Skip to main content
NCBI home page
The Yale Journal of Biology and Medicine logoLink to The Yale Journal of Biology and Medicine
. 2025 Mar 31;98(1):3–19. doi: 10.59249/QTVT7676

ROCK2 Downregulation in Pediatric Medulloblastoma Increases Migration and Predicts the Involvement of SHH Non-canonical Signaling

Rodrigo Guedes Hakime a,b, Luís Fernando Peinado Nagano c, María Sol Brassesco b,*
PMCID: PMC11899262  PMID: 40165809

Abstract

The participation of the Rho-associated protein kinases (ROCK1 and 2) in the regulation of actin cytoskeleton organization, cell adhesion, motility, and gene expression has been extensively investigated in many tumors of different histology. However, their pathogenic roles in medulloblastoma (MB) remain understudied, demanding a deeper appreciation of their participation in cancer cell dissemination and tumor progression. Herein, we show that ROCK2 is downregulated in MB tumor samples and functionally increases migration of cell lines belonging to the SHH subgroup. A comprehensive comparative bioinformatic scrutiny of differentially expressed genes within a list of ROCK2 candidate substrates, uncovered a network of 21 dysregulated genes from which DYPSL3 (dihydropyrimidinase-related protein 3) denoted a strong positive correlation. Enrichment analysis revealed SHH/RHOA/ROCK2/DYPSL3 as top hub genes and the intersection between two biological processes of most importance in MB: actin cytoskeleton remodeling and neuron development. Of note, evidence shows that both ROCK2 and DYPSL3, interact with RHOA and in many tumor types they act as tumor suppressors, mitigating cell spreading. Alternatively, their impaired activity leads to undifferentiated phenotypes and inappropriate cytoskeletal dynamics affecting cell shape, attachment to the extracellular matrix, and cell movement. In parallel, cell motility is considered a prototypical non-canonical response to SHH mediated by RHOA. Therefore, we propose a model in which the interplay between these pathways may lead to a perturbation of proper cytoskeletal dynamics that underpins cell migration.

Keywords: medulloblastoma, SHH, ROCK2, DYPSL3, migration

Introduction

The molecular classification of medulloblastoma (MB) [1] has opened horizons for directed therapies against specific genes or pathways dysregulated in each subgroup [2], as is the case of the smoothened (SMO) inhibitors vismodegib or sonidegib [3,4] under clinical trials for patients diagnosed with the SHH (sonic hedgehog) variant. However, even with the emergence of subgroup-directed therapies, the poor survival rates observed for younger children (> 3 years old), those with molecular markers that connote poor prognosis, along with the vexing toxicities associated with current treatment make the better understanding of the basic biology of this tumor a continuous subject of research. The inhibition of factors involved in the control of cell migration, for example, could decrease tumor metastasis, and consequently improve high risk MB treatment.

Altered expression or activity of the actin cytoskeleton remodelers ROCK1 and ROCK2 (Rho-associated kinase 1 and 2) have frequently been related to more invasive and metastatic phenotypes [5]. However, evidence shows that these kinases might also function as negative regulators of cancer cell migration in a context-defined manner [6,7].

Moreover, a wider scope of ROCK proteins in the orchestration of other fundamental cellular processes has been described, mainly as a result of dramatic alterations in global gene transcription after shRNA-knockdown or their pharmacological inhibition [8-14]. Such involvement in different cell signaling pathways is also supported by more than 140 ROCK-substrate candidate proteins previously identified through a kinase-interacting screening performed by Amano et al. [15].

Yet, data about ROCK kinases involvement in MB progression within the pediatric setting is scarce. Therefore, the present study aimed to verify the expression patterns of ROCK1 and ROCK2 in tumor samples, the functional consequences of their inhibition in vitro, as well as to investigate possible molecular interactions and associations with tumor progression through in silico analysis.

Methods

Tumor samples. Exclusively pediatric MB samples (n=24) were collected by the Pediatric Neurosurgery service from our institution’s Clinical Hospital (Ribeirão Preto Medical School—University of São Paulo, Brazil). All samples were obtained after the patient’s or their guardian’s consent. Five pediatric (<13 years old) samples of non-neoplastic cerebellum were used as controls. All procedures were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments. This research was submitted to and approved by the Institutional Research Ethics Committee (CAAE n° 45984615.3.0000.5407).

Cell lines. The pediatric MB cell lines DAOY and ONS76 (purchased from the American Type Culture Collection (ATCC) and Banco de Células do Rio de Janeiro, respectively) and UW402 and UW473 (kindly provided by Dr. Michael Bobola) were grown in an incubator under standard culture conditions (at 37 °C in 5% CO2). The four cell lines belong to the SHH molecular MB subgroup [16]. Cell line authentication was performed through short tandem repeat analysis prior to any experiment.

Drug and treatments. Hydroxifasudil (pan-ROCK inhibitor) and SR3677 (ROCK2 inhibitor) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Treatment concentrations consisted of 2, 4, and 8 µM Hydroxifasudil and 25, 50, and 100 nM SR3677. For all experiments, the drugs were added on the culture medium immediately before application to cells and the corresponding control cultures received equal volumes of the solvent dimethyl sulfoxide (DMSO).

In vitro assays. Cell growth was evaluated through the XTT® assay (Sigma, St. Louis, MO, USA) as previously described in Souza et al. [5]. Wound healing assays were performed according to Vieira et al. [17] and the migration rates were calculated as the distance traveled by the cells in this area over time. All tests were performed in quadruplicate in three independent experiments and data shown as mean ± standard deviation.

RNA extraction and qRT-PCR. Total RNA was extracted from tumor samples using the Trizol® Reagent (Invitrogen Inc, Carlsbad, CA, USA) and cDNA samples were synthesized using the High Capacity® kit (Applied Biosystems®, Foster City, CA, USA. qRT-PCR was Taqman®-based using ROCK1 and ROCK2 probes (Hs01127699-m1 and Hs00178154-m1, respectively) on a 7500 Real Time PCR System (Applied Biosystems, Waltham, MA, USA). The GUSB (glucuronidase beta) and HPRT (hypoxanthine guanine phosphoribosyl transferase) genes were used as endogenous controls and the ONS-76 cell line was used as calibrator. The relative quantification of gene expression was determined using the Livak & Schmittgen 2-ΔΔCT method [18]. The samples were classified according to Cruzeiro et al. [19].

