In vivo Synthesis of Biodegradable Polymers
We produced poly-(3-hydroxybutyrate) [P(3HB)] as a substrate to fold biOrigami. P(3HB) is a biopolymer with thermoplastic properties, meaning that it contracts in heat. Unlike polystyrene, P(3HB) is biodegradable. Building on previous iGEM teams’ work on P(3HB), we have contributed BioBricks that increase P(3HB) production and facilitate its extraction from Escherichia coli, and have processed the polymer into useful sheets. See our biobricks here:See our BioBricks
Introduction and Background
We are using recombinant E. coli to produce P(3HB) for folding biOrigami.
Since plastics on earth have deleterious environmental impacts, we wanted to focus on producing one biodegradable plastic that could be synthesized for use on the moon or on Mars.
P(3HB) is the most common polyhydroxyalkanoate, which is a group of fully biodegradable polymers with properties similar to those of petroleum-based plastics. It can be produced in a variety of microorganisms, and its synthesis as a compound for energy storage in vivo has been extensively studied . P(3HB) accumulates inside cells, and has been reported to occupy up to 90% (w/w) of the dry cell mass in certain species . However, P(3HB) has yet to become integrated into the plastics market on Earth due to high costs of both fermentation and purification . This summer, we focused on making the production, extraction, and use of P(3HB) more feasible on a lunar or Martian base. The main project was increasing P(3HB) yield; preliminary data was obtained for the lysis system; and, we were successfully able to make sheets of P(3HB) by improving on the protocol developed by the 2012 Tokyo Tech iGEM team.
(1) Increasing P(3HB) Yield
Increasing the yield of P(3HB) from a cell culture would decrease the time needed to complete the plastic production and extraction process by making smaller cultures more efficient. We have used the phaCAB operon from Ralstonia eutropha H16 that encodes the three genes required for P(3HB) production: PhaA, PhaB, and PhaC. Previous iGEM teams (Tokyo Tech 2012 and Imperial 2013) were able to successfully produce P(3HB) in vivo using BioBricks that encode these three genes. The pathway for P(3HB) production is as follows. First, 3-ketothiolase, encoded by PhaA, combines two molecules of acetyl-CoA to form acetoacetyl-CoA. This is then reduced by acetoacetyl-CoA reductase, encoded by PhaB, to form (R)-3-hydroxybutyl-CoA. This is then polymerized by PHA synthase, encoded by PhaC, forming poly-3-hydroxybuterate.
To increase the yield of P(3HB), we looked for a way to increase the amount of acetyl-coA, the precursor molecule to P(3HB) formation. Acetyl-CoA is composed of an acetyl group bound to coenzyme A. Coenzyme A consists of a beta-mercaptoethylamine group linked to pantothenic acid. On the Tokyo Tech 2012 iGEM team’s wiki, their results showed that adding pantothenic acid into their culture media led to increased yields of P(3HB). Therefore, we decided to increase the production of coenzyme A, which would then allow the cells to synthesize greater amounts of P(3HB).
Coenzyme A biosynthesis is a five-step process that is primarily regulated by the first enzyme in the pathway, pantothenate kinase (encoded by the gene PanK, which is also called CoaA) . Pantothenate kinase stringently controls amount of coenzyme A produced in E. coli because the pantothenate kinase produced by E. coli is significantly inhibited by coenzyme A and its thioesters (such as acetyl-coA). Therefore, since we wanted to increase coenzyme A synthesis, we had to find a way to get around this negative feedback inhibition. Fortunately, the pantothenate kinase made in Staphylococcus aureus does not experience feedback inhibition from coenzyme A or its thioesters . Indeed, this allows S. aureus to accumulate high levels of coenzyme A. We ordered the S. aureus pantothenate kinase sequence from IDT and inserted into the backbone PSB1C3 in front of the phaCAB operon designed by the Tokyo Tech iGEM team in 2012 with the hybrid promoter designed by the Imperial College iGEM team in 2013. The results section below shows that inserting the gene for pantothenate kinase allows for higher levels of P(3HB) production in E. coli.
