Skip to main content
NCBI home page
Cancer Science logoLink to Cancer Science
. 2026 Apr 27;117(7):1841–1851. doi: 10.1111/cas.70385

MAP17 Enhances Chemoresistance and Tumorigenicity of Glioblastoma‐Initiating Cells via the Canonical NF‐кB Pathway

Shenshen Dou 1, You Lee Son 1, Toru Kondo 1,✉
PMCID: PMC13327008  PMID: 42043964

ABSTRACT

Glioblastoma (GBM) is the most common malignant brain tumor in adults. Despite multimodal treatments, including surgery, radiation therapy, and temozolomide (TMZ) chemotherapy, the median survival remains poor at approximately 15 months. One reason for the therapeutic resistance is the existence of GBM‐initiating cells (GICs) within the tumor. Therefore, understanding the molecular insights of how GICs contribute to the therapy recurrence is crucial for developing new therapeutic strategies. Comparing expression profiles of TMZ‐resistant GICs (GICRs) with those of GICs, we identified membrane‐associated protein 17 (MAP17) as a new factor that is exclusively expressed in GICRs. We show that overexpression of MAP17 in GICs significantly increased their proliferation, TMZ resistance, and tumorigenicity, whereas its knockdown impaired these properties, indicating that MAP17 plays a critical role in both TMZ resistance and tumorigenicity of GICs. We also show that MAP17 increased the expression of anti‐apoptotic protein BCL2 through the activation of RELA‐dependent NF‐κB pathway in GICs. Furthermore, we demonstrate that overexpression of BCL2 increased TMZ resistance in GICs and their tumorigenicity, while its knockdown deprived these malignant characters in GICRs. Taken together, these findings identify a novel signaling pathway, MAP17‐NF‐κB‐BCL2, that controls TMZ resistance and tumorigenicity of GICs.

Keywords: B‐cell/CLL lymphoma 2 (BCL2), glioblastoma (GBM), glioblastoma‐initiating cell (GIC), membrane‐associated protein 17 (MAP17), temozolomide (TMZ)

MAP 17 is identified as highly expressed in TMZ‐resistant glioblastoma‐initiating cells (GICs). MAP17 enhances GIC proliferation, TMZ resistance, and tumorigenicity through NF‐κB/RELA‐dependent BCL2 upregulation.

graphic file with name CAS-117-1841-g006.jpg

Abbreviations

BCL2

B‐cell/CLL lymphoma 2

GBM

glioblastoma

GICRs

temozolomide‐resistant glioblastoma‐initiating cells

GICs

glioblastoma‐initiating cells

LGG

low‐grade glioma; MAP17, membrane‐associated protein 17

NF‐κB

nuclear factor‐kappa B

NOD/SCID

nonobese diabetic/severe combined immunodeficiency

TMZ

temozolomide

WNT7B

Wnt family member 7B

1. Introduction

Glioblastoma (GBM) is the most prevalent and aggressive primary brain tumor in adults, characterized by rapid growth and diffuse invasion [1, 2]. Survival rates of GBM patients have largely remained unchanged with approximately 5% at 5 years after diagnosis, even though treated with current therapies, including surgery, radiation, and temozolomide (TMZ) chemotherapy [2, 3]. Therefore, there is a strong demand for developing new therapeutic methods for GBM.

Many studies have indicated that cellular heterogeneity in GBM contributes to therapy resistance and recurrence [4, 5] while GBM‐initiating cells (GICs), which retain the characteristics of both stem cells and cancer cells, are thought to develop heterogeneity [6, 7, 8, 9, 10]. For instance, Kondo et al. proposed that heterogeneity of GICs generates distinct stem‐like subpopulations with variable sensitivity to therapy [4]. Consistently, Wu et al. revealed that single‐cell RNA sequencing of recurrent GBM identifies multiple therapy‐resistant stem‐cell‐like cell states [5], supporting a critical role of cellular heterogeneity in therapeutic resistance and recurrence. Therefore, it is essential to characterize GICs and identify factors/pathways specific to GICs to develop new therapeutic strategies against GBM.

Many mechanisms have been shown to be involved in TMZ resistance, including direct DNA demethylation by the DNA repair enzyme O6‐methylguanine‐DNA methyltransferase (MGMT), the rapid repair of N7 and N3‐methylated bases through the base excision repair (BER) pathway, and deficiencies in the mismatch repair (MMR) system that allow tumor cells to evade lethal DNA damage [11]. Independently, we previously established TMZ‐resistant GIC lines (GICRs) for identifying new chemicals, which specifically eliminated GICR [12, 13], and demonstrated two pyrimidine synthesis inhibitors, 1‐(3‐C‐ethynyl‐b‐D‐ribopentofuranosyl) uracil (EUrd) and 10580, induced cell death in GICRs in vitro and prevented GIC tumorigenesis. However, administration of these chemicals showed side effects in GIC tumor‐bearing mice. Therefore, it was needed to further characterize and identify new molecules/pathways that specifically act in GICRs.

To investigate molecular determinants associated with TMZ resistance in GICs, we reanalyzed expression profiles of GICs and GICRs, and identified the membrane‐associated protein 17 (MAP17, also known as PDZ domain‐containing 1‐interacting protein1, PDZK1IP1) as a potential candidate. MAP17 has been reported to be upregulated in multiple tumor types [14, 15, 16, 17], localized in the plasma membrane and the Golgi apparatus [18, 19], interacts with various PDZ domain‐containing proteins through its C‐terminal PDZ‐binding domain, and facilitates their transport across the plasma membrane [15, 19, 20]. MAP17 expression is restricted to the epithelial cells, such as the proximal tubular cells of the kidney and epidermal keratinocytes [15, 19, 20], while it increased in various cancers, including epithelial carcinomas, sarcomas, lymphomas, and GBM [15, 16, 17, 18, 21]. It was also shown that increased expression of MAP17 is frequently associated with aggressive tumor phenotypes and poor clinical outcomes [16, 20, 21, 22]. Nonetheless, there is no report showing the relationship between MAP17 and TMZ resistance.

