The Kill Switch
Design
While our team is excited to contribute towards the needs of our local community, there is hesitation from those in charge of managing the contamination because of the use of a modified bacteria. To ease the concerns of the community, we have added a kill switch to our overall design as an additional layer of biosecurity.
Last season, our team had significant engineering success in the assembly and transformation of a quorum sensing mazEF toxin/ antitoxin kill switch system. We began testing the efficacy of the design and found that we would like our kill switch to take effect in a narrower range of concentrations in order to incorporate this biosafety feature into the full biosensor. This season, our goal was to redesign the kill switch for improved efficacy.
Figure 1.1: Illustration of Original Kill Switch Circuit Design
We began by identifying the ways in which the kill switch could be modified and initially designed 3 different modifications based on the result of last season’s testing.
The first modification was a change in the ribosome binding site used.
Figure 1.2: Illustration of Modification 1 of Kill Switch, replacing medium RBS with stronger RBS in front of toxin gene
Part B0032 is a moderate ribosome binding site, compared to Part B0030 which is considered a strong ribosome binding site. By placing a stronger ribosome binding site before the coding region for the toxin, more of the toxin will be produced. This means that a higher concentration of cells must be present to produce more of the antitoxin, ensuring that the toxin can take over at low concentrations if the bacteria were to escape their testing kit. We can use this as one route to increase kill switch efficacy.
Modification two uses similar methods, in the sense that only certain parts are switched out.
Figure 1.3: Illustration 1 of Modification 2 of Kill Switch, placing a stronger general promoter in front of the portion of the circuit containing the toxin
Replacing the moderate promoter Part BBa_J23100 with the strong promoter BBa_J23199 will increase transcription of the portion of the genome that contains the toxin, increasing its production. This would also increase production of LuxI and LuxR, and in turn more HSL. Producing both more toxin and more activator might give useful insight as to how exact and effective the kill switch is altogether.
The third modification uses a much different method of modification. Last season, the qPCR results indicated that the coding region for the toxin may be degraded at its position directly at the end of the circuit. This redesign switches the order of the portions in the circuit, so that the toxin gene is in the middle rather than at the end, potentially preventing its possible degradation. However, Modification 3 remained in the design phase and efforts were focused on Modifications 1 and 2.
Figure 1.4: Illustration of Modification 3 of the Kill Switch. Switching the order of circuit segments could potentially protect the toxin gene from degradation.
Alignments were performed for each design to confirm that each circuit would assemble properly and DNA was ordered.
The DNA needed to transform the circuit into competent cells was ordered from TWIST Biosciences, and we began developing our modified kill switch.
A control kill switch to measure output which replaced the toxin with RFP was also designed to correspond to each modification.
Build
Go to TopAssembly and Transformation
The DNA was ordered from TWIST Biosciences and arrived freeze dried in a 96-well plate. For resuspension the plate was gently centrifuged and opened, and the DNA was resuspended in nuclease free Tris-EDTA (TE) buffer, pH 8.0 or 10 mM Tris-HCl, pH 8.0 to the desired concentration.
Table 1: DNA resuspension values for assembly prep
Once resuspended and combined with the plasmid backbone, the circuit was assembled using the HiFi Assembly method and then transformed into competent cells using the NEB High efficiency transformation protocol.
After transformation, the cells were plated on HSL enhanced agar plates and allowed to grow overnight.
Figure 2.1: Agar plates with colonies suspected to contain assembled plasmids with designs Mod1 and Mod1C
Figure 2.2: Agar plates with colonies suspected to contain assembled plasmids with designs Mod2 and Mod2C
Colony PCR and Gel Electrophoresis
Colony PCR was performed using 8 colonies chosen for Mod 1 and Mod 1C, and 4 colonies were chosen for Mod 2 and Mod 2C. PCR tubes containing master mix with 2X TAQ and VR/VF2 primers were inoculated from the selected colonies using sterile toothpicks and amplified using the PCR settings with an annealing temperature of 53℃, 155 second extension time, and 29 cycles. Resulting colonies should be about 2,600bp in size, if correctly assembled, for the modified kill switch and 2,900 for the control
Figure 3: Agar plates demonstrating the colonies chosen to perform colony PCR and gel electrophoresis for each Modification/ Control
After a Gibson assembly and colony PCR was performed, gel electrophoresis was run for each plasmid sample on a 0.8% agarose gel. These samples included newer modifications with their controls. The control samples used red fluorescent protein (RFP) in place of the toxin.
Figure 4.1: Gel electrophoresis for Mod1 Plate, colonies 1-8
Figure 4.2: Gel Electrophoresis, Mod1C Plate, colonies 1-8
Several of these are at or around the 3kbp mark (expected with the size of the assembly + ~200 bp for VF2 & VR.
