Development of our project design
Focusing on protein degradation method and the treatment of T2D, our project design mainly went through two major phases. After background researches and brainstorming, we first proposed to focus on GCGR degradation as a possible approach for T2D treatment. In this phase of our project, we designed the GCGR degrading module, which is a Trim21 based proteasome dependent system that could degrade GCGR highly specifically (Figure 2a). After trial experiment, this design generally worked as expected and showed promising protein degradation efficiency in HepG2 cells, phenotype changes on glucagon-induced glucose production was also observed.
With the preliminary results, we presented our data to experts and physicians for feedbacks on possible application of such approach on the treatment of T2D. With their feedbacks, we pushed our project design into a brand-new phase to achieve a ‘smart’ control of protein degradation and upgrade our test platform into ex-vivo primary cells. We also integrated virus-based delivery approaches to closely mimic the application condition (Figure 2b). In this phase of project, we designed a new glucose sensing module to pass the intracellular glucose concentration signal to the expression of GCGR degrading module. We also managed to pack our GCGR degrading module into adenovirus systems and characterized its function in primary mouse hepatocytes. Meanwhile, to avoid animal use and test the conjugation of two modules, we built a mathematical model to simulate how our circuit may contribute to the whole-body glucose homeostasis. Seeing forward, we’ll integrate two different modules to determine if our system works in in vitro and ex vivo condition as we predicted in the model.
Figure 2. A brief history of how we developed our project. (a). Schematic representation of our Phase I project design; (b). Schematic representation of our Phase II project design.
GCGR degrading module
The novel GCGR degrading module in our project was designed based on the previously reported Trim-away system. Trim-away is an antibody-dependent, proteasome-based system that is used to accurately deplete endogenous protein2. In such system, microinjected antibody specifically interacts with the target protein, while its FC domain being recognized and bound by the E3 ubiquitin ligase Trim21 at the same time. Once such complex is formed, Trim21 recruits the ubiquitin-proteasome system to the target protein, leading to its destruction (Figure 3a). Though such system showed impressive degrading efficiency and specificity comparing to DNA and RNA manipulation approaches, it is notable that the antibody used in such scheme requires complex techniques like microinjection to be administered into the cells, which limits the in vivo application of such method. Hence, a genetically encoded version of Trim-away would benefit the field by providing a virus/plasmid deliverable approach for rapid and accurate depletion of endogenous proteins.
In our iGEM 2018 project, we tried to meet such challenge by changing the antibodies into a FC-fused nanobody (Figure 3b). Such system can be easily genetic encoded, while is still limited by the number and available peptide sequences of nanobodies that are currently published. To further expand its application, we hereby demonstrate a ligand-receptor-interaction-based approach for protein degradation using genetically encoded parts. In the GCGR case, the ligand of GCGR, glucagon (Gcg), is fused with the IgG FC domain by a GS linker. By co-express such fusion protein with Trim21, we reasoned that GCGR, glucagon-FC and Trim21 may form a ternary complex which induces the ubiquitination and degradation of GCGR before it reaches the cell membrane (Figure 3c).
Figure 3. Schematic representation of how we developed our GCGR degrading module. (a). Antibody based original Trim-away system; (b). Nanobody based PR Predator system we reported in iGEM 2018; (c). Ligand-receptor-interaction-based PR Predator V2 system (in the case of GCGR, we named as GCGR Predator) in our project.
Glucose sensing module
Inspired by the expert feedbacks from our HP work as well as some literatures, we designed a novel glucose sensing system by rewiring the endogenous glucose sensing signal to our synthetic circuit by a hybrid promoter. Forming one of the most important metabolic organs, hepatocytes are capable of sensing multiple nutrients including glucose. Carbohydrate response element binding protein (ChREBP) is one of the major glucose responsive transcription factors in liver3. Under high glucose, ChREBP is dephosphorylated by Protein phosphatase 2 (PP2A) on its S196 residue. The dephosphorylated ChREBP is then capable of entering the nucleus and activate the gene expression of genes containing Carbohydrate-responsive element (ChoRE) sequence. The dephosphorylation on S196 by PP2A is highly related to the concentration of xylulose 5-phosphate (X5P), one of the metabolic products of glucose4,5 (Figure 4a).
