We began our project by first familiarizing ourselves with the climate crisis, carbon fixation pathways, and current initiatives in carbon sequestration. In this effort, we collectively read literature that cover the science behind climate change, natural carbon fixation pathways, synthetic carbon fixation pathways, and current efforts in carbon capture through synthetic biology [1].

From our literature search, we were introduced to the two main categories of carbon fixation - natural carbon fixation pathways and synthetic carbon fixation pathways [2]. Synthetic carbon fixation pathways are human designed pathways that are pieced together using characterized enzymes. Because of this, synthetic carbon fixation pathways are well-characterized and can have higher carbon fixation rates than natural pathways. In particular, one literature that we drew inspiration the most from is the development of the pyruvate carboxylase (PYC), oxaloacetate acetylhydrolase (OAH), acetate-CoA ligase (ACS), and pyruvate synthase (PFOR) (POAP) cycle which is the first reported in vitro minimized carbon fixation pathway [3]. The simplicity of the pathway, due to its minimized enzymatic reactions, and high reported carbon fixation rates makes it an enticing candidate for in vivo applications. When selecting the biological chassis, we had to consider the need for easy analysis of pathway impact on biological functions because in vivo implementation of minimized carbon fixation cycles is unexplored. It is important for us to identify how such a pathway can affect biological processes and in response, find ways to optimize pathway compatibility and activity. Combined with considerations on biocontainment, safety, and ease of engineering, we chose the minimal cell (JCVI-Syn3B) to be our biological platform. The minimal cell is both the world’s simplest and synthetic cell. This means the minimal cell is very well characterized and grows only in specific laboratory conditions [4]. Its genome also contains a cre-loxp site, known as the landing pad, that allows for efficient integration of genes into the genome. This prompted us to imagine a carbon fixation system that is afforded by the POAP cycle and the minimal cell.

After imagining a carbon fixation system, we were led to the design of MACS - A Minimal, Adapted Carbon Sequestration system. The minimal part of the system arises from the placement of a minimized carbon fixation cycle into a minimal cell. The adapted part comes from the adaptive laboratory evolution (ALE) experiments we performed on the minimal cell to increase compatibility with the POAP pathway. Our first design of MACS entailed the selection of optimal variants for the enzymes used in the POAP cycle. One challenge we identified was that the POAP cycle activity was reported at 50 degrees Celsius in vitro which is hostile to the growth of minimal cells [3]. The main reason why the POAP cycle operates at this high temperature is because the pyruvate:ferredoxin oxidoreductase (PFOR) enzyme requires high temperature to catalyze pyruvate synthesis. Our first design goal is then to screen for variants of PFOR enzyme that have lower operating temperatures for pyruvate synthesis. Using the Braunschweig Enzyme Database (BRENDA), we looked at all characterized versions of PFOR from different organisms and finalized on Desulfocurvibacter africanus PFOR which has reported pyruvate synthesis activities at 40 degrees Celsius [5]. Likewise, we did the same search for all enzymes involved in the POAP cycle, using operating temperatures around 40 degrees Celsius and pH around 5-7 as screening criterias, and finalized on the following variants of enzymes to build our POAP cycle. Despite using the africanus PFOR, we were still concerned with the conditions being hostile for the minimal cells. Thus, we initiated an adaptive laboratory evolution project aimed at evolving minimal cell thermal tolerance (TALE) which runs parallel to our plasmid design and implementation cycle.

To finalize our design, we had to attach appropriate Mycoplasma promoters to our genes. By using characterized expression levels of Mycoplasma promoters identified in a transcriptomics study, we selected promoters with intermediate strength. These promoters, which have been registered as basic parts, were attached to the genetic sequences of our genes, and these sequences were ordered from Twist Biosciences. Because some gene fragments are large, Twist Biosciences opted to synthesize our gene fragments, namely Acetyl-CoA synthetase (ACS) and PFOR, in fragments of 2 and 3 respectively. Additionally, primers designed to amplify our genes and perform crossover PCR were also ordered from Twist Biosciences. We designed our building step to constitute assembly of individual genes into distinct plasmids. This affords us the ability to screen for the activity and biological compatibility of individual genes before assembly into the full POAP pathway. Combination of all 4 genes from the very beginning may lead to difficult troubleshooting in the face of obstacles since we will be more uncertain of where and how things failed.

Our building efforts began upon the arrival of our ordered genes and primers. We first performed PCR on our genes to both obtain larger quantities of our gene fragments and to prepare for E. Coli plasmid assembly. Our second step is to perform crossover PCR to combine the two fragments of ACS into one and the 3 fragments of PFOR into one. We then began to assemble our plasmids by combining the pCC1BAC EcoRV fixed HindIII Exp Vector backbone (provided by the Suzuki Lab) with our gene inserts using E. Coli plasmid assembly. Successful assemblies will supply transformed E. Coli cultures with resistance to chloramphenicol which is used as a selection marker.