In silico analysis. The in silico analyses were performed using the Web-based genomics analysis and visualization platform R2 (Genomics Analysis and Visualization Platform - http://r2.amc.nl). To compare gene expression levels between pediatric medulloblastoma samples and non-neoplastic cerebellum samples, we used the Tumor Medulloblastoma Ependymoma-denBoer-51-MAS5.0-u133p2 (GEO ID: GSE74195) generated by de Bont et al. [20], which contains 27 MB pediatric samples and five samples of normal cerebellum tissue as controls.

To compare gene expression levels between MB molecular subgroups, other three different datasets were used: 1) Tumor Medulloblastoma-763-rma_sketch-hugene11t (GEO ID: GSE85217) generated by Cavalli et al. [21], which comprises 223 Shh-MB samples, 70 Wnt-MB samples, 144 Group3-MB samples, and 326 Group4-MB samples; 2) Tumor Medulloblastoma-Pfister-223-MAS5.0-u133p2 generated by Northcott et al. [22] containing 59 Shh-MB samples, 17 Wnt-MB samples, 56 Group3-MB samples, and 91 Group4-MB samples; and 3) Tumor Medulloblastoma-Pfister-167-fpkm-mb500rs1 also generated by Northcott et al. [22], composed of 46 Shh-MB samples, 16 Wnt-MB samples, 41 Group 3 MB samples, and 64 Group 4 MB samples. At all times, information on gene expression from adult tumor samples was excluded for our analysis.

For gene expression changes after ROCK2 inhibition we used microarray datasets generated by Boerma et al. [10] (GEO ID: GSE8686), who used the SLx-2119 ROCK2 inhibitor on primary cultures of human microvascular endothelial cells (HMVEC), human pulmonary artery smooth muscle cells (PASMC), and normal human dermal fibroblasts (NHDF). Similar analysis after ROCK2 overexpression was performed through the dataset generated by Zhou et al. [14] (GEO ID: GSE122145), using the human hepatocellular carcinoma cell line MHCC-97H.

Gene expression heatmaps were built on http://www2.heatmapper.ca/expression/. Protein-protein interactions and gene ontology (GO) enrichment analysis were performed using the public web server STRING v11.5 (available at https://string-db.org/) applying significance threshold of 0.05. Data were further plotted as chord and bubble graphs with the help of the SRplot online platform for data analysis and visualization (available at http://www.bioinformatics.com.cn/srplot).

Statistical Analysis

Mann Whitney’s nonparametric tests were used to analyze the dynamics of gene expression, with the aid of the software Graph Prism 5.0 (GraphPad Software, San Diego, CA, USA), and SPSS 21.0 (SPSS Inc. Chicago, IL, USA), considering the significance level of α = 0.05 for all tests performed. Correlations were evaluated by Spearman tests.

Results

ROCK2 Levels are Significantly Reduced in Pediatric Medulloblastoma Tumor Samples

Relative gene expression of ROCK1 and ROCK2 was analyzed through TaqMan based qRT-PCR in 24 pediatric MB clinical samples. As seen in Figure 1a, ROCK1 expression levels did not vary between groups (p<0.05). Conversely, a significant decrease in ROCK2 expression (p = 0.0019) was observed in tumor samples when compared to the non-neoplastic controls. These results were further validated in an independent cohort by in silico analysis using gene expression data from the work of de Bont et al. [20] (Figure 1b).

Figure 1.

Figure 1

Open in a new tab

a) Medulloblastoma gene expression profiles of ROCK1 and ROCK2 in pediatric tumor samples and the control group of non-neoplastic tissue; b) Validation of ROCK2 hypoexpression in a second cohort from data retrieved from the dataset Tumor Medulloblastoma Ependymoma-denBoer-51-MAS5.0-u133p2 (GEO ID: GSE74195) generated by de Bont et al. [20] accessed through the R2: Genomics Analysis and Visualization Platform (http://r2.amc.nl); c) Dependency data on seven MB cell lines imported from the DepMap consortium (CRISPR (DepMap 22Q2 Public+Score, Chronos; https://depmap.org/portal/). The data was plotted on a histogram where it is possible to see the vulnerability of each cell lines to ROCK1 and ROCK2 knockdown. Scores under zero denote gene dependency. Note that cell lines belonging to the SHH subgroup are not dependent on ROCK2 depletion; d) Treatment of cell lines with ROCK2 pharmacological inhibitors increases migration as detected by wound healing assays. The asterisks (*) represent p<0.05.

ROCK2 Inhibition Improves MB Migration In Vitro

To further validate the role of ROCK2 in MB, in parallel to gene expression assays, four different cell lines belonging to the SHH subgroup were treated with increasing concentrations of the panROCK inhibitor hydroxifasudil, and the ROCK2 selective inhibitor SR3677. Our results showed that cell viability was not altered, irrespective of the inhibitor, dose, and time of exposure (Appendix A: Supplementary Figure 1). In accordance, in silico analysis of MB vulnerability against ROCK2 depletion based on CRISPR (accessed through the Cancer Dependency Map portal - http://depmap.org) showed that the viability of four out of seven MB cell lines does not depend on ROCK2 expression (Figure 1c). Of note, the non-dependent cell lines belong to the SHH and 3-4 subgroups [23] Alternatively, the platform showed that ROCK1 knockdown, is more important for MB cell lines’ survival, albeit, in our system hydroxifasudil which has identical IC50 for both isoforms, was ineffective as well.

Nevertheless, migration under treatment with both inhibitors was increased, with more expressive results for UW473 and UW402 cells under hydroxifasudil treatment, in which the rate of gap closure was enhanced in the range of 30 to 50% (Figure 1d).

Potential ROCK2 Substrates are Differentially Expressed in MB

Our next step was to find out which genes may be affected by ROCK2 hypoexpression. Thus, we performed an initial screening considering the 68 candidate substrates for ROCK2 that were detected twice or more times in independent analyses by Amano et al. [15] through their kinase-interacting substrate screening (KISS) method. By using once again the dataset generated by de Bont et al. [20], we were able to identify 34 genes—besides ROCK2 itself—that are differentially expressed (DEGs) in MB samples when compared to control tissues. The results are presented in a heatmap with the expression data on individual samples (Figure 2a). Further clustering through a correlation heatmap allowed the identification of a potential regulation network in MB, formed by ROCK2 and 21 out of the 34 DEGs (Figure 2b).

Figure 2.

