Significance of CO2 donor on the production of succinic acid by Actinobacillus succinogenes ATCC 55618

Background Succinic acid is a building-block chemical which could be used as the precursor of many industrial products. The dissolved CO2 concentration in the fermentation broth could strongly regulate the metabolic flux of carbon and the activity of phosphoenolpyruvate (PEP) carboxykinase, which are the important committed steps for the biosynthesis of succinic acid by Actinobacillus succinogenes. Previous reports showed that succinic acid production could be promoted by regulating the supply of CO2 donor in the fermentation broth. Therefore, the effects of dissolved CO2 concentration and MgCO3 on the fermentation process should be investigated. In this article, we studied the impacts of gaseous CO2 partial pressure, dissolved CO2 concentration, and the addition amount of MgCO3 on succinic acid production by Actinobacillus succinogenes ATCC 55618. We also demonstrated that gaseous CO2 could be removed when MgCO3 was fully supplied. Results An effective CO2 quantitative mathematical model was developed to calculate the dissolved CO2 concentration in the fermentation broth. The highest succinic acid production of 61.92 g/L was obtained at 159.22 mM dissolved CO2 concentration, which was supplied by 40 g/L MgCO3 at the CO2 partial pressure of 101.33 kPa. When MgCO3 was used as the only CO2 donor, a maximal succinic acid production of 56.1 g/L was obtained, which was just decreased by 7.03% compared with that obtained under the supply of gaseous CO2 and MgCO3. Conclusions Besides the high dissolved CO2 concentration, the excessive addition of MgCO3 was beneficial to promote the succinic acid synthesis. This was the first report investigating the replaceable of gaseous CO2 in the fermentation of succinic acid. The results obtained in this study may be useful for reducing the cost of succinic acid fermentation process.


Background
Succinic acid, an intermediate in the cycle of tricarboxylic acid (TCA), is one of four-carbon platform chemicals for producing different kinds of petroleum derivatives and biodegradable polymers [1,2]. Succinic acid could be produced by chemical conversion and microbial fermentation [3]. Because of the rising price, the limited reserves of petroleum and the pollution of environment, the oil-based industries had been prompted a movement towards the bio-based chemicals, and the bio-based succinic acid production had drawn the attention from enterprises and research institutes [4,5].
As the end-product of the energy metabolism, succinic acid could be produced by many anaerobic microbes, such as Actinobacillus succinogenes, Anaerobiospirillum succiniciproducens, Mannheimia succiniciproducens, Escherichia coli, and other microbes [2,4,6,7]. Especially A. succinogenes ATCC 55618, which is a facultative anaerobe isolated from the bovine rumen [8]. In the production of succinic acid by A. succinogenes, one of the key factors is the supply of CO 2 . A higher concentration of CO 2 could increase the ratio of succinic acid concentration to the other acids production, the ratio of carbon recovery, and the yield of succinic acid [9]. When A. succinogenes and A. succiniciproducens were used for the production of succinic acid, as a kind of co-substrate of phosphoenolpyruvate (PEP)-carboxykinase in the TCA cycle, CO 2 could promote carbon flow toward the production of succinic acid [10,11]. For the other succinic acid production microorganisms such as E. coli and Mannheimia succiniciproducens, CO 2 was incorporated into the backbone of threecarbon compound to generate four-carbon oxaloacetate via PEP carboxylase to enhance the production of succinic acid [12,13].
Because of the poor solubility of gaseous CO 2 at 1 atm, many kinds of carbonate and bicarbonate salts were employed as indirect CO 2 donor to enhance the dissolved CO 2 concentration in fermentation broth. MgCO 3 was a preferable carbonate because the addition of MgCO 3 would not lead to a radical change of culture pH, and an increase of Mg 2+ concentration in fermentation broth showed little negative effect on the metabolism profile and morphology of succinic acid production strain [14]. Some investigators tried to demonstrate the relationship between extra CO 2 donors and succinic acid production [9,15,16]. But there were a few different features in physiological and biochemical characteristics among various kinds of succinic acid producing strains, and the current results were weak in promoting succinic acid production [15,16].
In this study, the dissolved CO 2 concentration and the addition amount of MgCO 3 were quantitatively determined to optimize succinic production by Actinobacillus succinogenes ATCC 55618. To calculate the dissolved CO 2 concentration in the fermentation broth, a mathematical model which considers culture pH, temperature, ionic strength, and salt concentration in the fermentation broth were developed. According to the model prediction and experimental verification, this work firstly demonstrated that the supply of gaseous CO 2 had no significant effect on succinic acid production when MgCO 3 was fully supplied.

