O23 - The Interplay Between Interstitial Fluid Flow and the Glycocalyx in Glioma Stem Cell Invasion

Introduction: : Glioblastoma (GBM) is a malignant astrocytoma with a low survival time of around 15 months from diagnosis with a near 100% recurrence rate and is characterized by extensive local metastasis in the brain. Glioma stem cells (GSCs) within the tumor cell population have previously demonstrated differing invasion responses from exposure to biophysical cues, especially interstitial fluid flow (IFF). IFF also causes GSC differentiation and therefore intratumoral heterogeneity, which contributes to treatment resistance. The glycocalyx is a sugar-rich outer coating on eukaryotic cells that mediates cellular interactions with other cells and their environment. It is known to be dysregulated in cancer and especially heterogeneously expressed in breast cancer, as well as involved in cancer stem cell invasive behavior. However, glycocalyx heterogeneity in GSCs and the role of specific glycocalyx components in IFF-mediated GSC invasion remain unexplored. Here, we evaluate glycocalyx biopolymer expression in GBM and quantify GSC invasive responses to IFF by altering the expression of glycocalyx components.

Materials and

Methods: : GBM patient survival analyses on hyaluronic acid synthase (HAS) and mucin (MUC) genes were performed using the TCGA GBM Database in Xena (University of California, Santa Cruz). Furthermore, RNASeq data from G34 patient-derived parental and subcolony cell lines was analyzed to evaluate pairwise differences in expression of HAS and MUC expression. Previously isolated patient-derived G34 GSC subcolonies (1-3, 3-3, and 4-3), which were confirmed to respond to IFF differently, were nuclear labeled to express fluorescent proteins Aquamarine (1-3), iRFP-670 (3-3), and mCitrine (4-3) and then encapsulated in PhotoHA-Collagen hydrogels that recapitulated the brain microenvironment and seeded in a tissue culture insert under static or flow conditions. Gels were then imaged as Z-stacks using confocal microscopy and G34 invasion was quantified across the top, middle, and bottom slice of the gel as well as the tissue culture membrane. Baseline glycocalyx biopolymer expression was also analyzed in the G34 parental cell line via flow cytometry. Pharmacological inhibition of specific glycocalyx biopolymers was also conducted and treated cells were incorporated into invasion assays. Conversely, parental G34s were subjected to static or flow conditions in hydrogel-based invasion assays and separated into distinct populations depending on their invasiveness, after which flow cytometry was used to determine glycocalyx biopolymer expression.

Results, Conclusions, and Discussions:: TCGA analyses indicated that all 3 HAS genes (HAS1, HAS2, HAS3) as well as MUC1 and MUC5B are significantly differentially expressed in GBM tumors compared to healthy patients. However, only HAS1 and MUC1 significantly decreased survival in GBM patients when upregulated in primary tumors. Upregulated MUC1 also significantly decreased survival in GBM patients with recurrent tumors. However, these results do not indicate the GBM tumor cell types (GSC/non-GSC) where each of these genes are differentially expressed. To clarify implications in GSCs, RNAseq analysis demonstrated significant differences in HAS2, HAS3, MUC1, MUC3A, MUC4, MUC17, and MUC20 between G34 types, potentially revealing different glycocalyx expression profile in GSCs compared to other GBM tumor cells once true gene expression levels are evaluated. Flow cytometry results revealed a marked expression of hyaluronic acid on G34 parental cells. Taken together, these results provide sufficient premise for further HA and mucin-based studies to investigate their effect on IFF-mediated GSC invasion both in vitro and in vivo. The results of these future studies will advance our understanding on how the glycocalyx affects GBM aggressiveness and potentially add to a list of precision medicine biomarkers for either GBM diagnosis or treatment targets.

Acknowledgements and/or References (Optional):: This work is supported by NCI Cancer Center Support Grant P30CA012197, NCI R37CA222563, and the Red Gates Foundation.