γ-Glutamyl transpeptidase-activable nanoprobe crosses the blood-brain barrier for immuno-sonodynamic therapy of glioma

Bo Li; Gengjia Chen; Huihai Zhong; Tan Li; Minzhao Lin; Huiye Wei; Qiaoyun Zhang; Qi Chen; Jinsheng Huang; Xintao Shuai

doi:10.1038/s41467-024-54382-z

. 2024 Nov 29;15:10418. doi: 10.1038/s41467-024-54382-z

γ-Glutamyl transpeptidase-activable nanoprobe crosses the blood-brain barrier for immuno-sonodynamic therapy of glioma

Bo Li ^1,^✉,^#, Gengjia Chen ^2,^#, Huihai Zhong ¹, Tan Li ¹, Minzhao Lin ¹, Huiye Wei ¹, Qiaoyun Zhang ¹, Qi Chen ¹, Jinsheng Huang ^3,^✉, Xintao Shuai ^1,^✉

PMCID: PMC11607351 PMID: 39613729

Abstract

Effective treatment against glioma remains challenging nowadays because the protective blood-brain barrier (BBB) impedes drug penetration into brain and the limited efficacy of conventional chemotherapy. While strong positively charged nanoparticles have good permeability through the BBB, they often come with the caveat of cationic toxicity to healthy tissues and organs during blood circulation. Here we show a neutrally charged nanoprobe with a surface decorated with γ-glutamyl moieties that can be cleaved by γ-glutamyl transpeptidase, an enzyme overexpressed on brain capillaries. Upon the cleavage, positively charged primary amines are generated, facilitating the effective crossing of the nanoprobe through BBB via the adsorption-mediated transcytosis pathway, while avoiding the caveat of cationic toxicity. In addition, when reaching the acidic tumor microenvironment, the nanoprobe co-encapsulating sonosensitizer and immune agonist swells, which results in an accelerated drug release under ultrasound irradiation to induce a combined immune response, ultimately leading to a robust anticancer effect. Overall, we report an effective drug delivery nanoplatform across the BBB for an enhanced therapy of glioma.

Subject terms: Drug discovery, Chemical modification, Biomedical materials

Chemotherapy of glioma is limited by delivering drugs across blood-brain barrier (BBB) to tumor sites in brain. Here, this group designs a neutrally-charged nanoprobe with γ-glutamyl moieties decorated on the surface that enabling the effective delivery of sonosensitizer and immune agonist R848 across BBB through adsorptive mediated transcytosis, thereby eliciting the anticancer effects.

Introduction

Glioma is the most fatal intracranial malignancy nowadays, with a median survival of only 12–18 months in adults and a 5-year survival of less than 5%^1,2. To date, despite great advancements have been made in medicine, e.g., the emerging sonodynamic therapy (SDT) or immunotherapy applied in clinics brings about new opportunities for cancer patients^3,4. These new cutting-edge therapies display limited anti-tumor efficiency in glioma. This grim outcome is caused by various reasons, chief among them is the existence of blood-brain barrier (BBB) that blocks over 98% of therapeutic agents from entering the brain⁵.

Therefore, increasing the permeability of drugs across the BBB is recognized as the first and most crucial step for glioma treatment. Previous studies have discovered that drugs could be delivered into brain via the vesicular transport, a vectorial movement across the cerebral endothelial cells from the luminal to the abluminal side⁶. This type of transport system, known as transcytosis, is further divided into the specific (receptor-mediated transcytosis, RMT)^7,8 and nonspecific (adsorptive-mediated transcytosis, AMT) processes^9,10. Although several strategies based on RMT or AMT pathways have been explored to enhance drug delivery into the brain so far, few of these strategies proved to be noninvasive and effective. For example, attaching a targeting ligand on the drug carrier to recognize the specific receptor on brain capillary endothelial cells (BCECs), i.e., the RMT-based drug delivery, is mainly employed to facilitate the transport of drugs across the BBB in recent years. Nevertheless, due to the large size of currently available nano-agents¹¹, less than 1% of drugs could be ferried into brain via RMT^12,13. In contrast, AMT provides another route for drugs to penetrate the BBB. In the systemic physiological environment, the glycocalyx, consisting of the sialo-glycoconjugates and heparan sulfate proteoglycans expressed on the luminal surface of BCECs, contributes to an overall negative charge of BBB¹⁴. Therefore, cationized nanodrugs could trigger the adsorption-mediated endocytosis, i.e., AMT-based drug delivery, by electrostatic interactions between the positively charged moieties of the nanodrugs and the negatively charged membrane surface regions of the BBB¹⁵. Previous studies have proven that coupling the protamine, poly-L-lysine, polyamines, or other cationic substances to nanodrugs could significantly increase their BBB-crossing permeabilities^9,10. Nevertheless, it is worth noting that cationic substances exhibit a high affinity for anionic cell membranes found in all living cells⁶. Thus, cationized nanodrugs designed for targeting a desired site may result in an unpredictable distribution throughout the body after intravenous injection, bringing about toxic effects including possible endothelial damage, membrane nephropathy, immune complex-mediated glomerulonephritis, tissue edema or inflammation, and so on^10,16.

Another critical challenge in developing treatments against glioma involves increasing the antitumor effect. Conventional chemotherapy exhibits a poor antitumor effect on the drug-resistant glioma cells¹⁷. For instance, although temozolomide can cross the BBB and is therefore commonly employed in clinics to treat gliomas, the response rate of this treatment is only approximately 40% of patients owing to the expression of O6-methylguanine DNA methyltransferase¹⁸ and even goes down to 20% for those suffering from higher-grade gliomas¹⁹. Noteworthily, the burgeoning immunotherapy, capable of stimulating the host immune system to destroy cancer cells, brings about new opportunities for the management of various malignancies²⁰. However, it has been believed for many years that there was negligible interaction between the immune system and the brain parenchyma since the BBB limited the escape of antigens from and the entry of lymphocytes into the central nervous system. Nevertheless, this long-held belief was challenged by recent studies that have demonstrated the activated T cells could cross the BBB²¹ and meanwhile lymphocytes could enter the brain by expressing specific adhesion molecules²². Despite this progress, it is still unclear whether the immune system could be activated to resist orthotopic glioma.

In this work, we develop a γ-glutamyl transpeptidase (GGT)-activable nanoprobe that maintains a neutral surface charge (±10 mV) in the blood circulation, aiming to utilize the efficient cross of BBB via AMT-based drug delivery while avoiding the toxic side effects of cationic nanodrugs. Apolipoprotein E (ApoE) peptide, the tandem dimer sequence of the receptor-binding domain of apolipoprotein E that can specifically bind to the low-density lipoprotein receptor (LDLR) highly expressed on BBB, is incorporated into the nanoprobe to endow it with the function of BCECs binding²³. Besides, the γ-glutamyl moieties, which can be cleaved by the GGT overexpressed on BCECs to generate positively charged primary amines^24,25, are introduced into the surface layer of the nanoprobe (Supplementary Fig. 1). On the one hand, the nanoprobe bound on BCECs is expected to continually cross the BBB via AMT pathway attributing to the increasing positive surface charge. On the other hand, once the sonosensitizer ILD and immune agonist R848 co-encapsulated nanoprobe targets the tumor site, the “small to large” transition may be triggered by the acidic microenvironment, which further accelerates drug release owing to the enhanced cavitation effect under low-frequency ultrasound (LFUS) irradiation. Meanwhile, the insonated ILD may exert an SDT effect on glioma cells to induce the immunogenic cell death (ICD), and then activate the immune response with the aid of R848 for synergistic therapy of glioma (Fig. 1).

Fig. 1 — The process performs as follows: (I) The nanoprobe binds on brain capillary endothelial cells (BCECs); (II) GGT enzyme expressed on the outer membrane of BCECs cleaves the γ-glutamyl moieties on the surface of the nanoprobe to generate positively charged primary amines, which promotes the nanoprobe crossing BBB through adsorption-mediated transcytosis (AMT) mechanism; (III) After targeting to the tumor site, the “small to large” expansion of the nanoprobe in response to the acidic microenvironment of tumor tissue enhances the low-frequency ultrasound (LFUS)-induced cavitation effect and the release of encapsulated drugs.

Results

Polymer synthesis and nanoprobe preparation

This study aims to solve the two main challenges for the inefficiency of current glioma treatment: (1) the protective BBB impedes drug penetration into the brain, and (2) conventional chemotherapy exerts limited efficacy. To this end, we proposed a strategy of constructing a GGT-activable nanoprobe as a drug delivery nanoplatform across the BBB for an enhanced anti-glioma effect of immuno-sonodynamic therapy. Considering that the sufficient ICD of glioma cell induced by SDT could better activate the immune response assisted by resiquimod (R848), we further endowed the nanoprobe with the function of ultrasonic response to promote drug release and cellular internalization based on our previous study²⁶. Thus, the enzyme and pH dual-sensitive polymer BEAGA₂-PEG_2k-PDHD was synthesized as outlined in Supplementary Fig. 1. Boc-NH-PEG_2k-NH₂ was used as a macroinitiator to trigger the ring-opening polymerization of N-carboxy anhydride of β-benzyl L-aspartate (BLA-NCA). The polymerization degree of the thus-obtained PBLA block was calculated to be 85 based on the integral values of the benzyl group and the known molecular weight of PEG (2 kDa). Perfluorododecanoic acid (C₁₁F₂₃-COOH) was coupled to the polymer terminal to ensure the nanocarrier has a high-loading content of ultrasonic contrast agent fluorocarbon²⁷. Subsequently, the PBLA block of the polymer was aminolyzed by diethylethylenediamine (DEA), histamine (His), and N, N-diisopropylamino ethylamine (DIP) at a molar ratio of 20:10:55 to endow the nanocarrier with a sensitivity to the acidic tumor²⁶. The γ-glutamyl moieties (BEAGA), which could be hydrolyzed to primary amines by the GGT enzyme, were incorporated into another terminal of the polymer. Additionally, lysine was introduced to increase the coupling sites between the polymer and BEAGA. All the polymer structures mentioned above were verified by ¹H NMR or ¹⁹F NMR spectral analyses, which demonstrated the successful synthesis of the enzyme and pH dual-sensitive polymer BEAGA₂-PEG_2k-PDHD (Supplementary Figs. 2–9). Meanwhile, the pH-sensitive polymer, abbreviated as N₃-PEG_3.4k-PDHD with an azido terminal for subsequent DBCO-modified binding peptide coupling, was synthesized and confirmed by ¹H NMR and ¹⁹F NMR spectra (Supplementary Figs. 10–14).

Next, ILD was synthesized and used as a sonosensitizer by coupling a derivative of indocyanine green (ICG) with the T1-weighted MRI contrast agent Gd-chelated DTPA according to our previous study (Supplementary Fig. 15)²⁸. In addition to the bimodal MR/fluorescence imaging capacity, reactive oxygen species (ROS) may be produced by the ICG moiety under LFUS irradiation. Besides, as an FDA-approved immunomodulator, R848 is a hydrophobic imidazoquinoline-like molecule that can bind to toll-like receptors 7 and 8 (TLR7/8) and has already been applied in clinical trials for treating skin cancers²⁹. Then, the double-emulsion method was employed to prepare the nanoprobe, which co-encapsulated the sonosensitizer ILD, immune agonist R848, and ultrasonic contrast agent fluorocarbon (PFP/PFB) using N₃-PEG_3.4k-PDHD and BEAGA₂-PEG_2k-PDHD at a molar ratio of 1:4. The formulation of the nanoprobe was optimized by varying the feeding ratios of the cargoes over the polymers, by which the nanoprobe had a maximum ILD-loading content of 7.71 ± 0.55% and a R848-loading content of 1.27 ± 0.23% (Supplementary Table 1). Owing to the coupling of C₁₁F₂₃-COOH to the polymer terminal, the nanoprobe exhibited a high loading content of PFP/PFB (4.15 ± 0.42%). Furthermore, the peptide ApoE-DBCO, capable of specifically binding to LDLR expressed on brain endothelial cells and glioma cells, was coupled to the azido group of nanoprobe through a Cu-free click cycloaddition reaction. This reaction has high selectivity, high yielding, and rapid reaction kinetics, which ensured the ApoE coupling at a molar density of 20% for high targeting efficiency to the BBB and the glioma tissue^30,31. This tailor-made nanoprobe, termed BEAGA/ApoE-PDHD, is expected to exhibit a dual-responsive ability to both the GTT enzyme and the LFUS irradiation. In detail, GGT may catalyze the hydrolysis of γ-glutamyl moiety on the surface of the nanoprobe to produce a primary amine head, converting the neutrally charged nanoprobe (±10 mV) to a robustly cationic one (~ +30 mV). The later nanoprobe may further swell in response to the acidic microenvironment after targeting the tumor site, heightening the LFUS-induced cavitation effect of BEAGA/ApoE-PDHD (Fig. 2a).

As confirmed in Fig. 2b and Supplementary Fig. 16, GGT-catalyzed hydrolysis was relatively rapid in the initial time and then began to slow down. After incubation with 10 U mL⁻¹ GGT for 12 h, the zeta potential of BEAGA/ApoE-PDHD increased from 8.4 ± 1.9 mV to 29.1 ± 3.8 mV, which was higher than that incubated with 1 or 5 U mL^–1 GGT, thus confirming the GGT sensitivity of the nanoprobe. Afterwards, the pH sensitivity of BEAGA/ApoE-PDHD was also investigated. As shown in Fig. 2d, the particle size of the nanoprobe swelled from 210.08 ± 14 nm to 574.67 ± 36 nm at 37 °C by adjusting the pH from 7.4 to 6.5. The protonation of polymer in an acidic solution caused the hydrophobic segments to become hydrophilic ones, which induced the nanoprobe core swelling (Fig. 2c). TEM analysis further verified the morphological and size changes of the nanoprobe along with the change in solution pH (Fig. 2e). The swell of nanoprobes lowered the vaporization threshold of the PFP/PFB, leading to PFP/PFB gasification that possessed high echogenicity²⁶. As shown in Fig. 2e, the LFUS imaging signal of the nanoprobe solution was significantly intensified at pH 6.5. The high sensitivity of the nanoprobe under the LFUS irradiation is crucial for its therapeutic applications, i.e., the LFUS-controlled drug release. As confirmed in Fig. 2f, the ILD release was slightly faster at pH 6.5 than that at 7.4, which was much promoted upon LFUS irradiation. This was because the acoustic radiation forces and acoustic cavitation forces created by the nanoprobe with high echogenicity under LFUS irradiation can induce pressure and shock waves, a phenomenon known as cavitation effect. Such an effect resulted in an unstable expansion of the nanoprobe and ultimately led to its violent collapse, which caused rapid drug release³². The LFUS-triggered quick release of drugs improved the drug uptake by tumor cells, thereby potentially enhancing the therapeutic effect of SDT (Fig. 2g). Notably, the encapsulation of the bimodal molecular probe ILD endowed the nanoprobe with T1-weighted MR imaging function (Fig. 2h, i), providing a powerful imaging technique for clinical diagnosis of gliomas. Similarly, the other two control nanoprobes, i.e., (1) ApoE-decorated ApoE-PDHD nanoprobe without GGT responsiveness and (2) GGT-responsive BEAGA-PDHD nanoprobe without ApoE decoration, were prepared using single polymer N₃-PEG_3.4k-PDHD and BEAGA₂-PEG_2k-PDHD, respectively.

