(542e) CRISPR Editing of the Cancer Glycocalyx to Tune Mechanically Regulated Migration in Glioblastoma Multiforme

Introduction: Glioblastoma multiforme (GBM) is one of the most aggressive cancer types that affect the brain and is characterized by a bulky presence of sugar glycocalyx, to mechanically tense its membrane, fuel aggression and prevent immune recognition. This glycocalyx is a potent mechanosensor that transmits mechanical and chemical environmental cues as well as all signals that must navigate this barrier to reach the cell. Any changes to the total composition of this sugar layer would in turn impact on the harmonious interaction between the cells and its environment. In more than 95% of cancer cells, glycocalyx is enriched/capped with bulk of negatively charged sialic acid sugar. This charged sugars mediate cell-cell, cell-matrix, sugar-sugar repulsion as well as modulation of cancer-immune interaction to further influence cancer detachment from the tumor mass, hence immune evasion and metastasis. All of these repulsive forces put together are amplified in the membrane vicinity of the tumor cell in pico-newton magnitude to activate integrin focal adhesion kinases known to increase tumor migration, invasion, and proliferation. So far, treatment options to prevent GBM metastasis or increase immune recognition have been ineffective as most patients do not survive 8 months post-diagnosis. In a bid to extend the length of survival in patients, modification of the glycocalyx remains a promising therapeutic target worthy of perturbation. GBMs expresses a complex mixture of glycoproteins, glycolipids, and mucins, and insight into the specific glycan that is homing the critical functional end group sugar (sialic acid) to mediate mechanical forces on its membrane as well as immune evasion is needed to effectively edit it. To this end, we hypothesize that CRISPR-Cas9 mediated editing of either the chain initiating enzymes of N-linked glycans (MGAT1), O-linked glycans (COSMC), or glycolipids (UGCG) can edit the glycocalyx in such a way that it de-escalates the membrane tension, disrupts the aggressive ability of GBMs to migrate, signal, and evade immune recognition.

Materials and Methods: CRISPR plasmids with gRNA sequences against human COSMC, UGCG and MGAT1 in the pSpCas9(BB)-2A-GFP (PX458) V2.0 vector were generated and COSMC-KO (U87-COSMC-KO), MGAT1-KO (U87-MGAT1-KO) and UGCG-KO (U87-UGCG-KO) single KO cell lines were created by chemically transfecting the WT U87s with these vectors using lipofectamine 3000^TM reagent. To characterize the KOs, the region surrounding the CRISPR target site was amplified from genomic DNA and specific deletions confirmed. Approximately 6 weeks after transfection, morphological changes were quantified by microscopy with firmly adherent, spindle like cells being more aggressive and mesenchymal GBMs and loosely attached, more rounded cells being pro-neural and less aggressive. Then we fashioned polyacrylamide gels at a range of stiffnesses (400-60,000 Pa) conjugated with Fibronectin and monitored if the knockouts migrate differently than the controls and if the stiffness affects the migration of the GBMs. Next, we checked to see if the signaling (FAK) is altered in any of the GBM knockouts through the use of tyrosine specific activated antibodies (Tyr397 FAK and pFAK) on both the wild type and mutants to check changes in signaling profiles and response to mechanical cues.

Results: Microscopic observation (Figure 1) showed a trend towards a more pro-neural phenotype, as expected, in each of U87-COSMC-KO, and U-87-UGCG-KO and U87-MGAT1KO. What is more surprising is that U87-COSMC-KO and U-87-UGCG-KO showed the most robust trend as they were detached, less aggressive, and had the most recognizable proneural phenotype. This phenotype is seen more often in healthy cells, an observation that is expected to correlate with glycocalyx content and size upon quantification.

Conclusions, and Discussions: while both MGAT1 and UGCG KOs adopted a relaxed tension morphology, COSMC appear to be the most critical sugars responsible for GBM aggression. However, our ongoing experiments are quantifying the changes in surface topography and glycocalyx architecture as well as signaling and migration associated with the phenotypic changes seen in each KOs with the hope of identifying mucin or glycolipids as key players in GBM aggression. We plan to inject each of the KOs in mice to study tumor development and survival. Finally, we will check for immune and drug susceptibility for each KOs. This could ultimately lead to new potential therapies that could increase immune recognition and in improved patient prognosis in GBM.

Acknowlegement: This work was funded by NIH-NIGMS GM143357, Center for Engineering MechanoBiology, CEMB CMMI: 15-48571, NJIT Faculty Seed Grant and New Jersey Health Foundation (NJHF), Grant number: PC:26-24, to the cellular motion lab.