To identify promising nanoporous MOF structures for siRNA delivery, a systematic study of the structure–adsorption performance relationship is essential. Yet to date, the discovery of promising novel porous materials for specific adsorption applications is happening by trial and error rather than by rational design. Tens of thousands of porous MOF structures have been reported in the literature and could potentially act as DDSs. However, most of them have only been tested for gas phase adsorption and studies of porous MOFs for drug delivery are extremely scarce, and generally do not include systematic drug adsorption, biocompatibility, targeting, and in vitro/in vivo studies.
Molecular simulation have been used widely for studying adsorption processes in porous materials, allowing the explanation and prediction of new experimental results. From a pool of ca. 1M entries of crystal structures at the CCDC, we looked for “look-for-MOF” criteria. From about 90,000 existing MOFs, only 9,000 were porous. In addition, we were able to filter only biocompatible metals such as Fe, Zr, Zn and Mg, reducing the number of MOFs to about 400 structures. The idea is to get a landscape of structure-property relationships to filter the best structures for RNA delivery.
Following MOF identification, we have achieved the synthesis of a series of biocompatible MOFs. We have also developed the synthesis of nano-sized Zr-MOFs with a wider pore size (> 2 nm), such as NU-1000, NU-901, PCN-222, PCN-224, PCN-333 and PCN-777, capable of encapsulating macromolecules such as siRNA. The nanoparticulate form of these MOFs has been optimised using solvothermal methods and the use of synthesis modulators, using monodentate capping ligands during the self-assembly of the MOFs. All the materials have been prepared with particle sizes smaller than 200 nm as required to ensure free circulation within the smallest capillaries.
The degradation of the MOFs has been studied under physiological conditions in phosphate buffered saline (PBS) as well as other different buffers (His, Tris, etc.) at 37˚C and for different incubation times. We have measured release times and have de-risked the experimental protocols to work with MOFs in drug delivery applications. We have also achieved the successful loading and protection from enzymatic degradation of a siRNA fragment – with a sequence of 22 nucleotides in length for HEK-293 cell lines expressing mCherry – on nano-sized NU-1000, allowing the delivery of siRNA effectively in the cytoplasm to knockdown gene expression.
We have worked on magnetic induction technology using Fe-based, magnetic nanoparticles (NPs). We have achieved the synthesis of core@shell Fe@Fe3O4 NPs with core diameters of ca. 8 nm. On the other hand, we have demonstrated the tuneable and versatile room-temperature incorporation of gold nanorods (AUNRs) into the NU-901 MOF by using the AuNRs as seeds for the MOF growth. We were able to achieve an excellent core-shell yield (>99%), with the first preliminary evidence for an NU-1000 phase grown at room temperature.
Performance of selected materials for drug delivery in cancer therapy has been evaluated by using in vitro studies. We evaluated the toxicity of the materials developed to better understand how MOFs cross the cell membranes. We have used an array of in vitro techniques, such as MTS and Annexin V cytotoxicity tests to study the biocompatibility of the Zr-MOFs named above and half-maximal inhibitory concentration (IC50) of siRNA/drug/calcein-loaded MOFs and free siRNA/drug/calcein on HeLa cells, J774.A1 and HEK-293. The basic idea is to use the MOF DDS to entrap and internalise the siRNA into the cellular endosomes, where the MOF will be subsequently degraded, causing the release of the siRNA into the cytosol. Once in the cytosol, the siRNA will bind the RNA-induced silencing complex (RISC).18 The activated siRNA-RISC complex will recognise the complementary mRNA, and the catalytic components of the complex, the endonucleases called Argonaute 2 (AGO2) proteins, will cleave the target mRNA sense strand. This will leave the antisense strand bound to the newly activated RISC-siRNA unit, which then will travel around the cells to find, bind to, and cleave any other complementary mRNA strands.
