What this is
- Glioblastoma (GBM) is a highly aggressive brain cancer with limited treatment options.
- Temozolomide (TMZ) is a standard chemotherapy, but its effectiveness is often hindered by the DNA repair protein .
- This research explores using technology to downregulate expression, enhancing glioma cell sensitivity to TMZ.
- The findings suggest a novel therapeutic approach for overcoming chemoresistance in GBM.
Essence
- effectively downregulates in glioma cells, enhancing their sensitivity to TMZ chemotherapy. This strategy shows promise for treating chemoresistant glioblastoma.
Key takeaways
- targeting of mRNA resulted in significant downregulation of both mRNA and protein levels in glioma cells. This downregulation was achieved in both established glioma cell lines and patient-derived gliomaspheres.
- The study demonstrated that downregulation of led to enhanced cytotoxic effects of TMZ, with significant increases in cell death observed in previously resistant glioma cells. This suggests that could be a viable strategy to improve treatment outcomes.
- Stable knockdown of using a lentiviral Cas13d system maintained enhanced sensitivity to TMZ in gliomasphere models, indicating the potential for long-term therapeutic applications.
Caveats
- In vivo delivery of the system remains a significant challenge, limiting immediate clinical applicability. Effective delivery methods that cross the blood-brain barrier need to be developed.
- The study lacks in vivo validation in orthotopic glioma models, which is necessary to confirm the translational efficacy and safety of this approach.
- The effectiveness of the system in the context of mismatch repair deficiency was not evaluated, which may affect treatment outcomes in some patients.
Definitions
- MGMT: A DNA repair enzyme that counteracts the effects of alkylating agents like TMZ, contributing to chemoresistance in glioblastoma.
- CRISPR-Cas13: A genome-editing technology that targets and degrades RNA, allowing for precise regulation of gene expression.
Simplified
Introduction
Glioblastoma (GBM) is the most common and aggressive primary brain cancer in adults [1]. Current standard of care combines fractionated radiation with the DNA alkylating agent temozolomide (TMZ), which provides a modest improvement in overall survival [2, 3]. TMZ exerts its cytotoxic effect by methylating guanine residues in DNA; however, the DNA repair protein O⁶-methylguanine-DNA methyltransferase (MGMT) reverses this alkylation, resulting in chemoresistance [3, 4].
MGMT expression is epigenetically regulated by its gene promoter methylation status [5]. Clinical sensitivity to TMZ is confined to ~ 40% of patients with MGMT promoter methylation and resultant low MGMT levels [5]. Conversely, patients with an unmethylated MGMT promoter express high MGMT, leading to chemoresistance and worse outcomes [5]. Therefore, strategies to overcome MGMT-mediated chemoresistance are of therapeutic importance.
Previous efforts to inhibit MGMT have yielded mixed results. RNA interference (RNAi) studies were hindered by lack of specificity and complex in vivo delivery [6–8]. Furthermore, small molecule inhibitors designed to directly block MGMT protein activity failed to provide significant clinical benefit and have been associated with adverse side effects [9, 10]. More recently, CRISPR-based technologies have emerged as novel tools for modulating gene expression. Our group and others have demonstrated that dCas9-based epigenetic editing induces MGMT promoter methylation and enhanced TMZ sensitivity [11–14]. However, epigenetic approaches are limited by slow kinetics and in vivo delivery of the large dCas9-fusion protein [15].
To address these limitations, we turned to the Class 2, Type VI CRISPR-Cas13 system, which directly targets and degrades RNA transcripts more efficiently than epigenetic editing and with higher specificity than traditional RNAi [16]. In this study, we selected the Cas13x and Cas13d variants, which are distinguished by their high on-target efficiency and minimal non-specific collateral RNA degradation [11, 17]. We hypothesized that directing these Cas13 variants to degrade MGMT mRNA would effectively sensitize chemoresistant glioma cells to TMZ. Here, we demonstrate in vitro in both established glioma cell lines and patient-derived gliomaspheres that Cas13-mediated knockdown of MGMT significantly enhances the cytotoxic effects of TMZ chemotherapy, presenting a novel therapeutic strategy for glioblastoma.
