Differential Transcriptomic Modulation by Histone Deacetylase Inhibitor SAHA in LUAD and LUSC

Background: Histone deacetylases (HDACs) drive non-small-cell lung cancer (NSCLC) progression, yet HDAC inhibitor (HDACi) responses vary by tumor subtype. We compared the pan-HDAC inhibitor SAHA (vorinostat) in two NSCLC models-HER2-mutant lung adenocarcinoma (NCI-H1299; TP53del, NRASQ61K) and EGFR/PI3K-driven squamous carcinoma (NCI-H1703; PDGFRAamp, PIK3CAE542K)-to uncover lineage-specific mechanisms. Methods: Cells were treated with SAHA (10 µM, 24 h), a clinically relevant dose yielding ~1 µM free drug in plasma. Strand-specific RNA-seq was aligned to GRCh38.p13 and analyzed with edgeR and DESeq2 (FDR < 0.05, |Log2FC|= 1). Pathway, protein-interaction, apoptosis (Annexin V/PI), and migration (scratch, trans-well + mitomycin C) assays complemented transcriptomics. Clinical relevance was assessed by correlating SAHA-regulated genes with disease-free survival in TCGA-LUAD and TCGA-LUSC via GEPIA2. Results: SAHA altered 1098 genes in H1299 and 1532 in H1703, with only 437 shared. In H1299, SAHA induced EMT and MAPK-feedback genes while repressing interferon/apoptosis pathways. In H1703, SAHA activated complement/ECM remodeling and suppressed cell-cycle regulators. Class II HDACs (HDAC4/6) were downregulated only in H1703. Functionally, H1703 exhibited greater apoptosis; H1299 showed stronger migration inhibition. SAHA-reversed gene signatures (e.g., HMMR, PLK1 in LUAD; PAPPA, SAMD11 in LUSC) were significantly associated with poor disease-free survival, defining HDAC9-HMMR/PLK1 and HDAC4/6-PAPPA/SAMD11 axes. Conclusions: SAHA elicits distinct subtype-specific transcriptional and phenotypic effects in NSCLC. These findings highlight the importance of genetic context in HDACi therapy and support tailored combinations-such as SAHA with MAPK inhibitors in LUAD or HDAC4/6-targeted approaches in LUSC-and the use of SAHA-modulated biomarkers for patient stratification. Overall design: Total RNA was extracted from cells grown in 6-well plates (2 × 105 cells per well, ~70 % confluence at harvest) using TRIzol reagent (Invitrogen) and purified with the RNeasy Mini Kit (Qiagen), including an on-column DNase I digestion. RNA purity (A260/280 ratio = 2.05 ± 0.04, A260/230 ratio = 1.98 ± 0.03) was measured using NanoDrop 2000, and RNA integrity (RIN = 9.6) was confirmed by an Agilent 2100 Bioanalyzer. High-quality RNA (500 ng per sample) was used to generate strand-specific poly(A) libraries with the NEBNext Ultra II Directional RNA Library Prep Kit. Libraries were pooled and sequenced on an Illumina NovaSeq 6000 platform (paired-end, 2×151 bp, median depth of 55 million reads per sample, Q30 >93% , ~130× transcriptome breadth).