(2) Facilitating P(3HB) Extraction
Since bacteria that naturally synthesize P(3HB) use the plastic as an energy storage mechanism, the plastic is produced inside the cell and stays there. Currently, the isolation of P(3HB) from the cells and culture media is one of the main bottlenecks in the P(3HB) production process – and the main source of its high cost of approximately $4-6 USD . Traditional isolation methods include the use of large amounts of chemicals that can be hazardous to human health and the environment, such as chloroform and sodium hypochlorite, or other problems such as large volumes of waste water produced or impurities in the polymers . Since one of our team’s goals has been to reduce up-mass of materials on space flights, we wanted to minimize the amounts of harmful (and heavy) chemicals that would need to be brought up as payload to extract the P(3HB).
We sought to accomplish this by introducing a lysis device into the PSB1C3 plasmid that contained the phaCAB operon and the gene for pantothenate kinase. The lysis device was developed by the 2008 Berkeley iGEM team. The device contains T7 phage antiholin under a weak promoter, and T7 phage holin and endolysin inducible under (L)-arabinose. This lysis device was not reported to kill entire populations of E. coli, and we saw this as an advantage for a continuous plastic-production system. We would be able to cause part of the population to lyse and release its P(3HB) into the media, and the rest of the cells can continue to grow. The rest of the population would continue growing and producing plastic. The plastic can be recovered from the media by causing the granules to aggregate, and then requiring lower amounts of solvents and less time to complete the purification process . We were able to obtain preliminary data showing that the lysis system works.
We also worked on an auto-lysis system that would not need to be induced by arabinose. This system furthers the goal of being able to have a culture that continuously produces plastic. This device was made by combining the existing lysis+panK+phaCAB with a quorum sensing system using the lux genes and AHL. When the cell culture grows dense and enough AHL are produced, they will bind to the luxR receptor, which then induces the lysis system. We were not able to test this system due to time constraints, but we confirmed the sequence and submitted the BioBrick.
(3) Improving the qualities of P(3HB)
Our plan to improve the qualities of P(3HB) was to add the gene propionate CoA transferase from Clostridium propionicum (Pct) to the P(3HB) construct. This gene is supposed to cause the bacteria to produce P(3HB-co-3HV), a similar polymer with properties that make it more useful in manufacturing (and perhaps folding). The Yale 2008 iGEM team had a BioBrick for that part; however, after several rounds of unsuccessful assembly and sequencing, we realized that the sequence from the Registry was inconsistent, and was likely why this part didn't work. We did not have time to order the correct sequence of this part synthesized, but this is a promising area for future efforts.
Data and Results
We have improved a BioBrick to allow for E. coli to accumulate more plastic.
Our first set of experiments focused on testing the benefits of adding the S. aureus panK gene in front of the Imperial College 2013 team’s device. We measured the amount of P(3HB) produced in vivo in two ways: by extracting the plastic and measuring its mass, and by staining the cultures and measuring the fluorescence on a flow cytometer and fluorometer.
To extract the plastic, we used sodium hypochlorite to dissolve the lyophilized cells, freeing the plastic. The protocol (posted below in greater detail) then calls for several wash steps to purify the plastic. Measuring the mass of the culture post-lyophilization and measuring the mass of the plastic once extracted, purified, and fully dried allowed us to determine the percentages of the cells that were taken up by P(3HB) granules and compare between devices. The addition of panK to the existing P(3HB) device allowed for, on average, a 23% increase in the amount of plastic accumulated in vivo as a percentage of dry cell weight.
P(3HB) produced in vivo and extracted using sodium hypochlorite. The tube on the left shows a negative control, a culture with no genes for plastic production. The tube in the middle was from a culture with the phaCAB operon with hybrid promoter from Imperial College iGEM team, 2013. The tube on the right shows the plastic produced by a culture with the panK gene added to the Imperial team's construct.