Materials and methods
For full reagent details and catalog numbers, please see Supplementary Table 4.
Plasmids and construction
We used the pCMV-Cas13X.1 plasmid (Addgene) to express Cas13x and its corresponding guide CRISPR RNAs (crRNAs) in glioma cells. For cloning, we synthesized five pairs of DNA oligonucleotides (Supplementary Table 1) encoding a non-specific control (NSC) crRNA and four MGMT-targeting crRNAs (Fig. 1-a). The annealed templates were cloned into the BbsI-digested vector to generate five distinct Cas13x-crRNA expression plasmids, which were confirmed by DNA sequencing using BigDye Terminator v3.1 chemistry (Applied Biosystems).
For Cas13d and crRNA expression, we used the lentiviral plasmid pLentiRNACRISPR_006 (Addgene). We synthesized five pairs of DNA oligonucleotides (Supplementary Table 2) with the same crRNA sequences used for the Cas13x system. These were cloned into the BsmBI-digested vector to create five unique Cas13d-crRNA expression plasmids, validated by DNA sequencing.
In vitro characterization and cleavage activity of the Cas13x/crRNA System AgainstmRNA. () Schematic of themRNA transcript, indicating crRNA target sites and RT-PCR primer binding sites. () Agarose gel electrophoresis of the five in vitro transcribed crRNAs (NSC,crR-1, -2, -3, and − 4). () Ruby stain showing total purified Cas13x-His protein after electrophoresis and transfer to a nitrocellulose membrane. () Western blot of the same membrane from (), probed with an anti-His antibody to confirm protein identity. () Cell-free cleavage assay. Total RNA from LN18 cells was incubated for 16-hours with pre-assembled Cas13x/crRNA ribonucleoprotein (RNP) complexes. RemainingmRNA was detected by RT-PCR (ACTB control). () Time-course ofmRNA cell free cleavage by the Cas13x/crRNA-3 RNP. () Box-and-whisker plots for all crR-1, -2, -3, -4 (23 bp) NCBI BLAST+ results using the transcriptome as a reference, permitting mismatches and gapped (indels) alignments. A few potential “collateral” hits are highlighted MGMT MGMT MGMT MGMT MGMT a b c d c e f g
Cas13x protein expression inand purification E. coli
Recombinant Cas13x protein was produced by transforming the pET28-6His-SUMO-Nsp10 vector into E. coli BL21(DE3) and inducing expression with 100 µM IPTG. Cells were grown at 37 °C until 600 nm (OD₆₀₀) ~ 0.5, then incubated for 5 h post-induction. Cas13x-His protein was purified from the cell lysate under native conditions using a Ni Spin Column (NEB). Purity and protein identification was validated by SYPRO™ Ruby Protein Blot Stain (Invitrogen) and with Western blot with Rabbit Anti-His antibody (Proteintech).
crRNA design, synthesis in vitro, and purification
The four MGMT-targeting crRNAs were identified and designed to prioritize target sites with high predicted RNA accessibility [18] and optimal GC content within the first 500 bases of MGMT mRNA (Fig. 1-a) [19]. We avoided regions with predicted high secondary structure complexity to ensure efficient binding of the Cas13-crRNA complex.
crRNA DNA templates, each including a 5’ T7 promoter sequence, were generated by PCR using five forward primers and one common reverse primer (Supplementary Table 3) and the Cas13x-crRNA-All-in-One vectors as template. The five crRNAs (NSC, MGMT crR-1, -2, -3, and − 4) were synthesized by in vitro transcription using the HiScribe™ T7 High Yield RNA Synthesis Kit (NEB). Products were purified with TRIzol™ reagent (Invitrogen) and validated by agarose gel electrophoresis.
For analysis of collateral activity, crRNAs were further evaluated for potential “off-target” sequence matches across the transcriptome (GENCODE v46) using Bowtie2 (v2.5.4, additional arguments: –local –very-sensitive-local -L 10 -N 1 –rdg 3,1 –rfg 3,1 -k 100) and the blastn algorithm from NCBI BLAST+ (v2.13.0, additional arguments: -task blastn-short -word_size 4 -gapopen 2 -gapextend 1 -perc_identity 50 -evalue 700).