Here, we investigated the functional relevance of MAP17 expression in GICRs. We found that MAP17 promotes TMZ resistance and tumorigenicity of GICs by activating the RELA‐dependent canonical NF‐κB pathway, leading to upregulation of the anti‐apoptotic protein B‐cell/CLL lymphoma 2 (BCL2). We further identified that MAP17‐mediated NF‐κB activation occurs through its interaction with RELA in GICs. These findings revealed a novel mechanism by which MAP17 drives GIC tumorigenicity and TMZ resistance.

2. Material and Methods

2.1. Animals and Chemical Reagents

The mouse experiments were performed according to protocols approved by the Animal Care and Use Committees of Hokkaido University. The nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice were obtained from CLEA Japan Inc. The chemicals and growth factors were purchased from Invitrogen and PeproTech, respectively, unless otherwise indicated.

2.2. Human Brain Tumors

Human glioma samples, GBM, diffuse midline glioma (DMG, WHO grade IV), desmoplastic infantile ganglioglioma (DIG, WHO grade I), pilocytic astrocytoma (PA, WHO grade I), and anaplastic oligodendroglioma (AO, WHO Grade III), were obtained from Hokkaido University Hospital with patients' consent, according to the guidelines of the Research Ethics Committee. Tumor samples, approximately 0.5 cm3, were washed twice with PBS and used for RT‐PCR.

2.3. Cell Culture

Human GICs, E3, E6, and E16, and their GICR lines, E3R, E6R and E16R, were cultured as previously described [23]. Due to contamination in the E3 culture, we conducted all of the experiments using the E6/E6R and E16/E16R pairs. The MTT assay was performed as previously described [12]. For MTT assay, the indicated n values in the figure legends represent technical replicates per condition within each experiment. Each experiment was independently repeated multiple times to ensure reproducibility.

2.4. Vector Construction

Vectors have been constructed as previously described [12, 13]. The detailed method is described in Data S1.

2.5. Tumorigenesis

A tumor‐bearing mouse test was performed as previously described [12, 13]. The detailed method is described in Data S1.

2.6. Immunochemistry

Tumors were sectioned and immunostained as described previously [12, 13]. Detailed information of antibodies is described in Data S1.

2.7. Gene Expression Based on GEPIA Analysis

Human GBM and lower‐grade glioma (LGG) data for MAP17 were obtained from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/), and normal brain tissue data were obtained from The Genotype‐Tissue Expression (GTEx, https://www.gtexportal.org/). Detailed information of the analysis is described in the Data S1.

2.8. Assessment of NF‐κB Activation

To monitor NF‐κB signal transduction, cells were transfected with the pGreenFire1‐NF‐κB Lentivector (System Biosciences), which expresses a destabilized copepod GFP under the control of NF‐κB response elements (NF‐kB‐RE). NF‐κB activity was quantified by measuring the mean fluorescence intensity of GFP signal using ImageJ software. Cells were also treated with the NF‐κB inhibitor pterostilbene [24] (Tokyo Chemical Industry) for 48 h at the concentrations indicated in the figure legend.

2.9. Gene Expression Analysis

RNA‐sequencing (RNA‐seq) analysis and PCR‐based analysis were performed as described previously [25, 26]. The detailed method is described in Data S1.

2.10. Protein Analysis

Western blotting analysis was performed as previously described [27]. The detailed method is described in Data S1.

2.11. Statistical Analysis

Data are presented as means ± SD values. Statistical significance was assessed using GraphPad Prism 9.5.1 (GraphPad). Data were analyzed by a two‐tailed Mann–Whitney U test or unpaired two‐tailed t‐test for two‐group comparison, and one‐way ANOVA with Tukey's post hoc test or two‐way ANOVA with Sidak's post hoc test for multiple‐group comparison, as indicated in the figure legends.

3. Results

3.1. Identification of MAP17 as a TMZ Resistance Factor

To identify factors associated with TMZ resistance, we re‐analyzed expression profiles of GICs and GICRs [12] (Figure S1). In our analysis of genes upregulated in GICRs, we concentrated on the membrane protein MAP17 and the secretion factor Wnt family member 7B (WNT7B). These elements are particularly suitable for therapeutic targeting using established strategies, including the application of specific antibodies [28, 29, 30, 31]. In addition, MAP17 has been shown to activate the AKT/PI3K pathway, which prevents drug‐induced cell death by increasing reactive oxygen species [22, 32], while WNT7B has been shown to promote TMZ resistance [33].

We first examined whether WNT7B increases TMZ resistance in GICs. E6 was cultured in the presence of TMZ at various concentrations of recombinant WNT7B, and cell viability was examined using MTT assay. However, there was little protective effect of WNT7B against TMZ toxicity in GICs (Figure S2A). High concentrations of WNT7B did not increase the viability of E6, even in the presence of low concentration of TMZ (Figure S2B). Together, these data suggested that WNT7B cannot induce TMZ resistance in E6.

To examine whether MAP17 is involved in TMZ‐resistance, we examined MAP17 expression in GICs and GICRs. Quantitative PCR analysis revealed that MAP17 expression was higher in GICRs than in their corresponding parental GICs (Figure 1A). We confirmed that MAP17 was nearly undetectable in GICs but significantly increased following TMZ treatment (Figure 1B). As MAP17 expression increased in GICRs in a TMZ concentration‐dependent manner, these data suggest that TMZ resistance is strongly correlated with MAP17 expression.

FIGURE 1.

