Team:PKU Beijing/Project/Assemble


Project > Assemble


First Stage Assembly


Several strategies are designed for the first stage assembly, the assembly includes three different parts that we constructed. One of them is the AND Gate Module (without output), the second is the Bistable module that we inherited from iGEM 2007 Team Peking, and the third part T7promoter CI acts as a bridge which is both the output of the AND Gate and the input to the bistable module.

fig1. AND Gate part
fig2. Bistable part
fig3. T7promoter CI part

Considering the size of the plasmid and replication origin compatibility of the plasmids, we draw up the following methods to assembly the system by making combinations among the available plasmids and those parts. What we wanted to do is to pick out a result that most approaches our need. And then further carry out mutation on the ribosome binding site for fine tuning if necessary.

The bistable switch is inherited from iGEM Team Peking 2007, it is constructed to have two reporter genes on CI and CI434 side, the CI434 side is GFP gene, and the CI side is RFP gene. We kept the two reporter genes to direct our assembly. The signal is given through the AND Gate, and test whether the fluoresence state is changed.

First Strategy: This strategy is picked because in this way the AND Gate is intact with its input and output on one plasmid. And that the bistable switch is on an independent plasmid so that it has no influence on other modules and vice versa. T7 promoter-CI is placed both upstream and downstream of the AND Gate, for the reason that they are different in their mutual influence.

fig8. First strategy---a
fig8. First strategy---b
fig8. First strategy---c
fig8. First strategy---d

The second strategy is shown below:

We make the T7promoter-CI part the same copy number as the bistable switch, to make the bistable switch easier triggered. However, this design may have larger disturbance on the bistable switch.

fig8. Second strategy---a
fig9. Second strategy---b
fig10. Second strategy---c
fig11. Second strategy---d

The third strategy is to put bistable switch together with the AND Gate. It is easier for the bistable module to be switched.

fig12. Third strategy---a
fig13. Third strategy---b
fig14. Third strategy---c
fig15. Third strategy---d

Finally, we have a 3 plasmid assembly. This design eliminates all the influences between either of the two parts.

fig16. Forth strategy---a
fig17. Forth strategy---b

The first and second method of assembly can well support second stage assembly, because the bistable switch is on relatively higher copy plasmid, coupling the next module can only have minor interference to it. The third method, although not so promising in further assembly, but we didn’t give up trying it.

For the final method, which works out (see result section), its further assembly is also promising. It is discussed in the Second Stage assembly section.


We chose the best AND gate we had constructed (pBad-SupD, pSal-T7ptag, RBS B0033), bistable switch and 6 different T7promoter-C1 (the difference is on the RBS: B0030, B0031, B0032, B0033, B0034 and J44001) to achieve the first stage of assembling.

At last, we have successfully constructed some of clonings in the first and second strategy, and the most of clonings in the third strategy (AND gate and Bistable on pSB4K5) and the last strategy (three plasmids). After testing the function of some cloning in the third strategy and last strategy, we got some promising results, and based on these results, we can continue to finish our whole project.

Results of the third strategy:

After constructed the AND gate and Bistable on pSB4K5, we then transformed T7-C1 (12 kinds in total: two plasmid, pSB3T3 & pSB6A1; six RBS) into the E.coli. Suspend the colonies contained two plasmids, and plate them with inducers or without inducers (control group). Use fluorescent stereoscope to test the expression of GFP and RFP of colonies. The best result we get is from strategy c (fig 14), Bistable is upstream of AND gate, and C1 with the RBS of B0030. The difference of control group and experiment group is distinguished (fig 18).

Fig 18. The best result of the third strategy. Fig 18a: control group, without induction. 18b: induced group: suspend the cells, add 5uL 1M salicylate and 5uL 1M arabinose, plate together with the cells. 18a and 18b are both merged pictures, for original individual figures, please follow this link


We found that the copy number of different plasmid and RBS of C1 can really influence the result. With strong RBS, such as B0034, some colonies become red without inducing, while with weak RBS, like B0033, no one colonies become red after saturated inducing. These data are not showed here. When assemble a complex system, the bridging part (C1 in this assembling) is crucial: it should adjust output of the previous circuit (AND gate in this assembling) to the proper input of the next circuit (Bistable in this assembling). Thus, we need to try a lot of conditions, different plasmids and different RBSs, to get satisfied results.

