Project Results

Construction of Cytb₅₆₂

Cytochrome b₅₆₂ is normally found in E.coli and forms part of the respiratory chain, where it is involved with electron transport. We obtained DNA encoding for cytochrome b₅₆₂from MG1655 using colony PCR. The PCR product was run on a gel to ensure that we had the correct length, which is 318 bp. The product was PCR purified and digested with HpaI and SalI, and was put into the pVS72 backbone to be used for functional validation. In parallel, DNA containing BioBrick-optimized sequence coding for cytochrome b₅₆₂ was ordered from IDT, and Gibson assembly was used to insert it into the pSB1C3 backbone. This pSB1C3 plasmid was obtained from the iGEM 2015 distribution kit in a construct that contained the RFP coding device (BBa_J04450). The plasmid was digested with restriction enzymes PstI and EcoRI to cut out the RFP gene, which normally makes the resulting colonies pink. The transformed cells were plated and colonies that appeared white were selected for plasmid miniprep. A diagnostic digest was used to check whether the Gibson assembly had worked. The products were run on a 0.8% agarose gel to confirm this.

Construction of Sup35-NM

The protein responsible for the structure of our nanowires, Sup35NM, originates from a yeast prion gene in S. cerevisiae (baker’s yeast). The DNA sequence, encoding for Sup35NM and optimized for the BioBrick standard, was ordered from IDT. Once we obtained the sequence, we used Gibson assembly to insert it into the pSB1C3 backbone. The same procedure was used to select for the successfully transformed cells, where we took the white colonies and used miniprep kits to obtain the purified plasmid. To check that we had the correct plasmid, we performed a diagnostic digest. The resulting fragments were run on an agarose gel and we confirmed the insert band size was consistent with the length we expected.

Construction of Sup35-NM fusion protein BioBricks

This part consists of Sup35-NM and CsgAss for exporting and self-assembling amyloid fibres. Together, these genes have the capacity of being exported into the extracellular space due to the CsgA signal sequence.

We have created two BioBricks that encode for fusion protein containing Sup35NM sequence. One BioBrick part consists of Sup35NM and CsgAss for exporting and self-assembling of amyloid fibres containing Sup35 monomeric units, the other BioBrick is made up of coding regions for the bipartite CgAss signal sequence, Sup35-NM, and cytochrome b₅₆₂.

We constructed the BioBricks by ligating the plasmid pSB1C3 with a fragment containing the Sup35NM with CsgAss through conventional cloning. The plasmid was derived from the iGEM standard parts distribution kit, which contained RFP in pSB1C3 backbone (red fluorescent protein BBa_J04450). We digested the RFP coding device plasmid with restriction enzymes EcoRI and PstI according to standard protocol. The insert fragments were synthetized by IDT. The ligation product were subsequently transformed into Top 10 cells. The advantage of the RFP-containing plasmid was that successful clones were easily identifiable. Clones that were potentially successful resulted in white colonies, while the unsuccessful ones, which still contained RFP, were pink. This enabled us to pick the white colonies, prepare cells in liquid cultures in LB broth, which were used to for plasmid mini-preps . Correct plasmid samples were confirmed by re-digestion with EcoRI and PstI, and run on an 0,8% agarose gel, to confirm whether the plasmids contained the desired fragments judged by band sizes.
The expected bands were approximately 1020bp and 2000bp for the BioBrick containing Sup35-NM and CsgAss as seen in Figure 1. While for the BioBrick containing Sup35-NM, CsgAss and cytochromeb₅₆₂ the bands were expected to be 1338bp and 2000bp as seen in Figure 2. The functions of these two fusion proteins encoded by our BioBricks were validated and confirmed by Congo red plate assays, AFM imaging, and conductivity measurements.

Figure 1: Agarose gel of the restriction digest of BBa_K1739002 in pSB1C3 plasmid backbone with EcoRI and PstI.

Figure 2: Agarose gel of the digest of BBa_K1739003 in pSCB13 plasmid backbone with EcoRI and PstI.

Validation

CONGO red

Congo red is a diazo dye, used to show the presence of our nanowires as it binds to amyloid. We grew VS45 cells that had been transformed with our fusion protein biobricks in pVS72 backbone . We included arabinose and IPTG in the media, as this induces amyloid formation in VS45. This plasmid contained the same genes as our biobrick and therefore gave a realistic representation of what we would expect. The cells expressing our fusion proteins self-assembled into amyloid form bright red colonies after a couple of day’s incubation. This shows that the fibres take around two days to grow. (picture here)
We plated our negative control cells (VS45 transformed with pVS105, wich contains sup35M that do not self-assemble into amyloid nano-wires) on Congo red plates, alongside VS45 cells expressing EviroWire. Our results indicate a clear visual validation of amyloid formation for our fusion proteins as the the colonies were more red than the negative control cells.

