SUMMARY
Glioblastoma remains a deadly cancer driven by invasion of tumor cells into the brain. Transcriptomic analyses have revealed distinct molecular subtypes, but mechanistic differences that explain clinical differences are not clear. Here, we show that, as predicted by the motor-clutch model for cell migration, mesenchymal glioma cells are more spread, generate larger traction forces, and migrate faster in brain tissue compared to proneural cells. Despite their fast migration and comparable proliferation rate in vitro, mice with mesenchymal tumors live longer than mice with proneural tumors, which was correlated with an immune response in the mesenchymal mice that included T cell-mediated killing of cancer cells, similar to human tumors. Thus, mesenchymal tumors have aggressive migration, but are relatively immunologically ‘hot’ which suppresses net proliferation. These two features counteract each other and may explain the lack of a strong survival difference between subtypes clinically, while also opening up new opportunities for subtype-specific therapies.
INTRODUCTION
Glioblastoma (GBM: WHO grade IV primary brain tumor) progression can be characterized in terms of tumor growth and spreading, two key parameters which are influenced by many of the hallmarks of cancer (Hanahan and Weinberg, 2011). In GBM, tumor spreading is driven by tumor cells’ ability to infiltrate healthy brain parenchyma, which prevents complete surgical resection and results in tumor recurrence (de Gooijer et al., 2018; Hoelzinger et al., 2007; Lefranc et al., 2005). Molecular and genetic analyses of human GBM have identified at least three distinct molecular subtypes: proneural, classical, and mesenchymal (Phillips et al., 2006; Verhaak et al., 2010; Wang et al., 2017). These subtypes were shown to strongly correlate with specific genetic alterations (Mesenchymal: NF1 loss; Classical: EGFRvIII; Proneural: PDGFRA) and cellular developmental states (Neftel et al., 2019; Patel et al., 2014; Verhaak et al., 2010; Wang et al., 2017). Despite accumulating evidence of distinct transcriptomic and genetic signatures, the characteristic mechanistic differences between such signatures, if any, have not been identified. As a result, it remains unclear how knowledge of the different subtypes should inform clinical decisions.
One intriguing correlate of subtype is the level of CD44 expression, a cell surface protein expressed on tumor and immune cells, which is known to play a role in cancer progression across a variety of cancers including GBM (Bhat et al., 2013; Klank et al., 2017; Mao et al., 2013; Mooney et al., 2016; Naor et al., 2002; Neftel et al., 2019; Ozawa et al., 2014; Pietras et al., 2014; Toole, 2009; Wang et al., 2017). In GBM, we previously showed that CD44 expression is a prognostic marker with a biphasic dependence: better outcomes are observed at both lower and higher levels of CD44 while poorer outcomes are observed at intermediate levels, an example of optimality and the ‘goldilocks’ phenomenon (Klank et al., 2017). In animal models, CD44 expression has further been shown to correlate with glioma cell migration in a biphasic relationship with a peak migration rate at intermediate expression level, which also correlated with the minimum in survival in both the animal model and human GBM (Klank et al., 2017). In addition, CD44 transcript levels are shown to vary across GBM molecular subtypes with elevated expression in mesenchymal tumors (Phillips et al., 2006; Pietras et al., 2014; Verhaak et al., 2010). CD44 expression in the mesenchymal tumors is, on average, closer to the CD44 level that corresponds to the minimum in patient survival than the proneural subtype (Klank et al., 2017). As an adhesion molecule, CD44 engages the extracellular matrix with the actin cytoskeleton through adapter proteins to mediate cell migration (Toole, 2009). This suggests that mesenchymal cells have a near-optimal level of CD44 adhesion molecules to serve as molecular “clutches” that resist myosin II motor forces, allowing them to migrate faster than proneural cells which on average have a lower, suboptimal level of CD44 clutches (see Figure 2B and 4E in Klank et al., 2017). This could then explain the slightly worse outcomes for mesenchymal patients and higher cell migration and invasion (Wang et al., 2017; Yoshida et al., 2012). In addition, it would predict that mesenchymal cells would have lager cellular spread area, be more polarized, and generate more traction force as they migrate. More generally, lower CD44 is indicative of an epithelial state and higher CD44 indicative of a mesenchymal state (Bloushtain-Qimron et al., 2008; Polyak and Weinberg, 2009), and so an increase in myosin motors and adhesions, either integrin- or CD44-mediated, may be driving the epithelial-to-mesenchymal transition (EMT) in a variety of cancers such as breast cancer (Mekhdjian et al., 2017).
Based on these previous results, we tested the hypothesis that a key mechanistic difference between GBM molecular subtypes is that proneural cells are slow migrating and mesenchymal cells are fast migrating. To address this question, we generated animal models recapitulating the transcriptomic signatures of human mesenchymal and proneural GBM in an immune competent background using perturbations of known GBM oncogenic pathways. Specifically, mesenchymal and proneural-like tumors were driven by SV40-large T (LgT) antigen, to mimic common inhibition of p53 and Rb signaling found in GBM (Ahuja et al., 2005; McLendon et al., 2008), in combination with either NRASG12V (NRAS) or PDGFB (PDGF), respectively, which resulted in mesenchymal or proneural transcriptomic features with only a single genetic change required to switch subtypes in a wild type mouse background. As predicted, CD44 expression was higher in NRAS-driven tumors and, consistent with our simulation predictions, ex vivo brain slice live imaging showed NRAS tumor cells migrate faster than PDGF tumor cells, and exhibit greater spreading, polarization, and force generation as well. Despite increased migration, the NRAS cohort had better survival than PDGF which was attributed to enhanced antitumoral immune response in NRAS tumors, consistent with increased immune cell infiltration in human mesenchymal GBM (Doucette et al., 2013; Hara et al., 2021; Wang et al., 2021, 2017). Overall our work identified a clinically actionable difference in migration mechanics between GBM subtypes and establishes an integrated biophysical modeling and experimental approach to mechanically parameterize and simulate distinct molecular subtypes in preclinical models of cancer.