Maintenance and preculture of Actinobacillus succinogenes
The strain of A. succinogenes ATCC 55618 was purchased from American Type Culture Collection (ATCC, Manassas USA), which was maintained in 20% glycerol at -70°C.
The plate was inoculated with the above strain and incubated at 37°C for 2 days. Preculture medium consisted of the following components (g/L): tryptone 17; soya peptone 3; glucose 2.5; NaCl 5; K 2 HPO 4 2.5, and culture pH was adjusted to 7.1-7.5. For the first preculture, 50-mL medium was prepared in a 250-mL anaerobic bottle, and then a colony from a plate culture was inoculated, and followed by 12-hour incubation at 37°C on a rotary shaker at 120 rpm. For the second preculture, 47.5-mL medium was prepared in a 250-mL anaerobic bottle, and inoculated with 2.5-mL first preculture broth, then followed by 12-hour incubation at 37°C on a rotary shaker at 120 rpm.

Fermentation in the stirred-tank bioreactor
The stirred-tank bioreactor used was a 5.0-L (working volume) BioFlo 110 New Brunswick Scientific (NJ, USA) agitated bioreactor with two six-bladed Rushton impellers (5.9-cm i.d.). The lower impeller was 2.5 cm above the reactor bottom, and the vertical distance between two impellers was 8.5 cm. The reactor was aerated through a ring sparger with a pore size of 1.0 mm, which was located 2.2 cm above the reactor bottom. The bioreactor was equipped with probes of pH (Mettler-Toledo GmbH, Switzerland), temperature and foam.
Four cultures were carried out simultaneously in the stirred-tank bioreactors with homogeneous cell source under well-controlled process conditions but under different culture conditions. The identical cell source and process conditions, other than the experimental condition, made it possible to perform accurate head-to-head comparisons. The results presented here were confirmed to be reproducible in another experiment (data not shown).

Model description and calculation
When a gas mixture of CO 2 and N 2 was supplied into the bioreactor and became liquid-gas phase equilibrium, the CO 2 dissolved concentration in broth at 1 atm could be described by the reduction of Henry's law: Where P CO2 is the CO 2 partial pressure (kPa) in gas mixture which is determined by the mixing ratio of CO 2 and N 2 , H is the Henry's constant for CO 2 in the fermentation broth (kPa·L/mol), and C CO2 is the dissolved CO 2 concentration in the fermentation broth (mol/L).
In Equation (5), the Henry's constant of CO 2 in the pure water was 4320 kPa L/mol [15]. T is the absolute temperature (K) in culture condition.
After combining all the parameters mentioned above into the Equation (5) and Equation (1), the model used for calculating CO 2 dissolved concentration in broth could be obtained when gaseous CO 2 was used as external CO 2 donor and the result is shown in Figure 1A. There is a correlated linear trend between CO 2 partial pressure and the dissolved CO 2 concentration in fermentation broth. The dissolved CO 2 concentration in the fermentation broth was 5.05, 10.11, 15.16 and 20.22 mM when CO 2 partial pressure was 25.33, 50.66, 75.99 and 101.33 kPa, respectively. And the maximal dissolved CO 2 concentration is 20.22 mM due to the solubility of gaseous CO 2 .
When MgCO 3 was added with the supply of pure gaseous CO 2 at 1 atm, CO 2 , HCO 3 -, and CO 3 2would become in equilibrium in the fermentation broth according to the following equations [15]: As reported in the previous study [19], the maximum solubility of MgCO 3 in water at 40°C was 139 mM.
According to Equation (1), (5), (6) and (7), the model used for calculating the dissolved CO 2 concentration in the fermentation broth could be obtained when both gas phase CO 2 and MgCO 3 were used as CO 2 donors, and the relationship between the addition amount of MgCO 3 and the dissolved CO 2 concentration under the CO 2 partial pressure of 101.33 kPa is shown in Figure  1B

