In order to control locust aggregation, we would construct guaiacol and serotonin degradation enzyme system and synchronized lysis circuit which can help the release of enzymes efficiently. Moreover, we also developed thermal response suicide switch to ensure biosafety.
Degradation enzyme system
We want to prevent the locust aggregation, so we search some information about it. According to a Science literature reported in 2009, locusts can receive the stimulus on their hind legs, and stimulate the expression of serotonin in the central nervous system. It changes locusts’ behaviors from suspicion to attraction, and thus begin to expand the group size [1]. In addition, according to a report in the Journal of Insect Physiology in 1973, guaiacol can effectively promote the aggregation behavior of locusts. The conclusion that guaiacol is used as a locust aggregation pheromone has been further confirmed in many subsequent articles [2, 3, 4]. These contents illustrate the important role of serotonin and guaiacol in the formation of locust plagues. It is worth noting that guaiacol is mainly synthesized by the intestinal microorganisms of locusts, and it accounts for a high proportion of all molecules with aggregation effects, which can reach 60%-90%; after entering the environment through feces, it can attract other locusts by volatilization [5, 6]. Moreover, the microbes in locust’s gut can also synthesize serotonin molecules [7]. Considering the high content of guaiacol and serotonin in the intestine, we set up a design to decompose guaiacol and serotonin in the intestine of locusts in order to destroy the aggregation of locusts and prevent locust plagues. Therefore, in order to degrade serotonin and guaiacol efficiently, we constructed a degradation enzyme system for guaiacol and serotonin [8].
For the degradation of guaiacol, the first step is mainly catalyzed by Cytochrome P450 (CYP), to generate catechol (Figure 1) through demethylation. This step of demethylation usually requires some redox chaperone proteins to assist. In an AMB Express. article in 2019, the authors did a very detailed study on the demethylation process of guaiacol. Their data shows that only using two enzymes from G+ bacteria Rhodococcus rhodochrous is enough for guaiacol degradation, namely Cytochrome P450 (CYP, WP_085469912 from R. rhodochrous J3) and Ferredoxin reductase (FR, WP_085469913 from R. rhodochrous J3). These two enzymes can effectively degrade guaiacol to produce catechol (Catechol) and formaldehyde (Figure 2), either with or without additional carbon source supplementation [9]. Therefore, we plan to transfer these two enzymes into E. coli to express and decompose the guaiacol in the intestines of locusts.
Figure 1. The degradation pathway of guaiacol [9]
Figure 2. The degradation effect of CYP and FR on guaiacol in the absence (left) and supplementary carbon source (right) [9]
Serotonin is a very common neurotransmitter. In the past few decades, its synthesis and degradation processes have been thoroughly studied. As shown in Figure 3, serotonin can be catalyzed by monoamine oxidase (MAO) to form 5-HIAL, and then form 5-HTOL under the catalysis of NADPH-dependent aldehyde reductase (ALDR) [10]. It should be noted that there are two subtypes of MAO, of which type A has a higher affinity for serotonin, and MAO-A will also be used as a functional element in our design.
Figure 3. Synthesis and degradation pathways of serotonin [10]
·Experiments
In order to complete the degradation enzyme system for guaiacol and serotonin, we constructed two plasmids PSB4K5_pTac_CYP_FR (PSHS4) and PSB4K5_pTac_MAO_ALDR (PSHS5) by molecular cloning. We transferred these two plasmids into E. coli BL21, separately. After IPTG induction, we found the target protein expressed, except MAO, by SDS-PAGE.
Figure 4. The process we do in order to determine the expressing of protein.
Figure 5. The analysis of SDS-PAGE of PSHS4 and PSHS5. Lane 1 and 2 are samples of PSHS4 and 5 by overnight LB culture. Lane 3 and 4 are PSHS4 and 5 after overnight culture, and then diluted to M9 and cultured to an OD approximately equal to 0.6. Lane 5 and 6 are PSHS4 and 5 adding IPTG for 3h under 37 degrees centigrade.
We can see the result obviously lane 5 and 6, which can prove that the systems we constructed are successful. However, we found that there are also some protein expressions before induction, which we thought was caused by the leakage of promoter. Moreover, we did not see the obvious expression of MAO, and we will find out the cause and optimize our system in future experiments.