RESULTS
Genetically induced high-grade glioma mouse models recapitulate the transcriptomic signatures of mesenchymal and proneural GBM
To characterize the mechanics of GBM subtypes, we utilized the Sleeping Beauty (SB) transposon-based gene transfer system to induce high grade gliomas in immunocompetent FVB/NJ-strain mice (Calinescu et al., 2015; Klank et al., 2017; Koschmann et al., 2016; Núñez et al., 2019; Wiesner et al., 2009). Constructs of plasmids carrying SB transposons with oncogenic driver transgenes (SV40-LgTA+NRASG12V or SV40-LgTA+PDGFB; here termed NRAS and PDGF, respectively) were used to model mesenchymal and proneural GBM tumors, respectively (Figure 1A). DNA plasmids carrying an SB transposon with encoded firefly luciferase and green fluorescent protein (GFP) transgenes, and expressing SB transposase were co-injected to allow for confirmation of successful gene transfer, detecting tumor development and monitoring tumor growth using bioluminescence imaging (BLI) and single cell tracking using fluorescence microscopy. Similar to human GBM, histological sections from these tumors exhibited highly mitotic tumor cells, necrosis, anaplasia, and perivascular infiltration and proliferation, (Figure 1B).
To assess whether the NRAS and PDGF tumors recapitulated the mesenchymal and proneural subtypes, respectively, we performed cross-species transcriptomic analysis using bulk RNA sequencing data from mouse and human tumors. Bulk RNA sequencing was performed on tumor tissues from both cohorts and on normal brain tissues (NBT) (NRAS N=4, PDGF N=4, and NBT N=3). IDH-WT human GBM transcriptomic profiles were retrieved from Broad GDAC Firebrowse (Brennan et al., 2013). Unsupervised hierarchical clustering of the mouse dataset (Table S1) revealed clear differences between normal tissue and tumor tissue and between NRAS and PDGF tumors (Figure S1A). Not surprisingly, gene ontology enrichment analysis, using EnrichR (Kuleshov et al., 2016), showed an enrichment of cell cycle related processes in tumor tissue specific gene cluster (817 genes) (Figure S1B) and neuronal processes in normal tissue cluster (1722 genes) (Figure S1D). Interestingly, the NRAS-specific cluster (1327 genes) was enriched with cytokine-mediated signaling and inflammatory response processes (Figure S1C).
To determine whether any variations observed between the two mouse cohorts were also present in human tumors, unsupervised hierarchical clustering was performed on both mouse and human tumor datasets and clusters were independently identified in both datasets (Table S2). Three and 10 gene clusters were identified in both mouse and human datasets, respectively, as shown in Figure 1C. We identified mouse cluster MC1 (n=1534 genes) as being significantly enriched in genes found in human cluster HC1 (n=1186 genes, p < 1x10-15), MC2 (n=414 genes) is significantly enriched in genes found in HC2 (n=1098 genes, p < 1x10-15), and MC3 (n=232 genes) is significantly enriched in genes enriched in HC4 (n=432, p < 1x10-15). These results show that conserved transcriptomic patterns distinguish subtypes of both mouse and human tumors.
To assess whether the transcriptional patterns present in the mouse tumor models represent previously described GBM subtypes, we compared the expression of identified mouse gene clusters within human GBM subtypes. We found that MC1, which is enriched in NRAS tumors, is significantly enriched in human mesenchymal GBM relative to proneural and classical GBMs (Figure 1D). In contrast, MC3, which is enriched in PDGF tumors, is significantly enriched in proneural GBM relative to mesenchymal and classical GBMs (Figure 1D). Furthermore, we found the expression of known mesenchymal and proneural genes and gene signatures are relatively elevated in NRAS and PDGF tumors, respectively (Figure 1E, 1F and S2). These results demonstrate that NRAS and PDGF tumors transcriptionally resemble mesenchymal and proneural GBMs, respectively.
Motor-clutch modeling of cell migration predicts NRAS/Mesenchymal tumor cells will migrate faster, have larger cell spread area, and generate more force than PDGF/Proneural tumor cells
To examine tumor cell migration, we used our cell migration simulator (Bangasser et al., 2017; Klank et al., 2017) to predict migration phenotypes in response to gene expression changes. The cell migration simulator is based on the motor-clutch model which incorporates actin-based protrusion dynamics, mass conservation, and force balances to reproduce cell polarization and random motility in 1D and 2D compliant microenvironments (Bangasser et al., 2013, 2017; Chan and Odde, 2008; Estabridis et al., 2018; Hou et al., 2019; Klank et al., 2017; Liu et al., 2019; Prahl et al., 2018, 2020). The number of adhesion/clutches and motors are key determinants of cell migration, with a relative balance being essential for efficient migration (Bangasser et al., 2013, 2017; DiMilla et al., 1991). Using a set of 54 cell migration genes expressed in the human U251 GBM cell line (Bangasser et al., 2017), NRAS tumors significantly upregulate transcription of adhesion and adapter genes (Figure S3A). A similar set of genes was also significantly upregulated in MES relative to PN GBM (Figure S3B). Both NRAS and MES tumors upregulated CD44 and its cognate adhesion adapter gene moesin (MSN), which mechanically links the CD44 cytoplasmic tail to F-actin (Fehon et al., 2010; Freeman et al., 2018; Legg and Isacke, 1998; Ponta et al., 2003; Toole, 2009; Tsukita et al., 1994; Yonemura, 1998). Notably, the levels of myosin motor genes were not differentially expressed in the mouse dataset, while, in the human dataset, MYH9 and MYO1C were modestly upregulated in MES tumors but to a lesser degree than adhesion molecules (Figure S3B). These results suggest NRAS/Mesenchymal tumor cells have a higher number of adhesion/clutches than PDGF/Proneural tumor cells and little to no change in the number of motors.
Based on these results, we simulated the effect of CD44 expression level on cell migration by simply adjusting the number of adhesion bonds (number of clutches, Nc) in the model (Bangasser et al., 2017; Klank et al., 2017). We used low clutches relative to motors (low adhesion) to simulate PDGF/Proneural cells and a medium level of clutches that balanced the number of motors (optimal adhesion) to simulate NRAS/Mesenchymal cells (Figure 2A). Our simulations show that lowering the number of adhesions, representing the PDGF/Proneural case, results in reduced cell migration, force transmission, cell spread area, and cell polarization due to an insufficient number of clutches relative to the number of motors, as shown in Figure 2. In the case where the number of clutches and motors are balanced, representing the NRAS/Mesenchymal case, where the number of adhesions is increased while holding the number of motors constant, simulated cells recover their ability to migrate, transmit forces, spread, and polarize across a range of substrate stiffnesses. Consequently, simulation results predict that NRAS/Mesenchymal tumor cells will migrate faster than PDGF/Proneural tumor cells due to increase of adhesion (i.e. CD44) expression in NRAS/Mesenchymal tumors and not due to small difference in molecular motor expression (Figure S3A and S3B). In addition, due to their higher number of clutches and balanced motor-clutch ratio, NRAS/Mesenchymal tumor cells are predicted to generate higher force, have larger spread area, and be more polarized.
