Overview
This section of our wiki describes how we iterated our design for the taurocholic acid (TCA) sensor, which serves as the overall "switch" of our probiotic platform.
Cycles #1-4 describe our attempts to improve the CmeR-pCmeR sensor design, which is based on TCA's ability to bind to CmeR and dissociate it from cmeO, therefore allowing downstream gene expression. We used the design-build-test-learn cycle four times to find the optimal positioning of cmeO, -35 region, and -10 region inside pCmeR, the appropriate RBS strength in cooperation with the promoter, as well as constitutive promoter upstream of CmeR (Figure 1A). At the end of Cycle #4, however, our experimental data and published literature collectively indicated that TCA has very limited ability to cross the plasma membrane and thus cannot effectively act as a cytoplasmic ligand for CmeR.
This limitation of the CmeR-based TCA sensors drove our design for another TCA sensor in Cycle #5, in which we incorporated the transmembrane detector TcpP as the core component of our TCA sensor. Further, we replaced TcpP's DNA-binding domain with that of CadC, which is compatible with E. coli, and expressed TcpH as a protection against TcpP degradation. In the presence of TCA, the CadC-TcpP fusion protein will dimerize and activate promoter pCadBA. (Figure 1B) At the end of Cycle #5, our experimental results showed that this TcpP-based TCA sensor succeeded in detecting TCA with a >20-fold dynamic range.
Figure 1. The plasmid designs of TCA sensors in our engineering process. A, plasmid designs of CmeR-based TCA sensors used in Cycles #1-4; the first 4 cycles aim to optimize the performance of this TCA sensor by tuning the switchable components (including the promoter upstream of CmeR, the promoter upstream of GFP, and the RBS upstream of GFP). B, plasmid design of the TcpP-based TCA sensor used in Cycle #5. Created by biorender.com.
Cycle #1
Design
Our TCA sensor was constructed to detect bile acids, which serve as an indicator of food presence. We first selected the TetR-family repressor CmeR as a critical component due to its strong, well-characterized binding affinity for TCA (with a dissociation constant of 1.5 μM) (Lei, 2011), making it ideal for modulating gene expression in response to TCA.
In our initial design, CmeR was placed under the control of the constitutive promoter J23119. J23119 was chosen for its ability to drive high-level transcription, ensuring substantial expression of CmeR. The CmeR-regulated promoter, pCmeR, was constructed from pQacR (BBa_J428050) by replacing the operator sequence qacO with cmeO (Nasr et al., 2022). GFP was placed downstream of pCmeR to enable monitoring of CmeR's regulatory function. When TCA binds to CmeR, it disrupts CmeR's inhibitory function, leading to the expression of GFP (Figure 2). This interaction provides a visual representation of our system's response to TCA levels.We call this plasmid pJ23119-CmeR_pCmeR(-10)-B0031-GFPmut2, in which the major functioning device is BBa_25F4UCQM.
Figure 2. The genetic circuit of the initially designed TCA sensor, pJ23119-CmeR_pCmeR(-10)-B0031-GFPmut2. Created by biorender.com.
Build
We integrated this design into the plasmid backbone pUC57-Kan (Figure 3), and ordered the plasmid from Tsingke.
During the synthesis process, the company was unable to produce the correct sequence. To address this problem, they explored various strategies, including conducting several rounds of experimental repairs, substituting with a special competent strain, and culturing at low temperatures. Nevertheless, after multiple failed attempts, the best result they produced still contained a point mutation in the sequence (Figure 4).
Figure 3. Plasmid map of pCmeR_GFPmut2_pUC57-Kan.
Figure 4. Sequence comparison of the designed and mutated plasmids. The red box indicates a nonsense mutation in the cmeR gene.
