Team:Evry Paris-Saclay/Contribution
Figure 1. Toehold switches principle (schematic inspired from iGEM Athens 2018).
## Contribution n°1: information learned from literature about the promoter we used in all our studies, the T7 promoter with GGG at 3' end
To drive the expression of our composite parts containing the rosewood switches or triggers, we used the strong promoter from T7 bacteriophage (taatacgactcactata) with GGG at 3' end.
The T7 RNA polymerase initiates the transcription at the first guanidine of this stretch of three G and it was shown that +1 GGG is one of the best +1, +2 and +3 base combinations at the transcription initiation for enhanced promoter strength [[2]](#ref2).
Twelve twin sequences of this promoter were found in the Parts Registry (BBa_G00113, BBa_K923003, BBa_K2150031, BBa_K2066537, BBa_K2310001, BBa_K2398556, BBa_K2290003, BBa_K2346008, BBa_K2512009, BBa_K2816005, BBa_K3172100, BBa_K3222000). We did not add this promoter to the Parts Registry again, but improved the annotations of all of them and chose only BBa_K2150031 as "our promoter" as it was the first part to be added in Parts Registry as a promoter (some other parts are not primer, simple DNA or phage).
## Contribution n°2: information learned from literature about some terminators
For efficient expression, the promoter and the terminator need to be compatible. As we chose to use the strong promoter from T7 bacteriophage (taatacgactcactata) with GGG at the 3' end for the expression of our parts, a terminator recognised by the T7 RNA polymerase is required.
The T7 Tphi transcriptional terminator is an obvious choice, but we finally settled on the SBa_000587 synthetic terminator (BBa_K3453000) as it was shown to be stronger than the wild type T7 Tphi terminator [[3]](#ref3).
Another possible choice would have been the BBa_B0015 double terminator which is one of the most used parts in the registry, being included in nearly 5000 composite parts in October 2020 and having 24 twins. BBa_B0015 is a composite terminator made by joining 2 other terminators, one derived from E. coli (BBa_B0010) and the other from the T7 phage (BBa_B0012). Unlike what one can guess from the name and origin, the E. coli terminator can terminate transcription by T7 RNA polymerase, but not the T7 TE terminator. Indeed, BBa_B0010 is the E. coli rrnB T1 terminator which was shown to be an efficient terminator for the E. coli RNA polymerase, but also for the phage SP6 and T7 RNA polymerases through two different mechanism: one involving an upstream hairpin structure and the other one a downstream sequence-specific signal [[4–6]](#ref4). However, the T7 TE terminator (BBa_B0012) is located at the end of the T7 DNA ligase gene which is in the early region of bacteriophage T7 genome [[7]](#ref7). This terminator is an efficient one for the E. coli RNA polymerase, but not for the T7 RNA polymerase [[8]](#ref8).
## Contribution n°3: new data from laboratory experiments on a toehold switch designed by a previous iGEM team: BBa_K2916068 and design of new part to better understand the system
As a reference system for toehold switches design and characterization, we decided to use a toehold switch designed by a previous iGEM team, and not one from a commercial source or published literature.
We choose BBa_K2916068, a toehold switch build by the EPFL 2019 team, for several reasons:
- it is specific for another plant species, the grapevine and not for a pathogen or a mammalian gene
- its reporter gene is sfGFP, the same as that for our rosewood biosensors
### Design
During the design phase of our project, we analysed BBa_K2916068 and realised that there wasn’t sufficient data from lab experiments about this part both on Parts Registry and on [the wiki of EPFL 2019 iGEM team](https://2019.igem.org/Team:EPFL). Our observations about this part are detailed below, together with the new parts we designed and characterized in order to better understand the system.
#### Observation n°1: the triggers
The sequence of the trigger used to activate translation of sfGFP from BBa_K2916068 is not available in Parts Registry (as is the case for all triggers designed by the EPFL 2019 iGEM team).
