Albumin-bound paclitaxel augment temozolomide treatment sensitivity of glioblastoma cells by disrupting DNA damage repair and promoting ferroptosis

The human U87-MG and LN229 cell lines used in this study were purchased from the American Type Culture Collection (ATCC). G353 and G393 are primary GBM cells isolated from fresh GBM tissues. Pathological information of primary cell patients was summarized in Table S1. All GBM samples were obtained from the Nanfang Glioma Center of Nanfang Hospital. GBM cells were cultured in Dulbecco's Modified Eagle Medium (Vivacell, C3113-0500) supplemented with 10% fetal bovine serum (BI Biotech, 04–001-1A) and maintained at 37℃ in a 5% CO₂ incubator. This study received approval from the ethics committee of Nanfang Hospital, Southern Medical University.

Cell viability assays were performed in 96-well assays at 10³ cells per well culture (5 replicate wells per group) with indicated drug treatment for 72 h. Cell viability evaluation was executed by the Cell Counting Kit-8 (CCK-8) assay, which was performed using CCK-8 kit (Cat.no. C6005; New Cell & Molecular Biotech) according to the manufacturer’s instructions.

For the colony formation assay, GBM cells were cultured in six-well plates with 200 cells per well in the presence or absence of indicated drug treatment for 24 h. Later, the medium was replaced with fresh growth medium. After 2 weeks, visible colonies were fixed with 100% methanol and stained with 0.1% crystal violet in 20% methanol for 15 min. The area of colonies was calculated using ImageJ software (1.48v, National Institutes of Health, USA).

GBM samples, nude mouse tumor tissues and GBM organoids were fixed in 10% formalin for 1 h, washed, and suspended in 70% ethanol. Histology and immunohistochemistry were performed, details are provided in the supplementary methods. For immunofluorescence, 5 × 10³ cells were grown on 20 mm confocal petri dishes and received indicated drug treatments for 24–72 h. Images were captured on a Carl-Zeiss confocal microscope. Zen Black software (Zeiss) and ImageJ were used for image capture and analysis, respectively.

The comet assay was performed with minor modifications as described [15]. The steps were performed according to kit instructions (WLA123; WanleiBio, Shenyang, China). All samples were stained and placed under the fluorescence microscope for observation. Data was analyses using the Comet Assay Software Project (CASP software).

For detecting the ferroptosis status of GBM cells after receiving indicated drug treatments. The total quantities of glutathione were measured using a GSH Assay Kit (Dojindo, G263), Intracellular chelatable iron was determined using the fluorescent indicator FerroOrange (Dojindo, F374). Intracellular lipid peroxides were determined using a Liperfluo-detection kit (Dojindo, L248).

Tumor tissue was collected from the operating room, suspended in ice cold culture media and brought to the lab on ice within 30 min from explantation. The protocol for constructing GBM organoids was referenced as previously described in the literature [16]. GBM organoids were plated on six-well, ultra-low adhesion plate (Corning; 3471). Plates were rotated at 120 rpm in a humidified incubator at 37°C, 5% CO2, and 21% oxygen. Short-Term GBM Organoid Mediumwas refreshed in organoid cultures every 48 h.

Statistical analysis of SPSS 23.0 was used in this study. Difference between groups were determined by the independent student’s t test or analysis of variance (one- or two-way ANOVAs). For all tests, P values less than 0.05 were considered to be statistically significant.

The viability of GBM cells was assessed by CCK-8 assays to determine the effects of TMZ and ABX. In this study, four GBM cell lines, namely U87-MG, LN229 and two primary GBM cell lines (G353 and G393), were selected for validation. The median inhibitory concentration (IC50) of TMZ and ABX was measured to determine the optimal drug combination. The concentration–response curves demonstrate that the IC50 values of TMZ in U87-MG, LN229, G353 and G393 cell lines were 673.69 µM, 1058.43 µM, 712.64 µM and 1122.52 µM respectively (Figure S1a). The IC50 values of ABX in U87-MG, LN229, G353 and G393 cell lines were 62.07 nM, 73.04 nM, 53.58 nM and 25.30 nM, respectively (Figure S1b). Overall, the sensitivity of GBM cells to the two drugs differed; ABX exhibited greater efficacy. Recognizing the side effects associated with the combination therapy of ABX and TMZ, we prudently chose a relatively low concentration of ABX (12 nM, approximately an IC₂₅-IC₃₀ concentration) and TMZ to conduct the following experiments in GBM cells. In U87 and G353 cells, we selected TMZ concentrations of 400µM corresponding to IC₂₅ to IC₃₀ levels, while for LN229 and G393 cells, we chose TMZ concentrations of approximately 800µM corresponding to IC₂₅ to IC₃₀ levels.

