Team:NCTU Formosa/Results

All of the toxin genes we chose were toxic to E. coli. Thus, during the cloning experiment, we could hardly get the intact sequence of toxin genes. Since they cause growth arrest in E. coli, most of the E. coli would go through natural selection with some mutation to lower the toxicity of the gene itself. Therefore, we expected that the natural selection (mutations) might happen during the cloning step (Table 1). However, it did not stop us from the progress of the functional test. It is expected that the naturally “modified” sequence would not lose its function but rather mild ones for survival. This logic supported the idea of the project design, for toxin genes were not to kill all the E. coli. Therefore, we went on to the subsequent functional test to check whether if there were functional colonies from the plates. Fortunately, all the toxin genes showed functional colonies. These colonies were sequenced, and the modified amino acid sequence of the four genes are listed below.

Table 1: Modified amino acid sequence of toxin genes. WT: Wild Type, MT: Mutation Type.

We submitted the parts with modified sequences for future iGEMers to easily passed on the work we have done for reference.

Cloning

We conducted colony PCR to verify that our target gene was correctly cloned into the E. coli BL21 (DE3).

Figure 1 was the electrophoresis results of the colony PCR with a marker on the left side and the target gene on the right side. The lengths are labeled beside each band. As a result, we successfully cloned five target genes, respectively, into E. coli.

Figure 1: Colony PCR result of toxin genes after cloning into E. coli BL21(DE3). (A) ccdB* BBa_K3256440 (B) MazF* BBa_K3256441 (C) YafQ* BBa_K3256442 (D) ydfD BBa_K3256443 (E) ChpBK* BBa_K3256444.

Functional Test

After confirming the cloning of target genes, we tested their function by measuring the O.D. value after IPTG induction when the O.D. value reached 0.3 and compared with the O.D. value of the control. The O.D. values were documented in every five minutes for seven hours in total. All experiments were completed in triplicates. (Figure 2~6)

Figure 2: Growth curve of E. coli BL21(DE3) with 500uM IPTG induction ccdB toxin gene (blue), and control (orange)

Figure 3: Growth curve of E. coli BL21(DE3) with 500uM IPTG induction MazF toxin gene (blue), and control (orange).

Figure 4: Growth curve of E. coli BL21(DE3) with 500uM IPTG induction YafQ toxin gene (blue), and control (orange).

Figure 5: Growth curve of E. coli BL21(DE3) with 500uM IPTG induction YdfD toxin gene (blue), and control (orange).

Figure 6: Growth curve of E. coli BL21(DE3) with 500uM IPTG induction ChpBK toxin gene (blue), and control (orange)

Calculating Toxicity of Toxin Genes

We first fit the control’s experiment data to the following equation, $$\frac{dB_{T}}{dt}= g\cdot B_{T} (1-\frac{B_{T}}{B_{Max}})$$ and we fit the induced data to another equation, $$\frac{dB_{T}}{dt}= g\cdot B_{T} (1-\frac{B_{T}}{B_{Max}})-T_{toxin}\cdot B_{N}\cdot [toxin]$$ Next, we compared the two equations to calculate Ttoxin, the toxicity of the toxin gene (Figure 7). In the end, we chose the ydfD gene, which had the most significant toxicity, and the ccdB gene, which had relatively weak toxicity, moving on to the functional test with mutagens.

Figure 7: Toxicity of different toxin genes. Ttoxin: The toxicity of toxin gene.

Mutagens Bioassay Function Test

According to the toxin gene functional test, with selected ccdB and ydfD for mutagens bioassay. To see whether our design can work as expected, we put the transformed E. coli into a different dose of mutagens (UV light and EtBr), respectively. Since the mutagen might cause growth inhibition, it was difficult to tell how strong the mutation rate caused by the mutagen was only through the raw data of O.D.600. Therefore, we took the maximum O.D value of the controlled group (which represented the none-induced E. coli treated with the same dose of mutagen) as 100%, and the minimum value is 0% to normalize the data, and we called the normalized data growth population percentage.

Functional Test of Mutagenic Bioassay on UV Light

In the environment, UVB (275～320nm) is the most carcinogenic wave band that humans usually contact according to the reference, so we chose it to be our physical mutagen for the functional test of bioassay. In the experiment, we exposed them to the different intensity of UV light for 5 seconds and induced the toxin gene when its O.D reached 0.3. After that, we documented the O.D. of the E. coli every 5 minutes for 7 hours. The results of the two engineered E. coli with ydfD and ccdB, respectively, under different UV light intensity, were shown below (Figure 9, 10).

Figure 8: The growth curve of ydfD with different intensity of UV Light.

Figure 9: The growth curve of ccdB with different intensity of UV Light.

Functional Test of Mutagenic Bioassay on EtBr

We chose EtBr, a chemical compound that causes mutation and usually seen in laboratories to be our chemical mutagen. First, we made them added different concentrations of EtBr for creating different mutagenic strengths and added 500uM IPTG to induce the toxin gene when its O.D reached 0.3. After that, we documented the O.D value of E. coli every 5 minutes for 7 hours. Below was the EtBr mutagenic effect on the two toxin genes.

Figure 10: The growth curve of ydfD with different dose of EtBr

Figure 11: The growth curve of ccdB with different dose of EtBr

From the experiment result, we can see that the curve of ydfD has a more significant drop than ccdB after we induced the toxin gene. This phenomenon can be explained, for the sequence of ccdB has been modified (Figure 11, 12). Therefore, its toxicity has been weakened compare with the intact ydfD sequence.

Mutation rate analysis

According to the growth curve of two mutagenic factors with two selected toxin genes respectively for building up mutagenic bioassay, we further converted the raw data above to mutation rate of the mutagen which we called M₂, we used the following differential equation from modeling to fit the experiment data: $$\frac{dB_{N}}{dt}=[ g\cdot B_{N}(1-\frac{B_{T}}{B_{Max}}) -T_{chem}\cdot B_{N}](1-M_{1}-M_{2})-T_{toxin}\cdot B_{N}\cdot [toxin]$$ $$\frac{dB_{M}}{dt}=[g\cdot B_{N}(1-\frac{B_{T}}{B_{Max}})-T_{chem}\cdot B_{N} ] (M_{1}+M_{2})+[g\cdot B_{M}(1-\frac{B_{T}}{B_{Max}})-T_{chem}\cdot B_{M}]$$ $$\frac{dB_{T}}{dt}=[g\cdot B_{T}(1-\frac{B_{T}}{B_{Max}})-T_{chem}\cdot B_{T} ] -T_{toxin}\cdot B_{N}\cdot [toxin]$$

We were first fixed the parameters from the isolated experiments and essays, and then used Matlab to estimate the mutation rate, M₂, with the most appropriate curve. Below is the linear regression of M₂ and dose of mutagen (Figure 12~15).

Figure 12: Linear correlation of mutation rate in UV intensity (ydfD)

Figure 13: Linear correlation of mutation rate in UV intensity (ccdB)

Figure 14: Linear correlation of mutation rate in EtBr concentration (ydfD)

Figure 15: Linear Correlation of Mutation Rate in EtBr concentration (ccdB)

We can see that the M₂ we calculate was a highly positive correlation with the dose of mutagens. It verified that by using our designed gene circuit and model, the mutagenic bioassay system we constructed is highly credible and can be further used for mutagenic studies for unknown substances. For the detail of how we build up the simulation equation and the verification of the growth curve model, please see the growth curve model.

Go to >>>

Contribution