High SOX9 Maintains Glioma Stem Cell Activity through a Regulatory Loop Involving STAT3 and PML

Citation: Aldaz, P.; Martín-Martín, N.; Saenz-Antoñanzas, A.; Carrasco-Garcia, E.; Álvarez-Satta, M.; Elúa-Pinin, A.; Pollard, S.M.; Lawrie, C.H.; Moreno-Valladares, M.; Samprón, N.; et al. High SOX9 Maintains Glioma Stem Cell Activity through a Regulatory Loop Involving STAT3 and PML. Int. J. Mol. Sci. 2022, 23, 4511. https://doi.org/10.3390/ ijms23094511 Academic Editor: Christina Piperi Received: 11 March 2022 Accepted: 18 April 2022 Published: 19 April 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). International Journal of Molecular Sciences Article High SOX9 Maintains Glioma Stem Cell Activity through a Regulatory Loop Involving STAT3 and PML Paula Aldaz 1, Natalia Martín-Martín2, Ander Saenz-Antoñanzas 1, Estefania Carrasco-Garcia 1,3, MaríaÁlvarez-Satta 1, Alejandro Elúa-Pinin 4, Steven M. Pollard 5, Charles H. Lawrie 6,7 , Manuel Moreno-Valladares 1,4, Nicolás Samprón1,4, Jürgen Hench 8, Robin Lovell-Badge 9, Arkaitz Carracedo 2,8,10,11 and Ander Matheu 1,3,7,* 1Group of Cellular Oncology, Biodonostia Health Research Institute, 20014 San Sebastian, Spain; [email protected] (P.A.); ander[email protected] (A.S.-A.); [email protected] (E.C.-G.); maria.alvar[email protected]g (M.Á.-S.); [email protected] (M.M.-V.); nicolas.sampr[email protected] (N.S.) 2Center for Cooperative Research in Biosciences (CIC bioGUNE), Basque Research and Technology Alliance (BRTA), Bizkaia Technology Park, 48160 Derio, Spain; [email protected] (N.M.-M.); [email protected] (A.C.) 3CIBER of Frailty and Healthy Aging (CIBERFES), Carlos III Institute, 28029 Madrid, Spain 4Donostia University Hospital, 20014 San Sebastian, Spain; [email protected] 5Centre for Regenerative Medicine & Edinburgh Cancer Research UK Centre, Institute for Regeneration and Repair, Edinburgh EH16 4UU, UK; [email protected] 6Group of Molecular Oncology, Biodonostia Health Research Institute, 20014 San Sebastian, Spain; [email protected] 7Ikerbasque, Basque Foundation for Science, 48009 Bilbao, Spain 8Institute of Pathology, University Hospital Basel, 48009 Basel, Switzerland; [email protected] 9The Francis Crick Institute, London NW1 1AT, UK; r[email protected] 10 Biochemistry and Molecular Biology Department, University of the Basque Country (UPV/EHU), 48940 Leioa, Spain 11 CIBER of Cancer (CIBERONC), Carlos III Institute, 28029 Madrid, Spain *Correspondence: ander[email protected]; Tel.: +34-943006073 Abstract: Glioma stem cells (GSCs) are critical targets for glioma therapy. SOX9 is a transcription factor with critical roles during neurodevelopment, particularly within neural stem cells. Previous studies showed that high levels of SOX9 are associated with poor glioma patient survival. SOX9 knockdown impairs GSCs proliferation, confirming its potential as a target for glioma therapy. In this study, we characterized the function of SOX9 directly in patient-derived glioma stem cells. Notably, transcriptome analysis of GSCs with SOX9 knockdown revealed STAT3 and PML as downstream targets. Functional studies demonstrated that SOX9, STAT3, and PML form a regulatory loop that is key for GSC activity and self-renewal. Analysis of glioma clinical biopsies confirmed a positive correlation between SOX9/STAT3/PML and poor patient survival among the cases with the highest SOX9 expression levels. Importantly, direct STAT3 or PML inhibitors reduced the expression of SOX9, STAT3, and PML proteins, which significantly reduced GSCs tumorigenicity. In summary, our study reveals a novel role for SOX9 upstream of STAT3, as a GSC pathway regulator, and presents pharmacological inhibitors of the signaling cascade. Keywords: glioblastoma; glioma stem cell; SOX9; therapy; transcriptome; STAT3; PML; pharmacological inhibition 1. Introduction Glioblastoma, IDH wildtype, (GB) is a grade IV diffuse glioma [ 1 ], and represents the most common and aggressive primary brain tumor class in adults, with a median survival of 15 months and a 5-year survival rate of less than 5% [ 2 ]. Conventional treatment consists of surgical bulk removal of the tumor, followed up by combined radiotherapy Int. J. Mol. Sci. 2022,23, 4511. https://doi.org/10.3390/ijms23094511 https://www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2022,23, 4511 2 of 20 and temozolomide (TMZ)-based chemotherapy [ 3 ]. However, current therapy protocols have low success, and patients almost always relapse, often in a more aggressive manner. This is partially due to a cellular hierarchy, in which a subpopulation of glioma stem cells (GSCs) contribute to tumor relapse [ 4 ] and therapeutic resistance [ 5 , 6 ]. Therefore, strategies aiming at the eradication of GSCs are potentially promising for improving the prognosis of patients with GB [7]. GSCs share properties with neural stem cells (NSCs). Studies in mouse models and human tumor specimens have demonstrated that NSCs are likely the cells of origin of human GB [ 8 – 10 ]. There is increasing evidence indicating that transcriptional and epigenetic pathways controlling normal NSCs activity also contribute to the regulation of GSCs [ 6 , 11 ]. Transcription factors that govern stem cell differentiation can potentially function as oncogenes by promoting the acquisition of transcriptional programs required for tumorigenesis, including epigenetic dysregulation [ 12 ]. Among them, several members of the SOX ( S ex-determining region Y (SRY)-b OX ) family of transcription factors have been identified to be essential for GB propagation and GSC activity [13–16]. SOX9 is a critical developmental regulator that plays an essential role in the establishment and maintenance of adult stem cells in a wide range of tissues, including the central nervous system [ 17 ]. Studies on SOX9 gain and loss of function in cell cultures and mice models revealed that it maintains adult NSCs [ 18 , 19 ]. Moreover, SOX9 facilitates the neoplastic transformation of different cell types including NSCs [ 20 – 22 ] and exerts a prooncogenic activity by controlling cancer stem cells (CSCs) in several tumor types, including GB [ 14 , 23 ]. In tumor biopsies, SOX9 expression is generally elevated and correlates with poor prognosis [ 20 , 24 ]. Although it is known that the SOX9 transcription factor plays a role in CSCs, the SOX9-related molecular downstream effectors in these cells in human samples remain poorly understood. In this study, we present evidence that SOX9 is a critical regulator for GSC maintenance. We reveal that STAT3 and PML are critical effectors regulated by SOX9. We found a genetic regulatory loop involving SOX9, STAT3, and PML that modulates GSC activity. Hence, pharmacological inhibition of the SOX9–STAT3–PML pathway may represent a novel treatment strategy for GB. 2. Results 2.1. High SOX9 Levels Correlate with Lower Patient Survival We first characterized the expression of SOX9 in glioma grade taking advantage of publicly available TCGA and Rembrandt