Skip to main content
NCBI home page
Neuro-Oncology logoLink to Neuro-Oncology
. 2015 Jun 1;18(1):37–47. doi: 10.1093/neuonc/nov091

USP1 targeting impedes GBM growth by inhibiting stem cell maintenance and radioresistance

Jin-Ku Lee 1,†, Nakho Chang 1,†, Yeup Yoon 1,†, Heekyoung Yang 1,†, Heejin Cho 1, Eunhee Kim 1, Yongjae Shin 1, Wonyoung Kang 1, Young Taek Oh 1, Gyeong In Mun 1, Kyeung Min Joo 1, Do-Hyun Nam 1, Jeongwu Lee 1
PMCID: PMC4677407  PMID: 26032834

Abstract

Background

Clinical benefits from standard therapies against glioblastoma (GBM) are limited in part due to intrinsic radio- and chemoresistance of GBM and inefficient targeting of GBM stem-like cells (GSCs). Novel therapeutic approaches that overcome treatment resistance and diminish stem-like properties of GBM are needed.

Methods

We determined the expression levels of ubiquitination-specific proteases (USPs) by transcriptome analysis and found that USP1 is highly expressed in GBM. Using the patient GBM-derived primary tumor cells, we inhibited USP1 by shRNA-mediated knockdown or its specific inhibitor pimozide and evaluated the effects on stem cell marker expression, proliferation, and clonogenic growth of tumor cells.

Results

USP1 was highly expressed in gliomas relative to normal brain tissues and more preferentially in GSC enrichment marker (CD133 or CD15) positive cells. USP1 positively regulated the protein stability of the ID1 and CHEK1, critical regulators of DNA damage response and stem cell maintenance. Targeting USP1 by RNA interference or treatment with a chemical USP1 inhibitor attenuated clonogenic growth and survival of GSCs and enhanced radiosensitivity of GBM cells. Finally, USP1 inhibition alone or in combination with radiation significantly prolonged the survival of tumor-bearing mice.

Conclusion

USP1-mediated protein stabilization promotes GSC maintenance and treatment resistance, thereby providing a rationale for USP1 inhibition as a potential therapeutic approach against GBM.

Keywords: deubiquitination, glioblastoma stem cells, targeted therapy

Glioblastoma (GBM) is the most lethal primary brain tumor. The current standard of care for GBM patients is maximal surgical resection followed by adjuvant concurrent radiation and the oral methylator temozolomide. Unfortunately, these therapies provide only palliation, and the prognosis for patients with GBM remains poor with a median survival of 14.6 months.1 In addition, recent trials with molecularly targeted agents including receptor tyrosine kinase inhibitors and antiangiogenesis inhibitors have been largely disappointing, with no significant impact on overall survival.2–4

GBM reveals a profound degree of inter- and intratumoral heterogeneity. Clonal diversity and cellular hierarchy likely contribute to the intratumoral heterogeneity of GBM. Accumulating evidence suggests that GBMs contain a subpopulation of highly tumorigenic stem-like cells termed glioblastoma stem-initiating cells (GSCs).5,6 While many unresolved issues regarding GSC surface markers (eg, reversibility of the stem-cell state, frequency, and cell of origin) exist, numerous studies have shown that GSCs significantly contribute to tumor propagation and recurrence, supporting the notion that GSCs are a critical target for curative therapy.7–10

Recently, a series of large-scale genomic studies have catalogued genomic alterations and deregulated signaling pathways in GBM.11–13 Since oncogenic signaling pathways appear to be redundant and complex, inactivation of a single oncogenic pathway may not be sufficient, in part due to various resistance mechanisms.14 The challenges inherent in developing more effective GBM treatments include GBM's resistance to standard treatments, its genetic complexity and molecular adaptability, and GSCs. Therefore, novel effective therapeutic options should meet the above challenges, enhance the efficacy of standard therapy, and target the signal hubs that can cripple multiple oncogenic pathways in GBM.

Protein ubiquitination regulates the degradation of proteins. It is a dynamic, 2-way process because the deubiquitinating enzymes (DUBs) can rescue the proteins from degradation. The human genome encodes ∼100 proteins with putative DUB activity.15 Although USPs are considered to be promising therapeutic targets, little is understood about the roles of USPs in GBM and the feasibility of USP targeting as an anti-GBM therapeutic approach.16 Here, we demonstrate that USP1 promotes self-renewal of GSCs and radioresistance by regulating ID1 and CHEK1, which are 2 critical regulators of stem cell maintenance, and DNA damage response. Furthermore, we show that genetic or pharmacological inhibition of USP1 impedes tumor growth and treatment resistance of GBM, providing a rationale for USP1 targeting as a potential anti-GBM therapy.

Materials and Methods

Human Glioblastoma Specimens and Derivative GSCs

Following informed consent, glioblastoma specimens were obtained from patients undergoing surgery in accordance with the Institutional Review Boards. Within hours after surgical removal, tumor specimens were enzymatically dissociated into single cells, following the procedures previously reported.17 Human neural progenitor cells (NPCs) were purchased from Lonza and cultured similarly to GSCs.

Intracranial Tumor Cell Injection

GBM cells were resuspended in 2 μL of Hank's Balanced Salt Solution (HBSS) and injected intracranially into the striatum of adult nude mice by using a stereotactic device (Kopf Instruments) (coordinates: 2 mm anterior, 2 mm lateral, 2.5 mm depth from the dura).18 All experiments followed the guidelines of the Animal Use and Care Committees.

