Difference between revisions of "Team:Evry Paris-Saclay/Peptide"

Line 42: Line 42:
 
.axel{
 
.axel{
 
   padding-top:15px;
 
   padding-top:15px;
 +
}
 +
 +
table{
 +
  font-size: 12px;
 
}
 
}
  

Revision as of 12:32, 13 October 2018


TO CONTACT US
Genopole Campus 1, Batiment 6, 91030 Evry Cedex, France
+33 7 69 96 68 31
igemevry@gmail.com

© Copyright 2018
Design & Developpement by
IGEM EVRY GENOPOLE

THE PEPTIDE

In our system, the hexapeptide SAIRGA is the signalling molecule used for cell-to-cell communication [1]. To perform its quorum sensing function in the natural phi3T phage infection, SAIRGA needs to be secreted out of the cell. It is produced as an immature pre-pro-peptide, AimP (BBa_K2279001 and BBa_K2675001), that upon secretion is cleaved extracellularly to remove the secretion signal and release the mature hexapeptide. The mature peptide then enters cells and binds to its receptor protein AimR (BBa_K2279000 and BBa_K2675000). Binding of SAIRGA to AimR blocks the activator function of AimR that, in turn, facilitates a switch from lysogenic-to-lytic viral cycle. By acting as a negative regulator of AimR, the SAIRGA signal makes the lysogeny-to-lysis switch of phi3T phage dependent on the “quorum” of Bacillus cells.

DESIGN

In phi3T phage, SAIRGA is produced with specific secretion and protease-cleavage tags [1] that allow secretion and extracellular processing by Bacillus. Since we had no further information about this protease, we could not investigate the ability of Escherichia coli to produce a similar enzyme. Consequently, to produce SAIRGA in E. coli, we have decided to replace this tag by an E. coli secretion signal. We have investigated three different secretion signals: OmpA (BBa_K208003), PelB (BBa_J32015) and Tat of csp2 gene of Corynebacterium glutamicum (Uniprot Q04985).

Part Name CDS Sequence Part Number Expression Cassette Part Number
AimP MKKVFFGLVILTALAISFVAGQQSVSTASASDEVTVASAIRGA BBa_K2675001 BBa_K2675041
OmpA-SAIRGA MKKTAIAIAVALAGFATVAQASAIRGA BBa_K2675002 BBa_K2675042
PelB-SAIRGA MKYLLPTAAAGLLLLAAQPAMASAIRGA BBa_K2675003 BBa_K2675043
Tat-SAIRGA MFNNRIRTAALAGAIAISTAASGVAIPAFASAIRGA BBa_K2675004 BBa_K2675044

To express them in E. coli, we have codon optimized all sequences for E. coli DH5α, have designed specific RBSs with the Salis RBS Calculator [2, 3] and have placed these peptides under the control of the constitutive strong promoter BBa_J23100. Thus, we generated four SAIRGA expression plasmids (BBa_K2675041, BBa_K2675042, BBa_K2675043 and BBa_K2675044).

EXPERIMENTAL SETUP

E. coli DH5α harboring the SAIRGA expression plasmids were grown at 30°C and 37°C in mineral salts medium [4] containing 2 mg/mL glucose and 35 μg/mL chloramphenicol (Figure 8). During the culturing, samples were prelevated, centrifuged and the supernatant was first sterilised by filtering through a 0.22 µm membrane, then loaded on a Pall Nanosep centrifugal device with Omega membrane (MWCO 3 kDa). The thus recovered low molecular weight fraction of the culturing media was analysed by LC-MS: ultra high pressure liquid column chromatography (UPLC) on a ACQUITY HSS T3 1.8 µM, 2.1 x 100 mm column (Waters) coupled with a Xevo G2-S Qtof mass spectrometer (Waters). The chromatography was performed using a gradient of using a (A) 1% formic acid in whater / (B) 1% formic acid in acetonitrile with the following program: from 0 to 5 min solvent B was increased from 2% to 70%, then it was maintained for 1 min at 70%; aftreward it was decreased in 30 sec to 2% and it was maintained at 2% for 3.5 min for column re-equilibration. The flow rate was 0.2 mL/min and the column temperature was set to 45°C. The ions were detected in resolution negative mode.

The pure synthetic SAIRGA peptide was readily detected by LC-MS. As shown in figures 1 and 2, a peak of 572 Da, which correspond to SAIRGA’s monoisotopic mass [M-H]- was observed at a retention time of 3.9 min (figure 3). The area under the curve is proportional to the concentration of SAIRGA (figure 4).

