Software error:

IPTG Inducible Bio-IcsA

Sequence and Features

Assembly Compatibility:

• 10
  COMPATIBLE WITH RFC[10]
• 12
  COMPATIBLE WITH RFC[12]
• 21
  INCOMPATIBLE WITH RFC[21]

  Illegal BamHI site found at 2278
  
• 23
  COMPATIBLE WITH RFC[23]
• 25
  INCOMPATIBLE WITH RFC[25]

  Illegal AgeI site found at 649
  Illegal AgeI site found at 1126
  
• 1000
  INCOMPATIBLE WITH RFC[1000]

  Illegal BsaI site found at 2814
  

Introduction

Figure 1: Schematic of the magnetic orientation of E. coli in the microfluidic device. Schematic shows magnetic beads connected to the bacteria through a biotin-streptavidin bond, and an applied magnetic field to orient the bacteria.

Having unorganized cells unipolarly pumping ions is not sufficient to generate a voltage difference. Instead, we needed to come up with a way to orient the E. coli along their old poles and stack them in a microfluidic device, much like the way you stack batteries in a flashlight. Diverse alternatives were explored in order to induce orientation of the cells in response to an external stimulus. We first considered and tested a chemotactic approach, making the E. coli swim toward a chemoattractant to orient them along their old pole. However, due to the intrinsic, random nature of E. coli’s general swimming patterns and and dynamics, any methods relying on the bacteria’s natural mode of locomotion in response to a stimulus were not feasible options for our intended outcome. In fact, many genera of motile bacteria, including E. coli, engage in a behavior called swarming when surrounded by a dense colony in a bulk fluid (Swiecicki et al.). During swarming, E. coli continually reorient themselves by random jostling with their neighbors, randomizing their directions within a few tenths of a second: for example, the cells may move laterally if they collide with adjacent cells or may reverse their direction abruptly as the motors that actuate the flagella change direction (Darnton et al.). Overall, the cells tend to separate from each other principally due to their difference in speeds or rotational diffusion.

Thus, we resorted to a mechanical solution in order to impose the desired highly-organized and super-oriented alignment. Specifically, we aimed to functionalize the cells with superparamagnetic microbeads and employ an applied, external magnetic field to induce the mechanical rotation of the cells towards the magnet, as determined by the magnetic force on the beads. Superparamagnetic microbeads were selected for this application due to their ability to increase in magnetization when a magnetic field is applied, and lose their magnetic properties upon removal of the field (Akbarzadeh et al.). This implies that no residual magnetization is observed, eliminating the potential risk of the beads interfering with the cells’ biological functions and behavior. From a literature review, we were able to determine that previous studies had shown the successful integration of a biotinylation site within the coding sequence of IcsA without disrupting or interfering with its function (May et al.). Through the integration of a biotinylation site on the IcsA aa 87, we were able to biotinylate the cells and subsequently enable the formation of biotin-streptavidin bonds with streptavinated paramagnetic microbeads in solution. The biotin-streptavidin bond, is one of the strongest found in nature, thus further confirming the stability and reliability found in the choice of this induced orientation method.

Experimental Design

Chemotaxis

Figure 2: Image of microfluidic device, empty and then filled with E. coli, taken with the Zeiss microscope at 20X.

Media:Orientation_figure3.mp4

The first method of orienting the E. coli we pursued was chemotaxis. At first we were discouraged because E. coli are peritrichous bacteria, meaning their cell surface is covered in flagella (Pratt et al.). This subsequently means that if we induced the cells to swim in one direction, there would be no guarantee with which side they would swim. Because our voltage construct theoretically localizes to the old pole of E. coli, having random directional swimming would mean that the ions would be pumped in a random direction relative to the microfluidic device. However, we then found a paper that found that 66% of the time, E. coli swim with their old poles behind them (Ping et al.). Although not completely polar, the skew of E. coli’s swimming pattern toward swimming with their old pole behind them was promising. We then conducted a chemotaxis experiment where we incubated E. coli strain W in M9 minimal media (“M9 Minimal Media Recipe”). The strain was selected for its flagella, as the normal lab strain NEB5ɑ does not have flagella, and M9 minimal media was used because it has less nutrients than LB, meaning the E. coli would theoretically swim toward to the more nutrient-dense media. We then pipetted approximately 15 μl of E. coli W in M9 minimal media into the inlet well of our microfluidic device, and 10 μl of LB, or lysogeny broth, into the outlet well. We then took the video displayed in link above. Unfortunately, we observed random motion of the E. coli, not directed movement toward the chemoattractant of LB. We therefore moved onto our second idea for orientation, mechanical orientation.

