Fig 2: Deleterious effects of RBBP9 knockdown in FA-HNSCC.a Ribbon representation of RBBP9 structure adapted from SWISS-MODEL (#O75884, O75884 | SWISS-MODEL Repository (expasy.org)), highlighting the reactive Serine residue (S75, in purple) of the enzyme’s catalytic triad, putative RB-interacting LxCxE motif (Residues 63, 65, 67 in aqua) and the mononucleotide-binding Rossmann fold in navy blue. b Representative fluorescent photomicrographs of fixed, propidium iodide stained cells in individual wells of 96-well plates (4x objective, IncuCyte®) after 96 h of RBBP9 knockdown, (as in validation experiments) demonstrating its selectively lethal effects in FA-HNSCC tumour cells. Ø = 6.58 mm, well diameter. c Confluence-based growth kinetics of FA-HNSCC tumour cell lines with RBBP9 knockdown for 96 h (top row). Note the higher proliferation rate of FA-reconstituted (hence ICL-repair proficient) VU-SCC 1131 + FANCC cell line, indicating conditional synthetic lethality of RBBP9 with FA-pathway loss. Similar growth curves of respective FA patient-matched hTERT-immortalised CAFs/Fs and the ARPE-19 hTERT WT, FA-A and FA-C model cell lines are shown in the lower row for comparison. Despite similar seeding densities, note the lag in growth curves of both FANCC-/- and FANCA-/- ARPE-19 hTERT cell lines, typical of cells harbouring FA-defects. n = 3 independent experiments in duplicates. Error bars = SD. NT non-targeting, UBB ubiquitin B, PLK1 Polo-like kinase 1. d, e Representative photomicrographs (at 72 h) of live FA-HNSCC cells IncuCyte®-imaged in 96-well plates for upto 100 h after RBBP9 knockdown, fed with the Caspase-3 fluorogenic substrate NucView 488®. Green fluorescent cells represent apoptosis. Caspase-3 activity maxima plotted as green object counts over time (in e, upto 100 h), reflect cell death as a function of time, indicating differences between the FA patient-derived HNSCC cell lines. n = 3 independent experiments in duplicates. Error bars = SD. Scale bar ˜150 µm. See also Supplementary Fig. 4 for full-frame images with scale bars. f, g Propidium Iodide exclusion flow cytometry plots of the FA-HNSCC and FA-normal ARPE-19 hTERT lines (quantification in g) to measure terminal cell death. Bars are a percentage of PI-positive (Dead) cells. RBBP9 knockdown (72 h in these experiments) does not affect untransformed ARPE-19 hTERT cells with or without FA-defects. n = 3 independent experiments in triplicates. Error bars = SD. Non-parametric two-way ANOVA with Holm–Sidak correction for multiple comparison was performed for statistical analysis. **p = 0.01.

Fig 3: Pharmacological targeting of RBBP9 enzymatic function in FA-HNSCC.a Chemical structure of the known irreversible RBBP9 inhibitor Emetine, an oxime ester, drawn using adobe illustrator. b Representative photomicrographs (right) of propidium iodide stained cells in individual wells of 96-well plates (4x objective, IncuCyte®) after 96 h of Emetine (15 and 25 nM) treatment, demonstrating its selectively lethal effects in FA-HNSCC tumour cells. ARPE-19 hTERT WT, FANCC-/- and FANCA-/- cells were included to denote human, diploid, untransformed cells with or without FA-proficiency, respectively, while ARPE-19 hTERT RBBP9-/- cells is a control to monitor any gross off-target lethality due to Emetine. Ø = 6.58 mm, well diameter. Generation and characterisation of ARPE-19 hTERT RBBP9-/- cells is presented in Supplementary Fig. 5. c Cell viability (CellTiter-Blue®) assay of the cell line panel as in a, after 96 h of 25 nM Emetine treatment. Note the general lethality in HNSCC cells, while all untransformed cells, patient-matched or otherwise, irrespective of their genotype, remain unaffected. Fluorescence values were normalised to untreated controls of the respective cell line. n = 3 independent experiments in duplicates. Error bars = SD. Non-parametric two-way ANOVA with Holm–Sidak correction for multiple comparison was performed for statistical analysis. **p = 0.01, ****p = 0.0001. d Representative photomicrographs (left) and the mean spheroid volumes (right) of control or Emetine-treated (500 nM, 48 h) treated 3D spheroid cultures of FA-HNSCC tumour cell lines, as well as human diploid retinal pigment epithelial ARPE-19 hTERT cell lines with a targeted disruption of the FANCA and FANCC genes, representing FA-defective untransformed cells. RBBP9-/- ARPE-19 hTERT cell line served as a control for Emetine’s activity. n = 2 independent experiments in triplicates. Error bars = SD. Scale bar = 200 um. Non-parametric two-way ANOVA with Holm–Sidak correction for multiple comparison was performed for statistical analysis. ***p = 0.001, ****p = 0.0001. e Immunoblot analysis (top) and densitometric quantification (bottom) of the isogenic VU- SCC 1131 (FA-deficient) and VU-SCC 1131 + FANCC (FA-proficient) cell line pair treated with 25 nM Emetine for 72 h. Levels of the DNA damage marker pS139 H2AX, and the activity of replication checkpoint kinases pS428 ATR and pS345 Chk1, collectively suggest that low-dose Emetine treatment induces cell death independent of DNA damage. n = 3 independent experiments. GAPDH was used as the loading control. Error bars = SD. **p = 0.01, ***p = 0.001. f Proposed biochemical network of RBBP9 based on protein–protein interaction(s) identified by co-immunoprecipitation (coIP) and MS analysis RBBP9-FLAG overexpressed in HEK293T cells. High-confidence interactors were shortlisted based on statistics (p value =0.05) and further enriched based on higher peptide count representation in RBBP9-FLAG co-IPs compared to FANCC-FLAG co-IPs. Dotted orange lines denote known physical interactions, and grey dotted lines co-expression. Functional annotation was performed using the online GO analysis tool Toppgene. Benjamini FDR =0.05 was deemed significant.