Targeting mitochondrial Serine Hydroxymethyltransferase (SHMT2) through RNA-based inhibitors: from riboregulation to therapeutic perspectives

Cancer development and progression is tightly linked to metabolic reprogramming, which allows malignant cells to sustain rapid proliferation and survive under stress. Among the enzymes that orchestrate this rewiring, mitochondrial serine hydroxymethyltransferase (SHMT2) plays a pivotal role by catalyzing the conversion of serine to glycine and generating one-carbon units for nucleotide biosynthesis, mitochondrial translation, and redox homeostasis. SHMT2 is frequently overexpressed in many malignancies and its upregulation correlates with poor prognosis. Previous studies from our laboratory revealed that SHMT2, like its cytosolic isoform SHMT1, can directly bind RNA and this binding inhibits the protein catalytic activity. This post-translational regulation is mediated by its own 5'-untranslated region (UTR2). This discovery opened a new therapeutic perspective based on exploiting RNA- protein interactions to selectively inhibit SHMT2. In this thesis, we sought to refine this strategy by designing and characterizing shorter inhibitory RNA molecules mostly, but not only, derived from the UTR2, aiming to identify minimal RNA fragments capable of maintaining both binding affinity and enzymatic inhibition of SHMT2, while being more suitable for pharmacological applications. A series of iRNA candidates were screened through in vitro binding (EMSA) and inhibition assays. These analyses revealed that RNA length and secondary structure jointly influence the interaction with SHMT2, with the most effective inhibition observed for longer sequences and those mitochondrially-targeted. Functional assays in lung adenocarcinoma cells (A549 and H1299) confirmed that several mitochondrially-targeted RNAs significantly reduced cancer cell viability, while their cytosolic counterparts were less to none effective. Among the shorter sequences tested, the mitochondrially targeted construct m4 emerged as the most potent inhibitor, consistently decreasing cell viability across multiple cancer models, including triple-negative breast cancer (MDA-MB-231) and glioblastoma (T98G). Moreover, SHMT2 inhibition was found to reduce cell motility under nutrient-deprived conditions, reinforcing the link between mitochondrial metabolism, energy balance, and cell migration. Finally, preliminary delivery studies explored two complementary nanocarrier systems -engineered humanized ferritin nanocages and PLGA-based nanoparticles- to assess their ability to transport the inhibitory RNA molecules into cells. Both platforms successfully mediated RNA uptake, and microscopy imaging revealed partial mitochondrial localization of the m4 RNA, supporting its potential for targeted intracellular delivery. Collectively, this work expands the understanding of SHMT2 riboregulation and demonstrates the feasibility of targeting this enzyme through rationally designed RNA molecules. The identification of m4 as a compact and potent mitochondrial inhibitor lays the foundation for future optimization of RNA-based therapeutics and delivery strategies aimed at selectively impairing SHMT2-driven malignancies.