cohorts. Herein, we found that SOX9 was higher in grade IV cases, linking its high levels to tumor malignancy (Figure 1A). Then, we analyzed the clinical relevance of SOX9 in GB by studying its expression in a cohort of 88 human GB patients from Donostia University Hospital and compared it with healthy brain tissue (Figure 1B). The levels of SOX9 mRNA were significantly upregulated in GB, where more than 80% of samples (71 out of 88 patients) showed overexpression (fold change higher than two) (Figure 1C). These results were also confirmed at the protein level since TMA data from Donostia Hospital and University Hospital Basel showed that SOX9 expression was elevated in GB samples (Figure 1D,E). Remarkably, survival analysis revealed that GB patients with high SOX9 displayed significantly poorer outcomes than patients with low expression. Thus, patient median survival was reduced from 24 to 13 months in patients with high SOX9 from Donostia Hospital (Figure 1F) and from 8 to 3.5 months in patients from Hospital Basel (Figure 1G). These results suggest SOX9 as a prognostic biomarker in GB. Int. J. Mol. Sci. 2022,23, 4511 3 of 20 Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 3 of 21 Figure 1. High levels of SOX9 correlate with poor survival:(A) SOX9 mRNA expression in grade II, III, and IV of glioma in TCGA and Rembrandt cohorts; (B) SOX9 mRNA expression in GB patients from the Donostia University Hospital (DUH; n = 88) relative to the mean expression in healthy brain tissue (n = 6); (C) percentage of patients with “overexpression” and “normal” expression of SOX9 in the DUH cohort; (D) representative images of TMAs for SOX9 staining in GB samples from the DUH (n = 47) and University Hospital Basel (UHB; n = 20) cohorts; (E) percentage of patients with “SOX9+” (less than 60% of SOX9-positive cells) and “SOX9++” (equal to or more than 60%) expression from TMAs shown in “C”; (F) Kaplan–Meier survival analysis in GB patients from the DUH cohort (n = 47; p = 0.003) and (G) UHB cohort (n = 20; p = 0.018) based on SOX9 protein expression levels determined by TMA. 2.2. SOX9 Upregulation Increases Tumorigenic Capacity of GSCs We previously demonstrated that SOX9 was overexpressed in oncospheres derived from conventional and patient-derived GSCs, both in vitro and in vivo [23]. In order to further ascertain the molecular and biological processes controlled by SOX9 in GSCs, we first overexpressed SOX9 in GNS166 patient-derived cells and also in U373-MG cells (Figure 2A) and performed in vitro and in vivo experiments. In vitro, SOX9 overexpression promoted increased cell growth (Figure 2B). In vivo, we observed an enhanced tumor formation capacity when SOX9-upregulated GNS166 cells were injected orthotopically in immunodeficient NOD-SCID mice. This led to a significant decrease in mice survival (p = 0.004), with an overall lifespan of 35 ± 7.8 days, compared with 116 ± 11.5 days of median survival in controls (Figure 2C). In addition, we injected control and SOX9-upregulated U373-MG cells subcutaneously in Foxn1nu/Foxn1nu nude mice, which notably increased their tumorigenic potential. Thus, 87.5% of mice inoculated with SOX9-upregulated U373MG cells developed tumors, compared with the 12.5% of mice injected with control cells. Tumors generated by U373-MG cells with overexpression of SOX9 appeared earlier (day 53 vs. day 84), and they had greater volume (Figure 2D,E). Immunohistochemistry analyses also showed that these tumors had higher SOX9 and increased proliferative capacity measured by Ki67 expression (Figure 2F). Additionally, SOX2 levels were also increased Figure 1. High levels of SOX9 correlate with poor survival: ( A )SOX9 mRNA expression in grade II, III, and IV of glioma in TCGA and Rembrandt cohorts; ( B )SOX9 mRNA expression in GB patients from the Donostia University Hospital (DUH; n= 88) relative to the mean expression in healthy brain tissue (n= 6); ( C ) percentage of patients with “overexpression” and “normal” expression of SOX9 in the DUH cohort; ( D ) representative images of TMAs for SOX9 staining in GB samples from the DUH (n= 47) and University Hospital Basel (UHB; n= 20) cohorts; ( E ) percentage of patients with “SOX9 + ” (less than 60% of SOX9-positive cells) and “SOX9 ++ ” (equal to or more than 60%) expression from TMAs shown in “C”; ( F ) Kaplan–Meier survival analysis in GB patients from the DUH cohort (n= 47; p= 0.003) and ( G ) UHB cohort (n= 20; p= 0.018) based on SOX9 protein expression levels determined by TMA. 2.2. SOX9 Upregulation Increases Tumorigenic Capacity of GSCs We previously demonstrated that SOX9 was overexpressed in oncospheres derived from conventional and patient-derived GSCs, both in vitro and in vivo [ 23 ]. In order to further ascertain the molecular and biological processes controlled by SOX9 in GSCs, we first overexpressed SOX9 in GNS166 patient-derived cells and also in U373-MG cells (Figure 2A) and performed in vitro and in vivo experiments. In vitro , SOX9 overexpression promoted increased cell growth (Figure 2B). In vivo , we observed an enhanced tumor formation capacity when SOX9-upregulated GNS166 cells were injected orthotopically in immunodeficient NOD-SCID mice. This led to a significant decrease in mice survival (p= 0.004) , with an overall lifespan of 35 ± 7.8 days, compared with 116 ± 11.5 days of median survival in controls (Figure 2C). In addition, we injected control and SOX9-upregulated U373-MG cells subcutaneously in Foxn1 nu /Foxn1 nu nude mice, which notably increased their tumorigenic potential. Thus, 87.5% of mice inoculated with SOX9-upregulated U373-MG cells developed tumors, compared with the 12.5% of mice injected with control cells. Tumors generated by U373-MG cells with overexpression of SOX9 appeared earlier (day 53 vs. day 84), and they had greater volume (Figure 2D,E). Immunohistochemistry analyses also Int. J. Mol. Sci. 2022,23, 4511 4 of 20 showed that these tumors had higher SOX9 and increased proliferative capacity measured by Ki67 expression (Figure 2F). Additionally, SOX2 levels were also increased suggesting enrichment of stemness activity (Figure 2F). Taken together, our results show that high SOX9 levels endorse tumorigenic capacity of GSCs. Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 4 of 21 suggesting enrichment of stemness activity (Figure 2F). Taken together, our results show that high SOX9 levels endorse tumorigenic capacity of GSCs. Figure 2. SOX9 upregulation leads to increased tumorigenicity: (A) representative immunoblot of SOX9 expression in indicated glioma cells infected with pWXL GFP (“Control”) and pWPXL-Sox9 (“SOX9”) plasmids. β-actin was used as loading control; (B) quantification of cell growth in control and SOX9 overexpression in U373-MG and GNS166 cells at day 5; (C) Kaplan–Meier curves representing NOD-SCID mice survival after stereotactic injection of GNS166 control and SOX9 GNS166 cells (n = 4).; (D) tumor volume at indicated time points after subcutaneous injection of control and SOX9 U373-MG cells in immunodeficient mice (n = 4); (E) representative image of tumors in both flanks (right flank: injection of SOX9-overexpressed cells; left flank: control cells). Tumors are indicated by arrows; (F) representative images of SOX9, SOX2, and Ki67 IHC staining from subcutaneously generated tumors in “Control” and “SOX9” groups. Scale bar corresponds to 100 µm. * p < 0.05, ** p < 0.01. 