NRAS/Mesenchymal tumor cells migrate faster than PDGF/Proneural tumor cells in brain tissue
To test our model prediction that NRAS/Mesenchymal cells migrate faster than PDGF/Proneural cells, we performed live cell imaging on tumor bearing mouse brain slices using confocal microscopy. Time-lapse images of GFP-positive tumor cells were used to track single cell migration and generate single cell trajectories. As shown in Figure 3A, 3B and Video S1, NRAS/Mesenchymal tumor cells appeared qualitatively to move farther, have larger spread area, and polarize to a greater extent than PDGF/Proneural tumor cells. Quantitative analysis of single cell trajectories confirmed that NRAS/Mesenchymal tumor cells have a higher random motility coefficient than PDGF/Proneural tumor cells (30.1 µm2 hr-1 vs 2.5 µm2 hr-1, p<0.001; see Figure 3C and S4A). In addition, morphological analysis of tumor cells revealed cell spread area and cell aspect ratio (i.e. polarization) are also higher in NRAS/Mesenchymal tumor cells than PDGF/Proneural tumor cells (406.6 µm2 vs 235.8 µm2, p<0.00001 and 2.1 vs 1.7, p<0.001, respectively, see Figure 3D, 3E, S4B and S4C). As predicted by our modeling, and the hypothesis that NRAS/Mesenchymal has balanced motors and clutches while PDGF/Proneural lacks sufficient clutches, we find NRAS/Mesenchymal cells migrate faster, are more spread, and are more polarized than PDGF/Proneural cells.
Migration phenotype is species and tumor microenvironment independent
To determine whether migration phenotype is cancer cell intrinsic as predicted by our modeling and not due to microenvironment differences, we generated three primary mouse lines grown as neurospheres from each cohort to investigate their migration phenotype outside their tumor microenvironment. Organotypic mouse brain slice culture was used to image tumor cell migration in healthy mouse brain slices (Liu et al., 2019). Dissociated mouse tumor cells were plated and allowed to invade and migrate in healthy mouse brain slices. Figure 4A shows representative fluorescence images of primary isolated cells in organotypic slice culture. Time-lapse imaging was used to track single cells and quantify their migration rates. Consistent with the ex vivo migration in intact tumor-bearing brain slices, random motility coefficient in normal mouse brain tissue is higher for primary NRAS/Mesenchymal tumor cells than primary PDGF/Proneural tumor cells (131.4 µm2 hr-1 vs 31.3 µm2 hr-1, p<0.00001; Figure 4B, S4D and Video S2). The area of cell spreading is also higher in primary NRAS/Mesenchymal tumor cells than PDGF/Proneural (1075.1 µm2 vs 838.8 µm2, p<0.001; Figure 4C and S4E). Furthermore, cell aspect ratio, the ratio of major and minor axis of a fitted ellipse, was trending higher in NRAS/Mesenchymal tumor cells than PDGF/Proneural but did not reach statistical significance (Figure 4D and S4F).
To assess the relevance of these results to human GBM, we tested the migration phenotype of six patient-derived xenograft (PDX) lines (three mesenchymal and three proneural) using the organotypic mouse brain slice culture. Figure 4E shows representative fluorescence images of PDX cells in organotypic slice culture. We found mesenchymal PDX cells migrate faster than proneural PDX cells (43. µm2 hr-1 vs 8.8 µm2 hr-1, p<0.00001; Figure 4F, S4G and Video S3). In addition, similar to our mouse models, mesenchymal PDX cells have larger area of cellular spreading and aspect ratio relative to proneural PDX cells (701.7 µm2 vs 564.6 µm2, p<0.00001 and 2.3 vs 2.0, p<0.001, respectively, see Figure 4G, S4H, 4H and S4I).
Traction strain energy is larger for NRAS/Mesenchymal cells than for PDGF/Proneural cells, consistent with model predictions
In addition, cell migration simulations predict NRAS/Mesenchymal tumor cells would have increased force generation as a result of higher number of clutches resulting in balanced myosin motors and clutches, relative to PDGF/Proneural tumor cells which would have insufficient clutches relative to motors (Figure 2D). Using the primary isolated mouse lines, traction force microscopy was used to measured traction strain energy generated by tumor cells on polyacrylamide hydrogels (Bangasser et al., 2017; Butler et al., 2002). Because transcriptomic analysis showed an overall higher level of adhesion molecules in NRAS/Mesenchymal tumor tissues, including CD44 and integrins, Type-I Collagen was used to coat polyacrylamide hydrogels (Figure S3A). Because of the overall higher level of adhesion Consistent with model predictions, NRAS/Mesenchymal tumor cells generate higher traction strain energy than PDGF/Proneural tumor cells across different substrate Young’s moduli (Figure 5A and 5B). Cell spread area is also higher in NRAS/Mesenchymal than PDGF/Proneural on polyacrylamide hydrogels (Figure 5C). In addition, NRAS/Mesenchymal cells exhibit stiffness sensitive cell spreading; cells on stiff substrate (4.6 and 9.3 kPa) were more spread than on soft substrate (0.7kPa, p <0.00001; see Figure 5C). Furthermore, we also examined force generation of mesenchymal and proneural PDX cells in mouse brain slices. Qualitative analysis of vasculature deformation is consistent with the model prediction that mesenchymal PDX cells generate larger deformations relative to proneural PDX cells, as shown in Figure 5D, 5E and Video S3.
NRAS/Mesenchymal mice have better survival and slower tumor growth rate
Since NRAS/Mesenchymal cells have nearly optimal CD44 expression, and therefore higher motility compared to PDGF/Proneural cells which have a suboptimal low level of CD44 expression (Klank et al., 2017), we asked whether the differences in migration rate, morphology and force generation correlate with disease progression and survival. Specifically, based on the faster migration in the NRAS/Mesenchymal cohort, we expected that these mice would progress faster and die sooner than PDGF/Proneural mice. To test this hypothesis, we measured survival times of tumor bearing mice and found that, opposite to our expectation, the NRAS/Mesenchymal cohort had better median survival than PDGF/Proneural cohort (NRAS N=21, PDGF N=24, 65 days vs. 35 days, log-rank test, p<0.0001; Figure 6A). To explain the difference in survival, we quantified in vivo tumor growth using bioluminescence imaging (BLI) of tumor-bearing mice. Consistent with their shorter survival, we found PDGF tumors grew twice as fast as NRAS tumors (Slopes: 0.127±0.01191 vs. 0.0716±0.00343 p <0.001, Figure 6B and 6C). Using mouse tumor neurospheres, we quantified mouse primary tumor cell line proliferation rates in vitro and found no significant difference between NRAS/Mesenchymal and PDGF/Proneural cells (Figure 6D). These results imply that an additional factor, besides proliferation or migration, enables the NRAS/Mesenchymal mice to live longer and their tumors to grow slower in vivo than PDGF/Proneural mice.