However, the [“Amplification” page of the EPFL 2019 wiki](https://2019.igem.org/Team:EPFL/Amplification) describes how the triggers were designed. The grapevine trigger was PCR-amplified using as template the Vitis vinifera TrnL-UAA and the forward primers EC_fwd1 or EC_fwd1_T7 and the EC_rev1 reverse primer (it should be noted that the sequences of all reverse primers listed in the [additional materials](https://2019.igem.org/wiki/images/8/8a/T--EPFL--additional_materials_v1.pdf) of this page are given as 3’-5’ sequences and not in the conventional 5’-3’ orientation).
We contacted the EPFL 2019 iGEM team, who kindly sent us a grapevine trigger sequence. However, this sequence was shorter than the one we deduced from [their wiki](https://2019.igem.org/Team:EPFL). The two trigger sequences are now available in the Parts Registry as BBa_K3453081 and BBa_K3453083.
#### Observation n°2: the ATG start codon and the BioBrick RFC[10] scar
As depicted in Figure 1, a toehold switch blocks translation initiation because the RBS and the ATG start codon are embedded in the middle of a hairpin structure maintained by base pairs before and after the start codon. The hairpin can be unfolded upon binding of a trigger RNA thereby exposing the RBS and the ATG start codon and thus permitting translation of the reporter protein. For this to happen, two important design precautions have to be taken: (i) the start codon should be in frame with the reporter gene, and (ii) there should be no in-frame stop codon before the reporter gene.
BBa_K2916068 is a composite part formed by BBa_K2916050, BBa_I746916, BBa_B0010 and BBa_B0012 and was specified with an RFC[10] scar between the parts. The RFC[10] scar TACTAG between the BBa_K2916050 (T7 promoter + grapevine switch 1.1) and BBa_I746916 (sfGFP) introduces a stop codon and leads to premature termination of translation (and thus no expected sfGFP production).
To circumvent this problem, we constructed BBa_K3453170 in which the RFC[10] scar TACTAG was removed by site directed mutagenesis on BBa_K2916068.
In addition and as a control for our experiments on rosewood toehold switches, we also made BBa_K3453171. In this part BBa_K2916050 (T7 promoter + grapevine switch 1.1) was placed upstream of an sfGFP fused to the LVA degradation tag. LVA degradation tag (AANDENYALVA) is an ssrA tag that accelerates protein degradation in E. coli at 37°C [[9]](#ref9) and is used in synthetic biology to rapidly degrade reporter genes and better observe the synthetic gene networks dynamics. Moreover, in our new part BBa_K3453171, the transcription termination was controlled by the strong SBa_000587 synthetic terminator for the T7 RNA polymerase [[3]](#ref3).
#### Observation n°3: the trigger binding site
The role of the trigger RNA is to bind to the toehold switch and thus unfold the hairpin structure thereby exposing the RBS and the ATG start codon and thus permitting translation of the reporter protein.
The binding between the trigger and the switch is dictated by RNA-RNA interactions between complementary base pairs.
In BBa_K2916050 (which is the T7 promoter together with grapevine switch 1.1 used to build the composite part BBa_K2916068), the trigger binding region is in the same direction as in the grapevine genome and of the trigger sequences derived from BBa_K3453081 and BBa_K3453083. As a result, this trigger / switch pair does not interact and is not functional.
However, the hairpin structure of the grapevine switch 1.1 is predicted to be unfolded by the reverse complements of the two triggers BBa_K3453082 and BBa_K3453084 (Table 1).
#### Positive and negative controls
As positive controls for the above described sfGFP expression cassettes, we designed three parts, all harbouring the sfGFP-LVAtag under the control of T7 promoter and the strong SBa_000587 synthetic terminator, but with different RBSes :
- In BBa_K3453104, the sfGFP-LVAtag was equipped with a custom made RBS (BBa_K2675017).
