEN K.K., Japan) according to the manufacturer’s recommendation. The quality of extracted RNA was determined by the optical density measurements at 260 and 280 nm, as well as from bands obtained on 1% formaldehyde agarose gel electrophoresis. The sequences of the primers used in the present study have been previously reported ¹⁶ exception the following:

mlc 5’ AGCAGACCAACGCGGGCGCG 3`

5` GACTATACGCAGGAAGGGCC 3`

gadA 5’ CGGATAAACCAAACCTGGTG 3’

5’ GAATTTATCCAGCGCATCGT 3’

yfiD 5’ AACTCTTTCTGGCTGCTGG 3’

5’ GATGGTCAGCTGCGGATAT 3’

Gene-specific primer pairs were designed following the criteria described by Sambrook and Russel (2001). The primers used in this study were synthesized at Hokkaido System Science Co. (Sapporo, Hokkaido, Japan). In all cases, the company confirmed the absolute specificity and purity of the primers.

RT-PCR reactions were carried out in a TaKaRa PCR Thermal Cycler (TaKaRa TP240, Japan) using Qiagen One Step RT-PCR kit (QIAGEN K.K., Japan). The 25 µl reaction mixture was incubated for 30 min at 50°C for reverse transcription (cDNA synthesis) followed by 15 min incubation at 95°C for initial PCR activation. The process was thereafter subjected to 30 cycles of amplification which consisted of a denaturing step (94°C for 1 min), annealing step (approximately 5°C below melting temperature of primers for 1 min), and an extension step (72°C for 1 min). Finally in the final extension, the reaction mixture was subjected to a temperature of 72°C for 10 min. To check for nucleic acid contamination from the reaction components, a negative control which lacks the template RNA was run concurrently in every round of RT-PCR. Thereafter, 5 µl of amplified products were run on a 1.8 % agarose gel. Gels were stained with 1 mg mL^-1 of ethidium bromide, photographed using a Digital Image Stocker (DS-30, FAS III, Toyobo, Osaka, Japan) under UV light and analyzed using Gel-Pro Analyzer 3.1 software (Toyobo, Osaka, Japan). In order to determine the optimal amount of input RNA, two-fold diluted template RNA were amplified in RT-PCR assay under identical reaction conditions to construct a standard curve for each gene product. When the optimal amount of input RNA was determined for each gene product, RT-PCR was carried out under identical reaction conditions to detect differential transcript levels of genes. The gene dnaA, which encodes E. coli DNA polymerase, was used as an internal control for the RT-PCR determinations since it is not subject to variable expression, i.e. abundant expression at relatively constant rate. The gene expressions are thus presented as relative values to that of dnaA. To calculate the standard deviation, RT-PCR was independently performed three times under identical reaction conditions. To ensure that the observed changes were statistically significant, the Student’s t-test was applied.

3. Results

Presented in Table 1 is the effect of culture pH on the fermentation characteristics. The results obtained indicates that a significantly higher amount of acetate was formed (p< 0.05), the cell yield was significantly lower (p< 0.05), and the specific glucose consumption rate was significantly lower (p< 0.1) at pH 5.5, as compared to the case at pH 7.0.

To clarify the phenomenon of E. coli metabolism under low pH, gene expressions were measured by RT-PCR. Figure 1 presents a comparison of gene expressions obtained at two different pH values. Results obtained indicates that significant up-regulation of the expression of rpoS (p< 0.10), gadA (glutamate decarboxylate gene) (p< 0.05) and acs (p< 0.05). These genes are known to be under the control of RpoS.

Figure 1 also shows the up-regulation of arcA gene expression (p< 0.1), where arcA gene product functions as a repressor of such genes as are involved in the TCA cycle under micro-aerobic condition . In accordance with the up-regulation of arcA, some of TCA cycle genes such as icdA (p < 0.1) and gltA were down-regulated (p< 0.05). Figure 1 also indicates observations from the expressions of the respiratory chain genes. While cydB was up-regulated (p< 0.1), cyoA was down regulated (p < 0.1). This may be due to the up-regulation of arcA since cydB operon is under the positive control of ArcA compared to cyoA which is under the negative control of ArcA . The fnr gene expression was also up-regulated, as well as arcA. This caused the down regulation of lpdA and aceE gene expressions. Figure 1 also indicates that the expression of crp gene, which codes for cAMP receptor protein Crp, was up-regulated (p < 0.1). Also, the expression of sdhC, which is known to be under the control of Crp was also up-regulated (p < 0.1) ¹⁷. Moreover, mlc gene expression was higher (p < 0.05), and ptsG gene expression was lower (p < 0.05). This is in line with previously published finding that that ptsG is repressed by Mlc ¹⁸.

