Scroll to reveal !!

Results

Overview

  • Introduction
  • In Silico Validation
  • 1) Molecular Docking
  • 2) Molecular Dynamics
  • 3) Directed Evolution
  • 4) Mathematical modeling
  • Lab Validation
  • 1) Bacterial Transformation
  • 2) PCR Amplification
  • 3) Digestion
  • 4) Ligation
  • 5) Culture & Colony PCR
  • 6) Lipofectamine 3000 Transfection
    • Fly to top

    Introduction

    SUPER was designed to be a universal platform for the eradication of autoimmune diseases, taking rheumatoid arthritis as a prototype for our platform, and also to be immunity-friendly by sparing the common side effects of the current treatment, which is immune suppression, so we decided to develop multilayer platform based on Mesenchymal Stem Cells secreting exosomes carrying a CRISPR-Cas system.

    On this page, we will discuss the latest results we have reached until this point to validate our approach in the following order:

    1- In-Silico Simulations

    2- Wet-lab Validation

    In-Silico validation:

    To simulate the different stages of our approach, we had to conduct several in-silico simulations to validate the efficiency of our approach before proceeding to the wet lab phase. These simulations include:

    • Conducting docking simulations to determine the best peptide to be included in the design of our Syn-Notch receptor.

    • Performing Molecular Dynamics (MD) simulations on our chosen complexes to determine their stability within a physiological environment.

    • Using directed evolution approaches to improve the function of our parts.

    • Modeling the action of the Syn-Notch using differential equations.

    1- Molecular Docking

    These simulations were conducted between the Anti-Citrullinated Peptide Antibody (ACPA) receptor of B-cells as the main target in association with three different proteins that are known to be targeted by the autoantibody. These proteins are:

    • Citrullinated Vimentin (CV).

    • The Cartilage Intermediate Layer Protein (CILP).

    • The Cyclic Citrullinated Peptide (CCP1).

    Figures 1,2,3. Demonstrate the 3D structure of the proteins after completing the docking simulations. These structures were visualized using the PyMOL Molecular Visualization System. (A) ACPA-CV, (B) ACPA-CILP, (C) ACPA-CCP1
    Figure 4. We found that the highest docking score was the ACPA and CCP1, with a score of (-298.80).

    These docking simulations were followed by subjecting our proteins to molecular dynamics simulations to analyze their interactions at the atomic and molecular levels, as well as visualize the energy changes of the binding process.

    Molecular Dynamics:

    Using the AMBER Molecular Dynamics Package on Google Colab, we’ve performed molecular dynamics simulations on all of the aforementioned complexes to test their stability in a perturbed environment, similar to what they would encounter within the cellular settings.

    ACPA-CCP1:

    The ACPA-CCP1 simulations showed the best results in terms of structural stability and flexibility. After 4 nanoseconds of simulation, the Root Mean Square Deviation (RMSD) was between 1.5-2.5 Å, which means that the deviation between the atoms of both proteins was minimal, indicating stability of the binding between the 2 proteins. The value of the Root Mean Square Fluctuation (RMSF) also ranged between 2-3 Å, which also indicates minimal fluctuation in the position of each of the residues forming the structure of each protein.

    Figure 5. A display of the MD results of the ACPA-CCP1 Complex

    ACPA-CV:

    As for the ACPA-CV complex, the MD results weren’t as good as with the ACPA-CCP1 Complex. It showed an RMSD value that fluctuated between 2 up to 4 Å, showing more deviation between the position of atoms in the 2 proteins and indicating less stable binding than the ACPA-CCP1 complex. The RMSF calculations were also slightly higher than the ACPA-CCP1 complex, fluctuating between 1-4 Å.

    Figure 6. A display of the MD results of the ACPA-CV Complex

    ACPA-CILP:

    Finally, the results of the ACPA-CILP showed constant increase in the RMSD value, reaching up to 7 Å. This indicates more deviation in the positions of the molecules that construct both proteins, resulting in a less stable binding process compared to the other two complexes.

    Figure 7. A display of the MD results of the ACPA-CILP Complex

    All of these simulations have directed our modeling process towards using the CCP1 protein, as it showed the highest docking score, in addition to being the most stable structure when bound with the ACPA, compared to the other proteins.