To verify this increase in production, we also quantified plastic production using fluorescence. Nile red is a fluorescent dye that has previously been used to show production of poly-hydroxyalkanoates in bacteria . It binds to P(3HB) and other lipids, such as those in the cellular membranes. Therefore, staining with Nile red is expected to show fluorescence even for the negative controls - the cultures that do not contain P(3HB); however the fluorescence will be much higher in cultures that do contain P(3HB). The second way we quantified plastic production was by measuring the fluorescence of cell cultures and dividing by the density of the cell cultures using OD600 – this allows us to compare the relative fluorescence of cultures containing our construct compared to cultures containing other constructs. We also used a flow cytometer to qualitatively show an increase in fluorescence in cells that contained panK. The addition of panK to the existing P(3HB) device caused, on average, a 34% increase of fluorescence from Nile red bound to plastic. We hypothesize that the slightly larger increase in fluorescence than increase in percentage of plastic could be because the cells that accumulate more plastic would be larger, and have a greater surface area of cell membrane, and therefore would bind to more Nile red dye. They would fluoresce more not just because they have accumulated more plastic, but also because their cell membranes are larger – therefore, the percentage increase in fluorescence and percent mass would not be the same.
Our lysis experiments depended heavily on having good negative controls. The cultures we were testing were cells that contained the lysis device and the genes for plastic production, and they were induced with 1M (L)-arabinose. The first negative control was a cell culture that contained the lysis device and the genes for plastic production, but it was not induced with arabinose. We would expect that culture to produce plastic, but not lyse, so there would be no plastic in the media. The second negative control was a cell culture that contained the genes for plastic production, but not the lysis device; this culture was induced with arabinose, however we would not expect it to lyse, and all of the plastic would stay inside. Initial tests showed that the cells with the lysis devices had less viable cultures the next day; however, we ran out of time before we were able to test the media for P(3HB).
Device construction. Parts for this project were obtained from the 2015 Distribution Plates, ordered from the Registry, or synthesized by IDT. BioBricks assembly was used to construct the devices.
Transforming bacteria. Chemically competent NEB5-alpha E. coli cells from New England Biolabs were used for all parts of this project. Plates containing successful transformants were stored at 4C.
Growing Plastic. Transformed cells were inoculated into Terrific Broth (TB) medium supplemented with chloramphenicol and shaken at 37C for 72 hours.
Extracting Plastic Powder. Liquid cultures of P(3HB) were pelleted by centrifugation at five-minute cycles at 4,000xg at 4C. The tubes with pellets were then placed at -80C for at least one hour, and then rushed down to the lyophilizer, where they were allowed to freeze dry for at least 24 hours. The freeze-dried pellets were then ground up and re-suspended in a 6-10% (v/v) sodium hypochlorite solution. This reaction is exothermic and significant pressure can build up; we had an unfortunate incident wherein the Eppendorf tubes all popped open in the incubator after only a few minutes. To avoid this, we then started to use 15mL Falcon tubes since the tops could be better secured. The first 20 minutes of reaction were carried out at room temperature with the lids only attached lightly to allow air to escape. Then, the lids were tightened and the tubes were shaken at 37C for an hour, with the air pressure being released every five to ten minutes. The samples were then centrifuged at 4,000xg at 4C for five minutes, the supernatant was discarded, and then the samples were re-suspended in 70% (v/v) ethanol. The samples were centrifuged at the same settings again, the supernatant was discarded, and they were re-suspended once more in distilled water. The samples were centrifuged one last time, the supernatant was poured off, and then the remaining water was allowed to evaporate overnight at 40-60C. The final mass of the plastic was then measured. This protocol was adapted from Hahn et al . .
Extracting Plastic Sheets. Liquid cultures of P(3HB) were pelleted by centrifugation at five-minute cycles at 4,000xg at 4C. The tubes with pellets were then placed at -80C for at least one hour, and then rushed down to the lyophilizer, where they were allowed to freeze dry for at least 24 hours. The freeze-dried pellets were then ground up and added to a sealable glass container of chloroform. We used half as much chloroform as media in the original culture (100mL chloroform if we had cells growing in 200 mL TB media). The contents of the container were stirred for 24 hours and then filtered into a flask using a mixed cellulose acetate filter (glass fiber filters also work well). At this point, the P(3HB) and lipids from cells are dissolved in chloroform is in the flask. This solution was allowed to evaporate slightly, and then was dropped into a container of methanol. At this point, the plastic precipitates out, and the methanol dissolves the remaining lipids and non-polar cell particles. This solution is then filtered using a glass fiber filter to let the methanol and lipids through, and keep the P(3HB) in the filter. The filter was then changed to a different flask, and chloroform was added to the filter to bring the plastic through. The contents of the flask were poured into a glass petri dish and the chloroform was allowed to evaporate, leaving a plastic sheet at the bottom of the petri dish.