Chemicals and Antibodies

Pimozide was purchased from Tocris. The following antibodies were used as primary antibodies: USP1 (Bethyl Laboratories; 1:1000); phospho-CHK1 (Ser296) phospho-CHK2 (Thr68), CHK1, CHK2, phospho-H2AX, PARP, caspase-3 (Cell Signaling; 1:1000); ID1, ID3 (SantaCruz); and β-actin (Sigma-Aldrich; 1:5000).

Lentivirus Production and Transduction

Lentiviral clones expressing shUSP1 or NT shRNA (SHC002) were acquired from Sigma-Aldrich. Lentiviruses were produced in 293FT cells with packaging mix (ViraPower Lentiviral Expression Systems, Invitrogen).

Neurosphere-forming Limiting Dilution Assay

GBM cells derived from GBM patient specimens and xenograft tumors were dissociated into single-cell suspensions and then plated into 96-well plates with various seeding densities (1–500 cells per well). Cells were incubated at 37°C for 2 weeks. At the time of quantification, each well was examined for formation of neurosphere-like cell aggregates.

Single-cell Gel Electrophoresis

DNA single- and double-strand breaks were determined and quantified by alkaline and neutral comet assay using the CometAssay Kit (Trevigen).

Quantitative Real-time Polymerase Chain Reaction

RNAs were extracted, and their complementary DNAs were synthesized (Invitrogen) per manufacturer’s instructions. Real-time PCR was performed on an ABI Prism 7900 sequence detection system (Applied Biosystems) according to the manufacturer's instructions using the following primers:

  • USP1     Forward: 5′-CCAATGAGAGCGGAAGGAGG-3′
    •      Reverse: 5′-CACCAATTATATCTAGACCAAAGCC-3′
  • CDKN1A (p21) Forward: 5′-TCACTGCTTGTACCCTTGTGC-3′
    •      Reverse: 5′-GGCGTTTGGAGTGGTAGAAA-3′
  • SOX2     Forward: 5′-TGCGAGCGCTGCACAT-3′
    •      Reverse: 5′-TCATGAGCGTCTTGGTTTTCC-3′
  • OLIG2    Forward: 5′-ATAGATCGACGCGACACCAG-3′
    •      Reverse: 5′-ACCCGAAAATCTGGATGCGA-3′
  • GFAP     Forward: 5′-TGTGTGAGTAAGAAGGGACCGCAA-3′
    •      Reverse: 5′-GCAGGGCATGACTTGTCCCATTT-3′
  • β-actin    Forward: 5′-AGAAAATCTGGCACCACACC-3′
    •      Reverse: 5′-AGAGGCGTACAGGGATAGCA-3′

Statistical Analysis

Data are presented as means ± standard deviation. For group comparison, 2-tailed Student t tests were conducted using Graphpad Prism software (version 5.01; GraphPad Software.)

Results

USP1 is Highly Expressed in Primary Human Glioma Tissues and GSCs

To interrogate the role of USPs in human gliomas, we first surveyed the mRNA expression levels of various USPs in glioma specimens utilizing the ONCOMINE and REMBRANDT databases.19,20 Due to the prominent expression of USP1 in gliomas relative to nontumor brain tissues, we chose to focus on USP1 (Fig. 1A). Those patients with high levels of USP1 mRNA in gliomas (> 2-fold) had significantly shorter survivals than the other patients (P < .01), and high-grade gliomas expressed high levels of USP1 mRNA (Fig. 1B and C). Consistent with mRNA data, immunoblot analysis confirmed that USP1 proteins were highly expressed in patients’ GBM specimens relative to nontumor brain tissues (Fig. 1D). Collectively, these data suggest a positive correlation between USP1 expression and glioma malignancy.

Fig. 1.

Fig. 1.

Open in a new tab

Ubiquitination-specific protease 1 (USP1) is highly expressed in primary human glioma tissues, and its expression correlates with poor survival. (A) Analysis of ONCOMINE datasets showing high expression of USP1 mRNA in GBM specimens (n = 146) compared with nontumor brain tissues (n = 35). P < .001 with unpaired t test. (B) Kaplan-Meier plot of glioma patients in the REMBRANDT database. More than 2-fold elevation of USP1 mRNA expression correlated with poor glioma patient survival (n = 70 USP1 high; n = 273 USP1 intermediate; P = .0017 vs all other samples with log-rank analysis). A positive correlation was found between USP1 mRNA levels and WHO glioma grade in the ONCOMINE database. (C) A positive correlation between USP1 mRNA levels and World Health Organization grade gliomas using the ONCOMINE database. The mRNA expression level of gliomas was represented as a fold-increase relative to that of normal brain tissue (set to 1). P < .01 by ANOVA. (D) Western blot analysis of USP1 in nontumor brain tissues and the patient-derived GBM specimens. “N” denotes nontumor sample, and the number represents the designated specimen. β-actin was used as a loading control. (E) Western blot analysis using CD133 and/or CD 15 positive/negative cell lysates from primary GBM and derivative xenograft tumors. β-actin was used as a loading control. (F) Western blot analysis of USP1 in GSC-enriched cells versus differentiated progenies. Differentiation was induced by culturing these cells in the presence of serum (10%) for 3 days. SOX2 (a GSC-specific transcription factor), and GFAP (an astroglial differentiation marker) were examined. β-actin was used as a loading control.