2.3. SOX9 Downregulation Decreases Stemness and Tumorigenic Capacity in GSCs To further address the impact of SOX9 on the regulation of GSCs, we knocked down SOX9 in GNS166 cells and conventional cells U373-MG and U251-MG (with intermediate and high basal levels of SOX9, respectively). Data from immunoblotting experiments confirmed an effective reduction in SOX9 levels accompanied by a marked diminishment of SOX2 expression (Figure 3A). In this context, we detected a significant decrease of more than 50% in cell growth in U373-MG and U251-MG cells (Figure 3B), as well as a reduction in the number of phospho-Histone H3 (P-H3)-positive cells in GNS166 cells (Figure 3C) in SOX9 knockdown cells. This impairment in cell proliferation was accompanied by a significant increase in senescence measured by cytoplasmic SA-β-gal activity (Figure 3D) and also by a decrease in apoptosis [25]. Moreover, SOX9 knockdown cells also presented markedly reduced stemness activity measured by decreased colony formation ability (Figure 3E), lower number of soft agar foci (Supplementary Material Figure S1A), increased expression of differentiation markers (Figure 3F), and reduced tumorsphere formation by more than 60% in both primary and secondary culture condition (Figures 3G and S1B). We performed in vivo experiments and found a notably reduced tumor formation and progression capacity for shSOX9 U373-MG cells in Foxn1nu/Foxn1nu nude mice. Thus, only 33% of mice inoculated with shSOX9 cells developed tumors, compared with 85% of controls, and they formed smaller tumors (Figure 3H). In line with this, shSOX9 tumors expressed lower levels of SOX9 and Ki67 than control tumors (Figure 3I). Similarly, SOX2 Figure 2. SOX9 upregulation leads to increased tumorigenicity: ( A ) representative immunoblot of SOX9 expression in indicated glioma cells infected with pWXL GFP (“Control”) and pWPXLSox9 (“SOX9”) plasmids. β -actin was used as loading control; ( B ) quantification of cell growth in control and SOX9 overexpression in U373-MG and GNS166 cells at day 5; ( C ) Kaplan–Meier curves representing NOD-SCID mice survival after stereotactic injection of GNS166 control and SOX9 GNS166 cells (n= 4); ( D ) tumor volume at indicated time points after subcutaneous injection of control and SOX9 U373-MG cells in immunodeficient mice (n= 4); ( E ) representative image of tumors in both flanks (right flank: injection of SOX9-overexpressed cells; left flank: control cells). Tumors are indicated by arrows; ( F ) representative images of SOX9, SOX2, and Ki67 IHC staining from subcutaneously generated tumors in “Control” and “SOX9” groups. Scale bar corresponds to 100 µm. * p< 0.05, ** p< 0.01. 2.3. SOX9 Downregulation Decreases Stemness and Tumorigenic Capacity in GSCs To further address the impact of SOX9 on the regulation of GSCs, we knocked down SOX9 in GNS166 cells and conventional cells U373-MG and U251-MG (with intermediate and high basal levels of SOX9, respectively). Data from immunoblotting experiments confirmed an effective reduction in SOX9 levels accompanied by a marked diminishment of SOX2 expression (Figure 3A). In this context, we detected a significant decrease of more than 50% in cell growth in U373-MG and U251-MG cells (Figure 3B), as well as a reduction in the number of phospho-Histone H3 (P-H3)-positive cells in GNS166 cells (Figure 3C) in SOX9 knockdown cells. This impairment in cell proliferation was accompanied by a significant increase in senescence measured by cytoplasmic SAβ -gal activity (Figure 3D) and also by a decrease in apoptosis [ 25 ]. Moreover, SOX9 knockdown cells also presented markedly reduced stemness activity measured by decreased colony formation ability (Figure 3E), lower number of soft agar foci (Supplementary Materials Figure S1A), increased expression of differentiation markers (Figure 3F), and reduced tumorsphere formation by more than 60% in both primary and secondary culture condition (Figures 3G and S1B). Int. J. Mol. Sci. 2022,23, 4511 5 of 20 Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 5 of 21 expression was also reduced (Figure 3I). Additionally, orthotopic injection of shSOX9 GNS166 cells revealed a significant increase in mice survival (p = 0.009) passing from ~100 days in controls to almost 200 days in shSOX9 (Figure 3J). Taken together, our results point out that SOX9 is a relevant mediator of malignant phenotypes of GB by modulating GSCs capacities. Figure 3. SOX9 knockdown impairs tumorigenicity: (A) representative immunoblot of SOX9 and SOX2 expression in glioma cells with SOX9 downregulation (shSOX9-1; sh1) and controls (pLKO). β-actin was used as loading control (n = 3); (B) quantification of cell growth in control and sh1 U373MG, U251-MG cells at day 5 (n = 3); (C) quantification of P-H3 + GNS166 cells infected with pLKO or sh1 plasmids (n = 3); (D) quantification of SA-β-gal + cells in sh1 and control cells (n = 3); (E) quantification of number of colonies formed from glioma cell lines infected with pLKO or sh1 plasmids (n = 3); (F) expression of Tuj1 and p27 KIP in GNS166 cells infected with pLKO or sh1 plasmids (n = 2); (G) quantification of primary tumorspheres after 10 days in culture (n = 3). Quantifications are expressed relative to pLKO; (H) tumor volume of subcutaneous tumors generated by pLKOand sh1infected U373-MG cells (n = 4); (I) SOX9, SOX2, and Ki67 IHC staining from subcutaneously generated tumors in “pLKO” and “sh1” groups. Scale bar corresponds to 100 µm; (J) Kaplan–Meier curves representing mice survival after stereotactic injection of pLKO and sh1 GNS166 cells. * p < 0.05, ** p < 0.01, *** p < 0.001. 