FIGURE 1

Open in a new tab

MAP17 expression is elevated in TMZ‐resistant GICs and correlates with poor prognosis in GBM patients. (A) MAP17 expression in GICs (E6 and E16) and GICRs (E6R and E16R). The gene expression levels relative to 18S are plotted by setting GICs to 1. n = 3 for each group. (B) MAP17 expression in GICs following treatment with increasing TMZ concentrations for 48 h. The gene expression levels relative to 18S are plotted by setting 0 μM TMZ to 1. n = 4 for each group. (C) MAP17 expression levels in human glioma samples. MAP17 mRNA levels were measured by qPCR and normalized to 18S. Each bar represents an independent tumor sample. Glioma subtypes included DMG, DIG, PA, AO, and GBM. Statistical analysis between non‐GBM (n = 11) and GBM (n = 18) samples was performed using a two‐tailed Mann–Whitney U test (**p = 0.0019**). (D) Exploratory analysis of MAP17 expression in GBM (n = 120 from TCGA), LGG (n = 290 from TCGA), and normal samples (n = 207 from GTEx) analyzed using GEPIA. (E) Exploratory Kaplan–Meier overall survival analysis of TCGA‐GBM cases stratified by MAP17 expression using GEPIA2; only cases with available survival information were included (n = 81 per group). (F) Representative immunofluorescence images showing co‐expression of MAP17 and NESTIN in human GBM samples. Three independent primary human GBM specimens (T982, T974, and T980) were immunostained for MAP17 (red) and NESTIN (green), with nuclei counterstained using Hoechst 33342 (blue). Arrows and arrowheads indicate MAP17 signals in the nucleus and cytoplasm, respectively. Quantification of MAP17 and NESTIN‐double positive cells (MAP17+/NESTIN+) is indicated. Scale bars, 100 μm. Error bars, ±SD. p < 0.05 by t‐test (A and D) and Log‐rank test (E). Different letters indicate statistically significant differences, p < 0.05 by one‐way ANOVA (B).

To investigate the clinical relevance of MAP17 in GBM, we assessed its expression in glioma specimens. We observed that MAP17 levels were significantly elevated in GBM compared to non‐GBM samples, DMG, DIG, PA, and AO (Figure 1C). Consistent with these findings, analysis of the TCGA dataset revealed significantly higher MAP17 expression in GBM relative to normal brain tissue, whereas no significant difference was observed in comparison with that in LGG (Figure 1D). In addition, patients with high MAP17 expression exhibited significantly worse prognosis, and their median survival time (1233 days) is shorter than that (2618 days) in patients with low MAP17 expression (Figure 1E). Immunohistochemical analysis revealed that approximately half of NESTIN+ cells in GBM were positive for MAP17 (Figure  1 F ), suggesting that a subset of GICs express MAP17 in GBM.

3.2. MAP17 Enhances Proliferation, TMZ Resistance, and Tumorigenicity in GICs

To characterize the functional impact of MAP17 on GICs, we overexpressed MAP17 in GICs because they have very low levels of protein. On the contrary, we knocked down MAP17 in GICRs because they have high levels of protein. We first established MAP17‐overexpressing GICs (MAP17‐GICs, Figure S3A) and confirmed overexpression at both the mRNA and protein levels. During the experiment, we observed that MAP17 overexpression increased proliferation of the cells. To quantify cell proliferation, we performed an MTT assay and confirmed that MAP17 overexpression enhanced cell proliferation approximately 1.5 and 1.8 times in E6 and E16, respectively (Figure 2A,B).

FIGURE 2.

FIGURE 2

Open in a new tab

MAP17 overexpression in GICs promotes cell proliferation, TMZ resistance, and tumorigenicity. (A,B) Cell proliferation measured by MTT assay in GICs and MAP17‐GICs. n = 3 for each group. (C,D) Cell viability measured by MTT assay in GICs (C) and MAP17‐GICs (D) after TMZ treatment for 48 h. n = 3 for each group. (E,F) Representative images of CC3 (green) in control GICs and MAP17‐GICs  following TMZ treatment for 48 h, with nuclei counterstained using Hoechst 33342 (blue). Quantification of CC3‐positive (CC3+) cells is shown in the right panel. n = 3 for each group. Scale bars, 50 μm. (G,H) H&E staining of brain tumor sections derived from mice implanted with GICs or MAP17‐GICs (left), and quantification of tumor area percentage (right). n = 4 for E6, MAP17‐E6, and MAP17‐E16 groups; n = 6 for E16 group. Scale bars, 0.5 mm. Error bars, ±SD. p < 0.05, p < 0.01, p < 0.001, p < 0.0001 by t‐test (A, B, E, F) and two‐way ANOVA (C, D).

To examine the TMZ resistance of MAP17‐GICs, we culture the cells in the presence of various concentrations of TMZ and then examined their survival using the MTT assay. As shown in Figure 2C,D, MAP17 overexpression exerted TMZ resistance in GICs. The half maximum inhibitory concentration (IC50) is approximately 838 μM and 399 μM in MAP17‐E6 and MAP17‐E16, respectively, whereas those in E6 and E16 were 381 μM and 281 μM, respectively. TMZ resistance of MAP17‐GICs was confirmed by the decreased number of cleaved Caspase 3 (CC3)‐positive cells (Figure 2E,F).

We then assessed the tumorigenicity of MAP17‐GICs. Five weeks after transplantation, we prepared brain sections, stained them with H&E, and found that MAP17‐GICs formed larger tumors, approximately twice the size of the control GICs (Figure 2G,H).

To confirm the above results, we established MAP17‐knockdown GICRs (shMAP17‐GICRs, Figure S3B), in which knockdown efficiency was confirmed by RT‐PCR. We then examined their proliferation, TMZ resistance, and tumorigenicity. As expected, MAP17 knockdown decreased GICR proliferation (Figure 3A,B) and resistance to TMZ (Figure 3C,D). The IC50 values in shMAP17‐E6R and shMAP17‐E16R were 235 μM and 219 μM, whereas those in E6R and E16R were 600 μM and 1286 μM, respectively. Furthermore, MAP17 knockdown significantly prevented GIC tumorigenesis in the brain (Figure 3E,F). Together, these data indicate that MAP17 is a crucial factor in TMZ resistance and tumorigenicity in GICs.

FIGURE 3.