Figure 2

Open in a new tab

a) Gene expression heatmap the 34 ROCK2 candidate substrates that were found differentially expressed by using once again the dataset generated by de Bont et al. [20], b) Correlation heatmap (k=3) of ROCK2 and the other 34 DEGs MB samples. The clustering establishes a potential ROCK2 regulation network in this tumor type. ROCK2 is highlighted in red among those genes.

Modulation of ROCK2 Leads to Altered DYPSL3 Expression

Further, to confirm whether the expression of these genes is altered by ROCK2, we analyzed two public datasets that modulated ROCK2 expression in different in vitro models. Boerma et al. [10] (GEO ID: GSE8686) dataset was used to verify the effect of the newly developed ROCK2 inhibitor SLx-2119 on genome-wide gene expression in HMVEC and PASMC cells. Alternatively, to contrast these results we analyzed how ROCK2 overexpression affects the expression levels of the genes contained in the potential regulatory network proposed in this work. For this purpose, we used the transcriptional profiling of the human hepatocellular carcinoma cell line MHCC-97H before and after ROCK2 transfection (GSE122145).

In ROCK2-depleted HMVEC cells, we were able to identify a significant decrease in the expression levels of the ADD3, DPYSL2, DPYSL3, and MARCS genes. Likewise, the inhibition of ROCK2 in PASMC lead to a reduction in the expression of DPYSL2, DPYSL3, DPYSL4, and SPTBN2. In turn, forced expression of ROCK2 in MHCC-97H was accompanied by an increase in the expression of DPYSL3, GOLGA3 and LIN7A. Through a Venn diagram, it is possible to observe that the expression of DPYSL3 was the only one that changed upon ROCK2 modulation in the three experimental settings (Figure 3a). Note that DPYSL3 expression levels are congruent with ROCK2 inhibition or overexpression, suggesting a positive correlation between genes (Figure 3b), what was confirmed through data analysis from three different datasets (p<0.05) (Figure 3c). Moreover, the three datasets showed reduced levels of both genes when expression in samples belonging to the SHH subgroup were compared to non-SHH variants (Figure 3d). Expression data on ROCK2 in our tumor samples showed a similar pattern (Appendix A: Supplementary Figure 2).

Figure 3.

Figure 3

Open in a new tab

a) Venn diagram, showing the number of genes among the 21 DEGs whose expression is altered after ROCK2 overexpression (OE) (GSE122145 MHCC-97H) or ROCK2 inhibition by SLx2119 (GSE8686 HMVEC; GSE8686 PASMC); b) Relative expression of DPYSL3 after ROCK2 modulation in both experimental settings; c) ROCK2 and DPYSL3 are positively correlated in data retrieved from three different datasets from R2 genomics platform: GSE85217, Pfister-167 and Pfister-223; d) Both genes show lower expression levels in SHH-MB when compared to non-SHH variants. For proper analysis in c and d, data from adult samples were excluded.

Enrichment Analysis of ROCK2 Candidate Substrates Denotes a Migration-associated Network Mainly Involved with the Actin Cytoskeleton

To further investigate protein-protein interactions between ROCK2 and its 21 candidate substrates and the biological processes in which they might participate the STRING v11.5 software was used considering curated databases, experimentally determined relationships, gene neighborhood, gene co-occurrence, text mining, co-expression, and protein homology. As shown in Figure 4a, the 24 nodes (SHH and RHOA were also included) generated 26 edges with and the average nodal degree was 2.17. Of note, SHH/RHOA/ROCK2/DYPSL3 represented top hub genes and the intersection between two biological processes of most importance in MB: actin cytoskeleton remodeling and neuron development. Indeed, GO enrichment analysis revealed that the genes are mainly related to actin filament bundle assembly (GO:0051017 - p=1.17e-06), actin crosslink formation (GO:0051764 - p=0.0296), regulation of actin filament organization (GO:0110053 - p=0.0086), cytoskeleton organization (GO:0007010 - p=0.00013), actin cytoskeleton organization (GO:0030036 - p=0.00036), actin filament capping (GO:0051693 - p=0.00036) and GO:0048666- Neuron development (p=0.00061). The number of connections between each biological process and individual genes in the network and gene counts within each enrichment category has shown in the form of a chord and bubble plots, respectively (Figure 4b and 4c).

Figure 4.

Figure 4

Open in a new tab

ROCK2 candidate substrates form a regulation network mainly involved with actin cytoskeleton remodeling and neuron development. a) Protein-protein interactions were accessed through the software STRING v11.5. The parameters evaluated were curated databases, experimentally determined, gene neighborhood, gene co-occurrence, text mining, co-expression, and protein homology. The interaction score was 0.400. The enrichment analysis was also performed on the same software. Network edges denote confidence and disconnected nodes were omitted. The two biological processes highlighted, neuron development and actin cytoskeleton organization, denote RHOA, ROCK2 and DPYSL3 as hub genes; b) Chord diagram generated through SRplot platform showing the enrichment and amount of connections between each biological process and individual genes in the network; c) GO enrichment bubble plot featuring gene count with each category: GO:0051764 - Actin crosslink formation (p=0.0296); GO:0048666 - Neuron development (p=0.00061); GO:0110053 - Regulation of actin filament organization (p=0.0086); GO:0007010 - Cytoskeleton organization (p=0.00013); GO:0030036 - Actin cytoskeleton organization (p=0.00036); GO:0051693 - Actin filament capping (p=0.00036); GO:0051017 - Actin filament bundle assembly (p=1.17e-06).

Discussion

ROCK isoforms are reported to share 65% overall identity and 92% identity in kinase domain, despite this, in the brain, ROCK1 is more expressed in glia, whereas ROCK2 is strongly expressed in neurons. Moreover, while it has been assumed that they functionally mediate overlapping roles, in varied tumor types many discrepancies or even opposite roles on adhesion, migration, proliferation, and gene expression have been described [24,25].

In MB, the cellular, physiological, and pathogenic roles of these kinases are not clear. Previously, by investigating public datasets, we described inconsistencies [5]. Herein, we show that ROCK2, but not ROCK1, is differently expressed in MB tumor samples. This isoform does not seem to be essential for SHH-MB viability as demonstrated through gene dependency assays. However, to further investigate the functional effects of interfering with ROCK2, in vitro assays were performed in the presence of two different pharmacological inhibitors. The results showed that, as proliferation was unaffected, both, Hydroxifasudil and SR3677 treatments lead to increased migration. These drugs and other inhibitors such as Y-33075 and Y-27632 have shown similar results in other tumor types [17,26,27]. Likewise changes in motility after ROCK2 knockdown have already been reported in glioblastoma stem cells [28], and causes a dramatic change in the localization of vinculin, phosphorylated caveolin-1 and tyrosine-phosphorylated proteins in colon cancer suggesting that alternative roles in regulating cytoskeletal plasticity and cell adhesion to extracellular matrix [29].