For further confirmation of synthesis, the PCR products were diluted 1:1000 and used in another PCR with primers internal to the kill switch.
For the Mod 1 assembly, these primer combinations were tried:
Table 2: Mod 1 assembly primer combinations
Amplification was completed with PCR settings 52℃ annealing temperature, 55 second extension time, and 21 cycles. This lower number of cycles was chosen because only 1000X amplification is required. These samples were resolved on a 0.8% agarose gel.
Fig 4.3: Gel Electrophoresis, Mod1-1 with primer mixes A, B, C, D
Fig. 4.4: Gel electrophoresis, Mod1-5 with primer mixes A, B, D, E
Fig. 4.5: Gel Electrophoresis, Mod1-8 with primer mixes A, B, C
The above gel results were to confirm the presence of the specific genes (MazE, MazF, LuxI, etc) within the kill switch modifications before being sent off for sequencing. The bands being located in their specific positions is indicative of the presence of the target segments that we were hoping to see.
Sequencing Data
From the colonies chosen to complete the colony PCR, liquid cultures were created by inoculating 3mL of LB + 3uL Chloramphenicol antibiotic (CM) enhanced with HSL. These liquid cultures were grown in the shaker overnight and the miniPREP procedure was performed to separate the DNA from the rest of the contents. This was sent to Michigan State University for sequencing.
Sequencing results were indicative of synthesis of the intended DNA circuits. In each of the following photos, the topmost gray line represents the expected sequence. Each of the smaller, gray and/or red segments represent different segments of the sequence that are chosen using different primers and areas of overlap to determine how closely the synthesized sequence matches the expected sequence. The areas of gray indicate a match with the expected sequence, while the red does not match. Overall, the significant areas of gray overlap between segments indicates that we did synthesize the intended circuit. Noting that there are some areas of significant red, it is important to keep in mind that the closer you are to the end of the sequence, the lower the quality of the reading as these ends can be damaged during processing.
Fig. 5.1-5.5: Visualization of sequencing results for various colonies
Fig 5.1: Clone Mod1-8
Fig. 5.2: Clone Mod1c-2
Fig 5.3: Clone Mod1c-8
Fig. 5.4: Clone Mod 2-1
Fig. 5.5: Clone Mod 2-2
Test
Go to TopNew Colonies/Liquid Cultures
New colony plates were created on HSL enhanced plates using the methods previously described, and the plates were allowed to grow overnight.
Fig. 6: Test plates for Mod1/1C, Mod2/2C, and the original kill switch/its corresponding control to use for testing. 2 colonies chosen per intended circuit.
The liquid cultures were created by using a sterilized inoculating loop to pick up bacterial colonies and inoculate the culture solution containing 3mL LB + 3uL CM enhanced with HSL to help drive the bacteria in the different colonies to grow in liquid cultures.
Colony Concentration Tests
From the liquid cultures, each of the 11 colonies was tested at a variety of concentrations to determine the efficacy and confirm the action of the kill switch. Liquid cultures were diluted into PBS solution and placed into a 1 cm cuvette. The cuvette samples were placed in a spectrophotometer to measure the OD600 of each sample to help calculate and determine their concentrations in liquid culture. An OD600 value of 1.0 indicates a concentration of 1x109CFU/mL; the OD600 of each culture was measured and used to calculate a number of microliters culture equivalent to 1x1010 CFU so that these cells could be pelleted.
Table 3: Calculation of uL liquid culture required to pellet an equivalent number of cells for each culture to set a standard starting concentration
The liquid cultures were placed in a microcentrifuge tube and spun at 12.0 rpm for 90 seconds and the supernatant was removed. The cells were resuspended in 1mL LB+CM to yield our intended starting concentration. A serial dilution to give a final concentration of 10 CFU/μL was completed.
Table 4: Calculation of concentration of cells pipetted per dilution
10 uL of each dilution was transplanted onto a 10uL HSL enhanced plate and a control plate in a grid pattern in order to determine at which concentration the cells are no longer able to survive in order to compare this to what was observed last season. These plates were incubated at 37℃ and allowed to grow overnight.
Liquid Culture Dilution
Using the OD 600 values and serial dilutions previously described, (which concentrations were used in the liquid culture plate and how was this test set up, table). A 96 well plate was used to test triplicate samples of dilution 3 as well as dilution 5 of each sample. This 96 well plate was then placed into the microplate absorbance reader to record results over about 16 hours.
qPCR
qPCR was performed from each of the 11 cultures. We began by pelleting cells via centrifugation at 12.0rpm for 90 seconds and resuspending in 1 mL water to an equal cell density of 1.0x109CFU/mL. Samples were boiled for 10 minutes, followed by 10 minutes in the freezer. This disrupts the cell membranes so that we can access the plasmid and the DNA in the cells. An RT and Control reaction for each of the 11 lysates was prepared using 2uL either RT or control mix, and 8uL lysate. These reactions were incubated at room temp for 2 minutes, 55℃ for 10 minutes, and 95℃ for 1 minute in order to prepare them for the qPCR reaction. The table below demonstrates the setup for the qPCR reaction.
