Identification of Reliable Housekeeping Genes for qRT-PCR Normalization in Neuroblastoma and Glioblastoma Cell Lines Infected with Lentivirus

Abstract

Background: Lentivirus infection significantly impacts gene expression in host cells, including the regulation of housekeeping genes, which are essential for normalization in quantitative reverse transcription PCR (RT-qPCR). This normalization reduces measurement errors arising from sample quality variations, RNA extraction methods, and experimental conditions. In the context of glioblastoma (GBM) and neuroblastoma (NB) — two malignancies characterized by poor prognoses and limited progression-free survival — it is crucial to identify stable reference genes to ensure accurate gene expression analysis. Objectives: Reliable normalization provides more trustworthy insights into the molecular mechanisms underlying tumor progression and may inform the development of targeted therapeutic strategies. Methods: The expression stability of eight housekeeping genes (RPL32, RPS23, GAPDH, 18S rRNA, TUB, ACTB, HPRT, and TBP) in U87 glioblastoma cells and seven genes (RPL32, HPRT, GAPDH, 18S rRNA, TUB, ACTB, and RPII) in SH-SY5Y NB cells was evaluated using RT-qPCR. Gene stability was analyzed using four statistical tools: GeNorm (pairwise variation-based ranking), NormFinder (model-based variance estimation), BestKeeper (standard deviation and CV analysis), and RefFinder (integrative ranking). Each experiment was performed in three biological replicates and analyzed in technical triplicate to ensure statistical robustness. Results: In SH-SY5Y cells, ACTB, RPL32, and RPII were consistently identified as the most stable reference genes across GeNorm, NormFinder, and BestKeeper analyses, while TUB was ranked as the least stable. In U87 cells, GeNorm ranked RPS23/HPRT as the most stable, NormFinder favored TUB/GAPDH, and BestKeeper prioritized ACTB/RPL32. Although slight differences in the gene rankings were observed among the different statistical tools, the overall selection of the most stable reference genes remained consistent. RefFinder’s integrative analysis resolved these discrepancies, identifying 18S (M = 0.18) and GAPDH as the most stable genes, and ACTB/HPRT as the least stable genes in U87. In SH-SY5Y, ACTB and RPL32 (M = 0.22) were the most stable genes, and TUB was the least stable (M = 2.45). Conclusions: This study provides a framework for reliable gene expression analysis in lentivirus-infected models. Our findings highlight the context-dependent stability of housekeeping genes, necessitating validation in diverse experimental settings (e.g., alternative viruses, primary cells) to ensure broader applicability. By emphasizing rigorous normalization, this work enhances reproducibility in gene expression studies and advances translational research in viral oncology and neuro-oncology, particularly for glioblastoma and NB therapeutics.