Overview

To understand the insect test in an easy way, the testee selection, and experiment design would be briefly described. The results consist of three parts-feeding assay pre-test, the fusion protein improvement test, and preference test.
**Note: The word “PANTIDE” in the following paragraph refers to the collection of all the toxin design in our project.

• Testee selection

  In the test of PANTIDE toxicity, we chose tobacco cutworm as the testee for the insect appetite tests because all the PANTIDE targets lepidopterans. We hope that we can observe the difference of the three toxins in vivo. Tobacco cutworms, which are one of the major pests in vegetable farms, impact on Crucifers in farmland around the world. In the serious pest damage, the density of tobacco cutworms is approximately up to two hundred million per ha^[1]. Therefore, we used five tobacco cutworms to mimic the much more severe situation in our insect experiment.

• Experiment Design

  All the experiments were to check the functions of PANTIDE for leaf protection, so the observation of the results would focus on the remained area of the leaf disks.
         In the following experiments, we used E. coli BL21 Rosetta-Gami strain to produce PANTIDE. We aimed to evaluate the remained leaf area applying PANTIDE. In this insect test, PANTIDE was diluted into three concentrations to observe the trends of dose response and to confirm the quality of PANTIDE. All the following experiments have three dilution ratios including 1/125 x, 1/25x and 1/5x. Three repeated tests were done in each experiment. See the procedure of insect experiment on Protocols.

• Hypothesis

  According to the mechanism of PANTIDE, PANTIDE is supposed to perform its toxicity in the nervous system of the larvae. We might see larvae paralyze and finally die. So that we may define PANTIDE as a biological pesticide.

Insect Test Result

Feeding Assay Pre-test

To know the qualitative dose response of PANTIDE, we prepared the samples with the sonicated LB solution containing PANTIDE-expressed E. coli Rosetta-Gami strain and diluted it with the three concentration. This experiment is the pre-test that shows us whether the amount of PANTIDE is sufficient enough to perform the toxicity against the larvae. We applied the sample onto the leaf disks and put five cutworms into the separate cabinets for feeding assays. The positive control in the experiment was to apply Bacillus thuringiensis, which is the most widely-used bioinsecticide. We preserved all the result of the remained leaves sealing with the glass paper and calculated the ratio of the remained area on the leaves. The collected data were analyzed by t–test. Here are the feeding assay results.

Feeding Assay Pre-test of Hv1a and Hv1a-lectin

Figure1. Remained leaf disks in the pre-test with the Hv1a and Hv1a-lectin.

Figure1 shows the pictures of the remained leaf disks after twelve hours of feeding assays. After we had done the feeding assays on tobacco cutworms with the three dilution ratio, we measured the area with the software ImageJ.

Figure2. The analysis of remained leaf area in the pre-test with the Hv1a and Hv1a-lectin.

Figure2 shows the ratio of the average remained area on the leaf disks. The higher the bar is, the larger the remained leaves area is. The observed phenomenon can be analyzed below. The p-value can be used to determine if two sets of data are significantly different from each other. The observed phenomenon can be analyzed below.

With the leaves applying E. coli for a negative control, the remained leaves area with PANTIDE were all more than that of E. coli. There is the significantly different with negative control from their p-value. It shows that the sonicated LB solution containing Hv1a/Hv1a-lectin(see the SDS-PAGE) decreases the appetite of tobacco cutworm.
• With the dilution of Hv1a/Hv1a-lectin, the remained leaves area decreased, which clearly shows the dose response.
• Compared Hv1a with Hv1a-lectin, the remained leaves area in the Hv1a-lectin is higher than that of Hv1a, which means the repellent efficiency of Hv1a-lectin is higher than that of Hv1a. There is significantly different between Hv1a and Hv1a-lectin from their p-value.
• Compared with the positive control, we can observe that the repellent efficiency of PANTIDE is equal or even higher than it. There is no significantly different with positive control from their p-value.This shows the efficiency of Hv1a as well. To sum up, the Hv1a and Hv1a-lectin work well in vivo. The effect can be roughly ranked as: Hv1a-lectin > Hv1a > E. coli > Negative control.

We get the similar result in the Sf1a/Sf1a-lectin and OAIP/OAIP-lectin.



Figure3. Remained leaf disks in the pre-test with the Sf1a and Sf1a-lectin.



Figure4. The analysis of remained leaf area in the pre-test with the Sf1a and Sf1a-lectin.



Figure5. Remained leaf disks in the pre-test with the OAIP and OAIP-lectin.



Figure6. The analysis of remained leaf area in the pre-test with the OAIP and OAIP-lectin.

The Fusion Protein Improvement Test

In this experiment, we compared the toxicity effect of the three toxin designs including Hv1a/Hv1a-lectin/Hv1a-lectin with GS linker to test the toxicity effect. (For more about the improved PANTIDE with GS linker, see the Link)



Figure7. Remained leaves in the GS-linker improvement test
with the Hv1a, Hv1a-lectin, and Hv1a-lectin GS-linker improved.



Figure8. The analysis of remained leaves in the GS linker improved test
with the Hv1a, Hv1a-lectin, and Hv1a-lectin GS linker improved.

To know how the length of the linker affect the function of the fusion protein, we constructed Hv1a-lectin with the original three-alanine linker and a longer GS linker. Besides, there is also a Hv1a without the linker between the repellent peptide and the lectin served as a comparative sample. Figure8 shows the pictures of the remained leaf disks after twelve hours of feeding assays. After we had done the feeding assays on tobacco cutworms with the three dilution ratio, we measured the area with the software imageJ. Figure9 shows the percentage of the average remained area on the leaf disks. The higher the bar is, the larger the remained leaves area is. The observed phenomenon can be analyzed below.

• We purified PANTIDE into four quantitative concentration with 0.4nM, 2nM, and 10 nM. The remained leaves area decreased as the concentration of PANTIDE decreased, which shows the dose response of PANTIDE.
• In comparison with the two other design, we could find out that the remained leaves area with improved Hv1a-lectin were all more than that of the original Hv1a-lectin and Hv1a. The result shows that by enhancing the length of the linker, the fusion protein works performs its toxicity better.

Preference Test

In the preference test, we wanted to test if the larvae will have any preference on the leaf disks. There was an interesting finding that tobacco cutworms prefer the negative control to the sonicated or purified PANTIDE. The results are shown in the following video.

In the two-hour test, we can see that most of the tobacco cutworms move to eat the negative control group, which proves the repellent effect of PANTIDE.

Conclusion

After the three experiments, we have three conclusions.

• PANTIDE performs its toxicity as a repellent in vivo. It is a noticeable finding that goes beyond our original hypothesis. The mechanism of the repellent effect is so far unknown.
• The fusion proteins of the three toxins(Hv1a/Sf1a/OAIP) with snowdrop lectin respectively enhance the repellent effect.
• The toxicity efficacy of PANTIDE is equal to Bacillus thuringiensis.
• With the elongation of the linker, the improved fusion protein has the best repellent effect.
• The tobacco cutworms have a preference on the leave disks without applying PANTIDE.

Although PANTIDE does not act as a biological pesticide as our hypothesis had mentioned, we may conclude that it is a biological insect repellent, which corresponds to our ideology of lowering the artificial impacts on the environment and simultaneously achieve effective pest control.