Figure 7: Visualization of qPCR setup
Learn
Go to TopColony Concentration Tests
With the designs from last year shown in the images below, we can see that it isn’t until the final dilution that we observe singular CFU that can be counted (lysate id#:8-11).
With the previous design, we saw too much growth at the lower concentrations and observing fewer CFU in the last 3 columns for Mod 1 and 2 is a good indication that we successfully improved the efficacy of the kill switch.
Visually, it appears that culture 5, Mod 2-1, is working with the best efficacy. On the plate enhanced with HSL, the culture has the ability to grow across the grid pattern, though single CFU can be observed at low cell concentrations. However, the control plate with no HSL demonstrates that the colonies cannot grow past the second column, demonstrating the efficacy of the kill switch as cells did not have the ability to survive at this lower concentration.
Figures 8.1-8.6: Serial dilutions 1-5 plated on 0.8% agarose
Figure 8.1: Mod 1/1C, HSL enhanced
Figure 8.2: Mod1/1C, Control
Figure 8.3: Mod2/2C, HSL enhanced
Figure 8.4: Mod 2/2C, Control
Fig. 8.5: Original Kill Switch/Control, HSL
Fig. 8.6: Original Kill Switch/ Control, Control
Liquid Culture Dilution
Figure 9.1: Mod 2-1, comparing OD600 of dilutions high vs. low cell concentration over 16 hours
Figure 9.2: Ɑ5 (original kill switch), comparing OD600 of dilutions high vs low concentrations over 16 hours
Liquid culture test results were analyzed using a 96 well microplate that was placed in the microplate reader to record the sample’s OD measurements over the course of about 16 hours. These are measured both at the high and low CFU concentration, where it can be seen that there is significantly less absorbance with the less concentrated samples. This is indicative that our kill switch is performing in the manner that it is supposed to by not rapidly replicating, which is a great improvement on the kill switch from the year prior.
qPCR
To measure the level of gene expression from each colony, qPCR was performed on each of the 11 lysates, testing 3 different primers each, with both an RT and control reaction. Below is the plot of amplification in Lysate 1, RT, with primers mazF, mazE, and LuxI. This plot demonstrates that each gene we attempted to amplify is present and accounted for.
Figure 10.1: qPCR Amplification Plot, Mod 1-5, RT reaction for mazF, mazE, and LuxI
The graph above shows the CT values for mazF (blue), mazE (orange), and LuxI (gray). All three of these primers are on the same RNA stand so their CT values should be very similar, which can be observed in the graph.
Figure 10.2: qPCR Dissociation Curve, Mod1-5, comparing gene expression of mazF, mazE, and LuxI
Each of the reactions ran only produced one peak on the dissociation curve which indicates that only the desired PCR product was produced. The peaks are all in similar locations showing that the reactions were of good quality.
Culture 5 contains Sample Mod 2-1. Visually from the colony tests, Mod 2 appears to function better than last cycle’s kill switch, which can be demonstrated quantitatively by comparing mazF expression in culture 5 (which contains Modification 2) to mazF expression observed in culture 9 (which contains last cycle’s design).
Fig. 10.3: Amplification plot comparing mazF expression in old vs. new circuit to demonstrate improvements in efficacy of design
As observed in the last cycle, culture 9 reveals that mazF is being degraded and no expression is seen at the end of qPCR, however, in this current cycle Culture 5 clearly demonstrates increased expression of the toxin gene, effectively proving the improved efficacy of Modification 2 showing evidence of engineering success. Further testing is needed to determine what mechanisms Mod 2 is acting upon to increase the efficacy of the circuit.
Future Steps
Go to TopFuture steps could include the assembly and implementation of a third design on our kill switch that can be further tested and compared to the previous two designs based on the efficacy, performance, and reproducibility of results. A series of trials will need to be completed to perform the necessary comparison to the data of previous designs. Essentially, the assembly and testing of a third design would only take place in the event that designs one and two did not perform to the expected results and would be accomplished by the rearrangement of the circuit to place the toxin in the middle of the sequence instead of the end to avoid degradation when attaching to the backbone. Another alteration we could apply includes the implementation of the enzyme lactonase into the original assembly, which can be added to the liquid culture and should prevent the bacteria present in the culture from sharing HSL molecules and would ultimately result in activation of the kill switch, thus shutting down the whole system. However, this would require purchasing the lactonase enzyme, and then cloning it before being applied to the circuit. This could potentially be a more efficient implication compared to previous designs due to the potential of the deletion of the HSL molecule all together, rather than lowering its concentration within the cells.