Design and Build
sRNA designer and Machine Learning
The first stage of the process is the design of the sequences, for this we used our software, “sRNA designer” to evaluate the likelihood of effective binding between our sRNA and the mRNA target, for this, the program starts by taking into consideration the most conserved regions within the gene we want to silence, and proceed to generate all the possible sequences available, additionally we compare it´s structural characteristics with a true sRNA as reported in literature and use thermodynamic parameters to evaluate this data to screen the sequences that posses the best results.
Figure 1.
Our team has made an important contribution since there are no software tools for sRNA designing in order to target a specific gene. The actual protocols for sRNA design mainly takes into account the targeting of one specific region as the main criteria for a given sRNA, and if used thermodynamic properties these are calculated using external tools. Our software provides a simple and elegant solution for the design of these constructs
On the other hand Machine learning is used as system that will use data provided by many sourcers to, once processed, be able to predict the up or downregulation of both sRNAs using a neural network and a pipeline such as below.
Figure 2.
gblocks assembly and cloning procedures
gBlocks are flanked with a set of M13 primers to amplificate the parts and obtain more DNA by this method, then the biobrick prefix and suffix sequences have restriction sites for EcoRI, XbaI, SpeI, PstI respectively allowing for an easier cloning procedure. We added forward and reverse Axtl primers for colony PCR of the parts once they clone into their respective cassettes.
Figure 3.
As the backbones, we selected PSB1A3 and PSB1C3 which lack an F1 ori needed for the sRNA package into bacteriophages, which are our selected method of delivery, as a consequence we added it downstream so we our sRNAs could be package by the M13 bacteriophage.
The HFQ binding domain was added in tandem with the sRNA so the protein could bind to the sRNA-mRNA complex and degrade the mRNA. The lac promoter was used in all the constructs intended to be used in E. coli together with the T1 terminator.
The genes selected for silencing are the following:
- CAM-R: This gene provides resistance against chloramphenicol.
- AMP-R: This gene provides resistance against ampicillin.
- RFP: This gene produces the recombinant red fluorescent protein.
- Type IV Pilus: A virulence factor in Pseudomonas aeruginosa responsible for attachment to host cells and biofilm formation.
- Mutant gyrA: This mutant gene has been identified in ciprofloxacin resistant Pseudomonas aeruginosa and plays a vital role in intrinsic resistance development.
For each gene a total of 4 constructs were designed, except for mutant gyrA and Type IV Pilus. We chose not to clone an sRNA designed for the same resistance gene found in the backbone in order to avoid possible interference.
All CAM-R constructs were digested using the EcoRI and SpeI enzymes and cloned into PSB1A3-RFP digested with EcoRI and Xba. Xba and SpeI have compatible sticky ends and therefore are able to ligate due to complementarity. In this case the RFP sequence was kept intact for it to work as a reporter gene.
All AMP-R constructs were digested using the EcoRI and SpeI enzymes and cloned into PSB1C3-RFP digested with EcoRI and Xba. In this case the RFP sequence was kept intact for it to work as a reporter gene.
The RFP constructs were digested using the EcoRI and PstI enzymes and cloned into PSB1A3-RFP digested with the same enzymes. In this case the RFP sequence was removed in the plasmid digestion to avoid interference with the sRNA.
The gyrA and Type IV Pili constructs were digested using the EcoRI and SpeI enzymes and cloned into PSB1A3-RFP digested with EcoRI and Xba. In this case the RFP sequence was kept intact for it to work as a reporter gene.
This double sRNA construct was digested with EcoRI and Spei enzymes and cloned into PSB1A3-RFP digested with EcoRI and Xba. The RFP sequence was kept intact for it to work as a reporter gene.
Figure 4.
Phage display of the Pf1 Protein
With the objective of infecting Pseudomonas aeruginosa, we designed a modification on the G3P protein of the M13 bacteriophage, which only infects E. coli. This bacteriophage has been used extensively for phage display and many chemical and genetic modification methods have been described [1].
The G3P protein is responsible for the penetration of the viral genome via F-pilus retraction. The C-terminal domain of G3P mediates the release of the membrane-anchored virion from the host cell. The N-terminal domain is responsible for the F-pilus recognition and interaction with the entry receptor tolA inducing the penetration of viral DNA [2].
The filamentous bacteriophage Pf1 is capable of infecting Pseudomonas aeruginosa strain K. The structure of this bacteriophage is very similar to the M13, and both have the minor coat protein G3P both serving as pilus absorbing proteins. G3P from Pf1 bacteriophage interacts with the type IV PAK pilus and there is evidence of a conserved pathway of infection between Pf1 and M13 bacteriophages [3].
While there is no similarity in the amino acid sequence between the G3P from Pf1 and M13 its C-terminal and N-terminal domains have similar activities, with the C-terminal domain being attached to the rest of the virion and the N-terminal domain being responsible for pilus recognition and viral DNA penetration induction [3].
Figure 5.
Modification of the G3P needs a linker sequence rich in glycine to improve stability and avoid undesired interactions between native and added peptides [1]. We designed a construct that keeps the C-terminal domain of the M13 G3P protein and replaces the N-terminal domain with one of the Pf1 bacteriophage. The purpose of this construct is to create an M13 bacteriophage with the ability to recognize and bind to the type IV PAK pilus and infect Pseudomonas aeruginosa strain K.