Oral Presentation Australasian RNA Biology and Biotechnology Association 2026 Conference

Modulating eIF5A-dependent mRNA translation to target cancer stem cells (145308)

Christina Bui 1 , Cerys McCool 1 2 3 , Vanina Rodriguez 1 3 , Shabarni Gupta 1 2 3 , Erin E Heyer 1 2 3 , Heloisa Helena H Milioli 1 2 3 , Christine L Chaffer 1 2 3
  1. The Garvan Institute of Medical Research, Darlinghurst, NSW, Australia
  2. St Vincent's Clinical School, UNSW Medicine, UNSW Sydney, Darlinghurst, NSW, Australia
  3. The Kinghorn Cancer Centre, Darlinghurst, NSW, Australia

Triple-negative breast cancer (TNBC) is a heterogeneous disease characterised by high risk of relapse, poor overall survival and a lack of targeted therapies. Within these tumours, cellular plasticity is a major driver of cancer heterogeneity, enabling cells to dynamically and reversibly switch between a non-cancer stem cell (non-CSC) state and a therapy-resistant, metastatic cancer stem cell (CSC) state. Critically CSCs serve as a reservoir for metastasis and disease recurrence. Despite their central role in treatment failure, the non-genetic programs that sustain the CSC state remain poorly understood.

Using integrative RNA-seq and quantitative proteomics on purified CSC (CD44high) and non-CSC (CD44low) TNBC populations we uncovered a distinct mRNA translation program sustaining the CSC state, alongside elevated expression of Eukaryotic Initiation Factor 5A (eIF5A). High eIF5A levels correlated with poor survival outcomes in TNBC clinical cohorts (TCGA, METABRIC). Functionally, eIF5A is uniquely activated by hypusination, an essential post-translational modification sequentially catalysed by deoxyhypusine synthase (DHPS) and deoxyhypusine hydroxylase (DOHH). Hypusinated eIF5A (eIF5Ahyp) regulates multiple layers in mRNA translation, including start-codon selection, and translation elongation (through ribosome pausing and the resolution of stalled ribosomes), and is recruited to stress-associated cytoplasmic condensates. We hypothesise that eIF5A hypusination drives a specialised translational program required to sustain the CSC phenotype, making it a critical therapeutic vulnerability in TNBC.

Pharmacological inhibition of DHPS using N1-guanyl-1,7-diaminoheptane (GC7) elicited state-specific responses. In CSCs, GC7 led to marked enrichment of ribosome biogenesis and assembly, rRNA processing and maturation and mRNA processing and splicing (NES >2.3), consistent with compensatory translational response, while suppressing cell cycle progression. In contrast, non-CSCs activated senescence signals via the induction of senescence-associated secretory phenotype. Functionally, GC7 (10 μM) reduced CSC proliferation and migration, impaired tumoursphere formation, and sensitised cells to chemotherapeutic agents doxorubicin and docetaxel.

Together, these findings identify eIF5Ahyp as a translational vulnerability in CSCs and provide a rationale for combining eIF5Ahyp inhibitors with chemotherapy to disrupt CSC-driven plasticity, chemoresistance and metastatic progression.