Engineering Success

Design

The purpose of our design was to efficiently and specifically express cytotoxic (Cyt) proteins that have insecticidal action against greenhouse pest larvae, effectively creating a next generation biopesticide. Our engineered microbe will express a cyt cassette under the influence of the IPTG-induced promoter for us to exert tunable gene expression. These Cyt proteins (Cyt1Aa and Cyt2Ba) kill multiple dipteran insect species through the proteolytic cleavage of the Cyt protoxin. This process occurs in the midgut of the insect target and results in pore formation of midgut cells. The idea of our project is that growers could inoculate their growth substrate at any time during their plants’ life cycle and allow our biopesticide to colonize the plant roots. Following the detection of a dipteran infestation of fungus gants, the greenhouse farmers would then spray the inoculated plant with an IPTG solution to express the cytotoxic proteins. Once the fungus gnats eats the root and our biopesticide that are expressing Cyt protein, it will kill the pest before it could cause further plant damage or spread unwanted microbial diseases.

The biological system’s bacterial chassis that has been selected is Bacillus subtilis. B. subtilis was used because of its natural adaptations to plants as it is an endophyte. Important factors include B. subtilis' ability to:

1. Sporulate and survive in situ application,
2. Colonize the root hairs of plants,
3. Release proteins extracellularly in the insect gut and soil,
4. Relative non-toxicity to humans, and
5. Its natural occurrence in the plant microbiome.

Our design was built through the gathering of sequences from Genbank, assembly of existing DNA parts, and synthesis. The pCG004 plasmid that we used in our project is an E. coli to B. subtilis shuttle plasmid that we obtained from addgene to express certain genes. The genetic circuitry we’ve constructed contains a strong artificial promoter (consisting of groE promoter, lac operator and gsiB ribosome binding site (P_grac)) and the cytotoxic genes cyt1Aa and cyt2Ba (Gilbert et al., 2017). These parts have been previously characterized to operate in B. subiltis (Cohen et al., 2008, Manasherob et al., 2001, MoBiTec, 2019; Soberón et al., 2013). Our wetlab goals are split up between Engineering Success and Proof of Concept but in short our specific outcomes were:

1. Engineering Success
1. Design a theoretically functional cytotoxic plasmid in snapgene
2. Clone the plasmids into E. coli
3. Extract and transform the plasmids into B. subtilis
4. Perform a cytotoxicity assay to ensure proper gene expression and function
5. Microscopy to validate fluorescence
2. Proof of Concept
1. Perform in vitro assays to make sure the mutants don’t have colonization defects
Biofilm formation
Sporulation assay
Seed colonization assay
2. Perform in vivo assays to quantify mutant efficacy
1. Root colonization assay
2. Perform a time-course toxicity assay
3. Perform a greenhouse field trial

For Engineering Success we will describe our process of creating and initially validating our strains before profiling their activity further on our Proof of Concept webpage. Our full set of data and protocols are open source and available for the public to view on our href="https://2022.igem.wiki/guelph/proof-of-concept">Proof of Concept and href="https://2022.igem.wiki/guelph/experiments">Experiments pages. The protocol that we developed and our cassette designs are available to the public so that our experiments can be repeated with the same analytical results.

Build

---

GEM Guelph 2022 built the Cyt1Aa and Cyt2Ba cassettes as a whole new component. The Pgrac feature is located within the pCG004 plasmid, therefore it is not needed in the Cyt1Aa or Cyt2Ba cassettes. For the Cyt1Aa cassette, our method of assembly included the addition of a helper protein (P20) upstream of the cyt1Aa sequence (BBa_K2938003) and mScarlet downstream of the cyt1Aa sequence (Figure 1). This helper protein drastically reduces the lethal effects of Cyt1Aa on the expression cell and was included in our cassette for this reason (Manasherob et al., 2001). For the Cyt2Ba cassette, the P20 protein is not needed, as Cyt2Ba readily forms inclusion bodies preventing cell death (Manasherob et al., 2001). To build the Cyt2Ba cassette, a BFP was included downstream of the cyt2Ba sequence (Figure 1). A purification and ligation of the Cyt1Aa and Cyt2Ba cassettes and pCG004 were cloned using a BsaI restriction enzyme and in this case, resulted in the retention of the BsaI sites. The retention of these sites does not impact the design specification or downstream functions of our build.