- In BBa_K3453105, the sfGFP-LVAtag was equipped with the synthetic stem-loop containing an RBS designed by Pardee *et al*. [[10]](#ref10) to be part of a standardized toehold switch (BBa_K3453005).
- In BBa_K3453172, the sfGFP-LVAtag was equipped with a truncated grapevine switch 1.1 (BBa_K3453072). The truncation in BBa_K3453072 compared to the grapevine switch 1.1 concerns the trigger binding site which is absent. Thus this part mimics the open (unfolded) state of this toehold switch.
As negative control, we used BBa_K3453103 in which no promoter and no RBS are present upstream of sfGFP-LVAtag.
### Build
The trigger sequences (Table 1) were placed under the control of the T7 promoter and followed by the strong SBa_000587 synthetic terminator and the resulting transcriptional units were synthesized and assembled in the high copy plasmid pSB1C3.
The sfGFP expression cassettes are summarised in Table 2:
- BBa_K2916068 was synthesized and assembled in the low copy plasmid pSB3T5
- BBa_K3453170 we constructed by site directed mutagenesis on BBa_K2916068 in pSB3T5 backbone according to Single-Primer Reactions IN Parallel (SPRINP) protocol [[11]](#ref11) with the primers 5’-GCTGGCGGCAGCGCAAAAGATGCGTAAAGGCGAAGAGC-3’ and 5’-GCTCTTCGCCTTTACGCATCTTTTGCGCTGCCGCCAGC-3’.
- All other constructs were assembled by Golden Gate in the low copy plasmid pSB3T5. For this, fragments containing the T7 promoter followed by different toehold switches and RBS were synthesized with flanking type IIS restriction sites, and then interested into one of the acceptor vectors BBa_K3453101, BBa_K3453102 or BBa_K3453103 in which Golden Gates adapters are present upstream of sfGFP-LVAtag with either BsaI, BsmBI or BbsI sites, respectively.
Table 1. Grapevine triggers.
| Grapevine trigger label | Part number | Transcriptional units’ part numbers |
|---|---|---|
| Grapevine trigger short | BBa_K3453081 | BBa_K3453181 |
| Grapevine trigger short reverse | BBa_K3453082 | BBa_K3453182 |
| Grapevine trigger long | BBa_K3453083 | BBa_K3453183 |
| Grapevine trigger long reverse | BBa_K3453084 | BBa_K3453184 |
| sfGFP expression cassettes’ part numbers | Description |
|---|---|
| BBa_K2916068 | T7 promoter + Grapevine Switch 1.1 + scar + sfGFP + scar + B0015 terminator |
| BBa_K3453170 | BBa_K2916068 without the scar between the Grapevine Switch 1.1 and sfGFP |
| BBa_K3453171 | T7 promoter + Grapevine Switch 1.1 + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453172 | T7 promoter + truncated Grapevine Switch 1.1 (no trigger binding site) + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453104 | T7 promoter + custom made RBS (BBa_K2675017) + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453105 | T7 promoter + standard toehold RBS stem-loop (BBa_K3453005) + sfGFP-LVAtag + SBa_000587 terminator |
Figure 2. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switch 1.1 in BBa_K2916068 and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least three measurements on independent biological replicates.
However, a low expression was observed when the BBa_K34533170 was tested in the presence of the two grapevine triggers, both in reverse direction (Figure 3). This confirms that the stop codon introduced by the RFC[10] scar TACTAG plays an important negative role in the function of the toehold switch, as the translation initiated at the AUG codon embedded in the hairpin structure is prematurely stopped before sfGFP translation. Moreover, we can clearly observe that the grapevine switch 1.1 can be activated by a trigger, but only if it is expressed in a reverse complement orientation.
Figure 3. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switch 1.1 in BBa_K34533170 and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least three measurements on independent biological replicates.
