Notebook
Our lab team kept an exceptional record of everything throughout the project. See the exact details of how our wet lab work was conducted.
Team:MiamiU OH/Design
Project Design
Major crop yield is globally declining while population growth is beginning to outpace global food production (1). Agricultural productivity must increase by up to 120 percent to meet this rising demand (2). Due to limits in cultivable land and clashes with global sustainability goals to reduce agricultural land consumption, the ideal solution is to increase crop yield at the level of the individual plant. Our team was inspired to tackle this problem through our iGEM project because of its rising urgency and our location in an agricultural area. The major barrier to increasing crop yield is the metabolic inefficiency of photosynthesis. The maximum observed photosynthetic efficiency in crops is less than 4 percent (3). This limit can stem from either of the 2 stages of photosynthesis: light reactions and Calvin-Benson-Bassham (CBB) cycle.
Our Focus
The native CBB cycle for the regeneration of RuBP. Our team focused on the CBB cycle; compared to the light reactions whose optimizations are notoriously difficult, this pathway provides the most direct target for metabolic engineering (4). We concluded the most promising portion of this cycle is the regeneration of intermediate RuBP (ribulose1,5-bisphosphate). The natural process for regeneration is often stalled due to the consumption of intermediates by other metabolic pathways. Our proposed solution is to reduce the number of intermediates within the cycle. By replacing the current pathway with one reliant on fewer intermediates and enzymes, we can create a more robust and continuous CBB cycle. This approach not only lowers intermediate siphoning from the pathway, but also decreases enzyme synthesis and maintenance costs.
We created two novel pathways to accomplish this goal. In both cases, these redesigned pathways replace the native pathway currently found in the cell, which is reliant on the enzyme sedoheptulose-bisphosphatase (SBPase).
One pathway utilizes the transaldolase enzyme, which already operates in the cell. By reversing its usual reaction we can skip the intermediate conversions from SBPase entirely. This pathway relies on fewer enzymes, and more importantly, uses fewer intermediates.
Unlike the transaldolase pathway, the glycolaldehyde pathway skips the entire natural pathway and truncates the 8-enzyme native pathway to one reliant on only 3 enzymes. This dramatically reduces the number of enzymes and skips over several intermediate steps. Instead, this pathway is reliant on one native and two nonnative enzymes.
Our Alternative Regeneration Pathways
Glycolaldehyde pathway: synthetic novel pathway
Our redesigned glycolaldehyde pathway involves the regeneration of RuBP by combining glyceraldehyde-3-phosphate (G3P) with glycolaldehyde to eventually form RuBP.
The glycolaldehyde pathway. In the native CBB cycle, every 6 completions of the cycle yield 1 G3P for a net fixation of 3CO2 molecules, but the remaining 5 produced G3P molecules are used to regenerate RuBP. The glycolaldehyde pathway instead converts 2 of the 5 G3P molecules into 3 glycolaldehyde. Fructose-6-phosphate aldolase, an enzyme in Escherichia coli, can combine 3 glycolaldehydes with the remaining 3 G3P to form 3 5-carbon arabinose-5-phosphates (Ar5P). Ar5P can then be directly converted to ribulose-5-phosphate, which reenters the normal regeneration cycle to become RuBP. This redesigned cycle has a major obstacle; the conversion of G3P to glycolaldehyde has no known enzyme recorded in the KEGG database. This does not, however, discredit the pathway; this enzyme could exist in uncharacterized cellular metabolisms, or be generated though accelerated evolution experiments. This factor means we would not have been able to easily analyze this pathway in vivo. However, the fact that this cycle can be analyzed in silico further illustrates the advantages that computational modeling can have for metabolic analyses.