Figure 1. Snapgene files of the Cyt1Aa and Cyt2Ba cassettes designed by iGEM Guelph 2022. The Cyt1Aa cassette is co-expressed with a 20kDa helper protein (P20) and mScarlet, while the Cyt2Ba cassette is co-expressed with BFP. Each Cassette is flanked by BsaI sites (not shown).

For the building stage, we used DH5alpha E.coli cells for amplification of the Cyt1Aa- (Cyt1Aa-pCG004) and Cyt2Ba-containing plasmid (Cyt2Ba-pCG004) (Figure 2). The Cyt2Ba protein readily forms inclusion bodies when expressed in both E. coli and B subtilis, therefore the cytotoxic effects on both strains will not be observed. After successful transformation in E. coli, our construct was moved to B. subtilis for final testing of Cyt1Aa and Cyt2Ba expression.



Figure 2. Assembly of the Cyt cassettes into pCG004. Two distinct populations of Cyt-containing plasmids were made. The Cyt1Aa-pCG004 construct is displayed on the left while the Cyt2Ba-pCG004 construct is on the right.

Because our destination plasmid is an E. coli to B. subtilis shuttle vector, we do not need to additionally modify the vector or express different promoters. The Cyt cassette was designed by our team and synthesized by IDT, and pCG004 was obtained from addgene. To move the construct from E. coli to B. subtilis we carried out a miniprep from overnight cultures of E. coli and transformed into B. subtilis. When transformed into E. coli, the transformants were plated onto ampicillin-containing agar. Only E. coli cells with the correct build will be able to grow on these agar plates. When transformed into B. subtilis, only cells that contained our build grew on chlroramphenicol-containing agar plates (Figure 3). The strains created and used in further experiments are listed below (Table 1).



Figure 3. Transformation of Cyt2Ba-containing plasmid into Bacillus subtilis 1A976.

Table 1. Strains Developed This Year

Complete Strain Name	Strain Name
E.coli DH5alpha	E.coli WT
E.coli DH5alpha pCG004	E.coli Empty Backbone
E.coli DH5alpha pCG004-mScarlet-Cyt1Aa	E.coli Cyt1Aa
E.coli DH5alpha pCG004-BFP-Cyt2Ba	E.coli Cyt2Ba
B.subtilis 1A976	B.subtilis WT
B.subtilis pCG004-GFP	B.subtilis Empty Backbone
B.substilis pCG004-mScarlet-Cyt1Aa	B.subtilis Cyt1Aa
B.subtilis pCG004-BFP-Cyt2Ba	B.subtilis Cyt2Ba

Test

---

The last part of confirming Engineering Success was to test and validate the gene expression of our construct. The assay that validated the effectiveness of our biopesticide revolved around identifying Cyt1Aa and Cyt2Ba’s efficacy in reducing the population of Dipteran insect larvae. We used wingless Drosophila melanogaster as the model insect population in our project, however other insect pests that fall under the order Diptera, such as Fungus gnats, leafminers, and shore files can also be applied to this project model. We modelled the functions of Cyt1Aa and Cyt2Ba through a cytotoxicity assay. Additionally, we wanted to characterise fluorescence of the strains as further on we hope to attempt a root colonization assay in our Proof of Concept and will image successful colonization using fluorescent microscopy.

Cytotoxicity Assay

The Cyt1Aa- and Cyt2Ba-containing plasmids (pCG004) will be induced in a B. subtilis culture. The culture was then used to soak cotton pads that were introduced to the D. melanogaster population (Figure 4) and the insects were scored dead if they are no longer moving for a few minutes and alive if they are moving continuously which was then recorded to quantify the percentage of mortality. Based on our preliminary literature review, we expected to see a decrease in the population size in the treatment conditions. Although we have changed the target species that the initial paper used, we still expect to observe a population decrease because D. melanogaster is within the Dipteran order.