FIGURE 3

Open in a new tab

MAP17 knockdown in GICRs reduces cell proliferation, TMZ resistance, and tumorigenicity. (A,B) Cell proliferation measured by MTT assay in GICRs or shMAP17‐GICRs. n = 3 for each group. (C,D) Cell viability measured by MTT assay in indicated cells after TMZ treatment for 48 h. n = 3 for each group. (E,F) H&E staining of brain tumor sections derived from mice implanted with GICRs or shMAP17‐GICRs (left), and quantification of tumor area percentage (right). n = 3 for each group. Scale bars, 0.5 mm. Error bars, ±SD. p < 0.05, p < 0.01, p < 0.001 by t‐test (A, B, E, F) and two‐way ANOVA (C, D).

3.3. MAP17‐Induced TMZ Resistance and Tumorigenicity Is Mediated by BCL2 Upregulation

To elucidate the molecular mechanisms underlying MAP17‐induced TMZ resistance, we compared RNA‐seq‐based gene expression profiles of MAP17‐GICs with those of GICs and found that BCL2, a well‐known anti‐apoptotic factor, was the top gene induced in MAP17‐GICs (Figure S4). We also found that BCL2 strongly increased in E6R, as summarized in Figure S4 (BCL2 row in the Log2FC(E6R/E6) column). There is much evidence showing that BCL2 acts as a major resistance factor against chemotherapy and radiotherapy in multiple cancer types [34, 35, 36].

After confirming that MAP17‐GICs increased BCL2 expression by qPCR and western blotting (Figure 4A,B, respectively), we established BCL2‐overexpressing GICs (BCL2‐GICs, Figure S5A,B). To verify that BCL2 is the main contributor to TMZ resistance in GICs, we cultured BCL2‐GICs and GICs in the presence of various concentrations of TMZ and examined their survival using the MTT assay. As shown in Figure 4C,D, BCL2‐GICs were strongly resistant to TMZ: the IC50 of BCL2‐E6 and ‐E16 were 390 μM and 988 μM of TMZ, whereas those of E6 and E16 were 161 μM and 530 μM, respectively. We also examined the tumorigenicity of BCL2‐GICs and observed that BCL2‐GICs formed larger tumors than controls in mouse brains (Figure 4E,F).

FIGURE 4.

FIGURE 4

Open in a new tab

BCL2 is upregulated in MAP17‐GICs, and its overexpression in GICs promotes TMZ resistance and tumorigenicity. (A) BCL2 mRNA expression in GICs and MAP17‐GICs. The expression levels relative to 18S are plotted by setting GICs to 1. n = 4 for E6 and n = 3 for E16. (B) Increased BCL2 protein in MAP17‐GICs. Representative Western blot images are shown on the left, and quantification normalized to β‐actin (loading control) is shown on the right. n = 3 for each group. (C,D) Cell viability measured by MTT assay in control GICs and BCL2‐GICs after TMZ treatment for 48 h. n = 4 for E6 groups (C) and n = 3 for E16 groups (D). (E,F) H&E staining of brain tumor sections derived from mice implanted with GICs or BCL2‐GICs (left), and quantification of tumor area percentage (right). n = 3 for each group. Scale bars, 0.5 mm. Error bars, ±SD. p < 0.05, p < 0.01, p < 0.001, p < 0.0001 by t‐test (A, B, E, F) and two‐way ANOVA (C, D).

To further validate the role of BCL2 in TMZ resistance, we knocked down BCL2 in MAP17‐GICs (shBCL2‐MAP17‐GICs, Figure S5C) and examined their TMZ resistance and tumorigenicity. As shown in Figure 5A,B, BCL2 knockdown deprived TMZ resistance in MAP17‐GICs. Unexpectedly, we observed that BCL2 knockdown markedly diminished the tumorigenesis of MAP17‐GICs (Figure 5C,D), suggesting that BCL2 is a critical factor for GIC tumorigenesis.

FIGURE 5.

FIGURE 5

Open in a new tab

BCL2 knockdown in MAP17‐GICs abolishes MAP17‐induced TMZ resistance and tumorigenicity. (A,B) Cell viability measured by MTT assay in shBCL2‐MAP17‐GICs or shNC‐MAP17‐GICs (negative control) after TMZ treatment for 48 h. n = 4 for each group. (C,D) H&E staining of brain tumor sections derived from mice implanted with shBCL2‐MAP17‐GICs or shNC‐MAP17 (left), and quantification of tumor area percentage (right). n = 5 for each group. Scale bars, 0.5 mm. Error bars, ±SD. p < 0.05, p < 0.01, p < 0.001, p < 0.0001 by two‐way ANOVA (A, B) and t‐test (C, D).

To examine the role of BCL2 in GIC tumorigenesis, we established shBCL2‐GICs (Figure S5D), although BCL2 expression in GICs is relatively low. We found that whereas the proliferation of shBCL2‐GICs was similar to that of shNC‐GICs (Figure S6A,B), their tumorigenicity was strongly prevented in the brain (Figure S6C,D): tumor size of shNC‐E6, shNC‐E16, shBCL2‐E6, and shBCL2‐E16 was 8.5%, 6.2%, 1.6%, and 1.8%, respectively. Immunostaining revealed increased BCL2 expression in GICs‐derived tumors, indicating that GICs began to express BCL2 during tumorigenesis (Figure S6E–G). Together, these results indicate that BCL2 is dispensable for TMZ resistance in GICs but contributes to their tumorigenesis.

3.4. MAP17 Induced BCL2 Expression Through the Canonical NF‐κB Pathway in GICs

We finally addressed how MAP17 induces BCL2 expression in GICs. It is well known that the NF‐κB signaling pathway activates BCL2 expression by directly binding the RELA‐containing transcription factor complex to the promoter [36, 37, 38, 39]. To examine whether MAP17 activates the NF‐κB signaling pathway in GICs, we used an NF‐κB reporter assay that induces GFP expression dependent on the activation of the NF‐κB signaling pathway. As shown in Figure 6A, GFP expression was significantly higher in MAP17‐GICs than in GICs. Using an NF‐κB inhibitor pterostilbene [24], we confirmed that BCL2 expression in MAP17‐GICs decreased in a concentration‐dependent manner (Figure 6B). There was no effect on BCL2 expression in GICs treated with pterostilbene (Figure S7A). These data indicated that BCL2 expression is regulated by the NF‐κB signaling pathway in GICs.