2.4. Transcriptomic Analysis Reveals STAT3 and PML as Mediators of SOX9 Activity in GSCs We addressed the molecular mechanisms underlying SOX9 function in GB. For that purpose, we compared expression microarray data from SOX9 knockdown GNS166 cells with control GNS166 cells. Cluster analysis showed differences in gene expression between both phenotypes being the levels of SOX9 gene the most decreased validating the experimental approach (Figure 4A and Supplementary Materials Table S1). Gene Figure 3. SOX9 knockdown impairs tumorigenicity: ( A ) representative immunoblot of SOX9 and SOX2 expression in glioma cells with SOX9 downregulation (shSOX9-1;sh1) and controls (pLKO). β -actin was used as loading control (n= 3); ( B ) quantification of cell growth in control and sh1 U373-MG, U251-MG cells at day 5 (n= 3); ( C ) quantification of P-H3 + GNS166 cells infected with pLKO or sh1 plasmids (n= 3); ( D ) quantification of SAβ -gal + cells in sh1 and control cells (n= 3); ( E ) quantification of number of colonies formed from glioma cell lines infected with pLKO or sh1 plasmids (n= 3); ( F ) expression of Tuj1 and p27 KIP in GNS166 cells infected with pLKO or sh1 plasmids (n= 2); ( G ) quantification of primary tumorspheres after 10 days in culture (n= 3). Quantifications are expressed relative to pLKO; ( H ) tumor volume of subcutaneous tumors generated by pLKOand sh1-infected U373-MG cells (n= 4); ( I ) SOX9, SOX2, and Ki67 IHC staining from subcutaneously generated tumors in “pLKO” and “sh1” groups. Scale bar corresponds to 100 µ m; ( J ) Kaplan–Meier curves representing mice survival after stereotactic injection of pLKO and sh1 GNS166 cells. * p< 0.05, ** p< 0.01, *** p< 0.001. We performed in vivo experiments and found a notably reduced tumor formation and progression capacity for shSOX9 U373-MG cells in Foxn1 nu /Foxn1 nu nude mice. Thus, only 33% of mice inoculated with shSOX9 cells developed tumors, compared with 85% of controls, and they formed smaller tumors (Figure 3H). In line with this, shSOX9 tumors expressed lower levels of SOX9 and Ki67 than control tumors (Figure 3I). Similarly, SOX2 expression was also reduced (Figure 3I). Additionally, orthotopic injection of shSOX9 GNS166 cells revealed a significant increase in mice survival (p= 0.009) passing from ~100 days in controls to almost 200 days in shSOX9 (Figure 3J). Taken together, our results point out that SOX9 is a relevant mediator of malignant phenotypes of GB by modulating GSCs capacities. Int. J. Mol. Sci. 2022,23, 4511 6 of 20 2.4. Transcriptomic Analysis Reveals STAT3 and PML as Mediators of SOX9 Activity in GSCs We addressed the molecular mechanisms underlying SOX9 function in GB. For that purpose, we compared expression microarray data from SOX9 knockdown GNS166 cells with control GNS166 cells. Cluster analysis showed differences in gene expression between both phenotypes being the levels of SOX9 gene the most decreased validating the experimental approach (Figure 4A and Supplementary Materials Table S1). Gene Ontology analysis revealed JAK2-mediated signaling among the top pathways significantly altered in response to SOX9 knockdown, together with interferon signaling, interleukin 17 (IL17) signaling, or growth hormones (Figure 4B). Janus kinase 2 (JAK2) is a tyrosine kinase that activates the signal transducer and activator of transcription 3 (STAT3) transcription factor, which has been involved in the progression of most types of cancers including GB, which is required for GSC growth and self-renewal [ 26 ]. In addition, activated STAT3 levels are known to correlate with promyelocytic leukemia (PML) gene expression in several tumor models [ 27 ], including breast cancer, where it acts as an upstream regulator of PML [ 28 ]. In this context, we hypothesized that STAT3 and PML could be part of a regulatory pathway that regulates GSC activity. We validated transcriptomic results and detected reduced levels of STAT3 (total STAT3 and p-STAT3), as well as PML in shSOX9 GNS166 cells (Figure 4C). Similar results were obtained in SOX9-silenced U251-MG cells (Figure 4C). Accordingly, moderately higher levels of STAT3, p-STAT3, and PML were detected in SOX9 overexpressing U373-MG cells (Figure 4D). These data suggest that SOX9 regulates STAT3 and PML expression in GB. We analyzed SOX9, STAT3/p-STAT3, and PML expression in a set of established glioma cell lines (Figure 4E) and patient-derived cells cultures (Figure 4F,G) by immunoblot. We observed a positive correlation among SOX9, p-STAT3, and PML protein levels in all cell lines except U373-MG cells (Figure 4E–G). This correlation was also identified in tumorspheres from U87-MG and U251-MG cells, where SOX9, p-STAT3, and PML expression was increased (Figure 4H). In addition, similar results were obtained in subcutaneous tumors generated from U373-MG secondary tumorspheres in which the number of positive cells and the intensity of staining is higher for SOX9, STAT3, and PML, compared with U373-MG parental cells (Figure 4I) [ 23 ]. In line with this, the expression of STAT3 and PML also was higher in grade IV clinical biopsies from publicly available TCGA and Rembrandt cohorts (Figure 4J). To further reinforce the link between SOX9, STAT3, and PML, we examined bioptic tissue. Remarkably, the association was further confirmed in GB samples from the Hospital Basel cohort, where we found a significant positive correlation between SOX9 with PML and p-STAT3, as well as between PML and p-STAT3 (Figure 4K). To reinforce this result, we compared SOX9 and STAT3, as well as PML and STAT3 mRNA expression in The Cancer Genome Atlas (TCGA) cohort (n= 580) [ 29 ] and again found a positive correlation ( ρ Spearman = 0.61 and 0.68, respectively) (Figure 4L). Together, these results reveal that SOX9, STAT3, and PML expression correlate in GB biopsies. Int. J. Mol. Sci. 2022,23, 4511 7 of 20 Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 7 of 21 Figure 4. SOX9 expression correlates with STAT3 and PML: (A) cluster analysis of GNS166 cells after transcriptomic analysis; (B) top canonical pathways altered in shSOX9 (sh1) GNS166 cells compared to controls (n = 3); (C) representative immunoblots of PML, p-STAT3, STAT3 and SOX2 expression in sh1 and pLKO GNS166 and U251-MG cells (n = 3); (D) representative immunoblot of PML, p-STAT3, and STAT3 expression in SOX9 and control U373-MG cells (n = 3); (E) representative immunoblot of SOX9, p-STAT3, and PML expression in indicated glioma cell lines (n = 3); (F) representative immunoblot of p-STAT3 and STAT3 expression in GNS166, GNS179, and GB1 patientderived glioma stem cells; (G) representative immunoblot of SOX9 and PML protein levels in Figure 4. SOX9 expression correlates with STAT3 and PML: ( A ) cluster analysis of GNS166 cells after transcriptomic analysis; ( B ) top canonical pathways altered in shSOX9 (sh1) GNS166 cells compared to controls (n= 3); ( C ) representative immunoblots of PML, p-STAT3, STAT3 and SOX2 expression in sh1 and pLKO GNS166 and U251-MG cells (n= 3); ( D ) representative immunoblot of PML, p-STAT3, Int. J. Mol. Sci. 2022,23, 4511 8 of 20 and STAT3 expression in SOX9 and control U373-MG cells (n= 3); ( E ) representative immunoblot of SOX9, p-STAT3, and PML expression in indicated glioma cell lines (n= 3); ( F ) representative immunoblot of p-STAT3 and STAT3 expression in GNS166, GNS179, and GB1 patient-derived glioma stem cells; ( G ) representative immunoblot of SOX9 and PML protein levels in patient-derived GSCs (23) (n= 3); ( H ) representative immunoblot of SOX9, p-STAT3, and PML expression in parental (“-”) and oncospheres (GSCs) from indicated cells; ( I ) representative images of SOX9, PML, and p-STAT3 IHC staining from tumors after subcutaneous injection of parental and tumorspheres from U373-MG cells. Scale bar corresponds to 50 µm; (J)STAT3 and PML mRNA expression in grades II, III, and IV of glioma in TCGA and Rembrandt cohorts; ( K ) TMA of p-STAT3, PML, and SOX9 in GB human samples from the Hospital Basel cohort (n= 20). Chi-squared ( χ2 ) test was used to compare frequency data and showed statistically significant differences between groups (p< 0.01 or 0.0001, respectively) with high and low staining of the indicated proteins; ( L ) correlation of STAT3 with SOX9 and PML mRNA expression in GB samples from the TCGA cohort (n= 580) ( ρ Spearman = 0.61 and 0.68, respectively). N.A., not available. ** p< 0.01, *** p< 0.001. 