In vitro cleavage activity assay (cell free) of Cas13x/-crRNA ribonuclear protein (RNP) MGMT
The cell-free cleavage activity of the Cas13x RNP was measured following Konermann et al. [20]. Briefly, RNP complexes (1 µg Cas13x-His, 1 µg crRNA) were incubated with 2 µg of total RNA from LN18 cells in RNA cleavage buffer. Reactions were incubated at 37°C for specified time points (1, 2, 4, 8, or 24 hours). The reaction was terminated with TRIzol™ reagent, and the RNA was purified. Purified RNA was used to synthesize first-strand cDNA (SuperScript™ III Reverse Transcriptase, Invitrogen) and the remaining MGMT transcript was measured by PCR using the REDTaq® ReadyMIX PCR Reaction Kit (Sigma-Aldrich) and MGMT-specific primers (Forward: 5’-TGCGCACCGTTTGCGACTTG-3’; Reverse: 5’-GGTTGCCTGCCAGGGCTGC-3’).
Culture of LN18 glioma cells and patient derived glioma stem-like cells (gliomaspheres)
LN18 glioma cells were sourced from American Type Culture Collection (Manassas, Virginia) and primary patient-derived GS104 and GS081 gliomaspheres were supplied by Dr. Nathanson (Department of Molecular and Medical Pharmacology, UCLA). GS104 and GS081 were derived from newly diagnosed IDH wild-type WHO grade IV GBM with an unmethylated MGMT promoter and were selected due to their relatively fast growth in culture following transfection [21].
LN18 cells were maintained in DMEM (GIBCO) supplemented with 10% FBS and penicillin/streptomycin (37 °C, 5% CO₂) (GIBCO). Gliomaspheres were maintained in serum-free Neurobasal medium (GIBCO) supplemented with 1X Anti-Anti (GIBCO), 1X B27 (GIBCO), 5 µg/mL heparin (Sigma-Aldrich), 40 µg/mL EGF (Millipore), 40 µg/mL bFGF (Millipore), 1.25X N2 Supplement (GIBCO), and 200 mM L-Glutamine (GIBCO). All cell types express MGMT due to an unmethylated promoter and were cultured as previously described [11, 22].
Cas13x transfection (LN18 cells)
Cas13x was transfected into LN18 cells as pre-assembled Cas13x-crRNA ribonucleoproteins (RNPs) using Lipofectamine™ CRISPRMAX™ Reagent (Invitrogen). For transient expression, the Cas13x-crRNA-All-in-One plasmid was transfected using X-tremeGENE™ HP DNA Transfection Reagent (Roche). Both were performed per the manufacturer’s protocols.
Cas13d lentiviral transduction (LN18 and gliomaspheres)
To create stable cell lines, the Cas13d-crRNA-All-in-One lentiviral vector was first packaged into infectious lentiviral particles in HEK293T cells, as previously described [11, 22]. LN18 cells were then infected with the resulting virus. GS104 and GS081 gliomaspheres were infected using a spin-down method to enhance efficiency [22]. Three days post-infection, the cells underwent a five-day selection with puromycin (1.0 µg/ml for LN18; 0.7 µg/ml for gliomaspheres) to establish stable cell lines. Puromycin selected LN18 cells were grown out for one week before use in subsequent experiments while selected GS104 and GS081 cells required three months to establish stability.
RT-PCR formRNA MGMT
Total cellular RNA was isolated using the RNeasy® Mini Kit (Qiagen). First-strand cDNA was synthesized from 0.5 µg of RNA using SuperScript™ III Reverse Transcriptase (Invitrogen). The resulting cDNA was used for PCR to quantify the relative abundance of the MGMT transcript.