Transaldolase pathway: redesigned native pathway
The redesigned transaldolase pathway for the regeneration of RuBP. The transaldolase pathway we designed skips a step of the native pathway by utilizing the reverse reaction of transaldolase, an enzyme already found in the cell. The current native pathway, also called the wildtype pathway, relies on ATP hydrolysis and the enzyme SBPase to convert sedoheptulose-1,7-bisphosphate (SBP). SBP is produced by the combination of two intermediates, erythrose-4-phosphate (E4P) and dihydroxyacetone phosphate (DHAP). Our proposed transaldolase pathway circumvents the dependence on these intermediates as well as the energy investment for the conversion to SBP. Instead of SBPase producing sedoheptulose-7-phosphate from sedoheptulose-1,7-bisphosphate, transaldolase can directly produce G3P and S7P from fructose-6-phosphate (F6P) and E4P if operating in reverse. This reverse reaction is typically performed in the oxidative portion of the pentose phosphate pathway and may be present at low levels for photosynthetic reactions in the wildtype cells. Additionally, the native pathway relies on the use of G3P twice, while our transaldolase pathway only uses this triose phosphate (3-carbon phosphate molecule) a single time, instead replenishing this intermediate through the transaldolase reaction. Therefore, our transaldolase cycle is most limited by F6P availability. Unlike pentose phosphates and triose phosphates on which the SBP pathway is most reliant, hexose phosphates (6-carbon phosphate molecules) like F6P are more easily accessible through pathways such as glycogen degradation. Ultimately, the transaldolase pathway provides a pathway hypothetically less dependent on intermediates with lower enzymatic need.
Of our two proposed pathways, the transaldolase pathway is the most feasible to assess in vivo. Therefore, this pathway was first analyzed in silico and has been the focus of all wet-lab efforts.
Computational Modeling
A more detailed description of computer modeling can be found under our Model page.
First, we set out to assess the implications our pathways may have on the growth and metabolism of the cells through computer-based modeling. All our pathways were able to be replicated in silico using a model developed by Jared Broddrick and his partners at UC-San Diego (5). Using multiple novel scripts, we then used these models to assess the pathway impact on cellular growth and metabolism. This allowed insight into the wet lab conditions that might optimize our pathways for biomass production, as well as clarify the expected and unexpected impacts of our imposed reactions. Although the glycolaldehyde pathway has the most obvious advantages, due to time and resource constraints of attempting to develop an enzyme that has not yet been characterized, the transaldolase pathway was chosen to be tested in vivo.
Wet Lab
A more detailed description of wet lab efforts can be found under our Notebook page.
Aims of Wet Lab
We aimed to remove SBPase activity to disrupt the natural regeneration pathway, thereby forcing the cell to use a redesigned regeneration pathway. SBPase is considered an essential enzyme in cyanobacteria, specifically due to its involvement in two vital metabolic reactions, only one of which is directly involved in RuBP regeneration. Therefore, we aimed to engineer cells without SBPase activity by first overexpressing suitable replacement enzymes. The enzyme transaldolase can circumvent the SBPase RuBP regeneration reactions, however, another challenge is that SBPase is a dual functional enzyme in cyanobacteria, also maintaining fructose-1,6-bisphosphatase (FBPase) activity. To mitigate this vital reaction loss, we also overexpressed an introduced FBPase. Our lab efforts began by first introducing our overexpression target genes with the goal of eventually deleting SBPase in these overexpression mutants.
We divided these genetic manipulations into 4 transformant lines. For overexpression, we created an overexpression of transaldolase, an overexpression of fructose-1,6-bisphosphatase, and transformant overexpressing both components. We were able to successfully develop all 3 of these overexpression mutants.
Although we did attempt to create SBPase deletion transformants, we expected these cells to be unviable due to the vital reaction. As expected, we were unable to create a fully segregated mutant (a mutant with all genome copies having the desired change) for complete deletion of SBPase.
Details of Plasmid Design, Mutant Generation and Growth Analysis
Wet lab began by designing plasmids to overexpress genes encoding transaldolase (tal) and fructose-1,6-bisphosphatase (fbp) as well as delete sedoheptulose-1,7-bisphosphatase (glpX). In total we created 4 plasmids for overexpression of tal, fbp, both, or the deletion of glpX, respectively. These plasmids and their components are detailed on the Contributions page.