Using Nile Red to Quantify Plastic Production. We stained cultures that had been growing for three days with Nile red to measure fluorescence. The final data is from three devices, and five cultures of each device. The devices were the panK plus phaCAB operon, just the phaCAB operon, and a negative control that contained the gene for LuxL (no plastic production). Our staining protocol was adapted from Kelwick et al. . An 80 ug/mL Nile red stock solution was prepared in DMSO. 1mL samples of cell culture were centrifuged at 8,000RPM at room temperature for five minutes and the supernatant was discarded. The cells were re-suspended in 1mL of phosphate buffer saline (PBS) and then centrifuged again at 8,000RPM for five minutes at room temperature. The supernatant was discarded and the cells were re-suspended in .5mL 35% (v/v) ethanol and rocked at room temperature for 15 minutes to fix the cells. These cultures were then centrifuged at 8,000RPM for five minutes at room temperature once more, and the supernatant was poured out and the remaining ethanol was pipetted out. 930mL of PBS and 70uL of the Nile red stock solution were added to the tubes, and the pellets were re-suspended and then rocked on ice for one hour. Once Nile red is added, exposure to light should be minimized. For the flow cytometry experiment, the samples were diluted 1:100 in PBS. For the plate reader experiments, the samples were diluted 1:10 in PBS. The OD600 was measured first, and then the fluorescence was measured. The cells were excited at 535nm and the emission was measured at 605nm, with a cut-off at 590nm. The relative fluorescent units (RFU) of each well were divided by the OD600, to standardize the final data.
 Anderson, A. J. and Dawes, E. A. Occurrence, Metabolism, Metabolic Role, and Industrial Uses of Bacterial Polyhydroxyalkanoates. Microbiological Reviews 54 (4), 1990. 450 – 472.
 Madison, L. L. and Huisman, G. W. Metabolic Engineering of Poly(3-Hydroxyalkanoates): From DNA to Plastic. Microbiol Mol Biol Rev 63 (1), 1999. 21 – 53.
 Kunasundari, B. and Sudesh, K. Isolation and recovery of microbial polyhydroxyalkanoates. eXPRESS Polymer Letters 5 (7), 2011. 620 – 634.
 Leonardi, R. et al. Coenzyme A: Back in action. Progress in Lipid Research 44, 2005. 125 – 153.
 Leonardi, R. et al. A Pantothenate Kinase from Staphylococcus aureus Refractory to Feedback Regulation by Coenzyme A. The Journal of Biological Chemistry 280, 2005. 3312 – 3322.
 Jacquel, N. et al. Isolation and purification of bacterial poly(3-hydroxyalkanoates). Biochemical Engineering Journal 39 (1), 2008. 15 – 27.
 Rahman, A. et al. Secretion of polyhydroxybuterate in Escherichia coli using a synthetic biological engineering approach. Journal of Biological Engineering 7 (24), 2013.
 Spiekermann, P. et al. A sensitive, viable-colony staining method using Nile red for direct screening of bacteria that accumulate polyhydroxyalkanoic acids and other lipid storage compounds. Arch Microbiol 171 (2), 1999. 73 – 80.
 Hahn, S. K. et al. Recovery and Characterization of Poly(3-Hydroxybutyric Acid) Synthesized in Alcaligenes eutrophus and Recombinant Escherichia coli. Applied and Environmental Microbiology 61 (1), 1995. 34 – 39.
 Kelwick, R. et al. A Forward-Design Approach to Increase the Production of Poly-3-Hydroxybuterate in Genetically Engineered Escherichia coli. PLOS One DOI: 10.1371/journal.pone.0117202, 2015.