Test
To investigate the mutations in cmeR, we performed in silico analyses to predict the transcription initiation and translation rates of the cmeR gene within the plasmid using the Promoter Calculator and RBS Calculator algorithms of De Novo DNA (LaFleur et al., 2022; Salis et al., 2009). First, the predicted transcription initiation rates across the entire plasmid revealed a significant peak of 40055 au at 1245 bp on the reverse strand, corresponding to the site of J23119-CmeR transcriptional unit (Figure 5). Second, a high translation rate of the full-length CmeR protein was also identified by prediction (Figure 6). They collectively confirmed that CmeR could be expressed at significantly high levels.
Figure 5. The De Novo DNA PromoterCalculator result of originally designed pJ23119-CmeR_pCmeR(-10)-B0031-GFPmut2. The black circle indicates the peak in transcription initiation rate of theJ23119-CmeR transcriptional unit.
Figure 6. Translation initiation rates ofthe mRNA encoding CmeR. The red circle suggests a peak in CmeR translationinitiation, indicating the translation of the full-length CmeR protein.
Learn
The combined insights from these in silico analyses suggest that the high transcription and translation levels of CmeR might result in a significant level of CmeR within the cell. Meanwhile, a high-copy plasmid also contributes to the elevated CmeR levels. This may have led to an excessively high level of the transcriptional repressor, potentially resulting in toxicity to the host cell and killing it (Barneda-Zahonero et al., 2015). Hence, we ultimately concluded that this was the reason why only the variants incorporating a mutation in the cmeR gene survived.
Cycle #2
Design
As indicated in the "learn" part of Cycle #1, we identified a possible reason for the infeasibility in synthesizing our target plasmid to be the usage of the J23119 promoter, which might lead to toxic levels of CmeR in cells with the correct sequence. Therefore, in this cycle, we aimed to address two problems: we tried to fix the unwanted stop codon and to turn down the transcriptional activity of the constitutive promoter upstream of CmeR. Since J23119 is the strongest member of a family of constitutive promoters, we decided to mutate promoter J23119 and choose a mutant with suitable transcriptional strength.
To achieve this, we designed primers that keep the same spacer sequences of J23119 but have degenerate nucleotides for the -35 and -10 regions (Figure 7). In doing so, we obtained a library of constitutive promoters of varying strength. We called this plasmid pJ23119(lib)-CmeR_pCmeR(-10)-B0031-GFPmut2. After switching to a promoter of weaker strength, correct transcription of the plasmid and suitable levels of expression of CmeR should be achieved in E. coli. Therefore, we hypothesized that after using a weaker promoter, CmeR expression can be well-tolerated by E. coli.
Figure 7. Degenerate nucleotides thatmodify J23119's -10 and -35 regions in the J23119 promoter library. The"Y", "R", "M", "W", and "K"notations are degenerate nucleotides that respectively represent thenucleotides: C or T, A or G, A or C, A or T, G or T.
Build
PCR and Golden Gate assembly using specially designed primers were performed to introduce mutations in the promoter sequence, creating a J23119 library, and to fix the stop codon into the original amino acid. Agarose gel electrophoresis (AGE) was performed, and the bands indicate the correctness of the components' and the assembled product's length (Figure 8), verified by whole-plasmid sequencing.
Figure 8. The AGE result of the PCRproducts (A and B) and Golden Gate assembly product (C) for the construction ofthe plasmid pJ23119(lib)-CmeR_pCmeR(-10)-B0031-GFPmut2. The bands indicate thecorrectness of the components' and the assembly product's length.
Test
The plasmid library of J23119 mutants was transformed into E. coli Trelief 5α, and solid inoculation was carried out. Most colonies had significant levels of green fluorescence. Because no TCA was added into the plate, the result contradicted our expectation that there would be no fluorescence, suggesting a high leaky expression of our design.
Learn
Our results suggest that while our mutant promoter from the J23119 library is successfully incorporated and can correctly express CmeR in E. coli, leaky expression of GFP is a severe issue. This may be caused by CmeR not being able to sufficiently block pCmeR from transcription factors' binding and elongation. In order to prevent leaky expression of GFP, it is necessary to enhance the repression effect of CmeR on pCmeR. This can be achieved by increasing the dominance of CmeR in its competition with transcriptional factors at PcmeR. However, CmeR expression should not be elevated due to its toxicity to E. coli at high levels. In the next cycle, we will try to alter the location of cmeO on pCmeR in order to enable CmeR to have complete control.