2.5. STAT3 Regulates GSC Activity and Its Pharmacological Inhibition Reduces Tumorigenicity To investigate the SOX9–STAT3–PML axis as a potential therapeutic target, we knocked down STAT3 in GNS166 and U251-MG cells using two different shRNA constructs (sh41 and sh43) and analyzed the resulting phenotypes. Immunoblot confirmed the reduced levels of STAT3 and p-STAT3, as well as SOX9 and PML (Figure 5A). STAT3 silencing led to a significant reduction of more than 50% in cell proliferation in both models (Figure 5B), which was associated with an increased number of senescent cells (Figure 5C). Moreover, STAT3 knockdown impaired the tumorsphere formation of U251-MG glioma cells (Figure 5D). These results confirm overlapping phenotypes in SOX9 and STAT3 knockdown glioma cells in vitro. We next moved to in vivo assays, which further confirmed these results. Thus, immunodeficient mice subcutaneously injected with sh41 or sh43 U251-MG cells did not form tumors (sh43) or formed them at a low percentage and with less volume (sh41) than control mice (Figure 5E). In addition, the limiting dilution assay revealed that STAT3 knockdown led to a reduced number of tumor-initiating cells (Figure 5F). Overall, our data reveal that STAT3 knockdown produces equal phenotypes of similar severity as those observed after SOX9 downregulation. Since established pharmacological STAT3 inhibitors could potentially be translated to clinical practice, we explored the effect of STAT3 inhibition on the SOX9–STAT3–PML axis and GB progression. We tested the specific STAT3 inhibitor STX-0119, which prevents the dimerization of STAT3 and its subsequent binding to DNA. We found that increasing doses (25, 50, and 100 µ M) of this drug significantly reduced viability in glioma and GSC cells (Figure 5G). In addition, the tumorsphere formation ability rate was significantly diminished in all cell cultures (Figure 5D). Remarkably, the pharmacological inhibition of STAT3 reduced SOX9 and PML expression levels, as well as the expression of SOX2, a stem cell marker (Figure 5I), mimicking the effects of gene silencing of STAT3. We have previously described that SOX2 controls SOX9 levels [ 23 ], so it might be of interest to unravel whether it plays a role in the SOX9–STAT3–PML axis. Int. J. Mol. Sci. 2022,23, 4511 9 of 20 Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 9 of 21 Figure 5. Genetic and pharmacological STAT3 inhibition in glioma cells: (A) representative immunoblots of p-STAT3, STAT3, PML, SOX9, and SOX2 in U251-MG and GNS166 cells in STAT3 knockdown (“sh41” and “sh43”) and control (“pLKO”) cells (n = 3); (B) quantification of cell growth in sh41 and sh43 cells compared with controls (n = 2); (C) quantification of SA-β-gal+ cells in STAT3 knockdown cells (n = 2); (D) quantification of tumorspheres (primary and secondary) in sh41 and sh43 U251-MG cells relative to pLKO condition (n = 2); (E) tumor volume representation at indicated time points after subcutaneous injection of U251-MG cells (n = 4); (F) ELDA plot of limiting dilution assay of sh41, sh43, and control U251-MG cells; (G) cytotoxicity exhibited by indicated glioma cells after treatment with increasing doses of STX-0119 for 72 h (n = 6); (H) quantification of tumorspheres in indicated cells after treatment with 50 and 100 µM STX-0119 (n = 3). Calculations were relative to untreated cells; (I) representative immunoblot of p-STAT3, STAT3, PML, SOX9, and SOX2 Figure 5. Genetic and pharmacological STAT3 inhibition in glioma cells: ( A ) representative immunoblots of p-STAT3, STAT3, PML, SOX9, and SOX2 in U251-MG and GNS166 cells in STAT3 knockdown (“sh41” and “sh43”) and control (“pLKO”) cells (n= 3); ( B ) quantification of cell growth in sh41 and sh43 cells compared with controls (n= 2); ( C ) quantification of SAβ -gal + cells in STAT3 knockdown cells (n= 2); ( D ) quantification of tumorspheres (primary and secondary) in sh41 and sh43 U251-MG cells relative to pLKO condition (n= 2); ( E ) tumor volume representation at indicated time points after subcutaneous injection of U251-MG cells (n= 4); ( F ) ELDA plot of limiting dilution assay of sh41, sh43, and control U251-MG cells; ( G ) cytotoxicity exhibited by indicated glioma cells after treatment with increasing doses of STX-0119 for 72 h (n= 6); ( H ) quantification of tumorspheres in indicated cells after treatment with 50 and 100 µ M STX-0119 (n= 3). Calculations were relative to untreated cells; ( I ) representative immunoblot of p-STAT3, STAT3, PML, SOX9, and SOX2 expression in U251-MG cells after treatment with 50 and 100 µ M STX-0119 for 72 h (n= 3). * p< 0.05, ** p< 0.01, *** p< 0.001. Int. J. Mol. Sci. 2022,23, 4511 16 of 20 in culture, spheres were counted using a light microscope. For secondary (2 ry ) generation, spheres were mechanically and enzymatically disaggregated with accutase (Life Technologies, Carlsbad, CA, USA) and then maintained in culture for another 10 days. 4.7. Senescence-Associated β-Galactosidase Staining To evaluate cellular senescence, the senescence-associated β -galactosidase (SAβ -gal) activity was measured using a commercial staining kit (9860S, Cell Signaling, Danvers, MA, USA) according to the manufacturer’s protocol. 4.8. Cell Viability Assay Cells were seeded in 96-well plates at a density of 2.5 × 10 3 cells per well and treated 24 h later with the indicated concentrations of STX-0119 (Sigma-Aldrich) for 72 h or arsenic trioxide (ATO; Sigma-Aldrich) for 48 h, in sextuplicates. As a control, cells were treated with the indicated solvent for each drug. Cells were then incubated with 0.25 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich) for 3 h. The formazan produced by viable cells was dissolved in 150 µ L of dimethyl sulfoxide (DMSO; Sigma-Aldrich), and the absorbance was determined at 570 nm in a microplate reader (Multiskan Ascent Thermo Electron Corporation, Waltham, MA, USA). 4.9. Quantitative Real-Time PCR Total RNA was extracted using Trizol (Life Technologies). Reverse transcription was performed from 1 µ g of total RNA using random primers and the High-Capacity cDNA Reverse Transcription Kit (Life Technologies) according to the manufacturer’s guidelines. Quantitative real-time PCR (qPCR) was then carried out using Power SYBR ® Green Master Mix (Thermo Scientific, Waltham, MA, USA), 10 mM of primers, and 20 ng of cDNA in an ABI PRISM 7300 thermocycler (Applied Biosystems, Waltham, MA, USA). The ∆∆ CT method was used for relative quantification of gene expression, using GAPDH as the reference gene. Primer sequences are available upon request. 