NRAS/Mesenchymal mice have increased immune response relative to PDGF/Proneural mice
Because mesenchymal GBMs are known to be relatively immunologically “hot” – which presumably confers a survival benefit due to an antitumoral immune response-relative to immunologically “cold” proneural GBMs (Doucette et al., 2013; Neftel et al., 2019; Wang et al., 2017), we assessed the extent to which mouse NRAS/Mesenchymal tumors induce an immunological response relative to PDGF/Proneural tumors and normal brain tissues. Using our mouse transcriptomic dataset and previously published GBM immune gene sets (Doucette et al., 2013), NRAS/Mesenchymal tumors were found to have increased expression of both immune activators and suppressors gene signatures similar to human mesenchymal GBM (Doucette et al., 2013), as shown in Figure 7A and 7B. Specifically, expression of immune cell marker genes such as Aif1, Itgam (microglia/macrophages) and Cd3 (T cells) are elevated in NRAS/Mesenchymal tumors relative to PDGF/Proneural tumors (Figure 7C upper panel) and in mesenchymal GBMs relative to proneural and classical GBMs (Figure 7C lower panel). Elevated expression of immune cell markers is indicative of increased immune cell infiltration in mesenchymal tumors. Consistent with the transcriptomic findings, IHC staining of NRAS/Mesenchymal and PDGF/Proneural tumor sections revealed significant levels of immune cell infiltration, including both microglia/macrophages and T-lymphocytes, in NRAS/Mesenchymal but much less so in PDGF/proneural tumors (Figure 7D). Furthermore, increased immune infiltration and activity in NRAS/Mesenchymal tumors was accompanied by increased cell killing as measured by granzyme B and cleaved caspase-3 staining (Figure 7D). Image clustering analysis was used to quantify CD3, IBA1, granzyme B, and cleaved caspase-3 staining, and statistically significant differences were observed between NRAS/Mesenchymal and PDGF/Proneural cohorts (Figure 7E-H). This anti-tumor response was also evident in the three instances where NRAS mice developed tumors and tumor regression was observed (Figure S5). Despite the anti-tumor immune response observed in NRAS/Mesenchymal tumors, transcriptomic analysis also revealed elevated relative expression of immune checkpoint genes including PDL1, CTLA4, and CD200R1 (Figure S6). Thus, the NRAS/mesenchymal tumors, like human mesenchymal GBMs, are relatively immunologically “hot” with evidence of both immune activation and immune suppression, as well as evidence of cell killing. Altogether, the enhanced immune cell-associated tumor cell killing provides a mechanism by which survival is extended in NRAS/Mesenchymal tumors despite their enhanced migration speeds relative to PDGF/Proneural tumors.
Brownian dynamics simulations explain NRAS/Mesenchymal and PDGF/Proneural tumor progression
To quantitatively describe the interplay between tumor cell migration and proliferation and anti-tumoral immune response in tumor growth, we developed a three-dimensional (3D) Brownian dynamics tumor simulator (BDTS) based on our original 1D Brownian dynamics simulator (Klank et al., 2018; Ray et al., 2018). The simulator takes into account anti-tumoral immune cells that infiltrate tissue, migrate, proliferate, encounter cancer cells, deliver cytotoxic agents, dissociate from cancer cells, undergo exhaustion, and, eventually, undergo apoptosis (or egress to lymphatics) as shown in Figure 8A. At the same time, cancer cells migrate, proliferate and undergo CTL-mediated death in the presence of anti-tumoral immune cells in the case NRAS/Mesenchymal tumors. In the case of PDGF/Proneural tumors, no anti-tumoral immune cells were simulated. Figure 8B shows simulation output at day 0 and day 16, which showed the observed behavior of overall faster growth of PDGF/Proneural tumors. In the NRAS/Mesenchymal tumor simulations, cancer cells appear more dispersed, whereas, in PDGF/Proneural simulations, cancer cells are less dispersed. Simulated tumor growths were plotted in Figure 8C and shown in Video S4. Using the parameters in Table S5, including the experimentally observed single cell migration speeds and neurosphere proliferation rates, simulated tumors qualitatively recapitulate the in vivo growth profile of NRAS/Mesenchymal and PDGF/Proneural tumors without parameter adjustment (Figure 8C).
DISCUSSION
Understanding glioma progression and the mechanism driving glioma cell migration is critical for the design of effective therapies. Here we developed high-grade glioma mouse models which capture the transcriptomic and the immune microenvironment changes associated with human proneural and mesenchymal GBMs. Using the mouse models and PDX lines, we defined a mechanistic difference in glioma cell migration which highlights a functional characteristic of GBM molecular subtypes. The migration difference was consistent with changes in cellular adhesion, notably by CD44, but not molecular motors such as myosin II motors. This finding points toward an anti-migratory therapy approach targeted against cellular adhesion as opposed to myosin motors. With the negative Phase III clinical trial of the integrin-inhibitor, Cilengitide for GBM (Stupp et al., 2014), it is possible that integrins may not be the major adhesion molecules utilized by glioma cells to migrate but instead they could utilize CD44. While anti-CD44 therapies have not been tried in GBM, an anti-CD44 monoclonal antibody therapy (RO5429083, Roche, Basel, Switzerland) has been investigated in Phase I trials in patients with solid tumors and with AML (Menke-van der Houven van Oordt et al., 2016; Vey et al., 2016). Therefore, an anti-CD44 therapy could provide clinical benefits by slowing glioma migration.
