Engineering
Overview
Our project began with a bioinformatic analysis, revealing that pancreatic cancer, particularly in gemcitabine-resistant cases, tends to upregulate EGFR and HER2 receptors. To address this, in the initial phase, we first designed a bispecific fusion protein that contains anti-EGFR [Zeronian et al, 2022] and anti-HER2 nanobody [Lamtha et al, 2021] joined by a Fc fragment [Schmidt et al, 1986], aimed to simultaneously inhibit both receptors on pancreatic cancer to induce cancer cell death. However, such complicated design led to poor solubility and failure. In Phase 2, we introduced a secretory signal peptide but still failed to achieve soluble expression. Finally, in Phase 3, we simplified the design by removing the Fc fragment, resulting in the successful expression of a soluble dual-targeting nanobody, namely the Panobody. We used AlphaFold to predict the structure of Panobody and performed molecular docking to confirm its specificity to both EGFR and HER2 receptors [Jumper et al, 2021]. Cell viability assays demonstrated that Panobody exhibited a cytotoxicity effect against wild-type pancreatic cancer cell lines and enhanced the cytotoxicity of gemcitabine in gemcitabine-resistant lines. Further detailed studies are now underway to explore the full potential of this bispecific nanobody for therapeutic applications.
BioBrick Design
Phase 1: Initial BioBrick Design
The first phase of our project involved the design of a novel bispecific fusion protein construct, referred to as pET-24d(+)-HER2-Fc-EGFR. This construct was designed to mimic the components of an IgG by joining two different nanobodies, one targeting EGFR and the other targeting HER2, by a flexible linker and fused to the Fc fragment of an antibody (Figure 1a). The Fc fragment plays a crucial role in enhancing the stability, half-life, and therapeutic efficacy of biologics. It helps extend the circulating life of the protein by interacting with the neonatal Fc receptor (FcRn), promoting recycling and reducing degradation [Gogesch et al, 2021]. Such design was expected to exhibit dual-receptor blockade on EGFR and HER2 of cancer cells to synergically eliminate cancer cells. However, this initial BioBrick design was found to be insoluble in the prokaryotic expression system.
Phase 2: Addition of Secretion Signal Peptide
To solve the insolubility problem, in Phase 2, we introduced the secretory peptide ssSTII1 S-13 to ahead of HER2-Fc-EGFR coding sequence, aiming to direct the protein to the periplasmic space for better folding and solubility in E. Coli [Zhou et al, 2016]. This new construct, pET-24d(+)-sec-HER2-Fc-EGFR (Figure 2a), was designed to guide the bispecific fusion protein into the secretory pathway of E. coli to facilitate proper protein folding. We expected that moving the protein into the periplasmic space could help resolve the aggregation issues and low scretion seen in previous phases.
Unfortunately, despite these efforts, the protein still aggregated and remained insoluble. The addition of the signal peptide did not lead to a successful expression of soluble bispecific fusion protein, indicating that even with secretion targeting, the complexity of the construct was too high for the bacterial system to handle effectively.
As a result of this phase, we concluded that further optimization of this expression system for the soluble production of the bispecific fusion protein in E. coli might not be feasible within our current timeframe and resources. Given the persistent challenges, we decided to simplify the construct by removing the Fc fragment in the next phase to focus on expressing a more manageable, functional bispecific nanobody construct. This shift marks our next attempt to overcome the solubility issues and move forward with a less complex, but still functional molecule.
Phase 3: Removal of Fc Fragment
In Phase 3, we simplified the construct by removing the Fc fragment, resulting in a smaller and less complex molecule consisting of only the two nanobodies, HER2 and EGFR, joined by a flexible linker. This new construct, pET-24d(+)-HER2-EGFR (Figure 3a), was designed to focus purely on the bispecific targeting function without the added complexity of the Fc region.
This simpler structure allowed us to successfully express the protein in a soluble form, marking a significant breakthrough in the project. We named this molecule “Panobody”. We suspected that the previously encountered solubility issues were likely due to the complexity introduced by the Fc fragment and the multiple disulfide bonds that hindered proper folding.
To ensure rigorous experimental controls, we also designed a scrambled sequence of Panobody (Figure 4a) as a negative control for the upcoming functional studies and in vitro experiments. This milestone enables us to move forward with testing the binding and efficacy of Panobody in targeting HER2 and EGFR.
Engineering Cycle
Our project followed an iterative engineering cycle consisting of three main stages: 1) Design and Build, 2) Test, and 3) Learn. Over the course of the competition, we went through three complete cycles in Phases 1 to 3, each time incorporating new insights, experimental results, literature reviews, and feedback to refine our design. Here's a breakdown of the engineering cycle for these phases:
Phase 1: The Bispecific Nanobody Fc Fusion Protein
1. Design and Build: In the initial phase, our goal was to design a bispecific fusion protein drug conjugate targeting both EGFR and HER2, which are upregulated in pancreatic cancer, especially in gemcitabine-resistant cases. We designed a construct consisting of two nanobodies—one for EGFR and one for HER2—linked by a Fc fragment from an antibody. The Fc fragment was expected to enhance the serum stability of the molecule. The plasmid was named pET-24d(+)-HER2-Fc-EGFR.