Surprisingly, a low sfGFP expression was observed in the positive control BBa_K34533172, where the sfGFP-LVAtag was equipped by a truncated grapevine switch 1.1 (BBa_K3453072) (Figure 4). The truncation in BBa_K3453072 compared to the grapevine switch 1.1 concerns the trigger binding site which is absent. Thus this part mimics the open (unfolded) state of this toehold switch. This suggests that the sequence upstream of the natural ATG of sfGFP has a negative influence on its expression. A second round of the Design-Build-Test-Learn (DBTL) cycle (see below) will uncover (some of) the reasons. One can also notice that the sfGFP-LVAtag expression in this positive control BBa_K34533172 was observed when no trigger sequence was present, which is expected as the truncated grapevine switch 1.1 contains only an RBS and a few codons in frame with the sfGFR-LVAtag gene.
Figure 4. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switch 1.1 in BBa_K3453172 and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least three measurements on independent biological replicates.
All these results (Figures 2, 3 and 4) are presented grouped in Figure 5 for easy comparison
Figure 5. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switches (Table 2) and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105 . The data and error bars are the mean and standard deviation of at least three measurements on independent biological replicates.
### DBTL cycle, round 2
### Design
As evidenced in Figure 5, the sequence upstream of the natural ATG of sfGFP in BBa_K34533172 has a negative influence on its expression. The same sequence is present in all BBa_K2916050 (T7 promoter + grapevine switch 1.1) derived parts analysed above BBa_K2916068, BBa_K34533170, BBa_K34533171 , BBa_K34533172). Indeed, translation initiation rates (Figure 6) predicted using the Salis lab RBS Calculator v2.1 [[12–15]](#ref12) show a lower rate for BBa_K34533172 compared to both positive controls BBa_K34533104 and BBa_K3453315.
Figure 6. Translation initiation rates predicted with the Salis' RBS Calculator 2.1 [12–15].
The different features of BBa_K2916050 presented in Table 3 may be responsible for this behaviour and among them the most obvious are the mRNA secondary structure and the RBS strength.
Table 3. Features of grapevine switch 1.1 and its derivatives 1.1.2 and 1.1.3 and the cognate truncated versions.
| Part number | Part name | Trigger binding site (a fragment) | Trigger binding site (b fragment) + 3nt | RBS stem-loop till ATG start codon | b’ fragment + TAT | Linker |
|---|---|---|---|---|---|---|
| BBa_K2916050 | Grapevine Switch 1.1 | CAGAAAACGATAATAAAAAAAGGA | TAGGTGCAGAGAC | CACTTTAAGACATGTAGAGGAGATTAAAGATG | TCTCTGCACCTA | CAGGTGCTGGCGGCAGCGCAAAAG |
| BBa_K3453072 | Truncated Grapevine Switch 1.1 (no trigger binding site) | - | - | CACTTTAAGACATGTAGAGGAGATTAAAGATG | TCTCTGCACCTA | CAGGTGCTGGCGGCAGCGCAAAAG |
| BBa_K3453073 | Grapevine Switch 1.1.2 | CAGAAAACGATAATAAAAAAAGGA | TAGGTGCAGAGAC | CACTTTAAGACATGTAGAGGAGATTAAAGATG | TCTCTGCACCTA | AACCTGGCGGCAGCGCAAAAG |
| BBa_K3453075 | Truncated Grapevine Switch 1.1.2 (no trigger binding site) | - | - | CACTTTAAGACATGTAGAGGAGATTAAAGATG | TCTCTGCACCTA | AACCTGGCGGCAGCGCAAAAG |
| BBa_K3453076 | Grapevine Switch 1.1.3 | CAGAAAACGATAATAAAAAAAGGA | TAGGTGCAGAGAC | GGACTTTAGAACAGAGGAGATAAAGATG | TCTCTGCACCTA | CAGGTGCTGGCGGCAGCGCAAAAG |
| BBa_K3453078 | Truncated grapevine switch 1.1.3 (no trigger binding site) | - | - | GGACTTTAGAACAGAGGAGATAAAGATG | TCTCTGCACCTA | CAGGTGCTGGCGGCAGCGCAAAAG |
Figure 7. Secondary-structure prediction of the Grapevine switch 1.1 with the ATG of the reporter gene. The prediction was realised using the NUPACK web server [[16]](#ref16) and graphically represented using the forna RNA secondary structure visualization tool [[18]](#ref18). Nucleotides were coloured to match the different segments in Figure 1. The secondary structure of BBa_K2916050 (without the T7 promoter) in dot-bracket notation is: ........................(((((((((((((..((((((....(.....)....))))))..)))))))))))))...((((((....)))))).....
