Our team embarked on an exciting journey to study the growth of two strains of Xanthomonas bacteria: Xanthomonas oryzae pv oryzae BXO43 from the Center for DNA Fingerprinting and Diagnostics (CDFD) and Xanthomonas oryzae pv. oryzae BXO1 from the Center for Cellular and Molecular Biology (CCMB). The aim was to optimize their growth conditions and conduct various assays to gain valuable insights.
Our first task was to create a suitable environment for these bacterial strains. We obtained glycerol stocks of Xanthomonas oryzae pv. oryzae BXO43 from CDFD. To facilitate their growth, we prepared rifampicin PSA plates and covered them with aluminum foil, graciously borrowed from Dr. Sreeramaiah Gangappa's Plant Biology lab.
We streaked Xanthomonas oryzae pv. oryzae BXO43 on two rifampicin PSA plates and placed them in a 28°C incubator for a week. Simultaneously, we received agar stabs of Xanthomonas oryzae pv. oryzae BXO1 from CCMB. These were streaked onto PSA plates without any antibiotics and also incubated at 28°C.
To provide the necessary growth medium, we attempted to prepare 100ml of PSA rich media three times, but each time, it got contaminated. Despite the setbacks, we persevered.
After seven days, we observed the growth of Xanthomonas BXO43 colonies . Unfortunately, one of the plates that originated from the rifampicin-containing colonies was contaminated. To rectify this, we streaked two LB agar plates from the remaining colonies and placed them in the 28°C incubator.
In parallel, we initiated primary cultures of Xanthomonas BXO1 from CCMB and kept them in an incubator shaker at 28°C for 24 hours. We encountered some complications due to issues with agar and LB concentrations from the previous day, but we persisted in our efforts.
One of the primary cultures showed turbidity after 24 hours, indicating growth, but the other one showed no signs of development. We also measured the optical density (OD) of the primary culture, which had reached approximately 0.3.
For further experimentation, we prepared three different growth media: LB, PS, and Nutrient broth, and autoclaved them. Subsequently, we initiated primary cultures of Xanthomonas BXO43 from CCMB in these three different media, originating from two different Xanthomonas-streaked plates, resulting in a total of six primary cultures. These were placed in the 28°C incubator at 250rpm.
With our cultures in place, we began our series of tests and assays. We streaked Xanthomonas oryzae BXO43 and BXO1 from previously streaked plates and closely monitored their growth at various temperatures and time points to optimize their growth conditions.
We also initiated a growth assay of the primary culture of Xanthomonas BXO1 in LB media at 28°C and 250rpm, recording their progress at different time intervals.
As our experiments continued, we gathered valuable data on the growth patterns of Xanthomonas oryzae BXO43 and BXO1 under various conditions. We learned about the challenges of media contamination and the importance of meticulously maintaining culture conditions.
Through our assays, we gained insights into the optimal growth conditions for Xanthomonas BXO43 and laid the foundation for further research and experiments aimed at better understanding these bacterial strains.
In the early stages of our project, we recognized that OMV (Outer Membrane Vesicle) isolation was the foundational step we needed to master. With a clear goal in mind, we embarked on a quest to find the most effective procedures for OMV isolation.
To guide us in this endeavor, we drew inspiration from the previous year's iGEM IISER TVM Team in 2022, who had worked extensively with hyper vesiculating strains as part of their project. We reached out to a former member of that team, Sneha PR, who kindly agreed to mentor us and help us obtain the hyper vesiculating E. coli strain we needed.
Simultaneously, we scoured the academic landscape to identify professors with expertise in hyper vesiculating E. coli strains and OMV isolation. Thanks to our Principal Investigator (PI), we reached out to these professors via email. Fortunately, our efforts bore fruit, as we secured assistance from several professors and obtained strains from their laboratories.
Our next step was to delve into the study of these strains and compare their Outer Membrane Vesicles. To do this, we needed a robust protocol for OMV isolation and purification. Seeking guidance, we met with Dr. Amirul Islam Mallick, who offered valuable insights into potential characterization studies we could undertake for OMVs
With guidance from Afruja, a PhD student from Dr. Amirul Islam Mallick's lab, we began the process of isolating OMVs.
Armed with our isolated OMVs, we proceeded to conduct various tests and characterizations. We employed scanning electron microscopy (SEM) and transmission electron microscopy (TEM) to explore the OMVs' morphology and structure. These tests yielded successful results, allowing us to gain deeper insights into our OMVs.
