Team:SZ-SHD/Proof Of Concept


As characterized in the design of our project, we aim to develop a biological pesticide missile using E. coli. A missile needs to contain explosives, or harmful ingredients that kills its victims. It has been recently discovered one of the Bt toxins (toxic proteins produced by Bacillus thuringiensis), CryCa1, display a high specifically and lethality against locusts. When mixed and used with chitinolytic enzymes, the toxicity of Cry7Ca1 would remarkably increase, as the protective layer of chitin on insects’ intestines were being breakdown.

These toxic proteins are all natural degradable and non-poisonous to other organisms, thus will contribute less to the ecology deterioration than chemical pesticides. Moreover, the fermentation is a cheap and portable alternative to industrial manufacture of chemical pesticides: use organic debris as medium for proliferation of toxin-producing bacteria evidentially have fewer requirements on equipment than carefully monitored chemical reactions.

Nevertheless, even toxins were produced by e. coli, little would release out into environment. Therefore, a trigger would be needed to break the shell of bacteria and release the toxins inside, which would best be regulated by the light intensity. Therefore, the UV inducible promoter characterized by team UT-Tokyo in 2011 has been used as the trigger of lysis mechanism (the T4 lysis system which has been characterized in Parts: BBa_K112000 ; BBa_K112012 ).

We have disintegrated the goal to several steps that would help to accomplish our design:

1. Construct plasmids for expression of toxic proteins.
2. Test the activity of chitinase.
3. Experiment the mortality of each toxins against locusts.
4. Evaluate the SulAp promoter activity by measure the downstream gene expression.
5. Measure the efficiency of T4 lysis.
6. Characterize the mechanism of UV-induced T4 lysis.

Expression of Bt toxin Cry7Ca1:

As the most important ingredient in our insecticide, the synthesized gene fragment for Cry7Ca1 has been amplified in vitro by PCR using primers Cry7Ca1_U and Cry7Ca1_D. Which then fixed with XhoI/ NcoI digested pET28a by SMART assembly to construct plasmid circuit pET28a-Cry7Ca1 (pSS3) for IPTG inductive expression in vivo (E. coli, strain BL21).

Cry7Ca1_F: tttaactttaagaaggagatataccATGGATAAACAGAATGATAGCGGCATAATT
Cry7Ca1_R: tggtggtggtggtggtggccgctgctGCACAGATTTTCCATCAGGAAC

300ml LB medium was inoculated with recombined BL21 and cultured in incubator shaker at 37℃, 220 rpm. When OD600 approaches 0.6, 1mM IPTG was added. Cells were harvested 8 hours later by centrifuging at 8000 rpm, 15 minutes at 4℃; then resuspend in Lysis buffer and disrupted by ultrasonic homogenizer. The bacterial lysate then spined at 12000rpm for 10min at 4℃ to remove cellular debris and supernatant undergoes Ni-NTA purification. Samples have been taken from bacterial lysate, elusion buffer flowthrough and non-induced culture. The SDS-PAGE of them have been performed where results are displayed in the figure below:

We have repeated the expression in DH10B, another E. coli strain which cannot metabolize arabinose hence was also used to evaluate the T4 lysis system. A dimmer but visible band could be observed.

Unfortunately, without official approval to carry out experiment on locusts, we did not manage to further investigate the toxicity of Cry7Ca1. That is why we derived a mathematical model to illustrate the production of Cry7Ca1 in vitro.

Characterization of Chitinase ifchit:

Then gene for ifchit, an endochitinase with molecular mass ~46kDa, from insect pathogenic fungi Isaria fumosorosea, have been codon-optimized and amplified via PCR, (primers: ifchit1_F, ifchit1_R) We assembled the plasmid pET28a-ifchit1 (pSS4) and transformed into strain BL21, allowing the expression to be regulated under pTac system.

Ifchit1_F: aactttaagaaggagatataccatgCTGAGCCTGCTGAAAAAAAGTATCG
Ifchit1_R: gtggtggtggtggtggtggccgctgctTTTCATGCCTTTTTTGATGCTATCG

When OD 600 ~0.6, 1mM IPTG was added and cultured for 10 hours at 37℃. Cells were harvested and resuspend in lysis buffer, and disrupted by ultrasonication. SDS-PAGE then performed to compare the protein composition before and after induction (detailed in our protocols and methods.):

The image shows no clear differences in bands between 50 and 40kDa, we cannot conclude that ifchit has been successfully synthesized under such condition.

