Part:BBa_K2485013

Arabinose Inducible Proteorhodopsin-cPT-RFP

This is our full expression construct for Rho-cPT-RFP, which we also refer to as the Voltage protein.

Sequence and Features

Assembly Compatibility:

• 10
  COMPATIBLE WITH RFC[10]
• 12
  INCOMPATIBLE WITH RFC[12]

  Illegal NheI site found at 1205
  
• 21
  INCOMPATIBLE WITH RFC[21]

  Illegal BamHI site found at 1144
  Illegal BamHI site found at 2353
  
• 23
  COMPATIBLE WITH RFC[23]
• 25
  INCOMPATIBLE WITH RFC[25]

  Illegal AgeI site found at 979
  
• 1000
  INCOMPATIBLE WITH RFC[1000]

  Illegal SapI site found at 961
  

Introduction

Currently the ISS relies primarily on solar arrays for power, turning to lithium ion batteries when not in direct sunlight. But given the high costs of sending massive objects into space and the hazardous volatility of current batteries, these may not be the most sustainable sources for power. Looking for solutions in nature, we were inspired by the electric eel, whose linearly oriented electrogenic cells unipolarly pump ions in a single direction to generate an electric potential. This inspired our idea of a biobactery. Using E. coli, we mimicked the two key characteristics of eel cells: (1) stimulus-mediated unidirectional ion transport to generate a voltage and (2) linearly oriented cells that propagate this potential, like batteries in sequence. Ultimately, we aimed to use high-density E. coli colonies within a microfluidic device to polarly pump hydrogen ions via optogenetic activation of an ion channel, bacteriorhodopsin, to produce a voltage difference across the device.

Construct Design

The Voltage protein we designed is a triple fusion protein. Its schematic is shown in Figure 1. Each domain has a distinct function. From the amino terminus to the carboxy terminus, these domains are the proteorhodopsin domain, the cPT domain, and the RFP domain. The first domain contains a green-light absorbing proteorhodopsin. We chose to use a proteorhodopsin rather than a channelrhodopsin or a bacteriorhodopsin, because eukaryotic channelrhodopsins are difficult to express in E. coli due to the requirement of post-translational modifications, and because bacteriorhodopsins form a hexagonal lattice structure which is very important to functionality, less so than the trimeric structure formed by proteorhodopsin.1,2

Figure 1: Schematic of the proteorhodopsin, cPT, RFP fusion protein. The proteorhodopsin is embedded in the inner membrane and linked to the cPT region with a GSx5 linker.

Proteorhodopsin is a transmembrane proton pumping protein endogenous to SAR86 γ-proteobacteria. In response to light, proteorhodopsin pumps protons out of the cell and helps its native organism establish the proton motive force necessary to make ATP.3 It does form a trimeric lattice structure in the inner membrane; however, it has been shown that the individual monomer contains all the necessary components to pump protons, albeit at a potentially lesser efficiency.4 We sourced the sequence for the green-light absorbing proteorhodopsin from the 2012 Caltech iGEM team part BBa_K773002; however, codons towards the carboxy terminus were slightly modified to enhance GC content for IDT synthesis.5 This is the first domain in the protein because the carboxy terminus of proteorhodopsin is intracellular, while the amino terminus is extracellular, and the remaining domains must be intracellular for functionality.6

The second domain of the Voltage protein is the cPT, or central polar targeting BBa_K2485011, region of the larger IcsA protein. In its native organism, Shigella flexneri, the IcsA protein localizes to the outer membrane of the old pole and is responsible of the motility of the organism within mammalian host cells via actin polymerization. The cPT domain within this larger protein, when fused to GFP, has been shown to form a unipolar focus in 64% of E. coli cells.7 It should be noted that this focus occurs intracellularly since neither the cPT domain, nor the GFP contain export signals.To fuse this domain to the carboxy terminus of the proteorhodopsin, we searched the literature for fusion proteins with proteorhodopsin. After being unable to find any previous attempts at this fusion, we decided to incorporate a flexible 5x Glycine-Serine linker between the proteorhodopsin and the cPT region.