Western blot for MGMT protein and densitometry analysis
To assess MGMT protein levels, Western blot was performed as previously described [11]. The following primary antibodies were used: anti-HA, rabbit, (1:1000, Sigma); anti-MGMT, mouse (1:1000, ThermoFisher); GAPDH, mouse (1:2000, Proteintech). Densitometry analysis of relative MGMT protein expression was performed in FIJI (ImageJ) by normalizing MGMT band intensities to GAPDH (same gel) and using the Analyze > Gels function. All groups (NSC, crR-2, crR-3, and crR-4) were run together in each gel replicate, for a total of three replicates across two gliomasphere lines (GS104, GS081) and compared using a one-way permutation analysis of variance (ANOVA).
Assay of cellular sensitivity to TMZ therapy
Cell viability following treatment was assessed using a standard MTT [3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2 H-tetrazolium bromide] (Invitrogen) assay. Cells were seeded into 24-well plates (2,300 or 4,600 cells/well) and treated with varying concentrations of TMZ (50, 100, or 200 µM) or a DMSO vehicle control (4 days for LN18, 5 days for gliomaspheres). Following treatment, cells were incubated with MTT solution (0.5 mg/mL in regular culture media). The resulting formazan product was solubilized with DMSO, and absorbance was measured at 560 nm (background subtraction at 660 nm) to quantify cell viability.
Statistical methods for MTT assays
Data were analyzed in Prism 9 via a student’s t-test or ANOVA with Tukey’s HSD post hoc test for multiple comparisons, where appropriate.
Live-cell imaging
Representative phase-contrast images of gliomasphere cell lines were acquired using a Life Technologies EVOS XL Core Imaging System (Life Technologies, Cat# AME3302) and captured using a 10x LWD (long working distance) phase objective at room temperature. Quantification of cell radii from >10 live-cell images for each condition were performed in FIJI using the Watershed and Analyze Particles functions.
Results
Verification ofmRNA cleaved by Cas13x/crRNAs in cell free condition MGMT
We first designed four unique MGMT-targeting crRNAs and a non-specific control (NSC) crRNA to target a 500-bp region near the 5’ end of the MGMT mRNA (Fig. 1-a). Their integrity was confirmed by agarose gel electrophoresis (Fig. 1-b).
The CRISPR-Cas13 family targets RNA [23]. We selected the Cas13x variant as our therapeutic effector due to its small molecular size, efficient on-target activity, and reduced off-target collateral activity [17]. We successfully expressed and purified the recombinant Cas13x protein from E. coli (Figs. 1-c, d).
To test the cleavage efficiency of our system, we assembled RNPs (purified Cas13x protein with synthesized crRNAs) and incubated them with total RNA from LN18 cells in a cell-free assay [24]. MGMT mRNA degradation was assessed by RT-PCR. All four MGMT-targeting RNPs efficiently cleaved the MGMT transcript compared to the NSC-RNP control (Fig. 1-e). The crRNA-3 RNP demonstrated the most potent cleavage activity. A time-course experiment revealed rapid cleavage by Cas13x/crRNA-3 RNP that occurred within the first hour (Fig. 1-f). To evaluate the selectivity of our system, we ran a highly permissive transcriptome-wide NCBI BLAST+ search for crR-1, -2, -3, and − 4, allowing for significant mismatches and gapped (indels) alignments (Fig. 1-g). No “off target” matching sequences were returned via a similar search with the Bowtie2 algorithm.
Cas13x/crRNA-3 RNP transfection in LN18 cells knocks down MGMT expression and enhances TMZ sensitivity in vitro
Given the potent guiding activity of crRNA-3, we proceeded with direct RNP transfection of LN18 cells. Knockdown of MGMT mRNA was nearly complete on day 1 post-transfection, with gradual recovery on days 2 and 3 (Fig. 2-a). A similar, slightly delayed reduction pattern was observed for MGMT protein, with recovery beginning after 72 h (Fig. 2-b).