Because GSCs are implicated in GBM malignancy, we next determined the levels of USP1 in GSC-enriched and depleted GBM cell populations (Fig. 1E). Patient-derived GBM cells were enriched with GSC markers (CD133 or CD15) and functionally validated by clonogenic assays and in vivo tumor formation assays.17,18,21,22 Percentages of CD133- or CD15-positive cells were 6.8% in 1228 tumors, 28.0% in 211 tumors, 38.2% in 308 tumors, and 10.2% in 308 derived xenograft tumors.22 SOX2 is a master regulator for stem cell maintenance in both normal and neoplastic stem cells.17,23,24 Similar to SOX2, USP1 proteins were highly expressed in CD133- or CD15-positive GBM cells compared with CD133- or CD15-negative cells. Culture of these cells with serum decreased the level of Sox2 while increasing GFAP (an astroglial differentiation marker). During this differentiation process, USP1 expression was markedly decreased in 3 different primary GBM cells tested (Fig. 1F). Together, these data indicate that USP1 is highly expressed in GBMs, particularly in enriched GSCs.

USP1 Targeting Decreases the Survival and Clonogenic Growth of GSCs

To investigate the functional roles of USP1 in GBM, we targeted USP1 using the lentivirus-expressing shRNA directed against USP1 and determined the effects on GBM cells (Fig. 2). USP1 knockdown in patient-derived GBM cells (131 and 827) dramatically decreased the cell number, as determined by short-term proliferation assays (Fig. 2A). To determine whether cell death is associated with USP1 knockdown, we performed immunoblot assays using antibodies against representative apoptosis marker proteins. Levels of PARP cleavage and active caspase 3 were significantly increased by USP1 knockdown (Fig. 2C). We also determined the effects of USP1 knockdown in NPCs. Expression levels of USP1 protein in NPCs were lower than in GBM cells. In contrast to GBM cells, USP1 knockdown did not induce the changes in cell proliferation or cell death of NPCs (Fig. 2B and C).

Fig. 2.

Fig. 2.

Open in a new tab

Ubiquitination-specific protease 1 (USP10 targeting impedes the survival and clonogenic growth of glioblastoma (GBM) cells. (A and B) Effect of USP1 knockdown on short-term proliferation of primary GBM cells (A) and normal human neural progenitor cells (NPCs). (B) Western blot analysis of USP1 in GBM cells transduced with either nontargeting (NT) shRNA or USP1 shRNA-expressing lentivirus. Two independent USP1 shRNAs effectively decreased the levels of USP1 protein and cell survival similarly. Cell numbers in both groups were counted and plotted. Data are means ± SD (n = 3). *P < .01. (C) Western Blot analysis of PARP and CASP3 using lysates isolated from GBM cells and NPCs with or without USP1 knockdown. Cleaved PARP and CASP3 were detected only in USP1 knockdown GBM cells. β-actin was used as a loading control. (D) In vitro tumorsphere-forming limiting dilution assays using 131 GBM cells either with the control NT shRNA or USP1 shRNA. Cells were plated with varying densities per well (1–100 cells per well; 24 wells per condition) and cultured for 2 weeks. The wells not containing spheroids were counted and plotted. Representative images of cells are shown. (E) Western Blot analysis of PARP and CASP3 using lysates isolated from GBM cells and NPCs treated with pimozide (5 μM). Cells were treated with pimozide and harvested at the indicated time points (4–24 h). β-actin was used as a loading control. (F) Short-term proliferation of GBM cells and NPCs treated with pimozide. Cells were treated with the indicated concentration of pimozide. Then cell numbers were counted and plotted relative to the initial cell number. *P < .001. (G) LDAs using 131 and 827 GBM cells treated with pimozide. Cells were plated with varying densities per well (1–200 cells per well; 24 wells per condition) and cultured for 2 weeks. Pimozide was added every 3 days.

Next, we determined the effect of USP1 knockdown on clonogenic growth of GBM cells by performing in vitro limiting dilution tumorsphere formation assays (Fig. 2D). GBM cells transduced with nontargeting shRNA readily formed spheroids, even when the initial seeding cells were only 1 or <10 cells per well. In contrast, more than 100 USP1 knockdown cells per well were required to form spheroids in at least 50% of wells (n = 24 wells per condition), indicating that USP1 targeting significantly impedes GBM clonogenic growth.

As small-molecule inhibitors are generally more feasible for clinical applications than genetic knockdowns, we determined the biological effects of pharmacological USP1 inhibition. Pimozide, a drug for treating chronic schizophrenic patients, has recently become known to inhibit USP1.25,26 To determine the effects of pimozide on cell proliferation, apoptosis, and clonogenic growth, we treated various GBM cells and NPCs with pimozide and performed the corresponding assays (Fig. 2E to G). Pimozide treatment induced caspase-3 activation and PARP cleavage in GBM cells but had little effect in NPCs (Fig. 2E). In limiting dilution assays, we have used about 1/10 of concentrations that were used in short-term proliferation and apoptosis assays because low concentration of pimozide for 10–14 days of treatment was sufficient to block clonogenic growth of GBM cells. While pimozide treatment significantly inhibited GBM growth in a concentration-dependent manner, its effects on NPCs were not evident (Fig. 2F and G). These data suggest that USP1 inhibition selectively affects GBM cells but not NPCs.