Cycle #3
Design
The plasmid design in this cycle was based on one variant (BBa_25A12OMH) of the J23119 library built in Cycle #2, called J23119(mut). To increase CmeR's ability to inhibit transcription, we moved the cmeO region in pCmeR to make it overlap with the -35 region (indicated by the red boxes in Figure 9), meanwhile changing the sequence in the -35 region to be compatible with this overlap (Figure 9). In this cycle, we also changed the GFP used, from GFPmut2 into superfolder GFP (sfGFP, BBa_K4850004), in order to achieve consistency across all reporter proteins used in our project.
Figure 9. The comparison ofpCmeR(-10)-B0031 and pCmeR(-35)-RBS(Bgl). The red boxes indicate the same cmeOsequence but at different locations. The blue boxes indicate the samespacer-(-10) sequence at the same locations.
The above change in the -35 sequence turns down the transcriptional rate of pCmeR by more than two-fold, which was verified by the in silico analysis in Figure 10. The transcriptional rate in this plasmid would even be lower considering the inhibition of CmeR, which occupies the transcriptional factor-binding region.
Figure 10. The comparison of predictedtranscriptional rates of GFP under pCmeR(-10) and pCmeR(-35). The yellow boxesand the yellow arrow indicate the pCmeR's location on the plasmid map. The redcircles indicate the peaks in transcriptional rates of the pCmeR region.
However, we did not want to lower the GFP expression in the presence of TCA. We therefore changed the upstream RBS sequence from B0031 into a stronger RBS, called RBS(Bgl) (Figure 9). This RBS was from the standard BglBrick plasmid library (Lee et al., 2011). We then ran an in silico analysis of translation rates, which verified the intended 15-fold increase in the RBS strength (Figure 11). All the above changes to the previous design eventually resulted in the plasmid named pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP, in which the major functioning device is BBa_25E1I21E.
Figure 11. The comparison of predictedtranslation rates of GFP downstream of B0031 and RBS(Bgl). Red circles indicatethe correct peaks in the translation of full-length GFP.
Further, we designed another plasmid, pJ23119(mut)-CmeR(2*TAA)_pCmeR(-35)-RBS(Bgl)-sfGFP (in which the major functioning device is BBa_256FO3O6), as a control to evaluate the effect of the CmeR protein. In this plasmid, most of the sequence is the same as the plasmid designed earlier in this section, except that it includes nonsense mutations on the second and third triplets of the cmeR sequence (Figure 12). This new plasmid will not express functional CmeR protein but has other parts the same, therefore serving as an appropriate control to pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP.
Figure 12. The comparison between sequences of CmeR and CmeR(2*TAA). The red box indicates the introduction of the 2*TAA nonsense mutation.
Build
PCR was performed to acquire most parts of pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP. For the region depicted in Figure 9, however, there was no ready-made template in previous plasmids, and it was too long (about 60 bp) to be carried in Golden Gate Assembly primers. Therefore, we annealed two ssDNA primers together to form our desired DNA segment, together with sticky ends compatible with other fragments in the Golden Gate Assembly (Figure 13).
Figure 13. Annealed DNA product. The top and bottom strands are two ssDNA primers.
We then performed the Golden Gate Assembly using 2 fragments from PCR and the above annealing product. Agarose gel electrophoresis (AGE) was performed, and the bands indicated the correctness of the PCR products' and the assembled product's lengths (Figure 14). The plasmid sequence was verified by whole-plasmid sequencing.
Figure 14. The AGE result of the PCRproducts (A) and Golden Gate assembly product (B) for the construction of theplasmid pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP. The bands indicate thecorrectness of the components' and the assembly product's length.