The overexpression plasmids work by integrating the new genes into neutral site I in the cyanobacterial genome. This neutral site has no functional or coding DNA sequences, making it a safe site to insert new DNA without disrupting other gene sequences important for cell life. In the plasmid, sequences corresponding to regions flanking the neutral site region allow matching of the plasmid and genome to insert our target gene into the site.
The deletion plasmid instead inserts the antibiotic resistance marker kanamycin into the target glpX gene, functionally deleting the gene by preventing functional mRNA transcription. Instead of the flanking sequences corresponding to the neutral site, they correspond to regions within and surrounding glpX. Cells in which this deletion is successful do not have SBPase enzymes, and instead express resistance to kanamycin, and thus can be selected for by growing cells on a plate containing kanamycin.
In order to replicate our plasmid constructs, we first transformed the plasmids into E. coli, as Synechococcus elongatus PCC 7942 does not maintain or replicate plasmids. Once replicated in E. coli, these plasmids could be used to transform S. elongatus. This cyanobacterial strain is readily transformable, meaning that it easily integrates and maintains the change in its genome.
We selected for successful cyanobacterial transformants using the antibiotic resistant genes encoded in the plasmids, now stable within the genome. For the overexpression plasmids, this antibiotic was spectinomycin and streptomycin, both found within successful overexpression mutants. The deletion plasmid contained both kanamycin resistance and gentamicin resistance, with successful deletion mutants only containing kanamycin. This exact process is detailed in the Notebook page. Briefly, successful deletion mutants contain only kanamycin resistance due to double crossover with the plasmid, whereas a single crossover results in both kanamycin and gentamicin being in a cell retaining SBPase function. We confirmed that the desired changes were made by amplifying the insertion site using PCR and ensured that it was the correct size through gel electrophoresis.
Cell transformants were then used to generate growth curves. Based on results from the computational modeling, we assessed growth under light conditions mimicking day and night cycles, as modeling predicted significant variations between constant light and realistic cycles.
References
1. Tilman D, Balzer C, Hill J, Befort BL. 2011. Global food demand and the sustainable intensification of agriculture. PNAS 108:20260–20264. (https://doi.org/10.1073/pnas.1116437108)
2. Ort DR, Merchant SS, Alric J, Barkan A, Blankenship RE, Bock R, Croce R, Hanson MR, Hibberd JM, Long SP, Moore TA, Moroney J, Niyogi KK, Parry MAJ, Peralta-Yahya PP, Prince RC, Redding KE, Spalding MH, Wijk KJ van, Vermaas WFJ, Caemmerer S von, Weber APM, Yeates TO, Yuan JS, Zhu XG. 2015. Redesigning photosynthesis to sustainably meet global food and bioenergy demand. PNAS 112:8529–8536. (https://doi.org/10.1073/pnas.1424031112)
3. Zhu X-G, Long SP, Ort DR. 2008. What is the maximum efficiency with which photosynthesis can convert solar energy into biomass? COBIOT 19:153–159. (https://doi.org/10.1016/j.copbio.2008.02.004)
4. Bailey-Serres, J, Parker, JE, Ainsworth, EA et al. 2019. Genetic strategies for improving crop yields. Nature 575 109–118. (https://doi.org/10.1038/s41586-019-1679-0)
5. Broddrick JT, Rubin BE, Welkie DG, Du N, Mih N, Diamond S, Lee JJ, Golden SS, Palsson BO. 2016. Unique attributes of cyanobacterial metabolism revealed by improved genome-scale metabolic modeling and essential gene analysis. PNAS 113:E8344–E8353. (https://doi.org/10.1073/pnas.1613446113)
Explore Next
Computational Modeling
Using programming we were able to test out theories on different pathways in order to help out the wet lab work. Explore the different pathways.
Contact Info
The main point of contact on the Miami University iGEM team is the principal investigator Xin Wang. Xin Wang can be inquired below. For additional contact information on the team, please refer to the Team page.
- xwang@miamioh.edu
- (513) 529-5427
- 212 Pearson Hall, 700 E. High St
Oxford, OH 45056
Special Thanks
This work was funded and made possible by the National Science Foundation!