Objectives: Proteolysis targeting chimera (PROTAC), a new approach to cancer therapy, involving the degradation of proteins exhibits higher potency compared to small-molecule inhibitors.[1] Its ability to sustain target depletion offers potential solutions to overcome drug resistance.[2, 3,4,5] However, we have yet to be able to image its multi-step action, whole-body biodistribution and pharmacokinetics- key parameters for the clinical translation of targeted protein degraders (TPDs). Here, we report a FAP-targeting fluorescent PROTAC using a trivalent linker. While PROTACs were initially developed to degrade intracellular proteins, they have since been shown to retain potency in a wide range of integral membrane proteins. We hypothesize that membrane-associated proteins that are capable of rapid ligand-induced internalizations, such as FAP, may remain viable target for PROTAC.

Methods: FAP inhibitors (FAPi) have high specificity, affinity, and selectivity for tumor-associated fibroblast in the tumor microenvironment (TME) but suffers from rapid clearance and low tumor residualization [6, 7,8], hampering its use for clinical therapy. Thus, FAP may benefit from TPD therapy using PROTAC technology. Initially inspired by the reported crystal structures of VHL–MZ1–BRD4BD2 (Figure 1A), the present design, synthesis and mechanistic characterization of trivalent PROTACs were reported to enhance targeted protein degradations. [9,10] A classic PROTAC is a bivalent construct composed of three components: (1) protein-of-interest (POI) targeting ligand, (2) E3-targeting ligand and (3) bivalent linker (Figure 1B). Here, an imaging PROTAC is a trivalent chemical construct is based on FAPi-46 which includes a fourth component- a near-infrared dye (Figure 1C). We have carried out chemical (high-performance liquid-chromatography (HPLC), liquid-chromatography mass-spectrometry (LC-MS), 1H- and 13C- nuclear magnetic resonance (NMR)) and photophysical characterizations. To assess its feasibility for cancer imaging, we tested for in vitro, ex vivo and in vivo using xenografts of human metastatic prostate cancer (PCa) model. Separately, detailed dose and time-dependent fluorescent imaging experiments using two standard prostate cancer (PCa) cell lines (PC-3 and DU145) was submitted as a poster abstract for the WMIC (abstract ID #78). Tumor-bearing DU145 mice were split into three groups (n = 3) of vehicle (sterile PBS), FAPi-PROTAC-FL (50 µg/mice (2.5 mg kg-1)) and blocking (500 µg FAPi/mice) group that were intravenously injected. Mice were imaged at various timepoints (15-60 minutes) and were sacrificed. Selected tissues (tumor, liver, spleen, muscle, and brain) were excised out and imaged. To evaluate the potential of FAPi-PROTAC-FL for imaging and therapy, in vitro degradation and functional assays has been characterized.

Results: Starting from commercially available precursors, trivalent PROTAC was synthesized in 5 synthetic steps (21% yield). While we were able to suspend FAPi-PROTAC-FL in PBS containing 40% PEG300 and <5% DMSO, we have observed high fluorescence accumulation in tail vein. In vitro (Figure 1E), in vivo (Figure 1F) and ex vivo (Figure 1G) tumor accumulation on DU-145 cell line was observed which were blockable (10×). Interestingly, our in vivo and ex vivo data (Figure 1F,G) showed FAPi-PROTAC-FL can cross the blood-brain-barrier (BBB). Hence, we are poised to further investigate FAPi-PROTAC-FL in orthotopic DU145 model. Trivalent FAPi-PROTAC degrades FAP in LNCaP tested cell line (Figure 1H). In summary, we report preclinical data on FAPi-PROTAC-FL, FAP-targeting fluorescent trivalent PROTAC that demonstrates a dual imaging and chemical degradation capabilities.

Conclusion: Our first-in-class trivalent PROTAC outlines an approach for dual use of PROTAC as an vivo imaging probe and a TPD. This enables the real-time monitoring of crucial events along the PROTAC degradation pathway, whole-body and ex vivo imaging of its biodistribution. It may inform quantifiable and relative amounts of cancer proteins that are about to undergo chemical degradation. Thus, this platform is also modifiable through site-specific conjugation which will allow access to other disease-promoting models and/or clinically relevant diagnostic probe.

Image/Figure:
Click to View

Image/Figure Caption:

Figure 1: (A) The exact crystal structure of BRD4^BD2-MZ1-VHL ternary complex (PDB code: 5T35)^[7]. The structure was examined by downloading it from PDB using the code 5T35 and visualizing structure using Chimera software, image shown is color adapted, otherwise it is unmodified from Gadd et al.^[7] (B) bivalent vs. trivalent PROTAC contructs. (C) Building block of FAPi-PROTAC-FL. (D). Photochemical characterization (top) and FAP-expression determination using western blot (WB) on PC-3, LNCaP and DU145. (E) In vitro cell fluorescence imaging of FAPi-PROTAC-FL on DU145. (F) Representative in vivo near infrared images of intravenously injected FAPi-PROTAC-FL (50 µg/mice) which were blockable using FAPi (10×). (G). Representative fluorescence images of ex vivo imaging showing tumor, muscle and brain of a vehicle or FAPi-PROTAC-FL injected mouse through tail vein injection.

Author