I3 - Gene Expression Profiling of 3D Lung Tumor Model Compared to 2D Culture and In Vivo Models

Introduction: : Lung cancer research traditionally relies on 2D cell culture and xenograft models to investigate tumor biology and evaluate therapeutic strategies. However, these models face limitations: 2D cultures fail to recapitulate the complex structure of tumors, while xenografts, though more biologically relevant, are labor-intensive and costly. To address these challenges, we developed a tunable 3D poly(ethylene glycol) (PEG)-based tumor model that mimics key biochemical and mechanical features of the lung tumor microenvironment (Fig. 1A-C). In this study, we compared the transcriptomic profile of lung cancer cells encapsulated in the hydrogel system to those grown in 2D culture or implanted subcutaneously in vivo.

Materials and

Methods: : Human A549 luciferase-expressing lung cancer cells were cultured in three systems of increasing complexity: (1) on 2D tissue culture plastic, (2) encapsulated in a 3D hydrogel composed of PEG-diacrylate, PEG functionalized with matrix metalloproteinase-degradable sites, and PEG functionalized with integrin binding sites, or (3) injected subcutaneously into NOD-SCID mice. In vitro cultures were maintained for 1 week, and in vivo tumors were grown for 3 weeks. Library preparation was done using Illumina RNA-Seq Libraries PolyA directional/stranded and samples were subjected to paired-end sequencing (50 million reads/sample) using the Illumina Novaseq at the University of Florida Health Cancer Center.

Results, Conclusions, and Discussions:: The PEG-based 3D tumor model demonstrated high global similarity to subcutaneous (SubQ) in vivo tumors, as evidenced by a principal component analysis (Fig. 1D), hierarchical clustering of the 1000 differentially expressed genes (DEGs) (Fig. 1E), and a higher number of non-differentially expressed genes (Fig. 1F). A Spearman’s correlation coefficient of 0.96 further supported the close resemblance between the 3D model to the subcutaneous tumor. To investigate biological pathways more accurately recapitulated in the 3D model, we identified genes that were not differentially expressed (p > 0.05) between the 3D and SubQ model but were differentially expressed in the 2D and SubQ models. This analysis revealed 2,829 genes uniquely preserved in the 3D model. KEGG pathway enrichment analysis of these genes identified Huntington’s disease as the only significantly enriched pathway. Closer examination showed that the 3D tumor model more faithfully reproduced the expression of genes associated with Huntington’s disease, which are also implicated in solid tumor progression. In contrast, the 2D model showed dysregulated expression of genes linked to cell mitochondrial activity, proliferation, and metastasis, whereas the 3D tumor model maintained expression levels comparable to the SubQ tumors. These findings suggest that the 3D PEG-based hydrogels offer a more biologically relevant platform than traditional 2D cell culture. Moreover, after just 1 week in culture, the 3D tumor model replicated aspects of subcutaneous tumors grown for 3 weeks, offering a time-saving and cost-effective alternative to animal studies. Future work will focus on further refining our method of analyzing non-DEGs to identify pathways that remain unchanged across models to guide the selection of the most appropriate system for specific biological questions.

Acknowledgements and/or References (Optional)::