Wet lab Modules
The design of the wet lab modules has undergone several iterations, leading to changes in the experimental setups.
Iteration 1
Design:
In the initial phase, we designed four gene constructs to demonstrate the core proof of concept in our wet lab experiments. These constructs were specifically engineered to address key aspects of our project:
- Expression of the LasR Transcription Factor: A gene construct was designed to test the expression of the LasR transcription factor.
- SERT Expression: Another construct aimed to validate the expression of SERT.
- Functionality Testing: A complex construct was created to assess the functionality of SERT, SNAT, and COMT proteins.
- pLasI Promoter Leaky Expression: The fourth construct was designed to investigate the leaky expression of the pLasI promoter.
During the design phase, we focused on conceptualizing these constructs to align with the project goals and experimental requirements.
Build (In Silico Assembly):
Following the design phase, we utilized bioinformatics tools such as �Snapgene� and �Benchling� for in-silico assembly. This step involved digitally assembling the gene constructs to assess compatibility and plan cloning strategies.
Test (Theoretical Validation):
To validate our designs, we engaged in comprehensive discussions with experienced peers, including seniors and Principal Investigators (PIs). These discussions were pivotal in scrutinizing the theoretical aspects of our construct designs. The insights provided by these professionals served as a critical form of theoretical testing, validating our designs before practical implementation.
Learn and Iterative Improvements:
Upon evaluation, we identified areas for improvement:
- Spacer DNA sequence Inclusion: It became evident that a spacer DNA sequence was necessary between the gene cassettes within a construct, which was initially overlooked.
- Fluorescent Protein Selection: We realized the impracticality of using GFP due to its interference with the plasmid backbone's sfGFP, hindering the distinction between empty and cloned vectors after cloning.
-
Missing Constructs: There was a gap in our prepared constructs, specifically regarding the expression analysis of Serotonin N-acetyl transferase and Caffeic Acid O-Methyl Transferase proteins.
Through this iterative DBTL cycle, we are refining our wet lab modules for a more comprehensive and effective implementation of our project.
Iteration 2:
Design:
Building upon the insights gained from the first iteration, our second design phase focused on refinement and expansion. We introduced a new gene construct, aiming to validate the expression of Serotonin N-Acetyl Transferase (SNAT) and Caffeic Acid O-Methyl Transferase protein (COMT). Additionally, we made strategic changes to enhance the constructs: replacing GFP with TagBFP to prevent Fluorescence Resonance Energy Transfer (FRET) by ensuring non-overlapping excitation and emission wavelengths.
Build (In Silico Assembly and Cloning Strategies):
In the build phase, we employed bioinformatics tools, including Benchling and Snapgene, for in-silico assembly. This step allowed us to assess the compatibility of genetic components and finalize cloning strategies. Attention was given to optimizing the assembly process and ensuring the integrity of our gene constructs.
Test (Expert Review and Validation):
To validate our enhanced designs, we sought expert opinions once again. Engaging with our seniors and Principal Investigators, we underwent a rigorous review process. They provided essential feedback, confirming the validity of our modifications and ensuring alignment with project objectives.
Learn:
Following expert evaluation, we identified areas that required further attention:
- LasR-pLasI Interaction: To comprehensively demonstrate the project�s proof of concept, we recognized the need for a construct specifically testing the ability of the LasR transcription factor to activate the pLasI promoter independently. This experiment was crucial for a detailed understanding of pLasI promoter functionality.
- SERT-pLasI Interaction: Considering the non-native nature of SERT in E. coli, we acknowledged the potential interference of SERT protein with the functioning of the pLasI promoter. A dedicated construct was required to investigate whether SERT could bind to the pLasI promoter, ensuring a thorough evaluation of promoter induction.
