Our concept involves targeting the STAT3 pathway responsible for solid tumor cell adhesion¹, using honokiol - a small molecule STAT3 inhibitor, specifically at the tumor site, with the help of a small molecule antagonist AM6358.

Fig. 1 The STAT3 pathway responsible for tumor cell adhesion ²

The project has been structured into the following major modules-

• Expression of Cannabinoid Receptor 1 on the T-cell membrane
• T-cell loading with honokiol
• Honokiol release at tumor site using AM6358 (antagonist molecule)

T-cells will be isolated from the patient's bloodstream and cultured in vitro. The Cannabinoid Receptor 1 (CB1) will then be expressed as a membrane bound receptor on the T-cell surface. We propose stable expression (genome integration) of CB1 achieved via retroviral based transduction or transposon mediated genome integration³. To model this in Escherichia coli, we propose the Ice Nucleation Protein (INP) to be employed (from Pseudomonas syringae). We designed a genetic construct based on a part created by iGEM CAU_China (BBa_K3279006). Ice nucleation proteins are expressed on the outer cell membrane of certain Gram-negative bacteria and function to catalyze ice formation of supercooled water⁴. We used this system to express our recombinant protein on the E. coli outer membrane.

The N-terminal domain amino acids of INPs are relatively hydrophobic and link the protein to the outer membrane via a GPI (glycosylphosphatidylinositol) anchor. The C-terminal domain of the protein is highly hydrophilic and exposed to the medium. The central part of INP comprises a series of repeating domains that act as templates for ice crystal formation. The N-terminal domain however, seems to be the only prerequisite for successful targeting and surface-anchoring, hence INP-N is what was employed here⁵.

The construct in a pET28a(+) expression vector:

We ordered the vector containing the INP-N+linker+CB1 construct from IDT in advance but unfortunately due to shipping issues, we didn’t receive it till the 10th of October, 2023. Our studies were therefore performed using the pUC57 vector that only contained the CB1 gene, which had been ordered via a different manufacturer. Additionally, there were also delays in the shipment of this vector and hence, this delayed the overall commencement of our wet lab experiments.
Fig. 2 The STAT3 pathway responsible for tumor cell adhesion

Through our docking simulations, we showed that honokiol has a high binding affinity for CB1. (Refer Docking Simulations section in Model). We propose to load the T-cells with honokiol in vitro prior to introduction into the patient's bloodstream. We plan to model this by incubating the recombinant bacterial culture with honokiol to effectively facilitate their binding. We also planned to perform a concentration dependent assay to confirm that the binding occurred by additionally incubating a control bacterial culture (not expressing CB1) with honokiol, centrifuging both cultures and measuring the absorbance of the supernatant in the range of honokiol via UV spectroscopy. We hypothesize that the supernatant of the recombinant bacterial culture would show the least absorbance implying that it has the least amount of free honokiol in the media i.e. a significant amount is bound to the recombinant CB1 receptor on the E. coli membrane. Another method that could be used to quantify the amount of free honokiol present in the media is Thin Layer Chromatography (TLC)⁶.

Figure 3: Our recombinant part

After loading the CB1 expressing T-cells with honokiol and reintroduction into the patient’s bloodstream via intravenous infusion, we aim to release the drug specifically at the tumor site using an antagonistic molecule- AM6358⁷. We theorize to release AM6358 either using a biocompatible nanocarrier - based delivery system, which is highly specific to the tumor in question⁸, so it directly travels to the site of action. It could possibly be placed in a biodegradable suture-based microsphere that can be used for targeted delivery⁹. AM6358 will then act on the CB1-honokiol conjugate attached to the T-cell. This leads to honokiol’s release from the receptor, allowing it to act on the tumor mass, disaggregating it and paving way for the CAR T-cells to effectively target all the tumor cells and thereby, elicit an effective immune response. By specifically targeting drug release at the tumor site, we propose that any off-target/side effects caused will be greatly reduced. Due to infrastructural constraints, we were unable to model this in vitro.

Figure 4: A pictorial representation of our model