Construct Design

In order to properly orient our E. coli, we needed to find a unipolarly localizing protein that could be mechanically manipulated to orient the E. coli. We found this solution in the protein IcsA, a protein that is native to Shigella flexneri and responsible for unidirectional actin polymerization to enable cellular motility (Goldberg et al., Bernardini et al.). IcsA always localizes to the old pole of the cell, and has been expressed in E. coli and shown to polarly localize in E. coli strains with a complete lipopolysaccharide in the outer membrane (Robbins et al.). The IcsA protein additionally consists of two parts; the passenger domain, or the region responsible for unipolar localization, and the beta barrel, the region responsible for actin polymerization in the endogenous protein (May et al.). IcsA has also been shown to be able to host a biotinylation site within its coding sequence without changing the functionality of the protein. This further makes IcsA an ideal candidate for the orientation construct because if the protein is biotinylated and localizes the old pole on the outer membrane, cells could be washed with streptavidin coated magnetic beads, allowing for a biotin-streptavidin bond between the cell and the magnetic bead. The E. coli would then be able to be oriented by applying a magnetic field in liquid culture.

Figure 3: Structure of IcsA, actin polymerization protein native to Shigella flexneri. Inflated image shows a closer examination of the cPT region of IcsA (Doyle et al.).

In our initial design of the orientation construct, we designed the coding sequence to be a conjugation between the IcsA passenger domain and GFP using 5X poly-glycine linker. A biotinylation site was encoded at amino acid 87, which has been shown to not disrupt the functionality of IcsA (May et al.). We used the IPTG inducible promoter from the Parts Registry, BBa_R0011 <https://parts.igem.org/Part:BBa_R0011>, to control the expression of our construct. We used the ribosome binding site from the Parts Registry, BBa_B0034 <https://parts.igem.org/Part:BBa_B0034>, to allow for translation of the protein. After the coding sequence of the protein, we included two terminators, BBa_B0010 <https://parts.igem.org/Part:BBa_B0010> and BBa_B0012 <https://parts.igem.org/Part:BBa_B0012>.

However, upon further literature research, we found that when the IcsA passenger domain has been conjugated to GFP, the protein no longer localizes to the outer membrane of E. coli, but instead to the intracellular space on the old pole of the cell (Charles et al.). To amend this flaw in the construct design and ensure unipolar localization to the outer membrane of the old pole of E. coli, we then planned to take out the GFP sequence and clone the beta barrel sequence back into the plasmid, as the full sequence of IcsA has been shown to localize to the old pole of E. coli. The beta barrel sequence was then also ordered as a secondary addition to the main orientation construct.

We also designed a construct to enhance biotinylation of our orientation construct. The motivation behind creating a biotinylation construct was to ensure that the biotinylation site on the IcsA protein would be biotinylated, as it needs to be able to bind strongly to the streptavidin-coated magnetic beads. To accomplish optimal biotinylation of IcsA, we designed our BirA construct. BirA is the protein responsible for biotinylation endogenously in E. coli, and acts by associating with biotin to form the holobirA complex, which then associates with the apoBCCP complex to transfer biotin in conditions of high biotin demand (“Bifunctional ligase/Repressor BirA”). For our biotinylation construct, we began with the Biobrick Prefix, followed by a rhamnose inducible promoter from the Parts Registry, BBa_K914003 <https://parts.igem.org/Part:BBa_K914003>. We next used the ribosome binding site from the Parts Registry, BBa_B0034 <https://parts.igem.org/Part:BBa_B0034>, to allow for translation of birA. The coding sequence for birA, obtained from the NCBI database, was codon optimized for E. coli and attached after the ribosome binding site, along with a 6X poly-His tag to allow for protein visualization. We then added the Parts Registry standard double terminator, BBa_B0015 <https://parts.igem.org/Part:BBa_B0015>, followed by the Biobrick suffix.