BBB-crossing ability of nanoprobe in vitro

The potential of the thus-obtained nanoprobes crossing BBB was then assessed in vitro. Western blotting analysis and immunohistochemistry were performed to investigate the GGT expression within the mouse brain²⁵. As shown in Fig. 3a, the expression levels of GGT protein were much higher in bEnd3 cells compared to RAW 264.7 and G422 cells, and mainly located on the capillary endothelium of the mouse brain (Fig. 3b), thus providing the potential of GGT in catalyzing BEAGA moieties present on our tailor-made nanoprobe to primary amines. Subsequently, an in vitro BBB model was established²⁸. A transwell filter with a mean pore size of 0.4 µm was seeded with a compact bEnd3 cerebral microvascular endothelial cell monolayer, while the bottom plate was seeded with a G422 glioma cell monolayer (Fig. 3c). When the transendothelial electrical resistance (TEER) value of the bEnd3 cell monolayer reached a stable level (Fig. 3d), the tight cell-cell junctions of this in vitro model could reasonably mimic the real BBB²⁸. As shown in Fig. 3e, f, after 4 h incubation, the G422 cells treated with BEAGA/ApoE-PDHD migrated from the apical chamber exhibited much stronger red fluorescence than those treated with BEAGA-PDHD or ApoE-PDHD. In other words, BEAGA/ApoE-PDHD crossed the tight BCECs monolayer much more effectively than BEAGA-PDHD and ApoE-PDHD did, which was then absorbed by the bottom plate-locating G422 cells. The in vitro BBB models composed of human or rat brain endothelial cells and glioma cells were further established (Supplementary Fig. 17). Similar results were obtained, which further confirmed the BBB-crossing potential of the BEAGA/ApoE-PDHD nanoprobe was universal and species-independent (Supplementary Fig. 18).

Fig. 3 — a Expression level of γ-glutamyl transpeptidase (GGT) in RAW 264.7, G422, and bEnd3 cells detected by western blot assay. b Immunohistochemical detection of GGT in capillary endothelium of the mouse brain (scale bar: 20 µm). c Schematic illustration of the in vitro BBB model. d Transendothelial electrical resistance (TEER) values of bEnd3 cell monolayer cultured in a Millipore Transwell 24-well Millicell Hanging Cell Culture Insert (0.4 µm PET; diameter: 6.5 mm), mean ± SD, n = 3 independent experiments. e The fluorescence images and (f) quantitative flow cytometry analysis of the down chamber-locating G422 cells incubated with BEAGA-PDHD, ApoE-PDHD or BEAGA/ApoE-PDHD having crossed the bEnd3 cell monolayer from the apical chamber (scale bar: 100 µm). g Z-series of confocal laser scanning microscope (CLSM) images showing the penetration of BEAGA/ApoE-PDHD pretreated with saline (control), 1, 5, or 10 U mL^–1 GGT in multicellular spheroids (scale bar: 200 µm; green: nanocarrier labeled by FITC; red: Cy3 dye standing for ILD). Right panel shows the representative fluorescent intensity profile towards the direction of white dotted arrows on spheroids.

The successful crossing of BEAGA/ApoE-PDHD through the in vitro BBB model encouraged us to further assess the BBB-crossing efficiency against various GGT concentrations using a three-dimensional cell sphere model composed of the GGT-absent bEnd3 cells and L929 as a basal cell. bEnd3 cell lines with permanent GGT knockdown (bEnd3-GGT/KD) were established using GGT-shRNA-Lentivirus and verified by western blot analysis (Supplementary Fig. 19). BEAGA/ApoE-PDHD nanoprobe with a FITC fluorophore-labeled PEG shell and a Cy3 dye-encapsulating core was pre-treated with different GGT concentrations, and then incubated with the cell spheres for half an hour. As shown in Fig. 3g, FITC or Cy3 fluorescence of BEAGA/ApoE-PDHD pre-treated with saline (control) or 1 U mL^–1 GGT was only detected on the cell sphere surface, suggesting that the nanoprobe could be endocytosed by a small number of peripheral cells, whereas BEAGA/ApoE-PDHD nanoprobe pre-incubated with 5 or 10 U mL^–1 GGT penetrated deeply into the cell sphere. In particular, the latter was completely absorbed by internal cells, resulting in a more uniform fluorescence distribution signal throughout the whole cell sphere. Profile analysis intuitively showed the differences in the penetration behavior of BEAGA/ApoE-PDHD nanoprobe treated with various concentrations of GGT. These results indicated that BEAGA/ApoE-PDHD penetrated the BBB-like endothelial tissue in a GGT concentration-dependent manner.

Potential to cross BBB and accumulate in glioma of mice

The BBB-crossing ability of various nanoprobes was directly evaluated using an intravital real-time confocal laser scanning microscope (CLSM) on the intracranial glioma-bearing mice. As shown in Fig. 4a, although the fluorescent ApoE-PDHD accumulated quickly alongside the wall of brain microvessels due to the binding function of ApoE, the fluorescence outside the cerebral vessels of mice injected with ApoE-free BEAGA-PDHD or ApoE-PDHD was hardly detected throughout the whole experiment time, suggesting that the ApoE decoration alone facilitated a low BBB-crossing efficiency³³. In contrast, the fluorescent BEAGA/ApoE-PDHD first accumulated on the wall of brain microvessels and then gradually permeated into brain parenchyma along with the increase in post-injection time. The extravasation of BEAGA/ApoE-PDHD from vasculature could be clearly observed at 60 min and much intensified at 120 min post-injection. The process of BEAGA/ApoE-PDHD crossing BBB was evidently shown in the Supplementary Movie 1. Three regions of interest (ROIs) of the brain parenchyma in each group were randomly chosen to quantitatively evaluate the BBB-crossing ability of various nanoprobes. Similarly, the mean fluorescence intensity of BEAGA/ApoE-PDHD in the selected regions increased sharply from 30 to 120 min post-injection (Fig. 4b). The result of the two-photon CLSM observation evidenced that the BBB-bindable and GGT-activatable BEAGA/ApoE-PDHD with an increasing positive surface charge under the catalysis of GGT could effectively cross BBB through AMT pathway.

Fig. 4 — a Sequential images of BEAGA-PDHD, ApoE-PDHD, and BEAGA/ApoE-PDHD nanoprobes gradually binding on and then crossing BBB on tumor-bearing mice (scale bar: 40 µm). Arrows indicate the enhanced fluorescent signals from ApoE-PDHD or BEAGA/ApoE-PDHD nanoprobe. b Change of the mean fluorescence intensity over time of BEAGA-PDHD, ApoE-PDHD, or BEAGA/ApoE-PDHD in the regions of interest (ROIs) of brain parenchyma, indicated as white rectangles in (a). Data are presented as mean ± SD, n = 3 ROIs of brain parenchyma. Statistical significance was calculated by a two-sided unpaired t-test, **p < 0.01. c In vivo fluorescence imaging of orthotopic G422 tumor-bearing mice i.v. administered various nanoprobes at different post-injection time points. d Ex vivo fluorescence imaging of main organs from mice in (c). e Confocal laser scanning microscope (CLSM) imaging of brain tissue sections from mice in (c). Blue: nuclei stained with DAPI; green: blood vessels stained using FITC-lectin; red: ICG of ILD encapsulated in various nanoprobes. White dotted lines mark the boundary between normal tissue and glioma (scale bar: 100 µm). f Transverse and (g) coronal sectional views of T1-weighted magnetic resonance (MR) imaging in pseudocolor mode on G422 glioma-bearing mice at pre-designed time points. White arrows indicated tumor location.

In consideration that ILD composed of ICG and DTPA-Gd subcomponents has both fluorescence and MR imaging functions, the in vivo fluorescence and MR imaging studies were employed to investigate the accumulation of nanoprobes on orthotopic glioma-bearing mice. In vivo fluorescence imaging showed that BEAGA/ApoE-PDHD exhibited a more efficient accumulation in tumor site than both the BEAGA-PDHD and ApoE-PDHD did (Fig. 4c). At 12 h post-injection, mice administered BEAGA/ApoE-PDHD showed much stronger ICG fluorescence in glioma than that injected with ApoE-PDHD, while mice receiving BEAGA-PDHD injection showed no detectable ICG fluorescence in tumor site (Supplementary Fig. 20a). Quantitative analysis of nanoprobe fluorescence in brain and other main organs was calculated, based on the ex vivo imaging shown in Fig. 4d. Although the nanoprobes were mainly distributed in liver, lung and kidney, about 15.57 ± 2.55% of BEAGA/ApoE-PDHD was still present in the brain, which was much higher than the levels reachable by both the ApoE-PDHD and BEAGA-PDHD (Supplementary Fig. 20b). Histological studies of isolated brains were performed and shown in Fig. 4e. Mice treated with BEAGA-PDHD showed very weak ICG fluorescence in tumor tissue, whereas mice receiving ApoE-PDHD showed strong ICG fluorescence just alongside the microvessel wall of tumor tissue. In contrast, mice injected with BEAGA/ApoE-PDHD exhibited bright red fluorescence that was evenly distributed within the tumor tissue rather than the normal tissue. The fluorescence signal intensities of ICG gradient from the blood vessel to the deep brain parenchyma tissue of mice treated with BEAGA/ApoE-PDHD were significantly increased, compared with that of mice treated with BEAGA-PDHD or ApoE-PDHD (Supplementary Fig. 21). Besides, the brain-tumor targetability of various nanoprobes were evaluated on U87 tumor-bearing nude mouse model and C6 tumor-bearing rat model and yielded consistent results (Supplementary Figs. 22 and 23), which confirmed that the brain tumor-targeting efficiency of BEAGA/ApoE-PDHD was not model dependent. In addition, the boundary identified by BEAGA/ApoE-PDHD probe between tumor tissue and normal brain tissue may provide a reference for intraoperative surgical management of glioma.

In vivo MRI scanning showed that mice tail vein injected with BEAGA-PDHD or ApoE-PDHD exhibited no appreciable increase in T1-weighted MRI intensity in the sectional view of transverse or coronal plane, whereas mice receiving BEAGA/ApoE-PDHD showed significantly enhanced T1-weighted MRI intensities in tumor site over 24 h postinjection (Fig. 4f, g and Supplementary Fig. 24). Quantitative analysis showed that mice treated with BEAGA/ApoE-PDHD had a 94.61% or 113.14% increase in the normalized T1-weighted signal intensity of tumor in the sectional view of transverse or coronal plane at 12 h post-injection, respectively. These results clearly indicated that the ingeniously designed BEAGA/ApoE-PDHD nanoprobe could effectively cross BBB through AMT pathway and then accumulate preferentially in tumor tissue owing to the targeting function of ApoE peptide³⁰.

In order to elucidate the clinical application potential of this nanoprobe, we further investigated the LDLR and GGT expression in various-grade human astrocytic gliomas (WHO grades I, II, III, and IV). As shown in Supplementary Fig. 25, GGT was strongly positive in the brain capillaries of all grades of human astrocytic gliomas, and higher-grade human astrocytic gliomas seemed to exhibit more GGT-positive expression. On the other hand, LDLR was overexpressed not only in brain endothelial cells but also in glioma cells. Thus, the incorporation of ApoE peptide into the nanoprobe endowed it with a dual-targeting ability to both the BCECs and the tumor cells. Noteworthily, other human malignancies such as hepatoma also exhibited high expression of GGT and LDLR in the vascular endothelial cells (Supplementary Fig. 26), revealing that BEAGA/ApoE-PDHD nanoprobe may be used to treat other tumors that overexpressed GGT and LDLR.

LFUS-triggered sonodynamic therapy in vitro

After verifying the effectiveness of the BBB-crossing and the tumor-targeting abilities of BEAGA/ApoE-PDHD, the therapeutic effect of drug-loaded BEAGA/ApoE-PDHD was assessed. Since ROS played a critical role in sonodynamic therapy^34,35, the intracellular ROS levels in G422 glioma cells incubated with R848-loaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@R848), ILD-loaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@ILD), and R848- and ILD- coloaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@ILD & R848) under LFUS irradiation were analyzed by 2’,7’-dichlorofluorescin diacetate (DCFH-DA) assay. As shown in Fig. 5a, the DCF fluorescence signal intensity in G422 glioma cells treated with BEAGA/ApoE-PDHD@ILD & R848 or BEAGA/ApoE-PDHD@ILD was much higher than that treated with BEAGA/ApoE-PDHD@R848 under LFUS irradiation. Moreover, the ROS generation in G422 glioma cells after various treatments was quantitatively analyzed with flow cytometry (Fig. 5b and Supplementary Fig. 27). The cells treated with ILD-loaded nanoprobe under LFUS generated a large amount of ROS, while cells treated with ILD and R848 co-loaded group showed the highest ROS level, which might be attributed to that the combination of ILD and R848 promoted the production of superoxide³⁶. These results demonstrated the effective ROS production of the sonosensitizer ILD irradiated by LFUS.

Fig. 5 — @R848, @ILD, and @R848 & ILD stand for BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@R848 & ILD, respectively. a Intracellular ROS generation of various formulation-treated G422 cells 1 h after low-frequency ultrasound (LFUS) irradiation (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min) evaluated by confocal laser scanning microscope (CLSM) observation (scale bar: 100 µm) or (b) flow cytometry analysis. c Viabilities of G422 cells incubated with different formulations for 12 h in the presence or absence of LFUS irradiation as determined by MTT assay. Data are presented as mean ± SD, n = 6 independent cell samples per group. Statistical significance was calculated by a two-sided unpaired t-test, ****p < 0.0001. d Calcein-AM and Propidium Iodide double staining assay of G422 cells 12 h after incubation with different formulations under LFUS irradiation (scale bar: 100 μm). e Quantitative analysis of G422 cell apoptosis using flow cytometry. f Mitochondrial membrane potential of G422 cells after various treatments monitored by CLSM observation (scale bar: 20 μm) or (g) flow cytometry analysis using JC-1 probe.

