---

In this page, we showcase the achievements we have made in the project. The following are the list of achievements that we have made throughout the project!

Cloning new Level 0 and Level 1 Parts

In our project, our goal is to clone level 0, 1 and 2 parts to build an antithetical integral controller. As we are utilising JUMP assembly, we have created 39 Level 0 and 37 Level 1 parts that will be a part of the antithetic integral controller. To learn more about the parts that we have the created, please refer to the Level 1 table in the Parts page. Unfortunately, we were not able to build any level 2 circuits (see below). This meant that we were not able to test any open loop circuits and all of our antithetical integral controller circuits that would have been perturbed.

Characterisation of P_BAD promoters

araC is one of the feedback species being utilised for our circuit, characterised by Meyer (2019). As we are considering using araC to activate the P_BAD promoter, we perform part improvement by designing a family of new P_BAD through optimisation. For more details about the results of P_BAD part improvement, please refer to the Part Improvement page.

Characterisation of the autoregulatory negative feedback loop

Autoregulatory negative feedback is a type of circuit design motif that exists in nature to improve system response. To compare the response of the autoregulatory negative feedback circuit with the antithetic integral cotroller, we characterise the autoregulatory negative feedback in response to perturbation. For more details, please refer to the Engineering Success page.

Downstream site insertion into Level 2 acceptor (pJUMP47-2A)

In order to create a perturbation circuit to test whether the antithetic integral controller can robustly adapt, we require 5 transcription units, thus we need to utilise the downstream site of the JUMP vector. During our project, we successfully clone the perturbation circuit into the downstream site through designing a 5' adapter and a 3' adapter. For more details on the success of the integration of the perturbation circuit into the downstream, please refer to the Engineering Success page.

StayGold GFP Characterisation

Purpose

As we are using a low copy number plasmid in our main project, we characterise a new fluorescent protein StayGold GFP which has a considerably better excitation coefficient, brightness and quantum yield compared to other GFPs. To test the fluorescence of a StayGold GFP, we built level 1 circuits but unfortunately we did not observe any fluorescence.

Design

To characterise StayGold GFP, we have decided to use a low-copy number Level 1 acceptor plasmid (pJUMP27-1A) as a backbone where we will make use of both the main cloning site and the downstream cloning site. To read more about how JUMP assembly works, please refer to the Rationale page in the main project.

Our circuit consists of 1 transcription unit in pJUMP27-1A backbone, consisting of PTet*, RBS, GFP and DT5. For GFP, in order to compare StayGold GFP with other GFPs available (Figure 1), we look through the Open Reporters in the Distribution Kit Plate 2 to find the 2 GFPs (afraGFP, efasGFP) that has the closest excitation and emission wavelength compared to StayGold GFP to allow better comparison. For the RBS, we have decided to characterise StayGold GFP with both B0033 (weak), B0032 (medium) and B0034 (strong) to see the effect (Figure 2). Quantifying the effect of the strength of RBS is important because fluorescent proteins form aggregates when present in high concentrations. DT5 is an ultrastrong double terminator - discussion of the choice and rationale of terminator usage is discussed in the Design page.

The backbone chosen for the characterisation of StayGold GFP is pJUMP27-1A because it is a low copy number plasmid. As discussed in the Rationale page, a lower copy number plasmid both reduce metabolic burden and noise.

The identifier of the constructs used in our characterisation of StayGold GFP in the following Table:

Table 1. ID of the constructs we have created for the characterisation of StayGold GFP

ID	Promoter	RBS	CDS	Terminator	Purpose
LVL1 sinI	PTet*	B0032	sinI	DT54	Negative control
41	PTet*	B0032	StayGold GFP	DT5	Characterisation of StayGold GFP with B0032
42	PTet*	B0033	StayGold GFP	DT5	Characterisation of StayGold GFP with B0033
43	PTet*	B0034	StayGold GFP	DT5	Characterisation of StayGold GFP with B0034
45	PTet*	B0032	afraGFP	DT5	Compare StayGold GFP with afraGFP
45	PTet*	B0032	efasGFP	DT5	Compare StayGold GFP with efasGFP
PB10	PBadwt	B0032	StayGold GFP	DT5	Troubleshoot: Characterise StayGold GFP with a different promoter

Build

We perform JUMP assembly to clone the transcription unit into pJUMP27-1A using the enzyme BsaI to clone the transcription unit into the main cloning site. For the protocol we used for JUMP assembly, please refer to the Experiments page.