Based on our previous work showing that a 4–5 day TMZ treatment is optimal for assessing viability changes in LN18 cells [11], we devised a double-delivery strategy to ensure sustained MGMT knockdown throughout the experiment (Fig. 2-c). The timing of the double-delivery strategy was optimized to prevent the recovery of down-regulated MGMT protein observed after 72 h of a single dose RNP dose (Fig. 2-b), thereby maximizing the targeting efficiency prior to assessing sensitivity to TMZ. LN18 cells treated with the Cas13x/crR-3 RNP showed a significant decrease in cell viability in the presence of TMZ compared to the non-specific control RNP (Fig. 2-d). These results demonstrate proof-of-concept for an RNA regulatory approach that sensitizes chemoresistant cells to TMZ using Cas13x.
![Cas13x/-crRNA RNP transfection reducesexpression and sensitizes LN18 cells to TMZ. () Time-course ofmRNA expression following RNP transfection (Cas13x protein alone, NSC RNP, orcrR-3 RNP).mRNA levels were assessed by RT-PCR (ACTB control). () Time-course of MGMT protein levels following RNP transfection as in (a) assessed by Western blot (GAPDH control). () Dual RNP delivery strategy with TMZ treatment. () Cell viability of LN18 cells following dual RNP transfection and TMZ treatment (0, 50, 100, 200 µM) measured by MTT assay. Data represent the summary of three transfections (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(3,64) = 107.2, p ≤ 0.0001; NSC vs. crR-3: F(1,64) = 145.2, p ≤ 0.0001; TMZ treatment and crR-3/NSC interaction: F(3,64) = 19.65, p ≤ 0.0001. Asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. crR-3, summarizing results at all 3 TMZ conditions: 50, 100, and 200 µM]. ns, not significant,,≤ 0.05; **,≤ 0.01; ***,≤ 0.001; and ****,≤ 0.0001 MGMT MGMT MGMT MGMT MGMT * p p p p a b c d](https://europepmc.org/articles/PMC12982228/bin/11060_2026_5500_Fig2_HTML.jpg.jpg "Click to view full size")
Cas13x/-crRNA RNP transfection reducesexpression and sensitizes LN18 cells to TMZ. () Time-course ofmRNA expression following RNP transfection (Cas13x protein alone, NSC RNP, orcrR-3 RNP).mRNA levels were assessed by RT-PCR (ACTB control). () Time-course of MGMT protein levels following RNP transfection as in (a) assessed by Western blot (GAPDH control). () Dual RNP delivery strategy with TMZ treatment. () Cell viability of LN18 cells following dual RNP transfection and TMZ treatment (0, 50, 100, 200 µM) measured by MTT assay. Data represent the summary of three transfections (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(3,64) = 107.2, p ≤ 0.0001; NSC vs. crR-3: F(1,64) = 145.2, p ≤ 0.0001; TMZ treatment and crR-3/NSC interaction: F(3,64) = 19.65, p ≤ 0.0001. Asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. crR-3, summarizing results at all 3 TMZ conditions: 50, 100, and 200 µM]. ns, not significant,,≤ 0.05; **,≤ 0.01; ***,≤ 0.001; and ****,≤ 0.0001 MGMT MGMT MGMT MGMT MGMT * p p p p a b c d
Lentiviral Cas13d system achieves stable MGMT knockdown in LN18 cells
We next sought to establish a stable MGMT knockdown in LN18 cells using lentiviral vector delivery strategy. Given the lack of an available lentiviral vector for the Cas13x variant, we pivoted to the Cas13d system, which has a similarly high on-target efficiency and low collateral activity [17]. We constructed lentiviral vectors expressing Cas13d and crRNAs (NSC, MGMT crR-2, -3, and − 4). Following infection and selection, all three MGMT-specific crRNAs led to a marked downregulation of both MGMT mRNA and protein compared to the NSC control (Figs. 3-a, b). Consistent with Cas13x-crRNA RNP results, stably transfected Cas13d-crRNA elicited a significant increase in TMZ sensitivity in LN18 cells (Fig. 3-c).