The cytotoxicities of BEAGA/ApoE-PDHD were then evaluated by MTT assay in G422 glioma cells. As shown in Supplementary Fig. 28, drug-loaded or -unloaded BEAGA/ApoE-PDHD showed negligible cytotoxicity in G422 glioma cells without LFUS irradiation. The cell viability still exceeded 90% at nanoprobe concentrations up to 800 µg mL⁻¹ after 12 h incubation. Conversely, after applying LFUS irradiation, viabilities of G422 cells incubated with BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 decreased to 50% ± 18% and 45% ± 12%, respectively (Fig. 5c), which could be attributed to the thus-generated ROS from ILD, as verified already in Supplementary Fig. 27. The living and dead G422 cells after SDT were further stained by Calcein-AM and Propidium Iodide (PI), respectively. As shown in Fig. 5d, G422 cells treated with BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 in the presence of LFUS irradiation exhibited bright PI-emitting red fluorescence representing dead cells. On the contrary, negligible red fluorescence was detected in cells treated with BEAGA/ApoE-PDHD@R848. The flow cytometry analysis quantitatively showed that the rates of apoptotic and necrotic cells induced by saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@ILD & R848 after LFUS irradiation were 5.07%, 3.90%, 56.69%, and 62.74%, respectively (Fig. 5e), thus further confirming the anti-tumor effect of the nanoprobe under LFUS irradiation.

As mitochondrion is an essential organelle of intracellular redox metabolism for cell survival, the membrane potential of mitochondrion reflects intracellular oxidative damage³⁷. The mitochondrion-specific fluorescent dye JC-1 is a commonly used indicator to monitor mitochondrial membrane potential based on the ratio between red (J-aggregates) and green (monomer) fluorescence. When the mitochondrial membrane potential remains high in healthy cells, the JC-1 dye accumulates on the cell membrane and emits red fluorescence. However, when the mitochondrial membrane potential decreases after oxidative damage, the JC-1 dye penetrates the cytoplasm and emits green fluorescence. As shown in Fig. 5f, bright red fluorescent dots were detected on the membrane of G422 cells treated with BEAGA/ApoE-PDHD@R848 under LFUS, whereas G422 cells treated with BEAGA/ApoE-PDHD@ILD or BEAGA/ApoE-PDHD@ILD & R848 after LFUS irradiation exhibited very weak red fluorescence on the membrane but strong green fluorescence in the cytoplasm, confirming the cell damage by oxidation. The percentages of J monomer-positive G422 cells treated with ILD-loaded BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 were 38.92% and 43.94%, respectively, which was much higher than that treated with ILD-free BEAGA/ApoE-PDHD@R848 via flow cytometry analysis (Fig. 5g).

Nanoprobe induced ICD under LFUS irradiation and activated immune cells

The above result confirmed that excessive ROS would damage and eventually kill tumor cells. In this process, endoplasmic reticulum stress was first triggered and subsequently led to the upregulation of a series of related “chaperone” proteins, such as calreticulin (CALR)³⁸. As verified in Fig. 6a, the CALR protein significantly increased in cells treated with BEAGA/ApoE-PDHD@ILD + LFUS group and BEAGA/ApoE-PDHD@ILD & R848 + LFUS group, which was positively related to the ROS generation. After staining the actin protein as cytoskeleton into red fluorescence, we found that the CALR protein migrated from the cytoplasm to the cell membrane. Meanwhile, two other damage-associated molecular patterns (DAMPs) biomarkers³⁹, i.e., adenosine triphosphate (ATP) and high mobility group box-1 (HMGB1), were also over-released from cells treated with either BEAGA/ApoE-PDHD@ILD + LFUS or BEAGA/ApoE-PDHD@ILD & R848 + LFUS (Fig. 6c, d).

Fig. 6 — @R848, @ILD and @R848 & ILD represent the BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@R848 & ILD, respectively. a Confocal laser scanning microscope (CLSM) images of calreticulin (CALR) exposure in G422 cells treated with different formulations (scale bar: 20 µm; blue: nuclei stained with DAPI; red: filamentous actin stained with phalloidin-Rhodamine; green: CALR labeled with FITC secondary antibody). b Schematic illustration of DCs maturation upon the stimulation of DAMPs induced by sonodynamic therapy (SDT). c, d The release of adenosine triphosphate (ATP) and high mobility group box-1 (HMGB1) in cell supernatant evaluated by ELISA. Data are presented as mean ± SD, n = 5 independent cell samples per group. e, f The expression levels of interleukin 12 (IL-12) and interferon-β (IFN-β) in CD11c⁺ BMDCs determined by ELISA. Data are presented as mean ± SD, n = 5 independent cell samples per group. No stimulus group referred to the BMDCs without treatment of glioma cell supernatant. g BMDCs maturation treated with various stimulations, determined by the expression of surface markers CD80 and CD86. For (c–f), statistical significance was calculated by a two-sided unpaired t-test. **p < 0.01; ***p < 0.001; ****p < 0.0001; and ns no significance.

The released DAMPs have been reported to enhance the immunogenicity and activate the innate immune system⁴⁰, during which an extra modulator is needed to provoke the potent adaptive immune responses. Therefore, we encapsulated R848, one of the agonists of TLR7/8, into the nanoprobe to activate antigen-presenting cells (APCs)²⁹. Liquid Chromatograph Mass Spectrometer (LC-MS) detection first confirmed that R848 remained intact and was successfully released after G422 ICD (Supplementary Figs. 29 and 30). Then the cell co-culture model was employed to examine the bone marrow-derived dendritic cells (BMDCs) maturation in vitro with the aid of R848 after SDT. Figure 6g showed that the combination of ILD and R848 after LFUS irradiation resulted in the highest level of DCs maturation (~56.29%) from immaturation, which was higher than that treated with BEAGA/ApoE-PDHD@R848 + LFUS or BEAGA/ApoE-PDHD@ILD + LFUS. Meanwhile, the matured DCs secreted multiple pro-inflammatory cytokines, essential for the activation of T cells to execute an antitumor immune response, including interleukin 12 (IL-12) and interferon-β (IFN-β) (Fig. 6e, f).

Overall, the insonated ILD exerted a sonodynamic effect on G422 cells, leading to the ICD and DAMPs release, including CALR, ATP, and HMGB1. The released DAMPs further stimulated APCs by binding to the pattern recognition receptors with the aid of R848⁴¹, which promoted the activation of APCs and up-regulated their co-stimulatory factors CD80 and CD86, the hallmarks of DCs maturation. As a result, the matured DCs secreted multiple pro-inflammatory cytokines, such as IL-12 and IFN-β, thus helping to convert the immunosuppressive tumor microenvironment into an immunogenic one^42,43. This conversion may ultimately elicit an adaptive antitumor immune response, leading to the elimination of tumor cells (Fig. 6b).

In vivo therapeutic efficacy on glioma-bearing mice

The successful delivery of the nanodrug into brain tissue encouraged us to further assess the potential of our BBB-crossing strategy in treating orthotopic glioma. Mice bearing G422 tumor were randomly assigned to 6 groups (n = 5) receiving treatments of saline, BEAGA-PDHD@ILD & R848, ApoE-PDHD@ILD & R848, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@ILD & R848, respectively. Different formulations were intravenously administered into mice at 6 days after tumor cell inoculation and carried out once every 3 days for a total of 4 times. Ultrasonic insonation was applied in all treatment groups at 12 h post-administration (Fig. 7a). The glioma growth was monitored every 3 days using the bioluminescence imaging, and the fluorescence intensities of tumor sites were calculated to compare tumor growth between different groups (Fig. 7b). As shown in Fig. 7c, mice treated with BEAGA/ApoE-PDHD@ILD & R848 exhibited the most effective tumor growth inhibition in comparison with all other treatment arms. 60% of mice in this group survived more than 36 days, while all mice in the other treatment groups died within 31 days (Fig. 7d). Specifically, compared with the BEAGA/ApoE-PDHD@ILD treatment group, mice treated with BEAGA/ApoE-PDHD@ILD & R848 showed no significant increase in tumor volume before and after drug withdrawal (15 days after tumor cell inoculation), which might be attributed to the strong immune response suppressing the tumor growth for a long time. As verified in Supplementary Fig. 31, BEAGA/ApoE-PDHD@ILD & R848 significantly increased the proportions of central memory (CD44⁺CD62L⁺) and effector memory (CD44⁺CD62L⁻) T cells, while decreasing the proportion of naive T cells (CD44⁻CD62L⁺), indicating that the activation of the immune system holds the capacity to exert a durable suppressive effect on glioma⁴⁴.

Irrespective of the treatment arm, no difference in the body weight of the mice was noted throughout the experiment (Fig. 7e). Hematoxylin/eosin (H&E) staining of main organ tissues from tumor-free mice displayed no arresting pathological changes (Supplementary Fig. 32), indicating the biosafety of the nanoprobes. Besides, Nissl staining of brain tissues after various treatments showed that no “dark” neuron in the cortex and hippocampus was caused by the ultrasonic irradiation, suggesting negligible neuron degeneration (Supplementary Fig. 33)²³. Moreover, expressions of the astrocytic marker GFAP (glial fibrillary acidic protein) and microglial cytoplasmic marker Iba1 (ionized calcium-binding adapter molecule 1) in the LFUS-irradiated brain tissues exhibited no significant difference compared with the control group (Supplementary Figs. 33 and 34). These results implied that the mechanical forces of ultrasonic cavitation did not cause irreversible inflammation and brain tissue damage.

H&E and Ki67 immunostaining were also carried out to investigate the histological changes of gliomas in mice. As shown in Fig. 7f, in line with the tumor growth inhibition, the tumor sections from mice receiving treatment of BEAGA/ApoE-PDHD@ILD & R848 after LFUS irradiation presented the least tumor cells and the lowest Ki67 expression levels. These results demonstrated that the effective BBB-crossing nanodrug BEAGA/ApoE-PDHD could remarkably improve the therapeutic effects of ILD in orthotopic glioma-bearing mice. Besides, the introduction of R848 into the nanodrug might further enhance the antitumor effect by activating the immune response.

Anti-tumor immune response on mice after SDT

The in vivo immune response after various treatments was subsequently investigated. In line with the in vitro immune activation process (outlined in Fig. 6b), the insonated ILD first caused the ICD and DAMPs (e.g., CALR, ATP, and HMGB1) release of G422 glioma cells (Fig. 8a and Supplementary Fig. 35a, b). The thus-generated DAMPs and R848 further induced the maturation of DCs, as shown in Fig. 8b and Supplementary Fig. 36a. The BEAGA/ApoE-PDHD@ILD & R848 treatment group exhibited the highest proportion of DC maturation (32.58 ± 3.42%) as measured by flow cytometry. The matured DCs, as the dominant partners to T cells, secrete cytokines such as IL-12 to link innate and adaptive immune responses⁴⁵. Based on these findings, we analyzed the expression level of IL-12 in tumors of mice treated with various formulations. The ELISA result showed that the IL-12 expression was significantly increased in gliomas treated with BEAGA/ApoE-PDHD@ILD & R848. In addition, IFN-β is also crucial for driving cytotoxic T lymphocyte activity and recruiting CD8⁺ T cells into tumor tissue (Fig. 8f). Immunohistochemistry staining showed that the CD8⁺ T cells were massively distributed in tumor and peri-tumor areas after BEAGA/ApoE-PDHD@ILD & R848 treatment, but few were detected in tumor of mice administrated other formulations (Fig. 8c). These findings were confirmed quantitatively by flow cytometry. As shown in Fig. 8d and Supplementary Fig. 36b, the proportion of CD8⁺ T cells over the total T cells (CD45⁺CD3⁺) within tumors treated with BEAGA/ApoE-PDHD@ILD & R848 was up to 26.81 ± 2.68%, which was several times higher than that of the control group; while the Foxp3⁺ Treg cell (CD45⁺CD3⁺CD4⁺) showed the opposite trend (Fig. 8e). Furthermore, among all treatment groups, the BEAGA/ApoE-PDHD@ILD & R848 nanodrug led to the highest up-regulation of tumor necrosis factor-α (TNF-α) and interferon-γ (IFN-γ) (Fig. 8f), both of which are necessary for effective antitumor immunity. In addition, although the BEAGA/ApoE-PDHD@ILD nanodrug with LFUS irradiation was able to kill glioma cells via SDT and release DAMPs in the first stage, it was powerless to recruit more T cells into the tumor site. This constraint appeared to be mitigated by the integration of R848 into the BEAGA/ApoE-PDHD@ILD nanodrug formulation, which conspicuously elicited a robust secretion of C-X-C motif chemokine ligand (CXCL)-9 and CXCL-10 chemokines (Supplementary Fig. 37a, b), thereby reinforcing the recruitment of T cells, particularly CD8⁺ cytotoxic T lymphocytes, into the tumor site. Moreover, we delved into the impact of BEAGA/ApoE-PDHD@ILD & R848 treatment on TAMs and neutrophils within glioma tissue. Post-treatment, a notable decline in neutrophil population was observed within the tumor, whereas the total TAM count remained stable but was accompanied by a marked increase in M1 macrophages (Supplementary Figs. 38 and 39). These results demonstrated that the BEAGA/ApoE-PDHD@ILD & R848 nanodrug could induce a superior DC maturation and further amplify the antitumor immune response.

Fig. 8 — a Confocal laser scanning microscope (CLSM) images of calreticulin (CALR) expression of tumor tissues after various treatments (scale bar: 50 µm; blue: nuclei stained with DAPI; red: CALR labeled with FITC secondary antibody). b Flow Cytometry (FCM) analysis of DCs maturation (gated on live CD45⁺CD11c⁺ cells) within tumors in various treatment groups. c Immunohistochemical analysis of the CD8⁺ T cells (brown) in tumor tissues (scale bar: 100 µm). d FCM analysis of tumor-infiltrating lymphocytes (gated on live CD45⁺CD3⁺ cells) and e Treg (gated on live CD45⁺CD3⁺CD4⁺ cells) after different treatments. f Expression levels of interleukin 12 (IL-12), interferon-β (IFN-β), tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ) in tumors detected by ELISA. Data are presented as mean ± SD, n = 5 mice per group. Statistical significance was calculated by a two-sided unpaired t-test. **p < 0.01; ***p < 0.001; ****p < 0.0001; and ns no significance.

Discussion

The blood-brain barrier acts like a bastion of iron which protects the central nervous system but also hinders the treatment of brain diseases such as glioma. To date, various strategies have been explored to enhance the efficiency of drug delivery into the brain. Thereinto, the intracranial drug entering pathways based on AMT and RMT are mostly employed. Unfortunately, neither of them has proven to be noninvasive and effective. On one hand, owing to the large size of currently available nano-agents, less than 1% of drugs could be ferried into the brain via RMT^12,13. Our previous study constructed a water-soluble molecular probe capable of effectively crossing BBB due to its small molecular size and the α_vβ₃ integrin receptor-mediated transcytosis²⁸. However, this strategy is incompetent for the delivery of the coupling group-absent drugs or those with an inherent particle size such as Abraxane. On the other hand, cationized nanoparticles could trigger the AMT pathway and facilitate the BBB crossing of drugs through an electrostatic interaction between the positively charged moieties and negatively charged BBB. However, this indiscriminate interaction with anionic cell membranes resulted in an unpredictable distribution of the cationized nanoparticles throughout the body after systemic administration, ultimately shortening the nanoparticles’ half-lives and leading to severe toxic effects^10,16. Besides, traditional chemotherapy performs a poor antitumor effect on drug-resistant glioma cells. As a result, gliomas tend to have a low cure rate, high mortality rate, and high recurrence rate.