After golden gate assembly, the construct is then chemically transformed into DH5a cloning strain and plated out. For details of the protocol, please refer to the Chemical transformation protocol on the Experiments page.

After overnight incubation on Kanamycin selection plates, we want to confirm the construct we want exists in the colony. To do this, we perform colony PCR and gel electrophoresis on a non-fluorescent colony that does not contain superfolder GFP and check whether it has the correct length of amplicon after colony PCR (cPCR). For the details of the protocol, please refer to the to the the Experiments page.

For the colony with the correct cPCR amplicon length on gel electrophoresis, the colony is being grown overnight and glycerol stocks and Miniprep are being done the next day. For details of the protocol, please refer to the Experiments page.

Although the existence of a correct band in colony PCR is a good indication that plasmid with the correct construct is present, the length of colony PCR of the correct construct and the control is very similar (both about 1300bp) and thus will be indistinguishable with a gel. Moreover, to make sure that there is no mutation within the construct, the sequence has to be verified through sequencing.

Test

Before we perform the plate reader assay, during the cloning process, we have already noticed some interesting features of StayGold GFP. For superfolder GFP, it has a high fluorescent intensity that can be seen with the naked eye on selection plates. Therefore, we have expected to be able to see StayGold GFP fluorescing in the selection plate because DH5a does not have tetR within them some Ptet* should not be repressed, thus StayGold GFP should be expressed. However, the colonies we picked and subsequently for sequence verification are all non-fluorescing. However, this may be because Ptet* is not a strong enough promoter itself to show fluorescent. Therefore, we believe doing a plate reader assay is the best choice to confirm the fluorescent of StayGold GFP.

After sequence verification, we transform the plasmid into the test strain. For this experiment, we are using the Marionette strain from Meyer (2019) where tetR is present in the cell strain. The tetR present in the cell strain represses Ptet* where the repression will be lifted when the inducer aTc is added. This allows ut to control whether the expression of StayGold GFP is turned on or off.

To transform the plasmid into Marionette strain, we have to use electroporation as Marionette strain itself is not competent. The electroporation protocol can be found on the Experiments page.

Experiment 1

To test out how the concentration of inducers will affect the expression of StayGold GFP, we have designed our plate reader assay with different amounts of aTc inducers on several of our different constructs. We have chosen the range of aTc inducers used based on the characterisation done in Meyer (2019) on the Ptet* promoter. In the experiment, we have put in 2 negative controls (no_cells, Marionette_LVL1sinI) and Marionette_sfGFP as positive control as superfolder GFP is known to fluoresce. The test plasmids we have decided to use is ID41 and ID42 which is a transcription unit consisting of B0032 and B0033 with StayGold GFP respectively. The aim of the experiment is to compare the fluorescent intensity of StayGold GFP with superfolder GFP and to compare the fluorescent intensity of StayGold GFP with different strengths of ribosome binding sites. The plate reader layout is shown in Figure 3.

Results of Experiment 1

Looking at the numerical data of the plate reader experiment. We can see that the fluorescent intensity of the superfolder GFP is a lot higher than that of the constructs containing StayGold GFP. To confirm this, a graph of the fluorescent intensity against time is plotted which is shown in Figure 4.

