Analysis

There are three major questions we have answer in result analysis

1. How much CO₂ (air) uptake by engineering E. coli?
2. How much CO₂ can be fixed in biomass?
3. If the pyruvate trend match biomass trend?

Amount of CO₂ uptake

At first, we set up a closed system to model CO₂ uptake by E. coli. After a period of time, concentration of CO₂ uptake reached a balance. The result can tell that more than 50% air CO₂ diffuse into E. coli in a closed system.

However, we cannot set a CO₂ utilization system in a closed system. No matter our application is continuous culture or fed-batch culture, air CO₂ concentration will be constant, which means a boundary condition. Therefore, we model again under open system. Under constant air CO₂ condition, CO₂ diffusion speed changed with time and we categorized them into three phases, which are steady climb phase, transition phase, and saturated phase. The time segments of these three phases depend on CA activity. We assume A, B, C as time intervals of three phases. The cover area in the result shows that different time interval refers to different amount of CO₂ uptake by E. coli. Its CO₂ uptake rate change with time as well. As a result, we collect the CO₂ uptake rate and then calculate with the xylose consumed rate and pyruvate produced rate. What’s more, the main reaction of CO₂ in engineered E. coli happened between RuBP to 3PGA, which is another part we are going to discuss.

1. Engineered E. coli with CA


Time A was about 10 min (or 700 s) with [CO₂ uptake] reach 1 mM, causing 25% CO₂ uptake in total.

Time B was about 1 hour (or 3500 s) with [CO₂ uptake] equals to [air CO₂], causing 75% CO₂ uptake in an hour.

Time C was about 1.5 hour (or 8000 s) for [CO₂ uptake] reach balance with the highest CO₂ uptake percentage, 90%.

Time interval	Xylose supplied	Concentration CO2 uptake (mM)	Xylose (mM/s)	Pyruvate (mM/s)	Total CO2 uptake percentage
A	0.2%	1	0.008	0.016	25%
B	0.2%	2.045	0.00264	0.00528	75%
C	0.2%	2.15	0.001131	0.002263	90%

Table 1

xylose consumed rate and pyruvate produced rate under different CO₂ uptake time interval with CA

2. Engineered E. coli without CA


Time A was about 10 min (or 700 s) with [CO₂ uptake] reach 1 mM, causing 25% CO₂ uptake in total.

Time B was about 1 hour (or 3500 s) with [CO₂ uptake] equals to [air CO₂], causing 75% CO₂ uptake in an hour.

Time C was about 1.5 hour (or 8000 s) for [CO₂ uptake] reach balance with the highest CO₂ uptake percentage, 88%.

Time interval	Xylose supplied	Concentration CO2 uptake (mM)	Xylose (mM/s)	Pyruvate (mM/s)	Total CO2 uptake percentage
A	0.2%	1	0.0036345	0.007504	28%
B	0.2%	2.045	0.0017486	0.003888	72%
C	0.2%	2.15	0.001631	0.00388	88%

Table 2

xylose consumed rate and pyruvate produced rate under different CO₂ uptake time interval without CA

From table 1 and table 2, we knew that the highest pyruvate produced rate happened under the lowest CO₂ uptake percentage. Besides, pyruvate produced rate of engineered E. coli with CA is higher than that of engineered E. coli without CA. Although our working space was open system that we cannot sense precise data of the change of CO₂ concentration. Through pyruvate produced rate, we can easily recognize which phase of CO₂ uptake and then figure out the percentage of total CO₂ uptake. It is the method we calculate how much CO₂ uptake by our engineered E. coli.



A



B



C

Fig 2. result of xylose and pyruvate under A, B, C, time interval

Actually, xylose consumed rate is slightly related to has little relationship with the CO₂ uptake rate, since that xylose metabolism wasn’t just a single pathway. What we can analysis is that the pyruvate produced rate being correlation with CO₂ uptake rate, which help us to define the question that how much CO₂ uptake by engineered E. coli. It can also fit with experimental data easily. Next, we discuss about the true CO₂ reaction in E. coli CO₂ utilization bypass pathway. Every single mole of CO₂ uptake will react with one mole of RuBP and then produce 2 mole of 3PGA.



