Search for: All records

Natural killer (NK) cell functionality is a strong indicator of favorable prognosis in cancer patients, making NK cells an appealing therapeutic target to prevent lymph node dissemination. We engineered liposomes that are conjugated with anti‐CD335 antibodies for NK cell targeting, and the apoptotic ligand TRAIL to kill cancer cells. Liposomes were made using a thin film hydration method followed by extrusion to approximately 100 nm in diameter and conjugation of proteins via thiol‐maleimide click chemistry. TRAIL/anti‐CD335 liposomes successfully bound to isolated NK cells. Once piggybacked to the surface of NK cells, these “Super Natural Killer Cells” were able to more effectively kill oxaliplatin‐resistant SW620 cells and metastatic COLO205 colorectal cancer cells via TRAIL‐mediated apoptosis compared to NK cells alone. Importantly, Super NK cells were more effective under physiological levels of fluid shear stress found in the lymphatics. Liposome biodistribution after intravenous administration confirmed the sustained presence of liposomes within the spleen and tumor draining mesenteric lymph nodes for at least 4 days. These results demonstrate the enhanced apoptotic effects of NK cells armored with liposomal TRAIL against clinically relevant colorectal cancer cells, providing the groundwork for in vivo treatment studies in mouse models of colorectal cancer metastasis.

 

Humans have the remarkable ability to recognize and acquire novel visual concepts in a zero-shot manner. Given a high-level, symbolic description of a novel concept in terms of previously learned visual concepts and their relations, humans can recognize novel concepts without seeing any examples. Moreover, they can acquire new concepts by parsing and communicating symbolic structures using learned visual concepts and relations. Endowing these capabilities in machines is pivotal in improving their generalization capability at inference time. We introduced Zero-shot Concept Recognition and Acquisition (ZeroC), a neuro-symbolic architecture that can recognize and acquire novel concepts in a zero-shot way. ZeroC represents concepts as graphs of constituent concept models (as nodes) and their relations (as edges). To allow inference time composition, we employed energy-based models (EBMs) to model concepts and relations. We designed ZeroC architecture so that it allows a one-to-one mapping between a symbolic graph structure of a concept and its corresponding EBM, which for the first time, allows acquiring new concepts, communicating its graph structure, and applying it to classification and detection tasks (even across domains) at inference time. We introduced algorithms for learning and inference with ZeroC. We evaluated ZeroC on a challenging grid-world dataset which is designed to probe zero-shot concept recognition and acquisition, and demonstrated its capability.