Iteration 3:
Design (Iterative Refinement):
Drawing on the insights from previous iterations, our final design phase comprised seven modules. The newly added designed modules aimed to check the following aspects:
- Expression and Binding Analysis: A module was dedicated to evaluating the expression of CUW_0748, the bacterial analog of the Human Serotonin Transporter (SERT), in E. coli. Additionally, it aimed to investigate the potential binding of SERT with the pLasI promoter.
- LasR Expression and Basal pLasI Activity: Another module focused on assessing LasR expression and analyzing the basal activity of the pLasI promoter in the presence of the LasR transcription factor alone.
Build (In Silico Assembly and Cloning Strategies):
In the build phase, we harnessed sophisticated bioinformatics tools, including Benchling and Snapgene, to conduct in-silico assembly. This step ensured the compatibility of genetic components and allowed for the meticulous planning of our cloning strategies.
Test (Expert Evaluation):
Our designs underwent thorough evaluation through consultations with esteemed seniors and Principal Investigators. Expert opinions were sought to scrutinize our gene constructs and explore potential assembly strategies. These discussions contributed significantly to the validation of our designs and the selection of optimal assembly techniques.
Learn and Implementation Strategy:
Upon expert evaluation, we made an informed decision to proceed with practical implementation. Considering various cloning strategies such as Biobrick assembly, Gibson Assembly, Golden Gate Assembly, and Infusion cloning, we opted for the Gate Gate assembly strategy. This strategic choice was made based on its perceived efficiency, especially given the complexity of inserting multiple components into the plasmid vector.
Testing of Module 4:
Design:
In the design phase, a gene construct was meticulously devised to address critical objectives:
- Verification of SERT and LasR Protein Functionality: The construct aimed to validate the functionality of SERT and LasR proteins within the bacterial cell.
- Evaluation of pLasI Activation with Serotonin+LasR Complex: External serotonin, facilitated by the expressed SERT membrane protein, was designed to bind to the LasR Transcription factor. The resulting LasR+Serotonin complex would induce the pLasI promoter.
Build:
During the build phase, essential Biobricks were synthesized de novo. All inserts were meticulously assembled into the pJUMP 28-1A plasmid using the Golden Gate assembly strategy. Precise cloning was achieved utilizing BsaI (Type IIS Restriction Enzyme) to insert the components into the plasmid.
Test:
The cloned plasmid was introduced into E. coli Dh5-alpha. Screening for positive colonies was performed through Colony PCR. Plasmids from the confirmed colonies were isolated from E. coli Dh5-alpha. Further validation included Restriction Digestion using EcoRI, SpeI, and NheI. The plasmid was subsequently transformed into E. coli Bl21 to assess mCherry expression. The BL21 cells were then inoculated in two media: one containing Serotonin and the other without Serotonin.
Learn:
Under the Learn phase, BL21 strain cells were observed using a Confocal microscope:
- In the Absence of Serotonin in the media: Observation revealed minimal mCherry fluorescence intensity, indicating the absence of LasR+Serotonin complex formation. Consequently, the pLasI promoter remained uninduced without serotonin. The observed fluorescence potentially stemmed from the leaky expression of the pLasI promoter.
- In the Presence of Serotonin in the media: Observations reveal increased mCherry fluorescence intensity indicating the functionality of the Serotonin transporter and the formation of the LasR+Serotonin complex.
Constructing the ODE Model of the Probiotic
Iteration 1
Design
Probiotic, We wanted to predict the outcomes of wet-lab results, experiments done on similar systems. We wanted to design an ODE model which extensively describes the system.
Build
A series of Ordinary Differential Equations (ODEs) were formulated to mathematically model the circuit's behavior as per the initial design. The following are the initial ODEs that were formulated after a good deal of literature review and discussion with experts. Certain equations like the induced expression of SNAT and COMT were taken from the iGEM Modeling webinars of 2020, 2021 & 2022 and by tracing back their literature.
