Introduction

Our team demonstrated engineering success through an iterative approach to metabolic engineering, resulting in the design of composite synthetic biology devices essential for achieving our goal of upcycling PET plastic. This undertaking was informed by our Integrated Human Practices work, where we consulted with Dr. Chitong Rao, Chief Scientist at Bluepha Co., Ltd.—a global leader in using synthetic biology for degradable bio-polymer production—and Dr. Joanne Sadler from the University of Edinburgh, whose work on producing vanillin from the microbial breakdown of PET served as initial inspiration for our project.

From these meetings, we learned that metabolic engineering to increase yields of useful products would be essential for making our concepts both economically viable and environmentally sustainable. This insight was further confirmed through our team's Life Cycle Analysis of our desired products and during the development of our business plan.

Our goal was to increase the production of poly(3-hydroxybutyrate) (PHB), a type of polyhydroxyalkanoate (PHA) that can be downstream depolymerized into beta-hydroxybutyrate (BHB). We chose this target for several reasons:

1. It is a type of PHA where Dr. Rao's expertise and the information he provided during our collaboration could be leveraged.
2. PHB can be depolymerized by an enzyme known to the team, as described by the 2024 Concordia iGEM team, thus enabling for the production of BHB as a high-value product from increased PHB levels.
3. We had already designed genetic circuits in both E. coli and P. putida to produce PHB, allowing our research to focus on enhancing this biosynthesis pathway familiar to the team.

Through literature research, we discovered that increasing the production of NADPH and Acetyl-CoA would result in increased PHB production via the PHB biosynthesis pathway (Shi et al., 1999). We utilized Flux Balance Analysis (FBA), a constraint-based metabolic modeling method, using several target genes, enzymes, and reactions identified from the literature (Lim et a., 2002; Zhang et al., 2020, Zhang et al., 2014).

The first iteration of the Design-Build-Test-Learn (DBTL) cycle tested whether incorporating certain genetic targets (identified from literature) would have the desired effect of increasing NADPH production by increasing metabolic flux. After verifying plausible targets, the second iteration determined whether our designed devices would elicit an increase in NADPH when modeled with expected enzyme concentrations and reaction fluxes.

Our engineering design cycle addressed the following questions:

1. Which potential gene/protein could be engineered into our devices that will increase the amount of NADPH available to the PHB biosynthesis pathway?
2. Will the engineering of new genes/proteins into these devices increase the amount of NADPH available to the PHB biosynthesis pathway when the solved enzyme concentrations and reaction rates are incorporated?

Iteration 1

Figure 1: DBTL cycle for Iteration 1.

Design

Suggested by the literature on engineering E.coli to improve poly(3-hydroxybutyrate) production, the team identified genes, enzymes, and reaction targets of which we hypothesized would be effective in increasing the production of Acetyl-CoA and NADPH. Subsequently, we hypothesized an increase in the production of PHB, a high-value product. (Zhang et al., 2014) In the metabolic engineering upregulation process, the reaction targets were G6PD and PDH, as they were proven to be responsible for the production of Acetyl-CoA and NADPH. (Lim et a., 2002; Zhang et al., 2020) The identified gene and enzyme targets included increasing zwf, increasing serA, and more.

Build

The team identified two models: iJN1463 for Pseudomonas putida KT2440 and iEC1356_Bl21DE3 for E. coli BL21(DE3). Using the model of Pseudomonas putida KT2440, we utilized Flux Balance Analysis, a constraints-based modeling, to test the effect of targets. To increase a certain gene or enzyme, the upper and lower bounds of the associated reaction need to be increased. Therefore, we identified the reactions that the selected target genes code for. The upper and lower bounds during the first iteration were theoretical values, as it was impossible to know the exact possible increase in the selected targets.

Test

We implemented a Python package, COBRApy, to conduct a flux balance analysis with the gene, enzyme, and reaction targets.

Learn

serA and zwf were shown to be successful in increasing the NADPH yield, though they had minimal effect on Acetyl-CoA yield. However, increasing NADPH yield can have the desired effect of downstream upregulation of key metabolites in PHB synthesis. Other targets showed insignificant changes in yields. Hence, serA and zwf were chosen as effective targets, specifically zwf being the most successful one.

Iteration 2

Figure 2: DBTL cycle for Iteration 2.

Design

From Iteration 1, the team designed genetic circuits using targets and standard parts, including inducible promoter systems T7 (E.coli) and Xysl (P.putida). The genetic circuits included serA and zwf as they were shown to be effective.

Build

To find yields of key enzymes, we aimed to find exact upper reaction bounds for our Constraint-Based Modeling. This was done through ODEs and enzyme kinetics for our designed circuits. We modelled PHGDH production by serA to obtain the maximum enzyme concentration. A similar work was done with the second target, zwf. Then, utilizing enzyme kinetics for our designed circuits, that value was converted to an upper bound that could be used in the FBA. Full working can be found here: Model

Test

To test how our system will behave after modifying the model, we used the upper bounds for our model from previous steps in the engineering cycle and conducted a second flux balance analysis for zwf.

Learn

The team learned that zwf increase is effective in increasing NADPH yield in our designed inducible systems. We confirmed our decision to utilize zwf in the experimentations, since it was also most effective in increasing NADPH and following PHB yield, a high-value product produced from PET plastic. We then proceeded to the final design based on data from FBA models.

Final Product (Designed Composite Parts)

Figure 3: ZWF gene (pink) which codes for Glucose-6-Phosphate-Dehydrogenase under the influence of T7 promoter. PhaC/A/B code for the biosynthesis of PHB. Bba_K2260002 is Phasin.

Figure 4: Zwf in P. putida under the Xysl/Pm expression system. SerA gene (pink) which codes for D-3-phosphoglycerate dehydrogenase. PhaC/A/B code for the biosyntehsis of PHB.

Through constraint-based metabolic modeling, our team demonstrated engineering success by showing that the increase in zwf and serA increased the amount of NADPH available to the PHB biosynthesis pathway. Moreover, our designed devices elicited an increase in NADPH when modeled with expected enzyme concentrations and reaction fluxes, thus confirming that, theoretically, our gene engineering result in increased PHB production.

References