USP1 Promotes ID1 Stability in Glioblatoma Cells

USP1 is known to rescue several proteins from ubiquitination-mediated protein degradation, including FANCD2, CHEK1, and ID1.27–29 Inhibiting DNA-binding (ID) proteins are transcription factors that antagonize the DNA-binding capacity of basic helix-loop-helix factors.30 IDs regulate cell cycle, cell differentiation, and self-renewal of various stem cells including GSCs.31,32 Because USP1-mediated protein stabilization depends on the cellular context, we determined whether USP1 regulates the levels of IDs in GBM.

USP1 targeting, either by shRNA-mediated knockdown or pimozide treatment, significantly reduced the level of ID1 in 3 different patient-derived GBM cells (Fig. 3A and B). ID1 reduction by pimozide treatment was evident within 4–8 hours, while ID3 level was not changed by USP1 inhibition. To test whether USP1 regulates the protein stability of ID1 in GBM cells, we treated cells with the protein synthesis inhibitor cycloheximide in combination with pimozide and determined ID1 expression by time-course immunoblot analysis. In the absence of new protein synthesis, the level of ID1 decreased to about 50% at a 2 hour time point. The combination of pimozide and cycloheximide significantly decreased the level of ID1 to <10% of the untreated control (Fig. 3C). Consistent with this, treatment with a proteasome inhibitor MG-132 increased ID1 levels in GBM cells, suggesting that ID1 protein is subject to proteasome-mediated degradation (Fig. 3D). To further confirm molecular interaction between USP1 and ID1, we performed immunoprecipitation experiments. ID1 co-immunoprecipitated with endogenous USP1 from the lysates derived from 448 and 827 GBM cells (Fig. 3E). Finally, ectopic expression of USP1 further increased ID1 levels in GBM cells (Fig. 3F). Collectively, these results suggest that USP1 positively regulates ID1 stability in GBM.

Fig. 3.

Fig. 3.

Open in a new tab

Ubiquitination-specific protease 1 (USP1) promotes ID1 stability in glioblastoma (GBM) cells. (A) Western blot analysis of ID1 and ID3 in GBM cells transduced with USP1 shRNA expressing lentivirus or treated with pimozide. β-actin was used as a loading control. (B) Immunofluorescence staining of ID1 in GBM cells either with NT shRNA or USP1 shRNA. Nuclei were labeled with DAPI. (C) Western Blot analysis of ID1 and ID3 in GSCs treated with cycloheximide (25 μg/mL media), pimozide (5 μM), or both in time-course experiments. Intensity of protein bands was quantitated by densitometry, normalized by the intensity of β-actin, and plotted. P and C represent pimozide and cycloheximide, respectively. (D) Western blot analysis of ID1 in 827 GBM cells with either nontargeting shRNA or shUSP1 in the presence or absence of MG-132 (20 μM). β-actin was used as a control. (E) Coimmunoprecipitation of USP1 and ID1 in 448 and 827 GBM cells. IgG represents a control antibody used for immunoprecipitation. (F) Western blot analysis of USP1 and ID1 after ectopic expression of USP1. V5 tagged USP1 proteins were expressed in 131 GBM cells by lentiviral transduction. Exogenous and endogenous USP1 were marked as Exo and Endo, respectively. (G) Quantitative reverse transcription PCR (RT-qPCR) analysis to determine mRNA expression of various genes in 448 and 827 GBM cells transduced with either NT shRNA or USP1 shRNA. Data are means ± SD (n = 3). *P < .01. (H) RT-qPCR analysis to determine mRNA expression of p21, Sox2, Olig2 and GFAP in USP1 overexpressing (O/E) cells. *P < .01. (I) The frequency of clonogenic cells was determined by LDA analysis and calculated by using Extreme Limiting Dilution Analysis (http:/bioinfo.wehi.edu.au/software/elda). *P < .001.

ID1 is known to increase the transcription of SOX2 while decreasing the cyclin-dependent kinase inhibitor p21WAF/CIP1 in stem-like GBM cells.33 Given the role of USP in ID1 stability, we determined expression levels of the differentiation status-associated genes in GBM cells with or without USP1 inhibition (Fig. 3G). Quantitative PCR analysis revealed that GBM cells expressing USP1 shRNA or treated with pimozide had significantly decreased levels of Sox2 and Olig2 compared with the control (Fig. 3G and Supplementary Fig. S1). In contrast, GFAP and p21 expression levels were increased by USP1 inhibition, suggesting that USP1 knockdown may be associated with loss of stem-like properties. Furthermore, ectopic expression of USP1 further increased the mRNA levels of Sox2 and Olig2 while decreasing GFAP and p21 (Fig. 3H). Finally, we determined the frequency of clonogenic cells in USP1-overexpressing GBM cells. Compared with the empty-vector expressing control cells, USP1-overexpressing GBM cells were more enriched with clonogenic cells (Fig. 3I). Together, these support the roles of USP1 in stem cell maintenance and clonogenic GBM growth.