Aiming to overcome this challenge, in the present study, we further proposed a BBB-bindable nanoprobe with a decoration of the well-designed γ-glutamyl moiety that could be activated by the GGT enzyme overexpressed on BCECs. Upon binding on BCECs, the γ-glutamyl moiety on the surface of the nanoprobe was continually cleaved by GGT, which generated positively charged primary amines for triggering the AMT pathway. In fact, we evidenced that the nanoprobe could efficiently bind on and then pass through the BBB of mice using an intravital real-time CLSM. Approximately 15% of the nanoprobe crossed BBB and entered the brain parenchyma, which was much higher than the levels reachable by the traditional pathways^12,13. By such means, this nanoprobe could transport the sonosensitizer ILD and the immune agonist R848 into the brain tumor site, which exerted a sonodynamic effect in the first stage and elicited an immune response subsequently. Furthermore, the ILD and R848 co-loaded nanodrug (i.e., BEAGA/ApoE-PDHD@ILD & R848) exhibited robust antitumor immunity. Compared to the control group, BEAGA/ApoE-PDHD@ILD & R848 significantly increased the proportion of activated DCs by approximately 20% within the tumor tissues. These DCs secreted multiple pro-inflammatory cytokines including IL-12 and IFN-β, which further recruited more tumor-infiltrating CD8⁺ T cells to precisely kill the glioma cells. As a result, mice administered BEAGA/ApoE-PDHD@ILD & R848 had more prolonged survival time than those in the control group even after drug withdrawal.

Taken together, this BBB-bindable and GGT-activatable nanoprobe in response to the special characteristics of BBB effectively crosses BBB through the AMT pathway to achieve a robust anti-tumor effect on glioma-bearing mice, while tactfully bypassing the cationic toxicity of cationized nanoparticles. Compared to the conventional well-documented nanodrugs, this strategy may highlight a tremendous potential in brain disease treatments which critically depends on effective BBB-crossing drug delivery.

Methods

Ethical statement

This research complies with all relevant ethical regulations. All surgical interventions conducted, as well as the subsequent postoperative care provided to the animals, were vetted and approved by the esteemed Institutional Animal Care and Use Committee of Sun Yat-sen University, Guangzhou, China. The assigned approval number of the ethical application of animal experiments is 20220512-00036. The experiment was designed without considering the sex of the animals, and female mice or rats were selected in this study to ensure gender uniformity. Human hepatoma tissue sections (SHXC2021YF01) and human glioma tissue microarray sections (HBraG149Su01) were obtained from Shanghai Outdo Biotech Co., Ltd. (Shanghai, China). The studies were reviewed and approved by the Ethics Committee of this company.

Materials

The sonosensitizer ILD containing T1 MR contrast agent Gd-chelated DTPA was synthesized according to our previous study²⁸. ApoE(159-167)₂ peptide with a dibenzocyclooctyne group on N-terminal (ApoE-DBCO, sequence: (LRKLRKRLL)₂C-DBCO, 95%) was obtained from Wuhan Holder Co., Ltd. (Wuhan, China). Boc-NH-(polyethylene glycol)-2000 amine (Boc-NH-PEG_2k-NH₂) and azido-(polyethylene glycol)-3400 amine (N₃-PEG_3.4k-NH₂) were purchased from GuangZhou Tanshui Technology Co., Ltd. Perfluorododecanoic acid (C₁₁F₂₃-COOH), resiquimod (R848), N, N-diethylethylenediamine (DEA), N, N-diisoprylamino ethylamine (DIP) and histamine (His) were purchased from Aladdin Industrial Corporation (Shanghai, China) and used as received. N-α, N-ε-di-Fmoc-D-lysine (Fmoc-Lys), Boc-L-glutamic acid 1-tert-butyl ester (Boc-Glu) and N-alpha-(9-fluorenylmethoxycarbonyl)-L-2-amino-butanoic acid (Fmoc-ABU-OH) were purchased from J&K Scientific Ltd. (Beijing, China). The dialysis bag was purchased from Green Bird Technology Development Co., Ltd. (Shanghai, China). Na₂CO₃ solid, anhydrous MgSO₄ solid, NaCl solid, and diethyl ether were purchased from Chemical Reagent Factory (Guangzhou, China). Petroleum ether, ethyl acetate, chloroform (CHCl₃), and dichloromethane (CH₂Cl₂) were dried over CaH₂ and distilled. Anhydrous dimethylsulfoxide (DMSO), anhydrous dimethylformamide (DMF), perfluoropentane (PFP), and 1,1,1,3,3-pentafluorobutane (PFB) were purchased from Sigma-Aldrich (Prague, Czech Republic).

Roswell Park Memorial Institute (RPMI) 1640 Medium, Dulbecco’s Modified Eagle Medium (DMEM), penicillin/streptomycin (100X), and fetal bovine serum (FBS) were purchased from the reputable Thermo Fisher Scientific. 4’,6-diamidino-2-phenylindole (DAPI), ROS probe 2’,7’-dichlorofluorescin diacetate (DCFH-DA) and mitochondrial membrane potential probe (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-imidacarbocyanine iodide, JC-1) were purchased from Beyotime (China). Live-Dead Cell Staining Kit was purchased from KeyGEN BioTECH (China). The mouse ELISA kits for ATP, HMGB1, IL-12, IFN-β, IFN-γ, CXCL-9, and CXCL-10 measurements were purchased from Shanghai Westang Bio-Tech Co., LTD (Shanghai, China). The mouse anti-GAPDH antibody (ab8245, 6C5, GR3275542-1), mouse anti-GGT antibody (ab55138, 1F9, GR1017756-3), rabbit anti-Calreticulin antibody (ab2907, GR3426057-2), rat anti-CD8 antibody (ab25478, 53-6.7, GR3181210-4), goat anti-rabbit IgG H&L (HRP) (ab6721), goat anti-rabbit IgG H&L (Alexa Fluor® 488) (ab150077), and goat anti-rabbit IgG H&L (Alexa Fluor® 555) (ab150078) were purchased from Abcam (Shanghai, China). The rabbit anti-Iba1 antibody (GB153502, AC221117065), rabbit anti-GFAP antibody (GB15100, AC231115002), rabbit anti-CD31 antibody (GB11063, AC230814049), rabbit anti-GGT antibody (GB113421, AC231008052), and rabbit anti-LDLR antibody (GB11369, AC231017018) were provided by Wuhan Servicebio Technology Co., LTD (Wuhan, China). The armenian hamster FITC anti-CD11c antibody (117306, N418, B356966), armenian hamster PE anti-CD80 antibody (104707, 16-10A1, B340153), rat APC anti-CD86 antibody (105011, GL-1, B323580), rat BV421 anti-CD45 antibody (103134, 30-F11, B370002), rat APC anti-CD3 antibody (100236, 17A2, B375907), rat FITC anti-CD4 antibody (100406, GK1.5, B374032), rat PerCP anti-CD8 alpha antibody (100732, 53-6.7, B386876), rat PE anti-CD62L antibody (104407, MEL-14, B242685), rat FITC anti-CD44 antibody (103006, IM7, B323775), Zombie Yellow™ (423103, B330652), rat BV510 anti-CD45 antibody (103138, 30-F11, B402130), rat APC/Cyanine7 anti-CD11b antibody (101225, M1/70, B397990), rat BV421 anti-F4/80 antibody (123137, BM8, B416422), rat PE anti-Ly-6G antibody (127607, 1A8, B397726), armenian hamster FITC anti-CD80 antibody (104705, 16-10A1, B354433), rat PE/Cyanine7 anti-CD206 antibody (141719, C068C2, B399589), and rat anti-CD16/32 antibody (101302, 93, B269215) were purchased from Biolegend Company (Beijing, China). The rat PE anti-Foxp3 antibody (12-5773-82, FGK-16s, 2510310) was purchased from Invitrogen, Fisher Scientific (Shanghai, China). All additional reagents employed in this study were sourced from commercial vendors and conformed to analytical purity standards or exceeded them.

Synthesis of BEAGA₂-PEG_2k-PDHD

Synthesis of Boc-NH-PEG_2k-PBLA₈₅

As outlined in Supplementary Fig. 1, the diblock copolymer Boc-NH-PEG_2k-PBLA₈₅ was synthesized by ring-opening polymerization of β-benzyl L-aspartate (BLA-NCA) using Boc-NH-PEG_2k-NH₂ as a macroinitiator according to a previous study⁴⁶.

Concisely stated, 1.00 g of Boc-NH-PEG_2k-NH₂ (0.50 mmol) was subjected to vacuum drying for 1 h at 70 °C within a 250 mL Schlenk flask, which was equipped with a magnetic stirrer. Once the temperature had safely decreased to 35 °C, 100 mL of anhydrous CH₂Cl₂ and a solution of 10.60 g of BLA-NCA (42.50 mmol) dissolved in 10 mL of anhydrous DMF were carefully introduced into the reaction system under an inert argon atmosphere. The reaction was allowed to proceed at a controlled temperature of 35 °C for 48 h. Subsequently, the reaction mixture was precipitated into a copious amount of chilled diethyl ether, filtered to separate the solid product, washed thoroughly with diethyl ether three consecutive times, and finally dried under vacuum until a constant weight was achieved (8.74 g, 0.45 mmol, 90% yield).

Synthesis of Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃

Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃ is synthesized by amidation of Boc-NH-PEG_2k-PBLA₈₅ and C₁₁F₂₃-COOH. First, the carboxyl group of C₁₁F₂₃-COOH was activated. 1.60 g of C₁₁F₂₃-COOH (2.60 mmol, 10 eq), 0.30 g of NHS (2.60 mmol, 10 eq), and 0.50 of EDC•HCl (2.60 mmol, 10 eq) were dissolved in 20 mL of distilled CHCl₃ in a 50 mL Schlenk flask under argon protection. The reaction was stirred for 12 h at room temperature. Afterwards, 50 mL of anhydrous DMSO containing 5.05 g of Boc-NH-PEG_2k-PBLA₈₅ (0.26 mmol, 1 eq) was added into the mixture. Then, 20 μL of distilled triethylamine was added and the reaction was allowed to proceed for another 2 days. Finally, the solution was dialyzed (MW cut-off: 3500 Da) against methanol for 2 days and dried under vacuum to yield white powder Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃ (5.01 g, 0.25 mmol, 95% yield).

Synthesis of Boc-NH-PEG_2k-PAsp(DEA-co-His-co-DIP)₈₅-C₁₁F₂₃ (Boc-NH-PEG_2k-PDHD)

A total of 5.00 g of Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃ (0.25 mmol) was dissolved in 20 mL of anhydrous DMSO under argon protection. Then, 0.58 g of DEA (5.00 mmol, 20 eq) was added and the solution was allowed to react for 8 h at 35 °C. Next, 0.28 g of His (2.50 mmol, 10 eq) was added and continued to stir for 12 h. Following the addition of DIP (1.80 g, 12.50 mmol, 50 eq), the reaction was allowed to proceed for an additional 12 h under identical conditions. Subsequently, the product was subjected to dialysis against methanol for 48 h, utilizing a membrane with a molecular weight cut-off of 3500 Da. The dialyzed solution was then concentrated via rotary evaporation, resulting in the isolation of Boc-NH-PEG_2k-PAsp(DEA-co-His-co-DIP)₈₅-C₁₁F₂₃, abbreviated as Boc-NH-PEG_2k-PDHD (4.89 g, 0.22 mmol, 90% yield).

Synthesis of Fmoc-Lys-PEG_2k-PDHD

First, the protective group of Boc-NH-PEG_2k-PDHD was removed. In brief, 5.00 g of Boc-NH-PEG_2k-PDHD (0.22 mmol) was dissolved in 20 mL of trifluoroacetic acid (TFA) at room temperature. After stirring for 2 h, TFA was removed by rotary evaporation and the residue was dissolved in DMSO. The solution was basified with 50 μL of triethylamine and dialyzed (MW cut-off: 3500 Da) against methanol for 48 h, then dried under rotary evaporation to yield NH₂-PEG_2k-PDHD. Afterwards, the carboxyl group of Fmoc-Lys was activated and then reacted with NH₂-PEG_2k-PDHD, using the same preparation method for Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃. Finally, white powder Fmoc-Lys-PEG_2k-PDHD was obtained (4.31 g, 0.19 mmol, 85% yield).

Synthesis of (Fmoc-ABU)₂-PEG_2k-PDHD

Similarly, the protective group of Fmoc-Lys-PEG_2k-PDHD was removed. A total of 5.00 g of Fmoc-Lys-PEG_2k-PDHD (0.22 mmol) was dissolved in 40 mL of anhydrous DMSO, followed by the addition of 10 mL of piperidine while maintaining a piperidine concentration of approximately 20% throughout the solution. After stirring for 0.5 h at room temperature, the solution was dialyzed (MW cut-off: 3500 Da) against methanol for 24 h, then dried under rotary evaporation to yield Lys-PEG_2k-PDHD. Afterwards, the carboxyl group of Fmoc-ABU-OH was activated and then reacted with Lys-PEG_2k-PDHD, using the same preparation method for Fmoc-Lys-PEG_2k-PDHD as mentioned above. Finally, white powder (Fmoc-ABU)₂-PEG_2k-PDHD was obtained (4.34 g, 0.19 mmol, 88% yield).

Synthesis of the enzyme and acid dual-responsive polymer BEAGA₂-PEG_2k-PDHD

First of all, the protective group of amine in the polymer (Fmoc-ABU)₂-PEG_2k-PDHD (4.57 g, 0.20 mmol) was removed by the addition of piperidine. Then, the carboxyl group of Boc-Glu was activated and reacted with the exposed amino groups in DMSO. The solution was dialyzed (MW cut-off: 3500 Da) against methanol for 24 h, and dried under rotary evaporation to yield (Boc-Glu-ABU)₂-PEG_2k-PDHD. The obtained polymer was next dissolved in TFA to remove the Boc group. After purification, the enzyme and acid dual-responsive polymer BEAGA₂-PEG_2k-PDHD was finally obtained. The relevant experimental procedures involved have been described above and will not be detailed here (4.05 g, 0.17 mmol, 85% yield).

Synthesis of N₃-PEG_3.4k-PDHD

Synthesis of the acid-responsive polymer N₃-PEG_3.4k-PDHD

The synthetic process of N₃-PEG_3.4k-PDHD was outlined in Supplementary Fig. 10 and was same as that of Boc-NH-PEG_2k-PDHD. Briefly, N₃-PEG_3.4k-NH₂ was used as a macroinitiator to trigger the ring-opening polymerization of BLA-NCA to obtain the diblock copolymer N₃-PEG_3.4k-PBLA₈₅. Then the carboxyl group of C₁₁F₂₃-COOH was activated and reacted with N₃-PEG_3.4k-PBLA₈₅ to yield N₃-PEG_3.4k-PBLA₈₅-C₁₁F₂₃. After aminolyzed by DEA, His, and DIP at a molar ratio of 20:10:55, the white powder N₃-PEG_3.4k-PAsp(DEA-co-His-co-DIP)₈₅-C₁₁F₂₃, abbreviated as N₃-PEG_3.4k-PDHD, was finally obtained.