Cloning

Bio-IcsA-GFP was synthesized by IDT as a three part construct. We ran the IDT constructs 1-3 on a gel and gel extracted for purification. These pure constructs were then PCR amplified to ensure we had enough DNA for assembly. We amplified the linear backbone pSB1C3 in preparation for Gibson assembly, meaning that we PCR-amplified the backbone to have 20 nucleotide overlaps into Bio-IcsA-GFP constructs 1 and 3. The extended linear backbone was PCR amplified with these overlaps and gel extracted. We then performed a 4-piece Gibson assembly with an hour long incubation time, and immediately transformed into NEB5-ɑ chemically competent cells. Once colonies grew, we performed colony PCR, selected the most promising bands from the gel images, and sent the miniprepped plasmid for sequencing. Bio-IcsA-GFP from colony 50 was then confirmed via sequencing.

To take out GFP from the construct, we performed PCR amplification, which linearized the plasmid without GFP. We then gel extracted the resulting band. IcsA beta barrel was synthesized by IDT in one piece. We ran the beta barrel construct on a gel, and gel extracted for purity. We then PCR amplified the beta barrel in preparation for Gibson assembly, meaning that we extended the construct to have 20 nucleotide overlaps into the linearized Bio-IcsA plasmid. We then ran a 2-piece Gibson assembly with a 15 minute incubation time to insert the beta barrel into the plasmid. We immediately transformed into chemically competent NEB5-ɑ cells. Once colonies grew, we performed colony PCR to screen for promising colonies, and selected those with the correct band length. We then sent the miniprepped plasmid for sequencing, when Bio-IcsA-full 50-2 was confirmed via sequencing.

We encountered two major sequencing abnormalities while cloning Bio-IcsA-full. One was the absence of 23 nucleotides from the promoter. Given that the promoter is only 54 nucleotides long, we sought to amend the gap in our promoter before pursuing any functionality testing. To restore the promoter, we designed primers that had a gene specific sequence on either side of the missing sequence, and then extended 11-12 nucleotides to cover the missing sequence. After linearizing the plasmid with these primers and gel extracted the result, we then phosphorylated the 5’ end of the linearized plasmid using NEB T4 Polynucleotide Kinase in preparation for blunt-end ligation. The blunt-end ligation was performed using NEB T4 Ligase in NEB T4 DNA Ligase Reaction Buffer at 16℃ overnight. The second sequencing abnormality we encountered was a point mutation within the IcsA passenger domain, which resulted in a stop codon in the reading frame. To solve this problem, we performed the Quikchange protocol, where we designed primers to include the desired base pair change (“Quikchange II”). We then ran 12 cycles of PCR using the designed primers, digested the methylated DNA with NEB DpnI, and transformed into chemically competent NEB5-ɑ cells. Once colonies grew, we performed colony PCR to screen for promising colonies, and selected those with the correct band length. We then sent the miniprepped plasmid to confirm via sequencing. After confirming that the plasmid had the correct sequence, we transformed it into E. coli strain W to ensure that IcsA would polarly localize.

Our second cloning project for the orientation construct involved assembling our biotinylation construct, birA. We synthesized the birA construct through IDT in one piece. Because the construct was in one piece, we chose to do a digestion/ligation into the pSB1C3 backbone instead of Gibson assembly. We digested using NEB EcoRI-HF enzyme and NEB PstI-HF enzyme in Cutsmart-HF buffer, and then ligated using NEB T4 DNA ligase in T4 DNA Ligase Reaction Buffer into the backbone pSB1C3. We then transformed into chemically competent NEB5-ɑ cells. Once colonies grew, we performed colony PCR to screen for promising colonies, and selected those with the correct band length. We then sent the miniprepped plasmid for sequencing, and confirmed that BirA 8 had the correct sequence.

Polar localization

Figure 5: Shows fluorescent tagged-IcsA expressed in E. coli NEB5a, a strain without a complete LPS. The image was taken at 40X in an Evos digital microscope.

Figure 6: Shows fluorescent tagged-IcsA expressed in E. coli strain W, a strain with a complete LPS, where in polar localization of IcsA can be visualized. The image was taken at 40X in an Evos digital microscope.