4.10. Immunoblot Immunoblots were performed as previously described [ 50 ]. The following primary antibodies were used: rabbit polyclonal anti-SOX9 (1:2000 dilution; Millipore, Burlington, MA, USA), mouse monoclonal anti-STAT3 (1:1000 dilution; Cell Signaling), rabbit monoclonal anti-phospho STAT3 (Tyr705) (1:1000 dilution; Cell Signaling), rabbit polyclonal anti-PML (1:1000 dilution; Bethyl laboratories, Montgomery, TX, USA), rabbit polyclonal anti-SOX2 (1:500 dilution; Millipore), and mouse monoclonal antiβ actin (1:2000 dilution; Sigma-Aldrich). Primary antibodies were detected with horseradish peroxidase (HRP)- linked antibodies: Goat anti-rabbit or goat anti-mouse (Santa Cruz Biotechnology, Dallas, TX, USA). Protein detection was performed using the NOVEX ® ECL system (Invitrogen, Carlsbad, CA, USA). 4.11. Immunofluorescence A total of 2 × 10 4 cells were seeded in Lab-Tek II Chamber Slides (ThermoFisher), fixed after 24 h with 4% formalin for 10 min at room temperature (RT), and blocked with phosphate-buffered saline (PBS) supplemented with 0.3% Triton X-100 (Sigma-Aldrich) and 5% FBS for 1 h at RT. Cells were then incubated for 2 h at RT with mouse monoclonal anti-Histone H3 (1:1000 dilution; Abcam, Cambridge, UK) and with Alexa Fluor ® 488 rabbit anti-mouse IgG (H + L) secondary antibody (1:500 dilution; A-11059, Invitrogen) for 1 h at RT. After that, slides were washed and mounted with a Vectashield Mounting Medium with DAPI (Vector Laboratories, Burlingame, CA, USA). Images were obtained with a Nikon Eclipse 80i microscope. Int. J. Mol. Sci. 2022,23, 4511 17 of 20 4.12. In Vivo Carcinogenesis Assays For subcutaneous injection, cells were resuspended in PBS. Briefly, 1 × 10 6 cells per condition were injected subcutaneously into both flanks of 8-week-old Foxn1 nu /Foxn1 nu nude mice. External calipers were used to measure tumor size; tumor volume was then estimated by the following formula: V = L × W 2× 0.5 (L = tumor length and W = tumor width). For xenotransplantation, GSCs were injected stereotactically into the frontal cortex of 6–8-week-old NOD-SCID mice. Briefly, GSCs were disaggregated with accutase and resuspended in PBS. Then, 1 × 10 4 or 1 × 10 5 cells in a final volume of 1 µ L were injected into the putamen using a stereotaxic apparatus (Kopf Instruments). For tumor initiation assays, U251-MG cells at a density of 5 × 10 4 or 5 × 10 5 cells per condition were injected subcutaneously into both flanks of 8-week-old Foxn1 nu /Foxn1 nu nude mice. To calculate the number of initiating cells, we used Extreme Limiting Dilution Analysis (ELDA) software (http://bioinf.wehi.edu.au/software/elda/, accessed on 10 March 2022). All animal handling and procedures were performed according to the ethical guidelines established by the Animal Experimentation Ethics Committee of Biodonostia Health Research Institute (CEEA14/016) and conducted in conformity with the European Union recommendations for animal experimentation specified in the Directive 2010/63/EU. 4.13. Immunohistochemistry Tumors generated in mice were dissected, fixed in 10% formalin for 48 h, and embedded in paraffin. Four micrometer-thick sections were stained with hematoxylin–eosin (H&E) using a Varistain Gemini ES machine (ThermoFisher, Waltham, MA, USA). For immunohistochemistry (IHC), sections were rehydrated and heated in citrate buffer pH 6 for 10 min for antigen retrieval. Endogenous peroxidase was blocked with 5% hydrogen peroxide in methanol for 15 min. Sections were incubated with the following primary antibodies: rabbit polyclonal anti-SOX9 (1:1000 dilution; Millipore), mouse monoclonal anti-phospho-STAT3 (Tyr705) (1:100 dilution; Cell Signaling), mouse monoclonal anti-PML (PG-M3) (1:200 dilution; Santa Cruz Biotechnology), rabbit polyclonal anti-SOX2 (1:500 dilution; Millipore), and rabbit polyclonal anti-Ki67 (1:1000 dilution; Abcam). Sections were then incubated with MACH 3 Rabbit HRP-Polymer (BioCare Medical). Staining was developed with 3,3’-diaminobenzidine (DAB) (Spring Bioscience, Pleasanton, CA, USA). IHC images were obtained with a Nikon Eclipse 80i microscope. 4.14. Transcriptome Analysis Expression microarrays were performed from 0.5 µ g of RNA from SOX9-silenced GNS166 cells and control GNS166 cells using the Gene Chip Human Genome U133 Plus 2.0 array (Affymetrix). Data were normalized by Robust Multiarray Average (RMA) using Affymetrix ® Expression Console ™ software V1.1. Differential gene expression analysis was carried out with Genespring software GeneSpring GX 14.9 (Agilent, Santa Clara, CA, USA). Pathway analysis was performed with the Interactive Pathway Analysis (IPA ® ) software (Ingenuity Systems, Redwood city, CA, USA, https://www.qiagenbioinformatics.com/ products/ingenuity-pathway-analysis, accessed on 10 March 2022). The data discussed in this publication have been deposited in NCBI’s Gene Expression Omnibus and are accessible in GSE181035. 4.15. Chromatin Immunoprecipitation Assay ASimpleChIP Enzymatic Chromatin IP Kit (Cell Signaling) was used for the chromatin immunoprecipitation (ChIP) assay. In detail, U251-MG cells harboring a doxycyclineinducible HA-PMLIV plasmid were grown in 150 mm dishes for 3 days and cross-linked with 35% formaldehyde for 10 min at RT. Then, cells were incubated for 5 min at RT after the addition of glycine, washed with PBS, and scraped into PBS with 100 µ M phenylmethylsulfonyl fluoride (PMSF). Pelleted cells were lysed and nuclei harvested. Nuclear lysates were digested with micrococcal nuclease for 20 min at 37 ◦ C and then sonicated in 500 µ L aliquots on ice for 3 pulses of 15 s using a Branson sonicator. Lysates were clarified, and chromatin Int. J. Mol. Sci. 2022,23, 4511 18 of 20 was stored at − 80 ◦ C until used. HA-Tag polyclonal antibody (Cell Signaling), rabbit polyclonal anti-PML (Bethyl laboratories), and normal rabbit IgG polyclonal antibody (Cell Signaling) were incubated overnight at 4 ◦ C with rotation. After that, protein G-conjugated magnetic beads were added and incubated for 2 h at 4 ◦ C. Washes and elution of chromatin were later performed and DNA quantification was carried out using a Viia7 Real-Time PCR System (Applied Biosystems) with SYBR Green and primers that amplify the predicted PML binding region to SOX9 promoter (chr17:70117013-70117409). Primer sequences used were forward 50-ccggaaacttttctttgcag-30and reverse 50-cggcgagcacttaggaag-30. 4.16. Data Availability Statement The authors confirm that the data supporting the findings of this study are available within the article and its Supplementary Materials. The microarray data that support the findings of this study are openly available in NCBI’s Gene Expression Omnibus repository. 