![Lentiviral delivery of Cas13d/MGMT-crRNA stably reduces MGMT expression and sensitizes LN18 cells to TMZ. () RT-PCR ofmRNA in transduced LN18 cells stably expressing Cas13d and crRNAs (ACTB control). () Western blot of MGMT protein in transduced LN18 cells (GAPDH control). () Cell viability (MTT assay) of stable cell lines after 96-hour treatment with increasing TMZ concentrations (0, 50, 100, or 200 µM). Data represent the summary of three transfections (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(3,188) = 236.0, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,188) = 53.74, p ≤ 0.0001; TMZ treatment and crR/NSC interaction: F(9,188) = 9.779, p ≤ 0.0001. Asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at all 3 TMZ conditions: 50, 100, and 200 µM]. ns, not significant,,≤ 0.05; **,≤ 0.01; ***,≤ 0.001; and,≤ 0.0001 a b c MGMT * p p p **** p](https://europepmc.org/articles/PMC12982228/bin/11060_2026_5500_Fig3_HTML.jpg.jpg "Click to view full size")
Lentiviral delivery of Cas13d/MGMT-crRNA stably reduces MGMT expression and sensitizes LN18 cells to TMZ. () RT-PCR ofmRNA in transduced LN18 cells stably expressing Cas13d and crRNAs (ACTB control). () Western blot of MGMT protein in transduced LN18 cells (GAPDH control). () Cell viability (MTT assay) of stable cell lines after 96-hour treatment with increasing TMZ concentrations (0, 50, 100, or 200 µM). Data represent the summary of three transfections (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(3,188) = 236.0, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,188) = 53.74, p ≤ 0.0001; TMZ treatment and crR/NSC interaction: F(9,188) = 9.779, p ≤ 0.0001. Asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at all 3 TMZ conditions: 50, 100, and 200 µM]. ns, not significant,,≤ 0.05; **,≤ 0.01; ***,≤ 0.001; and,≤ 0.0001 a b c MGMT * p p p **** p
Cas13d lentiviral system sensitizes primary patient derived gliomaspheres to TMZ treatment
We applied the Cas13d-crRNA lentiviral system to the primary patient-derived gliomasphere models GS104 and GS081. Stable populations were generated expressing Cas13d with an NSC or one of the MGMT-targeting crRNAs (crR-2, -3, or -4). All three MGMT-targeting crRNA vectors reduced both MGMT mRNA (Figs. 4-a, b) and protein levels (Figs. 4-c, d) with crR-3 demonstrating the most potent suppressive effect on both lines (Fig. 4-d). Assessing translational relevance, treatment with TMZ (100 and 200 µM) showed a significant increase in TMZ-induced cytotoxicity in gliomaspheres expressing MGMT crRNAs compared to NSC controls (Figs. 4-e, f). crR-3 expressing spheres showed the greatest enhancement of TMZ sensitivity. Additionally, long-term 14-day treatment of Cas13d-crR-3 GS104 gliomaspheres with 200 µM TMZ resulted in a greater cytotoxic response, as visualized by live-cell imaging (Fig. 4-g).
![Cas13d-mediated knockdown of MGMT sensitizes patient-derived gliomaspheres to TMZ. (,) RT-PCR ofmRNA in GS104 () and GS081 () gliomaspheres transduced with lentiviruses (control). () Western blot of MGMT protein in GS104 and GS081 gliomaspheres (GAPDH control). () Densitometry analysis of Western blot MGMT data from three independent experimental runs performed across two gliomasphere (GS104 and GS081) cell lines. All values are normalized to GAPDH expression; solid horizontal lines represent group means. Statistical analysis: one-way permutation ANOVA [ANOVA, RNA: F(3,8) = 17.31, p = 0.0008; asterisks indicate post hoc t-test result between crRNA and NSC with Benjamini-Hochberg multiple comparisons correction]. () Cell viability (MTT assay) of GS104 gliomaspheres after 96-hour treatment with increasing TMZ concentrations (0, 100, or 200 µM). Data represent the summary of three experiments from a single stable cell line (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(2,96) = 138.4, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,96) = 30.52, p ≤ 0.0001; TMZ treatment and crRNA/NSC interaction: F(6,96) = 7.743, p ≤ 0.0001; asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at both TMZ conditions: 100 and 200 µM]. () Cell viability (MTT assay) of GS081 gliomaspheres after 96-hour treatment with increasing TMZ concentrations (0, 100, or 200 µM). Data represent the summary of three experiments from a