Overview

CAU-China team aims to utilize rhizobium's microaerobic environment to make nodules an anaerobic biofactory, which will provide a new solution to the problems faced by anaerobic or hypoxic production. On this page, we show how experimental evidence and modeling results have validated our design and propose a plan for future validation.

WET LAB

Was the effect on chassis bacteria and plants successfully verified?

Proof of concept: The nitrogen fixation function of the bacteria deleted the phaC2 gene was not significantly altered, and was the same as that of the wild-type strain in terms of plant promotion. In contrast, chassis with the addition of the plasmid attenuated their own nitrogen fixation, leading to a decrease in the dry weight of the above-ground part.

Future steps: We did not study the effect of the modification on chassis at the molecular level. We found alterations in the morphological characteristics of the deleted phaC2 gene bacteriophage, but did not study the alterations in their metabolic processes. We need to further explore the stress of the modification process on the chassis and the effect on the plants in future experiments.

Was the phaC2 gene successfully deleted?

Proof of concept: The protein encoded by the phaC2 gene is an important enzyme in the synthesis of PHB from acetyl co-enzyme A. In order to increase the production of PUFA (namely polyunsaturated fatty acids), we attempted to delete the phaC2 gene to increase the amount of substrate required for its synthesis. We validated the deletion process in two ways: colony PCR and electron microscopy scanning. Colony PCR demonstrated the deletion of the phaC2 gene at the molecular level; and electron microscope scanning demonstrated the disappearance of PHB particles at the phenotypic level. Finally, the deletion of phaC2 gene was verified to result in an increase in fatty acid content using TLC technique.

Future steps: Due to time constraints, we were not able to perform electron microscope scanning of the bacterium in nodules to verify whether the PHB particles in the symbiotic system were reduced. Meanwhile, we found that the color of rhizobia changed after deletion of phaC2 gene, and we don't know whether it will affect their growth and development. We also need to quantify the change in fatty acid content using gas chromatography and liquid chromatography. This needs to be verified and improved in our subsequent experiments.

Were the regulatory modules successfully constructed?

Proof of concept:To reduce the stress of chassis, we set up regulation module to control the function of rhizobial synthesis in a time- and zone-specific manner. We first validated the glnK promoter, which is regulated by nitrogen concentration, and the nifH promoter, which is regulated by oxygen content. The change in fluorescence intensity proved that our prediction of promoter function was reasonable. Next, we constructed the regulation module and verified the integrity and correctness of the module at the molecular level using colony PCR. Then we utilized the fluorescent protein on the regulation module to verify the expression of the module under different conditions. At the fluorescent protein level, we successfully verified that the module was controlled by two environmental factors.

Future steps:We have not been able to validate our ideas in plant system, which is the biggest problem we have. The second is that we did not assay the sgRNA content at the transcriptional level, therefore we could not quantitatively verify whether the sgRNA is regulated. We also did not couple it to a synthetic module and therefore failed to clarify whether the regulation module controls downstream gene expression effectively.

Was the synthesis of the target product successfully completed?

Proof of concept:We selected two types of products, dipeptide aldehydes and PUFA (namely polyunsaturated fatty acids), for the validation of the synthesis module. Using real-time fluorescence quantitative PCR, we detected the expression of the bgc33 gene and the pfa biosynthetic gene cluster at the transcriptional level; using direct mass spectrometry, we detected the production of dipeptide aldehydes; using TLC, we found that the fatty acid content of the +pfa strain was increased; and using GC, we detected the synthesis of new long-chain fatty acids. These results successfully demonstrated the operability of our synthesis module.

Future steps:We were not able to identify the production of dipeptide aldehydes, and need to follow up with structural characterization of dipeptide aldehydes using LC-MS. As for the pfa biosynthetic gene cluster, although newly synthesized long-chain fatty acids were found, we still did not get our target product. Subsequent improvement of the synthesis module is needed to reduce attenuation at the transcriptional level, improve gene expression and increase the yield of the target product.

Was the suicide module successfully constructed?