The driving force of cell movement, provided mostly by the reorganization of the cytoskeleton, is indispensable for cell migration. Cytoskeleton remodeling is triggered by the activation of members of the Rho family of GTPases, such as RhoA, Rac1, and [30]. Of note, ROCK2 knockdown in glioblastoma cells resulted in increased migration with increased phosphorylation of Cdc42/Rac. Moreover, the authors observed decreased RhoA protein suggesting the presence of a feedback loop of ROCK expression on this GTPase [31].

Novel ROCK substrates are discovered constantly. To facilitate the understanding of how ROCK2 underexpression could be regulating migration, we investigated differentially expressed genes within the list of 68 candidate substrates described by Amano et al. [15], in a MB cohort. From those, 34 appeared dysregulated in this tumor, and 21 denoted a possible ROCK2 regulation network associated mostly with neuron development and actin remodeling. Moreover, the analysis of DEGs after ROCK2 modulation in different experimental settings, revealed a strong positive correlation with DPYSL3 (Dihydropyrimidinase-related protein 3) a member of the DPYSL gene family (Dpysl1-5) that encodes five cytosolic phospho-proteins involved in semaphorin/collapsin-induced cellular events [32]. Also called CRMP4 (collapsin response mediator protein 4), this protein is highly expressed in the developing and adult nervous system [33-35] and functions in a variety of cellular processes such as differentiation, neurite extension, and axonal regeneration [36-39]. More recently, a role as a regulatory molecule of the cellular adhesion complex has been confirmed, thus, DPYSL3 also regulates cell motility [40].

Indeed, DPYSL3 has been considered a metastasis suppressor in some tumor types [35,38,39,41,42]. For instance, as with ROCK2, its knockdown did not affect proliferation of lung cancer cells but induced clear morphology shifts into spindle and fibroblast-like shapes, increased expression of epithelial-to-mesenquimal transition (EMT) markers, such as TWIST1 and N-cadherin and significantly influenced migration and invasion in vitro. Moreover, inhibition of DPYSL3 promoted the progression of metastasis in C57BL/6 mice, a point that was validated when comparing its expression levels between samples from patients with stage I and stage IV lung cancer. Advanced tumors presented 2X lower expression at both mRNA and protein levels [43].

In neuroblastoma similar results were observed after DPYSL3 knockdown, with enhanced motility and increased expression of EMT markers, pointing to a role as a multifunctional signaling modulator. Moreover, the enhanced cell migration was accompanied by a disturbance of rib-like actin-structures in lamellipodia [44].

Of note, it was also shown that the dominant and shorter transcript of DPYSL3 (CRMP4a) directly binds to F-actin in modulating cytoskeletal organization, focal adhesion, and cell spreading, while physically and functionally interacts with RhoA to mediate neurite outgrowth inhibition, being the only member of the CRMP family proteins to interact with this GTPase protein in normal neural cells [44-46]. Moreover, in prostate cancer cells, Li et al., [47] demonstrated that CRMP4a sequesters RhoA, resulting in suppression of cytoskeletal organization, cell migration, and spreading, therefore in its absence, cell motility could be enhanced [47].

Even though the sequestration of RhoA by CRMP4a is not clear, CRMP proteins form multimers [48]. In this regard, CRMP2 has been reported to bind CRMP4 and CRMP5. CRMP2 is a key regulator of microtubule assembly and its phosphorylation inhibits axonal guidance [46]. ROCK2 has shown to phosphorylate CRMP2 at Thr-555 and thereby reduces its function in modulating neural cell motility and cancer cell migration [49,50]. Moreover, CRMP2 has shown to counteract with RhoA-induced neurite retraction in neuroblastoma, indicating that these proteins can inhibit RhoA action on neuronal morphology [51]. Likewise, is has been shown that CRMP4 can mediate structural interactions between the actin and tubulin cytoskeleton in neural cells coordinating guided movements of cells [46].

The importance of the Rho GTPases in regulating cell motility and axon guidance is well recognized. Indeed, many lines of evidence indicate that while axon attraction is mediated primarily by the Cdc42 and Rac, in neuronal cell lines activation of RhoA stimulates actinomysin contractility and stress fiber formation leading to cone growth collapse. Thus, by modulating the activation of different GTPases, axon guidance cues can determine whether a growth cone is attracted or repelled. Of note, among the different guidance receptors that regulate the Rho GTPases to effect changes in motility and migration promotion is SMO [52].

This is particularly interesting since mutations in SMO, together with PTCH1 and SUFU (all mutually exclusive) are the main mutations substantiating the tumor-driving role of the SHH pathway [53], defining the specific subgroup to which the four MB cell lines we used in in vitro assays are included.

Cell motility is a prototypical non-canonical response to SHH. Particularly, SHH/SMO-regulated Rho-dependent actin cytoskeleton affects the migration of neuronal and neuronal precursor cells and may play a similarly important role to that of GLI canonical activation [54-59].

Moreover, it seems that this Rho-to-actin pathway through the non-canonical activation of the SHH pathway is coupled by the heterotrimeric G protein Gi [52] whose activation also potentiates canonical signaling (GLI-dependent) [60]. Thus, an interplay between the SHH canonical and non-canonical pathways in the MB cell lines (Figure 5) allied to the dysregulation of other 20 actin-associated proteins (involved in capping, crosslinking, severing, and bundling) predicts a migration-associated network that may contribute to the cell dispersal observed herein through in vitro assays.

Figure 5.

Figure 5

Open in a new tab

ROCK2 downregulation mediates a migration-associated network in SHH-medulloblastoma leading to inappropriate cytoskeleton dynamics and motility through the loss of adhesion. Increased migration could also be influenced by an interplay with the non-canonical activation of the neurogenic pathway through SMO/RHOA. The figure was composed with the aid of illustrations from the SMART-servier Medical Art available at https://smart.servier.com/

It is well known that changes to both, the actin and microtubule cytoskeletons, are essential to neuron migration [61]. Of note, MB metastasis occurs almost exclusively by the spreading of cells through the cerebrospinal fluid to the spinal and intracranial leptomeninges, a process called leptomeningeal (LM) dissemination, which is virtually responsible for 100% of MB-associated deaths, and is primarily based on cell programs that mediate motility, invasiveness, and maintain tumor cells in a stem-like state [62].