4.17. Data Analysis Data are presented as the mean ± standard error of the mean (SEM) with the number of experiments (n) in parentheses. Mean values from quantitative variables were compared using Student’s t-test. Log-rank test was performed for Kaplan–Meier survival analyses. Correlations were calculated using the Spearman coefficient. Chi-squared (X 2 ) test was used to compare frequency data. Asterisks (*, ** and ***) indicate statistical significance (p< 0.05, p< 0.01 and p< 0.001, respectively). Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijms23094511/s1. Author Contributions: P.A., N.M.-M., A.S.-A. and E.C.-G. performed the experiments in glioma cells; A.E.-P. helped with stereotaxic experiments in vivo in mice; S.M.P. generated patient-derived cells; A.E.-P., M.M.-V., J.H. and N.S. collected clinical data and samples from patients, evaluated them, and performed the clinical experiments; C.H.L. performed the transcriptomic analysis; M.Á.-S. helped to write the manuscript; R.L.-B. and A.C. coordinated the experiments, analyzed results, and helped to direct the project; A.M. directed the project, contributed to data analysis, obtained funds, and wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data that support the findings of this study are included in the manuscript and Supplementary Materials. Microarray data are openly available in GEO database with access number GSE181035. Acknowledgments: P.A. and A.S.-A. were recipients of predoctoral fellowships from the AECC foundation and Carlos III Institute (ISCIII), respectively. M.Á.-S. holds a Sara Borrell postdoctoral contract from the ISCIII (CD19/00154). E.C.-G. was a recipient of a Stop Fuga de Cerebros postdoctoral fellowship and holds a Miguel Servet contract from the ISCIII (CP19/00085). We thank the Histology Platform of the Biodonostia Health Research Institute, The Neuro-Oncology Committee of Donostia University Hospital, and Basque Biobank for their help. This research was supported by grants from ISCIII and FEDER Funds (CP16/00039, DTS16/00184, PI16/01580, DTS18/00181, PI18/01612, CP19/00085), and the Industry and Health Departments of the Basque Country. Conflicts of Interest: The authors declare no conflict of interest. Int. J. Mol. Sci. 2022,23, 4511 19 of 20 References 1. Louis, D.N.; Perry, A.; Reifenberger, G.; von Deimling, A.; Figarella-Branger, D.; Cavenee, W.K.; Ellison, D.W. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: A summary. Acta Neuropathol. 2016,131, 803–820. [CrossRef] [PubMed] 2. Ostrom, Q.T.; Bauchet, L.; Davis, F.G.; Deltour, I.; Fisher, J.L.; Langer, C.E.; Pekmezci, M.; Schwartzbaum, J.A.; Turner, M.C.; Walsh, K.M.; et al. The epidemiology of glioma in adults: A “state of the science” review. Neuro Oncol. 2014 ,16, 896–913. [CrossRef] [PubMed] 3. Stupp, R.; Mason, W.P.; van den Bent, M.J.; Weller, M.; Fisher, B.; Taphoorn, M.J.B.; Belanger, K.; Brandes, A.A.; Marosi, C.; Bogdahn, U.; et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N. Engl. J. Med. 2005 ,352, 987–996. [CrossRef] [PubMed] 4. Chen, J.; Li, Y.; Yu, T.-S.; McKay, R.M.; Burns, D.K.; Kernie, S.G.; Parada, L.F. A restricted cell population propagates glioblastoma growth after chemotherapy. Nature 2012,488, 522–526. [CrossRef] 5. Bao, S.; Wu, Q.; McLendon, R.E.; Hao, Y.; Shi, Q.; Hjelmeland, A.B.; Dewhirst, M.W.; Bigner, D.D.; Rich, J.N. Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 2006,444, 756–760. [CrossRef] 6. Lan, X.; Joerg, D.; Cavalli, F.; Richards, L.; Nguyen, L.; Vanner, R.; Guilhamon, P.; Lee, L.; Kushida, M.; Pellacani, D.; et al. Fate mapping of human glioblastoma reveals an invariant stem cell hierarchy. Nature 2017,549, 227–232. [CrossRef] 7. Lathia, J.D.; Mack, S.C.; Mulkearns-Hubert, E.E.; Valentim, C.L.; Rich, J.N. Cancer stem cells in glioblastoma. Genes Dev. 2015 ,29, 1203–1217. [CrossRef] 8. Alcantara Llaguno, S.; Chen, J.; Kwon, C.H.; Jackson, E.L.; Li, Y.; Burns, D.K.; Parada, L.F. Malignant astrocytomas originate from neural stem/progenitor cells in a somatic tumor suppressor mouse model. Cancer Cell 2009,15, 45–56. [CrossRef] 9. Alcantara Llaguno, S.R.; Wang, Z.; Sun, D.; Chen, J.; Xu, J.; Kim, E.; Parada, L.F. Adult Lineage-Restricted CNS Progenitors Specify Distinct Glioblastoma Subtypes. Cancer Cell 2015,28, 429–440. [CrossRef] 10. Lee, J.H.; Lee, J.E.; Kahng, J.Y.; Kim, S.H.; Park, J.S.; Yoon, S.J.; Lee, J.H. Human glioblastoma arises from subventricular zone cells with low-level driver mutations. Nature 2018,560, 243–247. [CrossRef] 11. Patel, A.P.; Tirosh, I.; Trombetta, J.J.; Shalek, A.K.; Gillespie, S.M.; Wakimoto, H.; Cahill, D.P.; Nahed, B.V.; Curry, W.T.; Martuza, R.L.; et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 2014 ,344, 1396–1401. [CrossRef] [PubMed] 12. Suvà, M.L.; Riggi, N.; Bernstein, B.E. Epigenetic reprogramming in cancer. Science 2013,339, 1567–1570. [CrossRef] [PubMed] 13. Suva, M.L.; Rheinbay, E.; Gillespie, S.M.; Patel, A.P.; Wakimoto, H.; Rabkin, S.D.; Bernstein, B.E. Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells. Cell 2014,157, 580–594. [CrossRef] [PubMed] 14. de la Rocha, A.M.; Sampron, N.; Alonso, M.M.; Matheu, A. Role of SOX family of transcription factors in central nervous system tumors. Am. J. Cancer Res. 2014,4, 312–324. 15. Caren, H.; Stricker, S.H.; Bulstrode, H.; Gagrica, S.; Johnstone, E.; Bartlett, T.E.; Pollard, S.M. Glioblastoma Stem Cells Respond to Differentiation Cues but Fail to Undergo Commitment and Terminal Cell-Cycle Arrest. Stem Cell Rep. 2015 ,5, 829–842. [CrossRef] 16. MacLeod, G.; Bozek, D.A.; Rajakulendran, N.; Monteiro, V.; Ahmadi, M.; Steinhart, Z.; Kushida, M.M.; Yu, H.; Coutinho, F.J.; Cavalli, F.M.G.; et al. Genome-Wide CRISPR-Cas9 Screens Expose Genetic Vulnerabilities and Mechanisms of Temozolomide Sensitivity in Glioblastoma Stem Cells. Cell Rep. 2019,27, 971–986.e9. [CrossRef] 17. Pritchett, J.; Athwal, V.; Roberts, N.; Hanley, N.A.; Hanley, K.P. Understanding the role of SOX9 in acquired diseases: Lessons from development. Trends Mol. Med. 2011,17, 166–174. [CrossRef] 18. Cheng, L.-C.; Pastrana, E.; Tavazoie, M.; Doetsch, F. miR-124 regulates adult neurogenesis in the subventricular zone stem cell niche. Nat. Neurosci. 2009,12, 399–408. [CrossRef] 19. Scott, C.E.; Wynn, S.L.; Sesay, A.; Cruz, C.; Cheung, M.; Gomez-Gaviro, M.V.; Booth, S.; Gao, B.; Cheah, K.S.E.; Lovell-Badge, R.; et al. SOX9 induces and maintains neural stem cells. Nat. Neurosci. 2010,13, 1181–1189. [CrossRef] 20. Swartling, F.J.; Savov, V.; Persson, A.I.; Chen, J.; Hackett, C.S.; Northcott, P.A.; Grimmer, M.R.; Lau, J.; Chesler, L.; Perry, A.; et al. Distinct neural stem cell populations give rise to disparate brain tumors in response to N-MYC. Cancer Cell 2012 ,21, 601–613. [CrossRef] 21. Matheu, A.; Collado, M.; Wise, C.; Manterola, L.; Cekaite, L.; Tye, A.J.; Canamero, M.; Bujanda, L.; Schedl, A.; Cheah, K.S.; et al. Oncogenicity of the developmental transcription factor Sox9. Cancer Res. 2012,72, 1301–1315. [CrossRef] [PubMed] 22. Guo, W.; Keckesova, Z.; Donaher, J.L.; Shibue, T.; Tischler, V.; Reinhardt, F.; Weinberg, R.A. Slug and Sox9 cooperatively determine the mammary stem cell state. Cell 2012,148, 1015–1028. [CrossRef] [PubMed] 23. Garros-Regulez, L.; Aldaz, P.; Arrizabalaga, O.; Moncho-Amor, V.; Carrasco-Garcia, E.; Manterola, L.; Matheu, A. mTOR inhibition decreases SOX2-SOX9 mediated glioma stem cell activity and temozolomide resistance. Expert Opin. Ther. Targets 2016 ,20, 393–405. [CrossRef] [PubMed] 24. Wang, L.; He, S.; Yuan, J.; Mao, X.