To further evaluate the impact of combined ABX and TMZ treatment on overall survival and adverse events in tumor-bearing nude mice, we monitored their survival time and body weight. Kaplan–Meier survival curves demonstrated that only the concurrent therapy regimen, and not the sequential therapy regimen, significantly prolonged the survival time of tumor-bearing nude mice. The "TMZ + ABX^high" group exhibited the most favorable prognosis (Fig. 2e). The body weight of nude mice in each group was recorded and presented in Fig. 2f. The results showed no significant decrease after ABX administration compared with the TMZ alone group. Furthermore, we conducted histological analysis of the lungs, liver, kidneys, and spleen using H&E staining. The results showed no abnormal pathological changes in the combinational treatment group, indicating that there was no significant increase in toxicity after additional ABX treatment (Figure S5).

Subsequently, we performed western blotting and immunofluorescence assays to further substantiate our hypothesis. The total, nuclear, and cytoplasmic proteins of GBM cells were extracted following the indicated treatments. The western blot results demonstrated a reduction in total ERCC1 protein expression in the TMZ plus ABX treatment group compared to both mono-treatment and control groups (Fig. 4d). The subcellular distribution analysis revealed a significant reduction in nuclear expression of ERCC1 and XPC proteins following treatment with TMZ plus ABX, as compared to mono-treatment and control groups (Fig. 4e). The results of immunofluorescence staining demonstrated that XPC protein accumulated in the nucleus following treatment with TMZ, but was absent in both ABX and TMZ plus ABX groups (Fig. 4f).

Ferroptosis is a newly identified form of cell death that results from iron-dependent lipid peroxidation. We employed liperfluo, Ferrous iron (Fe²⁺), and Glutathione (GSH) detection kits, the most commonly utilized indicators for ferroptosis detection, to assess the ferroptosis status of each group. For live cell imaging, we observed a significant accumulation of Fe²⁺ (Fig. 5d) and lipid-reactive hydrogen peroxide (Fig. 5e) in the cytoplasm of GBM cells following combination treatment with TMZ and ABX, as compared to mono-treatment with TMZ. The combination treatment of TMZ and ABX resulted in a significant reduction in intracellular GSH levels compared to mono-treatment (Fig. 5f). Additionally, we have also identified the mRNA expression level of the PTGS2 gene, which has recently been established as a standard marker for ferroptosis in vitro [19]. The combination treatment resulted in a significant upregulation of PTGS2 mRNA levels in GBM cells compared to mono-treatment (Fig. 5g), as revealed by qRT-PCR analysis. Compared to TMZ mono-treatment, the viability of GBM cells was significantly reduced after 48 h of treatment with TMZ plus ABX. However, addition of a ferroptosis inhibitor, Fer-1, partially rescued GBM cell viability in the combination treatment group (Fig. 5h), suggesting that the combination therapy of TMZ plus ABX could induce ferroptosis in GBM cells and that ferroptosis may represent only a portion of the anti-tumor effects mediated by this combination. Next, we investigated the potential crosstalk between ferroptosis and DNA damage induced by combination therapy. Our findings from immunofluorescence and western blotting analyses revealed that Fer-1 treatment did not affect γ-H2AX expression or the number of γ-H2AX foci in GBM cells treated with TMZ plus ABX, indicating an absence of crosstalk between ferroptosis and DNA damage induced by combination therapy (Figure S10).