Figure 4. Set up of the Cytotoxicity Assay. Cyt1Aa- and Cyt2Ba-expressing B. subtilis and E. coli strains were soaked into cotton and had D. melanogaster exposed to it for 10 minutes and death was recorded.

For the parameters of the assay flies were separated into three trials using: B. subtilis empty backbone, B. subtilis-Cyt1Aa, B. subtilis-Cyt2Ba, E. coli-Cyt1Aa and E. coli-Cyt2Ba. We included our E. coli mutants to test if gene expression of the Cyt proteins could be expressed at notable levels even in lab strains. This is touching on our values of accessibility as we want as many iGEM teams to be able to address proof of concept biopesticide use in common lab strains (as E. coli and D. melanogaster are common model organisms, more so that B. subtilis and fungal gnats). After 10 minutes of exposure to the mutant strains, their killing activity was quantified using excel by the number of dead flies (Table 2) and the data plotted in the group below using Prism Graphpad Software (Figure 5).

Table 2.Results from the Cytotoixicty Assay

Raw Numbers	Replicate	B.subtilis Empty Backbone	B.subtilis Cyt1Aa	B.subtilis Cyt2Ba	E.coli Cyt1Aa	E.coli Cyt2Ba
# of D. melanogaster killed	1	0/3	2/3	7/7	1/8	2/5
# of D. melanogaster killed	2	0/4	1/3	4/5	2/4	3/6
# of D. melanogaster killed	3	1/3	3/3	1/3	2/3	2/4
Total Killed	-	1	6	12	5	7
Total Tested	-	10	9	15	15	15
Percentage Killed (%)	1	0	66	100	12.5	40
Percentage Killed (%)	2	0	33	80	50	50
Percentage Killed (%)	3	33	100	33	66	50
Average Percentage Killed (%)	-	11	66.3	71	42.8	46.7



Figure 5. Toxicity assay trials of Cyt-containing plasmids. Measurements were taken at the 10 minute mark and the average percentage of killed D. melanogaster was plotted finding the mean from triplicate trials and the Standard Error of the Mean (SEM) was calculated and shown in the error bars. Strain names can be found in Table 1 and the raw data for this graph can be found in Table 2.

Fluorescence Testing

For imaging and quantification of B. subtilis colonization of tomato plant roots, the fluorescent markers needed to be tested in the B. subtilis-Cyt strains (Table 1) and the B. subtilis controls. The strains were imaged in two experiments under conditions that induce fluorescence to test it.

UV Induced Fluorescence

The first test was simply imaging 5 mL cultures grown for 18 hours using a UV-wand to see levels of fluorescence that occurred (Figure 6).



Figure 6. Fluorescence of B. subtilis-Cyt strains when observed under UV light. The strains in order of Left to Right are as follows (A) B. subtilis WT, (B) B. subtilis Empty Bacbone (C)B. subtilis Cyt1Aa, (D)B. subtilis Cyt2Ba.

While the fluorescent colours of GFP, mScarlet and BFP could not be specifically determined, fluorescence was seen in all of our strains that could fluoresce while our WT did not fluoresce. This meant that we could continue to our next experiment, visualising fluorescence using a microscope.

Microscopic Observation of Fluorescence

The second test was one conducted using a microscope to verify the fluorescence of the four B. subtilis strains. Briefly, overnight strains were diluted to OD 0.2 in SD media with ITPG to induce gene expression. They were grown for 1 hour and imaged using the relevant filters in a 96-well plate using a Nikon microscope (Figure 7).



Figure 7.B. subtilis Cyt1Aa fluorescing red due to mScarlet expression as an example image.While B. subtilis Cyt1Aa was used as an example image, all of the other strains had images that demonstrate that the strains do indeed fluoresce the correct colours.

From the images it can be seen that the strains do indeed fluoresce the correct colours.

Learn

---

From the data we obtained, we learned that Cyt1Aa Cyt2Ba transformed B.subtilis and even E. coli can be an effective tool to reduce the population of greenhouse pests of the Dipteran order! Through the toxicity assays of the Cyt1Aa and Cyt2Ba cassettes were able to kill 66 and 71%, respectively, of the D. melanogaster insects tested. Although we anticipated population reduction to some degree as D. melanogaster is Dipteran, we did not expect the reduction to be this high. For this reason, we believe the Cyt2Ba protein and the Cyt1Aa cassette required no further improvement.

