Design

Plasmid Design

We identified the nitrite reductase proteins NrfA and NirB, and therefore the three operons that we wanted to work with, early on in the project.

Operon Breakdown

a. nirBD

• nirB → NirB → Large subunit
• nirD → NirD → Small subunit

NirB: Cytosolic nitrite reductase. Catalyzes the following reactions:

NH₄⁺ + 3 NAD⁺ + 2 H₂O → NO₂⁻ + 3 NADH + 5 H⁺
NH₄⁺ + 3 NADP⁺ + 2 H₂O → NO₂⁻ + 3 NADPH + 5 H⁺

b. nrfHAIJ

Produces four proteins, including one periplasmic nitrite reductase enzyme, NrfA, and other relevant enzymes.

c. nrfABCDEFG

Produces seven proteins, including one periplasmic nitrite reductase NrfA and other structural proteins.

NrfA: Periplasmic nitrite reductase. Catalyzes the following reaction:

6 Fe(III)-cyt c + NH₄⁺ + 2 H₂O → 6 Fe(II)-cyt c + NO₂⁻ + 8 H⁺

Vector Integration Strategy

We planned to insert these operons primarily into the pSEVA vector 648. Since all SEVA plasmids have the same order of restriction enzymes in their insertion region, we added restriction sites to the ends of the operons according to the order in which they were present in the plasmids.. Fortunately, all SEVA plasmids have the same order of restriction enzymes in their insertion region, so we added restriction enzymes to the operons accordingly.

We chose the enzymes EcoRI, BamHI, and HindIII due to their availability in our lab. We were getting our genes synthesized via TWIST Bioscience, and they had a limit of 5000 bp per plasmid. The nirBD operon fits neatly into this limit.

The nrf operons, however, are much bigger. We thus broke them up into two parts each, therefore building four gene fragments:

nrfABCD

nrfEFG

nrHA

nrfIJ

They could then be digested together and inserted into the plasmid. Following is a demonstration of how this would be achieved for nrfABCDEFG and pSEVA648.

Another point to note is the inducible nature of the XylS/Pm promoter. Since it is induced by benzoate, a plant root exudate, the engineered bacterium would only be able to express nitrite reductase when it is in close proximity to the plant roots. If the bacteria were to escape the rhizosphere, the operon would simply shut off, meaning it would populate the soil as any other common soil bacterium. Our pseudo-kill switch control mechanism was thus characterised.

Gene Editing

While our chassis was Pseudomonas putida KT2440, we were advised by Prof Umesh Varshney to demonstrate our plasmids in E.coli first for ease of handling. This meant that we had to optimise our genes for expression in both organisms separately.

Our method of choice was to use the Codon Adaptation Index (CAI), suggested to us by Ramesh Vaidyanathan at TWIST Bioscience. Each amino acid can be encoded by several different codons, but organisms often show specific biases, meaning they use certain codons more frequently than others for the same amino acid. These are called optimal codons. A gene with a high CAI score primarily uses these optimal codons in reference to that particular organism, suggesting it can be translated quickly and produce a large amount of protein. Conversely, a low CAI score indicates the use of rare codons, which can slow down translation and hinder protein production.

The sequences were then cross-checked for the presence of EcorI, BamHI, and HindIII within the genes, and we manually changed triplet codes to express the same amino acids when we found these instances.

Assay Designs

While designing the concentrations and compositions of the various assays, we kept the following broad questions in mind:

1. What C/N ratio(s) is/are best for our particular strain?
2. Is nitrite ammonification more preferred in broth or spread on agar? What can we infer from these results to qualitatively describe the oxygen-dependence of NarL, the native promoter vs XylS/Pm, our inducible promoter?
3. What, if any, is the threshold beyond which nitrite concentration/build-up proves detrimental (stunted growth, death, other observable phenomena)?
4. What, if any, is the threshold beyond which ammonium concentration/build-up proves detrimental (stunted growth, death, other observable phenomena)?
5. Could the presence of a nitrite transporter in P. putida KT2440 at least be inferred through the graph trends?

Biofilm assays

Our major goals with biofilm testing were two-fold:

1. Quantifying the native biofilm-production capabilities of P. putida KT2440.
2. Quantifying the biofilm-production capabilities of the transformed bacterium.

Since laboratory testing of biofilm production requires the use of a stressor, we planned to use tetracycline to induce biofilm formation.

Ammonium testing assays

Our major goal was to test for extracellular ammonium levels in the bacterial culture, since intracellular ammonium could be attributed to the assimilatory nitrate reduction pathway and simply calculated by subtracting initial total nitrogen amounts in the culture media. We decided to use the indophenol test with Berthelot’s reagent, the specifics of which are available in the Experimentation page.

Furthermore, to cross-check our results and verify their reproducibility, we designed assays using the traditional phenol reagent as well as a modified salicylic acid substitute.

Nitrite testing assays

We purchased the Merck nitrite testing kit for colorimetric analysis of nitrite amounts in the culture media pre- and post-bacterial treatment.

Soil bioremediation

We reached out to the researchers in the group Biofunctool at our institute to understand more about their work related to nitrate quantification in soil. This directed us to Berambadi village, one of their major testing sites, where they conducted periodic testing on soil samples. We have analyzed their work and displayed the statistics on our home page.

Water bioremediation

A major advantage of our design is its flexibility. Every piece of the puzzle can be switched out for something that could confer a different functionality. For example, we can easily switch to a facultative anaerobe or an obligate anaerobe to perform this pathway in water. Since most DNRA bacteria have been isolated from waterlogged fields, protein functionality would not be an issue.

Similarly, we can change the inducible promoter to respond to any inducer of our choice. This could tailor our solution to specific plants based on root exudates, and specific water bodies based on algal toxins or geographic location-specific molecules.

Kill Switch

We have theorized approaches to building a viable kill switch tailored to the deployment conditions we are considering. Read all about it on our biosafety and biosecurity page.