The final domain of the Voltage protein is a fluorescent reporter so that we can visualize the proteins localization. For this reporter, we chose to use FresnoRFP with an ex/em of 553/592.8 We chose this protein because its ex/em is sufficiently far away from the excitation frequency of 525nm of green-light absorbing proteorhodopsin, and we wanted to minimize excitation of the rhodopsin during visualization.3 To link the cPT region to the RFP, we used the same sequence as was used by Dr. Marcia Goldberg, M.D. from Massachusetts General Hospital who graciously gave us her plasmids which contained a cPT-GFP fusion. We showed that this fusion protein exhibited widespread unipolar localization in E. coli. In order to link cPT to GFP, the last four nucleotides of the cPT region was altered from “TCAT” to “ATCC.” We are unsure of the nature of this modification; however, it has been demonstrated to work efficiently.

By creating this fusion protein and demonstrating both the functionality of the proteorhodopsin and its unipolar localization, we hope to allow E. coli to generate a voltage differential across its length. Then, by aligning an entire culture of E. coli such that all the old poles, where the proteorhodopsin localizes, point in the same direction, each voltage will be additive like batteries in series. Although E. coli are very small, and proton diffusion is very fast, it has been shown that E. coli expressing proteorhodopsin will generate a voltage drop across the length of a cuvette in response to a unidirectional pulse of light.9 This is hypothesized to be a result of asymmetric excitation of the proteorhodopsin due to E. coli cells acting as a lens, focusing light on the far end of the cell. By controlling the expression of the proteorhodopsin to be unipolar, we hope to be able to generate this same sort of voltage without the need for the light to be unidirectional.

Experimental Design

Rhodopsin Functionality

Figure 2: In our experimental setup, we inserted a pH probe into our bacterial cultures and exposed them to a 2000W Xenon lamp, two minutes on and two minutes off for twelve minutes.

In order to assess the ability of the Voltage protein to pump protons, we grew liquid cultures of NEB5α with and without our plasmid containing the Voltage protein gene under the control of an arabinose inducible promoter. All cultures were grown in LB in the presence of 10µM all-trans retinal at 37°C in dark 15mL Falcon tubes. The cultures containing our plasmid were grown with 0.2% arabinose to induce the production of our protein and with chloramphenicol, as our plasmid also harbored a resistance gene to this antibiotic.

On testing day, the OD600 of each culture was measured. Based on this measurement, a calculated volume of culture was spun down such that when resuspended in 10mL of the testing medium, the OD600 was 0.5. Once the cells were pelleted, the growth medium was discarded leaving only the pellet. Approximately 25 minutes before testing, pellets were resuspended in 10mL of testing medium which was comprised of LB with sodium hydroxide added to adjust the pH to 9.04. We chose to work in an alkaline pH because the pH of the native ocean environment of SAR86 γ-proteobacteria is alkaline with a pH of ~8.2. 4

Figure 3: pH measurements were taken every 30 seconds to accuracies of three decimal places. Values were adjusted such that the data plotted is the pH difference from the start of the experiment.

For testing we illuminated each sample through 4 layers of green cellophane paper with a 2000W Xenon lamp with a UV filter, see Figure 2. Samples were exposed to alternating light and dark periods, 2 minutes each for a total of twelve minutes. pH measurements were recorded every 30 seconds with a pH meter accurate to 3 decimal places on a continuous read. Some samples were also exposed to varying concentrations of silver nitrate in an effort to stop cellular respiration which contributes to the proton motive force. If this proton motive force is already high, proteorhodopsin may be unable to pump protons according to Walter et al.10

Results of the testing show that the culture expressing the Voltage protein dropped pH significantly more than the control culture, see Figure 3. This demonstrates that the rhodopsin was pumping protons out of the E. coli in response to light. Interestingly, the addition of the respiration inhibitor, silver nitrate, did not seem to affect the pH drop. With the conclusion of this test, we confirmed that the rhodopsin domain is functional.

Unipolar Localization

Figure 4: Image of Voltage protein at 100x under fluorescent microscope. A slight unipolar focus forms in one E. coli cell.

To observe the localization of the Voltage protein, we used observed the E. coli transformed with the plasmid containing the Voltage protein with a fluorescence equipped microscope. After protein initial protein induction and observation, we were able to observe fluorescence; however, we were only able to observe slight polar localization at best, see Figure 4. We tried to vary the induction time, observing cultures after 30, 60, 120, and 180 minutes of induction by 0.2% arabinose, and we tried to vary the the amount of arabinose added from 0.01%, 0.1% and 0.2%, observing after 4 hours. Unfortunately, none of these attempts were successful in getting the Voltage protein to completely unipolarly localize. In future efforts, it may be useful to try different linkers between the rhodopsin and the cPT region.