RBS strength may also be may be responsible for the low sfGFP expression observed from all BBa_K2916050 (T7 promoter + grapevine switch 1.1) derived parts analysed during our 1st round or the DBTL cycle (BBa_K2916068, BBa_K34533170, BBa_K34533171, BBa_K34533172). To investigate this possibility we modified the RBS stem-loop region of BBa_K2916050 and transformed it into the ‘standard’ RBS stem-loop region (BBa_K3453005) of the Series B of toehold switch sensors for Zika virus detection [[10]](#ref10) and of the BioBits™ toeholds [[17]](#ref17) that was used in the positive control BBa_K3453105. The resulting part BBa_K3453076 was denoted the Grapevine Switch 1.1.3 and its features are presented in Table 3. A truncated version lacking the trigger binding site (BBa_K3453078) was also designed to mimic the open (unfolded) state of this toehold switch. It’s predicted translation initiation rate in the composite part BBa_K3453078 (Figure 6) was higher compared with the truncated Grapevine Switches 1.1 and 1.1.2.
### Build
The sfGFP expression cassettes containing the toehold switches listed in Table 3 are summarised in Table 4:
- BBa_K3453173, BBa_K3453174 and BBa_K3453175 were constructed by PCR with the primers 5'-ATACGTCTCACCTGGCGGCAGCGCAAAAGATG-3' and 5'-ATACGTCTCACAGGTTTAGGTGCAGAGACATCTTTAATCTCC-3' on BBa_K3453170, BBa_K3453171 and BBa_K3453172 respectively followed by a Golden Gate self circularisation in the presence of BsmBI
- BBa_K3453176 and BBa_K3453177 were constructed by PCR with the primers 5'-ATAGGTCTCACAGAGGAGATAAAGATGTCTCTGCACC-3' and 5'-ATAGGTCTCATCTGTTCTAAAGTCCGTCTCTGCACCTATCCTTTTTTTATTATCG-3' on BBa_K3453170 and BBa_K3453171 respectively followed by a Golden Gate self circularisation in the presence of BsaI
- BBa_K3453178 was constructed by PCR with the primers 5'-ATAGGTCTCACAGAGGAGATAAAGATGTCTCTGCACC-3' and 5'-ATAGGTCTCATCTGTTCTAAAGTCCCCCTATAGTGAGTCGTATTAC-3' on BBa_K3453172 followed by a Golden Gate self circularisation in the presence of BsaI
Table 4. Grapevine switches and controls designed in the 2nd round of the DBTL cycle.
| sfGFP expression cassettes’ part numbers | Description |
|---|---|
| BBa_K3453173 | T7 promoter + Grapevine Switch 1.1.2 + sfGFP + Scar + B0015 terminator |
| BBa_K3453174 | T7 promoter + Grapevine Switch 1.1.2 + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453175 | T7 promoter + truncated Grapevine Switch 1.1.2 (no trigger binding site) + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453176 | T7 promoter + Grapevine Switch 1.1.3 + sfGFP + scar + B0015 terminator |
| BBa_K3453177 | T7 promoter + Grapevine Switch 1.1.3 + sfGFP-LVAtag + SBa_000587 terminator |
| BBa_K3453178 | T7 promoter + truncated Grapevine Switch 1.1.3 (no trigger binding site) + sfGFP-LVAtag + SBa_000587 terminator |
Figure 8. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the truncated grapevine toehold switches (Tables 2 and 4) and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least four measurements on independent biological replicates.