Proof of concept:To prevent the escape of chassis, we set up a suicide circuit to guarantee biosafety in the form of a “AND gate”. We first validated the glnK promoter, which is regulated by nitrogen concentration, and the nifH promoter, which is regulated by oxygen content. The change in fluorescence intensity proved that our prediction of promoter function was reasonable. Next we constructed a suicide circuit and verified the integrity and correctness of the circuit at the molecular level using colony PCR. Then we utilized fluorescent proteins on the suicide circuit to verify the expression of the module under different conditions. At the fluorescent protein level, we successfully verified that the circuit was controlled by two environmental factors. And then we measured the OD value of the bacterial solution under different conditions of incubation and proved the normal expression of the toxin under high nitrogen. We also observed changes in the morphological characteristics of the colonies under different incubation time.

Future steps:First, we failed to validate the quantitative relationship between toxins and antitoxins at the protein level, thus making it impossible to determine the true effect of the toxins on the organism. Secondly, we failed to validate the suicide circuit at the plant level to know the true effect of our design. Finally, we need to further validate the growth of the bacteriophage under different conditions to rule out the effect of population effect on the suicide circuit.

DRY LAB

Can our software be used by others in the future?

Proof of concept: Others have used our software to verify that it is useful, and its performance is greatly enhanced compared with past tools. Also, it is friendly to non-professional users and works without a network connection. So, it seems that our software has great potential for application in the future. Our software will be a great help for teams that need to explore changes in substances in the pathway!

Future step: Although we have built a very convenient system, unfortunately, the amount of data we have collected is far from enough to allow users to retrieve almost all pathways. In the future, we will try to add more models to make the software more comprehensive and durable.

Can our nitrogen regulation system be applied to other rhizobia?

Proof of concept: In Rhizobium etli, the basic parts of nitrogen regulation have been validated in studies to be feasible. The literature demonstrates that the NtrC protein can bind to and function with the glnK/amtB operon. We predicted the protein structure by Alphafold3 and the protein-DNA docking by HDOCK, and the predicted results show that the same regulatory mechanism exists in our chassis S. fredii CCBAU45436, which indicates that the bacteria we chose has a similar nitrogen regulation mechanism as the one described in the literature.

Future steps: We only predicted the existence of a mechanism related to the NtrC protein as a nitrogen regulatory element in two species of rhizobia. To validate the broad spectrum of nitrogen regulatory systems in this project, we should expand our predictions to explore the regulatory mechanisms in more rhizobia in the future. In addition, we can utilize gromacs to accomplish further molecular dynamics simulations.

Can our model predict DHA yield?

Proof of concept: Our DHA yield prediction model simulates DHA production using Sinorhizobium, incorporating biochemical pathways and environmental factors. Initially, we developed a static FBA model, but time dynamics required transitioning to an ODE model. Sensitivity analysis highlighted that PUFA synthase, regulated by nitrogen levels, significantly affects DHA production, leading to the incorporation of a Nitrogen Absorption Model. This integration allows us to predict DHA yield under varying nitrogen conditions and optimize production based on environmental inputs.

Future steps: Due to the long growth cycle of plants and the difficulty of experiments, validating DHA yield experimentally is challenging. Future work will focus on refining the model with experimental data and exploring additional factors influencing DHA yield to further enhance prediction accuracy.

Can our suicide circuit guarantee the safety?

Proof of concept: In addition to wet lab validation, we also constructed a suicide circuit model to simulate the net expression of VapC toxin under four combinations of nitrogen and oxygen. We used ordinary differential equations (ODEs) to simulate the suicide circuit including gene transcription and translation, inhibition of transcription by sgRNA-directed Cas12k binding to DNA, protein degradation, and interactions. We finally obtained the trend of net toxin change with nitrogen and oxygen over the lifetime of rhizobium, finding that rhizobium would commit suicide directly after nitrogen fertilizer is applied with a spike of escaped toxin.

Future steps: Due to the choice of problems solved by ODEs, lots of parameters are required. However, we didn't find any parameters that fully fit our circuits in the existing studies, so we can only calculate them based on empirical values. We hope that in the future, we can get the specific values of these parameters through experiments and get results closer to reality.