Professor Lv Yongqin's team at Beijing University of Chemical Technology has developed a platform for chemical antibody molecular recognition elements targeting small molecules, proteins, and cells. They have also developed a manufacturing system combining chemical synthesis and AI-powered intelligent screening. Their research, titled Epitope-Imprinted Polymer Nanoparticles Enable Selective Recognition of Filamentous Vimentin for Tumor Imaging In Vivo, were published in Advanced Healthcare Materials on October 6, 2026. This platform, through AI-assisted design, high-throughput screening, and controllable chemical synthesis, overcomes the bottlenecks of traditional biological antibodies in balancing stability, cost, and yield. It has the potential for production at the hundred-kilogram scale, with some products costing approximately 1% of the cost of biological antibodies. Furthermore, it is room-temperature stable, has wide pH tolerance, and is not cold-chain dependent. The KD of specific targets is as low as the pM level, and its key performance is comparable to imported high-end biological antibodies. The platform can support therapeutics, in vitro diagnostics, separation and purification, and research product development, covering fields such as oncology, cardiovascular, and neurological diseases.
Vimentin is a cytoskeletal protein upregulated during epithelial-mesenchymal transition that can also be aberrantly displayed on the surface of multiple tumor cell types, making it a promising target for tumor imaging. Here, we develop epitope-imprinted polymer nanoparticles (nanoMIPs) for vimentin recognition through a two-step strategy comprising screening a library of non-imprinted nanoparticles to identify a lead formulation, followed by peptide-directed molecular imprinting. The resulting nanoMIPs exhibited enhanced apparent affinity for vimentin, with picomolar-range KD values, although these values represent model-dependent ensemble apparent affinities that may include contributions from multiple nanoparticle-ligand contacts and local rebinding rather than intrinsic monovalent affinities. The nanoMIPs preferentially bound cancer cells presenting cell-surface vimentin in vitro. The enhanced target binding and reduced off-target binding were consistent with an imprinting-associated improvement in molecular discrimination, although nonspecific physicochemical interactions were not completely eliminated. These findings extend epitope-guided nanoMIP design to structurally dynamic, cell surface-associated cytoskeletal targets, and support the development of synthetic receptors for tumor imaging. In vivo, the optimized nanoMIPs produced a tumor-associated fluorescence signal approximately 1.44 times that of the matched non-imprinted nanoparticles.
Paper link: https://doi.org/10.1002/adhm.71813

