A ROS-Responsive Dual-Targeting Drug Nanocarrier Serving as a GSI Synergist and Ferroptosis Sensitizer for T-Cell Acute Lymphoblastic Leukemia

Abstract

T-cell acute lymphoblastic leukemia (T-ALL) is a highly aggressive hematological malignancy for which targeted therapies remain underdeveloped. Oncogenic mutations in Notch1 occur in up to 75% of T-ALL patients. Although γ-secretase inhibitors (GSIs) can block Notch1 activation, their clinical application is limited by side effects and reduced sensitivity. Here, a self-assembling, reactive oxygen species (ROS)-responsive nanotherapeutic strategy-PHD/G-NPs-co-loaded with GSI and controlled released dihydroartemisinin (DHA), and modified with a CD38 antibody is reported. The CD38 antibody specifically targets T-ALL cells, while GSI selectively inhibits Notch1, resulting in a dual-targeting approach. GSI is released first, inhibiting Notch1 activation and inducing the death of a subset of T-ALL cells. To eliminate semi-quiescent T-ALL cells that escape initial therapy by elevating ROS levels, a ROS-sensitive DHA delivery system is employed to enhance ferroptosis and boost GSI efficacy. After elucidating the mechanism of action of PHD/G-NPs in T-ALL cells, PHD/G-NPs are combined with αPD-1, which triggers an anti-tumor immune response in vivo. This dual-targeting strategy using CD38-modified PHD/G-NPs enables controlled drug release, enhances ferroptosis, mitigates GSI-induced gastrointestinal toxicity, and improves therapeutic efficacy. This nanomedical approach offers a novel strategy for targeted T-ALL treatment.

Keywords: T‐cell acute lymphoblastic leukemia (T‐ALL); dihydroartemisinin (DHA); ferroptosis; γ‐secretase inhibitors (GSIs).

Scheme 1

A) The self‐assembling, ROS‐responsive polymer pDMA‐pEPEMA‐pHD (PHD‐NPs) was constructed, featuring a hydrophilic DMA block, a hydrophobic EPEMA block, and a ROS‐responsive bridge linking HEA and DHA. PHD‐NPs were used to encapsulate GSI and subsequently modified with a CD38 antibody to form CD38‐PHD/G‐NPs. B) The CD38 antibody enabled the targeted identification of T‐ALL cells and helped mitigate the severe side effects associated with GSI treatment. GSI inhibited Notch1 activation and induced the death of a subset of T‐ALL cells. In the surviving ROS^high‐semi‐quiescent T‐ALL cells following GSI treatment, DHA was released via ROS‐responsive bridge degradation. The ferroptosis induced by DHA was further enhanced through activation of the MAPK signaling pathway, which was upregulated post‐GSI treatment. Additionally, damage‐associated molecular patterns (DAMPs) released during ferroptosis, together with αPD‐1 administration, triggered an anti‐tumor immune response.

Figure 1

Follow‐up treatment with DHA enhances GSI cytotoxicity. A) Proliferative activity of T‐ALL cells treated with different concentrations of GSI (n = 3 independent experiments). ns, not significant versus 10 or 20 µm concentrations. B) Flow cytometric analysis of ROS levels in T‐ALL cells following treatment with GSI (20 µm) at different time points. C) Quantification of mean fluorescence intensity (MFI) for ROS detection as shown in (B) (n = 3 independent experiments). D,E) Cell viability (CCK8 assay) of Molt4 (D) and Jurkat (E) cells after different treatments: GSI alone for 24 h, DHA alone for 24 h, GSI+DHA for 24 h, pre‐treatment with GSI for 24 h followed by co‐treatment with DHA for 24 h (n = 3 independent experiments). *, versus The GSI+DHA group. F) Scheme of synergistic interaction between DHA and GSI. Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2

Preparation and characterization of PHD/G‐NPs and CD38‐PHD/G‐NPs. A) Schematic diagram illustrating the preparation of PHD/G‐NPs. B) Particle size distribution of PHD/G‐NPs evaluated by DLS. C) Representative TEM image of PHD/G‐NPs. Scale bar = 500 nm. D) Particle size, PDI, and zeta potential values of PHD/G‐NPs. E) Zeta potential changes of PHD/G‐NPs after incubation in pH 7.4 PBS (with or without 1 mm H₂O₂) (n = 3 independent experiments). F) TEM images of PHD/G‐NPs after incubation in pH 7.4 PBS (with or without 1 mm H₂O₂). Scale bar = 500 nm. G) In vitro release profile of DHA from PHD/G‐NPs after incubation in pH 7.4 PBS (with or without 1 mm H₂O₂). H) Schematic diagram of ROS‐responsive DHA release from PHD/G‐NPs. Data represent mean ± s.d.