Indeed, cells with altered cytoskeleton network display weaker mechanical properties (less cell stiffness and more elasticity) and higher cancer stemness, which is associated with frequent membrane protrusions and production of extracellular vesicles [63,64]. Moreover, defects in cellular morphogenesis not only contribute to tissue disruption and the acquisition of inappropriate migratory features but also with altered membrane trafficking, nuclear deformation, chromosome segregation defects (increasing genomic instability), and chemoresistance (to mitotic poisons, for instance) [65,66].

Also, two distinct cancer cell phenotypes have been detected in human LM metastasis, adherent and rounded-floating cells within the cerebrospinal fluid. The second group has shown to colonize mouse leptomeninges more rapidly and to be associated with shortened survival [67]. Of note, it has been demonstrated that mice models with constitutively activated SMO develop MB characterized by loss of neuronal differentiation, tumor cells with abnormal shapes and sizes, and indeed display LM tumor spread [68].

Moreover, there are reports of increased expression of the core transcription factors SOX2, NANOG, OCT3/4, and KLF4 in cases of SHH and SMO activation, all of which are essential for maintaining the stemness and self-renewal [69]. The contribution of ROCK2 downregulation in sustaining an undifferentiated state of cancer cells is supported by the ability of the pan-ROCK inhibitor Y-27632 to facilitate the growth of stem cells and increase self-renewal potential in vitro [70-73].

To date, most of the efforts to prevent SHH signaling have been directed on the development of SMO inhibitors. However, clinical trials have shown some limitations including early drug resistance by the acquisition of new missense mutations, the ineffectiveness for cases with aberrations in downstream effectors of the SHH signaling pathway, and premature growth plate fusion for young patients [3,4]. It is also likely that treatments have failed because alternative oncogenic events could be responsible for non-canonical SHH pathway activation [74]. Even though the relation between ROCK2/DYPSL3 and the non-canonical activation of SHH/RHOA with cytoskeletal dysfunction can only be hypothetical in our model, our data predicts an understudied pathway that deserves further investigation. Additional in vitro and in vivo preclinical models will be needed to scrutinize the occurrence of these intermolecular interactions to verify their potential to be exploited to treat MB or prevent its progression.

Acknowledgments

This work was supported by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo, Brazil): Grant given to MSB (2014/03877-3) and fellowship given to RGH (2014/19790-4).

Glossary

DEG

differentially expressed gene

HMVEC

human microvascular endothelial cells

LM

leptomeningeal

MB

medulloblastoma

NHDF

normal human dermal fibroblasts

PASMC

pulmonary artery smooth muscle cells

qRT-PCR

real-time quantitative reverse transcription polymerase chain reaction

Appendix A.

Author Contributions

Conceptualization: RGH, LFN, and MSB; Investigation and validation: RGH, LFN; Data curation and analysis: RGH, LFN; Writing-original draft preparation: RGH, LFN, and MSB; Writing-review and editing: MSB; Supervision: MSB; Project administration: MSB. MSB is supported by the National Council for Scientific and Technological Development (CNPq) (research productivity fellowship nº 303593/2022-9. All authors read and agreed to the publication of the manuscript.

