Ultrabright NIR-II Nanoprobes for Ex Vivo Deep Tissue Bioimaging: Protein Nanoengineering Meets Molecular Engineering

Introduction: : Near-infrared (NIR) fluorescence imaging has become a valuable tool for cancer diagnosis and treatment, offering high-resolution, real-time visualization in a non-invasive manner. Shifting the imaging window from the first near-infrared window (NIR-I, 750-950 nm) to the second near-infrared window (NIR-II, 1000-1700 nm) improves tissue penetration and reduces signal interference from light scattering and tissue autofluorescence. To improve upon traditional imaging methods, we combine molecular engineering and nanoengineering perspectives. From a molecular engineering standpoint, we demonstrate that quantum yield (QY) can be improved by modulating the amount and position of semiconducting polymers (SPs). From a nanoengineering perspective, we identify endogenous proteins that provide an effective matrix for encapsulating SPs into nanoparticles, not only amplifying the NIR-II fluorescence brightness but also providing active targeting capabilities. The encapsulation efficiency was tested for several functional proteins (albumins, hemoglobin, lactoglobulin, and transferrin) and the resulting protein-complexed nanoparticles were characterized by their physiochemical properties (e.g., size, charge, and colloidal stability under physiological conditions), as well as their photo- and biophysical properties (e.g., optical spectra, fluorescence intensity, quantum yield, molecular docking scores, and gel electrophoresis). We incorporated the optimal protein-complexed nanoparticles into tumor-mimicking phantoms for ex vivo studies in porcine lungs, stimulating cancer surgery. This integrated approach offers a novel method for developing ultrabright NIR-II imaging probes to enhance image-guided surgery and improve surgical outcomes.

Materials and

Methods: : A series of donor-acceptor-donor fluorophores were synthesized via Stille Coupling reaction by tuning thiophene and benzothiadiazole moieties (Fig. 2). These derivative candidates are referred to as T, TT, and 2T, depending on their thiophene unit orientation and number. Molecular docking was utilized to understand the interactions between functional proteins and T-derivative fluorophores. The optimized fluorophore-protein pairs were then screened for encapsulation efficiency using molecular docking scores. The synthesis of BSA@T derivatives and control, DSPE-PEG@T derivatives were synthesized using a modified nanoprecipitation technique, followed by ultracentrifugation to purify the nanoprobes from any unreacted dyes, proteins, or surfactants. Subsequently, physiochemical (hydrodynamic diameter, polydispersity index, and ζ-potential), and optical (fluorescence intensity, quantum yield, brightness, photostability) properties were analyzed for the optimized protein-encapsulated fluorophores. The in vitro biocompatibility and uptake were assessed in representative cancer cell lines (MCF-7a and 4T1 cells). The optimized candidate was tested in simulated fluorescence-based image-guided surgery (FIGS) using the IR VIVO imaging system, with tumor-mimicking phantoms embedded in ex vivo porcine lungs. Additionally, the optimized candidate was injected into ovine brain models to further evaluate the visualization abilities of the microvasculature, also employing the IR VIVO imaging system.

Results, Conclusions, and Discussions:: Based on our preliminary molecular docking screening, BSA was identified as the top candidate based on the docking scores. Using nanoprecipitation, BSA@T-derivatives and DSPE-PEG@T-derivatives were formed and showcased a uniform, mono-modal hydrodynamic diameter, and a polydispersity index (Fig. 3a). All the nanoprobes had spherical morphology as showcased by their transmission electron microscopy images (Fig. 3b). NIR-II images showcased that the BSA coating improved the brightness of BSA@T derivatives compared to DSPE-PEG@T derivatives and demonstrated a linear trend in increase of NIR-II intensity with increasing concentration (Fig. 3c). Using the optimized BSA@TT NP design, a brain ovine visualization study revealed that our experimental nanoprobe provided a more precise image of the microvasculature structure than the FDA-approved dye ICG (Fig. 3d, e). Across the three NIR-II imaging windows 925-1000 nm, >1000 nm, and >1250 nm, our BSA@TT NP exhibited signal-to-background ratios of 3.63, 3.113, and 3.09, respectively, outperforming ICG in all windows (Fig. 3f, g, h). To mimic an image-guided surgery setting, a tumor-mimicking phantom with BSA@TT solution was embedded into a porcine lung and imaged in the NIR-II window following excitation with an 808 nm laser. The fluorescence intensity of the simulated cancerous regions was compared to the normal lung tissue before and after resecting the tumor-covering tissue. The tumor to normal tissue ratio (T/NT) was 1.32 of the embedded tumor-mimicking phantom and 6.27 following the resection of phantom-covering tissue. A similar study will be conducted using tumor-mimicking phantoms embedded in porcine ovaries to mimic FIGS for ovarian cancer intervention. These comprehensive studies demonstrated that combining protein nanoengineering with molecular engineering techniques enhances the NIR-II fluorescence properties of nanoprobes, enabling precise deep tissue bioimaging in more clinically relevant scenarios, such as microvascular visualization or detecting buried tumor lesions (Fig. 1).

Acknowledgements and/or References (Optional):: Imaging studies on the IR VIVO system was funded by the Core Facility Support Award (grant number RP200572) from the Cancer Prevention and Research Institute of Texas (CPRIT) to the Imaging Core, Texas Tech University Health Sciences Center at Amarillo.