However, the molecular mechanism underlying ferroptosis induced by combination treatment remains unclear. From the proteomics data, we have observed a significant fold change in the levels of GPX4 and HO-1 proteins between TMZ and TMZ + ABX groups. Increasing evidence has confirmed the significant roles of GPX4 and HO-1 proteins in inducing ferroptosis. Subsequently, western blotting was performed to examine the expression levels of GPX4 and HO-1 proteins, revealing a significant downregulation of GPX4 expression and a significant upregulation of HO-1 expression in GBM cells treated with combination therapy compared to other groups (Fig. 5i). The aforementioned findings were further confirmed through HO-1 and GPX4 immunohistochemical staining in vivo. The findings validated that the combination therapy of TMZ and ABX significantly augmented HO-1 and reduced GPX4 expression in comparison to other groups (Figure S8).

Furthermore, we evaluated the antitumor efficacy and DNA damage response in three distinct GBM PDOs models following indicated treatments. The PDOs were observed under a microscope following various treatments (control, mono- and combination treatment) for a duration of five days. As depicted in Fig. 6d, the architecture of organoids exhibited a relaxed and even disintegrated state following treatment with TMZ and ABX combination. Moreover, the diameter of organoids was significantly reduced in the TMZ plus ABX group compared to other groups. The GBM PDOs from each group were collected and fixed for immunofluorescence staining of Ki-67 and γ-H2AX, respectively. Compared to the control and mono-drug treatment groups, the combination of TMZ and ABX significantly down-regulated Ki-67 expression and up-regulated γ-H2AX expression in organoids. This suggests that combination treatment can effectively inhibit organoid growth while inducing sustained DNA damage (Fig. 6e).

Finally, drug treatment sensitivity tests were performed on a cohort of GBM PDOs comprising 16 patients. The findings indicate that a combination treatment regimen comprising ABX and TMZ leads to an improved inhibition rate in over 80% of GBM cases (n = 13), with approximately 37.5% of GBM cases (n = 6) exhibiting a fold change in inhibition rate improvement exceeding 2 (Fig. 6f). The combination treatment group exhibited a significantly higher mean inhibition rate compared to the TMZ mono-treatment group (Fig. 6g). We also conducted an analysis on the correlation between genetic background and response rate in the PDOs cohort. Our findings indicate that GBM patients with IDH1/2 or ATRX mutation exhibit a nearly 100% response to combination treatment of ABX and TMZ (Fig. 6h), suggesting that this particular genetic profile may benefit from such a treatment regimen.

Additional file 1:Table S1. Clinical and genetic characteristics of patients enrolled for the establishment of primary glioblastoma cells.Additional file 2:Figure S1. The IC50 concentration value of TMZ (a) and ABX (b) in four GBM cell lines. Figure S2. TMZ and ABX combination index in four GBM cell lines was calculated by using the Chou–Talalay Index. Figure S3. GBM cell morphology changes were induced by prestimulation or combination of low dose ABX compared with TMZ alone. Figure S4. Cell viability of GBM cells with ABX pre-treatment for 24 h and followed with TMZ treatment at different concentrations. Figure S5. Evaluation of toxic side effects of drug combinations in vivo. Figure S6. Western Blot analysis showed the effect of ABX pre-stimulation on DNA damage in GBM cells. Figure S7. Top, change of γ-H2AX foci in GBM cells treated with TMZ (400 µM for U87-MG and G353; 800 µM for LN229 and G393) with or without low-dose ABX (12nM) for 48 h. Bottom, statistics on the number of γ-H2AX foci at various time points following drug elution. Scale bar, 10µm. Figure S8. Typical images of γ-H2AX (a), HO-1 (b) and GPX4 (c) expression immunohistochemical analysis in tumor tissue slices from GBM-bearing mice. Scale bar in (a), 50µm; scale bar in (b) and (c), 20µm. Figure S9. (a) Morphological changes of U87-MG cells treated with TMZ (400 µM) with or without low-dose ABX (12nM) for 72 h. (b) The expression of γ-H2AX between TMZ and TMZ+ABX groups by Western blotting. Figure S10. DNA damage response is not involved in ferroptosis-mediated efficacy of drug combination. Figure S11. Identification of PDOs related indicators.Additional file 3. Supplementary Methods.