$ \frac{dS_e}{dt} = LSP - k_{in}(S_e - S_i) $
$ \frac{d[LasR]}{dt} = \frac{k_{1LasR}.k_{2LasR}.C_N}{d_{mLasR}} - k_{f}[Si][LasR] +k_b[LS] - d_{2LasR}.[LasR] $
$ \frac{d[NAS]}{dt} = (k_{21}[SNAT])\frac{[S_i]}{[S_i]+k_{m1}} - (k_{22}[COMT])\frac{[NAS]}{[NAS]+k_{m2}} $
$ \frac{d[M_i]}{dt} = (k_{22}[COMT])\frac{[NAS]}{[NAS]+k_{m2}} - k_{eff}(M_i - M_e) $
$ [SNAT] = \frac{k_{2SNAT} \cdot k_{1LasI}}{d_{mSNAT}}\: \cdot C_N\: \cdot \left( \beta_{0SNAT} + \frac{(1 - \beta_{0SNAT}).[LS_i]^n}{[LS_i]^n + (K_{dLasI})^n } \right) $
$ [COMT] = \frac{K_{2COMT} \cdot K_{1LasI}}{d_{mCOMT}}\: \cdot C_N\: \cdot \left( \beta_{0COMT} + \frac{(1 - \beta_{0COMT}) \cdot [LS_i]^n}{[LS_i]^n + (K_{dLasI})^n } \right) $
$ \frac{d[LS_i]}{dt} = k_f[S_i][LasR] - k_b[LS_i] $
$ \frac{dM_e}{dt} = \frac{NV_c}{V_l}k_{eff}(M_i - M_e) $
Test
We did a parameter hunt with variety of methods. A rough protocol was followed for finding the parameters.
The goal is to scrutinise the output in reference to standard studies. The reference study considered was done by Byeon et al. Their study is about SNAT and COMT dual expression. Expected plots:
NAS vs Time
A range of simulations were conducted using Python to solve the ODEs. The parameters resulted in the following graphs Concentration(uM) vs Time(s):
The expected outcomes and the actual data didn�t match. The time scales and the trend do not match. Importantly SNAT and COMT have significant concentration differences.
Learn
We discussed these results and tried to troubleshoot the entire process. We seeked out for help from Experts in Computational Biology and Modelling. We cornered out the cause for this discrepancy to be inaccuracy in parameters.
Iteration 2
Design:
The ODEs were redesigned to the following after implementation of quasi-static assumptions.
Build:
We are left with 5 equations and 3 ODEs.
$S_e=S_i(=S)$
$M_e=M_i(=M)$
$L=\frac{\beta_1}{d_1} C_N$
${[SNAT]=\frac{\beta_2}{d_2}\left(\frac{1}{k_{e q}^n L^n+1}\right)}$
$\frac{\beta_2}{d_2}\left(\frac{1}{k_{e q}^n L^n+1}\right) ; k_{e q}=\frac{k_f}{k_b}$
$\frac{d N A S}{d t}=-k_1[S N A T]\left(\frac{S}{S+k_{m 1}}\right)$
$\frac{d M}{d t}=k_2[C O M T]\left(\frac{N A S}{N A S+k_{m 2}}\right)$
Test:
We determined that Extracellular serotonin, Intracellular serotonin and N-acetyl Serotonin(NAS) as the most important and tangible variables. Their plots in the modified model are:
With appropriate values considered for SNAT and COMT, the modified model fitted well into the reference experiment.
Learn:
Owing to the fact that our probiotic system is complex in nature with a myriad of factors involved at every step in the process, a little change in the value of parameters can give a completely different solutions. Discussions with our PIs were extremely helpful in this regard.
We learnt that exactly determining the dynamics exactly of each and every component of a complex system is not feasible. Even if the parameters seem accurate enough on their own regard, their inaccuracy compounds out over all the parameters in question.
To deal with complex systems it is necessary to approximate the system by conveniently and tactically avoiding the variables that are redundant to the total outcome of the system. Here we introduced the concept of quasi-static approximations into our model. Dr Raghunath O Ramabhadran played a key role in this endeavour of solving and approximating the system.
