Part:BBa_K3490020

Regulating the growth rate of bacteria.

Overview
For our project, we need the bacteria to stay alive in our contact lenses until it can be used. However, bacteria can’t live for a long time in such a small space, especially with such limited nutrients. To solve this problem, we designed a growth switch to put bacteria into hibernation for storage, which we can then resuscitate after exposure to external stimulation.

Fig.1. The big picture of growth switch

Toxin-antitoxin system TA System

The hicA-hicB locus located on E. coli chromosome is a toxin-antitoxin system (TA system). The TA system is composed of linked genes, encoding a toxic protein that can inhibit cell growth and an antitoxic protein which neutralize the toxic protein^{[1]^{. For our project, a kind of the TA system effect called the stress tolerance effect, is used to induce bacteria hibernation. By making use of this mechanism of the TA system, we designed a growth switch to regulate the growth of bacteria.}}

According to a research^{[2]^{, HicA is a toxin promoting mRNA degradation, leading to the hibernation of bacteria, while HicB is an antitoxin that can neutralize the effect of HicA. Hence, through regulating the expression of hicA and hicB genes, we can control the growth of bacteria, hibernating the bacteria before the contact lens are used.}}

The blue light-sensitive system: EL222

Since contact lens are sealed during the process of production, it is impossible to regulate the growth switch through chemical induction; hence, we turned to external stimuli such as light, temperature, etc. At first, we used EL222 to regulate the system^[3] (a blue light-sensitive protein), pBlind (a promoter activated by EL222), FLP, and FRT to change the orientation of our promoter^[4], which is activated by blue light (Fig.2). In other words, we have to illuminate the contact lens with blue light for 30 minutes before use.

Fig.2. The first version of design: BBa_K3490020
This design enables us to control the bacteria growth in three stages.
First, the production stage is when we are culturing our bacteria. Without any arabinose addition during culture, the bacteria will be able to grow normally. After the production process, we will put the bacteria and arabinose into the contact lens together, so the pBAD promoter will transcribe hicA, causing the bacteria to hibernate. When we need to resuscitate the bacteria, EL222 will be activated by blue light to induce pBlind promoter, and then FLP will be transcribed. FLP will act on the FRT sites to change the orientation of the constitutive promoter, as a result, the bacteria will start to transcribe hicB to neutralize HicA and thus continue to grow.
We replaced hicA and hicB with GFP and OFP gene to build a test plasmid because it is much easier to observe the color change than to count the CFU (Fig.3). If the color of fluorescence changes from green to orange, the experiment is successful to prove that the design is feasible.
However, during the experiment process, we had difficulty constructing both plasmid. With the doubt that the EL222-FLP system might be harmful to the growth of bacteria, we changed the design of blue-sensitive protein EL222 into thermo-sensitive protein CI857 on the plasmid pCP20 provided by our PI, Prof. Ng (Fig.4).

Fig.3. The plasmid map which changes the toxin-antitoxin gene into fluorescent genes

Fig.4. The heat-activated plasmid pCP20

The thermo-sensitive system: pCP20
Therefore, we employed heat-activation, as our new method to resuscitate the bacteria. The thermo-sensitive system can be activated through the degradation of CI857 protein when the temperature rises to 36-40 ºC.^[5] This time, the experiments of ligation and transformation were successful, but the color of GFP were not observed in the bacteria after transformation. Suspecting that the sequence might be mutated, we sent plasmid for DNA sequencing. As we expected, the sequencing region matched poorly to our GFP-OFP design. As shown in Fig. 4, one of the FRT sequences is severely mutated next to the promoter J23100(Fig.5). This suggests that our design had violated some fundamental principle. Since the FLP recombinase works by twisting the DNA to bring two FRT sites together and nicking them to generate insertion or deletion. We speculated that the distance between the two FRT sites is insufficient for the two FRT sites to form a loop in vivo. We then check for the optimal distance between two FRT sites for the FRT-FLP system to function. Previous report states that the optimal distance for the recombination to take place is around 200bp in vitro^[6], while our original design between two FRT sites is only 160bp, this may be the cause of malfunction.
Nevertheless, we had to come up with a new design.