After studied the research performed by Huang, Zhen, et al. [1], it was discovered that Isaria fumosorosea tend to grow in a lower temperature. Hence, we reduced the temperature used to 25℃ and 16℃ and re-do the IPTG induced expression & SDS-PAGE analysis.

We use Ni-NTA purification to extract the desired chitinase by adsorbing its 6*His-tag. But the electrophoresis suggested that there are multiple peptides present in the elusion buffer, whereas the length of them varied little in repeated groups. Nevertheless, we evaluated the chitinase activity (defined as the ability to hydrolyze chitin into its monomer N-acetyl-D-glucosamine (NAG) per hour. This monomer reacts with 3,5-dinitrosalicylic acid and form a red compound which absorbs light with 540nm wavelength [2][3].) of bacterial lysate which a negligible value was returned:

These results demonstrated that the lysate of bacteria barely hydrolyzed any colloidal chitin, which means faulty expressions may occurred, whereas the causes behind are still mysterious. Seems like the ifchit synthesis would be better if the temperature lowered below 16℃, but the yield would be much lower and invalid for large amount of industrial production. We have been convinced to aborted to continue the research around ifchit in future and decide to use another pre-investigated chitinase.

Characterization of Chitinase xncht:

Xncht is a chitinolytic enzyme with a prokaryotic origin from Xenorhabdus nematophila (strain 19061), which has been assembled onto pET28a for pTac-controlled expression.

The condition used for expressing this chitinase is much “wildly” compare to the previous one: 37℃ 4 hours, 220 rpm with 1mM IPTG. Cells harvested and lysed in lysis buffer for Ni-NTA purification and SDS-PAGE. The protein composition of each were compared:

The production of xncht over time was detailed in our mathematical model.

The chitinolytic activity assay Kit, (Solarbio, Beijing) was used to test the activity of xncht in bacteria lysate. By comparing to the absorbance of standard solutions with different NAG concentrations, we could predict the number of monomers produced from catabolic digestion of chitin. Detailed in our protocols and methods.

result of chitinase activity analysis. Control: bacteria lysate of uninduced culture; homogenate: lysate of recombined strain (only pET28-xncht) by ultrasonic disruption. The result shows a significant degradation of colloidal chitin when incubate with bacteria lysate which suggests a decent amount of chitinase activity: 0.43μMh-1 for homogenate, 0.29 μMh-1 for T4 lysate.

The result met our expectation and we believed it is important to investigate the activity of xncht when released by T4 lysis (see our suicide mechanism) instead of ultrasonication. The supernatant of self-lysing culture (co-transformed DH10B with pSS5 and pSS8, induced by 1mM IPTG then 1mM L-arabinose) have been extracted to test the chitinase activity within:

T4 lysate: culture supernatant of induced (IPTG+, arabinose+) co-transformed strain with both pET28a-xncht and pSB1C3-pBAD

result of chitinase activity analysis. Control: bacteria lysate of uninduced culture; homogenate: lysate of recombined strain (only pET28-xncht) by ultrasonic disruption; T4 (mediate cell death when L-arabinose present). Chitinase activity: 0.43μMh-1 for homogenate, 0.29 μMh-1 for T4 lysate

It was a shame that we cannot experiment the toxicity of this chitinase on locusts, due to ethic and safety considerations. But according to Mahmood, Saquib, et al., this chitinase have a significant mortality in larvae of H. armigera [4]. It could be assumed that would be lethal to locust as well.

Characterization of UV-inductive promoter SulAp:

This part has been studied and submitted by iGEM11_UT-Tokyo 2011 (BBa_K518010), which has been proved to have the potency to control the downstream gene expression when irradiated by UV. We aim to use this component to induced the T4 lytic protein production when exposed to UV light hence release toxic cytosol to environment (see our T4 lysis design).

By restriction endonuclease digestion and T4 ligation, we assembled the SulAp, eGFP gene onto the plasmid backbone pSB1C3:

The circuit was transformed into E. coli strain BL21, and cultured in 60ml M9 medium. When OD600 approached 0.6, medium was transfer into 12 sterilized petri dishes in bio-safety cabinet and irradiated under UV for different period of time. Then the medium was transferred back to EP tubes and cultured in 37℃. Samples were taken from each of them and fluorescence was measured.