single stable cell line (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(2,60) = 394.5, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,60) = 28.26, p ≤ 0.0001; TMZ treatment and crR/NSC interaction: F(6,60) = 7.220, p ≤ 0.0001; asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at both TMZ conditions: 100 and 200 µM]. () Live-cell images of Cas13d-NSC-GS104 and Cas13d-crR-3-GS104 gliomaspheres after 14-day TMZ treatment (scale bars = 400 μm). Statistical analysis: two-way permutation ANOVA with post hoc Student’s t-test. [ANOVA, TMZ treatment: F(1,5724) = 227.81, p < 0.001; crR-3: F(1,5724) = 34.68, p < 0.001; TMZ treatment and crR-3/NSC interaction: F(1,5724) = 110.52, p < 0.001; asterisks indicate post hoc pairwise results with Holm method multiple comparisons correction]. ns, not significant, *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001; and ****, p ≤ 0.0001 a b a b c d e f g MGMT ACTB](https://europepmc.org/articles/PMC12982228/bin/11060_2026_5500_Fig4_HTML.jpg.jpg "Click to view full size")
Cas13d-mediated knockdown of MGMT sensitizes patient-derived gliomaspheres to TMZ. (,) RT-PCR ofmRNA in GS104 () and GS081 () gliomaspheres transduced with lentiviruses (control). () Western blot of MGMT protein in GS104 and GS081 gliomaspheres (GAPDH control). () Densitometry analysis of Western blot MGMT data from three independent experimental runs performed across two gliomasphere (GS104 and GS081) cell lines. All values are normalized to GAPDH expression; solid horizontal lines represent group means. Statistical analysis: one-way permutation ANOVA [ANOVA, RNA: F(3,8) = 17.31, p = 0.0008; asterisks indicate post hoc t-test result between crRNA and NSC with Benjamini-Hochberg multiple comparisons correction]. () Cell viability (MTT assay) of GS104 gliomaspheres after 96-hour treatment with increasing TMZ concentrations (0, 100, or 200 µM). Data represent the summary of three experiments from a single stable cell line (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(2,96) = 138.4, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,96) = 30.52, p ≤ 0.0001; TMZ treatment and crRNA/NSC interaction: F(6,96) = 7.743, p ≤ 0.0001; asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at both TMZ conditions: 100 and 200 µM]. () Cell viability (MTT assay) of GS081 gliomaspheres after 96-hour treatment with increasing TMZ concentrations (0, 100, or 200 µM). Data represent the summary of three experiments from a single stable cell line (mean ± SEM). Statistical analysis: two-way ANOVA with Tukey’s post hoc test. [ANOVA, TMZ treatment: F(2,60) = 394.5, p ≤ 0.0001; NSC vs. crR-2, -3, -4: F(3,60) = 28.26, p ≤ 0.0001; TMZ treatment and crR/NSC interaction: F(6,60) = 7.220, p ≤ 0.0001; asterisks indicate post hoc Tukey’s multiple comparisons test result for NSC vs. each crRNA, summarizing results at both TMZ conditions: 100 and 200 µM]. () Live-cell images of Cas13d-NSC-GS104 and Cas13d-crR-3-GS104 gliomaspheres after 14-day TMZ treatment (scale bars = 400 μm). Statistical analysis: two-way permutation ANOVA with post hoc Student’s t-test. [ANOVA, TMZ treatment: F(1,5724) = 227.81, p < 0.001; crR-3: F(1,5724) = 34.68, p < 0.001; TMZ treatment and crR-3/NSC interaction: F(1,5724) = 110.52, p < 0.001; asterisks indicate post hoc pairwise results with Holm method multiple comparisons correction]. ns, not significant, *, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001; and ****, p ≤ 0.0001 a b a b c d e f g MGMT ACTB
Discussion
In this study, we demonstrated that targeting MGMT mRNA with the CRISPR-Cas13 system effectively reverses TMZ chemoresistance in preclinical in vitro models of MGMT-unmethylated glioblastoma. Using both Cas13x and Cas13d variants, we achieved significant MGMT downregulation in established glioma cell lines and primary patient-derived gliomaspheres. While the concept of inhibiting MGMT to enhance TMZ efficacy is well-established, this study is among the first to apply the direct RNA targeting CRISPR-Cas13x and Cas13d variants for this purpose in clinically relevant primary patient-derived glioma models [4, 16, 25]. This novel RNA-guided RNA knockdown strategy induced TMZ sensitivity, demonstrating proof-of-principle for TMZ refractory MGMT unmethylated glioblastoma.