Experiment 2 (Troubleshooting)

During Experiment 1, we did the experiment pipetting by hand which we have learnt is an error prone process. Moreover, as the first experiment shows that StayGold is not fluorescing, we have decided to repeat the experiment to troubleshoot the results. To do this, we have decided to conduct the experiment using both DH5a and the Marionette strain with all the constructs we have for StayGold, efasGFP and afraGFP with Ptet* with no inducer (aTc) added or with a concentration (2μM) leading to maximal response added. We have also included a construct with the Pbad wt promoter with StayGold GFP to confirm that it is not the promoter that is not working with and without arabinose. Therefore, for this second iteration of the experiment, we have 14 different culture conditions altogether which makes pipetting manually difficult and error prone. As a result, we have decided to make use of automation. We perform the experiment using OT-2 robot which allows automated pipetting with a Python script invented by Camillo Moschner, one of the instructors of the iGEM Team (Moschner et al., 2022). The plate layout is shown in the figure 4, a screenshot of the Python script we used to control the OT2 robot. As we can see, the plate layout is very complex and the experiment is benefiting hugely from automation.

The day before the plate reader experiment, we overnight culture the 12 different strains of cells from a streaked out plate into EZRDM for approximately 12 hours. After overnight culture, we pipette 14μL of the respective cell culture into PCR tubes and place it in the order shown in the screenshot of the Python script output into the robot (Figure 4). We also made EZRDM media with an appropriate Kanamycin and inducer concentration into Falcon tubes and place in the appropriate order as shown in the figure.

After preparation, the following plate reader settings were used to characterise StayGold GFP.

Results

The results again show that StayGold GFP does not fluoresce as expected. The following graphs are plotted as a comparison. We first plotted efasGFP, afraGFP and StayGold GFP (all RBS strengths) on the same plots under the same promoters (Ptet*) for both the DH5a strain (Figure 10) and the "Marionette" strain (Figure 11). The result shows that inside both Marionette strain (with aTc added) or DH5a, although efasGFP is fluorescing strongly and is saturating the plate reader at this gain, both afraGFP and StayGold GFP did not have a high fluorescence intensity.

To ensure aTc is performing what we expect of alleviating the repression of tetR on Ptet*, the time series of fluorescent intensity of efasGFP inside the Marionette strain with and without aTc is plotted on figure 12, showing that the addition of aTc does increase fluorescent intensity.

Apart from comparing the effects with different aTc concentrations added, we also use a Pbad promoter (PB10) to ensure that the previous low fluorescent intensity signal was not due to the Ptet* promoter. The following figure shows the fluorescent intensity of StayGold GFP under the Pbad wild type promoter with different concentrations of arabinose. As we can see, the fluorescent intensity did not significantly change even with the addition of arabinose.

Plotting the constructs containing StayGold GFP of different RBS strengths on the same graph for both DH5a (Figure 13) and the Marionette strain (Figure 14), we see that although the trace of B0032 and B0033 is very similar, the trace of B0034 is significantly higher. This may imply that putting StayGold GFP under an even stronger promoter may cause it to have a way higher fluorescent intensity.

DBTL cycle: Trouble shooting attempts

Throughout building our level 2 to create antithetic integral controllers we encountered a number of challenges in our cloning process. We initially successfully cloned 39 parts into level 0 plasmids and 37 transcriptional units into level 1 plasmids, however we were unable to complete any level 2 system. To overcome these cloning problems we tried implementing a number of troubleshooting techniques.

Success rate of the cloning process and optimisation strategies

Cloning our DNA parts into the level 0 plasmids was relatively straightforward however making and cloning the level 1 parts was substantially harder. The average overall success rate was about 25%. Once Golden Gate Assembled (GGA) and transformed, approximately 30% of the resulting colonies failed to show any non-fluorescent colonies (those which had inserted into backbone superfolder GFP (sfGFP) CDS) indicating that the golden gate protocol failed. The success rate within a given successful golden gate assembly to transformation varied dramatically from 10 to 90%. We attempted to optimise this by test out different cycle numbers and the time in each cycle (see overnight compared to same day GGA in protocols). However, little difference in golden gate assembly success occurred. From those non-fluorescent colonies about 50% had correct cPCR amplicon length. Often we accidently picked sfGFP colonies that were not fluorescing brightly. From these colonies, typically 2-3 were sequenced. Often the sequencing results showed that the transcriptional units were missing parts. Often these were RBS or promoters which were too small to distinguish on a cPCR gel given that they could be smaller than 100 bp.