References

  1. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella-Branger D, et al. The 2021 WHO classification of tumors of the central nervous system: A summary. Neuro-oncol. 2021. Aug;23(8):1231–51. 10.1093/neuonc/noab106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Milde T, Lodrini M, Savelyeva L, Korshunov A, Kool M, Brueckner LM, et al. HD-MB03 is a novel Group 3 medulloblastoma model demonstrating sensitivity to histone deacetylase inhibitor treatment. J Neurooncol. 2012. Dec;110(3):335–48. 10.1007/s11060-012-0978-1 [DOI] [PubMed] [Google Scholar]
  3. Robinson GW, Orr BA, Wu G, Gururangan S, Lin T, Qaddoumi I, et al. Vismodegib exerts targeted efficacy against recurrent sonic hedgehog - Subgroup medulloblastoma: Results from phase II Pediatric Brain Tumor Consortium studies PBTC-025B and PBTC-032. J Clin Oncol. 2015. Aug;33(24):2646–54. 10.1200/JCO.2014.60.1591 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Kieran MW, Chisholm J, Casanova M, Brandes AA, Aerts I, Bouffet E, et al. Phase I study of oral sonidegib (LDE225) in pediatric brain and solid tumors and a phase II study in children and adults with relapsed medulloblastoma. Neuro-oncol. 2017. Oct;19(11):1542–52. 10.1093/neuonc/nox109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. de Sousa GR, Vieira GM, das Chagas PF, Pezuk JA, Brassesco MS. Should we keep rocking? Portraits from targeting Rho kinases in cancer. Pharmacol Res. 2020. Oct;160:105093. 10.1016/j.phrs.2020.105093 [DOI] [PubMed] [Google Scholar]
  6. Peng L, Sang H, Wei S, Li Y, Jin D, Zhu X, et al. circCUL2 regulates gastric cancer malignant transformation and cisplatin resistance by modulating autophagy activation via miR-142-3p/ROCK2. Mol Cancer. 2020. Nov;19(1):156. 10.1186/s12943-020-01270-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bhandary L, Whipple RA, Vitolo MI, Charpentier MS, Boggs AE, Chakrabarti KR, et al. ROCK inhibition promotes microtentacles that enhance reattachment of breast cancer cells. Oncotarget. 2015. Mar;6(8):6251–66. 10.18632/oncotarget.3360 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Spencer C, Montalvo J, McLaughlin SR, Bryan BA. Small molecule inhibition of cytoskeletal dynamics in melanoma tumors results in altered transcriptional expression patterns of key genes involved in tumor initiation and progression. Cancer Genomics Proteomics. 2011;8(2):77–85. [PMC free article] [PubMed] [Google Scholar]
  9. Berenjeno IM, Bustelo XR. Identification of the Rock-dependent transcriptome in rodent fibroblasts. Clin Transl Oncol. 2008. Nov;10(11):726–38. 10.1007/s12094-008-0279-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Boerma M, Fu Q, Wang J, Loose DS, Bartolozzi A, Ellis JL, et al. Comparative gene expression profiling in three primary human cell lines after treatment with a novel inhibitor of Rho kinase or atorvastatin. Blood Coagul Fibrinolysis. 2008. Oct;19(7):709–18. 10.1097/MBC.0b013e32830b2891 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Harvey SA, Anderson SC, SundarRaj N. Downstream effects of ROCK signaling in cultured human corneal stromal cells: microarray analysis of gene expression. Invest Ophthalmol Vis Sci. 2004. Jul;45(7):2168–76. 10.1167/iovs.03-1218 [DOI] [PubMed] [Google Scholar]
  12. Stiles JM, Kurisetty V, Mitchell DC, Bryan BA. Rho kinase proteins regulate global miRNA expression in endothelial cells. Cancer Genomics Proteomics. 2013;10(6):251–63. [PubMed] [Google Scholar]
  13. Montalvo J, Spencer C, Hackathorn A, Masterjohn K, Perkins A, Doty C, et al. ROCK1 & 2 perform overlapping and unique roles in angiogenesis and angiosarcoma tumor progression. Curr Mol Med. 2013. Jan;13(1):205–19. 10.2174/156652413804486296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Zhou HM, Zhang JG, Zhang X, Fan GR, Liu GL, Li Q. Overlapping and unique roles played by ROCK1 and 2 in the modulation of coding and long noncoding RNA expression. BMC Genomics. 2019. May;20(1):409. 10.1186/s12864-019-5715-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Amano M, Hamaguchi T, Shohag MH, Kozawa K, Kato K, Zhang X, et al. Kinase-interacting substrate screening is a novel method to identify kinase substrates. J Cell Biol. 2015. Jun;209(6):895–912. 10.1083/jcb.201412008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Silva MO, de Sousa GR, Simões SC, Nicolucci P, Tamashiro E, Saggioro F, et al. Perillyl alcohol for pediatric TP53- and RAS-mutated SHH-medulloblastoma: an in vitro and in vivo translational pre-clinical study. Childs Nerv Syst. 2021. Jul;37(7):2163–75. 10.1007/s00381-021-05115-w [DOI] [PubMed] [Google Scholar]
  17. Vieira GM, Roberto GM, Lira RC, Engel EE, Tone LG, Brassesco MS. Prognostic value and functional role of ROCK2 in pediatric Ewing sarcoma. Oncol Lett. 2018. Feb;15(2):2296–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001. Dec;25(4):402–8. 10.1006/meth.2001.1262 [DOI] [PubMed] [Google Scholar]
  19. Cruzeiro GA, Salomão KB, de Biagi CA Jr, Baumgartner M, Sturm D, Lira RC, et al. A simplified approach using Taqman low-density array for medulloblastoma subgrouping. Acta Neuropathol Commun. 2019. Mar;7(1):33. 10.1186/s40478-019-0681-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. de Bont JM, Packer RJ, Michiels EM, den Boer ML, Pieters R. Biological background of pediatric medulloblastoma and ependymoma: a review from a translational research perspective. Neuro-oncol. 2008. Dec;10(6):1040–60. 10.1215/15228517-2008-059 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Cavalli FM, Remke M, Rampasek L, Peacock J, Shih DJ, Luu B, et al. Intertumoral Heterogeneity within Medulloblastoma Subgroups. Cancer Cell. 2017. Jun;31(6):737–754.e6. 10.1016/j.ccell.2017.05.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Northcott PA, Buchhalter I, Morrissy AS, Hovestadt V, Weischenfeldt J, Ehrenberger T, et al. The whole-genome landscape of medulloblastoma subtypes. Nature. 2017. Jul;547(7663):311–7. 10.1038/nature22973 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ivanov DP, Coyle B, Walker DA, Grabowska AM. In vitro models of medulloblastoma: choosing the right tool for the job. J Biotechnol. 2016. Oct;236:10–25. 10.1016/j.jbiotec.2016.07.028 [DOI] [PubMed] [Google Scholar]
  24. Iizuka M, Kimura K, Wang S, Kato K, Amano M, Kaibuchi K, et al. Distinct distribution and localization of Rho-kinase in mouse epithelial, muscle and neural tissues. Cell Struct Funct. 2012;37(2):155–75. 10.1247/csf.12018 [DOI] [PubMed] [Google Scholar]