AFM

Figure 3: An AFM image illustrating the production of nano-wires by our cells containing CsgAss and Sup35NM

Figure 4: Shows AFM imaging of our Envirowire plasmid. The nano-wires are visible although the reduced assembly efficiency of our designed amyloid nano-wire may be due to the incomplete folding of cytochrome b₅₆₂ as the heme needed for complete folding wasn't present in the growth media.

Figure 5: Illustrates a cluster of amyloid fibers, it is zoomed in from figure 4.

Figure 6: Shows AFM imaging of our negative control. No nano-wires are present

AFM is a form of scanning probe microscopy, which can be used to image biological specimen to high resolution and magnification. AFM was used to image cells expressing amyloid nano-wires. Colonies taken from agar plates were suspended and then placed onto a sample stage from which the scanning probe could operate, as described in the protocols section.
The first samples imaged (see Figure 3) were of the E. coliVS45 cells expressing CsgAss-SUP35NM protein, which had been induced to express and export the Sup35NM protein capable of forming amyloid fibrils. The cells exhibited long, distinct fibrils, which had a tendency to overlap and form a larger mass once they had detached from the cell. The nano-wires that assembled were around 20Å wide and varied in length. This shows that the protein could be exported and form part of a longer nano-wire.
The second samples we imaged (see Figure 4 and Figure 5) were of the same E. coli VS45 strain but were expressing a fusion protein containing CsgAss-SUP35NM fused with cytochrome b₅₆₂ in the C-terminus (CsgAss-SUP35NM-b₅₆₂). When these cells were induced to export the SUP35NM-b₅₆₂ fusion protein, the fibrils appeared to break off and form smaller curve-linear fibrillar aggregates much more frequently. In addition, it seems the protein assembles frequently as a larger cluster of amyloid rather than distinct fibrils seen in the sample with no cytochrome. Thus, we confirmed that the fusion protein was successfully exported and assembled, but our results show the fusion protein assembled less efficiently compared with their counterparts without the Cytochrome b₅₆₂.

Conductivity Measurements

To test conductivity of our wires, we inoculated agar with VS45 that expresses our EnviroWire fusion protein cyt b₅₆₂. The agar contained hemin, which is needed for cyt b₅₆₂ to fold properly outside the cell as it is exported in an unfolded form.
We used a two probes connected to a multimeter to measure resistance across 1 inch of the inoculated agar (see diagram in Figure 7). This was done for five consecutive days, due to fibrils forming at a much slower rate than cell growth. For the control sample, we used agar inoculated with VS45 cells that lack any amyloid-coding plasmid. The conductivity of cells expressing only Sup35NM without the cytochrome was also tested.
We measured resistance of our main sample, which was inoculated with VS45 that had been transformed with a plasmid containing our fusion protein seqence (CsgAss-Sup35NM-cyt b₅₆₂). The heme in the media is responsible for giving cyt b₅₆₂ its conductive properties, therefore we expected the resistance to be lower than our negative control samples. Conductivity was measured over a five-day period to see how the growth of amyloid fibres would be affected.
(See Table 1 for all the measurements and Figure 8 for the graph representation). Our readings were very erratic, alternating between approximately 13kΩ and 950kΩ. We probed the inoculated agar plates in five different positions each day, to see if there were spatial discrepancies in conductivity. The measurements varied significantly depending on where the gel was probed.
The results of this experiment are inconclusive, but in spite of the low levels of conductivity obtained from our amyloid-forming cells, there is evidence in the literature (ref) that cyt b₅₆₂ bound to similar fibres is able to produce intermittent conductive nano-wires. Graph 1 shows an increase in resistance over time, which can be expected as the gels progressively dry. When designing the envirowire biobrick, our research indicated that the distance between each cytochrome was within the correct range for electrons to travel. It may be that as the fibres collated, forming clumps rather than long fibres, only a small percentage of the cytochromes were exposed to the electrons flowing across. This could be tackled by purifying the product in the presence of heme, so that we have only the fibres and not the cells, which likely increase the resistance. This could then be laid out in the desired manner and the conductivity measured.
From the design aspect, we could potentially use sup35N, removing the M domain, and keep the cytochrome attached. Also, we could add the heme earlier on, to improve chances of cytochrome folding properly and thus increase conductivity. Improvements to be made when doing future tests include measuring resistance across different distances, as well as using a more specific set up. Conductivity could also be measured using STM, so that we can see the efficiency of electron transfer between individual nano-wire.

Figure 7: This circuit diagram is a representation of our conductive validation experiment.