Pre-inoculum of PSB4K5_pTac_CYP_FR strains and control PSB4K5 strains were cultured overnight in M9 medium with 10 g/L glucose at 30 °C with orbital shaking at 220 rpm. Guaiacol assimilation experiments were carried out in 250 mL shake flasks with 150 mL of M9 medium and 1.5 ml inoculum in the same conditions described for the pre-inoculum with guaiacol 5 mM. Aliquots were withdrawn regularly for measurement of optical density at 600 nm (OD600). Then we used TPR_China’ NAHR sensor analysis of the concentration of guaiacol.
Figure 6. The experimental process and the comparison of fluorescence intensity between 0h and 28h (the last one)
Figure 7. The degradation effect of CYP and FR on guaiacol in the absence (left) and supplementary carbon source (right)
OD values were measured after sampling at 0, 2, 4, 20, 24, and 28h after adding the final concentration of 5 mM, indicating that the bacteria were growing normally after adding guaiacol.
After the reaction with NAHR Senor of TRP team, which is a sensor that can sense small aromatic molecules, the higher the fluorescence value was, the higher the guaiacol was, and the overall trend of our curve was downward, but there was a rise in the middle, which may be due to the induction effect of the products produced by the degradation of guaiacol. The specific metabolic mechanism needs to be further studied. Time is limited, and we will continue to experiment in the future.
In the future, we determine the activity of enzymes with the synchronized lysis circuit (SLC) because we use this system to improve the degradation effect (more details about SLC can be seen in our page of synchronized lysis circuit in design).
·Modeling
We were unable to complete the experimental verification of the degradation enzyme system for guaiacol and serotonin due to time, so we established a mathematical model to verify the functionality of the experimental design.
In order to secrete enzymes into the locust's lumen efficiently, we introduce the synchronized lysis circuit (SLC). This system can promote the interaction between enzymes and substrates in locust’s lumen and improve the degradation effect (more details about SLC can be seen in our page of synchronized lysis circuit in design).
We assume that the change of protein concentration in the locust's lumen is related to the amount of gene expression, the rate of bacterial lysis, the rate of protein degradation, and the rate of locust intestinal peristalsis. We constructed a equation to describe the change of protein concentration and two equations to predict the change of the concentration of pheromone (more details can be seen in our model page):
Figure 8. The simulation result of protein concentration in locust’s lumen and the comparison of the concentration of Pheromone among with SLC and without it. The first graph shows the dynamics of enzymes. The graph two is the concentration of pheromones in locust with our system and the graph three is without our system.
According to the results showed above, we know that the proteins are released by SLC continuously, and we speculated that the concentration of enzymes can keep in an interval steadily, which is really effective for the degradation of serotonin and guaiacol.
All these simulation results show that our design with the four enzymes and SLC system can degrade the target molecules effectively. In the future, we hope to finish all molecular cloning and test the function of all plasmids in our laboratory as a demonstration.
Synchronized lysis circuit
In order to synthesize enzymes in the intestines of locusts to degrade serotonin and guaiacol efficiently, which can prevent locust plagues promptly, we introduced a synchronized lysis circuit in our design.
We referred to M. Omar Din et al. Nature, 2016, in which the authors designed a very sophisticated bacterial synchronized lysis circuit (SLC). The system contains activator plasmid to express LuxR and LuxI, and lytic plasmid to initiate cell lysis. When the number of bacterial populations proliferates to a certain number, the AHL accumulates and interact with LuxR, which forms LuxR-AHL complex. Then, while the concentration of LuxR-AHL complex reaches the threshold, the complex can bind to the Plux promoter and initiate the transcription of the lytic gene PhiX174E and the reporter gene GFP. So the bacteria will show fluorescence, and then lysis. In addition, after each bacterial lysis cycle, a small part of the surviving bacteria continues to proliferate and synthesize LuxR-AHL again, resulting in a proliferation-lysis cycle. This system is robust and adapts well to different environments. We hope to modify this gene circuit used in this paper and finally adapt it to the locust plague control system.
Figure 9: Synchronized lysis circuit (SLC) in M. Omar Din et al. Nature, 2016.
In our project, we made the following changes to the above experiment: We replaced some of the original documents. MAO&ADH and CYP & FR into the system, in order to degrade serotonin and guaiacol.
Figure 10: The design of our bacterial synchronized lysis circuit (SLC).
In order to verify the feasibility of SLC, we designed the following experiments to verify whether lysis protein Phix174E and quorum sensing components LuxR and Plux have normal functions:
1). Firstly, we need to test if the PhiX174E sequence that we ordered from the company works well. We designed a IPTG-induced PhiX174E expression plasmid to test if E. coli cells can suicide when adding IPTG.