2. Test: Using the plasmid as shown in Figure 5a, we have successfully transformed the plasmid into BL21(DE3) competent cells (Figure 5b) and expressed the protein (Figure 5c). However, the construct formed insoluble aggregates, resulting in the production of inclusion bodies. Despite various optimization attempts, we were unable to express the protein in a soluble form under these conditions.
3. Learn: From the aggregation of the protein, we hypothesized that the complexity of
the Fc fragment might still be contributing to the aggregation. This led us to rethink our approach,
leading into Phase 2 where we explored directing the protein through the bacterial secretory pathway.
Phase 2: Addition of Secretion Signal Peptide
1. Design and Build: To further improve the chances of obtaining a soluble protein, we incorporated a secretion signal peptide (ssSTII1 S-13L) to guide the protein to the periplasmic space of E. coli, where folding conditions are more favorable. This new construct, pET-24d(+)-sec-HER2-Fc-EGFR, aimed to promote proper folding and disulfide bond formation in the bacterial periplasm, potentially resolving the solubility issue.
2. Test: After expressing the construct in E. coli under conditions for periplasmic targeting, we tested the solubility of the protein. However, despite the addition of the secretion signal peptide, the protein remained insoluble and formed aggregates, even in the periplasmic space.
3. Learn: The failure to achieve soluble expression in this phase suggested that the
overall complexity of the molecule, particularly the Fc fragment, was still causing folding issues.
This led us to consider further simplification, ultimately deciding to remove the Fc fragment entirely
in Phase 3.
Phase 3: Removal of Fc Fragment
1. Design and Build: In this final phase, we simplified the construct by removing the Fc fragment entirely, leaving only the two nanobodies targeting EGFR and HER2, joined by a flexible linker. The new design, pET-24d(+)-HER2-EGFR, resulted in a smaller and simpler protein. We named this version Panobody, focusing on the bispecific targeting functionality without the added complexity of the Fc fragment.
2. Test: Upon expressing this simplified construct in E. coli, we were finally able to achieve soluble protein expression. This marked a significant success in our project, as the simplified design allowed for proper protein folding and solubility.
We also run our sample using LC-ESI-MS to confirm the molecular mass of the sample matches the predicted molecular mass of Panobody, 27851 Da.
3. Learn: The success of this phase taught us that the complexity introduced by the Fc fragment was the primary factor leading to aggregation and insolubility in earlier designs. By focusing solely on the bispecific nanobodies and removing unnecessary elements, we were able to produce a functional, soluble protein. This final construct, Panobody, provided a solid foundation for further functional and in vitro testing. In order to show the efficacy and robustness of our ovel dual-targeting Panobody, we have examined its effect alone and in combination with gemcitabine in our established gemcitabine-resistant pancreatic cancer cell line, PANC-1 by MTT assay followed by Bliss analysis. Details of the experiment results can be found in the Experiment page.
-
Gogesch, P., Dudek, S., van Zandbergen, G., Waibler, Z., & Anzaghe, M. (2021). The role of Fc receptors on the effectiveness of therapeutic monoclonal antibodies. International Journal of Molecular Sciences, 22(16), 8947.
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., ... & Hassabis, D. (2021). Highly accurate protein structure prediction with AlphaFold. nature, 596(7873), 583-589.
Lamtha, T., Tabtimmai, L., Bangphoomi, K., Kiriwan, D., Malik, A. A., Chaicumpa, W., ... & Choowongkomon, K. (2021). Generation of a nanobody against HER2 tyrosine kinase using phage display library screening for HER2-positive breast cancer therapy development. Protein Engineering, Design and Selection, 34, gzab030.
Schmidt, W. E., Jung, H. D., Palm, W., & Hilschmann, N. (1983). Three-dimensional structure determination of antibodies. Primary structure of crystallized monoclonal immunoglobulin IgG1 KOL, I. Hoppe-seyler's Zeitschrift fur Physiologische Chemie, 364(6), 713-747.
Zeronian, M. R., Doulkeridou, S., van Bergen en Henegouwen, P. M., & Janssen, B. J. (2022). Structural insights into the non-inhibitory mechanism of the anti-EGFR EgB4 nanobody. BMC Molecular and Cell Biology, 23(1), 12.
Zhou, Y., Liu, P., Gan, Y., Sandoval, W., Katakam, A. K., Reichelt, M. & Reilly, D. (2016). Enhancing full-length antibody production by signal peptide engineering. Microbial Cell Factories, 15, 1-11.