Home

Description

Background, Problem, Solution, etc

Engineering

Engineering Success

Wet Lab and Results

Our Wet Lab Procedures and Project Results

Contribution

Useful Contribution for Future iGEM Teams

Team

Team Description and Members

Attributions

Team Activities and Members Involved

Collaboration

Inter Team Collaborations

Human Practices

Silver and Integrated Human Practices

Safety Considerations

Safety Measures and Ethical Considerations

ENGINEERING

OVERVIEW

Here we describe the engineering ideas and the design process of our biosensor. We describe how we grounded our biosensor on a robust engineering framework based on a fluorescent protein biosensor, and the design choices to maximize our chances of success.

Biosensor Design Requirements

Biosensor Design Requirements

We would like to begin with the design requirements for our biosensor, which will influence the engineering design choices.

  • Specificity: The biosensor should accurately identify the specific bacterial strain it is designed to detect without cross-reactivity with other bacterial species or host biomolecules. This may require the careful selection and design of proteins that undergo unique conformational changes in response to target bacterial proteins or other biomarkers.
  • Sensitivity: The biosensor should be able to detect bacterial presence even at low concentrations to diagnose infections in their early stages.
  • Stability: Given the harsh conditions and lack of refrigeration in low-resource settings, the biosensor should be stable at a wide range of temperatures and humidity levels. It should also be resistant to degradation from microbial contamination or other environmental factors.
  • Ease of Use: The biosensor should be designed to be user-friendly, requiring minimal training for operation. Consider simple color change signals or other clear indicators of bacterial presence that can be easily interpreted.
  • Affordability: To be accessible in low-resource settings, the biosensor should be designed with low-cost materials and fabrication processes. (needs concrete value, e.g. < 1 USD / test)
  • Scalability: The design should be scalable for mass production and distribution without a significant increase in unit cost.
  • Rapid Response: The time from sample introduction to result should be minimized to facilitate fast diagnosis and treatment.
  • Portability: The biosensor should be compact and lightweight for easy transportation and use in field settings.
  • Safety: The biosensor should be safe to use and dispose of, with minimal risk of biohazard or chemical hazard.
  • Robustness: The biosensor should provide reliable results across a wide range of operating conditions and should not be overly sensitive to slight variations in handling or environmental conditions.



BIOSENSOR FRAMEWORK: MODULAR PROTEIN BIOSENSOR

The Institute for Protein Design has developed a new class of protein-based biosensors using de novo (from scratch) design principles. These biosensors function as modular molecular devices that switch from a non-luminescent state to a luminescent state upon the binding of a target molecule. The design is simplified by requiring only one target binding domain, which allows for direct readouts in solution. The sensors are highly sensitive and can detect various clinically relevant proteins and antibodies, including those associated with SARS-CoV-2. A specific sensor for the SARS-CoV-2 spike protein demonstrated a detection limit of 15 pM with a luminescence signal that was 50-fold higher than the background level [1]. The modularity and sensitivity of these biosensors highlight the potential for rapidly designing sensors for a wide array of targets. We hypothesized that this platform would fit well with our design requirements.

The figure above shows the two-component biosensor. The framework is composed of a "cage" protein that has a small bit of a nanoluciferase (caged SmBit) sequestered in an alpha helix of the cage. The rest of the luciferase is attached to a "key." When the key and the cage associate, they reconstitute the luciferase and create a fluorescent signal. Inactive and the active complexes are in thermodynamic equilibrium. The cage also has a caged target binding motif sequestered in its helix. When no target is present, the equilibrium favours the closed conformation, with just a background signal. However, when target is present at a sufficient concentration, it binds to the target binding motif on the cage and sterically blocks the sequesteration of the small bit of the luciferase and shifts the equilibrium to the right, giving out a fluorescent signal.

Binder Domains from Bacteriocins

The advantage of the biosensor framework described above is that it is a modular system. When we switch the binding domain, we can obtain a biosensor for a completely different target. In effect, the framework reduces the problem of biosensor design to a problem of finding a high-affinity binder to your target-of-interest. We hypothesized that a class of bacterial toxins, termed bacteriocins, would be perfect candidates for such high-affinity binders. Bacteriocins are proteinaceous or peptidic toxins produced by bacteria to inhibit the growth of similar or closely related bacterial strain(s). These molecules have sparked interest due to their potential application as natural preservatives and alternatives to traditional antibiotics. Bacteriocins are a diverse group of antimicrobial peptides. They have diverse mechanisms of action, such as permeabilizing the membrane and destroying the membrane potential, inhibiting protein synthesis, and degrading nucleic acids. There is a rich body of work characterizing these bacteriocins [8], and they have been found to have potentially very useful properties. For intstance, Certain bacteriocins have a very modular organization. For instance, colicins are typically organized into 3 modules:

  • (1) Binding Module: This module ensures specific and high-affinity binding to a receptor on the target cell. The specificity of bacteriocins often arises from their ability to recognize specific receptors on the surface of susceptible bacteria. This is crucial as it determines the bacteriocin's target range.
  • (2) Translocation Module: Once bound to the target, the bacteriocin needs to cross the bacterial membrane to exert its toxic effect. This is where the translocation module comes into play. It facilitates the transport of the bacteriocin across the membrane and into the target cell.
  • (3) Effector modules: This is the "business end" of the bacteriocin. Once inside the target cell, the effector domain exerts its toxic effect, leading to cell death. The exact mechanism of action varies among bacteriocins – some form pores in the target membrane leading to cell lysis, while others inhibit essential cellular processes, like DNA replication or protein synthesis.

    • We believed that the bacteriocin literature contains very useful biological and structural information that would form a solid foundation for our biosensor designs. The lucCage-lucKey two-component biosensor reduces the biosensor design problem down to finding suitable high-affinity binders against the target. We hypothesized that the high-affinity binder domains of bacteriocins, many of which have been extensively characterized, would be great binder domains to be inserted for our biosensors. Thus, we searched the bacteriocin literature for proteins that were: (1) high-affinity binders against a known bacterial surface protein, (2) well-characterized and whose structures were solved in complex with their cognate receptor, (3) suitable in their secondary structure (contained long alpha helices) for insertion into the biosensor. The following section describes the properties of the selected bacteriocins that are relevant to the engineering.



    1. Klebicin C Specifications

    Target: TolC Pump of Klebsiella pneumoniae

    TolC is one of the most prevalent drug efflux channels in bacteria and is frequently associated with multidrug resistance [7].

    Affinity: KlebC1-254 binds the TolC trimer with a Kd of 35 ± 7 nM

    Important notes for engineering:

    • Truncated KlebC51-254 bound TolC with an affinity approximately tenfold weaker (Kd = 368 ± 16 nM, N = 1.34 ± 0.12)
    • Pre-equilibrium fluorescence measurements revealed binding to be very slow, suggestive of a single-step process occurring with a bimolecular association rate constant (kon) of 1.9 ± 0.1 × 103 M−1 s−1. Ultra-slow association, such as that observed for the KlebC1-254–TolC complex, is often indicative of gross conformational changes accompanying complex formation.
    • This bacteriocin studied in the paper had significant activity against E. coli, which has implications for species specificity of the diagnostic. This sensor alone may not sufficiently distinguish E. coli and Klebsiella.
    • First 60 residues are disordered in solution

    PDB ID: 7NNA (in isolation), 7NG8 (in complex with TolC)

    Key paper: Housden, Nicholas G., et al. “Toxin Import through the Antibiotic Efflux Channel TolC.” Nature Communications, vol. 12, no. 1, July 2021. Accessed 12 Dec. 2022.

    2. Pyocin S2 Specifications

    Target: the iron transporter FpvAI of Pseudomonas

    FpvAI is a TonB-dependent transporter (TBDT) that actively imports the small siderophore ferripyoverdine. PyoS2 tricks the iron transporter FpvAI into transporting it across the outer membrane by a process that is remarkably similar to that used by its endogenous ligand, the siderophore ferripyoverdine.

    Affinity: Kd = 240 pM

    Important notes for engineering:

    • Deletion of the first 45 residues, including the PRR (Δ1–45 pyoS2NTD), decreased pyoS2NTD binding ∼1,000-fold.
    • There is some non-specific binding to TonB1 by the ß-hairpin. Direct binding of pyoS2NTD to TonB1 was shown by ITC (Kd ∼ 1 μM), as well as by cross-linking, and was abolished when the β-hairpin was deleted (Δ1–30 pyoS2NTD)
    • FpvAI was expressed in E. coli TNE012 cells (ompA−, ompB−, and tsx−) transformed with pNGH183 carrying the fpvAI gene from P. aeruginosa PAO1 with an E. coli ompF signal sequence. After isolation of the OM fraction, FpvAI was purified by anion exchange chromatography and SEC.
    • Direct binding of pyoS2NTD to TonB1 was shown by ITC (Kd ∼ 1 μM), as well as by cross-linking, and was abolished when the β-hairpin was deleted (Δ1–29 pyoS2NTD)

    PDB ID: 5ODW

    Key paper: White, Paul, et al. “Exploitation of an Iron Transporter for Bacterial Protein Antibiotic Import.” Proceedings of the National Academy of Sciences, vol. 114, no. 45, Oct. 2017, pp. 12051–56, https://doi.org/10.1073/pnas.1713741114.