Even though successful assemblies of our plasmid will lead to transformed E. Coli cultures growing on a plate containing an antibiotic, colonies should still be picked and have their plasmids mini-prepped and sequenced to confirm that our genes are unmutated and installed correctly. After mini-prepping all of our transformed E. Coli cultures and sending them off for sequencing by Plasmidsaurus, we confirmed that we have individual plasmids containing each component of the POAP cycle.

During our engineering cycle, we encountered multiple challenges in the combination of PFOR and ACS gene fragments to yield the final gene. In our troubleshooting, we tested different primer combinations and conditions for crossover PCR which eventually yielded us successful combinations. This learning experience gave us a better understanding on the principles of PCR, how certain primers may work better than others, and PCR conditions that can be fine tuned to gene concentration and size. In this process, we also learned the principles and protocol for E. Coli plasmid assembly which works similarly to Gibson Assembly. Lastly, our use of a copy control vector exposed us to a brand new field of plasmids which can have their rate of replication carefully controlled. Overall, we were able to improve on our initial protocol of crossover PCR and E. Coli plasmid assemblies and produce successfully assembled plasmids with desired gene inserts.

After obtaining assembled plasmids in E. Coli and confirming their fidelity, the very next step is to transform minimal cells with those plasmids. What’s particularly interesting is that the vector we are using to carry our genes of interest becomes a suicide plasmid once inside of the minimal cell. But because the plasmid contains a cre-loxp site, our gene of interest (along with an antibiotic resistance gene), can be integrated into the minimal cell genome which also contains a cre-loxp site known as the landing pad. This allows for the transfer of our gene of interest into the minimal cell, and we can use an antibiotic to select for the cells that underwent this transformation. In our design, we selected puromycin to be this selection marker. We also wanted a way to confirm that our gene is indeed integrated into the minimal cell. Thus, we designed primers that complements our gene and would produce PCR products from the minimal cell genome.

We proceeded to mini-prep more plasmids from E. Coli cultures that contain sequenced plasmids. These plasmids were used in Mycoplasma transformation, and we plated our minimal cell cultures onto SP4 agar plates containing puromycin.

Although we encountered challenges in transformation of certain plasmids, namely ACS and PFOR in the beginning, we eventually got transformed minimal cell colonies for all plasmids growing on puromycin SP4 agar plates. The growth of these colonies on puromycin plates alludes to the successful integration of our gene of interest into the minimal cell genome, but we needed a more concrete test. Thus, we performed PCR on our minimal cell culture using appropriate primers and ran them across an agarose gel. Their expected band size gives us confidence that our genes of interest are indeed integrated into the minimal cell genome.

In this engineering cycle, we learned new techniques such as Mycoplasma transformation as well as troubleshooting Mycoplasma transformation to get all of genes of interest integrated into the minimal cell. We were also exposed to the concept of suicide plasmids and became more familiar with the efficient integration of DNA using the cre-loxp system. The most valuable lesson we learned is a rigorous approach to troubleshoot and test things - including the use of PCR to confirm the presence of our gene in addition to just finding colonies on an antibiotic plate. From these lessons, we were able to continually improve our protocols, eventually getting transformations that failed in the beginning to work in the end.

Because the metabolites involved in the POAP cycle are mainly primary metabolites such as pyruvate and acetate, we wanted to develop an extraction and derivatization protocol that affords us untargeted metabolic profiling. Further, untargeted metabolic profiling also allows us to see the overall metabolic state of the minimal cell, and we can compare the profile of a transformed minimal cell with the untransformed minimal cell to observe the overall perturbations caused by one component of the cycle or by the overall cycle. With this goal in mind, we searched through literature to find reports of untargeted metabolite profiling. We found one literature in the Plant Journal that fitted our needs and adapted their procedure, with changes, for the minimal cell [6].

One thing we had to consider in our design for the extraction of minimal cell metabolites is that the minimal cells are fragile and can be ruptured easily. Because of this, we lowered the centrifugation speed to 2,800x RCF to ensure that cells can be pelleted but are not ruptured. Another thing we had to consider is the necessity of washing the cell pellets. SP4 media, the media that minimal cells grow in, is rich in cofactors, nutrients, and salts which can add lots of noise to a GC-MS profile. We must add extra steps of cell pellet washing to remove these contaminants which may add noise to the GC-MS analysis. In designing this wash step, we decided to use a TRIS-Saline buffer to control the pH and maintain osmosis so the cells do not rupture. This wash buffer was designed to be 20 mM TRIS and 200 mM NaCl tuned to a pH of 7.5. In the derivatization portion of the protocol, another design decision was made to replace the lyophilization of extracted metabolites by solvents to be speed-vacuumed. This is both to save time (lyophilization will take overnight whereas speed-vacuum can evaporate solvents in 2 hours) and allows us to treat 12 samples at the same time (speed-vacuum we had access to can treat 12 samples at a time). The protocol for minimal cell metabolite extraction and untargeted metabolite profiling by GC-MS can be found in our Experiments section in this wiki.