However, when the complete versions of the Grapevine Switches 1.1, 1.1.2 and 1.1.3 were controlling the sfGFP expression, no significant increase was observed (Figure 9). But, when using the sfGFP-LVAtag version, we were able to detect a 5 fold increase of MEFL/particle values (Figure 10) upon disruption of the small stem-loop structure in the linker region (compare the expression of sfGFP under the control of the Grapevine Switches 1.1 and 1.1.2 in BBa_K3453171 and BBa_K3453173 respectively).
Figure 9. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switches + sfGFP (Tables 2 and 4) and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least four measurements on independent biological replicates.
Figure 10. In vivo characterization of sfGFP expression by E. coli BL21 Star™(DE3) cells harbouring the grapevine toehold switches + sfGFP-LVAtag (Tables 2 and 4) and triggers (Table 1). The negative controls have been performed with an empty pSB3T5, pSB1C3 (no trigger) and BBa_K3453103 (no promoter, no RBS) and the positive controls with BBa_K3453104 and BBa_K3453105. The data and error bars are the mean and standard deviation of at least four measurements on independent biological replicates.
### Conclusion
We have characterized the function of BBa_K2916068, an sfGFP expression cassette under the control of a toehold switch built by a previous iGEM team. While doing so, we added to parts registry its cognate trigger without which this part cannot be tested. We also uncovered (some) of the reasons for its malfunction like the presence of an inframe stop codon in the reporter gene and successfully designed new parts that increase the fluorescence signal. Thus, BBa_K3453170, BBa_K3453171, BBa_K3453173, BBa_K3453174, BBa_K3453176 and BBa_K3453177 are improvements of BBa_K2916068.
However, all these six improved parts (BBa_K3453170, BBa_K3453171, BBa_K3453173, BBa_K3453174, BBa_K3453176 and BBa_K3453177) are functional in the presence of a trigger artificially expressed in E. coli in reverse orientation. They will most likely not be able to detect the presence of an RNA naturally expressed in grapevine because the trigger binding site is not in a compatible orientation. But that is another story, and we invite the reader to discover it on the [“Improvement”](https://2020.igem.org/Team:Evry_Paris-Saclay/Improvement) page of this wiki.
## Contribution n°4: information learned from sequence analysis about a collection of toehold switches BBa_K2916050 to BBa_K2916067 designed by a previous iGEM team
Following the sequence analysis and the experimental results described above under the sub-chapter “Contribution n°3”, we wondered if this was an exception, or the same design principles were applied to the other toehold switches designed by the same iGEM team and listed on [their wiki](https://2019.igem.org/Team:EPFL/Parts).
A “quick” look revealed that all 18 toehold switches (BBa_K2916050 to BBa_K2916067) have the trigger binding site in the same orientation as in the transcriptional orientation of their target genes (TrnL-UAA gene of grapevine, PlsC gene of Bois noir and 16S gene of Flavescence Doree). We improved the annotations of all these 18 parts by adding a note stating that it is unlikely that they will be able to detect the presence of an RNA naturally expressed in a cell because the trigger binding site is not in a compatible orientation.
Moreover, all composite parts containing expression cassettes of either sfGFP or catechol 2,3-dioxygenase (CDO) reporter genes (BBa_K2916068 to BBa_K2916086) were specified with an RFC[10] scar between parts. The RFC[10] scar TACTAG between the switch and the downstream gene introduces a stop codon and leads to premature termination of translation (and thus no reporter production). We improved the annotations of all these 18 composite parts too.
## References
[1] Green AA, Silver PA, Collins JJ, Yin P. Toehold switches: de-novo-designed regulators of gene expression. Cell (2014) 159: 925–939.