Figure 1: Mass spectrometry analysis of SAIRGA’s fraction obtained with various amounts of pure synthetic SAIRGA peptide.
Figure 2: Mass spectrometry analysis of SAIRGA’s fraction obtained with 5 µM of pure synthetic SAIRGA peptide (zoom on SAIRGA’s peak).
Figure 3: UPLC chromatograms obtained with various amount of pure synthetic SAIRGA peptide.
Figure 4: Standard curve of the area under the curve obtained by UPLC as a function of the concentration of SAIRGA.

RESULTS

LC-MS analysis of samples isolated from the fermentation broth of the E. coli strains harbouring a SAIRGA expression plasmid was performed.

As shown in Figures 5 and 6, the expression vector of OmpA-SAIRGA (BBa_K2675042) is able to produce a compound with a monoisotopic mass a peak of 572 Da, which correspond to SAIRGA’s monoisotopic mass [M-H]- at a retention time of 3.9 min. The size or the area under the peak is increasing with time (figure 7). This suggests that OmpA is an efficient E. coli secretion tag which allows the release of the mature peptide SAIRGA in the culturing media.

Figure 5: Mass spectrometry of culturing media taken after 2, 4, 6 and 8 hours of incubation of E. coli cells harboring the OmpA-SAIRGA expression cassette in pSB1C3 (BBa_K2675042). The negative control has been performed with en empty pSB1C3 (bottom red chromatogram). The positive control (pure synthetic SAIRGA at 5 µM) is represented on the top red chromatogram.
Figure 6: UPLC chromatograms of culturing media taken after 2, 4, 6 and 8 hours of incubation of E. coli cells harboring the OmpA-SAIRGA expression cassette in pSB1C3 (BBa_K2675042). The negative control has been performed with en empty pSB1C3 (bottom green chromatogram). The positive control (pure synthetic SAIRGA at 5 µM) is represented on the top red chromatogram.
Figure 7: SAIRGA production at 37°C by E. coli cells harboring the various SAIRGA expression plasmids (BBa_K2675041, BBa_K2675042, BBa_K2675043 and BBa_K2675044). The negative control has been performed with en empty pSB1C3.
Figure 8: Growth curves at 37°C of E. coli cells harboring the various SAIRGA expression plasmids (BBa_K2675041, BBa_K2675042, BBa_K2675043 and BBa_K2675044). The negative control has been performed with en empty pSB1C3.

The other parts (BBa_K2675041, BBa_K2675043 and BBa_K2675044) have not shown any peak at 572 Da, which suggests that the original Bacillus secretion tag, PelB and Tat are not able to release the mature peptide. Our hypothesis is that, in each case, the pro-peptide is produced and released but its secretion tag is not cleaved by the cell.

As shown in Figure 9, the cells which have been grown at 30°C have not shown any significant production of SAIRGA.

Figure 9: SAIRGA production at 30°C by E. coli cells harboring the various SAIRGA expression plasmids (BBa_K2675041, BBa_K2675042, BBa_K2675043 and BBa_K2675044). The negative control has been performed with en empty pSB1C3.

Moreover, after 24h of incubation, we were unable to detect the SAIRGA peptide in the culturing media (Figures 10 and 11). This is probably due to the exhaustion of the medium: the cells might have used SAIRGA as a source of amino acids.

Figure 10: UPLC chromatogram obtained with media taken after 24h of incubation of E. coli cells harboring the OmpA-SAIRGA expression cassette in pSB1C3 (BBa_K2675042).
Figure 11: Mass spectrometry analysis of the media taken after 24h of incubation of E. coli cells harboring the OmpA-SAIRGA expression cassette in pSB1C3 (BBa_K2675042).

CONCLUSION

We have successfully built a part, OmpA-SAIRGA (BBa_K2675002), which is able to produce and release the mature hexapeptide SAIRGA in the culturing media of E. coli cells harboring this part equipped with a RBS and under the control of a constitutive promoter (BBa_K2675042).

REFERENCES

[1] Erez Z, Steinberger-Levy I, Shamir M, Doron S, Stokar-Avihail A, Peleg Y, Melamed S, Leavitt A, Savidor A, Albeck S, Amitai G, Sorek R. Communication between viruses guides lysis-lysogeny decisions. Nature (2017) 541, 488-493.

[2] 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 Res (2014) 42, 2646-2659.

[3] Salis HM, Mirsky EA, Voigt CA. Automated design of synthetic ribosome binding sites to control protein expression. Nat Biotechnol (2009) 27, 946-50.

[4] Hall BG. Activation of the bgl operon by adaptive mutation. Mol Biol Evol (1998) 15, 1-5.