Our Model

Summary

One major task for us is to reduce the amount of by-products, such as acetic acid. They would occupy the metabolic resource for butyric acid and butanol synthesis and also result in impure final products. According to Prof. Wang in our interview, one of the best ways is ensuring a 1:1 ratio of butyric acid and butanol, which would even make the esterification reaction more chemically favorable. We considered this an important goal.

In this section, to investigate the key factors of balancing butanol and butyric acid production, we used the genome-scale mode (GSM) of C. tyrobutyricum to analyze the difference between the mode of butanol and butyric acid production. Two extreme situations, only butanol production and only butyrate production, were designed and simulated by a modified GSM (iCT583-adhE2). By comparing the flux changes in the network and the turnover rate of co-factors, we found that the butyryl-CoA dehydrogenase complex (BCD-ETF) involved in the ferredoxin loop is the key factor for driving butyrate and butanol synthesis. Moreover, the Rnf complex involved in ferredoxin oxidization was more beneficial for supplying NADH and improving butanol production, and regulating the ratio of involving in ferredoxin oxidization by HYD (hydrogenase) and Rnf complex could be a key point for balancing butanol and butyric acid production.

1. Introduction

C. tyrobutyricum is a native super-butyrate producer1. Due to the high-level metabolic flux of butyryl-CoA flux, it has been regarded as a great potential cell platform for butyric acid and butanol production2. And in our project, we also plan to use C. tyrobutyricum as a cell platform and design a process for butyl butyrate production. In theory, the esterification of butyl butyrate needs one molecular butyrate and one molecular butanol as substrates3. Therefore, how to balance the supply of precursor compounds is important issue. Based on previous studies, only heterogenous expression aldehyde/alcohol dehydrogenase (coding by adhE2 gene from C. acetobutylicum). in C. tyrobutyricum does not make the metabolic flux prefer to butanol production4. The mode of butyric acid production still is predominant. In our previous study, we provide an insight for explaining how C. tyrobutyricum produce butyrate couple with cell growth and H2 and CO2 emission. This phenotype could be explained by based on an energy conversion system, in which two ferredoxin loops make couple relationship between H2 and CO2 and butyrate production5. As shown in Fig. 1, in the pathway, oxidative ferredoxin can be reduced in the process of pyruvate oxidation and crotonyl CoA to butyral-CoA. Moreover, reductive ferredoxin can be oxidated by hydrogenase (Hyd) and H+/Na+-translocating ferredoxin:NAD+ oxidoreductase (Rnf complex) complex. However, these results did not further give an insight for why the predominance mode inclined to butyric acid production rather than butanol production. Therefore, in this part, we want to analyze the difference between the mode of butanol and butyric acid production in silico and find some targets or strategy for improving balance butanol and butyric acid production. It will be meaningful for further improvement of butyl butyrate production.

Fig.1 Schematic of the coupling metabolic pathways for butyrate synthesis5.

2. Methods

The genome-scale model (GSM) of C. tyrobutyricum (iCT583) of was used as a foundation to supply the butanol production pathway (Table,1) and the new named the new model iCT583-adhE2. For analyzing what is the key factor for driving the carbon flux distribution, we assumed two extreme situations, which were only butanol production and only butyrate production. We want to further understand the metabolic differences in these extreme situations. It will benefit to design the metabolic engineering strategy for controlling products distribution. The COBRAToolbox 2.0 was used to read and carry out the simulation program in the MATLAB environment 6 GLPK, a linear programming solver, was used for performing flux balance analysis (FBA) 7.

Table 1 The reaction for butanol production in iCT583-adhE2

Reaction ID	Reaction*	Function
CTT0060	Butanol[e] <=> Butanol[c]	Transport of butanol
CTE0006	Butanol[e] ->	Exchange of butanol
CTR0657	H+[c] + NADH[c] + Butyryl-CoA[c] -> NAD[c] + CoA[c] + Butyraldehyde [c]	Butyraldehyde dehydrogenase
CTR0658	H+[c] + NADH[c] + Butyraldehyde [c] -> NAD[c] + n-Butanol[c]	Butanol dehydrogenase

*[c] and [e] present Intracellular and extracellular metabolites, respectively.