Effect of CO 2 partial pressure
The significance of gaseous CO 2 partial pressure on succinic acid accumulation was studied by setting CO 2 partial pressure at 25.33, 50.66, 75.99 and 101.33 kPa during the whole fermentation process in the stirredtank bioreactors, which was controlled by adjusting the corresponding mixing ratio of CO 2 and N 2 at 25%, 50%, 75%, 100% (v: v) by gas mix controller (BioFlo110, New Brunswick Scientic NJ, USA), respectively, and the corresponding dissolved CO 2 concentration in the fermentation broth was 5.05, 10.11, 15.16, and 20.22 mM.
Effect of the supply of gaseous CO 2 and the addition of MgCO 3 The maximal dissolved CO 2 concentration in the fermentation broth was 20.22 and 139.00 mM when only gaseous CO 2 and MgCO 3 was supplied, respectively. In order to study the higher dissolved CO 2 concentration on the succinic acid production, the fermentations were conducted by adding MgCO 3 to enhance the dissolved CO 2 concentration. MgCO 3 was added to the broth after a separate sterilization before the inoculation. The effect of the supply of gaseous CO 2 and the addition of MgCO 3 on the fermentation process was studied by adding 2.92, 5.84, 11.68, and 23.35 g/L of MgCO 3 at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas), and its corresponding dissolved CO 2 concentrations in the fermentation broth were 54.97, 89.72, 159.22 and 159.22 mM, respectively. The dissolved CO 2 concentration maintained constant at 159.22 mM even when concentrations higher than 11.68 g/L of MgCO 3 were added at the CO 2 partial pressure of 101.33 kPa. The other culture conditions were the same as the above experiments.

Effect of the addition of higher amount of MgCO 3
Effect of the excessive addition amount of MgCO 3 was studied by adding 30, 40, 50 and 60 g/L of MgCO 3 at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas), and all the corresponding dissolved CO 2 concentration in the fermentation broth was 159.22 mM. The other culture conditions were the same as the above experiments.

Effect of CO 2 donor supply mode
According to the above results and Equation (5), the effect of CO 2 donor supply mode was studied by using two supply modes: 40 g/L MgCO 3 was used as the only CO 2 donor, and 40 g/L MgCO 3 was supplied at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas). The other culture conditions were the same as the above experiments.

Sampling, the determination of succinic acid production
For sampling, about 20-30 mL of broth was taken once from each reactor and the cell growth was monitored by measuring the optical density at 660 nm (OD 660 ). At an OD 660 of 1.0, A. succinogenes ATCC 55618 has a concentration of 0.626 g dry cell weight (DCW)/L. For succinic acid determination, 1 mL methanol and 1 mL acetonitrile were added to 1 mL fermentation broth to remove protein and the sample was kept at 4°C overnight. After centrifuging at 11, 000 rpm for 30 min, the supernatants were diluted and filtrated through 0.22 μm filter, then analyzed by high-performance liquid chromatography (HPLC, Dionex) using Maisch ReproSil-Pur Basic C18 column. The optimized mobile phase was 5 mM KH 2 PO 4 water solution, whose pH was adjusted to 2.8 by H 3 PO 4 . The column oven temperature was maintained at 40°C and the flow rate was 1 mL/min. The detection wave was 210 nm. Residual sugar level was assayed with phenolsulfuric acid method [20].