Figure 1 also shows that cra (catabolite repressor activator) gene expression was up-regulated (p<0.1) whereas glycolysis was repressed where cra gene product regulates the carbon flow in such a way that gluconeogenesis was activated. The down-regulations of pfkA, pykA and zwf gene expressions were partly due to up-regulation of cra. The gene expressions of fadR and iclR were also higher (p< 0.05, p< 0.1) since IclR is known to repress aceBAK where FadR activates iclR.

Table 2 shows the comparison of the fermentation data at two different pH values in the micro-aerobic condition. It was observed that higher amounts of acetate, formate, and lactate were produced, whereas lower amount of ethanol was produced at pH 5.5, as compared to the case at pH 7.0. Figure 2 compares the gene expressions at two different pH values. The gene expression patterns were similar to the observation in Figure 1 as it indicates that gadA (p<0.1) and fnr (p< 0.05) gene expression increased at pH 5.5, as compared to the case at pH 7.0. Also, arcA gene expression increased (p< 0.1) while the expressions of icdA (p< 0.05), aceE (p< 0.1) and mdh (p< 0.1) genes decreased. It was also observed that crp and mlc gene gene expression increased (p< 0.5) while the expressions of such genes as ptsG (p< 0.05), ptsH (p< 0.1), lpdA (p< 0.1) decreased. Under micro-aerobic conditions, additional changes were observed as yfiD and pflA gene expressions increased (p< 0.05). These genes are involved in formate formation. It was also observed that ldhA gene expression increased (p< 0.05) whereas adhE gene expression decreased (p< 0.1) at pH 6.0, as compared to observations made at pH 7.0.

Table 3 shows the combined effects of pH and temperature on the fermentation characteristics. It was observed that decreases in the specific glucose consumption rate were associated with increases in the specific acetate production rate. The specific CO₂ production rate and the cell yield significantly decreased at pH 6.0 and 42°C, as compared to the observation made at pH 7.0 and 37°C. Since cell growth was significantly depressed at pH 5.5 and 42°C, the culture pH at 42°C was set at 6.0. Comparing gene expressions observed in the study (Figure 3), rpoH gene expression was found to increase, as well as the up-regulation (p< 0.05) of the heat shock genes dnaK, groL, groS and ibpB at pH 6.0 and 42°C unlike the observations made at pH 7.0 and 37°C. Also, arcA and cydB and fnr gene expression increased (p< 0.5, p< 0.5 and p< 0.1, respectively) along with a decrease in the expressions of icdA and gltA (p< 0.05 and p< 0.1) at pH 6.0 and 42°C. Also, crp, sdhC and mlc gene expression increased p<0.05) while ptsG gene expression decreased (p< 0.05). It was also observed that lpdA gene expressions increased (p<0.1) It was also observed that pykF, and zwf gene expressions were down-regulated (p<0.05, and p< 0.1 respectively) at pH 6.0 and 42°C, as compared to those at pH 7.0 and 37°C.

4. Discussion

It is well known that the Gad gene system is needed for survival under low pH ¹⁹. Cells possess specific defense mechanisms against acid environments in which Gad system has been extensively studied because of its major role in the detoxification of acid-induced stress in E. coli. Figure 1 and Figure 2 indicate that gadA gene expression increased at pH 5.5 as compared to the case at pH 7.0. GadA and GadB are known to be induced at lower pH as compared to the case at pH 7. Under acid stress, the product GABA is exported by GadC. Anaerobiosis amplifies the acidic induction of amino acid decarboxylases. It has been noted that the acid-induced expression of amino acid decarboxylases is enhanced under anaerobic condition. The gad system (GadA/GadBC) neutralizes acidity and enhances survival in extreme acid; its induction during anaerobic growth may help protect alkaline-grown cells from the acidification resulting from anaerobic fermentation. Figure 1 shows that gadA was also up-regulated even at aerobic conditions.