3. Results and Discussion

For only butanol production, we constrained the butanol flux rate from 0 to 4 mmol/g CDW/h. and butyrate ethanol and acetate cannot produce. Then. the flux rate of other products and the turnover of cofactors (involved in ATP, NADH, NADPH and Ferredoxin) during the range of butanol flux rate from 0 to 4 mmol/g CDW/h were extracted and analyzed. As shown in Fig. 2A, we can see the cell growth rate and CO2 were increase with butanol production increase. The lactate was negative couple with butanol production. Notably, in this situation, there was not H2 emission. As shown Fig. 2B，we can see the turnover rate of NADH and ferredoxin were more significate change with increase of butanol production. It indicated that butanol production can couple with NADH and ferredoxin turnover.

Fig.2 In silico analysis of the only butanol production metabolic model in C. tyrobutyricum.

with increasing the specific butyrate production rate, (A) presents the changes in specific production rates for the major products; (B) is the change of the turnover rates of NADH, ATP, NADPH, and reduced ferredoxin; (C) presents the flux change of the key reaction in ferredoxin loop. (D) presents Schematic of the coupling metabolic pathways for only butanol production.

Some key reactions were labeled with the corresponding enzyme names and reaction IDs. BCD/ETF: butyryl-CoA dehydrogenase complex, CTR0417; HYD: hydrogenase, CTR0417, POR: pyruvate ferredoxin oxidoreductase, CTR0461, RNF: Rnf complex, CTR0447.

Through checking ferredoxin loop (Fig.2C and D), we found that reductive ferredoxin can only be oxidated by Rnf complex. Based on the stoichiometric coefficient, one reductive ferredoxin oxidated by Rnf complex can produce one NADH, but by hydrogenase will not produce NADH. No H2 emission means all ferredoxin oxidation in whole network were taken over by Rnf complex. It will provide extra NADH to butanol synthesis and make all cofactor turnover to a balance like in Fig2D. One glucose is covered to pyruvate produce 2 NADH by glycorlysis, then 2 pyruvates enter the first ferredoxin loop (composed by RNF complex and pyruvate oxidoreductase, POR) produce 2nadh, in which pyruvate covered to acetyl CoA coupled Rnf complex. Furthermore, converting crotonyl-CoA to buyryl-coA by butyryl-CoA dehydrogenase complex (BCD-ETF) can also be coupled with Rnf complex produce one NADH (the second ferredoxin loop). These NADH just match the consumption of NADH for butanol synthesis. These results demonstrated that forming ferredoxin loop with Rnf complex is key factor for providing adequate NADH to driving butanol production.

Same strategy for butyrate production, we found that H2 emission also coupled with butyrate production at the butyrate flux rate from 0 to 4 mmol/g CDW/h (Fig.3 A). It demonstrated that hydrogenase also take over a part of ferredoxin oxidation. Moreover, the increase of cell growth was more significantly than butanol production situation, when the butyrate flux rate at 4 mmol/g CDW/h (Fig.3 A). this means that the situation of only butyrate production is more suitable to cell growth. Moreover, the NADH turnover rate was not distinctly change compared with the situation of only butanol production (Fig. 3B). Further checking ferredoxin loop (HYD/POR and BCD-ETF/RNF), we found that in the situation of only butyrate production, the glycolysis can produce 2 NADH and Rnf complex oxidated about one third reductive ferredoxin to produce NADH, and about two thirds reductive ferredoxin were oxidated by hydrogenase (Fig. 3C and D). These NADH just matched the NADH consumption for butyrate production. This indication the demand of NADH was reduced in this situation.

Fig.3 In silico analysis of the only butyrate production metabolic model in C. tyrobutyricum.

with increasing the specific butyrate production rate, (A) presents the changes in specific production rates for the major products; (B) is the change of the turnover rates of NADH, ATP, NADPH, and reduced ferredoxin; (C) presents the flux change of the key reaction in ferredoxin loop. (D) presents Schematic of the coupling metabolic pathways for only butanol production.

Some key reactions were labeled with the corresponding enzyme names and reaction IDs. BCD/ETF: butyryl-CoA dehydrogenase complex, CTR0417; HYD: hydrogenase, CTR0417, POR: pyruvate ferredoxin oxidoreductase, CTR0461, RNF: Rnf complex, CTR0447.

4. Conclusion

Overall, in C. tyrobutyricum, BCD-ETF involved in ferredoxin loop is the key factor for driving butyrate and butanol synthesis. For butanol production, it needs more NADH supplement by Rnf complex involving in ferredoxin oxidization. However, in silico, the only butyrate production was the best model for cell growth and HYD involved in ferredoxin loop is key factor for driving butyrate and acetate production. Therefore, if we want to make a situation of only butanol and butyrate production, regulating the ratio of involving in ferredoxin oxidization by HYD and Rnf complex could be more important, excepting for enhancing the flux of BCD-ETF.