USP1 Promotes Glioblastoma Radioresistance

Radiation treatment decreases tumor burden and prolongs the survival of GBM patients, but tumor recurrence is essentially universal. The intrinsic radioresistance of GBM has been well recognized. Preferential activation of DNA checkpoint kinases, such as CHEK1, has been implicated in GSC radioresistance.8 As USP1 was implicated in DNA repair processes through its deubiquitination of CHEK1,34 we determined the role of USP1 in GBM radiation response. First, we inhibited USP1 by shRNA-mediated knockdown and determined the expression levels of CHEK1 in GBM cells (Fig. 4A). USP1 knockdown significantly decreased the levels of total CHEK1 and phosphorylated CHEK1 in 3 different patient-derived GBM cells. Similar to ID1, we performed time-course experiments to determine CHEK1 protein stability. In the presence of protein synthesis inhibitor, USP1 inhibition by pimozide treatment accelerated CHEK1 degradation compared with the control, suggesting that USP1 regulates the stability of CHEK1 protein GBM cells. Phosphorylated gamma histone 2AX (p-γH2AX) is an indicator of damaged DNA.35 USP1 inhibition increased the levels of p-γH2AX, while levels of total γH2AX remained constant (Fig. 4A and Supplementary Fig. S2). To further confirm the induction of p-γH2AX by USP1 targeting, we examined the p-γH2AX nuclei foci formation through immunofluorescence analysis in pimozide-treated GBM cells. Similar to USP1 knockdown, pimozide treatment significantly increased the number of p-γH2AX foci compared with vehicle-treated controls (Fig. 4B). These results suggest that USP1 targeting induces persistent DNA damage in GBM cells, presumably due in part to inhibition of the DNA repair process. To further verify the association between USP1 and DNA damage responses in GBM cells, we measured single- and double-strand DNA breaks by performing alkaline single-cell gel electrophoresis (comet) assays.36 Pimozide treatment significantly increased the number of comet-containing GBM cells. Notably, NPCs were much less sensitive to USP1 targeting as evidenced by no changes in γH2AX phosphorylation and comet assay results (Fig. 4C and D).

Fig. 4.

Fig. 4.

Open in a new tab

Ubiquitination-specific protease 1 (USP1) promotes glioblastoma (GBM) radioresistance. (A) Western Blot analysis of phosphorylated CHK1 (S296), total CHK1, and phosphorylated H2AX proteins in GBM cells transduced with the control shRNA (NT) and USP1 shRNA (sh). β-actin was used as a loading control. (B) Representative images of IF staining analysis using an antibody against phosphorylated histone 2AX (p-γH2AX). Eight hundred twenty-seven cells were treated with pimozide for the indicated periods (hours) and processed for IF analysis. DAPI was used to label nuclei. (C and D) Alkaline comet assays were performed at the indicated time points after pimozide treatment in GBM cells and NPCs. Representative images are shown in C. Arrows depict the cells with comet. Data from the comet assays are quantified in D. Data are means ± SD (n = 5) *P < .01. (E and F) IF staining analysis of p-γH2AX nuclei foci after treatment with NCS. One hundred thirty-one GBM cells were treated with NCS (100 ng/mL) for 3 hours. NCS-pretreated cells were further cultured with or without pimozide for an additional 1 or 24 hours. Cells with p-γH2AX nuclear foci staining were quantified. Data are means ± SD (n = 5) *P < .01. (G) Soft agar clonogenic assay of 448 GBM cells with irradiation and pimozide treatment. Surviving colonies were counted 2 weeks later, and surviving fractions of cells were calculated. Pimozide was added every 3 days. (H) LDA analysis of 448 cells that received irradiation alone, pimozide alone, or both.

The above data suggest that USP1 promotes GBM cell survival via facilitating the DNA damage-repair process, which raises the possibility that USP1 targeting may enhance the therapeutic efficacy of irradiation significantly. To test, we treated GBM cells with the radiomimetic drug neocarzinostatin (NCS) (100 ng/mL), cultured with or without pimozide, and performed immunostaining analysis of p-γH2AX at various time points (Fig. 4E and F). p-γH2AX nuclei foci were detected in most GBM cells 1 hour after NCS treatment but disappeared within 24 hours, indicating the efficient DNA damage repair in GBM cells. In sharp contrast, p-γH2AX nuclei foci were persistent in pimozide-treated GBM cells (Fig. 4E and F). Clonogenic regrowth of GBM cells after irradiation is a gold standard for determining radioresistance. We irradiated GBM cells with varying doses of radiation, with or without pimozide, and monitored the formation of clonogenic colonies (Fig. 4G and H). GBM cells that received irradiation and pimozide treatment failed to survive compared with the mock or single-treatment group (Fig. 4G). Results from both soft agar-based standard clonogenic assays and in vitro limiting dilution assays revealed that combination treatment of irradiation and pimozide significantly decreased clonogenic growth of GBM.

USP1 Targeting Impedes Gblioblastoma Growth in Combination with Irradiation

To determine the role of USP1 in vivo, we generated orthotopic xenograft tumor models and targeted USP1 function either by shRNA-mediated knockdown or pimozide treatment (Fig. 5). We implanted the cells with or without USP1 knockdown into the brains of SCID mice (50 000 cells per mouse). While the animals bearing GBM cells with nontargeting shRNA died earlier (31 days of median survival for 827; 120 days for 131), the animals injected with USP1 knockdown cells survived significantly longer. Kaplan-Meier survival plots showed a significant increase in the survival of mice bearing USP1 knockdown cells (P < .0001 by log-rank test) (Fig. 5A and B). In a parallel experiment, mice from both groups were euthanized, and the brains were examined for histological analysis (n = 3 for each group). Whereas massive tumors were observed in animals injected with control cells, no macroscopic tumors were detected in animals injected with USP1 knockdown GBM cells (Fig. 5A and B).