TEM image of E.coli Δnlp strain
SEM image
In addition to SEM and TEM, we also performed dynamic light scattering (DLS) to determine the size of the OMVs, further enriching our understanding of these crucial structures.
Through meticulous planning, collaboration with mentors and professors, and hands-on experimentation, we successfully achieved OMV isolation and purification. Our journey in the world of OMVs had just begun, and we had gained valuable knowledge about the procedures and characterization techniques involved.
This phase of our project had set a strong foundation for the subsequent stages, where we would delve deeper into the study of OMVs and their potential applications. With each step, we continued to learn, adapt, and refine our approach, eager to unravel the mysteries of these tiny vesicles and their significance in the field of synthetic biology.
It all started when we stumbled upon a fascinating review article by Tracy Palmer titled "Targeting of proteins to the twin-arginine translocation pathway." In this intricate world of protein transport, we uncovered the Twin-Arginine Translocation pathway. Excited and intrigued by its potential, we began to explore the possibility of utilizing this pathway for our protein of interest: fnCas12a. Our goal was to transport fnCas12a to the periplasm, a crucial step in our project.
Our research unveiled the Tat ABCE complex as the ideal candidate for this ambitious endeavor. To dive deeper, we turned to Uniprot, where we unearthed valuable insights into Tat A, Tat B, Tat C, and Tat E. Armed with knowledge about their sequences, structures, and functions, we were ready to embark on the journey of designing a genetic circuit that would utilize these components.
Designing the genetic circuit was the next logical step. To construct our vision, we sought the necessary genetic elements from the iGEM registry, including a promoter and ribosome binding site (RBS). The circuit took shape as follows: Promoter + RBS + TatA + RBS + TatB + RBS + TatC + RBS + TatE + Terminator. This intricate sequence needed to be carefully assembled for it to work effectively.
With the sequence in hand, we faced the challenge of ordering gene fragments. Our chosen suppliers, Twist Bioscience and IDT, had specific limits on the length of sequences they could synthesize. Thus, we divided the sequence accordingly. Our assembly method of choice was NEB Hi-fi DNA Assembly, following a protocol that ensured each sequence had a 30-base pair overlap with its neighboring fragments. Additionally, the first and last fragments had overlapping sequences with the plasmid pSB1C3.
We didn't leave anything to chance. Before proceeding, we rigorously verified the assembly in silico using Snapgene. After thorough review and double-checking, we placed our order with Twist Bioscience, setting the stage for the next phase.
While waiting for the gene fragments to arrive, we prepared the pSB1C3 plasmid from the iGEM Distribution Kit 2023. This involved reviving the plasmid, establishing a primary culture, and conducting plasmid isolation, ensuring that the concentration was satisfactory.
Once the gene fragments arrived, we subjected them to a critical examination. Gel electrophoresis was our chosen method for this task. We compared the bands to the NEB 1 kb plus DNA ladder, meticulously checking for quality and integrity. With the fragments deemed suitable, we moved forward to the next crucial step.
We needed to linearize the plasmids to facilitate the upcoming assembly. For this, we designed primers with overlapping base pairs with the first and last fragment. Both forward and reverse primers were carefully selected after in-silico validation, ensuring precision in our approach.
The journey was not without its challenges. We initiated a gradient PCR at various temperatures, ultimately observing bright bands at 62 and 64 degrees Celsius during gel electrophoresis.
Encouraged by these results, we proceeded with PCR at 64 degrees Celsius for gel extraction, creating three reactions of 25ul each. Gel electrophoresis confirmed the presence of bright bands, affirming the success of our procedure.
However, we encountered unexpected obstacles during the PCR clean-up process using the NEB Monarch Kit. Multiple attempts were made, optimizations were considered, such as extending the elution time and increasing the incubation period, but the problem persisted. After careful examination, we discovered that the PCR Clean-Up kit was nearing its expiration date and had issues.
Adapting to the situation, we embraced the Qiagen Gel Extraction Protocol, with valuable assistance from the Protein Engineering Lab at IISER Kolkata and the Plant Biology Lab under Dr. Sreeramaiah Gangappa. The collaboration proved fruitful as we successfully completed the gel extraction. With concentration and 260/280 ratio within the desired range, we proceeded with the NEB Hi-fi DNA Assembly.
The assembly was followed by transformation into E. coli DH5alpha competent cells, and later into BL21DE3 Electrocompetent cells. With successful transformation and plasmid isolation, we ran further checks using gel electrophoresis, confirming our progress. Finally, we performed Colony PCR for verification and sent samples for sequencing to validate our efforts.