Furthermore, the upregulation of CD44 in mesenchymal tumors is supportive of the existing literature which defines CD44 as a marker of cancer stem cell and EMT (Bloushtain-Qimron et al., 2008; Polyak and Weinberg, 2009; Ponta et al., 2003). During EMT, cancer cells take a more mesenchymal migratory phenotype to allow them to migrate through dense ECM and metastasize (Chaffer and Weinberg, 2011). Our results associate enhanced migration in mesenchymal glioma cells with increased traction forces due to increased adhesion molecules expression ‘clutches’ such as CD44. Similarly, in breast cancer cells, TGF-β-induced EMT is associated with increased traction forces and clutch number (Mekhdjian et al., 2017). Interestingly, downregulation of NF1, a negative regulator of Ras, in epithelial breast cancer cells and Schwann cells also induces expression of transcription factors related to EMT (Arima et al., 2010). Moreover, in our study, NRASG12V expression was used to mimic NF1 downregulation and inactivation in mesenchymal GBM (Krusche et al., 2016; Verhaak et al., 2010). Ras hyperactivation of MAPK pathway is required for EMT but not PI3K activation by Ras (Janda et al., 2002). Altogether, our results implicate EMT in enhanced glioma cell migration and force transmission associated with increased molecular clutches through, and suggest upregulation of clutches, either integrins or CD44, as a conserved feature of EMT across a range of cancers.
Despite the faster migration of the NRAS/Mesenchymal cells, the anti-tumoral immune response within the NRAS/Mesenchymal mouse model is able to slow disease progression and improve survival despite enhanced migration relative to the PDGF/Proneural mouse model. Such an anti-tumor response could potentially be used to slow disease progression and improve clinical outcome for GBM patients. In both mouse and human GBM, mesenchymal tumors are immunologically ‘hot’ relative to the immunologically ‘cold’ proneural tumors (Doucette et al., 2013; Wang et al., 2017). Despite the presence of immune cells within mesenchymal tumors, immune suppression leads to tumors ultimately prevailing against the anti-tumoral immune response. Based on these findings, we propose an immune checkpoint inhibition strategy, in combination with an anti-migratory therapy, targeting mesenchymal GBM but not proneural GBM.
Our study utilizes an integrated, state of the art experimental approach to study GBM progression and model GBM molecular subtypes by switching a single oncogenic driver (NRASG12V ↔ PDGFB), in an immunocompetent background without the need for genetically engineered mouse strains or further breeding. Using live cell and brain slice imaging, we identify key mechanical differences between mesenchymal and proneural tumor cells, with mesenchymal cells have larger cellular spread area, generate larger forces, and migrate faster. The functional differences were all predicted by a motor-clutch model for cell adhesion and migration where mesenchymal cells have an optimal level of CD44-mediated adhesion (clutches) relative to myosin motors, while proneural cells lack sufficient CD44 to match the myosin motor activity. Despite the faster migration, NRAS/Mesenchymal mice live longer, consistent with the presence of an anti-tumoral immune response that is lacking in PDGF/Proneural mice (Fig. 8D), dynamics that are readily captured computationally with little parameter adjustment using a 3-D Brownian dynamics tumor simulator (BDTS). Overall, this work establishes an integrated in vivo genetic and biophysical modeling framework to connect animal model and human transcriptionally-defined subtypes to fundamental mechanistic understanding, which has the potential to enable a new modeling-centric approach to clinical translation with application in a wide range of cancers (Brubaker and Lauffenburger, 2020).
MATERIALS AND METHODS
Generation of mouse tumor models
All animal studies were conducted according to guidelines approved by the Institutional Animal Care and Use Committee at the University of Minnesota. All animals were housed in a daily monitored animal facility. FVB/NJ strain of mice were used in this study. Malignant gliomas were induced in neonatal mice by DNA plasmid injection into the right lateral ventricle as described previously (Calinescu et al., 2015; Wiesner et al., 2009). Briefly, neonatal mice were injected with 1 µg of plasmid DNA mixed with polyethyleneimine (jetPEI, Polyplus, Berkeley, CA), and 5% dextrose in a total volume of 2 µL at a rate of 0.7 µL/min. The following four plasmids were used (1:1:1:1) ratio: empty vector, pT2/C-Luc/PGK-SB100, pT/CMV-LgTAg-IRES-GFP, pT2/Cag-NrasV12 or pT2/Cag-mPDGF. Animals were monitored daily for morbidity by bioluminescent imaging.
Immunohistochemistry of mouse tumor sections
Formalin fixed and paraffin embedded (FFPE) mouse brain tissues were used to prepare 4 µm thick slides. FFPE tissue slides were stained with hematoxylin and eosin (H&E) or IHC using standard methods. Table S6 contains a list of antibodies and reagents used for antigen retrieval, blocking and detection.
Quantification of IHC staining of mouse tumor sections
Immunohistochemistry data was quantified by counting the number of pixels in an image that were positively DAB stained. To avoid user bias and subjective counting, k-means clustering was used to identify pixels representing areas of positive DAB, hematoxylin staining, and background. In this implementation, every pixel in an analyzed image is assigned to one of four clusters, each cluster representing a different component in the image: positive DAB-stained areas (brown), positive hematoxylin counter-stained areas (blue), unstained tissue (light blue), and background glass slide (beige). Digital images of equal sizes (2000x2000 pixels) of DAB stained and hematoxylin counterstained tumor samples were converted from RGB to the HSV color model.
Using a custom written MATLAB algorithm, user input is used to define areas representing the four components (positive DAB stain, hematoxylin counterstain, unstained tissue, and background glass slide). These points are used as the initial estimates for the centroid locations of each of the four clusters. The squared Euclidean distance between each pixel’s HSV coordinates and the HSV coordinates of each cluster’s centroid is computed. Each pixel is then assigned to the cluster with the minimum squared Euclidean distance to the cluster centroid. Cluster centroids are recalculated as the mean of the HSV coordinated of all current members. This process is iterated until the centroid of each cluster is stable. The number of pixels in the positive DAB-stained cluster is used to quantify percent positive pixels in Figure 7E-H.
Transcriptional profiling of mouse tumors
Mice were euthanized in a CO2 chamber and perfused transcardially with isotonic saline. Mouse brains were extracted and GFP goggle (#FHS/EF-2G2; BLS-ltd, Budapest, Hungary) was used to dissect GFP-positive tumor tissues from NRAS and PDGF mouse brains. Two matched normal brain tissues were collected from brain regions away from the tumor and an additional normal brain tissue sample was collected from a health FVB adult mouse. All samples were immediately placed in RNALater solution (Sigma, St. Louis, MO) for 24 hours then flash frozen in liquid nitrogen and stored in -80 °C for downstream processing. RNA extraction and sequencing were performed at the University of Minnesota Genomics Center (UMGC, Minneapolis, MN). RNA was extracted using RNAeasy Plus Universal Mini kit (Qiagen, Venlo, Netherlands) and libraries were prepared using TruSeq stranded mRNA (Illumina, San Diego, CA).