It was demonstrated in the current study that Fnr is the major activator of yfiD expression, but that the extent of Fnr-mediated activation could be modulated by the indirect oxygen sensor ArcA. Moreover, in proteomic analysis, the expression of yfiD::lac reporter fusion and the intracellular content of YfiD were found to be high during growth at low pH for both aerobic and anaerobic conditions. Figure 1 and Figure 2 show higher expression of yfiD at both aerobic and micro-aerobic conditions.

It has been shown that YfiD, a homologue of pyruvate formate lyase, was induced to high levels at pH 4.4 and induced two-fold higher by propionate at pH 6. Both chemicals at low pH cause internal acidification. At neutral or alkaline pH, YfiD was virtually absent and YfiD is, therefore, a strong candidate for response to internal acidification. It has been reported that the expression of cytochrome o is repressed by acid stress. Figure 1 - Figure 3 also indicates the down-regulation of cyo gene expression, which may be also partly due to up-regulation of arcA gene expression. Therefore, there is a complex relationship between yfiD, cyoA pH and oxygen level.

The up-regulation of sdhC gene at lower pH under both aerobic and anaerobic conditions (Figure 1 and Figure 2) is subject to complex regulatory mechanisms in relation to the respiratory metabolism. The up-regulation of sdhC is also partly due to the up-regulation of crp gene. The ptsG gene expression was down-regulated at lower pH as compared to the case at neutral pH (Figure 1 - Figure 3). This may be due to the up-regulation of mlc gene expression, a phenomenon also observed at temperature up-shift ²¹.

Figure 1 and Figure 2 show that aceA is up-regulated. This may be repressed by the up-regulation of arcA gene expression. One of the reason why aceA increased may be due to the fact that isocitrate lyase (AceA) was induced and showed substantially greater induction in acid or in base condition than at pH 7.

Figure 1 and Figure 2 also indicates the up- regulation of rpoS at lower pH. At low pH, acetate as a permeant acid is driven into the cell down the pH gradient and will thus reach far greater cytoplasmic concentrations in acidic media ^{22, 23}. E. coli cells initially produce fermentation products such as acetate and other acidic fermentation products which induce stationary-phase stress proteins ^{24, 25}. The growth-phase dependent sigma factor RpoS regulates several components of resistance to both acetate and formate under micro-aerobic condition, which can re-enter the cell and reach deleterious concentrations at low external. At low pH, components of pyruvate dehydrogenase complex (PDHc) gene such as lpdA was repressed, whereas fermentative pathways gene pflA, as well as ackA genes were up-regulated thus leading to more formate and acetate production. It should be noted that ldhA is induced by acid in order to produce more lactate instead of acetate.

5. Conclusion

The regulation of cytoplasmic pH in bacteria has long been studied. Little is known about the pH homeostasis of Escherichia. coli yet E. coli is known to move away from areas of low external pH or weak acids. This indicates that E. coli are capable of sensing difference in external pH and reacting to these changes. The present research result clarified the mechanism of metabolic changes upon pH down-shift together with temperature up-shift in relation to gene expressions of heat shock genes, global regulators and the metabolic pathway genes. In particular, the pH down-shift caused the up-regulation of rpoS gene, which in turn caused the up-regulation of gadA gene expression, and finally protected the cell from intracellular acidification. It was also found that the pH down-shift caused the up-regulation of fnr gene expression and activated yfiD, as well as pflA gene expression leading to the production of higher amounts of formate. Moreover, it was shown that the pH down-shift caused other TCA cycle genes to be repressed due to up-regulation of arcA gene. This was indirectly caused by lower dissolved oxygen concentration that led to the production of higher amounts of acetate. These findings are potentially useful for a variety of applications such as temperature-induced heterologous protein productions and solid state fermentation (SSF). In particular, the use of low pH and high temperature conditions in SSF for improved lactate production is suggested since ldhA gene is induced upon pH down-shift.

Acknowledgements

Authors are grateful to Professor Kazuyuki Shimizu, Department of Bioscience and Bioinformatics, Kyushu Institute of Technology, Japan for donation of strain, reagents and equipments and his critical suggestions of manuscript writing.

Conflict of Interest

The authors declare that they have no competing interests.

References