Cloning

The entire Voltage gene construct was synthesized in two parts by IDT gBlocks with a 20nt overlap between the first and second fragment. Upon receipt of the of the fragments, they were each gel purified and re-amplified by PCR. This PCR product was then gel purified again yielding extremely pure gene fragments. These fragments were then spliced together by overlap extension yielding one linear piece of DNA which contained the entire construct.

To clone this construct into our backbone, pSB1C3, we PCR amplified and gel purified the backbone with ~20nt inserts into the construct. Then, the full construct was inserted into the backbone by Gibson Assembly. The Gibson product was then transformed into NEB5α cells. 54 of the resulting colonies were screened with colony PCR using iGEM primers VF2 and VR. Two of these colonies contained inserts of the expected length, and these colonies were grown overnight in liquid culture to be mini-prepped and sequenced. Sequencing confirmed that the insert was successfully cloned into the backbone, but that there was a 20nt gap in the cPT region.

To fix this gap, the plasmid was PCR amplified and gel purified(?) with primers containing the missing region in overhangs. The resulting linear piece of DNA was reannealed via blunt end ligation. This product was then transformed into NEB5α cells. Colonies containing inserts of the right length were selected via colony PCR and grown overnight in liquid culture to be mini-prepped and sequenced. Sequencing results confirmed that the missing segment had been reinserted in the correct location yielding a complete, sequence confirmed construct in the pSB1C3 backbone.

Conclusion

In conclusion, we attempted to engineer a protein which both unipolarly localizes and pumps protons in E. coli. In this way, each E. coli cell will operate essentially like a single battery, generating a voltage along its length in response to light. By orienting an entire culture of E. coli such that the unipolar localization of the Voltage protein points in the same direction for all cells, the voltages would be additive like batteries in series generating a measurable voltage drop. In the end, we were able to demonstrate that, in the fusion protein, the rhodopsin maintained functionality, but we were unable to demonstrate complete, or replicable unipolar localization.

References

1. Hou, Sing-Yi, et al. “Diversity of Chlamydomonas Channelrhodopsins.” Photochemistry and Photobiology, vol. 88, no. 1, 2011, pp. 119–128., doi:10.1111/j.1751-1097.2011.01027.x.
   
2. Yamashita, Hayato, et al. “Role of trimer–trimer interaction of bacteriorhodopsin studied by optical spectroscopy and high-Speed atomic force microscopy.” Journal of Structural Biology, vol. 184, no. 1, 2013, pp. 2–11., doi:10.1016/j.jsb.2013.02.011.
   
3. Bamann, Christian, et al. “Proteorhodopsin.” Biochimica et Biophysica Acta (BBA) - Bioenergetics, vol. 1837, no. 5, 2014, pp. 614–625., doi:10.1016/j.bbabio.2013.09.010.
   
4. Hussain, Sunyia, et al. “Functional Consequences of the Oligomeric Assembly of Proteorhodopsin.” Journal of Molecular Biology, vol. 427, no. 6, 2015, pp. 1278–1290., doi:10.1016/j.jmb.2015.01.004.
   
5. “Part:BBa_K773002.” Part:BBa K773002, parts.igem.org/Part:BBa_K773002.
   
6. Stone, Katherine M., et al. “Structural Insight into Proteorhodopsin Oligomers.” Biophysical Journal, vol. 104, no. 2, 2013, pp. 472–481., doi:10.1016/j.bpj.2012.11.3831.
   
7. Doyle, Matthew Thomas, et al. “A small conserved motif supports polarity augmentation of Shigella flexneri IcsA.” Microbiology, vol. 161, no. 11, Jan. 2015, pp. 2087–2097., doi:10.1099/mic.0.000165.
   
8. “ProteinPaintbox®.” ATUM, www.atum.bio/products/protein-paintbox#2.
   
9. Sineshchekov, Oleg A., and John L. Spudich. “Light-Induced intramolecular charge movements in microbial rhodopsins in intact E. coli cells.” Photochemical & Photobiological Sciences, vol. 3, no. 6, 2004, p. 548., doi:10.1039/b316207a.
10. Walter, J. M., et al. “Light-Powering Escherichia coli with proteorhodopsin.” Proceedings of the National Academy of Sciences, vol. 104, no. 7, Feb. 2007, pp. 2408–2412., doi:10.1073/pnas.0611035104.