Preparation of nanoprobes

Considering the pharmacodynamic concentrations of R848 and ILD as well as the low encapsulation rate of R848, we used ILD and R848 as the feeding ratio of 5:1 (m/m) to prepare the nanoprobe BEAGA/ApoE-PDHD@ILD & R848^29,47. In detail, a 1:4 molar ratio of N₃-PEG_3.4k-PDHD and BEAGA₂-PEG_2k-PDHD (a total weight of 40 mg) as well as 1 mg of R848 were dissolved in 0.5 mL of DMSO. Then, 200 μL of deionized water containing 5 mg of ILD was slowly dropped into the solution under sonication in an ice bath, followed by the addition of 200 μL of PFP/PFB (v/v = 2:1)²⁶. Subsequently, the emulsions were dripped into 20 mL of deionized water, while being sonicated in an ice bath to maintain a low temperature. Large aggregates were removed using a syringe filter with a pore size of 0.45 μm. The filtered mixture was then dialyzed against PBS (pH 7.4) at 15 °C for 12 h, ensuring the effective removal of DMSO and any unreacted polymer. Afterwards, ApoE-DBCO, in an equimolar amount to N₃-PEG_3.4k-PDHD, was introduced into the solution and allowed to incubate for an additional 4 h to facilitate conjugation. The resulting solution was concentrated and purified by washing it three times with PBS (pH 7.4) using a MILLIPORE Centrifugal Filter Device, set at a molecular weight cut-off of 100 kDa, a rotational speed of 18 g, and a temperature of 4 °C. The solution was then passed through a syringe filter with a pore size of 0.45 μm, eliminating any remaining large aggregates. The enzyme and acid programmed responsive nanoprobe BEAGA/ApoE-PDHD@ILD & R848 was finally yielded. Similarly, polymer N₃-PEG_3.4K-PDHD was treated likewise to obtain BBB-bindable nanoprobe ApoE-PDHD@ILD & R848 after adding ApoE-DBCO. And BEAGA₂-PEG_2K-PDHD was used to prepare enzyme-responsive nanoprobe BEAGA-PDHD@ILD & R848.

Characterization of polymers and nanoprobes

¹H NMR and ¹⁹F NMR spectra were recorded on a Varian Unity 400 MHz Spectrometer (Bruker, Switzerland) in various deuterated solvents. Dynamic light scattering (DLS) on a 90 Plus/BI-MAS instrument (Brookhaven Instruments Corporation, USA) was used to measure the size and zeta potential of nanoprobe at pH 7.4 or 6.5. Transmission electron microscopy (TEM) images were obtained from a model H-7650 TEM (Hitachi Ltd, Tokyo, Japan) operated at 120 kV. The samples were prepared by depositing a precise drop of the sample solution (10 μL, 1 mg/mL) onto a copper grid that had been previously coated with an amorphous carbon layer. Following this, a minute quantity of phosphotungstic acid aqueous solution (2 wt%) was dispensed onto the grid surface. After allowing for a brief interaction period of 1 min, excess solution was gently blotted away using a piece of filter paper. The copper grid was left to dry overnight within a desiccator, maintaining a controlled, moisture-free environment, prior to being subjected to observation.

Next, the GGT-catalyzed surface charge enhancement of BEAGA/ApoE-PDHD was investigated. GGT (1, 5, or 10 U/mL) was added to the solution of BEAGA/ApoE-PDHD (2 mg) in 1 mL of PBS solution. The solution was dialyzed (MW cut-off: 3500 Da) against 50 mL of PBS (pH = 7.4) with gentle shaking (75 rpm) at 37 °C in an incubator shaker (Shanghai Yiheng Scientific DKZ, China). At certain time intervals (0.5 h, 1 h, 3 h, 6 h, 9 h, and 12 h), a 20 μL aliquot was taken out from the dialysis bag and measured by the 90 Plus/BI-MAS instrument.

The in vitro ultrasound imaging was executed utilizing a clinical ultrasound scanning system (Acuson Sequoia 512, Siemens, USA) operating under the Power Doppler mode. For this purpose, 1 mL aliquots of the nanoprobe solution (5 mg/mL), were dispensed into designated sample wells within a custom-fabricated 2% (w/v) agarose phantom, with the pH adjusted to either 7.4 or 6.5 to assess the impact on imaging capabilities. The Power Doppler imaging was achieved by employing a broadband 15L8-w high-frequency linear transducer, operating at a frequency of 10 MHz and transmitting a power level of 18 dB48⁴⁸. The focal zone was precisely positioned at the center of each well to ensure optimal imaging resolution. Horizontal scans of the nanoprobe distribution were then acquired using the agarose gel model as an imaging medium.

Afterwards, the content of the sonosensitizer ILD encapsulated in nanoprobe was measured using a Unico UV-2000 UV-vis spectrophotometer (Shanghai, China) at the wavelength of 755 nm based on the standard absorption curve of ILD. The gadolinium (Gd) content was determined by an Optima 5300 DV ICP-AES (Perkin Elmer® Inc., USA) at the wavelength of 342 nm. The PFP/PFB content was detected using a gas chromatography-mass spectrometry analysis system (GC-MS, Trace GC 2000-DSP, USA) based on our previous study²⁶. The loading content of R848 in nanoprobe was investigated at the wavelength of 254 nm by high-performance liquid chromatography (HPLC) equipped with a 1260 Infinity II LC System (Agilent Technologies Inc., Santa Clara, CA, USA). Briefly, the pre-weighed freeze-dried nanoprobe was redissolved in 0.1 M HCl solution. R848 was extracted after the addition of methanol. The organic phase was separated by centrifugation at 3100 × g for 15 min and analyzed with a C18 column (50 mm × 2.1 mm, 1.7 μm) at 45 °C. The mixture of acetonitrile and 0.1% formic acid solution (95: 5, v/v) was chosen as mobile phase at a flow rate of 0.3 mL/min.

Furthermore, the in vitro drug release behavior of nanoprobe was analyzed at 37 °C. Two mL of solution containing 20 mg of nanoprobe was adjusted to pH 7.4 or 6.5 and introduced into a dialysis bag (MW cut-off: 14 kDa). The dialysis bag was placed into 20 mL of fresh PBS (pH 7.4 or 6.5) with gentle shaking (75 rpm) at 37 °C in an incubator shaker. Insonation was applied if needed (2 MHz, 2.0 W/cm², 20% duty cycle, 2 min). At specific time intervals (0.5 h, 1 h, 2 h, 4 h, 6 h, 8 h, 12 h, and 24 h), 3 mL aliquots of the solution external to the dialysis bag were withdrawn for UV-vis spectroscopic analysis, promptly followed by replenishment with an equal volume of fresh buffer solution. Subsequently, the cumulative percentages of the released drugs were graphically represented against time.

Finally, the MRI sensitivity of the nanoprobe was evaluated by measuring the T1 relaxation time on a clinical MR system (Ingenia 3.0 T; Philips Medical Systems, Best, Netherland). Nanoprobe dissolved in PBS at pre-designed concentrations was added to a 96-well detachable plate. An inversion recovery spin echo sequence was carried out with the following parameters: TR = 1500 ms; TE = 20 ms; FOV, 80 mm × 80 mm; matrix, 228 mm × 289 mm; voxel, 0.4 mm × 0.4 mm; slice thickness, 2 mm; reconstruction matrix, 512; IR delay, 400 ms; NSA, 1. Regions of interest (ROIs, mean size, 30 mm²) were checked to obtain the T1 relaxation times. Furthermore, the T1 relaxivity value (r₁) was derived by computing the slope of the linear plots constructed by plotting the reciprocal of T1 (1/T1) against the Gd concentration, employing a linear least-squares regression analysis method.

Cell culture

The mice glioma cells (G422), rat glioma cells (C6, Catalog No. CCL-107 (ATCC)), human glioma cells (U87, Catalog No. HTB-14 (ATCC)), and luciferase protein gene recombinant mice glioma cells (G422-Luc) were purchased from Tongpai Biotechnology Co., Ltd (Shanghai, China). Mouse brain microvascular endothelial cells (bEnd3, Catalog No. CRL-2299 (ATCC)), human brain microvascular endothelial cells (HBMEC, Catalog No. #1000 (ScienCell)), rat brain microvascular endothelial cells (RBMEC, Catalog No. #R1000 (ScienCell)), and mouse leukemia cells of monocyte macrophage (RAW 264.7, Catalog No. TIB-71 (ATCC)) were obtained from Zhong Qiao Xin Zhou Biotechnology Co., Ltd (Shanghai, China). The GGT-shRNA-Lentivirus (sequence (5’ to 3’): GGTTGGCCAATACCACCATGT) used to construct stable GGT knockdown bEnd3 cells (bEnd3-GGT/KD) were designed by Suzhou Jima Gene Co., Ltd. (Suzhou, China). G422 and G422-Luc cells were cultured in RPMI 1640 medium. C6, bEnd3, U87, and RAW 264.7 cells were cultured in a DMEM (low glucose) medium. HBMEC cells were cultured in an endothelial cell medium and RBMEC were cultured in an endothelial cell medium-rat. All cell lines were tested negative for mycoplasma contamination by using MycAwayTM-Color One-Step Mycoplasma Detection Kit. The cells were cultured at 37 °C in various mediums supplemented with 10% FBS and 1% penicillin-streptomycin in a humidified incubator of 5% CO₂. And the cells for all experiments were in logarithmic growth phase.

Western blot assay

BEnd3, bEnd3-GGT/KD, HBMEC, RBMEC, G422, and RAW 264.7 cells were seeded in a 6-well plate and cultured overnight. The cells were separated from culture medium and washed with PBS three times when the cell density reached 80%. The protein samples were prepared by adding loading buffer and denatured at 100 °C for 10 min. Then the sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gel was used to load protein samples and ran at 90 V for 2 h. Subsequently, the proteins were efficiently transferred from the gel onto a PVDF membrane utilizing a dedicated transfer buffer. To minimize non-specific binding, the membrane was then blocked with 5% bovine serum albumin (BSA) for a duration of 1 h at ambient temperature. Following blocking, the membrane was incubated overnight at 4 °C with gentle agitation in the presence of the primary anti-GGT antibody. Post-incubation, the membrane was washed three times with TBST buffer to remove unbound antibodies. Next, the secondary antibody, which was conjugated to HRP, was diluted in the blocking buffer and incubated with the membrane for 1 h at room temperature. The membrane was again subjected to three washes with TBST buffer. The proteins were visualized using the appropriate enzyme substrate solution according to the manufacturer’s instructions and imaged by a chemiluminescence detection system.

In vitro BBB model and cellular internalization studies

An in vitro BBB model was constructed to verify the efficiency of various nanoprobes crossing BBB. bEnd3 cells were seeded into polycarbonate 24-well minicell® Hanging Cell Culture Insert of 0.4 μm pore size (Millipore, USA) at a density of 5 × 10⁴ cells per well. The transepithelial electric resistance (TEER) was monitored every day using an epithelial voltohmmeter (Millicell-RES, Millipore, USA). The tight junctions of the in vitro model could be regarded to mimic the real BBB when the TEER values reached a stable level²⁸. The minicells with bEnd3 cells were hung upon the wells of a 24-well plate while G422 cells were cultured in the plate. The nanoprobes, i.e., BEAGA/ApoE-PDHD@ILD & R848, ApoE-PDHD@ILD & R848, and BEAGA-PDHD@ILD & R848, were added into the medium of minicells and incubated at 37 °C. Insonation was applied if needed (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min). After incubation for another 2 h, the nanoprobes internalized in G422 cells after crossing BBB constructed by bEnd3 were observed under CLSM and analyzed using flow cytometry assay.

To evaluate the BBB-crossing efficiency of various nanoprobes on other cellular models, the transwell filter with a mean pore size of 0.4 µm was seeded with a compact HBMEC monolayer or a compact RBMEC monolayer, and the bottom plate was correspondingly seeded with a U87 cell monolayer or a C6 cell monolayer. The in vitro BBB models were treated likewise. The nanoprobes internalized in U87 cells or C6 cells after crossing BBB constructed by HBMEC or RBMEC were observed under CLSM and analyzed using flow cytometry assay.

Penetration of nanoprobes in 3D cell spheroids

The 96-well plate was pre-coated with agarose. bEnd3-GGT/KD cells with stable GGT knockdown were established using GGT-shRNA-Lentivirus according to our previous study⁴⁹. In detail, bEnd3 cells were seeded in a 6-well plate and incubated overnight to reach 60-70% confluency. The GGT-shRNA-Lentivirus was diluted in DMEM with 10% FBS and supplemented with polybrene at a final concentration of 5 µg/mL to enhance transduction efficiency. The virus-containing medium was added to the cells and incubated for 24 h. Afterwards, the medium was replaced with fresh DMEM, and the cells were incubated for an additional 48 h. To select for successfully transduced cells, puromycin was added to the medium at a concentration of 1 µg/mL, and the selection was continued for 14 days until non-transduced cells were eliminated. The knockdown of the GGT gene was then verified by western blot to evaluate protein expression.

Next, L929 cells and bEnd3-GGT/KD cells were seeded in the 96-well plate at a ratio of 3:1 for 7 days to prepare the cell spheroids. The culture medium was changed every 3 days. The 3D cell spheroids were incubated with BEAGA/ApoE-PDHD@ILD & R848 pre-treated with various concentrations of GGT when the spheroids grew up to 500 μm. LFUS irradiation was carried out if required. After washing with PBS, the received cell spheroids were imaged under the CLSM.

Intracellular ROS generation

ROS generation of nanoprobes under insonation was determined by DCFH-DA. G422 cells were seeded at a density of 1.0 × 10⁶ cells per well into a 12-well plate and cultured at 37 °C overnight. After incubated with saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 at an ICG concentration of 50 μg/mL for 2 h, insonation was applied to the BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 treatment groups (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min). 1 h later, 1 mL of DCFH-DA was added and incubated with G422 cells for 30 min. After washing with fresh PBS three times, the intracellular ROS was observed under CLSM and quantitatively investigated by flow cytometry analysis.

Cell viability and apoptosis assay

The cell viabilities against nanoprobes were measured using MTT assay. G422 cells were seeded onto a 96-well plate at a uniform density of 5 × 10³ cells per well. The cells incubated overnight were treated with saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@ILD & R848 at an ICG concentration of 50 μg/mL. After further incubation for 12 h, insonation was applied if needed (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min). 12 h later, the cell viabilities were determined by Cell Counting Kit-8 (CCK8). Meanwhile, the cells were stained with Calcein-AM/Propidium Iodide (PI) to detect live and dead cells under CLSM.