FIGURE 6.

FIGURE 6

Open in a new tab

MAP17 induces BCL2 expression via NF‐κB activation, potentially through interaction with the RELA subunit. (A) NF‐κB‐RE‐driven GFP activity in GICs and MAP17‐GICs. Cells were transfected with an NF‐κB‐RE‐GFP reporter vector, and GFP signals were quantified 48 h post‐transfection. n = 3 for each group. (B) BCL2 expression in MAP17‐GICs following treatment with pterostilbene for 48 h, measured by RT‐PCR. GAPDH was used as an internal control. n = 3 for each group. Representative RT‐PCR images are shown below. MAP17‐E16 cells were not analyzed at pterostilbene concentrations of 40 μM or higher due to markedly reduced cell viability. (C,D) Effect of shRELA or shRELB on BCL2 expression in MAP17‐GICs. n = 3 for each group. (E) Co‐immunoprecipitation (IP) of MAP17 and NF‐κB/RELA subunit. GICs were transfected with expression plasmids of empty or FLAG‐tagged MAP17 as indicated. Cell lysates were immunoprecipitated with anti‐FLAG antibody (IP:FLAG) and control IgG (IP:IgG), followed by Western blot with antibodies against FLAG and RELA. Cell lysates used in IP are shown in upper panels (INPUT). (F) Schematic of the MAP17‐NF‐κB/RELA‐BCL2 axis in GICs. Error bars, ±SD. p < 0.05 by one‐way ANOVA (B). p < 0.05, p < 0.01 by t‐test (A, C). Different letters indicate statistically significant differences (B).

As both canonical (RELA) and non‐canonical (RELB) NF‐κB pathways have been reported to contribute to anti‐apoptotic responses [37, 38, 39], we examined which pathway is involved in MAP17‐dependent BCL2 induction in GICs. We established MAP17‐GICs expressing either shRELA or shRELB (Figure S7B) and examined their BCL2 expression. We observed that shRELA, but not shRELB, significantly reduced BCL2 expression in MAP17‐GICs (Figure 6C,D), indicating that MAP17 primarily induces BCL2 expression through the canonical pathway in GICs.

Because the PDZ‐domain is known to act as an interactive module for binding with many signaling factors including RELA [40, 41, 42, 43, 44], we examined whether MAP17 directly associates with RELA. Co‐immunoprecipitation assay revealed that FLAG‐tagged MAP17 was physically associated with endogenous RELA in GICs (Figure 6 E and Figure S7C), suggesting that MAP17 directly activates the canonical NF‐κB pathway.

Although BCL2 expression was undetectable in cultured GICs, we found that the cells acquired its expression during tumorigenesis (Figure S6E), suggesting that an unknown factor/mechanism prevents the canonical pathway. Since RELB acts as a competitor for the RELA pathway [45], we addressed whether RELB knockdown induces BCL2 expression in GICs. We found that while RELA knockdown did not alter BCL2 expression, RELB knockdown increased BCL2 expression in GICs (Figure S7D,E), indicating that RELB prevents RELA‐dependent BCL2 expression in GICs in culture [46].

4. Discussion

MAP17 is known to be diffusely expressed in multiple cancers and is associated with poor prognosis [16, 20, 21, 22]; however, its functional role in GBM recurrence and therapy resistance remains unexplored. In this study, we identified MAP17 as a novel regulator of TMZ resistance and tumorigenicity in GICs. MAP17 overexpression significantly enhanced cell proliferation, TMZ resistance, and tumorigenicity in GICs, whereas knockdown markedly suppressed these malignant phenotypes. We identified BCL2 as a key downstream effector of MAP17, as its knockdown effectively abolished MAP17‐driven phenotypes. Mechanistically, MAP17 directly associates with RELA, activating the canonical NF‐κB signaling pathway and inducing BCL2 expression. Together, these data reveal a novel MAP17‐NF‐κB/RELA‐BCL2 axis as a critical pathway for GIC malignancy (Figure 6F).

MAP17 has been reported to activate pro‐tumorigenic signaling pathways by regulating signaling factors through its intracellular PDZ‐domain [22, 43, 47]. For instance, MAP17 overexpression has been shown to activate the Notch pathway by sequestering Numb, increasing stemness in cancer cells [43]. Using the yeast two‐hybrid system, Kocher et al. have shown that PDZK1, the main associated factor of MAP17, interacts with cMOAT, an organic anion transporter associated with multidrug resistance, increasing the resistance to chemotherapies [47]. In addition, we demonstrated that MAP17 induces BCL2 expression via the canonical NF‐κB pathways. These findings suggest that MAP17 is a critical player in tumorigenesis and drug resistance.

We found that GICs acquire MAP17 expression during brain tumorigenesis, suggesting that microenvironmental cues, such as hypoxia, metabolic stress, and inflammatory factors, induce its expression. Indeed, The Human Gene Database GeneCards (https://www.genecards.org/cgi‐bin/carddisp.pl?gene=PDZK1IP1&keywords=PDZK1IP1) shows that the promoter contains SMAD5/SMAD4 and STAT5 binding sites, suggesting that various types of cytokines and growth factors, including BMP and Wnt, induce MAP17 expression in GICs. In addition, the promoter contains TEAD1/2/3/4 binding sites, suggesting that MAP17 is induced by the Hippo signaling pathway with YAP, a sensor of extracellular stiffness and membrane tension [48]. Thus, it is of interest to unveil the detailed molecular mechanism regulating MAP17 expression during tumorigenesis.