To test the polar localization of our biotinylated IcsA, we used a fluorescent streptavidin dye to bind to the outer membrane of our E. coli. We chose this method of verifying polar localization because fusing IcsA with a fluorescent protein causes it to localize intracellularly as opposed to extracellularly. With a generous gift from Thermo Fisher Scientific, we used Thermo Fisher Scientific’s Streptavidin, Alexa Fluor 405 conjugate to visualize the polar localization of IcsA (“Streptavidin, Alexa Fluor”). For this protocol, we grew up cultures of both NEB5ɑ and strain W of E. coli to compare how IcsA localized in strains with or without an LPS. The cultures were induced with IPTG at 0.2 mM for 5 hours. They were then spun down and resuspended in DPBS twice, and then incubated at 4 ℃ on a rotator for 30 minutes. The samples were then visualized at 20X and 40X under an Evos microscope. As expected, there was significantly more protein visualization and polar localization in strain W compared to NEB5ɑ, as shown in Figure 5, 6. For future experimentation to maximize the number of cells expressing IcsA and polar localization, the amount of fluorescent streptavidin dye could be varied, the level of IPTG induction could be manipulated, and the time given for the protein to fold could be studied.

Magnetic bead attachment

Figure 7: Visualization of magnetic bead attachment in magnetic separation rack, where pellet of cells and magnetic beads can be seen.

Figure 8: One sample clearly showing the pellet of magnetic beads and cells.

Media: Orientation_figure9.mp4

The second assay we used to validate the functionality of our IcsA construct was magnetic bead attachment. To do so, we purchased MojoSort streptavidin coated magnetic beads with a diameter of 30 nm that would be attached to our E. coli to orient them under a magnetic field (“MojoSort™ Streptavidin Nanobeads”). The small size of the magnetic beads was chosen specifically to counteract diffusion in the biobactery setup, as we wanted E. coli packed tightly together to promote unidirectional ion flow. We grew up liquid cultures of our IcsA-expressing E. coli overnight in both NEB and W strains, and measured the OD600 to determine the volume needed for 10^9 cells. The cells were then washed twice with DPBS + 0.1% Tween 20, and then incubated with 5% or 10% of the magnetic bead solution to facilitate binding. After incubation on ice for 30 minutes, the cells with magnets were spun down and resuspended in the DPBS-Tween 20 solution. They were then placed in NEB magnetic separation rack for 5 minutes. At this step, we could see the magnetic beads and a small cell pellet separate out from solution closest to the magnet on the rack (Figure 5). The supernatant was taken out of the tube, and the magnet and cell pellet was then resuspended in the DPBS solution. Both the supernatant and the pellet were spun down and resuspended in a final volume of 100 μl of DPBS. Both solutions were prepared for microscopy. Because the beads are too small to visualize under the Evos microscope, we placed a strong magnet found in the teaching lab at the top of the slide to induce orientation of our E. coli, using their movement as the indicator for the magnetic bead attachment. We took the video in displayed in the link above as evidence of the magnetic bead attachment, where we saw clear taxis of the E. coli toward the strong magnet. With these results, we have proven that we are able to attach magnetic beads to our E. coli via a biotin-streptavidin bond between IcsA and streptavidin covered magnetic beads.

Conclusion

With both the successful polar localization and magnetic bead attachment, we have proved that we can orient our IcsA-expressing E. coli along their old poles using a magnetic, induced orientation method. We proved that IcsA is expressed on the outer membrane, predominantly on the old pole of full LPS-containing E. coli, as well as the ability to conjugate magnetic beads to IcsA via a biotin-streptavidin bond. This novel method to orient E. coli along their old poles not only applies to our biological battery, but could also be used for other engineering applications, such as biological manufacturing. For further experimentation, more experiments could be done to increase polar localization of IcsA, particularly with optimizing the protocol with the fluorescent streptavidin dye. Furthermore, visualizing the E. coli conjugated to the nanoscale magnetic beads in the microfluidic device, and ensuring that we could pull them into the channel using a strong magnet would be promising for being able to create a biological battery. Together with the polarly localized rhodopsin construct, we are excited about the prospect of creating an E. coli-powered battery, given that we have proved our ability to orient our E. coli with a magnet.