The result of eGFP expression after irradiated by UVC, unit: fluorescence per OD. eGFP-: non-recombinant BL21 irradiated under UVC (without pSS1); eGFP- UV-: non-recombinant not exposed under UV. 0min: recombinant BL21, no UV exposure; 0.5min: recombinant BL21 irradiated under UV for 30 seconds; 1min: recombinant BL21 irradiated under UV for 1 min; 2min: the recombinant BL21 irradiated under UV for 2 min.

The figure above illustrated an average increase of 3-4 times in fluorescence/OD after UV radiation than before, showing that downstream gene expression could be boosted by UV induction, therefore, valid for the regulation of T4-lysis mechanism.

However, we failed to transform BL21 with pSS2 we constructed.

When plated onto agar medium, no colonies developed. We collude our failure to be cause by the leaky expression of SulAp that kills the bacteria quickly. Thus, we have to test the efficiency of T4 lysis under regulated by another promoter.

Evaluation of T4 lysis (suicide) mechanism:

We recombined pBAD promoter (L-arabinose regulated promoter, BBa_I0500) and T4 lysis gene with onto pSB1C3 as vector and transformed into DH10B (an E. coli strain that does not perform arabinose metabolism) for this experiment. We have mathematically modeled this process, click here!

The cells were cultured in LB medium till OD600 ~0.6, then 1mM arabinose was added to induce cell death, measure by OD 600 over time:

The cell death resulted after 1mM arabinose added when OD600 ~0.6. Control: E. coli DH10B strain contain plasmid pSB1C3; pSB1C3-pBAD-T4: DH10B contain pSB1C3-pBAD-T4 (pSS5). OD600 dropped over time indicated the bacterial death resulted by this induction.

Significant cell death can be observed after induction, though we doubt the feasibility of this mechanism to be used to release protein. As a result, a test was carried out on the co-transformed strain DH10B with both pSS8 and pSS5:

Recombined E. coli was cultured at 37℃ till OD ≈ 0.6, then induced by 1mM IPTG for 4 hours. The cell were harvested by spinning at 6000 rpm 20min at 4℃ and supernatant has been discarded. M9 was used to resuspend cells and incubate in 37℃ shakers for recovery (~30min), and 1mM arabinose was added to the culture to induce the T4 lysis. Which 1ml culture was sampled each time and centrifuged to remove bacterial debris and the absorbance of 280nm light of supernatant was compared to calculate (using online protein spectrophotometer: the abundance of xncht released.

the abundance of xncht in supernatant after induction by arabinose (1mM) measured over 42 hours, a positively correlation between induction time and protein concentration in supernatant was observed, with PMCC = 0.48


To conclude, we succeed in the expression of both toxins, which it has been shown that xncht activity is high in both homogenate and T4 lysate. Moreover, SulAp and the T4 lysis devices was tested separately and proven to be valid for cell lysis and protein release mechanism.

On the other hand, a few of our end goals yet accomplished at the end of our experiment: results of ifchit was not as expected and we did not have time to repeat the investigation of protein-release mechanism by UV-inducible T4 lysis. Based on these imperfections, we came with the future plan:

Future plan:


1. improve the SulAp-T4 circuit, with extra LexA gene rise the average inhibition level of SulAp and reduce leaky expression.
2. Study the activity of xncht released by UV-induced T4 lysis.
3. Study the xncht activity under different pH and temperature. A recommended test by prof. Zhang.
4. Assemble plasmid pSS9, ifchit with GFP tagged at C’ terminal. This can help us study whether the primary structure of ifchit was broken during translation. (if not, the purified fragment should emit green fluorescence).


Experiment on locusts, feed locusts with lysate supernatant containing Bt/ ifchit1, (compare with control groups: boiled lysate, ddH2O, chemical pesticides) and record the LC50 (mortality) of each. This test was suggested by prof. Wang and prof. Zhang.

>End goals:

We got the idea of this new end goal from interviews with prof. Zhang:
1. Study the LC50 against locusts with different ages (or sizes)
2. Seal the toxin and lysis genes onto E. coli genome DNA to prevent horizontal gene transfer (contaminations of gene pool.)

The Team Collaboration Special thanks Contribution Description Design Model Result Basic Composite Implementation Awards Judging Form Education Integrated HP Experiment Safety Lab Notebook Close About Us The Project Parts Judging Human Practices Lab