We also learned that the strains are able to be fluorescent and are ready for use in a root colonization imaging assay. Something that we noticed with E. coli Cyt2Ba was that we did not observe any fluorescences in white light which is to be expected. However, the E. coli Empty Backbone and E. coli Cyt1Aa had visual indicators of its fluorescence (Figure 8a?). This occurrence can be explained by two occurrences. The first is that B. subtilis expression plasmids often have leaky expression in E. coli cells (Tran et al., 2020). This often results in the unexpected and oftentimes high expression of proteins, also explaining our E. coli-Cyt mutants’ ability to kill D. melanogaster (Tran et al., 2020). The second subsequent occurrence that can explain the visualization of the GFP and mScarlet in E. coli Empty Backbone and E. coli Cyt1Aa and not E. coli Cyt2Ba is that the BFP’s absorption maxima is around 360 nanometers in contrast to the 569 and 501 nanometers of mScarlet and GFP respectively. The wavelength of white light is between 400 and 700 nanometers and can explain why we saw our fluorescent proteins in E. coli Cyt1Aa and E. coli Empty Backbone. For these two reasons, we think a key improvement should be the incorporation of a BFP whose excitation maxima also falls within the white light spectra so that it could serve as a quick visible indication that our Cyt2Ba-containing plasmid was within our transformed E.coli cells, to match our Empty Backbone and Cyt1Aa strains (Lambert, 2022).



Figure 8. Differences in E.coli transformed colony colour. The left panel includes white E.coli Cyt2Ba colonies. The right panel shows, varying degrees of red E.coli Cyt1Aa patches.

Proof of concept

To determine the toxic effects of Cyt2Ba on the Drosophila population, we performed a toxicity assay. We will also be performing accessory experiments (biofilm formation and accumulation, sporulation and seed coating assays) to ensure B.subtilis is behaving as it normally would in the soil environment, however the toxicity assy is our key assay. For the toxicity assay, we need to observe a decrease in the Drosophila population. For the sporulation, germination and biofilm assays the presence of the downstream BFP will be used to determine normal B.subtilis behaviour. These assays will be carried out with controls such as untransformed B.subtilis cells and B.subtilis cells that contain the undigested pCG004 plasmid containing the optional GFP dropout. Additional information on the results of the modelling tests can be found on our proof of concept page.

References:

Bacillus subtilis expression vector pht01 (PGRAC01 type ). MoBiTec Molecular Biotechnology. (2019).

Cohen, S., Dym, O., Albeck, S., Ben-Dov, E., Cahan, R., Firer, M., & Zaritsky, A. (2008). High-resolution crystal structure of activated Cyt2Ba monomer from bacillus thuringiensis subsp. israelensis. Journal of Molecular Biology, 380(5), 820–827.

Gilbert, C., Howarth, M., Harwood, C. R., & Ellis, T. (2017). Extracellular self-assembly of functional and tunable protein conjugates from bacillus subtilis. ACS Synthetic Biology, 6(6), 957–967.

Lambert, T. (2022). CFP at FPbase. FPbase.

Manasherob, R., Zaritsky, A., Ben-Dov, E., Saxena, D., Barak, Z., & Einav, M. (2001). Effect of accessory proteins P19 and P20 on cytolytic activity of cyt1aa from bacillus thuringiensis subsp. Israelensis in escherichia coli. Current Microbiology, 43(5), 355–364.

Soberón, M., López-Díaz, J. A., & Bravo, A. (2013). Cyt toxins produced by bacillus thuringiensis: A protein fold conserved in several pathogenic microorganisms. Peptides, 41, 87–93.

Tran, D. T., Phan, T. T., Doan, T. T., Tran, T. L., Schumann, W., & Nguyen, H. D. (2020). Integrative expression vectors with Pgrac promoters for inducer-free overproduction of recombinant proteins in bacillus subtilis. Biotechnology Reports, 28.