Team:DTU-Denmark/Design Promoter

Promoter Design

During the design of the promoters we had many considerations as to making them as useful as possible. One of the highest priorities for our project was to ensure that our promoter parts could be used by others as well.

Compatability with cloning standards

Our software was designed to produce promoters that would be compatible with a large range of cloning systems and be easy to de novo synthesize.

We have designed the LEAP promoters not just to be compatible with the new iGEM Type IIs RFC[1000] standard, but with a range of other assembly standards as well. The domestication of promoters was done by making a prioritized list of standards to be compatible with. Using this prioritized list, the scoring system seen in table 1 was implemented into the software used to design the promoters.
This scoring system resulted in a large number of the promoters in the final promoter library being compatible with some of the most widely used Type IIs standards, as seen under the design considerations sections on the pages of their respective parts.

Table 1: Standards and associated weights for making the synthetic promoters usable in the aforementioned standards.
Enzyme/Cloning system Penalty Justification
SwaI enzyme 2 SwaI domestication can be used for linearisation of vectors for genomic integration. This is something we have considered working on in the near future
MoClo 1.5 MoClo (level 0) is a widely used assembly standard, and have previously formed the basis of other standards such as the iGEM Phytobrick standard
Mobius 1 Mobius is a new assembly standard using the rare cutter AaRI along with the BsaI enzyme also used in MoClo[1]
GoldenBraid 0.3 An alternative to MoClo, but seem to mostly be used in plant synthetic biology
Chi 0.2 E.coli recombination hotspot. Removal makes long term maintenance marginally more convenient.

Design rules

For comparison, we have analysed a range of 15 different native promoters from Aspergillus spp. genes that our model has based the synthetic promoters upon. For 8 of those native promoters, we found one or more Type IIs sites from either BsaI or SapI, making these native promoters incompatible with the RFC[1000] standard. Domestication of these native promoters is not trivial, as we might ruin transcription factor binding sites when changing the nucleotide sequence of the promoter. We can get arround this by including the domestication of the synthetic promoters in the design from the very beginning and thereby produce the best performing promoters that are still compatible with the different assembly standards.

We have also designed the software to produce promoters that are easily synthesizable by de novo synthesis. This was done by creating some design rules with the algorithm that would keep the global GC% content of the promoters within a certain threshold (For the LEAP promoters, the threshold was set between 25% and 65% GC content), as well as prevent the promoters from having A/T or G/C homopolymers. This threshold can be changed depending on the constraints of the specific situation.

Promoter characteristics

With the design rules for the synthetic promoters described above, it was possible to look at designing promoters with different characteristics. According to our feedback from Novozymes and Zymergen, a useful promoter characteristic for large scale productions is the ability to have a low gene expression during exponential growth while having a strong gene expression in stationary phase. This inspired us to look for promoters with different dynamics, that would be active under different phases of cell growth. By looking at RNA-seq data from A.niger in exponential vs stationary phase, we were able to identify genes that had low activity in the exponential phase and high activity in the stationary phase, and vice versa for additional promoter coverage. We expected the synthetic promoters based on these selected genes to be active during the same phases of cell growth as the native genes they were based on.

Table 2: The table show examples of genes showing interresting qualities based on respective RNA-seq data. These among others were therefore chosen as starting points for our model.
Gene name (systematic) Reasoning
mstA Should be high constitutive but regulated by sugar availability
gpdA A strong constitutive promoter that is commonly used in industry
glaA High activity in exponential phase, low activity in stationary phase
hfbD Should be most active in stationary phase

Introducing noise

In order to fulfill the last major requirement for the synthetic promoters, we decided that the LEAP library should contain promoters of different strengths. Rather than having the strength of the promoter being exclusively determined by the strengths of the native promoters, on which they were modeled, we wanted to test the hypothesis that we could create some weaker promoter variants of the same synthetic promoter. As explained on our modeling page, this was ultimately done by introducing varying levels of stochastic noise to the consensus promoter. By designing promoters from the same native promoter, but with different strengths, the promoters should be able to express the same reporter in different quantities, and simultaneously maintain the same characteristics with regards to activity in different growth phases.



[1] A. I. Andreou and N. Nakayama, “Mobius assembly: A versatile golden-gate framework towards universal DNA assembly,” PLoS One, 2018.

The logos of our three biggest supporters, DTU Blue Dot, Novo Nordisk fonden and Otto Mønsted fonden The logos of all of our sponsors, DTU, BioNordica, Eurofins Genomics, Qiagen, NEB New England biolabs, IDT Integrated DNA technologies and Twist bioscience>