Methods

All PCR reactions for cloning were done using NEB Q5 or NEB OneTaq, with the protocol provided on the NEB website with annealing temperatures determined by the NEB Tm calculator (“Q5® High-Fidelity DNA Polymerase,” “OneTaq® DNA Polymerase”). All gels were run at 100V for 50 minutes using a 1% agarose gel. To gel extract or PCR cleanup, we used the Zymogen DNA Clean & Concentrator kit and the Zymoclean Gel DNA Recovery Kit (“DNA Clean & Concentrator™-5,” “Zymoclean™ Gel DNA Recovery Kit”). For DNA assembly we used NEB EcoRI - HF and NEB PstI - HF for digestion and T4 DNA ligase for ligation, and the NEB Gibson Assembly Master Mix or GeneArt Seamless Cloning and Assembly (“Gibson Assembly® Master Mix,” “GeneArt® Seamless Cloning & Assembly”). To isolate plasmid DNA, we used the Zyppy Plasmid Miniprep Kit (“Zyppy™ Plasmid Miniprep Kit”). We used Elim Biopharmaceuticals for our sequencing needs (“ELIM BIOPHARM”).

We used AlexaFluor 405 streptavidin conjugate for the visualization of our biotinylated IcsA construct (“Streptavidin, Alexa Fluor 405 conjugate”), a generous gift from Thermo Fisher Scientific. We used streptavidin coated magnetic beads from BioLegend to validate the magnetic activity of the E. coli (“MojoSort™ Streptavidin Nanobeads”). We used the Zeiss AxioImager Z1 light microscope and the Evos XL Digital Inverted Microscope for microscopy (“Axio Imager 1 for Life Science Research,” “EVOS® Digital Microscopes”).

The microchannel devices were designed in SolidWorks. The design (see Figure 1) consisted of two wells (inlet well and outlet well) of diameter 2mm for the placement of probes to measure electric potential and for the insertion of E. coli. The two wells were connected by a microchannel (width 150 um). The devices were prototyped with a biocompatible silicone elastomer, PDMS (Sylgard 184, Dow Corning, Midland MI). The channel structure of the chips was formed by soft lithography: a negative master mold for the the channels was fabricated with a UV-curable epoxy (SU8 by MicroChem, Newton MA) by conventional contact lithography. Liquid PDMS pre-polymer, in a 10:1 ratio of catalyst and resin, was vigorously mixed, degassed in a vacuum chamber and poured onto the mold to a thickness of about 2 -3mm and cured in an 80°C oven for 90 minutes. Following the curing step, the elastomer was peeled off and cut into individual chips. The holes at the end of each channel well punched using a 2mm luer stub. The chips were then baked for 30 more minutes at 150°C to increase hardness and Young’s modulus of the PDMS, to ensure that the micrometer-scale features were more stable against spontaneous collapse. The chips were then sealed to a #4 microscope cover glass by plasma wand high radio-frequency bonding. Before use, the channels were primed to increase hydrophilicity and reduce surface tension. The priming involved overnight incubation in a solution of 17:2:1 nuclease-free diH2O : Tween-20 (0.1%) : BSA.

