Deep Learning-Based TEM Image Analysis for Fully Automated Detection of Gold Nanoparticles Internalized Within Tumor Cell.

Transmission electron microscopy (TEM) imaging can be used for detection/localization of gold nanoparticles (GNPs) within tumor cells. However, quantitative analysis of GNP-containing cellular TEM images typically relies on conventional/thresholding-based methods, which are manual, time-consuming, and prone to human errors. In this study, therefore, deep learning (DL)-based methods were developed for fully automated detection of GNPs from cellular TEM images. Several models of "you only look once (YOLO)" v5 were implemented, with a few adjustments to enhance the model's performance by applying the transfer learning approach, adjusting the size of the input image, and choosing the best optimization algorithm. Seventy-eight original (12,040 augmented) TEM images of GNP-laden tumor cells were used for model implementation and validation. A maximum F1 score (harmonic mean of the precision and recall) of 0.982 was achieved by the best-trained models, while mean average precision was 0.989 and 0.843 at 0.50 and 0.50-0.95 intersection over union threshold, respectively. These results suggested the developed DL-based approach was capable of precisely estimating the number/position of internalized GNPs from cellular TEM images. A novel DL-based TEM image analysis tool from this study will benefit research/development efforts on GNP-based cancer therapeutics, for example, by enabling the modeling of GNP-laden tumor cells using nanometer-resolution TEM images.

Uncloaking cell-impermeant gold nanorods via tumor microenvironmental cathepsin B facilitates cancer cell penetration and potent radiosensitization.

Major impediments to conveyance of intravenously administered drugs to tumors are biofouling, opsonization, and rapid clearance from the circulation by macrophages and reticuloendothelial phagocytes. Cloaking nanoparticles with stealth epilayers partly overcomes these hurdles but it also foils interactions with tumor cells. Here, we describe the synthesis, characterization, and validation of smart gold nanorods (GNRs) that spontaneously transform from inert passengers in the blood stream to active cell-penetrating nanoparticles within tumors to potently sensitize tumors to radiation therapy. Intrinsically cationic and cell-penetrating GNRs were shielded from phagocytosis with a cloaking polyethylene glycol epilayer containing an intervening cleavable peptide. In the absence of an external trigger, this epilayer is clipped off by the tumor microenvironmental protease, cathepsin B, in colorectal cancers to uncloak and expose the free-circulating native unPEGylated GNR that is readily internalized by cancer cells and turn into immovable small clusters of GNRs. Selective uncloaking of GNRs in the tumor reduced off-target toxicity confirmed by hematologic, biochemical, and histopathological analysis of blood, serum, and normal organs, respectively. Subsequent irradiation led to significant tumor growth delay and improved survival of mice. By addressing multiple barriers to efficient transport and cellular internalization of nanoparticles, our results demonstrate that clinically meaningful radiosensitization can be achieved with rationally designed GNRs.

Synthesis and characterization of gadolinium-decorated [60]fullerene for tumor imaging and radiation sensitization.

PURPOSE: The excellent contrast of high atomic number (Z) elements compared to soft tissues has advanced their use as contrast agents for computed tomographic imaging and as potential radiation sensitizers. We evaluated whether gadolinium (Gd) could serve as such a theranostic agent for high-resolution magnetic resonance imaging (MRI) due to its paramagnetic properties and radiosensitization due to its high Z. MATERIALS AND METHODS: To improve the relaxivity of Gd, we coupled it to [60]fullerene, an elemental carbon allotropic nanoparticle that seamlessly traverses physiological barriers . By adding serinol, an aliphatic alcohol derived from amino acid serine, we turned [60]fullerene, which is otherwise insoluble in water, into a highly water-soluble derivative and decorated it externally with a payload of chelated gadolinium ions. RESULTS: When [60]fullerene was functionalized in this manner with two gadolinium ions (Gd2C60), it displayed considerably higher T1 relaxivity at 4.7 T than the commercially used MRI contrast agent, Magnevist, (18.2 mM-1s-1 vs. 4.7 mM-1s-1). Attempts to increase this even further via decoration of [60]fullerene with 12 gadolinium ions was unsuccessful due to a poor water solubility. However, the current formulation of Gd2C60 did not result in any appreciable radiosensitization. CONCLUSION: Our results show a successful generation of a novel contrast agent via decoration of fullerene with two chelated Gd ions. Though this formulation was not successful in generating radiosensitization, other chemical modifications can be further explored to increase radiosensitization potential.