Figure 1 Schematic representation of CRISPR Self Check-In operation for controlled genome editing. In disease cells, CRISPR Self Check-In controls nucleus translocation by itself for precise genome editing by sensing disease-specific miRNA and being cleaved off mRNA bridge mimetics containing NES.

In this project, we will develop the “mRNA bridge mimetics” platform, which comprises of a synthetic complementary strand of miRNA that can be degraded by endogenous miRNA-Ago complexes in the cytoplasm to confer the ability of self-control to the CRISPR Cas9 protein. The precise programmability of mRNA bridge mimetics enables Cas9 to accomplish Self Check-In to the nucleus by sensing the disease-specific miRNA and cleaving off the nucleus export signal (NES) on it (Fig. 1). The CRISPR Self Check-In was retained and degraded in the cytoplasm within 24 hours if there were no corresponding miRNAs for cleaving off the programmed mRNA bridge mimetics in the cells. Thus, this strategy is especially promising for applications in which disease-independent off-target editing must be minimized. 

Figure 3 Schematic representation of Synthetic Epigenetic Enzyme for locus-specific transcription regulation

In this project, we will develop a new synthetic epigenetic enzyme capable of targeting a specific locus and methylating a histone for up- or down-regulating gene expression. We hypothesized that the synthetic epigenetic enzyme protein would activate opsin photoreceptor genes, thereby compensating rhodopsin function and recovering retinal degradation. Currently, we successfully designed and constructed the novel synthetic epigenetic enzyme, and the production was optimized. The function of the synthetic enzymeas as a transcript activator was also confirmed so far in vitro. Next, we will confirm the opsin upregulation and therapeutic efficacy in vivo. Together, the novel synthetic enzyme developed in this study could provide an innovative platform for safe and controlled genome regulation in diseases.

Figure 4 Schematic representation of 3D Immunocompetent cancer model for HT drug screening

In this project, we will develop in vitro the tumor microenvironment containing M2 macrophages using the 3D architecture of cancer cells. First, we will find the different transcript and protein regulation between 2D and 3D cancer cells. Next, we will confirm the key molecules contributing M2 polarization in TME. Then, we will genetically modify either cancer or stromal cells to induce TAM in the 3D cancer model. The system will be developed as an HT drug screening platform by integrating AI technology. Altogether, the simple strategy to form in vitro 3D immunocompetent TME developed in this study will broaden the understanding of communication between cancer and macrophages and contribute to the precise high-throughput evaluation of cancer-targeting drugs.

Figure 5 Schematic representation of AI-based drug discovery

Artificial intelligence (AI) capable of integrating a large amount of related information to predict the therapeutic relationships such as disease treatment of known drugs, gene expression, and drug-target binding information, has great potential for use in next-generation drug development. Even though many promising therapeutic targets were found, only a limited number of drugs were developed due to the lack of effective drug design and screening strategy. In this project, we are developing AI-driven new drug candidates.