-G.; Cao, Y.; Zong, J.; Tu, Y.; Zhang, Y. Oncogenic role of SOX9 expression in human malignant glioma. Med. Oncol. 2012,29, 3484–3490. [CrossRef] 25. Aldaz, P.; Otaegi-Ugartemendia, M.; Saenz-Antoñanzas, A.; García-Puga, M.; Moreno-Valladares, M.; Flores, J.M.; Gerovska, D.; Arauzo-Bravo, M.J.; Samprón, N.; Matheu, A.; et al. SOX9 promotes tumor progression through the axis BMI1-p21(CIP). Sci. Rep. 2020,10, 357. [CrossRef] Int. J. Mol. Sci. 2022,23, 4511 20 of 20 26. Sherry, M.M.; Reeves, A.; Wu, J.K.; Cochran, B.H. STAT3 is required for proliferation and maintenance of multipotency in glioblastoma stem cells. Stem Cells 2009,27, 2383–2392. [CrossRef] 27. Hubackova, S.; Krejcikova, K.; Bartek, J.; Hodny, Z. Interleukin 6 signaling regulates promyelocytic leukemia protein gene expression in human normal and cancer cells. J. Biol. Chem. 2012,287, 26702–26714. [CrossRef] 28. Martin-Martin, N.; Piva, M.; Urosevic, J.; Aldaz, P.; Sutherland, J.D.; Fernandez-Ruiz, S.; Carracedo, A. Stratification and therapeutic potential of PML in metastatic breast cancer. Nat. Commun. 2016,7, 12595. [CrossRef] 29. Brennan, C.W.; Verhaak, R.G.; McKenna, A.; Campos, B.; Noushmehr, H.; Salama, S.R.; Davidsen, T. The somatic genomic landscape of glioblastoma. Cell 2013,155, 462–477. [CrossRef] 30. Lallemand-Breitenbach, V.; Jeanne, M.; Benhenda, S.; Nasr, R.; Lei, M.; Peres, L.; de The, H. Arsenic degrades PML or PMLRARalpha through a SUMO-triggered RNF4/ubiquitin-mediated pathway. Nat. Cell Biol. 2008,10, 547–555. [CrossRef] 31. Adam, R.C.; Yang, H.; Rockowitz, S.; Larsen, S.; Nikolova, M.; Oristian, D.S.; Polak, L.; Kadaja, M.; Asare, A.; Zheng, D.; et al. Pioneer factors govern super-enhancer dynamics in stem cell plasticity and lineage choice. Nature 2015 ,521, 366–370. [CrossRef] [PubMed] 32. Wang, Z.; Xu, X.; Liu, N.; Cheng, Y.; Jin, W.; Zhang, P.; Wang, X.; Yang, H.; Liu, H.; Tu, Y. SOX9-PDK1 axis is essential for glioma stem cell self-renewal and temozolomide resistance. Oncotarget 2017,9, 192–204. [CrossRef] [PubMed] 33. Gao, J.; Zhang, J.-Y.; Li, Y.-H.; Ren, F. Decreased expression of SOX9 indicates a better prognosis and inhibits the growth of glioma cells by inducing cell cycle arrest. Int. J. Clin. Exp. Pathol. 2015,8, 10130–10138. [PubMed] 34. Rahaman, S.O.; Harbor, P.C.; Chernova, O.; Barnett, G.H.; Vogelbaum, M.A.; Haque, S.J. Inhibition of constitutively active Stat3 suppresses proliferation and induces apoptosis in glioblastoma multiforme cells. Oncogene 2002 ,21, 8404–8413. [CrossRef] [PubMed] 35. Tan, M.S.Y.; Sandanaraj, E.; Chong, Y.K.; Lim, S.W.; Koh, L.W.H.; Ng, W.H.; Tang, C. A STAT3-based gene signature stratifies glioma patients for targeted therapy. Nat. Commun. 2019,10, 3601. [CrossRef] [PubMed] 36. Birner, P.; Toumangelova-Uzeir, K.; Natchev, S.; Guentchev, M. STAT3 tyrosine phosphorylation influences survival in glioblastoma. J. Neurooncol. 2010,100, 339–343. [CrossRef] [PubMed] 37. Ito, K.; Bernardi, R.; Morotti, A.; Matsuoka, S.; Saglio, G.; Ikeda, Y.; Rosenblatt, J.; Avigan, D.E.; Teruya-Feldstein, J.; Pandolfi, P.P. PML targeting eradicates quiescent leukaemia-initiating cells. Nature 2008,453, 1072–1078. [CrossRef] 38. Carracedo, A.; Weiss, D.; Leliaert, A.K.; Bhasin, M.; de Boer, V.; Laurent, G.; Adams, A.C.; Sundvall, M.; Song, S.J.; Ito, K.; et al. A metabolic prosurvival role for PML in breast cancer. J. Clin. Investig. 2012,122, 3088–3100. [CrossRef] 39. Amodeo, V.; Deli, A.; Betts, J.; Bartesaghi, S.; Zhang, Y.; Richard-Londt, A.; Ellis, M.; Roshani, R.; Vouri, M.; Galavotti, S.; et al. A PML/Slit Axis Controls Physiological Cell Migration and Cancer Invasion in the CNS. Cell Rep. 2017,20, 411–426. [CrossRef] 40. Ashizawa, T.; Miyata, H.; Iizuka, A.; Komiyama, M.; Oshita, C.; Kume, A.; Akiyama, Y. Effect of the STAT3 inhibitor STX-0119 on the proliferation of cancer stem-like cells derived from recurrent glioblastoma. Int. J. Oncol. 2013,43, 219–227. [CrossRef] 41. Akiyama, Y.; Nonomura, C.; Ashizawa, T.; Iizuka, A.; Kondou, R.; Miyata, H.; Sugino, T.; Mitsuya, K.; Hayashi, N.; Nakasu, Y.; et al. The anti-tumor activity of the STAT3 inhibitor STX-0119 occurs via promotion of tumor-infiltrating lymphocyte accumulation in temozolomide-resistant glioblastoma cell line. Immunol. Lett. 2017,190, 20–25. [CrossRef] [PubMed] 42. Fuh, B.; Sobo, M.; Cen, L.; Josiah, D.; Hutzen, B.; Cisek, K.; Bhasin, D.; Regan, N.; Lin, L.; Chan, C.; et al. LLL-3 inhibits STAT3 activity, suppresses glioblastoma cell growth and prolongs survival in a mouse glioblastoma model. Br. J. Cancer 2009 ,100, 106–112. [CrossRef] [PubMed] 43. Jensen, K.V.; Cseh, O.; Aman, A.; Weiss, S.; Luchman, H.A. The JAK2/STAT3 inhibitor pacritinib effectively inhibits patientderived GBM brain tumor initiating cells in vitro and when used in combination with temozolomide increases survival in an orthotopic xenograft model. PLoS ONE 2017,12, e0189670. [CrossRef] [PubMed] 44. Zhou, W.; Cheng, L.; Shi, Y.; Ke, S.Q.; Huang, Z.; Fang, X.; Chu, C.-W.; Xie, Q.; Bian, X.-W.; Rich, J.N.; et al. Arsenic trioxide disrupts glioma stem cells via promoting PML degradation to inhibit tumor growth. Oncotarget 2015 ,6, 37300–37315. [CrossRef] [PubMed] 45. Verhaak, R.G.W.; Hoadley, K.A.; Purdom, E.; Wang, V.; Wilkerson, M.D.; Miller, C.R.; Ding, L.; Golub, T.; Jill, P.; Alexe, G.; et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 2010,17, 98–110. [CrossRef] [PubMed] 46. Wang, Z.; Sun, D.; Chen, Y.-J.; Xie, X.; Shi, Y.; Tabar, V.; Brennan, C.W.; Bale, T.A.; Jayewickreme, C.D.; Laks, D.R.; et al. Cell Lineage-Based Stratification for Glioblastoma. Cancer Cell 2020,38, 366–379.e8. [CrossRef] [PubMed] 47. Gangoso, E.; Southgate, B.; Bradley, L.; Rus, S.; Galvez-Cancino, F.; McGivern, N.; Güç, E.; Kapourani, C.-A.; Byron, A.; Ferguson, K.M.; et al. Glioblastomas acquire myeloid-affiliated transcriptional programs via epigenetic immunoediting to elicit immune evasion. Cell 2021,184, 2454–2470.e26. [CrossRef] 48. Capper, D.; Jones, D.T.W.; Sill, M.; Hovestadt, V.; Schrimpf, D.; Sturm, D.; Koelsche, C.; Sahm, F.; Chavez, L.; Reuss, D.E.; et al. DNA methylation-based classification of central nervous system tumours. Nature 2018,555, 469–474. [CrossRef] 49. Pollard, S.M.; Yoshikawa, K.; Clarke, I.D.; Danovi, D.; Stricker, S.; Russell, R.; Bayani, J.; Head, R.; Lee, M.; Bernstein, M.; et al. Glioma stem cell lines expanded in adherent culture have tumor-specific phenotypes and are suitable for chemical and genetic screens. Cell Stem Cell 2009,4, 568–580. [CrossRef] 50. Matheu, A.; Klatt, P.; Serrano, M. Regulation of the INK4a/ARF locus by histone deacetylase inhibitors. J. Biol. Chem. 2005 ,280, 42433–42441. [CrossRef]