The construction of pJ23119(mut)-CmeR(2*TAA)_pCmeR(-35)-RBS(Bgl)-sfGFP was done in a similar way, with two fragments derived from PCR and one fragment being the annealing product. We then performed the Golden Gate Assembly and agarose gel electrophoresis. The bands indicated the correctness of the PCR products' and the assembled product's lengths (Figure 15). The plasmid sequence was verified by whole-plasmid sequencing.
Figure 15. The AGE result of the PCR products (A) and Golden Gate assembly product (B) for the construction of the plasmid pJ23119(mut)-CmeR(2*TAA)_pCmeR(-35)-RBS(Bgl)-sfGFP. The bands indicate the correctness of the components' and the assembly product's length.
Test
We transformed both plasmids into E. coli Trelief 5α for cloning, and the one without the 2*TAA mutation into E. coli Nissle 1917 for the kinetics assay. The detailed plate setup in the assay is shown in Figure 16.
Figure 16. Plate layout of kinetics assay for pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP.
The results of the kinetics assay were used to calculate the fluorescence / ABS600 curve. The normalized curve of the groups and the stationary phase result both showed no statistical differences as TCA concentration changes (Figure 17).
Figure 17. GFP expression of pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP in the kinetics and the stationary phase. Fluorescence / ABS600 was used to represent GFP expression. Error bars represent SD. The red box indicates the data used in plotting the stationary phase result.
To explore the underlying reason for this unexpected result, we looked at the solid inoculation result of pJ23119(mut)-CmeR(2*TAA)_pCmeR(-35)-RBS(Bgl)-sfGFP, the plasmid with the 2*TAA mutation and without functional CmeR. All colonies showed little or no green fluorescence.
Learn
The results suggest that integrating cmeO directly into the -35 region of pCmeR weakens the promoter too severely. As a result, transcription was minimal regardless of whether CmeR was functional or inhibited by TCA. This explains why pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP showed little induction by TCA, and why the 2*TAA control showed little fluorescence. While this design improved repression potential, it also disrupted RNA polymerase recruitment, leaving the promoter unable to generate measurable GFP expression.
To maintain repression while preserving basal promoter strength, cmeO should be shifted into the spacer region between -35 and -10 in future cycles. This placement would still allow CmeR to effectively block transcription when bound, but also permit RNA polymerase binding and initiation once CmeR is released by TCA.
Cycle #4
Design
The results from Cycle #3 revealed that placing cmeO directly within the −35 region of pCmeR largely lowers promoter activity to the extent that no measurable GFP expression could be induced, regardless of CmeR function. To balance promoter strength and repression efficiency, in Cycle #4 we redesigned the regulatory region by shifting cmeO into the spacer between the -35 and -10 sites (Figure 18). This positioning was chosen because it allows CmeR to physically obstruct RNA polymerase progression when bound, while still permitting basal promoter activity (by not disrupting the key sequence to recruit transcription factor) and full transcriptional activation once CmeR is released by TCA. This change led to the plasmid called pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP, in which the major functioning device is BBa_25RC9Z6P.
Figure 18. The comparison of pCmeR-RBS region between pCmeR(-10)-B0031, pCmeR(-35)-RBS(Bgl), and pCmeR(spacer)-RBS(Bgl). The red boxes indicate the movement of the same cmeO sequence as we improved our design.
The above change in the -35 sequence increases the transcriptional rate of pCmeR by more than 4-fold, even larger than the original design of pCmeR(-10), which was verified by the in silico analysis in Figure 19. We expected that this design could allow high GFP expression when CmeR is not blocking cmeO.
Figure 19. The comparison of predicted transcriptional rates of GFP under pCmeR(-10), pCmeR(-35), and pCmeR(spacer). The yellow boxes and the yellow arrow indicate the pCmeR's location on the plasmid map. The red circles indicate the peaks in transcriptional rates of the pCmeR region.