The Cas13-based RNA regulatory approach offers several advantages over other gene-targeting technologies. Unlike permanent genomic editing with Cas9, which carries the risk of irreversible off-target DNA mutations, Cas13 targets and degrades transient mRNA transcripts, making its effects reversible with a potentially more favorable safety profile [5, 15]. The Cas13 system is more direct and rapid compared to the multi-step dCas9-based epigenetic editing [11, 23, 26]. Furthermore, the smaller size of Cas13 variants like Cas13x, compared to the bulky dCas9-fusion proteins, may facilitate the development of optimized delivery systems that cross the blood-brain-barrier [20]. Additionally, so long as the designed guide RNAs have no sequence homology outside of the intended target genes, both Cas13x and Cas13d have been shown to exhibit little collateral activity, especially when compared to alternatives like RNAi and the use of small-molecule inhibitors [16].
Despite this proof-of-principle, our study has several limitations. The most significant challenge remains in vitro and in vivo delivery. Although lentivirus successfully established stable in vitro knockdown, such transduction strategies are unsuitable for clinical brain translation. Difficulties in delivering Cas13x RNPs to gliomasphere models highlight the need for safe, efficient, and tumor-restricted delivery vehicles (e.g., lipid nanoparticles or adeno-associated viruses) that can cross the blood-brain barrier [25, 27–29]. Notably, our study lacks in vivo validation in an orthotopic glioma xenograft model that would further establish translational efficacy, delivery, and safety of this system. In addition, while we did not identify any off target matches to our MGMT crRNAs without introducing features known to interfere with Cas13 activation (i.e., large mismatches or indels) [17] a comprehensive differential gene expression analysis with RNA-Seq will be necessary to confirm the absence of collateral activity in this system before its use in vivo and beyond.
Secondly, a notable observation in this study is that measurable chemosensitization was primarily achieved at TMZ concentrations (100–200 µM). While these levels exceed the typical steady-state concentrations found in patient CSF (~ 5–15 µM), they are consistent with established in vitro glioblastoma research standards. A systematic review by Poon et al. (2021) of over 212 studies found that IC50 values for glioma lines are frequently reported in the range of 200 µM to 1000 µM [30]. Furthermore, while the utilization of other Cas13 variants may enhance the potency of TMZ chemosensitization, we sought to characterize “on target only” variants to maximize the safety margin for future clinical application in the brain and to exploit the well-characterized MGMT methylation clinical biomarker in GBM.
Lastly, this study did not evaluate the effectiveness of the Cas13 system within the context of mismatch repair (MMR) deficiency. While MGMT promoter methylation remains the gold-standard biomarker for TMZ response, MMR expression is required to translate severe DNA damage into apoptosis [31]. Furthermore, at initial diagnosis, MMR deficiency is found in only ~ 4% of GBM cases and is often an acquired resistance mechanism occurring in ~ 25% of recurrent GBM [32]. Accordingly, our study only assessed primary cell lines from newly diagnosed MGMT-unmethylated glioma patients with confirmed MMR proficiency [21]. Future clinical implementation may benefit from the use of MMR proficiency as a secondary biomarker of therapeutic response.
In conclusion, CRISPR-Cas13-mediated MGMT RNA regulation is a novel strategy for MGMT-mediated chemoresistance. The speed, efficiency, and specificity of the Cas13 system make it a promising alternative to previous MGMT-targeting methods. While significant challenges, particularly in vivo delivery, remain, our findings provide a strong rationale for the continued development of Cas13-based therapeutics to overcome chemoresistance in glioblastoma.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1
Supplementary Material 2