Next-generation sequencing was performed on the prepared RNA libraries using an Illumina HiSeq 2500 device in high output mode and generated 51 bp reads with an approximate depth of 20 million paired reads per sample. Mapping and expression calculations were generated using the rnaseq-pipeline of Gopher-pipelines (https://bitbucket.org/jgarbe/gopher-pipelines), which executed TopHat2 (Kim et al., 2013) and Cuffnorm (Trapnell et al., 2010) using the UCSC mm10 version of the mouse reference genome. Fastq files and the Cuffnorm output were deposited at Gene Expression Omnibus (GSE161154).
Human GBM transcriptomic data
UNC RNASeqV2 level 3 expression (normalized RSEM) profiles of 171 samples (TCGA-GBM) were retrieved from Broad GDAC Firebrowse (Brennan et al., 2013). IDH status and subtype information were added to each sample based on the Wang et al. classification (Wang et al., 2017). For downstream analysis, 147 IDH-WT samples were used (57 Classical; 52 Mesenchymal; 38 Proneural).
Clustering analysis of mouse and human expression profiles
To analyze the transcriptional profiles of mouse and human datasets, a value of 0.1 was added to all FPKM and RSEM values to minimize the impact of inaccurate low values (Scott et al., 2018). The expression data was log transformed and mean centered and transcripts with Standard Deviation > 1 were clustered using average linkage hierarchical clustering in MATLAB. Pearson correlation was used as the similarity metric. A custom written MATLAB script was used to systematically identify transcriptional clusters within each dataset. In mouse dataset, a correlation greater than 0.5 and > 100 transcripts were used to identify gene clusters. Whereas, in human dataset, a correlation greater than 0.2 and >100 transcripts were used. Fisher’s exact test was used to compare cluster memberships in Figure 1C. All genes identified within each cluster are listed in Table S2.
To quantify the relative expression of gene clusters in Figure 1D and subtype gene signature set in Figure S2C and S2D, the average relative expression of each gene set was computed for all samples. The average relative expression from each sample was plotted and used to calculate mean relative expression within each mouse cohort and within human GBM molecular subtype.
Generation of mouse primary tumor lines
For each line, a tumor bearing mouse was euthanized and transcardially perfused with isotonic saline. Tumor tissue was collected and minced with a scalpel. Minced tumor fragments were incubated with PBS for 15 min at 37 °C. Tumor fragments were further dissociated by mixing them up and down using a 1000 μl micropipette. Finally, tumor suspension was passed through a 40 μm sterile cell strainer (Thermo Fischer Scientific, Waltham, MA) and filtrate was spun down and plated on a Matrigel (354230; Corning, Corning, NY) coated T-75 flasks (Corning, Corning, NY) using NSC media which consisted of DMEM/F12 (Gibco 11320033; Thermo Fischer Scientific, Waltham, MA) with 1X B-27 supplement (Gibco 12587010, Thermo Fischer Scientific, Waltham, MA) and 1X penicillin/streptomycin (Corning, Corning, NY). Twenty ng/ml EGF (PeproTech, Rocky Hill, NJ) and FGF (PeproTech, Rocky Hill, NJ) were added to the cell culture media every 2-3 days. Cells were cultured in a 37 °C 5% CO2 incubator. Once a confluent layer was achieved, cells were detached using 0.25% Trypsin EDTA (Corning, Corning, NY) and frozen down for later use.
Once tumor lines were established, cells were grown as neurospheres using NSC media and Ultra-Low Attachment 6-well plates (Corning, Corning, NY). Neurospheres were dissociated using accutase (Innovative Cell Technologies, San Diego, CA). In total, six different mouse primary tumor lines were established: three NRAS and three PDGF.
Patient-derived xenograft (PDX) cell line culture
The patient-derived xenograft (PDX) cell lines were taken from the Mayo Clinic GBM PDX collection (managed by Dr. Jann Sarkaria, Mayo Clinic, Rochester, MN). Three mesenchymal PDX lines (GBM 16, 39 and 44) and three proneural PDX lines (GBM 64, 80 and 85) were selected to study their migration in organotypic mouse brain slice. Cells were cultured on Matrigel (354230; Corning, Corning, NY) coated tissue culture flasks in a 37 °C 5% CO2 incubator. NSC media was used to culture PDX cell lines and 20 ng/ml EGF and FGF were added to the cell culture media every 2-3 days.
Ex vivo confocal imaging of tumor-bearing brain slices
Tumor-bearing mice were sacrificed when bioluminescence signals were around 5 × 107 radiance (p/sec/cm2/sr). Mice were euthanized in a CO2 chamber and perfused transcardially with isotonic saline. Mouse brains were extracted and kept in chilled artificial cerebrospinal fluid (124 mM NaCl, 2.5 mM KCl, 2.0 mM MgSO4,1.25 mM KH2PO4, 26 mM NaHCO3, 10 mM glucose). Coronal brain slices of thickness 300 µm were prepared using a vibratome (Leica Biosystems, Buffalo Grove, IL). Only one slice was used for live-cell imaging. Isolectin GS-IB4 (Alexa Fluor 568 Conjugate; Molecular Probes, Eugene, OR) was used to label the vasculature.
Before imaging, the slice was washed and transferred into a No. 0 glass bottom 35 mm culture dish (P35G-0-20-C; MatTek, Ashland, MA). A tissue culture anchor (SHD 42-15; Warner Instruments, Hamden, CT) was placed on top of the slice to prevent movement during imaging. The slice was then imaged on a Zeiss LSM 7 Live swept-field laser confocal microscope (Zeiss, Oberkochen, Germany) at 15-minute intervals for up to 20 hours in humidified 5% CO2 air at 37 °C. Images were collected with a 20x objective lens (Plan-ApoChromat 20X, 0.8 NA, Zeiss, Oberkochen, Germany). The number of Z stacks of several regions of interest was adjusted to ensure that the data acquisition of one frame in the time series was completed under 15 minutes (10-20 planes with 10 µm z-step was typically used). Maximum intensity projections from multiple Z stacks were used to generate 2D images for quantitative morphological and trajectory analysis. Images were registered by an affine transformation using ImageJ StackReg plug-in (École Polytechnique Fédérale De Lausanne) to account for stage drift and tissue relaxation during time-lapse imaging.