Fig.5. The sequencing result of sfGFP-FRT-J23100-FRT-OFP, J23100 is highlighted, and the red arrow indicates FRT sequence flanking J23100
The change of FRT mechanism
FRT sites can either change the orientation of the gene or delete the gene depending on the orientation of the two FRT sequences^[7]. Since the design of changing the promoter direction failed, we choose to delete the gene, which is more common in general lab settings. For the new design, we still used the plasmid pCP20 for heat-activation, but we put hicA gene between the FRT sites for deletion on the plasmid pSB1C3 (Fig 6.a), and the plasmid pSB3K3 with hicB gene was also designed to continuously produce HicB (Fig 6.b). Because pSB3K3 is a low copy number plasmid, the expression of HicB is less than HicA, therefore, the bacteria will hibernate. After heat activation, hicA gene will be deleted while hicB gene is still expressed, so the bacteria can resuscitate. For these three plasmids, we can regulate the growth of bacteria, meeting our need to design a growth switch.

Fig.6.a. The design of BBa_K3490030

Fig.6.b. The design of BBa_K3490031
The experiment of kanamycin-resistant gene deletion: Prove of design
Due to time constraint, we can only demonstrate this model using pKD3 plasmid, which carries a kanamycin-resistance gene flanked between two FRT sites provided by our PI, Prof. Ng. If the bacteria cannot survive at LB plate with kanamycin after heat-activation, we can demonstrate that the gene between the FRT sites was deleted, indicating that our design is feasible. For the experiment, we co-transformed pKD3 and pCP20 into E. coli DH5α, and cultured a colony in 37℃ for 8 hr and transfer to 42℃ for 1 hr to heat-activate the bacteria, and then spread on the LB plate (dilute 105 times) and culture for 12 hr. After the colony grew, we screened the colony on a kanamycin plate to check whether the bacteria will grow or not. Results show that most of the bacteria can not survive on the Kanamycin plate but can still survive on the LB plate without kanamycin (Fig.7, 8.), which means the antibiotic-resistance gene was deleted successfully. With this data, we are convinced that the design of pSB3K3 with hicA gene is feasible.

Fig.7. The bacteria can grow on the LB plate without antibiotics (the left one), but when we pick the colony from it to the kanamycin plate (the right one), there is almost no colony on the plate.

Lane1: the uncut plasmid with activation of the FRT system
Lane2: the plasmid cut by HindIII with activation of the FRT system
Lane3: the uncut plasmid without activation of the FRT system
Lane4: the plasmid cut by HindII without activation of the FRT system
LaneM: the marker

Fig.8. The plasmid pattern before and after activating FRT genes checked by DNA gel electrophoresis.

Sequence and Features

Assembly Compatibility:

10
INCOMPATIBLE WITH RFC[10]
Illegal EcoRI site found at 1512
Illegal SpeI site found at 1830
12
INCOMPATIBLE WITH RFC[12]
Illegal EcoRI site found at 1512
Illegal NheI site found at 7
Illegal NheI site found at 30
Illegal NheI site found at 2789
Illegal SpeI site found at 1830
21
INCOMPATIBLE WITH RFC[21]
Illegal EcoRI site found at 1512
Illegal BamHI site found at 2729
23
INCOMPATIBLE WITH RFC[23]
Illegal EcoRI site found at 1512
Illegal SpeI site found at 1830
25
INCOMPATIBLE WITH RFC[25]
Illegal EcoRI site found at 1512
Illegal SpeI site found at 1830
Illegal NgoMIV site found at 142
Illegal AgeI site found at 367
1000
INCOMPATIBLE WITH RFC[1000]
Illegal BsaI site found at 2225
Illegal BsaI.rc site found at 2265