Figure 3

Effects of PHD/G‐NPs on inducing cell death in vitro. A) Quantification of MFI for ROS detection by flow cytometric analysis (n = 3 independent experiments). B) Cell viability of Molt4 (left) and Jurkat (right) cells treated with different nanoparticles measured by CCK8 assay (n = 3 independent experiments). *, versus The GSI‐NP group. #, versus The PHD‐NP group. C) Representative cell death analysis of T‐ALL cell lines after treatment with various formulations (20 µm). D,E) Statistical histograms showing the live cell percentages (n = 3 independent experiments) for Molt4 (D) and Jurkat (E) cells from (C). F,G) Statistical histograms showing the percentages of necrosis and early apoptosis in Molt4 (F) and Jurkat (G) cells from (C). Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. ns. p > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 4

PHD/G‐NPs induced facilitated ferroptosis by activating MAPK. A) Heatmap showing the relative expression of genes enriched in the Notch signaling pathway. B) KEGG enrichment analysis of differentially expressed genes (DEGs) compared with the control group. C) Representative TEM images of Molt4 cells after different treatments showing the degree of ferroptosis. Scale bar = 500 nm. D) Mean fluorescence value (MFV) analysis of C11 BODIPY staining in Molt4 (left) and Jurkat (right) cells following different treatments, assessed by flow cytometry (n = 3 independent experiments). E) Statistical histograms showing relative MDA levels in Molt4 (left) and Jurkat (right) cells after different treatments (n = 3 independent experiments). F) Immunofluorescence analysis of Fe²⁺ in Molt4 (left) and Jurkat (right) after the different administrations. Scale bar = 12.5 µm. Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

Effects of PHD/G‐NPs on anti‐leukemia activity and ferroptosis in vivo. A) Experimental schedule for in vivo antileukemic studies in Notch1‐induced T‐ALL mice. B,C) Photographs showing spleen size (B) and corresponding statistical histograms of spleen weight (C) across eight treatment groups. Scale bar = 1 cm. D) Flow cytometric analysis of GFP+ T‐ALL cell percentages in spleen samples. E) Flow cytometric analysis of GFP+ cell percentages in bone marrow samples. F) Microscopy images of H&E‐stained sections of major organs following different treatments. Scale bar = 150 µm. G) Microscopy images of H&E and PAS staining of the ileum at the end of treatment. Scale bar = 75 µm. H) Immunohistochemical analysis of MDA (left) and GPX4 (right) expression levels in femur tissue following different treatments. Scale bar = 75 µm. n = 5 mice per group. Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 6

Anti‐tumor immune responses of PHD/G‐NPs combined with αPD‐1 in vivo. A) Experimental schedule for different treatments in Notch1‐induced T‐ALL mice: 1, PBS (CON); 2, αPD‐1; 3, PHD/G‐NP; 4, PHD/G‐NP combined with αPD‐1. B,C) Photographs showing spleen size (B) and corresponding statistical histograms of spleen weight (C) across the four groups. Scale bar = 1 cm. D) Flow cytometric analysis of GFP+ T‐ALL cell percentages in spleens. E) Flow cytometric analysis of the GFP+ cell percentages in bone marrow samples. F,G) Flow cytometric analysis of the CD3+CD8+ T cell percentages among GFP‐ cells in spleen (F) and bone marrow (G). H–K) Statistical histograms showing percentages of CD3+CD4+ T cells (H), CD3+CD8+ T cells (I), CD86+CD80+ DCs (J), and NK1.1+ NK cells (K) among GFP‐ cells in spleen. L–O) Statistical histograms showing percentages of CD3+CD4+ T cells (L), CD3+CD8+ T cells (M), CD86+CD80+ DCs (N), and NK1.1+ NK cells (O) among GFP‐ cells in bone marrow. n = 5 mice per group. Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 7

Targeted delivery and effects of CD38‐PHD/G‐NPs in T‐ALL CDX model. A) Schematic diagram illustrating the preparation of CD38‐PHD/G‐NPs. B) Representative TEM image of CD38‐PHD/G‐NPs. Scale bar = 500 nm. C) Particle size distribution of CD38‐PHD/G‐NPs evaluated by DLS. D) FTIR spectra of PHD/G‐NPs and CD38‐PHD/G‐NPs. E,F) Comparison of particle size and PDI E) and zeta potential F) between PHD/G‐NPs and CD38‐PHD/G‐NPs measured by DLS (n = 3 independent experiments). G) Fluorescence images of T‐ALL cell lines incubated with free Cy5.5, Cy5.5‐NPs, and CD38‐Cy5.5‐NPs for 1, 2, or 4 h. Red indicates Cy5.5; blue indicates nuclei. Scale bar = 75 µm. H) in vivo fluorescence imaging of Molt4‐bearing mice treated with free Cy5.5, Cy5.5‐NPs, or CD38‐Cy5.5‐NPs at different time points. I) Ex vivo fluorescence imaging of femurs and major organs from Molt4‐bearing mice 24 h post‐administration. J) Flow cytometric analysis of Cy5.5 uptake in bone marrow cells from Molt4‐bearing mice at 24 h post‐administration. K) Schematic diagram of the xenotransplantation experiment. L) Flow cytometric analysis of human CD45+ T‐ALL cell percentages in spleens. M) Flow cytometric analysis of human CD45+ cell percentages in bone marrow. N) Body weight changes of mice treated with different nanoparticle formulations. O) Microscopy images of H&E staining, PAS staining, and immunohistochemical analysis of lysozyme expression in the ileum following different treatments. Scale bar = 75 µm. P) Kaplan–Meier survival curves of mice treated with different nanoparticle formulations. n = 5 mice per group. Data represent mean ± s.d. Two‐tailed Student's t‐tests were used to assess statistical significance. Survival analysis was performed using Kaplan–Meier survival plots and log‐rank tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.