Instead of simply using endogenous ChREBP binding promoters, we chose to design a novel hybrid promoter to obtain improved robustness and extensibility. This promoter contains several ChREBP binding sites and a basic mini promoter, which shows minimum transcriptional activity in off condition. To enable robust ChREBP binding among different species, we integrated previously reported ChREBP ChIP-Seq data in both human and mouse to obtain reserved binding motif. Motif enrichment analysis provided us a minimum sequence of CHREBP binding site. Hence, we reasoned that a glucose sensitive transcriptional activation can be achieved by repeating such binding motif several times upstream of the minimum promoter (Figure 4b).
Figure 4. Schematic representation of the glucose sensing module design. (a) Integrating endogenous ChREBP pathway for glucose sensing.; (b). In-silco-assisted design of ChREBP activating promoter sets.
Experiments we used to validate our design
As a proof of concept, we put the GCGR degrading module under the control of CMV promoter to test if Trim21 can successfully bind and degrade GCGR with the help of Gcg-FC. To validate this, we transfected the CMV-GcgFC-2A-Trim21 cassette into HepG2 cells and performed immunoprecipitation experiment to determine if Trim21 interacts with GCGR. Also, we performed immune-blotting assay to test the cellular GCGR load. Moreover, we performed glucagon-stimulated-glucose-production assay and glycogen staining to validate if the depletion of GCGR generate expected phenotype. Quantitative PCR is also used to monitor the expression level of several gluconeogenesis related genes.
To further demonstrate that our protein degrading system can be used in vivo, we rebuilt the mouse codon optimized CMV-GcgFC-2A-Trim21 cassette and packed this expression device into adenovirus, which is then used to deliver the circuit into mouse primary hepatocyte. Experiments similar to those we performed in HepG2 cell line were also performed on primary hepatocyte to see if our circuit can work in an ex-vivo condition.
As for the glucose sensing device, we constructed different promoters with different number of CHREBP binding sites. Using these promoters to drive firefly luciferase expression in HepG2 cells, we could determine if glucose is capable of inducing transgene expression. We also tested our glucose sensing module in a GFP based reporter system, proving this glucose sensing device is adaptable to different downstream genes. Further demonstration was performed on mathematical modeling integrating two modules and human glucose homeostasis model, we further validated our circuit in HepG2 cells by introducing our designed glucose sensing promoter derived GcgFC-2A-Trim21 cassette to test if it works in a glucose level dependent manner.
Reference
1 Auslander, S. & Fussenegger, M. From gene switches to mammalian designer cells: present and future prospects. Trends Biotechnol 31, 155-168, doi:10.1016/j.tibtech.2012.11.006 (2013).
2 Clift, D. et al. A Method for the Acute and Rapid Degradation of Endogenous Proteins. Cell 171, 1692-1706.e1618, doi:10.1016/j.cell.2017.10.033 (2017).
3 Hiromi Yamashita, M. T. M. S. R. K. B. W. J. H. W. S. D. A. K. U. A glucose-responsive transcription factor that regulates carbohydrate metabolism in the liver.
4 Kawaguchi, T., Osatomi, K., Yamashita, H., Kabashima, T. & Uyeda, K. Mechanism for Fatty Acid “Sparing” Effect on Glucose-induced Transcription: REGULATION OF CARBOHYDRATE-RESPONSIVE ELEMENT-BINDING PROTEIN BY AMP-ACTIVATED PROTEIN KINASE. Journal of Biological Chemistry 277, 3829-3835, doi:10.1074/jbc.M107895200 (2002).
5 Jois, T. et al. Deletion of hepatic carbohydrate response element binding protein (ChREBP) impairs glucose homeostasis and hepatic insulin sensitivity in mice. Molecular metabolism 6, 1381-1394, doi:https://doi.org/10.1016/j.molmet.2017.07.006 (2017).