The culmination of our rigorous work and collaboration led to the creation of our assembled product, a vital component for the upcoming double transformation with another circuit, marking another chapter in our scientific journey.
One of the most essential roadblock we had was to do outer membrane vesicles released from gram negative bacteria show interspecies communication. The important question we had was - Is it possible that outer membrane vesicles from E.coli can fuse or interact with Xanthomonas. After a deep literature review we found out studies that have shown interactions between Outer membrane vesicles between same species, quite few papers were shown on interactions between interspecies communication.
We came across literature studies where interaction between membrane vesicles with mammalian cell lines was shown. Hence, we decided to use Lipophilic Dye to stain Outer membrane vesicles and study the interaction with Xanthomonas using Flow Cytometry assay and Confocal microscopy.
The basic design layout of the experiment was built with the help of Scientific Officer Tamal Ghosh. Further, Dr Bidisha Sinha suggested us to use FM1-43 membrane staining dye to stain the outer membrane of Xanthomonas and to better analyze the interaction.
The Confocal microscopy experiments set the base for further future GFP localization testing experiments as well and isolation and interaction studies.
Flow Cytometry Results We first used Rhodamine B dye to stain Outer membrane vesicles and unstained Xanthomonas oryzae.. Then incubated them together for 1 hour in 1:1 ratio of volume. A significant amplification curve was obtained when Xanthomonas oryzae. were incubated with OMV’s suggesting that some interaction was happening. A triplicate of the following experiments was done which showed us same results hence, we decided to proceed and try confocal microscopy.
This cycle was crucial for us to understand that outer membrane vesicles carry the possibility of potential gene transfer agent. We further went on to perform experiment with GFP loaded outer membrane vesicles and positive results were obtained.
Confocal microscopy results The same sample preparation method was done for confocal microscopy as well but first we proceeded with just DAPI as the mounting medium to stain the nucleiod of Xanthomonas oryzae.. And the following positive results were obtained:
Figure : Control Sample of Xanthomonas oryzae.. With DAPI staining
Figure 2: Interaction of Rhodamine stained OMV’s with DAPI stained Xanthomonas, but the localisation can only be observed in membranes.
Then we discussed the results with various microscopy experts and Dr Bidisha Sinha helped us in providing the FM1-43 dye for membrane staining which further showed the following positive results.
This cycle was crucial for us to understand that outer membrane vesicles carry the possibility of potential gene transfer agent. We further went on to perform experiment with GFP loaded outer membrane vesicles and positive results were obtained.
Imaging of just OMV’s show no signal, the aggregate of OMV’s that come together to interact with Xanthomonas show amplification in signal obtained.
We can improve upon this experiment by understanding the size limitation transfer of these proteins and further the above protocol can be used to perform TEM imaging which can give higher resolution images on the interaction taking place between OMV’s and Xanthomonas oryzae..
The idea actually came up during one of our conversations with Prof Susmita Roy when she pointed out that her lab has recently developed this new ML approach for predicting mutational responses in viral genome recently in light of covid outbreak and whether we can incorporate that for our bacterial pathogen. After digging up some literature, we went onboard with the idea. In our study, we employed a feedforward backpropagation neural network characterized by a specific model architecture denoted as 3–8–8–1. This architecture encompasses four distinct layers. The initial layer accommodates three input features, facilitating the input data's entry into the network.Subsequently, we integrated two hidden layers, each containing eight neurons, which play a pivotal role in feature transformation and representation learning. The final layer, consisting of a solitary neuron, corresponds to the network's output and directly addresses the prediction of our target variable.
To test out the validation of the model, we used the trained model to make predictions of a validation sequence. Initially we used the Torch package in python to implement the model. But the prediction did not come as expected. The probabilities were coming in order of 10^-08. It was time for debugging now.
After a lot of debugging and switching from Torch to Keras package, we finally tested our model for predictions. And this time we were successful. The model predictions against the validation file were giving probabilities of the order of 0.1. With a user defined cutoff of 0.15, we checked with actual predictions and found out the model predicted a total of 57 potential positions that can mutate in the future version of XopN. Out of these 57 predicted mutations, the model predicted 9 residues that have actually mutated in the validation sequence.
With the successful prediction of mutational sites, we learned that the features selected for estimating the random mutation worked just as well in bacteria as in viruses. The overall low mutational probability for bacteria might be a relic of the less amount of data availability for model training as compared to the viral case. Also we learned what could be the possible mutations for a given XopN sequence, which in turn can be very helpful for modifying the biocontrol in near future.