Fig 3. result of RuBP and 3PGA during CO₂ uptake

Since that RuBP and 3PGA are just intermediate products in metabolism, their concentration is quite low. Besides, results of three CO₂ uptake time interval showed similar. We still can see that 3PGA produced is 2 times larger than RuBP produced. We then calculate their produced rate in three CO₂ uptake time interval.

Time interval	RuBp produced rate (mM/s)	3PGA produced rate (mM/s)
A	4.25E-09	6.35E-10
B	3.89E-09	8.42E-09
C	1.76E-09	3.71E-09

Carbon metabolism flux

The main metabolic pathway of xylose in E. coli is PP pathway and glycolysis. As for recombinant E. coli, it has multiple xylose metabolic pathways, and we can simplify them into original pathway and CO₂ Bypass pathway. Therefore, we need to define the percentage of xylose, which is consumed by engineered E. coli, entering CO₂ bypass pathway and utilize CO₂.

It takes a lot of time to get absolute metabolic flux of CO₂ in engineered E. coli and require feed of ¹³CO₂ during cultivation. Since the metabolic flux of the original metabolic pathway is quite stable, the relative metabolic flux of CO₂-utilization over that of the original metabolic pathway could show a quantitative understanding on the CO₂ utilization efficiency. This relative value was MFI_CO2^[1], a term as metabolic flux index of CO₂-utilization pathway in heterotrophic engineered E. coli. This is the percentage of xylose that we mentioned above.

To define the MFI_CO2, we use CO₂ fixed by the CO₂ bypass pathway, noted as b, divided by the 3PGA generated from the central pathway, noted as a. We also assume c is mol of 3PGA¬0 and d is mol of 3PGA’ that channels into downsteam metabolism. After metabolism, (a+b) mol of 3PGA₀ and b mol of 3PGA’ are generated.

Besides, X and Y represent the actual 3PGA detected from the original pathway and CO₂ bypass pathway, which show in 3PGA₀ and 3PGA’ in the Fig 1., respectively. In the experiment, we use ¹³C-labeled CO₂ and unlabeled sugar to get the amount of 3PGA₀ and 3PGA’. However, it was reported that 3.45% of unlabeled 3PGA, which is noted as 3PGA’, will convert to its isotopic during the culturing E. coli strains in medium. Eventually, we concluded these situation into two equations.

$${y = b + 3.45\% \times (a+b) - d ......(1)}$$

$${x = (1-3.45\%) \times (a+b) - c ....(2)}$$

Since d/c = y/x under a metabolic steady-state, we derive equation (1) and (2) into a final relationship between a, b, x, and y.

$${MFI(Metabolic \ flux \ index) = {b \over a} = {{0.97y-0.03x} \over {1.03x-0.97y}}}$$

As a result, we only need the amount of 3PGA₀ and 3PGA’ to calculate MFI_CO2. Through modelling, we supply 0.4% xylose and 5% CO₂ to get the data of 3PGA₀ and 3PGA’, which helps us to adjust the rate between xylose and CO₂ sources.