  25. Yoneda A, Multhaupt HA, Couchman JR. The Rho kinases I and II regulate different aspects of myosin II activity. J Cell Biol. 2005. Aug;170(3):443–53. 10.1083/jcb.200412043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Bachtler N, Torres S, Ortiz C, Schierwagen R, Tyc O, Hieber C, et al. The non-selective Rho-kinase inhibitors Y-27632 and Y-33075 decrease contraction but increase migration in murine and human hepatic stellate cells. PLoS One. 2023. Jan;18(1):e0270288. 10.1371/journal.pone.0270288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Yang S, Kim HM. ROCK inhibition activates MCF-7 cells. PLoS One. 2014. Feb;9(2):e88489. 10.1371/journal.pone.0088489 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Wong SY, Ulrich TA, Deleyrolle LP, MacKay JL, Lin JM, Martuscello RT, et al. Constitutive activation of myosin-dependent contractility sensitizes glioma tumor-initiating cells to mechanical inputs and reduces tissue invasion. Cancer Res. 2015. Mar;75(6):1113–22. 10.1158/0008-5472.CAN-13-3426 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wei L, Surma M, Shi S, Lambert-Cheatham N, Shi J. Novel Insights into the Roles of Rho Kinase in Cancer. Arch Immunol Ther Exp (Warsz). 2016. Aug;64(4):259–78. 10.1007/s00005-015-0382-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Ananthakrishnan R, Ehrlicher A. The forces behind cell movement. Int J Biol Sci. 2007. Jun;3(5):303–17. 10.7150/ijbs.3.303 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mertsch S, Thanos S. Opposing signaling of ROCK1 and ROCK2 determines the switching of substrate specificity and the mode of migration of glioblastoma cells. Mol Neurobiol. 2014. Apr;49(2):900–15. 10.1007/s12035-013-8568-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Schmidt EF, Strittmatter SM. The CRMP family of proteins and their role in Sema3A signaling. Adv Exp Med Biol. 2007;600:1–11. 10.1007/978-0-387-70956-7_1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mukherjee J, DeSouza LV, Micallef J, Karim Z, Croul S, Siu KW, et al. Loss of collapsin response mediator Protein1, as detected by iTRAQ analysis, promotes invasion of human gliomas expressing mutant EGFRvIII. Cancer Res. 2009. Nov;69(22):8545–54. 10.1158/0008-5472.CAN-09-1778 [DOI] [PubMed] [Google Scholar]
  34. Wu CC, Chen HC, Chen SJ, Liu HP, Hsieh YY, Yu CJ, et al. Identification of collapsin response mediator protein-2 as a potential marker of colorectal carcinoma by comparative analysis of cancer cell secretomes. Proteomics. 2008. Jan;8(2):316–32. 10.1002/pmic.200700819 [DOI] [PubMed] [Google Scholar]
  35. Gao X, Li LY, Rassler J, Pang J, Chen MK, Liu WP, et al. Prospective Study of CRMP4 Promoter Methylation in Prostate Biopsies as a Predictor For Lymph Node Metastases. J Natl Cancer Inst. 2017. Jan;109(6):djw282. 10.1093/jnci/djw282 [DOI] [PubMed] [Google Scholar]
  36. Li LC, Okino ST, Zhao H, Pookot D, Place RF, Urakami S, et al. Small dsRNAs induce transcriptional activation in human cells. Proc Natl Acad Sci USA. 2006. Nov;103(46):17337–42. 10.1073/pnas.0607015103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang HF, Takenaka K, Nakanishi A, Miki Y. BRCA2 and nucleophosmin coregulate centrosome amplification and form a complex with the Rho effector kinase ROCK2. Cancer Res. 2011. Jan;71(1):68–77. 10.1158/0008-5472.CAN-10-0030 [DOI] [PubMed] [Google Scholar]
  38. Zhou W, Xie P, Pang M, Yang B, Fang Y, Shu T, et al. Upregulation of CRMP4, a new prostate cancer metastasis suppressor gene, inhibits tumor growth in a nude mouse intratibial injection model. Int J Oncol. 2015. Jan;46(1):290–8. 10.3892/ijo.2014.2705 [DOI] [PubMed] [Google Scholar]
  39. Li K, Pang J, Cheng H, Liu WP, Di JM, Xiao HJ, et al. Manipulation of prostate cancer metastasis by locus-specific modification of the CRMP4 promoter region using chimeric TALE DNA methyltransferase and demethylase. Oncotarget. 2015. Apr;6(12):10030–44. 10.18632/oncotarget.3192 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kawahara T, Hotta N, Ozawa Y, Kato S, Kano K, Yokoyama Y, et al. Quantitative proteomic profiling identifies DPYSL3 as pancreatic ductal adenocarcinoma-associated molecule that regulates cell adhesion and migration by stabilization of focal adhesion complex. PLoS One. 2013. Dec;8(12):e79654. 10.1371/journal.pone.0079654 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Gao X, Pang J, Li LY, Liu WP, Di JM, Sun QP, et al. Expression profiling identifies new function of collapsin response mediator protein 4 as a metastasis-suppressor in prostate cancer. Oncogene. 2010. Aug;29(32):4555–66. 10.1038/onc.2010.213 [DOI] [PubMed] [Google Scholar]
  42. Li C, Jiang W, Hu Q, Li LC, Dong L, Chen R, et al. Enhancing DPYSL3 gene expression via a promoter-targeted small activating RNA approach suppresses cancer cell motility and metastasis. Oncotarget. 2016. Apr;7(16):22893–910. 10.18632/oncotarget.8290 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Yang Y, Jiang Y, Xie D, Liu M, Song N, Zhu J, et al. Inhibition of cell-adhesion protein DPYSL3 promotes metastasis of lung cancer. Respir Res. 2018. Mar;19(1):41. 10.1186/s12931-018-0740-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Rosslenbroich V, Dai L, Baader SL, Noegel AA, Gieselmann V, Kappler J. Collapsin response mediator protein-4 regulates F-actin bundling. Exp Cell Res. 2005. Nov;310(2):434–44. 10.1016/j.yexcr.2005.08.005 [DOI] [PubMed] [Google Scholar]
  45. Alabed YZ, Pool M, Ong Tone S, Fournier AE. Identification of CRMP4 as a convergent regulator of axon outgrowth inhibition. J Neurosci. 2007. Feb;27(7):1702–11. 10.1523/JNEUROSCI.5055-06.2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Khazaei MR, Girouard MP, Alchini R, Ong Tone S, Shimada T, Bechstedt S, et al. Collapsin response mediator protein 4 regulates growth cone dynamics through the actin and microtubule cytoskeleton. J Biol Chem. 2014. Oct;289(43):30133–43. 10.1074/jbc.M114.570440 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Li C, Xu H, Xiao L, Zhu H, Zhang G, Wei W, et al. CRMP4a suppresses cell motility by sequestering RhoA activity in prostate cancer cells. Cancer Biol Ther. 2018;19(12):1193–203. 10.1080/15384047.2018.1491507 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wang LH, Strittmatter SM. Brain CRMP forms heterotetramers similar to liver dihydropyrimidinase. J Neurochem. 1997. Dec;69(6):2261–9. 10.1046/j.1471-4159.1997.69062261.x [DOI] [PubMed] [Google Scholar]
  49. Arimura N, Inagaki N, Chihara K, Ménager C, Nakamura N, Amano M, et al. Phosphorylation of collapsin response mediator protein-2 by Rho-kinase. Evidence for two separate signaling pathways for growth cone collapse. J Biol Chem. 2000. Aug;275(31):23973–80. 10.1074/jbc.M001032200 [DOI] [PubMed] [Google Scholar]
