Measurement: Newcastle
This year we have been working alongside iGEM teams from across the globe.
We have been working closely with teams from Newcastle, Glasgow and Purdue to help each other improve our projects
from both in and outside the lab.
Part of the Newcastle iGEM team’s project this year
involved an experiment centred around the creation of biological electronic
components. Newcastle asked our team if we could help them out by finding the
thermal conductivity of different growth media. With the help of our biophysicist
supervisor Ryan Edgington, we came up with a plan to measure the conductivity.
Using the apparatus we had available, we discovered that the thermal conductivity of LB and M9 broth to be roughly
the same as water. The conductivity of water at room temperature is about 598.4 $\frac{mW}{Km}\text{ }$(mili
watt per metre kelvin).We found the conductivity of LB and M9 to be (605 $\pm$ 20) $\frac{mW}{Km}\text{ }$ and (570 $\pm$ 30) $\frac{mW}{Km}\text{ }$ respectively
You can read more about our method here.
Software:Purdue
Our team helped Purdue with this by logging data for the 260
iGEM teams of 2015 and critiquing ease of use and effectiveness of the database. For each team
we documented a summary of what their project was about, their track, number of team members, chassis,
research benchmarks, finished parts, the presence or absence of kill switches, medals and any awards
and nominations. We tagged the teams summaries with keywords to make finding a project much easier.
We gave Purdue feedback on the design, layout and how
easy this database was to use to help them improve on what they had done so far.
Parts:Glasgow
We collaborated with Glasgow iGEM 2016 to test the efficiency of the T7 Promoter we were using to construct the KillerRed, KillerOrange and Lysozyme kill switches. We new that it was leaky and we speculated that it was reducing the efficiency of our project but we needed proof that the leakiness of the promoter could affect our project. In return they gave us a DH5α.Z1 strain in the hopes it would improve the efficiency of our promoter. After subsequent testing we were unable to express our KillerRed and KillerOrange proteins in this strain. This is the report they sent us for the test of the T7 promoter:
Exeter and Glasgow iGEM 2016 Collaboration:
KillerRed and KillerOrange Promoter Efficiency Experiment
Methods
First, we transformed the plasmids for testing promoter efficiency into the E. coli strain DH5α.Z1:
- J04450 (RFP with a lac-repressible promoter) in pSB1C3
- lac-repressible promoter + KillerRed in pSB1C3
- lac-repressible promoter + KillerOrange in pSB1C3
Next, we set up 5ml LB broth overnight cultures in boiling tubes with loose caps at 37°C shaking at 225rpm of:
- DH5α, no plasmid
- DH5α, J04450-pSB1C3
- DH5α, KillerRed-pSB1C3
- DH5α, KillerOrange-pSB1C3
- DH5α.Z1, no plasmid
- DH5α.Z1, J04450-pSB1C3
- DH5α.Z1, KillerRed-pSB1C3
- DH5α.Z1, KillerOrange-pSB1C3
We set up each of these both with 1mM IPTG and without, so 16 overnights in total, with 25μg/ml chloramphenicol for the strains with plasmids.
The next day, we spun down 500μl of each overnight in 1.5ml eppendorfs, resuspended in 1ml PBS buffer (PBS gives less background fluorescence than LB broth), and transferred to cuvettes for OD600 measurements. For the fluorescence measurements, we used a Typhoon FLA 9500 with the samples in a 96-well plate. Each of the 16 samples were pipetted into 3 wells with 300μl each. The settings on the Typhoon were: 1) excitation laser at 532nm and emission filter above 575nm (so would detect any wavelength above 575nm) and 2) excitation laser at 473nm and emission filter above 575nm. Both were across the whole 96-well plate, but the second lower excitation wavelength was included in case the first was too high to excite KillerOrange.
Results
For each of the wells, a fluorescence value in arbitrary units (au) was calculated using ImageQuant software, where we set the 3 wells of PBS only blanks as 0% fluorescence, and the well with the highest value as 100% to convert au into percentage. Next, we normalised for any natural fluorescence in the cells by subtracting the average of the “cells only” wells (DH5α, and Dh5α.Z1, both with and without IPTG) from all wells, and then corrected for the difference in OD600 values between samples by dividing the normalised fluorescence measurements by the OD600 values. The table below shows the average across the 3 wells of each of the 16 samples.
Graph of these averages. The error bars are standard deviation but are very small because the 3 replicates for each sample are technical replicates, so do not show the variation that would be seen with biological replicates (3 different colonies for each of the 16 samples).
Fluorescence scan image from the Typhoon with labels for which samples are in each well.
These data indicate that there is no difference in fluorescence between either KillerRed or KillerOrange and the cells only control either with or without induction with IPTG. There could be several reasons for this, including the light was not intense enough to excite the fluorescent proteins, however no fluorescence from this type of the test with a laser for excitation would be unlikely. It is also possible that no protein is being produced, which could be due to insufficient IPTG. However, the RFP in J04450 under the control of the lac-repressible promoter R0010 clearly shows that in the DH5α.Z1 strain, there is less fluorescence without IPTG, than with IPTG. This is not a perfect control for the concentration of IPTG used unless KillerRed and KillerOrange also have the R0010 promoter. Interestingly, in the DH5α strain, there is no significant difference between RFP fluorescence with or without IPTG – this is due to DH5α not having a functional copy of LacI, the lac repressor, therefore lac-repressible promoters are not “OFF”, so cannot be switched “ON” by IPTG induction.