Distinct transcriptional subtypes of GBM such as pro‐neural, neural, classical, and mesenchymal display variable molecular signatures and treatment responses [49, 50]. Since our GICs, E6 and E16, exhibit pro‐neural and mesenchymal characteristics, respectively [51], there is a limitation to generalize our findings to other GBM subtypes. However, both GICs shared a common MAP17‐NF‐κB/RELA‐BCL2 axis; they displayed differential responses in the extent of BCL2 upregulation induced by MAP17, in their intrinsic TMZ resistance, and in their sensitivity to NF‐κB inhibition. Therefore, future investigations should address whether these differences influence the therapeutic efficacy of targeting MAP17 and whether MAP17‐driven BCL2 regulation exerts a common effect across other GBM subtypes.

A recent study in pancreatic cancer demonstrated that high MAP17 expression is associated with poor prognosis and therapy resistance, while also revealing that MAP17‐high tumors harbor specific vulnerabilities that can be exploited through combination treatment strategies [21]. Together with our mechanistic identification of a MAP17‐NF‐κB/RELA‐BCL2 axis in TMZ‐resistant GICs, these findings suggest that MAP17‐based stratification may represent a practical strategy to identify targetable vulnerabilities and inform combination treatment approaches in therapy‐resistant GBM.

In conclusion, this study identifies MAP17 as a critical regulator of chemoresistance and tumorigenicity in GICs. We demonstrate that MAP17 promotes TMZ resistance and malignant phenotypes through activation of a canonical NF‐κB/RELA‐BCL2 signaling axis, providing mechanistic insight into therapy‐resistant tumor states in GBM. Although direct targeting of MAP17 remains challenging, our results, together with emerging evidence from other cancer types, suggest the potential of MAP17 expression as a stratification marker to guide the development of rational combination therapeutic strategies across MAP17‐high tumors.

Author Contributions

Shenshen Dou: formal analysis, investigation, visualization, writing – original draft. You Lee Son: data curation, methodology, resources, validation, visualization, writing – original draft, writing – review and editing. Toru Kondo: conceptualization, funding acquisition, methodology, project administration, resources, supervision.

Funding

This work was supported in part by AMED Practical Research for Innovative Cancer Control Grant No. 17ck0106236h0002 (TK); JSPS KAKENHI Grant‐in‐Aid for Scientific Research Grant No. 20H03559 (TK); the Joint Research Program of the Institute for Genetic Medicine Hokkaido University (TK); JSPS KAKENHI Grant‐in‐Aid for Scientific Research Grant No. 21K06240 and 24K09511 (YLS).

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: cas70385‐sup‐0001‐Supinfo.docx.

CAS-117-1841-s002.docx (30.9KB, docx)

Figure. S1: Top 50 upregulated genes in TMZ‐resistant GICs.

Figure. S2. Evaluation of the effect of WNT7B on TMZ response in GICs.

Figure. S3. Validation of MAP17 overexpression in GICs and MAP17 knockdown in GICRs.

Figure. S4. Top 30 genes upregulated by MAP17 overexpression in GICs.

Figure. S5. Validation of BCL2 overexpression in GICs and BCL2 knockdown in MAP17‐GICs and parental GICs.

Figure. S6. Evaluation of BCL2 function in parental GICs in vitro and in vivo.

Figure. S7. NF‐κB subunit‐specific regulation of BCL2 in parental GICs.

CAS-117-1841-s001.docx (1.6MB, docx)

Acknowledgments

We thank Jiahui Hou for assistance with mouse experiments, Daisuke Yamashita for providing GICs, and Shigeru Yamaguchi for human glioma specimen.

Data Availability Statement

The authors have nothing to report.