Fig. 5.

Fig. 5.

Open in a new tab

Ubiquitination-specific protease (USP) targeting sensitizes glioblastoma (GBM) to irradiation and prolongs the survival of tumor-bearing mice. (A and B) Kaplan-Meier survival plots of mice injected with 827 (A) or 131 (B) GBM cells transduced with either NT shRNA or USP1 shRNA expressing lentivirus (10 mice for group, 827; 8 mice for group, 131). P value was determined by log-rank test. P < .01 for both experiments. Mice were euthanized 1 month (for 827) and 4 months (for 131) after injection, and their brains were sectioned and stained with hematoxylin and eosin. Bar represents 100 microns. (C) Kaplan–Meier plot of the mice bearing 559 xenograft tumors after irradiation and/or pimozide treatment. Once tumors were formed, mice received mock treatment, pimozide alone (10 mg/kg body weight, p.o. 5 times/wk), radiation alone (IR; 2 Gy daily for 5 days), and combination. Mice with combination treatment survived significantly longer than any other group. P < .001 with log-rank test. (D) Immunohistochemical analyses on the sections from 559 derived orthotopic tumors. Positively stained cells for p-γH2AX are labeled with brown color. Bars represent 50 microns for immunohistochemistry (IHC). (E) Quantitation of the cells immunopositive for p-γH2AX shown are in D. Error bars represent SD. *P < .01. (F) IHC for CD133 and SOX2 on the sections from xenograft tumors. Positively stained cells for p-γH2AX are labeled with brown color, and hematoxylin was stained for the background. Bars represent 50 microns. (G) The percentage of CD133 (blue) or SOX2 (red) positive cells shown in F was demonstrated as a bar graph. Data are represented as means ± SD. *P < .01 **P < .001.

To evaluate the therapeutic efficacy of USP1 targeting in a more clinically relevant setting, we used pimozide instead of genetic knockdown. Drug delivery into brain tumor tissues is a critical issue in anti-GBM therapy, in part due to the blood-brain barrier. Because pimozide has been used in clinic as a neuroleptic drug, the pharmacokinetic data and safety of this drug have been well established, including the penetrance of pimozide to the brain.37 We established orthotopic GBM tumors in nude mice and performed the fractionated irradiation regimen (2 Gy daily for 5 consecutive days).38 With the doses used in this study, irradiation or pimozide alone marginally increased the survival of tumor-bearing mice. Notably, however, mice receiving both irradiation and pimozide survived almost 2 times longer than any other group (median survival, 49 days; P < .001 with log-rank test). To determine in the vivo effects of pimozide, irradiation, and the combination on tumor cell biology, we harvested tumors and performed immunohistochemistry analyses. First, The level of p-γH2AX in xenograft tumors was significantly increased in the pimozide-treated group compared with the untreated control, with a further increase in the combination group (Fig. 5D and E) suggesting that pimozide enhances DNA damage in GBM tumor. Second, we determined the frequencies of SOX2- and CD133-positive cells in the xenograft tumors (Fig. 5F and G). Pimozide treatment notably decreased the expression of both SOX2 and CD133 in GBM xenograft tumors, and its inhibitory effect was further enhanced when radiotherapy was combined. Together, these results demonstrate that molecular targeting of USP1 in GBM impedes stem cell maintenance and radioresistance in patient-derived GBM models and further suggests that USP1 targeting is a novel therapeutic approach.

Discussion

In this study, we have shown that USP1 promotes stem cell maintenance and radioresistance of GBM tumor via stabilization of ID1 and CHEK1. Targeting USP1 attenuated self-renewal and survival of GBM cells and sensitized GBM cells to irradiation. Importantly, USP1 targeting spared normal neural progenitors, supporting its potential utility as a GBM-specific therapeutic approach.

There are several potential explanations for our finding that USP1 targeting elicits strong antitumorigenic effects while sparing NPCs. Disruption of USP1 resulted in increased sensitivity to DNA cross-linkers and hypersensitivity to mitomycin C, suggesting that inhibiting USP1 might sensitize cells to DNA damage by decreasing the cells' ability to repair or tolerate DNA lesions.39 Since genomic instability is a general characteristic of human tumor, cancer cells tend to have higher levels of endogenous replication-associated lesions than those present in normal cells.40 Alternatively, high levels of USP1 in GBM tumors and GSCs compared with normal counterpart may explain differential sensitivities between GBM cells and NPCs. In agreement with this, we found a marginal decrease of ID1 in NPCs after USP1 inhibition (data not shown). Although ID1 is a critical regulator for NPCs, ID1 might be functionally redundant with other ID proteins in NPCs.41

DUBs are overexpressed or activated in various cancers, and they contribute to tumor progression and recurrence. However, the roles of DUBs in GBM growth and cellular hierarchy are largely unknown. Preferential expression of USP1 in GSCs warrants further studies to elucidate the underlying molecular mechanism. USP1 promoted the stability of ID1 and CHEK1. Interestingly, the levels of phosphorylated CHEK1 appeared to be more robustly modulated by USP1 than those of total CHEK1. Previous studies showed that ID1 and phosphorylated CHEK1 proteins were highly expressed in stem-like GBM cells relative to the bulk tumor. Therefore, it is tempting to speculate that USP1-mediated protein stability is a key regulator for maintaining stemness in these cells. Protein levels of USP1 positively correlated with the levels of ID1 and CHEK1 in human GBM specimens (Supplementary Fig. 3).