Similar to what we have done in Cycle #3, we additionally designed a plasmid without full-length CmeR, called pJ23119(mut)-CmeR(2*TAA)_pCmeR(spacer)-RBS(Bgl)-sfGFP (in which the major functioning device is BBa_25UQXNE7) , by introducing a 2*TAA mutation in the second and third amino acid loci (the same change as in Figure 12). This serves as an appropriate control to pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP.
Build
PCR and Golden Gate Assembly were performed to build the correct plasmids. Agarose gel electrophoresis (AGE) was performed, and the bands indicated the correctness of the PCR products' and the assembled products' lengths (Figure 20). Similar to the building of pJ23119(mut)-CmeR_pCmeR(-35)-RBS(Bgl)-sfGFP in Cycle #3, we annealed two ssDNA primers together as the third fragment in the Golden Gate assembly of pJ23119(mut)-CmeR(2*TAA)_pCmeR(spacer)-RBS(Bgl)-sfGFP. The sequence of plasmids was verified by whole-plasmid sequencing.
Figure 20. The AGE result of the PCR products (A) and Golden Gate assembly product (B) for pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP; the AGE result of the PCR products (C) and Golden Gate assembly product (D) for pJ23119(mut)-CmeR(2*TAA)_pCmeR(spacer)-RBS(Bgl)-sfGFP. The bands indicate the correctness of the components' and the assembly product's length.
Test
The plasmid with 2*TAA mutation was transformed into E. coli Trelief 5α, and solid inoculation was performed. Since no full-length CmeR was expressed and therefore no repressor was blocking GFP expression, all colonies showed green fluorescence as expected. This shows that pCmeR's transcriptional rate is high enough without the binding of CmeR to activate downstream expression.
However, in our qualitative test of pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP, the group with TCA++ showed no induction compared to the group without TCA, and neither of the two groups showed green fluorescence (Figure 21).
Figure 21. Qualitative test of pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP using TCA. TCA (++), 1mM TCA added. TCA (-), no TCA added.
Comparing the TCA- group of the above qualitative test and the solid inoculation of the plasmid with 2*TAA mutation, we could conclude that the presence of CmeR protein could successfully bind to cmeO and block downstream gene expression. The problem, then, might lie in the interaction between CmeR and TCA.
Therefore, we set up another qualitative test of pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP using salicylate, another inducer of pCmeR. In previous characterizations of CmeR in literature, adding 1mM salicylate can increase downstream expression by 8-fold (Nasr et al., 2022). As shown in Figure 22, the tube in the presence of salicylate showed higher green fluorescence.
Figure 22. Qualitative test of pJ23119(mut)-CmeR_pCmeR(spacer)-RBS(Bgl)-sfGFP using salicylate. Salicylate (++), 1mM salicylate added. Salicylate (-), no salicylate added.
Learn
The results above confirmed several important aspects of the CmeR-pCmeR sensor system. First, as already mentioned in the "test" part, we could conclude that CmeR successfully binds to cmeO and blocks downstream gene expression. The test with salicylate as the inducer further helped us conclude that CmeR stops blocking downstream gene expression when binding to its ligands. However, although structural studies and literature indicate that TCA can bind to CmeR with high affinity and change its 3D structure (Lei, 2011), our experimental data showed that TCA was not able to turn on repression in vivo. This confirmed that the regulatory logic of our construct was correct, but that TCA itself was not available to cytoplasmic CmeR.
Taken together, we concluded that the CmeR-pCmeR system cannot be directly used as a TCA sensor inside E. coli because of TCA's poor membrane permeability, which was tested in literature (Elkins & Savage, 2003). This learning marks a turning point in our design strategy: instead of attempting further optimizations of intracellular CmeR regulation, we needed to incorporate a sensor that detects extracellular TCA.
Cycle #5
Design
The "test" and "learn" parts in Cycle #4 drove our design of the second TCA sensor, which is based on the transmembrane sensor TcpP.
