Description

Food is the essential need for our body and the basic requirement for life. But pest can lead to up to 40% of production reduction and 220 trillion dollars loss, so as human beings, we have been called for pesticides for decades.

Pesticides play a crucial role in controlling pests and diseases, helping to boost crop yields and ensuring agricultural stability. In recent years, the demand for pesticides in global agricultural production has significantly increased, driven by the rapid growth in global population and the rising need for food. However, the heavy reliance on chemical pesticides has introduced several pressing challenges.

Prolonged and excessive use of chemical pesticides has led to a substantial increase in pest resistance. As pests evolve to become resistant, farmers are forced to apply pesticides more frequently and in higher quantities to maintain efficacy. This resistance not only diminishes the effectiveness of pest control but also escalates production costs. The spread of pesticide resistance has become a significant threat to global agriculture, potentially undermining food security.

The extensive use of chemical pesticides has also contributed to the problem of pesticide residue. Although some of the hypertoxic pesticides have been banned in some areas of the world, they remain in the environment and in the body of plants given the reason that they can’t be degraded, which leads to bad influences in crop production, human health, and global commercial.

In response to these challenges, the pesticide industry is undergoing a transformation toward more sustainable practices. Innovations such as biopesticides and Integrated Pest Management (IPM) are being promoted to reduce the dependence on chemical pesticides and mitigate their environmental impacts. Compared to traditional pesticides, biopesticides offer a more environmentally friendly alternative, helping to curb pesticide resistance and protect ecosystems.

In conclusion, while pesticides are indispensable for modern agriculture, their overuse has led to resistance problems and environmental degradation. The global pesticide industry must adopt more sustainable management practices to ensure the long-term balance between agricultural productivity and environmental protection, for which, we focused on Spinosad.

To solve the problems brought by chemical pesticides, our project focused on spinosad, a kind of macrolide pollution-free and high-efficiency biological insecticide extracted from the fermentation broth of \( Saccharopolyspora\:spinosa \), which can quickly paralyze pests and eventually lead to death. Its insecticidal rate is comparable to that of chemical pesticides with higher security rate and no cross-resistance to commonly used insecticides.

In conclution, it is a low-toxicity, high-efficiency and broad-spectrum insecticide so that we can put it into pollution-free vegetable and fruit production application.

Here we summerize some advantages of spinosad compared to traditional pesticides.

Unlike many conventional synthetic pesticides, Spinosad is classified as an environmentally friendly option. It is produced through fermentation and does not persist in the environment. Once applied, Spinosad breaks down photochemically when exposed to light, reducing the risk of long-term environmental contamination. Moreover, it binds to soil particles, which minimizes its leaching potential and lowers the risk of groundwater contamination.

Spinosad degrades into harmless byproducts such as carbon dioxide and water, making it highly biodegradable. Its degradation in the soil is rapid, and it is unlikely to accumulate in ecosystems, which is a common issue with many synthetic pesticides.

Spinosad exhibits low toxicity to mammals, birds, and many beneficial insects, making it safer for non-target species, including humans. This is particularly important in agricultural settings where worker safety and minimizing exposure risks are key concerns. It has been shown to have a much lower acute toxicity than traditional chemical insecticides like organophosphates and carbamates.

Spinosad is highly selective, targeting a wide range of harmful insect pests while sparing beneficial insects such as ladybugs, predatory wasps, and spiders. This makes it an excellent tool for integrated pest management (IPM) programs, as it helps maintain natural pest control populations in agricultural ecosystems.

Spinosad works through a novel mechanism, targeting the nervous system of insects. This unique mode of action helps in managing pests that have developed resistance to other classes of insecticides, such as organophosphates or pyrethroids. This reduces the likelihood of cross-resistance and extends its effectiveness against various resistant pest populations.

In summary, Spinosad offers a sustainable and effective alternative to traditional chemical insecticides, providing strong pest control while minimizing the environmental and health risks typically associated with conventional pesticides. These features make it an attractive alternative in the growing push for greener agricultural practices.

Enzymes, as essential components of life, have long been a research focal point. It play an irreplaceable role in catalyzing biochemical reactions within the \( Saccharopolyspora\:spinosa \), and further participating in determining the purity and unit yield of Spinosa. One of their fundamental characteristics, optimal temperature, serves as a key link between biological activity and temperature. However, due to the need for numerous repetitive experiments in wet labs, the currently available data is very scarce.

Regarding to this, we are inspired that is we use a deep learning model to predict the optimal temperature of key enzymes instead of ecperimenting, we can get more accurate data in a shorter time, so that we can improve the metabolic pathways of \( Saccharopolyspora\:spinosa \) with the goal of increasing purity and unit yield of spinosad.

So this leads to our project: HEATMAP.

In this project, we constructed a deep learning model to predict the optimal temperatures of enzymes in the body of \( Saccharopolyspora\:spinosa \) and compared them to the actual fermentation temperatures to identify enzymes with large differences. These enzymes became our key targets for optimization.

Meanwhile, based on the GEM model, we added enzyme and temperature constraints to constructe the etcGEM model to identify the key enzymes with yield limitations. Targeted evolution of these key enzymes brings their optimal temperatures closer to the actual fermentation temperatures, thus improves productivity.

Regarding the above issues, our proposed technical approach primarily revolves around the classic design, construction, testing, and learning cycle system in synthetic biology, which is DBTL in short.

During the learning phase, we extensively collect enzyme sequences and corresponding optimal temperature data for training AI. Simultaneously, we gather multi-omics data for Streptomyces coelicolor, including genomics, metabolomics, and proteomics.

Next, we build our deep learning model and train it. Additionally, based on the multi-omics data, we construct an ecgem model to identify key enzymes in the multiple antibiotic biosynthesis pathways. In the later stages, we enhance the ecgem model by incorporating AI-predicted optimal temperature data as a variable, creating a more advanced etcGEM model.

Subsequently, for enzymes with significant discrepancies between predicted optimal temperature and actual cultivation temperature, we employ methods like gene editing to modify their temperature range. This adjustment ensures alignment with the cultivation temperature, thereby enhancing catalytic efficiency.

Finally, using experimental data and experience, we further optimize our model through iterative learning to improve prediction accuracy. Based on this technical roadmap, we can guide microbial engineering using our model.

[1]  Santos, V. S. V., & Pereira, B. B. (2019). Properties, toxicity and current applications of the biolarvicide spinosad. Journal of Toxicology and Environmental Health, Part B, 23(1), 13–26. https://doi.org/10.1080/10937404.2019.1689878

[2]  Li, G., Hu, Y., Jan Zrimec et al. Bayesian genome scale modelling identifies thermal determinants of yeast metabolism. Nat Commun 12, 190 (2021). https://doi.org/10.1038/s41467-020-20338-2

[3]  Sánchez BJ, Zhang C, Nilsson A, Lahtvee PJ, Kerkhoven EJ, Nielsen J. Improving the phenotype predictions of a yeast genome-scale metabolic model by incorporating enzymatic constraints. Mol Syst Biol. 2017 Aug 3;13(8):935. doi: 10.15252/msb.20167411. PMID: 28779005; PMCID: PMC5572397.