Pimozide has been used clinically to treat schizophrenic patients,25 but it induces adverse effects including akinesia, depression, anorexia, tremor, and extrapyramidal symptoms with high-dose chronic exposure. To avoid the unwanted side effects in mice, we treated mice with a relatively low dose of pimozide (10 mg/kg body weight per day). This regimen was sufficient to induce the increased DNA damage response and the decrease the SOX2-positive cells within tumors, suggesting antitumor effects exerted by pimozide treatment. However, we observed the dramatic survival benefit in tumor-bearing mice in the group receiving both pimozide and irradiation but not in those receiving single treatment only. More thorough in vivo studies to determine adequate drug concentrations and regimens are required for the repositioning of pimozide or other more specific USP1 inhibitor as a GBM radiosensitizer.

In summary, we have provided evidence that USP1 promotes critical regulators of DNA damage repair and stem cell maintenance in GBM cells and showed that genetic and pharmacologic inhibition of USP1 markedly decreased the clonogenic growth of GBM cells and tumor growth in xenograft tumor models without normal cell toxicity. These data support that USP1-mediated protein stabilization is a key regulatory mechanism for GSC maintenance and treatment resistance of GBM and that USP1 inhibition can be an effective therapeutic approach against GBM.

Supplementary Material

Supplementary material is available online at Neuro-Oncology (http://neuro-oncology.oxfordjournals.org/).

Funding

This work was supported by a grant from the Korea Health Technology R&D project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI09C1552); the Global Frontier Project grant (NRF-2012M3A6A-2010-0029781) of National Research Foundation funded by the Ministry of Science, ICT and Future Planning (MSIP) of Korea; Samsung Medical Center grant GFO1140011 (D-H.N) and NIH R01 NS082312 and R01 NS083767 (J.L.).

Conflict of interest statement. There is no conflict of interest for all authors.

Supplementary Material

Supplementary Data
supp_18_1_37__index.html (872B, html)

