TraI Improvement Assay

Introduction

In the previous wiki page (Read TraI Assay page), we describe that the productivity of C8 in E. coli depends on the culture temperatures. However, to complete our co-culture system, the current 3OC8HSL (hereafter C8) productivity at 37℃ was not enough to transmit the AHL signal to human cells, because human cells are usually grown at 37℃. Therefore, we tried to mutate the traI gene and increase the productivity of C8 at 37℃.

Summary of experiment

TraI has not been characterized extensively, and thus, it is unclear what kind of mutations is appropriate for the above purpose. A preceding study describes that, in the case of LuxI, the amino acid substitution at the 34th and 63rd positions (both are substitutions from glutamate to glycine; E34G and E63G) increase the productivity of C6 (2). Since TraI has homology to LuxI over the entire amino acid sequences, we speculate that the same amino acid substitutions in TraI can increase the productivity of C8. Also, we here describe the modification of the culture conditions and the host strain choice for increased C8 production.

Results

When amino acid sequences of TraI and LuxI were aligned using the clustal W program (1), the E34 and E63 residues of LuxI were found to correspond to K34 and Q63 residues of TraR. According to this information, oligonucleotide primers to create TraI-K34G, TraI-Q63G, and TraI-K34G,Q63G mutants were designed. The primer sequences are shown in Fig. 1. The mutations were introduced to the pSB1C3-based traI plasmid using the inverse-PCR method, and successful introduction of the mutations were confirmed with Sanger sequencing.

Fig. 1 Sequences of the primers. Note that each primer set is divergent for inverse-PCR

The sequences of traI mutants and wild-type are shown in Fig. 2. The plasmids we used are shown in Fig. 3 and 4. The Sender and the Reporter strains were prepared in the same way as described in the previous wiki page (Read TraI Assay page).

Fig. 2 Sequence of traI wild type gene and mutant

Fig. 3 Structure of the plasmids used for creating the “Reporter” strain.

Fig. 4 Structure of the plasmids used for creating the “Sender” and “Mutated Sender” strains. One of these plasmids was used for preparing each transfromant

In the experiments shown in this page, one additional modification was made in experimental conditions; 1 microM of SAM (S-adenosylmethionine; structure is shown in Fig. 5) was added to the culture of the Sender. Since C8 is synthesized from SAM and ACP (acyl carrier protein) through the action of TraI in bacterial cells (3), we expected that the addition of SAM may increase the productivity.

Fig. 5 Chemical structure of SAM (S‐adenosylmethionine)

The result of C8 production using the TraI wild-type and the mutants is shown in Fig. 6. "W.T." means native traI.
The RFU value of the TraI(K34G)-expressing cells was about 3-fold higher than that of the TraI-expressing cells. E. coli introduced empty vector was used as Negative Control.
Other mutant did not show improvement of C8 production (data was not shown).
When these RFU values were converted to C8 concentrations using the calibration curve obtained in the reagent assay (Read TraI Assay page), they were calculated as 28 nM and 42 nM, respectively.

Fig. 6 Improvement of C8 production by the K34G mutant (37℃ culture)

Also, the other modification was made concerning the host strain of the Sender. The preceding iGEM study has shown that the amount of AHL produced by the luxI gene highly depends on the host strain; depending on the used strains as the Sender, there was approximatly 100-fold difference in cell number for obtaining the same activation level of the lux promoter (4). Therefore, we here used the other strain, MG1655hapB, as the Sender.
Note that, in the experiments of the previous wiki page (Read TraI Assay page ), only the DH5α strain was used as a host. The MG1655hapB strain is a mutant of MG1655 (the representative wild-type K12 strain) and has higher membrane permeability for hydrophobic compounds compared to the parent.
As a result (Fig. 7), we found that amount of C8 produced is dependent on strains. The RFU value in the MG1655hapB strain was about 3-fold higher than that in the DH5α strain. E. coli introduced empty vector was used as Negative Control.
When these RFU values were converted to C8 concentrations using the calibration curve obtained in the reagent assay (Read TraI Assay page), they were calculated as 28 and 37 nM, respectively.

Strain dependence of AHL production
We found that amount of C8 production is depend on E. coli’s strain. RFU is 2-fold higher than DH5α.
Calculated from the graph obtained in the reagent assay, C8 concentration of DH5α culture was 28 nM and MG1655hapB culture was 36 nM.

Fig. 7 Strain dependence of C8 production

Discussion

In the previous study, it was considered that the E34G mutation of LuxI most likely enhances the interactions between the enzyme and the ACP substrate. Therefore, we thought that K34G mutation of TraI also has the same effect.
It was also showed that the MG1655hapB strain produced more C8 than the DH5α strain. We speculate that the difference in permeability of hydrophobic compounds through the cell membrane is the main reason for this result.
Taken together, we conclude that increasing the productivity of C8 at 37℃ was successful. Notably, generation and functional identification the mutant traI (TraI(K34G)) meet the medal criteria of ”parts improvement”, because the wild-type traI parts was registered in iGEM parts collection earlier (Read BBa_K553001 page). However, further improvement of C8 production is necessary to transmit the signal from bacteria to human cells. Such improvement is possible through tuning the experimental conditions further.

Appendix: Material and Method

Reference

"> (1). http://www.genome.jp/tools-bin/clustalw
(2). Pavan Kumar Reddy Kambam, Daniel J. Sayut, Yan Niu, Dawn T. Eriksen, Lianhong Sun (2008) Directed evolution of LuxI for enhanced OHHL production. Biotechnology and Bioengineering Volume 101, Issue 2 1 October 2008 Pages 263-272
(3). MATTHEW R. PARSEK, DALE L. VAL, BRIAN L. HANZELKA, JOHN E. CRONAN, E. P. GREENBERG (1999) Acyl homoserine-lactone quorum-sensing signal generation. Proc. Natl. Acad. Sci. USA Vol. 96, pp. 4360-4365, April 1999 Biochemistry
(4). https://2007.igem.org/wiki/index.php/Chiba/Quorum_Sensing