Live-cell imaging of tumor cells in organotypic brain-slice culture
Healthy mouse brain slices were prepared using the same method as tumor-bearing brain slices above. For experiments using GFP-positive mouse primary tumor lines, neurospheres were dissociated using accutase (Innovative Cell Technologies, San Diego, CA). The protocol of grafting cancer cells into the brain slice was described in details in our previous publication (Liu et al., 2019). Briefly, after creating a single cell suspension, 300,000 cells in 3 mL of media were plated onto the brain slice. The cells were co-cultured with the brain slice for 4 hours at 37 °C and 5% CO2 before imaging to promote cell infiltration into the brain slice. Phenol-free NSC media +2% FBS (Gibco, Thermo Fischer Scientific, Waltham, MA) was used. The slices (Isolectin GS-IB4 stained) were washed several times using cell culture media and transferred into a No. 0 glass bottom 6-well plate (P06G-0-20-F; MatTek, Ashland, MA). The slices were imaged on a confocal microscope with a 10X objective lens (Plan-ApoChromat 10X, 0.45 NA, Zeiss, Oberkochen, Germany). Similar imaging protocol as mentioned above was applied.
For PDX cells, 500,000 – 800,000 cells were stained using DiO membrane dye (V22886; Thermo Fisher Scientific, Waltham, MA) for 5 minutes and then washed twice before plating onto the brain slice inside a 35mm tissue culture dish. The grafting of the cells to the brain slice was similar to the mouse primary cells described above. The slices were imaged on a confocal microscope using a 20X objective lens (Plan-ApoChromat 20X, 0.8 NA, Zeiss, Oberkochen, Germany). Similar imaging protocol was applied.
For both mouse primary tumor cells and PDX cells, we also acquired the maximum intensity projections from multiple Z stacks and performed image registration for further analysis for cell migration and morphology. Images were registered by an affine transformation using ImageJ StackReg plug-in (École Polytechnique Fédérale De Lausanne) to account for stage drift and tissue relaxation during time-lapse imaging.
Single cell tracking and morphology analysis
Single cell migration was tracked as previously described using a custom-written image segmentation algorithm in MATLAB (Bangasser et al., 2017; Klank et al., 2017). Using cell centroid coordinates, the mean squared displacement (MSD) of the cell trajectories over time was calculated using the time interval overlap method (Dickinson and Tranquillo, 1993). To quantify the dispersion of cells, the MSD over time was used to calculate the random motility coefficient µ according to the equation (MSD(t)=4µt; assuming 2-D geometry). Using segmented cell regions, cell area and cell aspect ratio, defined as the ratio between the major and minor axis length of a fitted ellipse, were measured for each individual tracked cell. Distributions of random motility coefficients, cell area and cell aspect ratio for the different conditions were compared using the Kruskal-Wallis test, which is a non-parametric rank-based test.
Bioluminescence imaging and analysis
Animals were monitored for tumor development and progression using noninvasive bioluminescence imaging. Oncogene-injected animals were injected intraperitoneally with 100 µl of 28.5 mg/ml luciferin (GoldBio, St. Louis, MO) prior to imaging. Mice were then anesthetized using 3% isofluorane and imaged on an IVIS50 or IVIS100 instrument (Xenogen, Alameda, CA). Images were acquired ten minutes after injection with five minutes exposure time (Xenogen LivingImage Software, Alameda, CA). To avoid saturation, exposure time was reduced appropriately in fully grown tumors and accounted for in the analysis. BL images were processed using a custom written MATLAB algorithm where background signal was subtracted and pixels away from the tumor were set to zero. BL signal from each animal was then normalized to the initial time point when tumor was first detected.
Quantification of proliferation of mouse primary tumor line
To measure proliferation rate, 200,000 cells from each line were plated into an ultra-low adhesion 6-well plate and grown as neurospheres. Growth factors were added every 2-3 days. At day six, neurospheres were dissociated and cells counted. After counting, the remaining cells were replated and resumed growing as neurospheres. Cells were also counted and replated at day nine and day 13. Experiment was repeated three times using each of five mouse primary tumor lines (three NRAS and two PDGF tumor lines). For each replicate, the average cell count for each cohort was calculated using the cell count from the different corresponding tumor lines.
Traction force measurements
Traction force measurements of mouse primary tumor lines were performed using traction force microscopy on polyacrylamide gels embedded with 0.2 µm crimson fluorescent beads (Thermo Fischer Scientific, Waltham, MA) and coated with Type-I Collagen (Corning, Corning, NY). Collagen coated polyacrylamide gels of varying Young’s modulus were prepared as previously described (Bangasser et al., 2017; Wang and Pelham, 1998). Briefly, 0.7, 4.6, and 9.3 kPa polyacrylamide polymer mixture with fluorescent beads were cast onto a No. 0 glass bottom dish then coated with Type-I Collagen using Sulfo-SANPAH (Thermo Fischer Scientific, Waltham, MA). Mouse primary tumor cells were dissociated from neurospheres and plated on prepared gels at low density (1-5 cells/mm2) using NSC media +2% FBS.
To measure force transmission, Traction Force Microscopy (TFM) was performed as previously described (Bangasser et al., 2017). Briefly, Nikon TiE and Ti2 epifluorescence microscopes were used to image fluorescent bead positions before and after cell detachment via trypsin. A Zyla 5.5 sCMOS camera (Andor Technology, Belfast, United Kingdom) and a 40x/0.95NA Ph2 lens with 1.5x intermediate zoom (60x total magnification, 110 nm spatial sampling) was used. Cells were maintained at 37 °C and 5% CO2 for the duration of imaging using an Oko lab Bold Line top stage humidified incubator (Okolab, Ottaviano, Italy). At each stage position, a phase contrast image of the cell was acquired. Next, an image of fluorescent beads at the top surface of the gel was captured using a 575/25 nm LED and eGFP/mCherry filter set with LED fluorescence illumination from a SpectraX Light Engine (Lumencor, Beaverton, OR). Media in dishes was carefully removed, cells were detached with 0.25% trypsin/EDTA (Corning, Corning, NY), and fluorescence images of beads in the absence of cells were acquired at saved stage positions.
Using a previously described method (Bangasser et al., 2017), the displacement field was determined using particle image velocimetry (PIV) using the before and after bead images. A window size of 80-pixels (8.8 µm) square was used in PIV and a final lattice spacing of 20 pixels (2.2 µm) was achieved. Stress and displacement vectors were obtained by solving the inverse Boussinesq problem in Fourier space (Butler et al., 2002). By integrating the product of the stress and displacement vectors over the entire image, substrate strain energy was determined as previously described (Bangasser et al., 2017).