  50. Ricard D, Rogemond V, Charrier E, Aguera M, Bagnard D, Belin MF, et al. Isolation and expression pattern of human Unc-33-like phosphoprotein 6/collapsin response mediator protein 5 (Ulip6/CRMP5): coexistence with Ulip2/CRMP2 in Sema3a- sensitive oligodendrocytes. J Neurosci. 2001. Sep;21(18):7203–14. 10.1523/JNEUROSCI.21-18-07203.2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Leung T, Ng Y, Cheong A, Ng CH, Tan I, Hall C, et al. p80 ROKalpha binding protein is a novel splice variant of CRMP-1 which associates with CRMP-2 and modulates RhoA-induced neuronal morphology. FEBS Lett. 2002. Dec;532(3):445–9. 10.1016/S0014-5793(02)03736-5 [DOI] [PubMed] [Google Scholar]
  52. Polizio AH, Chinchilla P, Chen X, Kim S, Manning DR, Riobo NA. Heterotrimeric Gi proteins link Hedgehog signaling to activation of Rho small GTPases to promote fibroblast migration. J Biol Chem. 2011. Jun;286(22):19589–96. 10.1074/jbc.M110.197111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Kool M, Jones DT, Jäger N, Northcott PA, Pugh TJ, Hovestadt V, et al. ICGC PedBrain Tumor Project . Genome sequencing of SHH medulloblastoma predicts genotype-related response to smoothened inhibition. Cancer Cell. 2014. Mar;25(3):393–405. 10.1016/j.ccr.2014.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Polizio AH, Chinchilla P, Chen X, Manning DR, Riobo NA. Sonic Hedgehog activates the GTPases Rac1 and RhoA in a Gli-independent manner through coupling of smoothened to Gi proteins. Sci Signal. 2011. Nov;4(200):pt7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Kotulak-Chrząszcz A, Kmieć Z, Wierzbicki PM. Sonic Hedgehog signaling pathway in gynecological and genitourinary cancer (Review) [Review] [Review]. Int J Mol Med. 2021. Jun;47(6):106. 10.3892/ijmm.2021.4939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Kasai K, Takahashi M, Osumi N, Sinnarajah S, Takeo T, Ikeda H, et al. The G12 family of heterotrimeric G proteins and Rho GTPase mediate Sonic hedgehog signalling. Genes Cells. 2004. Jan;9(1):49–58. 10.1111/j.1356-9597.2004.00701.x [DOI] [PubMed] [Google Scholar]
  57. Carballo GB, Honorato JR, de Lopes GP, Spohr TC. A highlight on Sonic hedgehog pathway. Cell Commun Signal. 2018. Mar;16(1):11. 10.1186/s12964-018-0220-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Fukushima N, Weiner JA, Chun J. Lysophosphatidic acid (LPA) is a novel extracellular regulator of cortical neuroblast morphology. Dev Biol. 2000. Dec;228(1):6–18. 10.1006/dbio.2000.9930 [DOI] [PubMed] [Google Scholar]
  59. Chinchilla P, Xiao L, Kazanietz MG, Riobo NA. Hedgehog proteins activate pro-angiogenic responses in endothelial cells through non-canonical signaling pathways. Cell Cycle. 2010. Feb;9(3):570–9. 10.4161/cc.9.3.10591 [DOI] [PubMed] [Google Scholar]
  60. Ho Wei L, Arastoo M, Georgiou I, Manning DR, Riobo-Del Galdo NA. Activation of the Gi protein-RHOA axis by non-canonical Hedgehog signaling is independent of primary cilia. PLoS One. 2018. Aug;13(8):e0203170. 10.1371/journal.pone.0203170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Wittmann T, Waterman-Storer CM. Cell motility: can Rho GTPases and microtubules point the way? J Cell Sci. 2001. Nov;114(Pt 21):3795–803. 10.1242/jcs.114.21.3795 [DOI] [PubMed] [Google Scholar]
  62. Fults DW, Taylor MD, Garzia L. Leptomeningeal dissemination: a sinister pattern of medulloblastoma growth. J Neurosurg Pediatr. 2019. Feb;23(5):613–21. 10.3171/2018.11.PEDS18506 [DOI] [PubMed] [Google Scholar]
  63. Wang EJ, Chen IH, Kuo BY, Yu CC, Lai MT, Lin JT, et al. Alterations of Cytoskeleton Networks in Cell Fate Determination and Cancer Development. Biomolecules. 2022. Dec;12(12):1862. 10.3390/biom12121862 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Yu W, Sharma S, Rao E, Rowat AC, Gimzewski JK, Han D, et al. Cancer cell mechanobiology: a new frontier for cancer research. J Natl Cancer Cent. 2021. Dec;2(1):10–7. 10.1016/j.jncc.2021.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Hall A. The cytoskeleton and cancer. Cancer Metastasis Rev. 2009. Jun;28(1-2):5–14. 10.1007/s10555-008-9166-3 [DOI] [PubMed] [Google Scholar]
  66. Aseervatham J. Cytoskeletal remodeling in cancer. Biology (Basel). 2020. Nov;9(11):385. 10.3390/biology9110385 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Remsik J, Chi Y, Tong X, Sener U, Derderian C, Park A, et al. Leptomeningeal metastatic cells adopt two phenotypic states. Cancer Rep (Hoboken). 2022. Apr;5(4):e1236. 10.1002/cnr2.1236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Hatton BA, Villavicencio EH, Tsuchiya KD, Pritchard JI, Ditzler S, Pullar B, et al. The Smo/Smo model: hedgehog-induced medulloblastoma with 90% incidence and leptomeningeal spread. Cancer Res. 2008. Mar;68(6):1768–76. 10.1158/0008-5472.CAN-07-5092 [DOI] [PubMed] [Google Scholar]
  69. Berrino C, Omar A. Unravelling the Mysteries of the Sonic Hedgehog Pathway in Cancer Stem Cells: Activity, Crosstalk and Regulation. Curr Issues Mol Biol. 2024. May;46(6):5397–419. 10.3390/cimb46060323 [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Wattanapanitch M, Klincumhom N, Potirat P, Amornpisutt R, Lorthongpanich C, U-pratya Y, et al. Dual small-molecule targeting of SMAD signaling stimulates human induced pluripotent stem cells toward neural lineages. PLoS One. 2014. Sep;9(9):e106952. 10.1371/journal.pone.0106952 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Tilson SG, Haley EM, Triantafillu UL, Dozier DA, Langford CP, Gillespie GY, et al. ROCK inhibition facilitates in vitro expansion of glioblastoma stem-like cells. PLoS One. 2015. Jul;10(7):e0132823. 10.1371/journal.pone.0132823 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Ohata H, Ishiguro T, Aihara Y, Sato A, Sakai H, Sekine S, et al. Induction of the stem-like cell regulator CD44 by Rho kinase inhibition contributes to the maintenance of colon cancer-initiating cells. Cancer Res. 2012. Oct;72(19):5101–10. 10.1158/0008-5472.CAN-11-3812 [DOI] [PubMed] [Google Scholar]
  73. Lai WH, Ho JC, Lee YK, Ng KM, Au KW, Chan YC, et al. ROCK inhibition facilitates the generation of human-induced pluripotent stem cells in a defined, feeder-, and serum-free system. Cell Reprogram. 2010. Dec;12(6):641–53. 10.1089/cell.2010.0051 [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Pietrobono S, Gagliardi S, Stecca B. Non-canonical Hedgehog Signaling Pathway in Cancer: Activation of GLI Transcription Factors Beyond Smoothened. Front Genet. 2019. Jun;10:556. 10.3389/fgene.2019.00556 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Yale Journal of Biology and Medicine are provided here courtesy of Yale Journal of Biology and Medicine

RESOURCES