    The simulation of the behavior of the 3 aforementioned complexes was important to determine the best protein that can effectively identify and bind with the ACPA receptor of Auto-reactive B-cells. The selected protein can then be integrated into the design of our Syn-Notch receptor, which can significantly improve the specificity of our approach, while maintaining the stability and flexibility of the binding between the 2 structures. This ensures a high level of specificity of our receptor, which is very important as this represents one of the safety measures of our system that allows for precise and specific targeting of the ACPA-presenting B-cells while avoiding normal B-cells.

    CD19-CD19 Ligand:

    As a proof-of-concept of our approach, our wet lab work was conducted through the binding between the CD19 receptor of B-cells and the CD19-ligand that is integrated in the structure of the Syn-Notch receptor. This is due to the fact that to work with ACPA-presenting cells, we would have to extract them from samples of RA patients, so we decided to test the validity of our approach before proceeding to the next phase.

    We’ve also performed docking and MD simulations for the CD19-CD19 Ligand to analyze the binding between them, as well as ensuring the stability of their structures within a simulation of a physiological system. The docking score between the 2 structures was -311.10. However, when the structures were submitted for MD simulations, their RMSD reached a value of 5 Å, as well as the values of their RMSF.

    Figure 8. The 3D structure of the docking between the CD19 and the CD19 Ligand, visualized using PyMOL
    Figure 9. A display of the MD results of the CD19-CD19 Ligand Complex.

    Directed Evolution:

    To improve the features of the proteins used in our project, we turned to the concept of directed evolution, where random mutations can be applied to protein residues in order to direct the protein into expressing better features. The impact of these mutations is then measured in relation to their surrounding residues, and on the protein structure as a whole.

    Using an online tool called EVcouplings, we subjected our proteins to generate mutant variants at different positions of the protein. The effect of these mutations is then measured independently, as well as on the global protein structure, which is called epistatic fitness. Both these parameters correspond to the overall evolutionary fitness of the formed mutant variants, which contributes to improving their functions.

    Herein, we’re presenting the mutational landscape that was generated by EVcouplings for the proteins that are most important in the performance and safety of our therapeutic approach. We’ve chosen to apply these mutations to 4 proteins that are implemented in different aspects of the therapeutic pathway. These include:

    • CD19 Ligand: as it is involved in the primary step of recognizing the auto-reactive B-cells.
      • Figure 10. Depicts the mutational landscape of the CD19 Ligand, generated by the EVcouplings software. It shows that the mutant variant associated with the highest epistatic fitness was (P152S), while the highest independent score was related to the (P162G).

    • The (CD63-L7Ae) loading system: which is designed for loading our therapeutic cargo in the exosomal delivery system once the recognition process is achieved. Afterwards, these exosomes start delivering our cargo to the auto-reactive cells to terminate their activity.
      • Figure 11. Depicts the mutational landscape of the CD63-L7Ae, generated by the EVcouplings software. It shows that the mutant variant associated with the highest epistatic fitness was (N117K), while the highest independent score was related to the (L18N).

    • The therapeutic cargo itself: represented in the CRISPR/Cas12k system: whose role is to target the B-cell Activating Factor Receptor (BAFF-R) gene which is essential for the survival and differentiation of B-cells. This results in initiating an apoptotic response that destroys the auto-reactive cells.
      • Figure 12. Depicts the mutational landscape of the Cas12k, generated by the EVcouplings software. It shows that the mutant variant associated with the highest epistatic fitness was (A8C), while the highest independent score was also related to the (A8C).

    • Our (DART-VADAR) safety switch: this safety system is designed so that our CRISPR/Cas12k cargo can only perform its action within auto-reactive B-cells, further improving the specificity and safety of our approach.
      • Figure 13. Depicts the mutational landscape of the ADAR enzyme, generated by the EVcouplings software. It shows that the mutant variant associated with the highest epistatic fitness was (Q173E), while the highest independent score was also related to the (A182G).

      Mathematical Modelling:

      Binding Model

      The model confirms the use of CCP1( cyclic citrullinated peptide 1) as an external domain for our Syn-Notch receptor as it has the highest docking score for binding to BCR (B-cell receptor) from VIM (vimentin) and CV (citrullinated vimentin) external domains.So, this implies that CCP1 is more stable and has low dissociation rate after binding to BCR; however, VIM and CV show moderate to high dissociation rate after binding to BCR as show in the graphs.

      • CCP1 binding model to BCR
        • Figure 14.The graph describes the binding between the CCP1 to BCR to form a stable binding state with minimal dissociation.