2). Secondly, we should test if the LuxR and plux promoter work well. We designed two exquisite experiments which you can see later.
3). Finally, we would combine these genes together, to test if our bacteria can enter the proliferation-lysis cycle.
Here are the details:
1. Construct the plasmid:
pSHS3_pSB1C3_pTac_Phix174E: Gibson assembly
Ligate pTac（~290bp）and BBa_J23104_K2541001_ Phix174E (~450bp) to pSB1C3
This plasmid is used to test if Phix174E works well.
pSHS6_pSB1C3_pTac_LuxR_plux_sfGFP: Gibson assembly
Ligate pTac_LuxR_Plux_sfGFP（~3.6Kb）and pSB1C3 backbone（~2Kb）
This plasmid is used to test if LuxR and plux promoter work well.
Figure 11: The electrophoretogram of colony PCR of pSHS6.
pSHS7_ pSB1C3_pTac_LuxR_Plux_Phix174E: Gibson assembly
Ligate pTac_LuxR_Plux_Phix174E (~2kb), and backbone PSB1C3（~2Kb）
This plasmid is the combination of LuxR, plux promoter and PhiX174E sequences, which is the final set up in our design.
2. Verify the function of Phix174E lysis protein
In this summer, we did a lot of Gibson assembly experiments to ligate the fragments with PhiX174E. However, we didn’t get one living E. coli colony yet. We thought the reason might be that all the colonies with right sequence of pTac-PhiX174E die due to the leaked expression of pTac promoter. Though we didn’t do the experiments, the clear LB plates verified the lysis function of PhiX174E from the side of the “dead” colonies.
Figure 12: The predicted result of the bacterial expression of pSHS3. After the addition of IPTG, the expression of PhiX174E, which is a self-lysis gene, would kill the bacteria, resulting in the failure of bacterial growth. Bacteria can grow normally without induction
If the construction of pSHS3 plasmid is successfully, we will set up two kind of experiments to test the function of this protein, Figure 12 shows a predicted and ideal result by our imagination.
3. Verify the function of LuxR and Plux
We ligated the pSHS6 and pSHS7 plasmids successfully and the Figure 13 shows the sequencing result of the two plasmids. So we designed several methods to test if Plux and LuxR works well in our experiments.
Figure 13. Sequencing results of Plux-eGFP in pSHS6 and plux-PhiX174E in pSHS7 plasmids.
Figure 14: The results of pSHS6 (left) and pSHS7 (right).
Our experiment uses pSHS6 to test the quorum sensing system by fluorescence, and use pSHS7 to test if the quorum sensing system can induce cell lysis effect. In figure 14(left), if the fluorescence value in the tube increases with time, and the fluorescence value in the tube with IPTG and atc is higher than no inducer or only one inducer, it proves that the bacteria emit fluorescence after reaching the threshold, therefore the experiment is successful. In figure 14(right), if the OD value in the tube increases with time, and the OD value in the tube with IPTG and atc is lower than no inducer or only one inducer, it proves that the bacteria lyse and die after reaching the threshold, so the experiment success.
In our experiments, the results are different from the ideal ones. The possible reason for the unsuccessful experiment is that we built the plasmid on pSB4K5 before, the copy number is relatively low, it is difficult to produce sufficient amounts of LuxI, LuxR, cannot perform effective Quorum sensing reaction.
4. Combine the quorum sensing system and PhiX174E and verify the function of SLC
Though we didn’t get good results of PhiX174E and the quorum sensing system in this summer, we also designed the experiments about how to test if SLC could work. We used the MATLAB software to model the SLC circuit and the results supports our SLC is OK and our design can work quite well.
In our design, after verifying the effectiveness of PhiX174E, LuxR and Plux, we will assemble the gene circuit shown in Figure 10, and verify whether E. coli can show the oscillatory growth effect of SLC like Figure 15 shown, and whether it can release degradation enzymes.
Figure 15: The oscillating effect of bacterial growth.
Thermal response suicide switch
After we construct the system which can prevent the locust aggregation, we planned to research a suicide switch, so that those bacteria can suicide when we do not need them. Taking into account the seasonal characteristics of the locust plague, there are outbreaks in spring and less in summer. We propose that temperature can be a very good solution if a safety switch is needed.