TcpP is responsible for bile salt sensing in Vibrio cholerae; in the presence of ligands, TcpP dimerizes and acts as a transcriptional factor, therefore activating downstream gene expression (Xue et al., 2016). Its subcellular localization (on the membrane with its ligand-binding domain oriented towards the extracellular side) addresses the problem identified in Cycle #4 with the previous TCA sensor.
We also integrated a protective protein, TcpH, to slow down the degradation of TcpP by a RseP homolog in E. coli, as suggested by past research (Yang et al., 2013; Chang et al, 2021). This TcpP-TcpH TCA sensor has shown high compatibility to E. coli after being integrated into a previously well-characterized, widely-applicable synthetic receptor platform called EMeRALD (short for Engineered Modularized Receptors Activated via Ligand-induced Dimerization) (Chang et al, 2021). In the design of this TcpPH-EMeRALD system, since the transcriptional factor in V. cholerae might not be compatible with promoters and RNA polymerases in E. coli, the DNA-binding domain of the native TcpP was replaced by the DNA-binding domain of CadC, a membrane-bound pH sensor in E. coli that regulates the promoter pCadBA (Lindner & White, 2014). The fusion protein CadC-TcpP, composed of the ligand-binding and transmembrane domains of TcpP and the DNA-binding domain of CadC, keeps the function both as a TCA sensor and as a transcriptional factor endogenous to E. coli. This TcpPH-EMeRALD system has shown a promising sensing ability in E. coli (Chang et al, 2021).
In our plasmid design, we put CadC-TcpP and TcpH downstream of two different constitutive promoters, P9 and apFAB339, because the relative expression levels of CadC-TcpP and TcpH proteins under this combination of promoters were tested to show the optimal sensing performance (Chang et al, 2021). We placed sfGFP as the reporter protein downstream of pCadBA, which will be expressed upon binding with CadC DNA-binding domain in the presence of TCA. (Figure 23) This eventually gives rise to the major functioning device BBa_25XJ1XUG in the plasmid pTcpP-sfGFP.
Figure 23. Plasmid design of TcpP-based TCA sensor. Created by biorender.com.
Build
We ordered the CadC-TcpP and TcpH sequences from Tsingke. PCR and Golden Gate Assembly were then performed to build the correct plasmids. AGE was performed, and the bands indicated the correctness of the assembled product's length (Figure 24).
Figure 24. The AGE result of the Golden Gate assembly product for the construction of the plasmid pTcpP-sfGFP. The bands indicate the correctness of the components' and assembly product's length.
Test
We transformed both the TcpP-based TCA sensor into E. coli Trelief 5α for cloning, and into E. coli Nissle 1917 for the kinetics assay. The detailed plate setup in the assay is shown in Figure 25.
Figure 25. Plate layout of the kinetics assay for TcpP-based TCA sensor.
The results of the kinetics assay were used to calculate the fluorescence / ABS600 curve. The normalized curve of the groups and the stationary phase result indicate that fluorescence level is positively correlated with TCA concentration with low concentrations of TCA (below 125 μM). Above 125 μM, however, the fluorescence level stays the same regardless of TCA concentration. Overall, this sensor has achieved a >20-fold dynamic range. (Figure 26)
Figure 26. GFP expression of TcpP-based TCA sensor in the kinetics and the stationary phase. Fluorescence / ABS600 was used to represent GFP expression. Error bars represent SD. The red box indicates the data used in plotting the stationary phase result.
Learn
Our sensor design finally succeeded at detecting changes in TCA concentrations. Although this sensor works only at low TCA concentrations, it is sufficient to act as a TCA sensor in the feline gut, considering the small amount of bile acid diluted into the entire intestinal lumen (Washizu et al., 1990). However, there are many possible future experiments and improvements. First, we could set up a 2*TAA-mutated version of this TcpP-based TCA sensor, just like what we have done in Cycles #3-4, to test the theoretical limit of this sensor's dynamic range. Further optimizations could be done by tuning RBS strengths for the three RBS used in the plasmid, as well as engineering the positioning of different sub-components of pCadBA. Even without further engineering and optimization, this sensor has already shown its promising functionality to be integrated into our overall platform.