References

1. Michaelis Menten Kinetics in bio – physic wiki, web : http://www.bio-physics.at/wiki/index.php?title=Michaelis_Menten_Kinetics
2. citric acid cycle from Brenda, web : https://www.brenda-enzymes.org/pathway_index.php?ecno=&brenda_ligand_id=Alpha-ketoglutarate&organism=Escherichia+coli&pathway=citric_acid_cycle&site=pathway
3. U. Sauer, J. E. Bernhard, The PEP—pyruvate—oxaloacetate node as the switch point for carbon flux distribution in bacteria. FEMS Microbiology Reviews, Volume 29, Issue 4, 1 September 2005, Pages 765–794.
4. O. Mugihito, S. Hideaki, T. Yukihiro , M Noriko, S. Tatsuya, O. Masahiro, I. Ayaaki, S. Kenji, Kinetic modeling and sensitivity analysis of xylose metabolism in Lactococcus lactis IO-1. Journal of Bioscience and Bioengineering VOL. 108 No. 5, 376–384, 2009.
5. W. Akira, N. Keisuke, H. Tomohiro, S. Ryohei, Reaction mechanism of phosphoribulokinase from a cyanobacterium, Synechococcus PCC7942. Photosynthesis Research 56: 27–33, 1998
6. G. B. Guillaume, D. F. Graham, T. J. Andrews, Despite slow catalysis and confused substrate specificity, all ribulose bisphosphate carboxylases may be nearly perfectly optimized Proc Natl Acad Sci U S A. 2006 May 9; 103(19): 7246–7251.
7. L. Yun, A. M. Keith, Determination of Apparent Km Values for Ribulose 1,5- Bisphosphate Carboxylase/Oxygenase (Rubisco) Activase Using the Spectrophotometric Assay of Rubisco Activity. Plant Physiol. (1991) 95, 604-609
8. Rong-guang Z, C. Evalena A., Alexei S., Tatiana S., Elena E., Steven B., Cheryl H. A., Aled M. E., Andrzej J., and Sherry L. M. Structure of Escherichia Coli Ribose-5-Phosphate Isomerase: A Ubiquitous Enzyme of the Pentose Phosphate Pathway and the Calvin Cycle Structure, Vol. 11, 31–42, January, 200
9. Inês L., Joana F., Christine C., Sandra M., Nuno S., Nilanjan R., Anabela C., and Joana T. Ribose 5-Phosphate Isomerase B Knockdown Compromises Trypanosoma brucei Bloodstream Form Infectivity PLoS Negl Trop Dis. 2015 Jan; 9(1): e3430.
10. Singh2006 TCA mtu model1. SBML2LATEX. Web : http: //www.ra.cs.uni-tuebingen.de/software/SBML2LaTeX
11. J. Shen, Modeling the glutamate–glutamine neurotransmitter cycle, Front. Neuroenergetics, 28 January 2013
12. X. Feng, H. Zhao, Investigating xylose metabolism in recombinant Saccharomyces cerevisiae via 13C metabolic flux analysis, Microb Cell Fact. 2013; 12: 114.
13. D. Runquist, M. Bettiga, Increased expression of the oxidative pentose phosphate pathway and gluconeogenesis in anaerobically growing xylose-utilizing Saccharomyces cerevisiae, Microbial Cell Factories 2009, 8:49
14. Kalle Hult rev 2005, 2007 Linda Fransson Department of Biotechnology KTH, Stockholm, Enzyme kinetics, An investigation of the enzyme glucose-6- phosphate isomerase
15. Model name: “Mosca2012 - Central Carbon Metabolism Regulated by AKT”, SBML2LATEX. Web : http: //www.ra.cs.uni-tuebingen.de/software/SBML2LaTeX
16. M. Ettore, A. Roberta, M. Carlo, B. Annamaria, C. Gianfranco, M. Luciano, Computational modeling of the metabolic states regulated by the kinase Akt, Front. Physiol., 21 November 2012
17. E. G. Jacqueline, P. L. Christopher, R. A. Maciek, Comprehensive analysis of glucose and xylose metabolism in Escherichia Coli under aerobic and anaerobic conditions by 13C metabolic flux analysis, Metabolic Engineering Volume 39, January 2017, Pages 9-18
18. N. N. Ulusu, C. Şengezer, Kinetic mechanism and some properties of glucose-6- phosphate dehydrogenase from sheep brain cortex, Türk Biyokimya Dergisi [Turkish Journal of Biochemistry–Turk J Biochem] 2012; 37 (4) ; 340–347
19. H. Stefania, M. Katy, C. Carlo, M. Morena, D. Franco, 6-Phosphogluconate Dehydrogenase Mechanism EVIDENCE FOR ALLOSTERIC MODULATION BY SUBSTRATE, J Biol Chem. 2010 Jul 9; 285(28): 21366–21371.
20. K. Nielsen, P.G. Sørensen, F. Hynne, H. G. Busse, Sustained oscillations in glycolysis: an experimental and theoretical study of chaotic and complex periodic behavior and of quenching of simple oscillations, Biophysical Chemistry 72 (1998) 49–62
21. UniProtKB - A0RV30 from web : https://www.uniprot.org/uniprot/A0RV30