References

  • 1.Stupp R, Hegi ME, Mason WP 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 Oncol. 2009;10(5):459–466. [DOI] [PubMed] [Google Scholar]
  • 2.Fonkem E, Wong ET. NovoTTF-100A: a new treatment modality for recurrent glioblastoma. Expert Rev Neurother. 2012;12(8):895–899. [DOI] [PubMed] [Google Scholar]
  • 3.Gilbert MR, Dignam JJ, Armstrong TS et al. A randomized trial of bevacizumab for newly diagnosed glioblastoma. N Engl J Med. 2014;370(8):699–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Chinot OL, Wick W, Mason W et al. Bevacizumab plus radiotherapy-temozolomide for newly diagnosed glioblastoma. N Engl J Med. 2014;370(8):709–722. [DOI] [PubMed] [Google Scholar]
  • 5.Dirks PB. Brain tumor stem cells: the cancer stem cell hypothesis writ large. Mol Oncol. 2010;4(5):420–430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Singh SK, Hawkins C, Clarke ID et al. Identification of human brain tumour initiating cells. Nature. 2004;432(7015):396–401. [DOI] [PubMed] [Google Scholar]
  • 7.Meacham CE, Morrison SJ. Tumour heterogeneity and cancer cell plasticity. Nature. 2013;501(7467):328–337. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bao S, Wu Q, McLendon RE et al. Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature. 2006;444(7120):756–760. [DOI] [PubMed] [Google Scholar]
  • 9.Liu G, Yuan X, Zeng Z et al. Analysis of gene expression and chemoresistance of CD133+ cancer stem cells in glioblastoma. Mol Cancer. 2006;5:67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chen J, Li Y, Yu TS et al. A restricted cell population propagates glioblastoma growth after chemotherapy. Nature. 2012;488(7412):522–526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Brennan CW, Verhaak RG, McKenna A et al. The somatic genomic landscape of glioblastoma. Cell. 2013;155(2):462–477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Verhaak RG, Hoadley KA, Purdom E et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Sturm D, Bender S, Jones DT et al. Paediatric and adult glioblastoma: multiform (epi)genomic culprits emerge. Nat Rev Cancer. 2014;14(2):92–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Stommel JM, Kimmelman AC, Ying H et al. Coactivation of receptor tyrosine kinases affects the response of tumor cells to targeted therapies. Science. 2007;318(5848):287–290. [DOI] [PubMed] [Google Scholar]
  • 15.Nijman SM, Luna-Vargas MP, Velds A et al. A genomic and functional inventory of deubiquitinating enzymes. Cell. 2005;123(5):773–786. [DOI] [PubMed] [Google Scholar]
  • 16.Cheng C, Niu C, Yang Y et al. Expression of HAUSP in gliomas correlates with disease progression and survival of patients. Oncol Rep. 2013;29(5):1730–1736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lee J, Kotliarova S, Kotliarov Y et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines. Cancer Cell. 2006;9(5):391–403. [DOI] [PubMed] [Google Scholar]
  • 18.Lee J, Son MJ, Woolard K et al. Epigenetic-mediated dysfunction of the bone morphogenetic protein pathway inhibits differentiation of glioblastoma-initiating cells. Cancer Cell. 2008;13(1):69–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Madhavan S, Zenklusen JC, Kotliarov Y et al. Rembrandt: helping personalized medicine become a reality through integrative translational research. Mol Cancer Res. 2009;7(2):157–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Rhodes DR, Yu J, Shanker K et al. ONCOMINE: a cancer microarray database and integrated data-mining platform. Neoplasia. 2004;6(1):1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kim E, Kim M, Woo DH et al. Phosphorylation of EZH2 activates STAT3 signaling via STAT3 methylation and promotes tumorigenicity of glioblastoma stem-like cells. Cancer Cell. 2013;23(6):839–852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Son MJ, Woolard K, Nam DH et al. SSEA-1 is an enrichment marker for tumor-initiating cells in human glioblastoma. Cell Stem Cell. 2009;4(5):440–452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Gangemi RM, Griffero F, Marubbi D et al. SOX2 silencing in glioblastoma tumor-initiating cells causes stop of proliferation and loss of tumorigenicity. Stem Cells. 2009;27(1):40–48. [DOI] [PubMed] [Google Scholar]
  • 24.Favaro R, Appolloni I, Pellegatta S et al. Sox2 is required to maintain cancer stem cells in a mouse model of high-grade oligodendroglioma. Cancer Res. 2014;74(6):1833–1844. [DOI] [PubMed] [Google Scholar]
  • 25.Opler LA, Feinberg SS. The role of pimozide in clinical psychiatry: a review. J Clin Psychiatry. 1991;52(5):221–233. [PubMed] [Google Scholar]
  • 26.Chen J, Dexheimer TS, Ai Y et al. Selective and cell-active inhibitors of the USP1/UAF1 deubiquitinase complex reverse cisplatin resistance in non-small cell lung cancer cells. Chem Biol. 2011;18(11):1390–1400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Williams SA, Maecker HL, French DM et al. USP1 deubiquitinates ID proteins to preserve a mesenchymal stem cell program in osteosarcoma. Cell. 2011;146(6):918–930. [DOI] [PubMed] [Google Scholar]
  • 28.Guervilly JH, Renaud E, Takata M et al. USP1 deubiquitinase maintains phosphorylated CHK1 by limiting its DDB1-dependent degradation. Hum Mol Genet. 2011;20(11):2171–2181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Yang K, Moldovan GL, Vinciguerra P et al. Regulation of the Fanconi anemia pathway by a SUMO-like delivery network. Genes Dev. 2011;25(17):1847–1858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Perk J, Iavarone A, Benezra R. Id family of helix-loop-helix proteins in cancer. Nat Rev Cancer. 2005;5(8):603–614. [DOI] [PubMed] [Google Scholar]
  • 31.Ying QL, Nichols J, Chambers I et al. BMP induction of Id proteins suppresses differentiation and sustains embryonic stem cell self-renewal in collaboration with STAT3. Cell. 2003;115(3):281–292. [DOI] [PubMed] [Google Scholar]
  • 32.Lasorella A, Benezra R, Iavarone A. The ID proteins: master regulators of cancer stem cells and tumour aggressiveness. Nat Rev Cancer. 2014;14(2):77–91. [DOI] [PubMed] [Google Scholar]
  • 33.Ciarrocchi A, Jankovic V, Shaked Y et al. Id1 restrains p21 expression to control endothelial progenitor cell formation. PloS One. 2007;2(12):e1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Nijman SM, Huang TT, Dirac AM et al. The deubiquitinating enzyme USP1 regulates the Fanconi anemia pathway. Mol Cell. 2005;17(3):331–339. [DOI] [PubMed] [Google Scholar]
  • 35.Celeste A, Fernandez-Capetillo O, Kruhlak MJ et al. Histone H2AX phosphorylation is dispensable for the initial recognition of DNA breaks. Nat Cell Biol. 2003;5(7):675–679. [DOI] [PubMed] [Google Scholar]
  • 36.Malyapa RS, Bi C, Ahern EW et al. Detection of DNA damage by the alkaline comet assay after exposure to low-dose gamma radiation. Radiat Res. 1998;149(4):396–400. [PubMed] [Google Scholar]
  • 37.Mothi M, Sampson S. Pimozide for schizophrenia or related psychoses. Cochrane Database Syst Rev. 2013;11:CD001949. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Joo KM, Kim J, Jin J et al. Patient-specific orthotopic glioblastoma xenograft models recapitulate the histopathology and biology of human glioblastomas in situ. Cell Rep. 2013;3(1):260–273. [DOI] [PubMed] [Google Scholar]
  • 39.Oestergaard VH, Langevin F, Kuiken HJ et al. Deubiquitination of FANCD2 is required for DNA crosslink repair. Mol Cell. 2007;28(5):798–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646–674. [DOI] [PubMed] [Google Scholar]
  • 41.Nam HS, Benezra R. High levels of Id1 expression define B1 type adult neural stem cells. Cell Stem Cell. 2009;5(5):515–526. [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

Supplementary Data
supp_18_1_37__index.html (872B, html)
supp_nov091_nov091supp_figs.pdf (387.4KB, pdf)

Articles from Neuro-Oncology are provided here courtesy of Society for Neuro-Oncology and Oxford University Press

RESOURCES