      Designing an engineered exosomes

      After binding the Syn-Notch receptor to its target, its internal domain will start a cascade of signals to produce our CRISPR system coated with exosomes. The exosomes have a receptor to guide them toward the autoreactive B-cells. This receptor has the same composition of the external domain of the Syn-Notch receptor. But to make sure that we use the suitable initiation signal for exosome secretion we simulate it using:

      Mathematical modelling:

      The model confirms the use of the ZF21.16 VP64 signal, from the internal domain of Syn-Notch receptors, to be effective enough to produce engineered exosomes.

      Figure 15.The graph describes the relation between activation of the internal domain of Syn-Notch receptors and its effect on the level of our engineered exosomes.

      By comparing internal domains of other receptors like Chimeric antigen receptor (CAR) ,Antibody scaffold receptor, receptors Activated Solely by Synthetic Ligands (RASSLs), and the exosomes produced after their activation upon binding to their targets, the result was not satisfactory as the signals were not specific to activate our circuit to produce the sufficient amount of our engineered exosomes.

      • Activation model of CAR internal domain and its effect on the level of exosomes.
        • Figure 16.The graph describes the relation between activation of the internal domain of CAR and its effect on the level of our engineered exosomes.

      • Activation model of Antibody scaffold receptor internal domain and its effect on the level of exosomes
        • Figure 17.The graph describes the relation between activation of the internal domain of Antibody scaffold receptors and its effect on the level of our engineered exosomes.

      • Activation model of RASSLs internal domain and its effect on the level of exosomes.
        • Figure 18.The graph describes the relation between activation of the internal domain of RASSLs and its effect on the level of our engineered exosomes.

      Lab Validation

      Step (1) Bacterial Transformation

      We chose the pcDNA 3.1- vector to deliver our genetic circuits to the cells. It was ordered from Thermo Fisher with a 20µg size, which was not enough to deliver all the circuits. So, we used bacterial transformation to amplify the vector.

      Figure 19.This figures illustrates the process of bacterial transformation and selected colonies as Plating of different constructed pCDNA on ampicillin agar plates All plates were incubated at 37 °C for 16 h. The equivalent of 50 µl of bacterial cells after transformation was plated on each plate.

      Step (2) PCR Amplification

      In order to amplify all the DNA parts, we used PCR amplification to reach the desired concentration by using specific forward and reverse primers, running the parts on gel electrophoresis, and then measuring the specific concentration of the running part using Real-Time PCR as shown in the following figures.

      Figure 20

      0.8 % agarose gel electrophoresis of PCR; each well contain one part in the following order, as shown in the following table:

      ID Part Name ID Part Name
      P1 Syn-Notch P6 CX-43
      P2 Booster-1 P7 Guide RNA-BFP-switch
      P3 Booster-2 P8 First MS2
      P4 Loading System P9 Sensor + second MS2 and Cas12k
      P5 Exosomes receptor P10 MCP-ADAR-RFP

      Step (3) Digestion

      pcDNA restriction digestion by Ecor1 and BamH1 :

      We used Ecor1 and BamH1 restriction enzymes as our sites in the plasmid for the ligation within the parts, followed by the running of gel as shown in the following figure:

      Figure 21. 0.8% agarose gel electrophoresis of digested pcDNA3.1(-) ; well P1 represents the uncut vector, respectively; well P2 represents the cut vector, respectively; and M represents 1Kb DNA ladder.

      The amplified genetic parts were labeled with IDs and separated into 3 plasmids:

      Plasmid-1

      Part Length Restriction enzymes
      Syn-Notch 2740 bp
      • EcoRI
      • Hind-3
      Booster gene-1 2490 bp
      • Hind-3
      • Pst-1
      Booster gene-2 2725 bp
      • Pst-1
      • BamHI
      pcDNA 3.1- (Backbone) 5427 bp
      • EcoRI
      • BamHI
      Total Length 13162 bp
      Table-1: Plasmid-1 consists of a pcDNA 3.1- plasmid vector ligated with 3 genetic parts : Syn-notch , Booster gene-1, and Booster gene-2. This plasmid is responsible for the expression of the Syn-Notch receptor on the surface of the transfected cell. The total length of the plasmid after ligation is 13162 bp which will be tested after ligation by gel electrophoresis.