    3. Colicin E3 Specifications

    Target: BtuB in E. coli, outer membrane cobalamin transporter

    Affinity: Kd = 0.9 ± 0.2nM

    Important notes for engineering:

    • Because of the predominantly β-strand structure of BtuB, it was possible using far-UV circular dichroism (CD) to measure the changes in helical content of R135, caused by binding to BtuB. Upon complex formation, the helical content of R135 decreased by 12 ± 5%.
    • This is one of the most promising candidates because of the alpha-helical secondary structure of the binding domain.
    • Binder domain is residues 356-411 of ColE3 in the crystal structure.

    PDB ID: 1UJW

    Key paper:Kurisu, Genji, et al. “The Structure of BtuB with Bound Colicin E3 R-Domain Implies a Translocon.” Nature Structural & Molecular Biology, vol. 10, no. 11, Oct. 2003, pp. 948–54, https://doi.org/10.1038/nsb997.

    DESIGN OF PYOCIN-BASED PSEUDOMONAS SENSOR

      For this design, we integrated the pyocinS2 binding domain into the N-terminus of the lucCage. To be able to successfully integrate the pyocin domain, we followed the following steps: (1) All bacteriocins bind through the N-terminal, but the lucCage default integration is the C-terminal. Thus, we reversed the sequence of the lucCage and the lucKey to be able to insert the bacteriocin from the N-terminus. (2) The pyocinS2 binder domain has a ß-hairpin that interacts with TonB and results in unspecific binding. For the design, we truncated the ß-hairpin to prevent such unspecific interactions. (3) Once we had a candidate sequence, we used AlphaFold to ensure that the alpha helices form a continuous helix with no kinks or breaks.

    Figure 1a: Remove ß-hairpin (Δ1–28), depicted in yellow. The bacteriocin is in purple and the receptor is white.

    Figure 1b: AlphaFold model of the designed biosensor sequence. PyocinS2 binder domain is in blue. The alpha helix of the bacteriocin binder smoothly attaches to the lucCage helix.

      DESIGN OF COLICIN-BASED E.COLI SENSOR

      The colicin E3 binder domain that binds to BtuB is particularly suited to integration into the biosensor because its secondary structure is mostly alpha helical.

      Figure 2a: The secondary structure of the binding region of colicin E3 is mostly alpha helical. The receptor is coloured with Adaptive Poisson-Boltzmann Solver (APBS) electrostatics.

      Figure 2b: AlphaFold model of the designed biosensor sequence. PyocinS2 binder domain is in blue. The bacteriocin domain (blue) nicely packs with the lucCage domain.

      REFERENCES

    • [1] Quijano-Rubio, A., Yeh, HW., Park, J. et al. De novo design of modular and tunable protein biosensors. Nature 591, 482–487 (2021). https://doi.org/10.1038/s41586-021-03258-z
    • [2] Zhang, J.Z., Yeh, HW., Walls, A.C. et al. Thermodynamically coupled biosensors for detecting neutralizing antibodies against SARS-CoV-2 variants. Nat Biotechnol 40, 1336–1340 (2022). https://doi.org/10.1038/s41587-022-01280-8
    • [3] Housden, Nicholas G., et al. “Toxin Import through the Antibiotic Efflux Channel TolC.” Nature Communications, vol. 12, no. 1, July 2021, https://doi.org/10.1038/s41467-021-24930-y. Accessed 12 Dec. 2022.
    • [4] White, Paul, et al. “Exploitation of an Iron Transporter for Bacterial Protein Antibiotic Import.” Proceedings of the National Academy of Sciences, vol. 114, no. 45, Oct. 2017, pp. 12051–56, https://doi.org/10.1073/pnas.1713741114.
    • [5] Housden, Nicholas G., et al. “Directed Epitope Delivery across the Escherichia Coli Outer Membrane through the Porin OmpF.” Proceedings of the National Academy of Sciences, vol. 107, no. 50, Nov. 2010, pp. 21412–17, https://doi.org/10.1073/pnas.1010780107.
    • [6] Kurisu, Genji, et al. “The Structure of BtuB with Bound Colicin E3 R-Domain Implies a Translocon.” Nature Structural & Molecular Biology, vol. 10, no. 11, Oct. 2003, pp. 948–54, https://doi.org/10.1038/nsb997.
    • [7] Koronakis, Vassilis, et al. “Structure and Function of TolC: The Bacterial Exit Duct for Proteins and Drugs.” Annual Review of Biochemistry, vol. 73, no. 1, June 2004, pp. 467–89, https://doi.org/10.1146/annurev.biochem.73.011303.074104.
    • [8] Kleanthous, C. Swimming against the tide: progress and challenges in our understanding of colicin translocation. Nat Rev Microbiol 8, 843–848 (2010). https://doi.org/10.1038/nrmicro2454
    Access Wiki Tools View Source Code