Results and discussion
Effect of CO 2 partial pressure As one of the direct substrates for the biosynthesis of succinic acid, CO 2 could affect the metabolic flux and the mass distribution of succinic acid [8,21]. The quantitative determination of the dissolved CO 2 concentration in the fermentation broth is beneficial to study the impact of CO 2 partial pressure on the production of succinic acid. Song et al. [15] and Lee et al. [16] reported that succinic acid production could be enhanced by increasing CO 2 partial pressure in the fermentation of M. succiniciproducens and A. succiniciproducens. Therefore, it was necessary to investigate the effect of CO 2 partial pressure on the accumulation of succinic acid by A. succinogenes ATCC 55618.
The effect of CO 2 partial pressure on the succinic acid production is shown in Figure 2A. When the CO 2 partial pressures were 25.33, 50.66, 75.99, and 101.33 kPa, the dissolved CO 2 concentrations in the fermentation broth calculated using Equation (5) (Table 1). And at the CO 2 partial pressure of 101.33 kPa, the maximal dissolved CO 2 concentration achieved was 20.22 mM, which was the highest dissolved CO 2 concentration when only gaseous CO 2 was supplied. The dissolved CO 2 concentration was increased with the increase of the partial pressure when gaseous CO 2 was used as sole CO 2 donor. The succinic acid productions were 8.84, 10.21, 10.44, and 10.97 g/L as obtained on 48 hour at the CO 2 partial pressure of 25.33, 50.66, 75.99, and 101.33 kPa, respectively, and its corresponding productivities were 0.18, 0.21, 0.22, and 0.23 g/L per hour. This indicated that when gaseous CO 2 was used as the sole CO 2 donor, CO 2 partial pressure showed no significant effect on the succinic acid accumulation. On the contrary, as reported by Lu et al. [22] and Samuelov et al. [23], a higher available CO 2 concentration could cause higher succinic acid production by increasing the activity of PEP carboxykinase. These indicated that when gaseous CO 2 was used as the sole CO 2 donor, the available dissolved CO 2 concentration was not high enough to increase the production of succinic acid in the fermentation of A. succinogenes.
As shown in Figure 2B, the patterns of acetic acid production at various CO 2 partial pressures were similar. The concentrations of other by-products such as formic acid, lactic acid and ethanol were relatively constant at around 5.0, 11.0 and 2.0 g/L, respectively, regardless of the levels of the dissolved CO 2 in the broth. Figure 2C shows the time profile of residual sugar under various CO 2 partial pressures. The glucose concentration at the CO 2 partial pressure of 101.33 kPa was decreased faster than that at other CO 2 partial pressures during the first 24 hours. The yield of succinic acid against glucose was around 0.21 g succinic acid/g glucose when gaseous CO 2 was used. That means the CO 2 partial pressure showed no significant effect on the succinic acid yield. And there was no significant effect on the cell growth. The OD 660 was between 6.0 and 6.7 when gaseous CO 2 partial pressure was 25.33, 50.66, 75.99, and 101.33 kPa ( Table 1).