For cell apoptosis assay, G422 cells were seeded in a 12-well plate at the density of 1 × 10⁶ cells per well and incubated overnight. Various formulations including saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@ILD & R848 (ICG dose: 50 μg/mL) were added and incubated for another 12 h. Insonation was applied to the BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 treatment groups (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min). 12 h later, cells were stained with Annexin V-FITC/PI dyes and further quantitatively measured by flow cytometry (FCM, NovoCyte 3000, Agilent).

Measurement of mitochondrial potential

Mitochondrial membrane potential was monitored by the fluorescent probe JC-1. G422 cells were treated in the same way as the apoptosis assay and collected. The JC-1 working solution was added and incubated for 10 min without light⁵⁰. After centrifugation (350 × g) for 5 min at room temperature, the supernatant was discarded. The cells were washed twice with 2 mL of medium and resuspended in 400 μL of medium for flow cytometry analysis.

For CLSM analysis, G422 cells were stained with JC-1 working solution for 15 min at 37 °C. Afterwards, PBS rinsing was carried out to completely remove the needless JC-1 dye. The cells were stained with DAPI for CLSM observation (DAPI: Ex/Em = 405/455 nm; monomer: Ex/Em = 514/529 nm; J-aggregate: Ex/Em = 585/590 nm).

Expression level of calreticulin (CALR), adenosine triphosphate (ATP), and high mobility group B1 (HMGB1) in vitro

The expression of CALR was analyzed through immunocytochemistry and subsequently visualized under CLSM. Prior to the analysis, G422 cells were seeded at a density of 1 × 10⁶ cells/well in a 12-well plate and allowed to adhere and proliferate overnight under optimal culture conditions. The cells were treated with saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD, and BEAGA/ApoE-PDHD@ILD & R848 for 12 h. Insonation was applied to the BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 treatment groups (2 MHz, 2.0 W/cm², 20% duty cycle, 5 min). Afterwards, the cells were fixed with 4% paraformaldehyde to preserve their structural integrity, followed by a blocking step to minimize non-specific binding. The cells were incubated with the anti-calreticulin antibody at 4 °C overnight. The FITC secondary antibody was then used to stain cells for 1 h. Afterwards, the actin was stained with the Actin-Tracker at a 1/100 dilution for 1 h and the nuclei were labeled with DAPI. The cells were finally observed using CLSM.

On the other hand, cells were treated in the same way to determine the expression level of ATP and HMGB1 through enzyme-linked immunosorbent assay (ELISA). Cell supernatants were collected by centrifugation at 12,400 × g for 20 min and analyzed with an ATP ELISA Kit or HMGB1 ELISA Kit following the manufacturer’s protocol.

Bone marrow-derived dendritic cells (BMDCs) maturation

First of all, a Liquid Chromatograph Mass Spectrometer (LCMS-2020, SHIMADZU, Japan) was carried out to verify the intactness and successful release of R848 after G422 ICD. The supernatant of G422 cells after treatment of BEAGA/ApoE-PDHD plus LFUS irradiation mentioned above was collected and dried up. The residue was dissolved in methanol, filtrated, and detected using the LC-MS. For HPLC measurement, a mixture of acetonitrile and 0.1% formic acid solution (95: 5, v/v) was used as an eluent at a flow rate of 0.3 mL/min. The content of R848 was investigated at the wavelength of 254 nm. A standard solution of R848 at a concentration of 10 μg/mL was employed as the control group. The experiment was performed using a C18 column (Thermo Hypersil GOLD) at the temperature of 45 °C. The mass spectrum of R848 in G422 cell supernatant after HPLC separation was obtained based on the following parameters: In Source Type, H-ESI; Spray Voltage, static; Positive Ion (V), 3200; Sheath Gas (Arb), 40; Aux Gas (Arb), 5; Ion Transfer Tube Temp (°C), 300; Vaporizer Temp (°C), 350. Experiments were repeated three times. Data were analyzed using Xcalibur (Ver.4.3) software.

Afterwards, the 6-week-old KM mice were euthanized and sterilized with 75% ethanol. The femurs and tibias of the mice were removed and the muscle was dissected from the bone. The bone marrow was flushed with the cold RPMI 1640 medium and filtrated through a 70-μm cell strainer. The bone marrow cells were collected by centrifuging at 350 × g for 5 min after the lysis of red blood cells. Then, the bone marrow cells were resuspended and cultured at a density of 1 × 10⁶ cells per well with the completed RPMI 1640 medium in the presence of 20 ng/mL GM-CSF. The cell culture media was half-changed every 3 days. The BMDCs were collected on the 6th day from the well by gently pipetting up and centrifuging at 300 × g for 5 min. The harvested BMDCs were seeded in a new plate and incubated with the supernatant of glioma cells receiving treatments of saline, BEAGA/ApoE-PDHD@R848, BEAGA/ApoE-PDHD@ILD and BEAGA/ApoE-PDHD@ILD & R848 under LFUS irradiation. BMDCs that have not been treated with glioma cell supernatant were served as a control group for comparison (no stimulus group).

The maturation of BMDCs was analyzed using FCM and ELISA. For FCM analysis, the BMDCs were collected and incubated with anti-CD16/32 antibody for 30 min to block Fc receptors. And then the cells were stained with fluorescently labeled antibodies such as CD11c, CD80, and CD86. After rinsed and resuspended, the cells were analyzed by FCM to calculate the percentage of maturation markers. Meanwhile, the cell supernatants were collected to measure the levels of cytokines including IL-12 and IFN-β using the ELISA Kits according to the manufacturer’s protocol.

Animal models

Four-week-old female KM mice were purchased from Guangdong Medical Laboratory Animal Center. G422-Luc cells were implanted into the brain striatum of mice to establish the orthotopic glioma model. In brief, mice were first anesthetized by injection of 2.5% pentobarbital sodium (3.5 mL/kg) intraperitoneally and restrained. 5.0 × 10⁵ G422-Luc cells resuspended in 5 μL of PBS were injected into the right striatum of mouse brain (2.0 mm lateral and 3.5 mm depth) at a speed of 0.5 μL/min using a mouse adapter. Mice with their wound sterilized and seamed were carefully housed in a specific pathogen-free (SPF) environment. To establish orthotopic U87 tumor-bearing nude mouse or C6 tumor-bearing rat model, U87 cells or C6 cells were respectively implanted into the brain striatum of 4-week-old female nude mice or 8-week-old female Sprague Dawley rats purchased from Guangdong Medical Laboratory Animal Center and treated likewise. The Ethics Committee stipulated a criterion of a maximum permissible tumor diameter of 2 cm, conjoined with a condition that the tumor’s weight ought not to surpass 10% of the total body weight of the mice. None of the tumor volumes of mice in the current study transgressed these regulations. During the experiment, all animals were subjected to regulated environmental conditions, including a precisely maintained temperature of 22 °C, humidity levels ranging between 40% and 50%, and a strictly adhered to light/dark cycle of 12 h each, with unrestricted access to both water and food.

In vivo biodistribution of nanoprobes

In vivo fluorescence imaging

The in vivo biodistribution of nanoprobes in tumor-bearing KM mice was first investigated through a small animal fluorescence imaging system (In-Vivo Imaging System FX Pro, Carestream Health Inc., New Haven, CT, USA). KM mice (n = 3) with G422 cells inoculated in their brain for 6 days were tail vein injected with BEAGA/ApoE-PDHD@ILD & R848, ApoE-PDHD@ILD & R848 or BEAGA-PDHD@ILD & R848 (500 μg/kg ICG), respectively. The hair of mice was removed. At pre-designed time points (pre, 1 h, 3 h, 6 h, 12 h, and 24 h), fluorescence images were recorded with an excitation wavelength of 720 nm and an emission wavelength of 790 nm. Subsequently, the mice were humanely euthanized, and their vital organs including the heart, liver, spleen, lungs, and kidneys, along with the brain, were excised for ex vivo fluorescence imaging. The brain tissues were further subjected to the preparation of frozen sections to investigate the distribution of nanoprobes in the brain. The in vivo biodistribution of nanoprobes in U87 tumor-bearing nude mice or C6 tumor-bearing rats was investigated through the same experimental procedure, followed by their brain tissues being subjected to prepare frozen sections.

In vivo MR imaging

MRI scan was next carried out to evaluate the in vivo imaging ability of nanoprobes on mice 6 days after tumor cell inoculation. Mice (n = 3) anesthetized by intraperitoneal injection of 2.5% pentobarbital sodium (3.5 mL/kg) were scanned on a 3.0 T MR scanner equipped with a 50 mm × 65 mm 8-channel phased-array mouse coil (Shanghai Chenguang Medical Technologies Co., Ltd., Shanghai, China). Subsequently, they were i.v. injected with BEAGA/ApoE-PDHD@ILD & R848, ApoE-PDHD@ILD & R848, or BEAGA-PDHD@ILD & R848 at a dose of 4.5 mg Gd/kg body weight. At pre-designed time points (pre, 1 h, 3 h, 6 h, 12 h and 24 h), a fast spin echo sequence was performed to acquire T1-weighted MR images: TR = 400 ms; TE = 11 ms; FOV, 60 mm × 60 mm; matrix, 248 mm × 246 mm; voxel, 0.24 mm × 0.24 mm; slice thickness, 1 mm; flip angle, 90°; NSA, 2.

Intravital real-time observation by two-photon CLSM

The real-time two-photon CLSM (Olympus FVMPE-RS, Japan) was applied to directly observe the processes of nanoprobes binding to the cerebral vessels and then crossing BBB on tumor-bearing mice. DiI was encapsulated into nanoprobes to make them visible under CLSM. Mice (n = 3) were anesthetized by intraperitoneal injection of 2.5% pentobarbital sodium (3.5 mL/kg). Their heads were restrained and polished using a hand-held cranial drill (RWD Life Science, China) under a dissection microscope to acquire a small skull region (~4 mm in diameter)⁵¹. The mice were then injected intravenously with DiI-loaded nanoprobes (500 μg/kg DiI), i.e., BEAGA/ApoE-PDHD@DiI, ApoE-PDHD@DiI and BEAGA-PDHD@DiI, and dripped with a drop of water on the thinned-skull cranial window. Next, the height of the objective lens was exactly adjusted to touch the water droplet on mice placed on a thermoplate of CLSM. A pre-programmed program was carried out at pre-designed time points (30 min, 60 min, 90 min, and 120 min). Images were acquired from Olympus FV1000 software and analyzed by Image J.

Anticancer efficacy of nanoprobes

In vivo anticancer effect

The in situ G422-Luc tumor-bearing mice were randomly assigned to 6 groups (n = 5) after tumor cells transplanted for 6 days, and received treatments of saline (control), R848 and ILD co-loaded BEAGA-PDHD (BEAGA-PDHD@ILD & R848), R848 and ILD co-loaded ApoE-PDHD (ApoE-PDHD@ILD & R848), R848 loaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@R848), ILD loaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@ILD), R848 and ILD co-loaded BEAGA/ApoE-PDHD (BEAGA/ApoE-PDHD@ILD & R848) at an ICG dose of 5 mg/kg body weight, respectively. Different formulations were intravenously administered into mice every 3 days for a total of 4 times. Mice in all treatment groups were exposed to ultrasonic insonation (2.0 MHz, 2.0 W/cm², 20% duty cycle, 10 min) at 12 h post-administration of nanoprobes. Mice were raised for overall survival observation with their body weights recorded and growth rates of intracranial glioma cells monitored by bioluminescence imaging.

In another separate experiment, mice subjected to the same treatment strategy were sacrificed 17 days after tumor cell inoculation, followed by their brains being excised, fixed with 4% paraformaldehyde, sliced coronally into sections, and finally subjected to hematoxylin/eosin (H&E) staining and immunohistochemistry (IHC) staining of Ki67.

Furthermore, 4-week-old tumor-free KM mice were randomly assigned to 6 groups (n = 3) and received the same treatment strategy as that of the tumor-bearing mice. Then, mice were perfused with ice-cold PBS and 4% paraformaldehyde (PFA) to prepare paraffin sections of brains. The sections were subjected to immunofluorescence staining using rabbit anti-Iba1 or rabbit anti-GFAP or stained with methylene blue according to standard protocol.

Evaluation of tumor-infiltrating immune cells and memory CD8⁺ T cells

The brain tumors were excised from the mice receiving different treatments and dissociated into single-cell suspensions through the mechanical method. The cells were filtrated through a 70-μm cell strainer and washed with cold PBS buffer to remove cell clumps and debris. After blocking Fc receptors, the cells were incubated with surface marker antibodies including CD45, CD3, CD4, and CD8 in the dark for 30 min at 4 °C. Then the cells were washed with the FACS staining buffer to remove the unbound antibodies, and subjected to intracellular staining of nuclear Foxp3 protein according to the manufacturer’s protocol. After fixation and permeabilization, the cells were incubated with the anti-Foxp3 antibody for 1 h at 4 °C. The unbound antibody was removed by centrifuging at 450 × g for 5 min. In addition, the DCs were stained with surface markers of CD45, CD11c, CD80 and CD86. To identify other myeloid populations such as neutrophils and tumor-associated macrophages (TAMs), the single-cell suspensions were stained with the antibody cocktails including CD45, CD11b, Ly-6G, F4/80, CD80 and CD206. The immune cell populations were identified using FCM. For memory CD8⁺ T cell detection, the spleens of mice (n = 3) were harvested and dissociated into single-cell suspensions. The splenocytes were stained with surface markers of CD3, CD8, CD44 and CD62L.

Clinical human specimens

Human hepatoma tissue sections and human glioma tissue microarray sections were obtained from Shanghai Outdo Biotech Co., Ltd. (Shanghai, China). The human tissues were incubated with anti-GGT antibody at a 1:1000 dilution to detect the expression of GGT or incubated with anti-LDLR antibody at a 1:100 dilution to detect the expression of low-density lipoprotein receptor (LDLR) by IHC staining.

Statistics and reproducibility

The experiments including TEM observation, western blot analysis, immunohistochemical assessments of GGT and CD8⁺ T cells expression, two-photon CLSM examination, dual staining with Calcein-AM and PI for viability analysis, histological analysis of tumor tissues, and CLSM-based visualization of nanoprobe internalization by tumor cells and 3D cell spheroids, intracellular ROS generation, nanoprobe distribution in brain tissue sections, alterations in mitochondrial membrane potential, and CALR exposure in tumor cells were repeated three times. All remaining experiments were repeated at least threefold to ensure reproducibility and statistical significance. The investigators were not blinded to allocation during experiments and outcome assessment. Results were expressed as the means ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA, USA). Comparison between groups was calculated by the two-sided unpaired t-test. A log-rank (Mantel–Cox) test was conducted to compare the survival curves. For all graphs, *p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(6.2MB, pdf)}

Peer Review File^{(6.4MB, pdf)}

41467_2024_54382_MOESM3_ESM.pdf^{(79.6KB, pdf)}

Description of Additional Supplementary Files

Supplementary Movie 1^{(19.2MB, mp4)}

Reporting Summary^{(4.4MB, pdf)}

Source data

Source Data^{(11.3MB, xlsx)}

Acknowledgements

This research was supported by the National Natural Science Foundation of China (82302359, B.L.; 51933011, X.S.; 52373164, J.H.), Natural Science Foundation of the Guangdong Province (2023A1515011822, B.L.; 2021A1515111006, B.L.; 2024A1515030079, J.H.), Key Areas Research and Development Program of Guangzhou (202007020006, X.S.), China Postdoctoral Science Foundation (2021M703763, B.L.; 2024M753755, G.C.) and the Shenzhen Science and Technology Program (JCYJ20220530144816037, J.H.). The authors also thank TosanBio company for their help in redrawing the schematic diagram.