Sequencing
Another reason there may not be any KillerRed or KillerOrange protein produced, is mutations in the promoter. This was something we encountered when attempting to clone a promoter in front of the toxin from the toxin-antitoxin system we were working with. If a protein is toxic to produce, any cell which is producing less or no protein will grow faster than a cell which is producing the toxic protein. This means a mutated, non-functional promoter will have a proliferative advantage during transformation. So, as we were sending our BioBricks for registry for submission, we decided to sequence the minipreps of KillerRed and KillerOrange as well with the registry standard pSB1C3 sequencing primer VF2, to check for any mutations. The results are shown below in screenshots of a plasmid editor software called ApE.
Both plasmids had the BioBrick prefix, and the correct sequence for both KillerRed and KillerOrange open reading frame, according to the papers cited on the Exeter 2016 iGEM wiki. The sequence between the prefix and the ATG start codon, we checked against lac-repressible promoters in the iGEM registry. We found a match to R0184, which is a T7 lac-repressible promoter. T7 promoters require T7 polymerase to be transcribed, as they are not recognised by E. coli polymerases. This results confirms the result of the fluorescence measurements. No KillerRed or KillerOrange protein was observed by fluorescence, as neither gene was being transcribed by either DH5α or DH5α.Z1 as neither strain produces the required T7 polymerase. A protein overexpression E. coli strain such as BL21 which has the T7 polymerase gene inserted into its genome is designed to use T7 promoters would have been able to express these KillerRed and KillerOrange constructs.
Sequencing result for KillerRed:
ATAAAATAGGCGTATCACGAGGCAGAATTTCAGATAAAAAAAATCCTTAGCTTTCGCTAAGGATGATTTCTGGAATTCGCGGCCGCTTCTAGAGTACTTAATACGACTCACTATAGGGGAATTGTGAGCGGATAACAATTCCCCTCAAGAAATAATTTTGTTTAACTTTAAACCTAAAGAGGAGAAAAATGGGCAGTGAAGGTGGTCCTGCGCTTTTCCAGTCAGACATGACCTTCAAAATTTTCATTGACGGTGAAGTTAATGGACAGAAATTTACGATCGTAGCCGATGGCTCAAGCAAATTCCCACATGGGGACTTCAATGTCCACGCCGTGTGCGAAACAGGCAAATTACCCATGAGCTGGAAGCCGATTTGTCATTTGATTCAGTACGGGGAGCCTTTTTTCGCTCGTTACCCAGATGGAATTTCTCACTTTGCCCAGGAGTGTTTTCCCGAAGGACTGTCTATCGATCGTACCGTGCGCTTTGAAAACGACGGTACTATGACCTCGCATCATACCTATGAATTAGACGATACATGCGTGGTAAGTCGTATCACGGTAAACTGCGACGGTTTTCAACCTGATGGCCCAATCATGCGTGACCAGTTGGTCGATATCCTGCCTAATGAAACCCATATGTTCCCGCATGGGCCAAATGCGGTCCGCCAATTAGCATTCATCGGGTTCACGACTGCGGACGGCGGACTTATGATGGGGCATTTTGACTCTAAGATGACCTTTAACGGTTCGCGCGCGATTGAAATTCCTGGGCCGCACTTTGTGACGATTATTACAAAGCAAATGCGTGATACATCTGACAAACGCGACCACGTCTGTCAACGTGAAGTCGCTTACGCACATTCAGTGCCTCGCATTACCAGTGCGATCGGTTCAGATGAGGACTGATAACTGCCCAGGCATCAAATAAAACGAAAGGGTCAGTCGAAAACT
Sequencing result for KillerOrange:
TATAAAATAGGCGTATCACGAGGCAGAATTTCAGATAAAAAAAATCCTTAGCTTTCGCTAAGGATGATTTCTGGAATTCGCGGCCGCTTCTAGAGTACTTAATACGACTCACTATAGGGGAATTGTGAGCGGATAACAATTCCCCTCAAGAAATAATTTTGTTTAACTTTAAACCTAAAGAGGAGAAAAATGATGGAATGCGGCCCGGCGCTGTTTCAGAGCGATATGACCTTTAAAATTTTTATTGATGGCGAAGTGAACGGCCAGAAATTTACCATTGTGGCGGATGGCAGCAGCAAATTTCCGCATGGCGATTTTAACGTGCATGCGGTGTGCGAAACCGGCAAACTGCCGATGAGCTGGAAACCGATTTGCCATCTGATTCAGTGGGGCGAACCGTTTTTTGCGCGCTATCCGGATGGCATTAGCCATTTTGCGCAGGAATGCTTTCCGGAAGGCCTGAGCATTGATCGCACCGTGCGCTTTGAAAACGATGGCACCATGACCAGCCATCATACCTATGAACTGAGCGATACCTGCGTGGTGAGCCGCATTACCGTGAACTGCGATGGCTTTCAGCCGGATGGCCCGATTATGCGCGATCAGCTGGTGGATATTCTGCCGAGCGAAACCCATATGTTTCCGCATGGCCCGAACGCGGTGCGCCAGCTGGCGTTTATTGGCTTTACCACCGCGGATGGCGGCCTGATGATGGGCCATCTGGATAGCAAAATGACCTTTAACGGCAGCCGCGCGATTGAAATTCCGGGCCCGCATTTTGTGACCATTATTACCAAACAGATGCGCGATACCAGCGATAAACGCGATCATGTGTGCCAGCGCGAAGTGGCGCATGCGCATAGCGTGCCGCGCATTACCAGCGCGATTGGCAGCGATCAGGATTGATGACTGCCCAGGCATCAATTAAAACGAAAGGCTCAGTCGAAAAC
Conclusion:
Glasgow iGEM did fantastic work for us, providing us with detailed analysis of the T7 promoter and suggestions for improving the efficiency of our project. Whilst their data on the DH5alpha Z1 strain is accurate and in accordance with subsequent research and advice, we have since noted there is expression of KillerRed and KillerOrange in DH5alpha in lab tests.