Stochastic cell migration simulator
The previously described (Bangasser et al., 2017; Klank et al., 2017) cell migration simulator (CMS v1.0) was used to simulate cells migration dynamics in response to changes in cell adhesion. The parameters used in the simulations are presented in Table S3. The number of adhesive clutches (Nc) was adjusted to model the change in adhesion observed between PDGF/Proneural and NRAS/Mesenchymal tumors. Nc of 2500 and 7500 clutches were used to simulate PDGF/Proneural and NRAS/Mesenchymal tumor cells, respectively. Four hours of cell dynamics were simulated and the first hour was excluded from analysis to allow the system to reach steady state. Analysis was performed using a ten-minute sampling interval as previously described (Bangasser et al., 2017).
Brownian dynamics tumor simulator (BDTS)
The Brownian dynamics tumor simulator was used as previously described with modifications (Klank et al., 2018; Ray et al., 2018). In the present study, we extended the BTDS to 3-dimentional tumors and incorporated immune cells’ dynamics as shown schematically in Figure 8A. Briefly, simulations started with 27 cancer cells, modeled as rigid sphere with radius (rcancer), placed in a 3x3x3 grid where the distance between each cancer cell (center-center) is 3*rcancer. In simulation including immune response, eight T cells, also modeled as rigid sphere with radius (rCTL), were included, and each was placed 1.5*rcancer away from a randomly selected cancer cell. At each simulation time step of 1 min, cancer and T cells are allowed to move randomly and grow as spheres with a linear volumetric growth rate. Movement and growth are rejected if the newly assigned space is already occupied by a like cell (i.e. no-overlap enforced between cancer cells and between T cells). However, a cancer cell and T cell contact occurs when a proposed cell movement put the distance between cell centers less than or equal (rCTL+rcancer). The duration of the contact is (1/kdissoc), in this case 10 minutes. For every contact, both cancer cell and T cell take a “hit” that reduces their hit points (HP) by one. Both cancer cell and T cell have limited HP and once HP is depleted (equals 0), the cell dies or become exhausted. For NRAS simulations, T cells were added to the tumor simulator and T cell-mediated killing was simulated. For PDGF simulation, only cancer cells were simulated. Cancer cell motility was estimated from ex vivo brain slice imaging of tumor cells (Figure 3C) and proliferation rate was estimated from the in vitro proliferation of mouse primary tumor lines (Figure 7D). The rest of parameters were estimated based on previous published work or used as an adjustable parameter (see Table S5).
Statistical analysis
Fisher’s exact test was used to compare the mouse and human transcriptomic clusters. One-way analysis of variance was used to compare transcript levels. Analysis of covariance (ANCOVA) was used to compare between the two regression lines in Figure 5C. Rank test, Kruskal-Wallis, one-way analysis of variance was used to compare single cell behaviors and IHC quantifications. Where appropriate, a subsequent Dunn-Sidak test for multiple comparisons was performed.
DATA AND CODE AVAILABILITY
All data and codes are available on the Odde laboratory website (oddelab.umn.edu) or on request from the corresponding author. Fastq files and the Cuffnorm output were deposited at Gene Expression Omnibus (GSE161154).
AUTHORS CONTRIBUTION
GAS, BLK, BRT, SSR, DAL and DJO contributed to study initiation, conception and design.
GAS, CJL, BCB and DJO contributed to writing the manuscript GAS, BCB and JMF contributed to developing mouse tumors
GAS, SKR and ALS contributed to the analysis of transcriptomic data. GAS ran and analyzed the cell migration simulations
GAS and CJL contributed to the acquisition and analysis of glioma cell migration
GAS established tumor lines and performed traction force measurements
GAS, BCB and HBC contributed to imaging and analysis of histological sections
NG, PCR, DM and DJO contributed to the design and implementation of the Brownian Dynamics Tumor Simulator
All authors contributed to the revisions of the manuscript
CONFLICT OF INTEREST STATEMENTS
DAL is the co-founder and co-owner of several biotechnology companies including NeoClone Biotechnologies, Inc., Discovery Genomics, Inc. (recently acquired by immusoft, Inc.), B-MoGen Biotechnologies, Inc. (recently acquired by Bio-Techne Corporation), and Luminary Therapeutics, Inc. DAL holds equity in, serves as a Senior Scientific Advisor for and Board of Director member for Recombinetics, a genome editing company. DAL consults for Genentech, Inc., which is funding some of his research. The business of all these companies is unrelated to the contents of this manuscript.
Video S1. Related to Figure 3. GFP-positive NRAS and PDGF tumor cells migrating in tumor bearing brain slices over 10 hours. Green shows tumor cells and magenta shows blood vasculature. Scale bar: 100 µm.
Video S2. Related to Figure 4A. GFP-positive NRAS and PDGF primary tumor cells migrating in mouse organotypic brain slice over 16 hours. Green shows tumor cells and magenta shows blood vasculature. Scale bar: 100 µm.
Video S3. Related to Figure 4D. Proneural and mesenchymal human PDX GBM cells migrating in mouse organotypic brain slice over 16 hours. Green shows tumor cells stained using green DiO membrane dye and magenta shows blood vasculature. Scale bar: 100 µm.
Video S4. Related to Figure 8. Brownian dynamics tumor simulator output showing simulated tumor growth.
Table S1. Related to Figure 1. FPKM expression of mouse tumors and healthy brain tissues.
Table S2. Related to Figure 1C. List of genes identified in each cluster (MC1, MC2, MC3, HC1, HC2, HC4).
Table S3. Related to Figure 2. List of cell migration simulator parameter values.
Table S4. Related to Figure 4. Characteristics of patient-derived xenograft (PDX) lines used in this study.
Table S5. Related to Figure 8. List of Brownian dynamics tumor simulator parameter values.
Table S6. List of antibodies and reagents used in IHC staining.
ACKNOWLEDGMENTS
The authors would like to thank Drs. Chris Wilke and Clark C. Chen for helpful discussion. This work was supported by National Institutes of Health grant U54 CA210190 to SSR, DAL and DJO and U54CA210180 to JNS. DAL acknowledges the American Cancer Society Research Professor grant, the John and Jean Hedberg Brain Tumor Fund, and the Children’s Cancer Research Fund. The authors acknowledge the Minnesota Supercomputing Institute (MSI) and University of Minnesota Genomic Center at the University of Minnesota for providing resources that contributed to the research results reported within this paper. We also acknowledge the Comparative Pathology, Cancer Bioinformatics, and Cytogenomics Shared Resources at the Masonic Cancer Center at the University of Minnesota for services.
Footnotes
↵+ co-senior authors