References

1. Akbarzadeh, Abolfazl, et al. “Magnetic Nanoparticles: Preparation, Physical Properties, and Applications in Biomedicine.” Nanoscale Research Letters, vol. 7, 2012, p. 144. PMC.
2. Bernardini, M. L., et al. “Identification of IcsA, a plasmid locus of Shigella flexneri that governs bacterial intra- and intercellular spread through interaction with F-Actin.” Proceedings of the National Academy of Sciences, vol. 86, no. 10, Jan. 1989, pp. 3867–3871., doi:10.1073/pnas.86.10.3867.
3. Biolabs, New England. “Gibson Assembly® Master Mix.” New England Biolabs: Reagents for the Life Sciences Industry, www.neb.com/products/e2611-gibson-assembly-master-mix#Product%20Information.
4. Biolabs, New England. “OneTaq® DNA Polymerase.” New England Biolabs: Reagents for the Life Sciences Industry, www.neb.com/products/m0480-onetaq-dna-polymerase.
5. Biolabs, New England. “Q5® High-Fidelity DNA Polymerase.” New England Biolabs: Reagents for the Life Sciences Industry, www.neb.com/products/m0491-q5-high-fidelity-dna-polymerase#Product%20Information
6. Charles, M., et al. “Polar targeting of Shigella virulence factor IcsA in Enterobacteriacae and Vibrio.” Proceedings of the National Academy of Sciences, vol. 98, no. 17, 2001, pp. 9871–9876., doi:10.1073/pnas.171310498.
7. Darnton, Nicholas C., et al. “Dynamics of Bacterial Swarming.” Biophysical Journal, vol. 98, no. 10, 2010, pp. 2082–2090., doi:10.1016/j.bpj.2010.01.053.
8. 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.
9. Goldberg, Marcia B., and J A. Theriot. “Shigella flexneri surface protein IcsA is sufficient to direct actin-Based motility.” Proceedings of the National Academy of Sciences, vol. 92, no. 14, Mar. 1995, pp. 6572–6576., doi:10.1073/pnas.92.14.6572.
10. May, Kerrie L., et al. “Self-Association of the Shigella flexneri IcsA autotransporter protein.” Microbiology, vol. 158, no. 7, 2012, pp. 1874–1883., doi:10.1099/mic.0.056465-0.
11. Ping, Liyan. “The Asymmetric Flagellar Distribution and Motility of Escherichia coli.” Journal of Molecular Biology, vol. 397, no. 4, 2010, pp. 906–916., doi:10.1016/j.jmb.2010.02.008.
12. Pratt, Leslie A., and Roberto Kolter. “Genetic analysis of Escherichia coli biofilm formation: roles of flagella, motility, chemotaxis and type I pili.” Molecular Microbiology, vol. 30, no. 2, 1998, pp. 285–293., doi:10.1046/j.1365-2958.1998.01061.x.
13. Robbins, Jennifer R., et al. “The making of a gradient: IcsA (VirG) polarity in Shigella flexneri.” Molecular Microbiology, vol. 41, no. 4, 2002, pp. 861–872., doi:10.1046/j.1365-2958.2001.02552.x.
14. Swiecicki, Jean-Marie, et al. “From swimming to swarming: Escherichia coli cell motility in two-Dimensions.” Integrative Biology, vol. 5, no. 12, 2013, p. 1490., doi:10.1039/c3ib40130h.
15. “Axio Imager 1 for Life Science Research.” Zeiss, www.zeiss.com/microscopy/us/products/light-microscopes/axio-imager-1-for-biology.html
16. “Bifunctional ligase/Repressor BirA.” Uniprot, 27 Sept. 2017, www.uniprot.org/uniprot/P06709.
17. “DNA Clean & Concentrator™-5.” Zymo Research, www.zymoresearch.com/dna/dna-clean-up/pcr-dna-clean-up-concentration/dna-clean-concentrator-5.
18. “ELIM BIOPHARM.” Elim Biopharmaceuticals, www.elimbio.com/.
19. “EVOS® Digital Microscopes.” Electron Microscopy Sciences, www.emsdiasum.com/microscopy/products/digital/evos.aspx.
20. “GeneArt® Seamless Cloning & Assembly.” Thermo Fisher Scientific, www.thermofisher.com/us/en/home/life-science/cloning/seamless-cloning-and-genetic-assembly/geneart-seamless-cloning-and-assembly.html.
21. “MojoSort™ Streptavidin Nanobeads.” BioLegend, Inc., www.biolegend.com/en-us/products/mojosort-streptavidin-nanobeads-11877.
22. “M9 Minimal Media Recipe (1000 mL).” The Lab Rat, www.thelabrat.com/protocols/m9minimal.shtml.
23. “Quikchange II.” Agilent, www.genomics.agilent.com/en/Site-Directed-Mutagenesis/QuikChange-Lightning/?cid=AG-PT-175&tabId=AG-PR-1162.
24. “Streptavidin, Alexa Fluor 405 conjugate.” Thermo Fisher Scientific, www.thermofisher.com/order/catalog/product/S32351.
25. “Zymoclean™ Gel DNA Recovery Kit.” Zymo Research, www.zymoresearch.com/dna/dna-clean-up/zymoclean-gel-dna-recovery-kit.
26. “Zyppy™ Plasmid Miniprep Kit.” Zymo Research, https://www.zymoresearch.com/dna/plasmid-dna-purification/zyppy-plasmid-miniprep-kit.