Effect of the supply of gaseous CO 2 and the addition of MgCO 3
The maximal dissolved CO 2 concentration is limited by the solubility of gaseous CO 2 when it was supplied as the sole CO 2 donor. In order to investigate the effect of higher dissolved CO 2 concentration on succinic acid production, the fermentations were conducted by adding MgCO 3 at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas) to enhance the dissolved CO 2 concentration in the fermentation broth.
The effect of the supply of gaseous CO 2 and the addition of MgCO 3 on the fermentation process was studied by adding 2.92, 5.84, 11.68, and 23.35 g/L of MgCO 3 at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas), and the corresponding dissolved CO 2 concentrations was 54.97, 89.72, 159.22, and 159.22 mM. Because of the solubility of MgCO 3 and CO 2 , the maximal dissolved CO 2 concentration of 159.22 mM was obtained under the addition of 11.68 g/L MgCO 3 with 100% CO 2 gas. Even more than 11.68 g/L of MgCO 3 , the dissolved CO 2 concentration maintained constant at 159.22 mM. As shown in Figure 3A, the highest succinic acid productions were 15.26, 15.94, 25.86 and 36.84 g/L as obtained on 48 hour with the addition of 2.92, 5.84, 11.68 and 23.35 g/L MgCO 3 , respectively, and its corresponding productivity was 0.32, 0.33, 0.54 and 0.77 g/L per hour,  respectively. The maximum succinic acid production was increased from 25.86 to 36.84 g/L when the addition amount of MgCO 3 was increased from 11.68 to 23.35 g/L, while the dissolved CO 2 concentration was maintained constant. It can be concluded that the higher dissolved CO 2 concentration was beneficial for the succinic acid biosynthesis. But the dissolved CO 2 concentration was not the only factor affecting succinic acid synthesis; the excessive MgCO 3 also had positive effect. As a kind of neutralization reagent, MgCO 3 could promptly neutralize the organic acid produced during the fermentation process. But when only 11.68 g/L of MgCO 3 was added, the saturated state of MgCO 3 would be lost quickly because of the rapid accumulation of organic acid during the fermentation. And when the addition amounts of MgCO 3 exceeded 11.68 g/L, there will be excessive solid MgCO 3 precipitate. Even if organic acids accumulate, MgCO 3 solution can also keep saturated. Equation (6) and (7) indicated that the dissolved concentrations of HCO 3 -, CO 3 2and CO 2 could be enhanced with the addition of MgCO 3 in the fermentation broth. However, MgCO 3 may not be used as CO 3 2-donor because there were few reports that CO 3 2could be used directly as substrate by succinic acid producing microorganisms. Although HCO 3 and CO 2 could be used as the co-substrate of PEP carboxylase and improve the production of succinic acid [23], HCO 3 was much less permeable to lipid cell membrane than the uncharged CO 2 molecule because is a kind of polar molecular, and there was no HCO 3 transporter on the membrane of A. succinogenes which could deliver HCO 3 from the broth into the cell [24]. So the higher concentration of HCO 3 could not promote the production of succinic acid. And MgCO 3 may be used as indirect CO 2 molecule donor to promote the production of succinic acid in the fermentation process of A. succinogenes. On the other hand, when the levels of dissolved CO 2 reached 159.22 mM, there would be insoluble MgCO 3 , and that could cause turbid broth. The cells were spread uniformly in the broth, which was helpful to eliminate the cell flocculation and indirectly promoting the succinic acid biosynthesis.
As shown in Figure 3B, the patterns of acetic acid production at the dissolved CO 2 concentration of 54.97 and 89.72 mM were similar. However, when the levels of dissolved CO 2 reached 159.22 mM, the acetic acid production was significantly enhanced. It was distinct from other reports. In the fermentation of M. succiniciproducens, the levels of dissolved CO 2 showed little effects on the acetic acid accumulation [15]. The CO 2 concentration has been shown to regulate the levels PEP carboxykinase pathway at high CO 2 levels, and PEP carboxykinase levels rise [23]. However, the enhanced PEP carboxylation may cause higher glucose consumption rate. This effect may cause more metabolic flow by PEP to pyruvic acid, and further to acetic acid. Meanwhile the production of formic acid, lactic acid and ethanol almost not be improved may be because the raised PEP carboxykinase activity competitively inhibited these key enzymes such as pyruvate formatelyase, lactate dehydrogenase and ethanol dehydrogenase.
The patterns of cell growth with the addition of different concentration of MgCO 3 were similar, and the OD 660 was between 9.7 and 10.9 (Table 1). Figure 3C shows the time profile of residual sugar under various addition amount of MgCO 3 . When 23.35 g/L of MgCO 3 was added, glucose was consumed faster than the other conditions between 12 and 24 hour, which corresponded well to the succinic acid accumulation. The yield of succinic acid against glucose was 0.27, 0.26, 0.38, and 0.45 g succinic acid/g glucose when the addition amount of MgCO 3 was 2.92, 5.84, 11.68 and 23.35 g/L, respectively. This indicated that the higher dissolved CO 2 concentration could effectively improve the yield of succinic acid against glucose.