Results of the last strategy (three plasmids):

We have constructed the three modules, AND gate, Bistable and T7-CI, on three different plasmids: pSB4K5, pSB1A2/pSB6A1 and pSB3T3. Then we transformed them into E.coli one by one. Then the colonies contained three plasmids were picked into 3 mL liquid LB and incubated until the OD600 value reached 0.4~0.6. At this time, we used different combinations of different concentration of arabinose and salicylate to induce. Every two hours we discarded 2mL LB culture, and added 2mL new LB into it, and the concentration of inducers should be kept. Repeating this step for 3 times, and then cultivated the culture overnight to ensure saturated induction and interaction between different modules. The results were tested by flowcytometry, controls that constitutively expresses GFP and RFP are used to determine the red cell region and green cell region(fig 19). Flowcytometry data of the first stage system is analysis by counting the cells that falls in both of the regions.

Fig 19. The RFP, GFP controls and definition of different states areas. The Y axis denotes the strength of RFP fluorescence, and the X axis means the strength of GFP fluorescence. Fig 19a is the result of RFP control, so we enclosed a polygon area as the state of RFP in Bistable Switch (CI side). Fig 19b is the result of GFP control, depended on which we defined the area as the state of GFP.

Taking advantages of the defined enclosed areas, we can count the cell number in either states of Bistable switch. And this is a criterion to assess the quality of assembling. We have tested a number of assemblies and different combinations of various concentrations of arabinose and salicylate.

At last, the best results are from the assembling system, which is made of AND gate on pSB4A5, Bistable on pSB1A2 and CI on plasmid pSB3T3, with the RBS of J44001. The concentrations of inducers are 10^-4M arabinose and 10^-6M salicylate.

Fig 20. The best result of the three plasmids strategy. The Y axis denotes the number of cells in the defined enclosed areas (figs on the left side for GFP area and figs on the right side for RFP area). The X axis means the strength of G(R)FP fluorescence. Fig 20a: control group, without inducing. Fig 20b: arabinose single inducing group, the concentration is 10^-4M. Fig 20c: salicylate single inducing group, the concentration is 10^-6M. Fig 20d: arabinose and salicylate double inducing group, the concentration of arabinose is 10^-4M and of salicylate is 10^-6M

We calculated red to green cell number ratio. The histogram of the Red/Green ratio under four conditions (no induction, 2 x singal induction, double induction) are shown in Fig21:

Fig21. The columns are the red to green cell number ratio values. From left to right, it is the no induction data, arabinose singal induction data, salicylate singal induction data and the double induction data. Double induction data shows significant difference from that of singal induction and no induction. Although the arabinose induced is to some extent leaky, it shows great difference form the double induction group.

Second Stage Assembly (In Process)


After the first stage assembly was finished, we designed the further step based on what we have. It is shown in the result section of the first stage assembly that the three plasmids assembly turn out to work. We designed the second stage assembly as below:

First the OR Gate-GFP module can be put on the same plasmid of the bistable switch, which has a relatively higher copy origin so that the output can be easier visualized. Actually, there is another consideration for this kind of design. Because the PhiR73 delta system is used by 2007 Cambridge team to make an amplifier, placing the PhiR73 delta(with an amber mutation) on a relatively lower copy plasmid than the output module can gain more chance that the OR gate is activated to the a similar extent when exposed to PhiR73 delta as when directly activated by salicylate. (Detailed description is on the OR Gate and Output page).

And then, the PhiR73 delta expression set, which is controlled by CI434 repressible CI activatable promoter is placed on the pSB3T3 plasmid together with the T7 promoter-CI construct. In order to reduce influence on each other, we can use the reversed standard to assembly PhiR73 delta expression set. (Learn more about the reversed standard). We designed the PhiR73 delta expression set on a medium copy plasmid, because the reason we mentioned above, and another reason is that the SupD expression set has a lower copy than the PhiR73 delta expression set. As our earlier experience tell us, supD is much easier to be leaky, while the other half of the AND Gate can be controled by adjusting its rbs. SupD on a lower copy plasmid gives us more chance.

The final assembly introduced another GFP reporter (the OR Gate - GFP set), so we want to delete the GFP gene from the bistable device, while the RFP gene is kept to track the change of memory state.

Actually, our second stage assembly can be viewed as the construction of an AND Gate: what we need to adjust is the rbs of PhiR73(amber mutation). Input to the PhiR73 is from the bistable module, and the input to supD is the same as the first AND Gate.