Team:Aix-Marseille/Design
Before beginning our project on S.cerevisiae, it is crucial to verify the enzymatic activity of our PUL proteins and to prove that our engineering is workable. This is why we decided to break down the project into two successive engineering phases. In the first phase, presented to the iGEM 2020 competition, we design a proof-of concept system where an engineered E.coli W3110 strain would secrete the PUL enzymes, allowing rhamnose production from ulvan directly into the culture medium. Because of COVID19 situation, we could not have access to the lab and focussed our work on the design of the genetic circuit in E.coli, and on a biosensor E.coli strain to carry our rhamnose biosensing system.
1) THE 4 ENZYMES OF INTEREST
For the proof of concept, we would use a E.coli W3110 mutant for rhaT et rhaB strain (unable to metabolize rhamnose). Our design is composed of 4 genes of Formosa Agariphila which code for enzymes that can degrade the polysaccharides of the wall of Ulves.sp : Ulvan. Those genes are regrouped in a Polysaccharide Utilization Loci (PUL) in Formosa agariphila but we would choose to start our construction with 4 of these ones because our goal is to degrade mostly the polysaccharides into rhamnose like it’s explain in our project description.
Two of those genes code endo-acting ulvan lyase so they generate the same product and can have a synergistic function together. In Formosa agariphila, both enzymes are secreted via a type IX secretion system, and are active in the extracellular medium.
Figure 1: diagram representing the enzymatic activity of our proteins of interest on ulvanes (sulphated polysaccharides composing the ulves). The first step is performed by the two ulvanes lyases (P30_PL28 and P10_PLnc) which cleaves the glycosidic linkage between Rha3S and GlcA or IduA. The second step performed by the glycosyl hydrolase is the cleavage of the unsaturated uronic acid at the non-reduced end of the oligosaccharides. The last step is the desulfatation of L-rhamnose at the 3-position. So, finally thanks to our enzymes we can obtain monomers of rhamnose.
- P30_PL28: Endo-acting ulvan lyase : This gene codes for an enzyme which catalyze rapid dissolution of insoluble ulvan : this enzyme cleaves the glycosidic linkage between Rha3S and GlcA or IduA, producing oligosaccharides with unsaturated uronic acid at the non-reducing end.
- P10_PLnc: Endo-acting ulvan lyase : the enzyme coded by this gene dissolves soluble ulvan oligomers at the cell surface. The localisation of this protein is not sure for now between outer membrane or periplasmic. This enzyme cleaves also the glycosidic linkage between Rha3S and GlcA or IduA, producing oligosaccharides with unsaturated uronic acid at the non-reducing end.
- P33_GH105: Unsaturated 3S-rhamnoglycuronyl hydrolase : The enzyme coded by this gene is periplasmic and works together with ulvan lyases to fully degrade the ulvan polymer. It catalyzes specifically the cleavage of the unsaturated uronic acid at the non-reduced end of the oligosaccharides generated by the ulvan lyase.
- P36_ S1_25: Bifunctional sulfatase/alpha-L-rhamnosidase : The enzyme coded by this gene is periplasmic and is a bifunctional enzyme containing sulfatase and alpha-L-rhamnosidase activities involved in ulvan degradation. The sulfatase is the part of this protein that interests us. It specifically desulfates L-rhamnose at the 3-position. The architecture of this protein with 2 functions is very rare.
2) BIOBRICKS CONSTRUCTIONS
To observe the secretion of the proteins coded by those genes we would add a secretion domain protein inspired by the project of iGEM Paris Pasteur team of 2018. Each gene cited previously would be integrated with three sequences in a plasmid :
- a alpha-haemolysin signal sequence (hylA) to secret the enzymes out of the cells.
- a TEV cleavage sequence to delete the peptide signal
- a sequence tag different for each to facilitate enzyme extraction and detect their extracytoplasmic concentration
Figure 2 : Figure illustrating the construction of the 4 plasmids for further analysis of each gene. Only the part before the dotted line would be changed in each plasmid. A different tag is added at each gene. A His Tag, a HA Tag, a Flag Tag and a Strep Tag in the C-terminal side of P30_PL28, P10_PLnc, P36_GH105, P33_S1_25 respectively.