Effect of the addition of higher amount of MgCO 3 Effect of the higher addition amount of MgCO 3 was studied by adding 30, 40, 50 and 60 g/L of MgCO 3 at the CO 2 partial pressure of 101.33 kPa (i.e., 100% CO 2 gas), and all the corresponding dissolved CO 2 concentration in the fermentation broth were 159.22 mM.
As shown in Figure 4A, the pattern of succinic acid production under various addition amount of MgCO 3 within the range of investigation was similar. The maximal succinic acid production of 53.55, 61.92, 61.48, and 58.05 g/L was obtained with the addition of 30, 40, 50, and 60 g/L MgCO 3 , respectively, and its corresponding productivity was 0.74, 0.86, 0.85 and 0.81 g/L per hour. When the addition amount of MgCO 3 exceeded 40 g/L, the production and productivity of succinic acid were kept almost constant, but the specific productivity was decreased. This indicated 40 g/L MgCO 3 was enough for improving the accumulation of succinic acid. Similarly, Du et al. [25] reported when the addition amount of MgCO 3 exceeded 30 g/L, there was no significant change on the production of succinic acid.
The significance of addition amount of MgCO 3 on acetic acid accumulation was studied. As shown in Figure  4B, there was no significant effect on the biosynthesis of acetic acid. Similarly, the concentrations of other by-products such as formic acid, lactic acid and ethanol were relatively constant at around 5.0, 11.0 and 2.0 g/L, respectively, regardless of the addition amount of MgCO 3 .
The cell growth patterns under the addition of MgCO 3 were quite similar (Table 1). Figure 4C shows that the time profile of residual sugar under various addition amount of MgCO 3 . The yield of succinic acid against glucose was 0.56, 0.60, 0.64, and 0.63 g succinic acid/g glucose when the addition amount of MgCO 3 was 30, 40, 50 and 60 g/L, respectively. It seemed that to obtain a higher yield of succinic acid against glucose, the addition amount of MgCO 3 should no be more than 50 g/L.
Effect of CO 2 donor supply mode Gaseous CO 2 was widely used as external CO 2 donor and anaerobic environment maintenance agent in succinic acid fermentation process. Calculated from Equation (5, 6, 7), the dissolved CO 2 concentration (139.00 mM) under 40 g/ L MgCO 3 was just decreased by 12.76% comparing with that obtained under the addition of 40 g/L MgCO 3 with 100% CO 2 gas. This suggested that the gaseous CO 2 may be removed by the addition of MgCO 3 . In order to testify this proposal, the effect of CO 2 donor supply mode was studied by using two supply modes: 40 g/L MgCO 3 was used alone; 40 g/L MgCO 3 was supplied with 100% CO 2 gas.
As shown in Figure 5A, after 72 h incubation, the production of succinic acid reached 56.14 and 60.38 g/L when 40 g/L MgCO 3 was used as the only CO 2 donor and 40 g/L MgCO 3 was supplied at the CO 2 partial pressure of 101.33 kPa, and the corresponding productivity was 0.80 and 0.84 g/L per hour. The succinic acid production was just decreased by 7.03% without the supply of gaseous CO 2 . As shown in Figure 5B, the acetic acid production was decreased by 17.91% without the supply of gaseous CO 2 . Figure 5C clearly shows the time courses of sugar consumption under different CO 2 supply modes were similar. The yield of succinic acid against glucose was 0.54 g succinic acid/g glucose when 40 g/L MgCO 3 was used alone, and the yield was 0.58 g succinic acid/g glucose when MgCO 3 was supplied with 100% CO 2 . And there was no significant effect on the cell growth whether gaseous CO 2 was used.

Conclusions
In this study, an effective CO 2 quantitative mathematical model was developed to calculate the dissolved CO 2 concentration in the broth during the fermentation of Actinobacillus succinogenes ATCC 55618. The model offered a quantitative method for screening the suitable CO 2 donor form and addition amount for the production of succinic acid. There was no significant effect of CO 2 partial pressure on the production of succinic acid when gaseous CO 2 was used as the sole CO 2 donor. But when gaseous CO 2 was used with MgCO 3 , higher amount of MgCO 3 was more effective on promoting the succinic acid synthesis. And the maximum succinic acid production of 61.92 g/L was obtained at 159.22 mM dissolved CO 2 concentration, which was supplied by 40 g/L MgCO 3 with 100% CO 2 gas. And it was concluded that the supply of gaseous CO2 was not essential when 40 g/ L of MgCO 3 was added in the fermentation medium. This is the first report investigating the replaceable of gaseous CO2 in the fermentation of succinic acid. The results obtained in this study may be useful for reducing the cost of succinic acid fermentation process.