Author contributions

B.L., G.C. and X.S. conceived and designed the project. B.L., H.Z. and T.L. synthesized and characterized the materials. B.L., G.C., M.L., H.W., Q.Z. and Q.C. performed all the cell and animal experiments. B.L., J.H. and X.S. discussed the results, analyzed the data, and wrote the paper with the help of all authors. J.H. and X.S. supervised the research. All authors read and approved the final version of the manuscript.

Peer review

Peer review information

Nature Communications thanks Zhaohui Wang, Raziye Piranlioglu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The figures of NMR spectra are summarized in Supplementary Figs. 2–9, 11–14 in the Supplementary Materials and the Source Data file. The mass spectrum is summarized in Supplementary Fig. 30 in the Supplementary Materials and the Source Data file. All other data supporting the findings of this study are available within the article, Supplementary Information or Source data files. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Bo Li, Gengjia Chen.

Contributor Information

Bo Li, Email: libo65@mail2.sysu.edu.cn.

Jinsheng Huang, Email: huangjsh55@mail.sysu.edu.cn.

Xintao Shuai, Email: shuaixt@mail.sysu.edu.cn.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-024-54382-z.

References

1.Sung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.71, 209–249 (2021). [DOI] [PubMed] [Google Scholar]
2.Sun, S. et al. Neuropilin-1 is a glial cell line-derived neurotrophic factor receptor in glioblastoma. Oncotarget8, 74019 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Huang, P. et al. Metalloporphyrin-encapsulated biodegradable nanosystems for highly efficient magnetic resonance imaging-guided sonodynamic cancer therapy. J. Am. Chem. Soc.139, 1275–1284 (2017). [DOI] [PubMed] [Google Scholar]
4.June, C. H., O’Connor, R. S., Kawalekar, O. U., Ghassemi, S. & Milone, M. C. CAR T cell immunotherapy for human cancer. Science359, 1361–1365 (2018). [DOI] [PubMed] [Google Scholar]
5.Min, H. S. et al. Systemic brain delivery of antisense oligonucleotides across the blood–brain barrier with a glucose‐coated polymeric nanocarrier. Angew. Chem. Int. Ed.59, 8173–8180 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Hervé, F., Ghinea, N. & Scherrmann, J.-M. CNS delivery via adsorptive transcytosis. AAPS J.10, 455–472 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Terstappen, G. C., Meyer, A. H., Bell, R. D. & Zhang, W. Strategies for delivering therapeutics across the blood–brain barrier. Nat. Rev. Drug Discov.20, 362–383 (2021). [DOI] [PubMed] [Google Scholar]
8.Qiao, R. et al. Receptor-mediated delivery of magnetic nanoparticles across the blood–brain barrier. ACS Nano6, 3304–3310 (2012). [DOI] [PubMed] [Google Scholar]
9.Bickel, U. Antibody delivery through the blood-brain barrier. Adv. Drug Deliv. Rev.15, 53–72 (1995). [PubMed] [Google Scholar]
10.Bickel, U., Yoshikawa, T. & Pardridge, W. M. Delivery of peptides and proteins through the blood–brain barrier. Adv. Drug Deliv. Rev.46, 247–279 (2001). [DOI] [PubMed] [Google Scholar]
11.Etame, A. B., Smith, C. A., Chan, W. C. & Rutka, J. T. Design and potential application of PEGylated gold nanoparticles with size-dependent permeation through brain microvasculature. Nanomed. Nanotechnol. Biol. Med.7, 992–1000 (2011). [DOI] [PubMed] [Google Scholar]
12.Wiley, D. T., Webster, P., Gale, A. & Davis, M. E. Transcytosis and brain uptake of transferrin-containing nanoparticles by tuning avidity to transferrin receptor. Proc. Natl Acad. Sci. USA110, 8662–8667 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Cox, D. S., Raje, S., Gao, H., Salama, N. N. & Eddington, N. D. Enhanced permeability of molecular weight markers and poorly bioavailable compounds across Caco-2 cell monolayers using the absorption enhancer, zonula occludens toxin. Pharm. Res.19, 1680–1688 (2002). [DOI] [PubMed] [Google Scholar]
14.Vorbrodt, A. Ultracytochemical characterization of anionic sites in the wall of brain capillaries. J. Neurocytol.18, 359–368 (1989). [DOI] [PubMed] [Google Scholar]
15.Kumagai, A., Eisenberg, J. B. & Pardridge, W. Absorptive-mediated endocytosis of cationized albumin and a beta-endorphin-cationized albumin chimeric peptide by isolated brain capillaries. Model system of blood-brain barrier transport. J. Biol. Chem.262, 15214–15219 (1987). [PubMed] [Google Scholar]
16.Nell, L. J. & Thomas, J. W. Frequency and specificity of protamine antibodies in diabetic and control subjects. Diabetes37, 172–176 (1988). [DOI] [PubMed] [Google Scholar]
17.Lim, M., Xia, Y., Bettegowda, C. & Weller, M. Current state of immunotherapy for glioblastoma. Nat. Rev. Clin. Oncol.15, 422–442 (2018). [DOI] [PubMed] [Google Scholar]
18.Wu, W. et al. Glioblastoma multiforme (GBM): an overview of current therapies and mechanisms of resistance. Pharmacol. Res.171, 105780 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Chang, S. M. et al. Temozolomide in the treatment of recurrent malignant glioma. Cancer100, 605–611 (2004). [DOI] [PubMed] [Google Scholar]
20.Cheng, Y. et al. An intelligent biomimetic nanoplatform for holistic treatment of metastatic triple-negative breast cancer via photothermal ablation and immune remodeling. ACS Nano14, 15161–15181 (2020). [DOI] [PubMed] [Google Scholar]
21.Chen, X. et al. Microglia-mediated T cell infiltration drives neurodegeneration in tauopathy. Nature615, 668–677 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Zhu, Y. et al. Neuron-specific SALM5 limits inflammation in the CNS via its interaction with HVEM. Sci. Adv.2, e1500637 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Li, B. et al. Anchoring microbubbles on cerebrovascular endothelium as a new strategy enabling low‐energy ultrasound‐assisted delivery of varisized agents across blood‐brain barrier. Adv. Sci.10, 2302134 (2023). [DOI] [PMC free article] [PubMed]
24.Zhou, Q. et al. Enzyme-activatable polymer–drug conjugate augments tumour penetration and treatment efficacy. Nat. Nanotechnol.14, 799–809 (2019). [DOI] [PubMed] [Google Scholar]
25.Hanigan, M. H. & Frierson, H. F. Jr Immunohistochemical detection of gamma-glutamyl transpeptidase in normal human tissue. J. Histochem. Cytochem.44, 1101–1108 (1996). [DOI] [PubMed] [Google Scholar]
26.Zhang, L. et al. Size-modulable nanoprobe for high-performance ultrasound imaging and drug delivery against cancer. ACS Nano12, 3449–3460 (2018). [DOI] [PubMed] [Google Scholar]
27.Li, G. et al. Fluorinated chitosan to enhance transmucosal delivery of sonosensitizer-conjugated catalase for sonodynamic bladder cancer treatment post-intravesical instillation. ACS Nano14, 1586–1599 (2020). [DOI] [PubMed] [Google Scholar]
28.Li, B. et al. Molecular probe crossing blood–brain barrier for bimodal imaging–guided photothermal/photodynamic therapies of intracranial glioblastoma. Adv. Funct. Mater.30, 1909117 (2020). [Google Scholar]
29.Chen, P.-M. et al. Modulation of tumor microenvironment using a TLR-7/8 agonist-loaded nanoparticle system that exerts low-temperature hyperthermia and immunotherapy for in situ cancer vaccination. Biomaterials230, 119629 (2020). [DOI] [PubMed] [Google Scholar]
30.Jiang, Y., Zhang, J., Meng, F. & Zhong, Z. Apolipoprotein E peptide-directed chimeric polymersomes mediate an ultrahigh-efficiency targeted protein therapy for glioblastoma. ACS Nano12, 11070–11079 (2018). [DOI] [PubMed] [Google Scholar]
31.Jewett, J. C. & Bertozzi, C. R. Cu-free click cycloaddition reactions in chemical biology. Chem. Soc. Rev.39, 1272–1279 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Liu, H.-L., Fan, C.-H., Ting, C.-Y. & Yeh, C.-K. Combining microbubbles and ultrasound for drug delivery to brain tumors: current progress and overview. Theranostics4, 432 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Pardridge, W. M. Blood–brain barrier delivery. Drug Discov. Today12, 54–61 (2007). [DOI] [PubMed] [Google Scholar]
34.Liang, S. et al. Harnessing nanomaterials for cancer sonodynamic immunotherapy. Adv. Mater.35, 2211130 (2023). [DOI] [PubMed]
35.Liang, S. et al. A robust narrow bandgap vanadium tetrasulfide sonosensitizer optimized by charge separation engineering for enhanced sonodynamic cancer therapy. Adv. Mater.33, 2101467 (2021). [DOI] [PubMed] [Google Scholar]
36.Yanagisawa, S. et al. Oxidative stress augments toll-like receptor 8 mediated neutrophilic responses in healthy subjects. Respir. Res.10, 1–13 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Dedov, V., Cox, G. & Roufogalis, B. Visualisation of mitochondria in living neurons with single-and two-photon fluorescence laser microscopy. Micron32, 653–660 (2001). [DOI] [PubMed] [Google Scholar]
38.Yuan, H. et al. Multivalent bi-specific nanobioconjugate engager for targeted cancer immunotherapy. Nat. Nanotechnol.12, 763–769 (2017). [DOI] [PubMed] [Google Scholar]
39.Carpio, M. A., Lopez Sambrooks, C., Durand, E. S. & Hallak, M. E. The arginylation-dependent association of calreticulin with stress granules is regulated by calcium. Biochem. J.429, 63–72 (2010). [DOI] [PubMed] [Google Scholar]
40.Zheng, P. et al. Ultrasound-augmented mitochondrial calcium ion overload by calcium nanomodulator to induce immunogenic cell death. Nano Lett.21, 2088–2093 (2021). [DOI] [PubMed] [Google Scholar]
41.Lynn, G. M. et al. In vivo characterization of the physicochemical properties of polymer-linked TLR agonists that enhance vaccine immunogenicity. Nat. Biotechnol.33, 1201–1210 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Barton, G. M. & Medzhitov, R. Toll-like receptor signaling pathways. Science300, 1524–1525 (2003). [DOI] [PubMed] [Google Scholar]
43.Marabelle, A., Kohrt, H., Caux, C. & Levy, R. Intratumoral immunization: a new paradigm for cancer therapy. Clin. Cancer Res.20, 1747–1756 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Zhang, Y. et al. Stimulation of tumoricidal immunity via bacteriotherapy inhibits glioblastoma relapse. Nat. Commun.15, 4241 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Rufo, N., Garg, A. D. & Agostinis, P. The unfolded protein response in immunogenic cell death and cancer immunotherapy. Trends Cancer3, 643–658 (2017). [DOI] [PubMed] [Google Scholar]
46.Li, B. et al. MRI-visible and pH-sensitive micelles loaded with doxorubicin for hepatoma treatment. Biomater. Sci.7, 1529–1542 (2019). [DOI] [PubMed] [Google Scholar]
47.Wu, T., Liu, Y., Cao, Y. & Liu, Z. Engineering macrophage exosome disguised biodegradable nanoplatform for enhanced sonodynamic therapy of glioblastoma. Adv. Mater.34, 2110364 (2022). [DOI] [PubMed] [Google Scholar]
48.Yin, T. et al. Nanobubbles for enhanced ultrasound imaging of tumors. Int. J. Nanomed.7, 895–904 (2012). [DOI] [PMC free article] [PubMed]
49.Xiao, Z. et al. Nanodrug simultaneously regulates stromal extracellular matrix and glucose metabolism for effective immunotherapy against orthotopic pancreatic cancer. Nano Today44, 101490 (2022). [Google Scholar]
50.Chen, G. et al. Theranostic nanosystem mediating cascade catalytic reactions for effective immunotherapy of highly immunosuppressive and poorly penetrable pancreatic tumor. Sci. China Chem.65, 1383–1400 (2022). [Google Scholar]
51.Yang, G., Pan, F., Parkhurst, C. N., Grutzendler, J. & Gan, W.-B. Thinned-skull cranial window technique for long-term imaging of the cortex in live mice. Nat. Protoc.5, 201–208 (2010). [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 Information^{(6.2MB, pdf)}

Peer Review File^{(6.4MB, pdf)}

41467_2024_54382_MOESM3_ESM.pdf^{(79.6KB, pdf)}

Description of Additional Supplementary Files

Supplementary Movie 1^{(19.2MB, mp4)}

Reporting Summary^{(4.4MB, pdf)}

Source Data^{(11.3MB, xlsx)}

Data Availability Statement

[CR1] 1.Sung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.71, 209–249 (2021). [DOI] [PubMed] [Google Scholar]

[CR2] 2.Sun, S. et al. Neuropilin-1 is a glial cell line-derived neurotrophic factor receptor in glioblastoma. Oncotarget8, 74019 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Huang, P. et al. Metalloporphyrin-encapsulated biodegradable nanosystems for highly efficient magnetic resonance imaging-guided sonodynamic cancer therapy. J. Am. Chem. Soc.139, 1275–1284 (2017). [DOI] [PubMed] [Google Scholar]

[CR4] 4.June, C. H., O’Connor, R. S., Kawalekar, O. U., Ghassemi, S. & Milone, M. C. CAR T cell immunotherapy for human cancer. Science359, 1361–1365 (2018). [DOI] [PubMed] [Google Scholar]

[CR5] 5.Min, H. S. et al. Systemic brain delivery of antisense oligonucleotides across the blood–brain barrier with a glucose‐coated polymeric nanocarrier. Angew. Chem. Int. Ed.59, 8173–8180 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR6] 6.Hervé, F., Ghinea, N. & Scherrmann, J.-M. CNS delivery via adsorptive transcytosis. AAPS J.10, 455–472 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Terstappen, G. C., Meyer, A. H., Bell, R. D. & Zhang, W. Strategies for delivering therapeutics across the blood–brain barrier. Nat. Rev. Drug Discov.20, 362–383 (2021). [DOI] [PubMed] [Google Scholar]

