Abstract:
Multivariate metal–organic frameworks (MOFs) with iso-reticular architectures and spatially programmable functional groups represent an attractive platform for the development of sequence-dependent materials. By enabling precise control over the type, ratio, and spatial arrangement of functional groups, multivariate MOFs offer opportunities to encode complex structure–function relationships that resemble those found in biological macromolecules. However, the construction of extensive multivariate MOF libraries that allow function-targeted material selection, analogous to directed evolution in proteins, remains a significant challenge.
Here, we report a biomimetic strategy to design multivariate MOFs (bMTV-MOFs) by incorporating amino acid–derived functional linkers into the reticular framework. This approach introduces protein-like binding cavities, chemical diversity, and tunability into otherwise rigid MOF structures. By enumerating all possible combinations of nine functionalized organic linkers, a comprehensive library of 1089 MOFs was generated, encompassing both single-component and multivariate entries. These bMTV-MOFs were synthesized in a combinatorial manner and systematically evaluated for drug delivery performance.
Verteporfin (VP), a clinically approved photosensitizer for photodynamic therapy (PDT), was selected as a model therapeutic cargo to assess structure–function relationships within the MOF library. Among the constructed materials, tertiary bMTV-MOFs exhibited superior VP loading capacity and controlled release behavior compared with pristine MOFs. The enhanced performance is attributed to the cooperative effects of multivariate functional groups in forming biomimetic binding cavities that promote favorable host–guest interactions.
In vitro studies demonstrated that VP-loaded tertiary bMTV-MOFs induced strong photodynamic effects in multiple cancer cell lines, including B16F10, 4T1, and K7M2. Notably, the privileged A1F6H3 and A3D1H6 bMTV-MOFs achieved a 23–68-fold enhancement in PDT efficacy relative to VP delivered by pristine MOFs, primarily through apoptosis-mediated cell death.
Overall, this work demonstrates that biomimetic multivariate MOF libraries enable function-targeted material selection and provide a versatile platform for precision drug delivery and photodynamic therapeutic applications.