Each sequence are codon-optimized for E.coli host by Twist and doesn’t include EcoRI, XbaI, SpeI, PstI and NotI sites.
If we could go to the lab, we would use those protocols.
Once integrated in the strain we will induce the gene expression by adding IPTG. The Plac promoter IPTG-dependant, is activated and begins transcription of the inducer genes fused with hlyA transport signal. Desired protein is secreted through the membrane via Type 1 secretion system (T1SS). Finally, thanks to the tag sequence, we would purify the 4 proteins with an affinity column. After these production steps, we wish to quantify protein concentration thanks to HPLC-MS. The last step would be to test the enzymes activity on Ulvan polysaccharides.
After those tests, we would integrate the 4 PUL biobricks into a genetic circuit composed of two compatible plasmids, one containing 3 PUL genes and the second containing the last PUL gene (P33_S1_S25). This design is motivated by the large size of some of the PUL genes. Initially we would use inducible Plac promoters which can be turned-on by IPTG, for expressing our chimeric genes, but we might need to optimize the circuit using other promoters.
Over-expressed 4 enzymes passing through the same secretion system, risks cluttering it and preventing optimal secretion in the medium. In this case, we have those to overexpress the genes coding for the type I secretion system protein.
Moreover, the TEV protease will be overexpressed in the culture medium to cleave HlyA_tag, allowing maturation proteins and to activate their catalytic activity. The removal secretion domain would be confirmed using SDS-PAGE and Western-Blot.
Finally, we would analyse the enzymatic activities by examining the products of ulvan degradation using HPLC-MS.
Figure 3 : Diagram representing the transformation of the E.Coli strain W3110 mutant rhaB and rhaT by the two plasmids carrying our different parts.
F.agariphila and E.coli probably have different secretion systems. Among the 4 enzymes, P33-GH105 et P36-S1-25 present a periplasmic origin in the reference strain. This structure wasn’t present in E.coli strain. To ensure export of these proteins into the extracellular medium, it is necessary to add a signal sequence (HlyA) in the C-terminal end of these genes. We would use the part product by iGEM Paris-Pasteur 2018 team which should permit to secrete enzymes directly using E.coli type I secretion system.
The type I secretion system is composed of :
- an inner membrane protein HlyB
- a periplasmic channel protein HlyD
- an outer membrane protein TolC
- A signal sequence alpha-haemolysin HlyA
This system allows passage of proteins in one step across the two cellular membranes.But there are two major problems when the system must secreted recombinant proteins :
- First, the signal sequence HlyA is not cleaved when crossing the channel, but it needs to be cleaved to obtain a functional protein in the medium. That is why we need to fuse a cleavage signal TEV (Tobacco Etch Virus protease) between gene and signal sequence. As we co-express TEV protease, the signal sequence will be eliminated once it is out of the cell, and our genes can be activated.
- Secondly,E.colihasn’t hlyB and hlyD genes. Thus, we need co-transformed our bacteria with another plasmid pVDL 9.3, bearing hlyB and hlyD sequences, in order to secrete the 4 enzymes out of the cell.
Figure 4: Diagram representing the synthesis and export of a recombinant protein by the type 1 secretion system
To detect precisely the amount of rhamnose in the production middle, we have decided to use a biosensor inspired by the biobricks of iGEM Paris Bettencourt 2012 team but a little bit ameliorated.
This biosensor is an E.coli strain with a plasmid that contains a rfp gene under control of a pRhaBAD promoter, inducible by L-rhamnose. In the same plasmid we can find an antibiotic resistance gene and a rhaS gene under control of a pPAMP promoter.
RhaS being the activator of the pRhaBAD promoter, which becomes active when it binds the rhamnose, its constitutive presence in the plasmid in addition with its presence in E.coli chromosome will allow the increase of the detection signal.
When RhaS binds the L-rhamnose, it will bind the pRhaBAD promoter and activate the expression of GFP. Thus, in presence of L-rhamnose by our production strain, we will have a green fluorescence dose-dependant of the amount of rhamnose. Thanks to this we will know if our production is efficient or not.
Figure 5 : Illustration showing the detection of rhamnose by our biosensor by fluorescence measurement.