[CR8] 8.Qiao, R. et al. Receptor-mediated delivery of magnetic nanoparticles across the blood–brain barrier. ACS Nano6, 3304–3310 (2012). [DOI] [PubMed] [Google Scholar]

[CR9] 9.Bickel, U. Antibody delivery through the blood-brain barrier. Adv. Drug Deliv. Rev.15, 53–72 (1995). [PubMed] [Google Scholar]

[CR10] 10.Bickel, U., Yoshikawa, T. & Pardridge, W. M. Delivery of peptides and proteins through the blood–brain barrier. Adv. Drug Deliv. Rev.46, 247–279 (2001). [DOI] [PubMed] [Google Scholar]

[CR11] 11.Etame, A. B., Smith, C. A., Chan, W. C. & Rutka, J. T. Design and potential application of PEGylated gold nanoparticles with size-dependent permeation through brain microvasculature. Nanomed. Nanotechnol. Biol. Med.7, 992–1000 (2011). [DOI] [PubMed] [Google Scholar]

[CR12] 12.Wiley, D. T., Webster, P., Gale, A. & Davis, M. E. Transcytosis and brain uptake of transferrin-containing nanoparticles by tuning avidity to transferrin receptor. Proc. Natl Acad. Sci. USA110, 8662–8667 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Cox, D. S., Raje, S., Gao, H., Salama, N. N. & Eddington, N. D. Enhanced permeability of molecular weight markers and poorly bioavailable compounds across Caco-2 cell monolayers using the absorption enhancer, zonula occludens toxin. Pharm. Res.19, 1680–1688 (2002). [DOI] [PubMed] [Google Scholar]

[CR14] 14.Vorbrodt, A. Ultracytochemical characterization of anionic sites in the wall of brain capillaries. J. Neurocytol.18, 359–368 (1989). [DOI] [PubMed] [Google Scholar]

[CR15] 15.Kumagai, A., Eisenberg, J. B. & Pardridge, W. Absorptive-mediated endocytosis of cationized albumin and a beta-endorphin-cationized albumin chimeric peptide by isolated brain capillaries. Model system of blood-brain barrier transport. J. Biol. Chem.262, 15214–15219 (1987). [PubMed] [Google Scholar]

[CR16] 16.Nell, L. J. & Thomas, J. W. Frequency and specificity of protamine antibodies in diabetic and control subjects. Diabetes37, 172–176 (1988). [DOI] [PubMed] [Google Scholar]

[CR17] 17.Lim, M., Xia, Y., Bettegowda, C. & Weller, M. Current state of immunotherapy for glioblastoma. Nat. Rev. Clin. Oncol.15, 422–442 (2018). [DOI] [PubMed] [Google Scholar]

[CR18] 18.Wu, W. et al. Glioblastoma multiforme (GBM): an overview of current therapies and mechanisms of resistance. Pharmacol. Res.171, 105780 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Chang, S. M. et al. Temozolomide in the treatment of recurrent malignant glioma. Cancer100, 605–611 (2004). [DOI] [PubMed] [Google Scholar]

[CR20] 20.Cheng, Y. et al. An intelligent biomimetic nanoplatform for holistic treatment of metastatic triple-negative breast cancer via photothermal ablation and immune remodeling. ACS Nano14, 15161–15181 (2020). [DOI] [PubMed] [Google Scholar]

[CR21] 21.Chen, X. et al. Microglia-mediated T cell infiltration drives neurodegeneration in tauopathy. Nature615, 668–677 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.Zhu, Y. et al. Neuron-specific SALM5 limits inflammation in the CNS via its interaction with HVEM. Sci. Adv.2, e1500637 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Li, B. et al. Anchoring microbubbles on cerebrovascular endothelium as a new strategy enabling low‐energy ultrasound‐assisted delivery of varisized agents across blood‐brain barrier. Adv. Sci.10, 2302134 (2023). [DOI] [PMC free article] [PubMed]

[CR24] 24.Zhou, Q. et al. Enzyme-activatable polymer–drug conjugate augments tumour penetration and treatment efficacy. Nat. Nanotechnol.14, 799–809 (2019). [DOI] [PubMed] [Google Scholar]

[CR25] 25.Hanigan, M. H. & Frierson, H. F. Jr Immunohistochemical detection of gamma-glutamyl transpeptidase in normal human tissue. J. Histochem. Cytochem.44, 1101–1108 (1996). [DOI] [PubMed] [Google Scholar]

[CR26] 26.Zhang, L. et al. Size-modulable nanoprobe for high-performance ultrasound imaging and drug delivery against cancer. ACS Nano12, 3449–3460 (2018). [DOI] [PubMed] [Google Scholar]

[CR27] 27.Li, G. et al. Fluorinated chitosan to enhance transmucosal delivery of sonosensitizer-conjugated catalase for sonodynamic bladder cancer treatment post-intravesical instillation. ACS Nano14, 1586–1599 (2020). [DOI] [PubMed] [Google Scholar]

[CR28] 28.Li, B. et al. Molecular probe crossing blood–brain barrier for bimodal imaging–guided photothermal/photodynamic therapies of intracranial glioblastoma. Adv. Funct. Mater.30, 1909117 (2020). [Google Scholar]

[CR29] 29.Chen, P.-M. et al. Modulation of tumor microenvironment using a TLR-7/8 agonist-loaded nanoparticle system that exerts low-temperature hyperthermia and immunotherapy for in situ cancer vaccination. Biomaterials230, 119629 (2020). [DOI] [PubMed] [Google Scholar]

[CR30] 30.Jiang, Y., Zhang, J., Meng, F. & Zhong, Z. Apolipoprotein E peptide-directed chimeric polymersomes mediate an ultrahigh-efficiency targeted protein therapy for glioblastoma. ACS Nano12, 11070–11079 (2018). [DOI] [PubMed] [Google Scholar]

[CR31] 31.Jewett, J. C. & Bertozzi, C. R. Cu-free click cycloaddition reactions in chemical biology. Chem. Soc. Rev.39, 1272–1279 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR32] 32.Liu, H.-L., Fan, C.-H., Ting, C.-Y. & Yeh, C.-K. Combining microbubbles and ultrasound for drug delivery to brain tumors: current progress and overview. Theranostics4, 432 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Pardridge, W. M. Blood–brain barrier delivery. Drug Discov. Today12, 54–61 (2007). [DOI] [PubMed] [Google Scholar]

[CR34] 34.Liang, S. et al. Harnessing nanomaterials for cancer sonodynamic immunotherapy. Adv. Mater.35, 2211130 (2023). [DOI] [PubMed]

[CR35] 35.Liang, S. et al. A robust narrow bandgap vanadium tetrasulfide sonosensitizer optimized by charge separation engineering for enhanced sonodynamic cancer therapy. Adv. Mater.33, 2101467 (2021). [DOI] [PubMed] [Google Scholar]

[CR36] 36.Yanagisawa, S. et al. Oxidative stress augments toll-like receptor 8 mediated neutrophilic responses in healthy subjects. Respir. Res.10, 1–13 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Dedov, V., Cox, G. & Roufogalis, B. Visualisation of mitochondria in living neurons with single-and two-photon fluorescence laser microscopy. Micron32, 653–660 (2001). [DOI] [PubMed] [Google Scholar]

[CR38] 38.Yuan, H. et al. Multivalent bi-specific nanobioconjugate engager for targeted cancer immunotherapy. Nat. Nanotechnol.12, 763–769 (2017). [DOI] [PubMed] [Google Scholar]

[CR39] 39.Carpio, M. A., Lopez Sambrooks, C., Durand, E. S. & Hallak, M. E. The arginylation-dependent association of calreticulin with stress granules is regulated by calcium. Biochem. J.429, 63–72 (2010). [DOI] [PubMed] [Google Scholar]

[CR40] 40.Zheng, P. et al. Ultrasound-augmented mitochondrial calcium ion overload by calcium nanomodulator to induce immunogenic cell death. Nano Lett.21, 2088–2093 (2021). [DOI] [PubMed] [Google Scholar]

[CR41] 41.Lynn, G. M. et al. In vivo characterization of the physicochemical properties of polymer-linked TLR agonists that enhance vaccine immunogenicity. Nat. Biotechnol.33, 1201–1210 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Barton, G. M. & Medzhitov, R. Toll-like receptor signaling pathways. Science300, 1524–1525 (2003). [DOI] [PubMed] [Google Scholar]

[CR43] 43.Marabelle, A., Kohrt, H., Caux, C. & Levy, R. Intratumoral immunization: a new paradigm for cancer therapy. Clin. Cancer Res.20, 1747–1756 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Zhang, Y. et al. Stimulation of tumoricidal immunity via bacteriotherapy inhibits glioblastoma relapse. Nat. Commun.15, 4241 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Rufo, N., Garg, A. D. & Agostinis, P. The unfolded protein response in immunogenic cell death and cancer immunotherapy. Trends Cancer3, 643–658 (2017). [DOI] [PubMed] [Google Scholar]

[CR46] 46.Li, B. et al. MRI-visible and pH-sensitive micelles loaded with doxorubicin for hepatoma treatment. Biomater. Sci.7, 1529–1542 (2019). [DOI] [PubMed] [Google Scholar]

[CR47] 47.Wu, T., Liu, Y., Cao, Y. & Liu, Z. Engineering macrophage exosome disguised biodegradable nanoplatform for enhanced sonodynamic therapy of glioblastoma. Adv. Mater.34, 2110364 (2022). [DOI] [PubMed] [Google Scholar]

[CR48] 48.Yin, T. et al. Nanobubbles for enhanced ultrasound imaging of tumors. Int. J. Nanomed.7, 895–904 (2012). [DOI] [PMC free article] [PubMed]

[CR49] 49.Xiao, Z. et al. Nanodrug simultaneously regulates stromal extracellular matrix and glucose metabolism for effective immunotherapy against orthotopic pancreatic cancer. Nano Today44, 101490 (2022). [Google Scholar]

[CR50] 50.Chen, G. et al. Theranostic nanosystem mediating cascade catalytic reactions for effective immunotherapy of highly immunosuppressive and poorly penetrable pancreatic tumor. Sci. China Chem.65, 1383–1400 (2022). [Google Scholar]

[CR51] 51.Yang, G., Pan, F., Parkhurst, C. N., Grutzendler, J. & Gan, W.-B. Thinned-skull cranial window technique for long-term imaging of the cortex in live mice. Nat. Protoc.5, 201–208 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

γ-Glutamyl transpeptidase-activable nanoprobe crosses the blood-brain barrier for immuno-sonodynamic therapy of glioma

Bo Li

Gengjia Chen

Huihai Zhong

Tan Li

Minzhao Lin

Huiye Wei

Qiaoyun Zhang

Qi Chen

Jinsheng Huang

Xintao Shuai

Abstract

Introduction

Fig. 1. Schematic illustration of the γ-Glutamyl transpeptidase (GGT) enzyme-activatable nanoprobe efficiently crossing the blood-brain barrier (BBB) for immuno-sonodynamic therapy of glioma.

Results

Polymer synthesis and nanoprobe preparation

Fig. 2. Characterization of the nanoprobe BEAGA/ApoE-PDHD.

BBB-crossing ability of nanoprobe in vitro

Fig. 3. Penetration of BBB-mimicking endothelial tissue and the sequent cellular uptake of nanoprobes.

Potential to cross BBB and accumulate in glioma of mice

Fig. 4. Evaluation of the BBB crossing ability of various nanoprobes in vivo.

LFUS-triggered sonodynamic therapy in vitro

Fig. 5. In vitro anticancer effects of BEAGA/ApoE-PDHD nanoprobe.

Nanoprobe induced ICD under LFUS irradiation and activated immune cells

Fig. 6. Damage-associated molecular patterns (DAMPs) release and bone marrow-derived dendritic cells (BMDCs) maturation after BEAGA/ApoE-PDHD treatment.

In vivo therapeutic efficacy on glioma-bearing mice

Fig. 7. In vivo therapeutic effect on mice bearing brain orthotopic G422-Luc glioma.

Anti-tumor immune response on mice after SDT

Fig. 8. In vivo antitumor immune response in glioma-bearing mice after immuno-sonodynamic treatment.

Discussion

Methods

Ethical statement

Materials

Synthesis of BEAGA2-PEG2k-PDHD

Synthesis of Boc-NH-PEG2k-PBLA85

Synthesis of Boc-NH-PEG2k-PBLA85-C11F23

Synthesis of Boc-NH-PEG2k-PAsp(DEA-co-His-co-DIP)85-C11F23 (Boc-NH-PEG2k-PDHD)

Synthesis of Fmoc-Lys-PEG2k-PDHD

Synthesis of (Fmoc-ABU)2-PEG2k-PDHD

Synthesis of the enzyme and acid dual-responsive polymer BEAGA2-PEG2k-PDHD

Synthesis of N3-PEG3.4k-PDHD

Synthesis of the acid-responsive polymer N3-PEG3.4k-PDHD

Preparation of nanoprobes

Characterization of polymers and nanoprobes

Cell culture

Western blot assay

In vitro BBB model and cellular internalization studies

Penetration of nanoprobes in 3D cell spheroids

Intracellular ROS generation

Cell viability and apoptosis assay

Measurement of mitochondrial potential

Expression level of calreticulin (CALR), adenosine triphosphate (ATP), and high mobility group B1 (HMGB1) in vitro

Bone marrow-derived dendritic cells (BMDCs) maturation

Animal models

In vivo biodistribution of nanoprobes

In vivo fluorescence imaging

In vivo MR imaging

Intravital real-time observation by two-photon CLSM

Anticancer efficacy of nanoprobes

In vivo anticancer effect

Evaluation of tumor-infiltrating immune cells and memory CD8+ T cells

Clinical human specimens

Statistics and reproducibility

Reporting summary

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Data availability

Competing interests

Footnotes

Contributor Information

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

Synthesis of BEAGA₂-PEG_2k-PDHD

Synthesis of Boc-NH-PEG_2k-PBLA₈₅

Synthesis of Boc-NH-PEG_2k-PBLA₈₅-C₁₁F₂₃

Synthesis of Boc-NH-PEG_2k-PAsp(DEA-co-His-co-DIP)₈₅-C₁₁F₂₃ (Boc-NH-PEG_2k-PDHD)

Synthesis of Fmoc-Lys-PEG_2k-PDHD

Synthesis of (Fmoc-ABU)₂-PEG_2k-PDHD

Synthesis of the enzyme and acid dual-responsive polymer BEAGA₂-PEG_2k-PDHD

Synthesis of N₃-PEG_3.4k-PDHD

Synthesis of the acid-responsive polymer N₃-PEG_3.4k-PDHD

Evaluation of tumor-infiltrating immune cells and memory CD8⁺ T cells