      Plasmid-2

      Part Length Restriction enzymes
      Loading System 1483 bp
      • EcoRI
      • Hind-3
      Exosomal Receptor 1158 bp
      • Hind-3
      • Pst-1
      CX-43 2250 bp
      • Pst-1
      • BamHI
      pcDNA 3.1- (Backbone) 5427 bp
      • EcoRI
      • BamHI
      Total Length 10079 bp
      Table-2: Plasmid-2 consists of a pcDNA 3.1- plasmid vector ligated with 3 genetic parts : Loading system, Exosomal receptor, Cx43. This plasmid will be responsible for producing engineered exosomes in the transfected cell. The total length of the plasmid after ligation is 10079 bp which will be tested after ligation by gel electrophoresis.

      Plasmid-3

      Part Length Restriction enzymes
      Guide RNA-BFP-switch 2725 bp
      • EcoRI
      • Hind-3
      New MS2 496 bp
      • Hind-3
      • Xbal
      Rest of MS2 and Cas12k 2484 bp
      • Xbal
      • Kpn-1
      MCP-ADAR-RFP 2065 bp
      • Kpn-1
      • BamHI
      pcDNA 3.1 (Backbone) 5427 bp
      • Eco-RI
      • BamHI
      Total Length 12845 bp
      Table-3: Plasmid-3 consists of a pcDNA 3.1- plasmid vector ligated with 4 genetic parts : Guide RNA-BFP-switch, New ms2, Cas12k, and MCP-ADAR-RFP. This plasmid is responsible for producing the killing signal for the auto-reactive B-cell. The total length of the plasmid after ligation is 12845 bp which will be tested after ligation by gel electrophoresis.

      We performed the double digestion method for the nine parts in the prefix and suffix of each part with its specific restriction enzyme and applied these parts to gel electrophoresis as shown in the following figure.

      Figure 22. 0.8% agarose gel electrophoresis of digested gBlocks; wells P1-P2-P3 represent our first three parts, which are the contents of the first plasmid, respectively; wells P4-P5-P6 represent three parts, which are the contents of the second plasmid, respectively; wells P7-P8-P9-P10 represent four parts of the third plasmid, respectively; and M represents 1Kb DNA ladder.

      Step (4) Ligation

      The genetic parts were ligated according to their function into three plasmid vectors. The first plasmid is carrying the Syn-Notch receptor and Booster genes; the second plasmid is carrying the loading system, exosomal receptor, and CX-43, finally, the third plasmid is carrying the CRISPR-Cas system and its safety switch.

      Each ligation process was made through sequential ligation.

      Figure 23

      Step (5) Culture & Colony PCR

      After the ligation step, we did culture of the ligated product to specifically select the optimum colonies to screen it using Colony PCR to make sure that our parts were correctly ligated in the plasmid.

      Plasmid-1

      Figure 24.Cell culture plate of transformed pCDNA vector containing insert parts.
      This plasmid contains
      • (Syn-notch)
      • (Booster gene-1)
      • (Booster gene-2)

      Plasmid-2

      Figure 25.Cell culture plate of transformed pCDNA vector 2 containing insert parts.
      This plasmid contains
      • (Loading system)
      • (Exosomal receptor)
      • (CX43)

      Plasmid-3

      Figure 26.Cell culture plate of transformed pCDNA vector containing insert parts.
      This plasmid contains
      • (Guide RNA-BFP-switch)
      • (New ms2)
      • (Rest of ms2 and cas12k)
      • (MCP-ADAR-RFP)
      Then we selected the best colonies for the colony selection process and screened them with colony PCR to make sure that our parts were correctly ligated with the vector.
      Figure 27. This figure illustrates the selection of colonies for our three plasmids. All plates were incubated at 37 °C for 16 h. The equivalent of 50 µl of bacterial cells.

      And this image illustrate our plates collection

      Figure 28

      Then we did Colony PCR to make sure that our plasmid were correctly ligated with its parts, as shown in the following figure:

      Figure 29. 0.8% agarose gel electrophoresis of digested gBlocks; wells P1-P2-P3 represent our first three parts, which are the contents of the first plasmid, respectively; wells P4-P5-P6 represent three parts, which are the contents of the second plasmid; respectively; wells P7-P8-P9-P10 represent four parts of the third plasmid; and respectively M represents 1Kb DNA ladder.

      Step (6) Lipofectamine 3000 Transfection

      And we transfected our plasmids into the Wi-38 intended to validate our approach through the following main steps : structural validation to reflect the successful expression of our biological parts structurally to be presented at the intended site, in the desired amount, and at the right time. and functional validation to ensure their ability to perform what was intended to be done and characterize their action and performance.