In 2018, the Jilin_China iGEM team designed four temperature-responsive gene switches based on the RNA stem-loop structure, including hot-start, heat-inhibition, cold-start, and cold-inhibition. Among them, the hot-start switch meets our needs. They tested it in DH5alpha bacteria. The data showed that when the temperature is low (≤ 35 degrees), the gene expression level is low, and when the temperature is high (≥ 37 degrees), the gene expression level is higher [11], basically in line with our needs. The advantages of the thermal response switch are: no complicated receptor protein-promoter design is required; the startup conditions are simple, the cost is very low, and the seasonal response is very valuable. Theoretically, we only need to put this sequence after the promoter, before the suicide gene can complete the suicide switch effect.
Figure 16. The structure of thermal response suicide switch (left) and reporter gene expression level at different temperatures (right).
·Expriments
We constructed two plasmids in order to complete the thermal response suicide switch successfully, pSB1C3_BBa_J23104_K2541001_sfGFP (PSHS1) and PSB1C3_BBa_J23104_K2541001_Phix174E (PSHS2).
PSHS1 can generate fluorescence to verify that the thermal response suicide switch is effective. This is the connection between the fragment PUC57_BBa_J23104_K2541001_sfGFP (~2k) and the vector PSB1C3 (~2k) by T4 ligase.
PSHS2 is designed to verify that the bacteria can suicide when the temperature increasing, which is ligated by the fragment PUC57_BBa_J23104_K2541001_Phix174E (~500bp) with the pSB1C3 vector (~2.1k) through Gibson Assembly.
·Results
Each component of bacteria with PSHS1 was divided into 3 biological controls, which were placed at 30℃, 33℃, 37℃, 42℃ with the negative control, positive control and blank control LB, and shaking at 220 rpm. We took samples at 0h, 4h, 8h and 20h respectively, used a microplate reader for spectrophotometer and fluorescence test, and recorded the data in OD600 and Fluorescence 485/510 (Excitation/Emission). We calculated the value of Normalized Fluorescence changes with the equation referenced from Jilin_2018 iGEM team at different temperatures and different time.
Figure 17. The process we do in order to test the function of PSHS1.
We expressed the data are a figure:
Figure 18. Normalized Fluorescence of PSHS1 changes with time at different temperatures. The abscissa represents the time, and the ordinate is the Normalized Fluorescence value. From the image, it can be seen that the Normalized Fluorescence increases with the increase of time. The value of Normalized Fluorescence in 42 ℃ increases most obviously, which proves that the suicide switch is effective at high temperatures.
After we verified the function of K2541001 to sense high temperature, we planned to test if high temperature could induce the lysis process of bacteria with PSHS2, which contains a lysis protein PhiX174E as the downstream of K2541001. In the next experiment, each component of PSHS2 was divided into 3 biological controls, which were shaken at 220 rpm at 30℃, 33℃, 37℃, and 42℃. Samples were taken at 0h, 2h, 4h, 8h, and 20h respectively. We recorded the data in OD600 through microplate reader, and plotted the growth status of PSHS2 at different temperatures over time:
Figure 19. The growth of PSHS2 over time at different temperatures. The abscissa represents time, and the ordinate represents the growth of bacteria. It can be seen from the image that the OD value of the bacteria which is putted in 42°C drops at higher temperatures, which proves that temperature control suicide is effective.
In the 20h, the OD value of the bacteria which is putted in 42°C did not drop significantly as we imagine. Because the temperature in shaking incubator is lower than normal temperature, which may influence the result. Therefore, in the future, we would adjust equipment and measure OD again.
References
[1]. M. L. Anstey et al., Science. 323, 627-629 (2009)
[2]. D. J. Nolte et al., Journal of Insect Physiology. 19, 1547-1554 (1973)
[3]. B. Torto et al., Journal of Chemical Ecology. 20, 1749-1762 (1994)
[4]. D. Obeng-Ofori et al., Journal of Chemical Ecology. 20, 2077-2087 (1994)
[5]. S. Fuzeau-Braesch et al., Journal of Chemical Ecology. 14, 1023-1033 (1988)
[6]. R. J. Dillon et al., Nature. 403, 851 (2000)
[7]. J. Nemcsk et al., Neurochemistry International. 9 (4), 505-510 (1986)
[8]. M. O. Din et al., Nature. 536, 81-85 (2016)
[9]. J. García‑Hidalgo et al., AMB Express. 9 (34) (2019)
[10]. M. Bortolato et al., Handbook of Behavioral Neurobiology of Serotonin. Chapter 2.4, 203-218 (2010)
[11]. http://parts.igem.org/Part:BBa_K2541001