[2] Imburgio D, Rong M, Ma K, McAllister WT. Studies of promoter recognition and start site selection by T7 RNA polymerase using a comprehensive collection of promoter variants. Biochemistry (2000) 39: 10419–10430.
[3] Temme K, Hill R, Segall-Shapiro TH, Moser F, Voigt CA. Modular control of multiple pathways using
engineered orthogonal T7 polymerases. Nucleic Acids Research (2012) 40: 8773–8781.
[4] Christiansen J. The 9S RNA precursor of Escherichia coli 5S RNA has three structural domains: implications for processing. Nucleic Acids Research (1988) 16: 7457–7476.
[5] Hartvig L, Christiansen J. Intrinsic termination of T7 RNA polymerase mediated by either RNA or DNA. The EMBO journal (1996) 15: 4767–4774.
[6] Kwon YS, Kang C. Bipartite modular structure of intrinsic, RNA hairpin-independent termination signal for phage RNA polymerases. The Journal of Biological Chemistry (1999) 274: 29149–29155.
[7] Dunn JJ, Studier FW. Complete nucleotide sequence of bacteriophage T7 DNA and the locations of T7 genetic elements. Journal of Molecular Biology (1983) 166: 477–535.
[8] Dunn JJ, Studier FW. The transcription termination site at the end of the early region of bacteriophage T7 DNA. Nucleic Acids Research (1980) 8: 2119–2132.
[9] Purcell O, Grierson CS, Bernardo M di, Savery NJ. Temperature dependence of ssrA-tag mediated protein degradation. Journal of Biological Engineering (2012) 6: 10.
[10] Pardee K, Green AA, Takahashi MK, Braff D, Lambert G, Lee JW, Ferrante T, Ma D, Donghia N, Fan M, Daringer NM, Bosch I, Dudley DM, O’Connor DH, Gehrke L, Collins JJ. Rapid, low-cost detection of Zika virus using programmable biomolecular components. Cell (2016) 165: 1255–1266.
[11] Edelheit O, Hanukoglu A, Hanukoglu I. Simple and efficient site-directed mutagenesis using two single-primer reactions in parallel to generate mutants for protein structure-function studies. BMC biotechnology (2009) 9: 61.
[12] Salis HM, Mirsky EA, Voigt CA. Automated design of synthetic ribosome binding sites to control protein expression. Nature Biotechnology (2009) 27: 946–950.
[13] Espah Borujeni A, Channarasappa AS, Salis HM. Translation rate is controlled by coupled trade-offs between site accessibility, selective RNA unfolding and sliding at upstream standby sites. Nucleic Acids Research (2014) 42: 2646–2659.
[14] Espah Borujeni A, Salis HM. Translation initiation is controlled by RNA folding kinetics via a ribosome drafting mechanism. Journal of the American Chemical Society (2016) 138: 7016–7023.
[15] Espah Borujeni A, Cetnar D, Farasat I, Smith A, Lundgren N, Salis HM. Precise quantification of translation inhibition by mRNA structures that overlap with the ribosomal footprint in N-terminal coding sequences. Nucleic Acids Research (2017) 45: 5437–5448.
[16] Zadeh JN, Steenberg CD, Bois JS, Wolfe BR, Pierce MB, Khan AR, Dirks RM, Pierce NA. NUPACK: Analysis and design of nucleic acid systems. Journal of Computational Chemistry (2011) 32: 170–173.
[17] Huang A, Nguyen PQ, Stark JC, Takahashi MK, Donghia N, Ferrante T, Dy AJ, Hsu KJ, Dubner RS, Pardee K, Jewett MC, Collins JJ. BioBits™ Explorer: A modular synthetic biology education kit. Science Advances (2018) 4: eaat5105.
[18] Kerpedjiev P, Hammer S, Hofacker IL. Forna (force-directed RNA): Simple and effective online RNA secondary structure diagrams. Bioinformatics (2015) 31: 3377–3379.