References

  • 1. Omuro A. and DeAngelis L. M., “Glioblastoma and Other Malignant Gliomas: A Clinical Review,” Journal of the American Medical Association 310 (2013): 1842–1850. [DOI] [PubMed] [Google Scholar]
  • 2. Stupp R., Mason W. P., van den Bent M. J., et al., “Radiotherapy Plus Concomitant and Adjuvant Temozolomide for Glioblastoma,” New England Journal of Medicine 352 (2005): 987–996. [DOI] [PubMed] [Google Scholar]
  • 3. Stupp R., Hegi M. E., Mason W. P., et al., “Effects of Radiotherapy With Concomitant and Adjuvant Temozolomide Versus Radiotherapy Alone on Survival in Glioblastoma in a Randomised Phase III Study: 5‐Year Analysis of the EORTC‐NCIC Trial,” Lancet Oncology 10 (2009): 459–466. [DOI] [PubMed] [Google Scholar]
  • 4. Kondo T., “Glioblastoma‐Initiating Cell Heterogeneity Generated by the Cell‐Of‐Origin, Genetic/Epigenetic Mutation and Microenvironment,” Seminars in Cancer Biology 82 (2022): 176–183. [DOI] [PubMed] [Google Scholar]
  • 5. Wu H., Guo C., Wang C., et al., “Single‐Cell RNA Sequencing Reveals Tumor Heterogeneity, Microenvironment, and Drug‐Resistance Mechanisms of Recurrent Glioblastoma,” Cancer Science 114, no. 6 (2023): 2609–2621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Singh S. K., Clarke I. D., Terasaki M., et al., “Identification of a Cancer Stem Cell in Human Brain Tumors,” Cancer Research 63 (2003): 5821–5828. [PubMed] [Google Scholar]
  • 7. Singh S. K., Clarke I. D., Hide T., and Dirks P. B., “Cancer Stem Cells in Nervous System Tumors,” Oncogene 23 (2004): 7267–7273. [DOI] [PubMed] [Google Scholar]
  • 8. Kondo T., “Brain Cancer Stem‐Like Cells,” European Journal of Cancer 42 (2006): 1237–1242. [DOI] [PubMed] [Google Scholar]
  • 9. Vescovi A. L., Galli R., and Reynolds B. A., “Brain Tumour Stem Cells,” Nature Reviews. Cancer 6 (2006): 425–436. [DOI] [PubMed] [Google Scholar]
  • 10. Liu G., Yuan X., Zeng Z., et al., “Analysis of Gene Expression and Chemoresistance of CD133+ Cancer Stem Cells in Glioblastoma,” Molecular Cancer 5 (2006): 67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Singh N., Miner A., Hennis L., and Mittal S., “Mechanisms of Temozolomide Resistance in Glioblastoma: A Comprehensive Review,” Cancer Drug Resist 4, no. 1 (2021): 17–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Tsukamoto Y., Ohtsu N., Echizenya S., et al., “Chemical Screening Identifies EUrd as a Novel Inhibitor Against Temozolomide‐Resistant Glioblastoma‐Initiating Cells,” Stem Cells 34 (2016): 2016–2025. [DOI] [PubMed] [Google Scholar]
  • 13. Echizenya S., Ishii Y., Kitazawa S., et al., “Discovery of a New Pyrimidine Synthesis Inhibitor Eradicating Glioblastoma‐Initiating Cells,” Neuro‐Oncology 22 (2020): 229–239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Garcia‐Heredia J. M. and Carnero A., “Dr. Jekyll and Mr. Hyde: MAP17's Up‐Regulation, a Crosspoint in Cancer and Inflammatory Diseases,” Molecular Cancer 17 (2018): 80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Kocher O., Comella N., Gilchrist A., et al., “PDZK1, a Novel PDZ Domain‐Containing Protein Up‐Regulated in Carcinomas and Mapped to Chromosome 1q21, Interacts With cMOAT (MRP2), the Multidrug Resistance‐Associated Protein,” Laboratory Investigation 79 (1999): 1161–1170. [PubMed] [Google Scholar]
  • 16. Guijarro M. V., Leal J. F., and Fominaya J., “MAP17 Overexpression Is a Common Characteristic of Carcinomas,” Carcinogenesis 28 (2007): 1646–1652. [DOI] [PubMed] [Google Scholar]
  • 17. Garcia‐Heredia J. M., Otero‐Albiol D., Perez M., Perez‐Castejon E., Munoz‐Galvan S., and Carnero A., “Breast Tumor Cells Promotes the Horizontal Propagation of EMT, Stemness, and Metastasis by Transferring the MAP17 Protein Between Subsets of Neoplastic Cells,” Oncogene 9 (2020): 96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Jaeger C., Schaefer B. M., Wallich R., and Kramer M. D., “The Membrane‐Associated Protein pKe#192/MAP17 in Human Keratinocytes,” Journal of Investigative Dermatology 115 (2000): 375–380. [DOI] [PubMed] [Google Scholar]
  • 19. Lanaspa M. A., Giral H., Breusegem S. Y., et al., “Interaction of MAP17 With NHERF3/4 Induces Translocation of the Renal Na/Pi IIa Transporter to the Trans‐Golgi,” American Journal of Physiology. Renal Physiology 292 (2007): 230–242. [DOI] [PubMed] [Google Scholar]
  • 20. Perez M., Praena‐Fernandez J. M., Felipe‐Abrio B., et al., “MAP17 and SGLT1 Protein Expression Levels as Prognostic Markers for Cervical Tumor Patient Survival,” PLoS One 8 (2013): e56169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Verdugo‐Sivianes E. M., Martinez‐Perez J., Navas L. E., Saez C., and Carnero A., “Efficacy of NAMPT Inhibitors in Pancreatic Cancer After Stratification by MAP17 (PDZK1IP1) Levels,” Cancers (Basel) 17 (2025): 17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Carnero A., “MAP17, a ROS‐dependent oncogene,” Frontiers in Oncology 2 (2012): 112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Yamashita D., Kondo T., Ohue S., et al., “miR‐340 Suppresses the Stem‐Like Cell Function of Glioma‐Initiating Cells by Targeting Tissue Plasminogen Activator,” Cancer Research 75 (2015): 1123–1133. [DOI] [PubMed] [Google Scholar]
  • 24. Priego S., Feddi F., Ferrer P., et al., “Natural Polyphenols Facilitate Elimination of HT‐29 Colorectal Cancer Xenografts by Chemoradiotherapy: A Bcl‐2‐ and Superoxide Dismutase 2‐Dependent Mechanism,” Molecular Cancer Therapeutics 7 (2008): 3330–3342. [DOI] [PubMed] [Google Scholar]
  • 25. Kondo T. and Raff M., “The Id4 HLH Protein and the Timing of Oligodendrocyte Differentiaiton,” EMBO Journal 19 (2000): 1998–2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Nakatani Y., Kiyonari H., and Kondo T., “Ecrg4 Deficiency Extends the Replicative Capacity of Neural Stem Cells in a Foxg1‐Dependent Manner,” Development 146 (2019): dev168120. [DOI] [PubMed] [Google Scholar]