Module 1B with Anderson promoter

This was particularly a problem for module 1 B’s with Anderson promoters. Our design for our antithetic integral controller circuit involves an Anderson promoter constitutively expressing Z1 in Level 1 Module B. Without these circuits we couldn’t build our level 2 circuits. When we are trying to clone in the transcription unit into Level 1 Module B, the promoter and RBS are removed from the assembly such that only the CDS and the Terminator are present. This has happened consistently throughout the cloning process where out of the 20+ tries for various different types of promoters and RBS, only 2 of the assemblies are successful with the whole transcription unit being cloned in. In all cases, both the RBS and promoter were missing.

We find this strange given JUMP assembly has ensured that the overhang of CDS is not compatible with the overhang of the backbone thus CDS and the backbone cannot be stitched together. In silico cloning in Snapgene further suggested that the backbone should not be able to be stitched directly with the CDS and terminator.

We have also found that the transformation efficiency of these plasmids into cells are very low where almost all of the cells on plates contains superfolder GFP, suggesting that cloning was unsuccessful. When we are trying to pick the non-fluorescent colonies, in a lot of the cases, we have resulted in a liquid culture that does not grow in Kanamycin culture and have no bands when running gel electrophoresis after colony PCR. This means that these non-fluorescent colonies do not contain the cloning product we want as well. Typically this indicates that the antibiotic selection has failed but this was controlled for by a negative/positive controls as well as replacing the antibiotics stocks.

A variety of strategies were used to tackle this problems. Firstly, we decided to try double and then quadruple the concentration of promoter and RBS to see whether the situation has improved. However, this yields the same result. To test whether it was a problem with module B, we cloned Pbad_wt for our characterisation circuit. Out of the three we tried, two worked. However the success of these cloning reactions remained low. We further hypothesised that since we were using strong constitutive promoters could have resulted in metabolic toxicity. One attempt to mitigate this was to grow the plates at 30C for 48 hrs rather then 37C for 12 hrs which should reduce the toxicity. However, this resulted in little additional success. Furthermore, even weak Anderson promoters (J23105) and weak RBS (B0033), in which high toxicity/metabolic load were not to be expected, also failed. A final strategy to make the level 1 parts was to use a linear GGA protocol instead of plasmid GGA. We had to design overhang adapter sequences with the appropriate level 1 restriction sites. The parts were then assembled using linear GGA, PCR amplified and purified. We identified bands of correct size but were unable to fully assemble the level 1 plasmid.

Level 2 cloning difficulties

Similar wto the level 1 circuits we were only ever to assemble two out of the four level 1 transcriptional units into a level 2 system. We tried approximately 20 cloning runs, trialling different golden gate protocols and restriction enzymes (Bsmb1 and Esp1) but all failed. In all cases only module A and B were found and attached unexplainably directly to the backbone.

References

1. Meyer, A.J., Segall-Shapiro, T.H., Glassey, E. et al. Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nat Chem Biol 15, 196–204 (2019). https://doi.org/10.1038/s41589-018-0168-3


2. Marcos Valenzuela-Ortega, Christopher French, Joint universal modular plasmids (JUMP): a flexible vector platform for synthetic biology, Synthetic Biology, Volume 6, Issue 1, 2021, ysab003, https://doi.org/10.1093/synbio/ysab003


3. Hirano, M., Ando, R., Shimozono, S. et al. A highly photostable and bright green fluorescent protein. Nat Biotechnol 40, 1132–1142 (2022). https://doi.org/10.1038/s41587-022-01278-2
4. 
   Moschner Camillo, Wedd Charlie, Hardo Georgeos, Bakshi Somenath (2022). The iBioFoundry: Automated, Low-Cost, High-Throughput Experimentation (IWBDA manuscript accepted)