With our developed extraction and profiling protocol, we processed multiple transformed minimal cell samples and developed standards to locate the derivatized product in the GC-MS trace. In our first round of experiment, we encountered two challenges that allowed us to learn and refine our protocol. The first challenge was the addition of organic solvents into plastic eppendorf tubes and the extraction of metabolites in these tubes. The concern is that organic solvents may strip plastic particles which may lead to noise later in the GC-MS analysis that can cover up smaller metabolic signals. Another challenge we encountered is the need for optimization of the GC-MS method. Because we are performing an untargeted metabolite analysis, metabolites of different size and polarities are screened with the GC-MS, from compounds that elute very quickly like pyruvate to compounds that elute much slower like lipids and cholesterol. We needed a GC-MS method that lends coverage for both the fast eluting metabolites and the slow ones, while also making sure that no metabolites are eluted so slowly such that they appear in the trace of a new run (which was actually observed initially in which slow-eluting metabolites of an old run appeared in the early elution phases of a new sample run).

From our two challenges encountered during the testing phase, we learned more about what needs to be considered in the extraction of metabolites, derivatization, and screening. Surprisingly, the GC-MS profiles from our first experiment were very clean with no noise. Thus, it is a pleasant surprise that extraction in plastic eppendorf tubes does not introduce plastic contaminants. In the case that extraction in plastic tubes will lead to contaminants, we could attempt to extract in glass vials by moving the cell pellets with a pipette tip to a glass vial prior to lyophilization. We can also avoid the centrifugation step post-extraction and just be very careful in picking up only the solvent and pipette it over a small filter before collection. From our challenges with the GC-MS profiling, we changed the settings of the GC with the help of our secondary PI, Dr. Immo Burkhardt, to have a broad coverage of derivatized metabolites. We also added an ethyl-acetate wash step in between each sample run to ensure that no metabolites from the previous run is eluted in a new run.

A primary identified problem with the minimal carbon fixating cell design was that the cell cannot withstand temperature extremes where PFOR operates in the correct direction. Research to overcome this issue began with looking at PFOR variants that could operate at lower temperatures. A solution appeared after consulting with synthetic biology experts.

On November 7, 2022, we consulted Dr. Clyde Hutchinson about our design bottleneck. Dr. Hutchinson suggested that we try to grow the minimal cell colonies at slowly increasing temperatures. The idea would be to select out the temperature resistant mutants through natural selection.

On November 11, 2022, we consulted with Dr. John Glass about Dr. Hutchinson’s idea. Dr. Glass suggested a serial passaging experiment where the current JCVI Syn3.0B cell is initially grown at 39oC. Using a thermocycler machine, we can use the heat gradient setting to find a thermal bottleneck temperature.

In designing our adaptive laboratory evolution (ALE) experiment, we consulted Dr. Kim Wise who recently collaborated with a researcher from the Palsson Lab at UC San Diego on an ALE project. From learning how the Palsson Lab automated their adaptive laboratory evolution project, we proposed our evolution project to be initially set at a bottleneck temperature in which cell growth is significantly hindered but present. The SP4 media contains phenol red which changes the color of the media from red to yellow when the media becomes acidic. This can be correlated to cell growth and concentration because minimal cells excrete lactic acid which decreases media pH. We would passage at a ratio of 1 part cell culture to 10 part fresh media when the SP4 media turns yellow. When overnight color change occurs at this passage ratio, we will increase the growth temperature to a new bottleneck temperature that slows the growth rate again.

Starting in December, we began our thermal adaptive laboratory evolution (TALE) project. We started at 42°C, after observing significantly hindered growth rate of minimal cells at this temperature, and continually propagated our cells when the media turned yellow. When overnight growth at 42°C was observed, we determined a new bottleneck temperature using a temperature gradient experiment run with PCR tubes and a thermocycler. We also had multiple lines of minimal cells evolving at the same time to increase the likelihood of beneficial mutations appearing and being selected out.

After a little under 4 months of continual evolution, we obtained strains of minimal cells that can grow overnight at 44.2°C. This is tested through culturing these evolved minimal cells using a 1:10 passage ratio in a waterbath that is measured to be 44.2°C water bath and observing the media change from red to yellow overnight. Aside from obtaining a strain of minimal cell that can optimize the activity of an implemented POAP cycle, we also extracted gDNA of both our evolved minimal cells and ancestral minimal cells. These gDNA were sent off for sequencing in the hopes of identifying the genes and mutations that give rise to thermal tolerance.

From this project overall, we have learned the techniques of running an adaptive laboratory evolution and how this experiment can lead to the development of desired phenotypes. Over the progress of TALE, we also improved our protocol to overcome challenges. For a majority of time, TALE experiments were ran in eppendorf tubes and the low headspace posed a higher risk of contamination. In response, we began conducting the experiment instead in 5 mL tubes with much greater headspace. Our switch from using heat blocks to waterbaths also allowed our cultures to maintain a more uniform and steady temperature throughout the experiment. We are also in the process of characterizing genes and mutations that are responsible for thermal tolerance in the minimal cell. This characterization work is planned to eventually result in a manuscript that investigates thermal tolerance in the minimal cell and Mycoplasma in general.