Our project, the story of “Snow White” is constructed based on mazEF system, which is one of toxin-antitoxin (TA) system on E. coli genomic DNA. At the same time, we are interested in other TA systems and we carried out assay using yafNO system.

2-1. Confirming YafO function as toxin on agar plates

We first confirmed YafO function by observing formation of colonies on agar plates. This experiment was carried out using E. coli where YafO expression can be induced by arabinose and E. coli without yafO gene. Construction of plasmids used in this experiment is shown as Fig. 2-1-1. We inoculated four types of E. coli differently on agar plates with or without arabinose.

2-2. Toxin-antitoxin assay

From previous experiment, we confirmed that YafO works as toxin. Next, we confirmed whether the cell growth recovers with antitoxin YafN after inhibition from YafO. Construction of plasmids used in this experiment is shown as Fig. 2-2-1. We prepared E. coli where YafO expression can be induced by arabinose and YafN expression by lactose. As comparisons, we also carried out same experiment with E. coli containing no yafO gene, no yafN gene, or none of them. These E. coli were respectively cultured in media with arabinose in order to express YafO, then IPTG were added to the cultures in order to express YafN. We confirmed YafN function against YafO by measuring turbidity and RFU (relative fluorescence units) of GFP.

3-1. Confirming YafO function as toxin on agar plates

Four types of E. coli shown as Fig. 2-1-1 were inoculated on agar plates with or without 0.2% arabinose, and incubated at 37°C. As a result, E. coli containing plasmid (a) and one containing plasmid (c) didn’t form any colonies, although all types of E. coli formed colonies on agar plate containing no arabinose (A). From this result, cell growth was inhibited by inducing expression of YafO.

Particularly, like E. coli containing plasmid (d), E. coli containing plasmid (c) formed fluorescent colonies on agar plate (A), and didn't form any colonies on agar plate like E. coli containing plasmid (a) on agar plate (B). These results insist that genes on plasmid (c) were working for sure.

3-2. toxin-antitoxin assay

Four types of E. coli (shown as Fig. 2-1-1) were inoculated in liquid media, respectively. When turbidity of culture reached 0.03, arabinose was added (final concentration 0.02%) to each culture. After two hours incubation with arabinose, IPTG was also added (final concentration 2 mM). Time-dependent change of RFU and turbidity is shown as Fig. 3-2-1. The graph (A) shows that, even though E. coli containing plasmids (a) has yafN gene, it couldn’t make the cell growth recover, like one containing plasmids (c) (lack yafN gene). Also, from graph (B), no recovery of RFU was shown on E coli containing plasmid (a), and its time-dependent change of RFU was similar to that of turbidity.

　　From the above result, it was clarified that growth of E. coli cells was repeatedly controlled by expression of mazE. The results of this experiment are very useful in our project, and it is expected to lead to new biotechnological applications.

On the other hand, we couldn’t see YafN working as antitoxin against yafO in this experiment. After inducing expression of YafO by arabinose, we induced expression of YafN by IPTG, but turbidity and RFU of the culture stayed constant (Fig. 3-2-1). In mazEF system, expression of antitoxin MazE made the cell growth recover from the inhibition by toxin MazF. These results insist that yafNO system has different mechanisms.

From these results, yafNO system couldn’t carry out “Stop&go” of E. coli cell growth, and we considered that this system can’t be treated the same as mazEF system, which is the basis of our project.