  • 27. Moriguchi T., Takeda S., Iwashita S., et al., “Ecrg4 Peptide Is the Ligand of Multiple Scavenger Receptors,” Scientific Reports 8 (2018): 4048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Hou J., Uejima T., Tanaka M., et al., “EVA1‐Antibody Drug Conjugate Is a New Therapeutic Strategy for Eliminating Glioblastoma‐Initiating Cells,” Neuro‐Oncology 27 (2025): 682–694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Gilbert M. R., Dignam J. J., Armstrong T. S., et al., “A Randomized Trial of Bevacizumab for Newly Diagnosed Glioblastoma,” New England Journal of Medicine 370 (2014): 699–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Smith L. M., Nesterova A., Ryan M. C., et al., “CD133/Prominin‐1 Is a Potential Therapeutic Target for Antibody‐Drug Conjugates in Hepatocellular and Gastric Cancers,” British Journal of Cancer 99 (2008): 100–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Vora P., Venugopal C., and Salim S. K., “The Rational Development of CD133‐Targeting Immunotherapies for Glioblastoma,” Cell Stem Cell 26 (2020): 832–844. [DOI] [PubMed] [Google Scholar]
  • 32. Guijarro M. V., Leal J. F., and Fominaya J., “MAP17 Inhibits Myc‐Induced Apoptosis Through PI3K/AKT Pathway Activation,” Carcinogenesis 28 (2007): 2443–2450. [DOI] [PubMed] [Google Scholar]
  • 33. Yu T., Wang X., and Zhang Y., “EZH2 Interacts With HP1BP3 to Epigenetically Activate WNT7B That Promotes Temozolomide Resistance in Glioblastoma,” Oncogene 42 (2023): 461–470. [DOI] [PubMed] [Google Scholar]
  • 34. Youle R. J. and Strasser A., “The BCL‐2 Protein Family: Opposing Activities That Mediate Cell Death,” Nature Reviews. Molecular Cell Biology 9 (2008): 47–59. [DOI] [PubMed] [Google Scholar]
  • 35. Czabotar P. E., Lessene G., Strasser A., and Adams J. M., “Control of Apoptosis by the BCL‐2 Protein Family: Implications for Physiology and Therapy,” Nature Reviews. Molecular Cell Biology 15 (2014): 49–63. [DOI] [PubMed] [Google Scholar]
  • 36. Kaloni D., Diepstraten S. T., Strasser A., and Kelly G. L., “BCL‐2 Protein Family: Attractive Targets for Cancer Therapy,” Apoptosis 28 (2023): 20–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Catz S. D. and Johnson J. L., “Transcriptional Regulation of Bcl‐2 by Nuclear Factor Kappa B and Its Significance in Prostate Cancer,” Oncogene 20 (2001): 7342–7351. [DOI] [PubMed] [Google Scholar]
  • 38. Vogler M., Braun Y., Smith V. M., et al., “The BCL2 Family: From Apoptosis Mechanisms to New Advances in Targeted Therapy,” Signal Transduction and Targeted Therapy 10 (2025): 91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Karin M., Cao Y., Greten F. R., and Li Z. W., “NF‐kappaB in Cancer: From Innocent Bystander to Major Culprit,” Nature Reviews. Cancer 2 (2002): 301–310. [DOI] [PubMed] [Google Scholar]
  • 40. Haselager M., Thijssen R., West C., et al., “Regulation of Bcl‐XL by Non‐Canonical NF‐kappaB in the Context of CD40‐Induced Drug Resistance in CLL,” Cell Death and Differentiation 28 (2021): 1658–1668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Tomita S., Chen L., and Kawasaki Y., “Functional Studies and Distribution Define a Family of Transmembrane AMPA Receptor Regulatory Proteins,” Journal of Biological Chemistry 275 (2000): 13056–13061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Lee H. J. and Zheng J. J., “Wnt/β‐Catenin Signaling Pathway and Its Role in Cancer,” Cell Communication and Signaling 8 (2010): 8.20509869 [Google Scholar]
  • 43. Garcia‐Heredia J. M., Lucena‐Cacace A., Verdugo‐Sivianes E. M., Perez M., and Carnero A., “The Cargo Protein MAP17 (PDZK1IP1) Regulates the Cancer Stem Cell Pool Activating the Notch Pathway by Abducting NUMB,” Clinical Cancer Research 23 (2017): 3871–3883. [DOI] [PubMed] [Google Scholar]
  • 44. Garcia‐Heredia J. M. and Carnero A., “The Cargo Protein MAP17 (PDZK1IP1) Regulates the Immune Microenvironment,” Oncotarget 8 (2017): 98580–98597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Navarro H. I., Daly A. E., Rodriguez B., et al., “NF‐κB RelB Suppresses the Inflammatory Gene Expression Programs of Dendritic Cells by Competing With RelA for Binding to Target Gene Promoters,” Cell Discovery 11 (2025): 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Dong F., Li R., Wang J., et al., “Hypoxia‐Dependent Expression of MAP17 Coordinates the Warburg Effect to Tumor Growth in Hepatocellular Carcinoma,” Journal of Experimental & Clinical Cancer Research 40 (2021): 121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Kocher O., Comella N., Tognazzi K., and Brown L. F., “Identification and Partial Characterization of PDZK1: A Novel Protein Containing PDZ Interaction Domains,” Laboratory Investigation 79 (1999): 1161–1172. [PubMed] [Google Scholar]
  • 48. Cai X., Wang K. C., and Meng Z., “Mechanoregulation of YAP and TAZ in Cellular Homeostasis and Disease Progression,” Frontiers in Cell and Development Biology 9 (2021): 673599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Verhaak R. G., Hoadley K. A., Purdom E., et al., “Integrated Genomic Analysis Identifies Clinically Relevant Subtypes of Glioblastoma Characterized by Abnormalities in PDGFRA, IDH1, EGFR, and NF1,” Cancer Cell 17 (2010): 98–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Singh S., Dey D., Barik D., et al., “Glioblastoma at the Crossroads: Current Understanding and Future Therapeutic Horizons,” Signal Transduction and Targeted Therapy 10 (2025): 213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Wang Z., Zou P., Chen Z., et al., “Novel Strategy to Target Glioblastoma‐Initiating Cells Using a Braintropic Adeno‐Associated Virus Carrying a miR‐Dependent Genome‐Editing System,” British Journal of Cancer 132 (2025): 1100–1109. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: cas70385‐sup‐0001‐Supinfo.docx.

CAS-117-1841-s002.docx (30.9KB, docx)

Figure. S1: Top 50 upregulated genes in TMZ‐resistant GICs.

Figure. S2. Evaluation of the effect of WNT7B on TMZ response in GICs.

Figure. S3. Validation of MAP17 overexpression in GICs and MAP17 knockdown in GICRs.

Figure. S4. Top 30 genes upregulated by MAP17 overexpression in GICs.

Figure. S5. Validation of BCL2 overexpression in GICs and BCL2 knockdown in MAP17‐GICs and parental GICs.

Figure. S6. Evaluation of BCL2 function in parental GICs in vitro and in vivo.

Figure. S7. NF‐κB subunit‐specific regulation of BCL2 in parental GICs.

CAS-117-1841-s001.docx (1.6MB, docx)

Data Availability Statement

The authors have nothing to report.

Articles from Cancer Science are provided here courtesy of Wiley

RESOURCES