References
↑ Barneda-Zahonero, B., Collazo, O., Azagra, A. et al. The transcriptional repressor HDAC7 promotes apoptosis and c-Myc downregulation in particular types of leukemia and lymphoma. Cell Death Dis 6, e1635 (2015). https://doi.org/10.1038/cddis.2014.594
↑ LaFleur, T. L., Hossain, A., & Salis, H. M. (2022). Automated model-predictive design of synthetic promoters to control transcriptional profiles in bacteria. Nature Communications, 13(1). https://doi.org/10.1038/s41467-022-32829-5
↑ Lei, H. T., Shen, Z., Surana, P., Routh, M. D., Su, C. C., Zhang, Q., & Yu, E. W. (2011). Crystal structures of CmeR-bile acid complexes from Campylobacter jejuni. Protein Science: A Publication of the Protein Society, 20(4), 712–723. https://doi.org/10.1002/pro.602
↑ Lee, T.S., Krupa, R.A., Zhang, F. et al. BglBrick vectors and datasheets: A synthetic biology platform for gene expression. J Biol Eng 5, 12 (2011). https://doi.org/10.1186/1754-1611-5-12
↑ Salis, H. M., Mirsky, E. A. & Voigt, C. A. (2009). Automated design of synthetic ribosome binding sites to control protein expression. Nature Biotechnology, 27(10), 946–950. https://doi.org/10.1038/nbt.1568
↑ Nasr, M. A., Timmins, L. R., Martin, V. J. J., & Kwan, D. H. (2022). A Versatile Transcription Factor Biosensor System Responsive to Multiple Aromatic and Indole Inducers. ACS Synthetic Biology, 11(4), 1692–1698. https://doi.org/10.1021/acssynbio.2c00063
↑ Elkins, C. A., & Savage, D. C. (2003). CbsT2 from Lactobacillus johnsonii 100-100 is a transport protein of the major facilitator superfamily that facilitates bile acid antiport. Journal of molecular microbiology and biotechnology, 6(2), 76–87. https://doi.org/10.1159/000076738
↑ Xue, Y., Tu, F., Shi, M., Wu, C. Q., Ren, G., Wang, X., Fang, W., Song, H., & Yang, M. (2016). Redox pathway sensing bile salts activates virulence gene expression in Vibrio cholerae. Molecular microbiology, 102(5), 909–924. https://doi.org/10.1111/mmi.13497
↑ Yang, M., Liu, Z., Hughes, C., Stern, A. M., Wang, H., Zhong, Z., Kan, B., Fenical, W., & Zhu, J. (2013). Bile salt-induced intermolecular disulfide bond formation activates Vibrio cholerae virulence. Proceedings of the National Academy of Sciences of the United States of America, 110(6), 2348–2353. https://doi.org/10.1073/pnas.1218039110
↑ Chang, H.-J., Zúñiga, A., Conejero, I., Voyvodic, P. L., Gracy, J., Fajardo-Ruiz, E., Cohen-Gonsaud, M., Cambray, G., Pageaux, G.-P., Meszaros, M., Meunier, L., & Bonnet, J. (2021). Programmable receptors enable bacterial biosensors to detect pathological biomarkers in clinical samples. Nature Communications, 12, 5216. Programmable receptors enable bacterial biosensors to detect pathological biomarkers in clinical samples | Nature Communications
↑ Lindner, E., & White, S. H. (2014). Topology, dimerization, and stability of the single-span membrane protein CadC. Journal of molecular biology, 426(16), 2942–2957. https://doi.org/10.1016/j.jmb.2014.06.006
↑ Washizu, T., Ikenaga, H., Washizu, M., Ishida, T., Tomoda, I., & Kaneko, J. J. (1990). Bile acid composition of dog and cat gall-bladder bile. The Japanese Journal of Veterinary Science, 52(2), 423–425. https://doi.org/10.1292/jvms1939.52.423