Figure 8: Graph showing the average log resistance of our 5 different heme inoculated samples over a time period of 5 days. There are samples containing plasmid with fusion protein CsgA_ss - Sup35 NM with cytb₅₆₂ and 4 other control samples over the period of 5 days. For each day, several repeat resistance measurements were taken (maximum 5). Average electrical resistances were calculated for each measurement.

Table showing conductivity measurements over a 5 day period for 5 different plates

Table 1: The average resistance calculated for our main heme inoculated cell samples containing plasmid with fusion protein CsgA_ss - Sup35 NM with cytb₅₆₂ and 4 other control samples over the period of 5 days. For each day, several repeat resistance measurements were taken (maximum 5). Average electrical resistances were calculated for each measurement. The values above are in 1 decimal places. Note: For some of the days, not enough measurements were taken to account into the average calculations.

The Future

While carrying out the range of experiments and protocols to produce our four BioBricks we were presented with a variety of obstacles. The images produced from the atomic force microscopy (AFM) of VS45 expressing EnviroWire fusion protein (CsgAss –Sup35NM-Cytb₅₆₂) showed few distinct amyloid fibrils, and instead it appeared that our fusion protein was forming aggregates. We hypothesize that this is due to cytochromeb₅₆₂ being exported in an unfolded state and blocking the stacking interaction required for amyloid fibril formation. Proper folding occurs exogenously and requires our bacteria to be grown in heme-containing medium. We postulate that growing our bacteria in a medium containing heme before the imaging would allow cytochrome folding, and thus amyloid fibre formation, providing a more complete image of our fusion protein in its intended state.

Our BioBrick encoding the bipartite CsgA signal sequence and Sup35NM protein can be modified into a fusion protein by future iGEM teams. This is possible because Sup35 in its native state has two functional domains, the N and C terminal, separated by the M domain (Frederick et al., 2014; Glover et al. 1997). This facilitates the addition of a new functional domain at the C terminus of Sup35-NM to form a fusion protein with the potential to perform a diverse range of functions that harness self-assembling amyloid fibres.

One of the future applications of our BioBrick would be to produce a biofilm on a rigid surface on which the amyloid fibres would grow. The E.coli could then be removed from the surface and only the fibres would be left to conduct electricity.
Another of the future results that we hope will be achieved is to use our nanowires to conduct electricity which would be produced by the bacteria themselves, by transferring electrons from the bacterial electron transport chain to the amyloid fibre. In this way it could function as a battery.

References

Fujiwara, T., Fnkumori,, Y. and Yamanaka, T. (1993). Halobacterium halobium Cytochrome b-558 and Cytochrome b-562: Purification and Some Properties. J. Biochem., 113, pp.48-54.

[1] Glover, J., Kowal, A., Schirmer, E., Patino, M., Liu, J. and Lindquist, S. (1997). Self-Seeded Fibers Formed by Sup35, the Protein Determinant of [PSI+], a Heritable Prion-like Factor of S. cerevisiae. Cell, 89(5), pp.811-819.

[2] Mathews, F. S., Bethge, P. H., & Czerwinski, E. W. (1979). The structure of cytochrome b₅₆₂ from Escherichia coli at 2.5 A resolution. Journal of Biological Chemistry, 254(5), 1699-1706.

[3] Robinson, C., Liu, Y., Thomson, J., Sturtevant, J. and Sligar, S. (1997). Energetics of Heme Binding to Native and Denatured States of Cytochrome b 562 †. Biochemistry, 36(51), pp.16141-16146.

[4] Sivanathan, V. and Hochschild, A. (2012). Generating extracellular amyloid aggregates using E. coli cells. Genes & Development, 26(23), pp.2659-2667.

[5] Sivanathan, V. and Hochschild, A. (2013). A bacterial export system for generating extracellular amyloid aggregates. Nat Protoc, 8(7), pp.1381-1390.

[6] Tessier, P. and Lindquist, S. (2009). Unraveling infectious structures, strain variants and species barriers for the yeast prion [PSI+]. Nat Struct Mol Biol, 16(6), pp.598-605.

[7] Wickner, R., Edskes, H., Shewmaker, F. and Nakayashiki, T. (2007). Prions of fungi: inherited structures and biological roles. Nature Reviews Microbiology, 5(8), pp.611-618.

[8] Xavier, A., Czerwinski, E., Bethge, P. and Mathews, F. (1978). Identification of the haem ligands of cytochrome b₅₆₂ by X-ray and NMR methods. Nature, 275(5677), pp.245-247.

[9]Frederick, K., Debelouchina, G., Kayatekin, C., Dorminy, T., Jacavone, A., Griffin, R. and Lindquist, S. (2014). Distinct Prion Strains Are Defined by Amyloid Core Structure and Chaperone Binding Site Dynamics. Chemistry & Biology, 21(2), pp.295-305.