Faxiang Yin; Xin Jin; Ligong Zhang; Qiang Xie; Jun Qian

doi:10.4143/crt.2025.849

Articles

Page Path: HOME > J Korean Cancer Assoc > Ahead-of print articles > Article

Original Article Salidroside-Loaded, TMTP1-Modified Cancer Stem Cell–Derived Exosomes Reprogram the PI3K/AKT/mTOR Axis to Overcome PD-1 Resistance in Breast Cancer: Faxiang Yin^1,2, Xin Jin², Ligong Zhang², Qiang Xie², Jun Qian^1,2; DOI: https://doi.org/10.4143/crt.2025.849
Published online: January 14, 2026

¹Anhui Medical University, Hefei, China

²Department of Oncology Surgery, First Affiliated Hospital of Bengbu Medical University, Bengbu, China

Correspondence: Jun Qian, Anhui Medical University, Department of Oncology Surgery, First Affiliated Hospital of Bengbu Medical University, No. 287 Changhuai Road, Bengbu 233000, China,
Tel: 86-13955215036, E-mail: byqianjun01056@163.com

• Received: August 8, 2025 • Accepted: January 11, 2026

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

1,141 Views
70 Download

Full Article

Download PDF

Abstract

Purpose
This study aimed to elucidate how salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) rewire the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) axis to remodel the immune microenvironment and reverse acquired programmed cell death-1 (PD-1) resistance in breast cancer.
Materials and Methods
Cancer stem cell (CSC)–derived exosomes were surface-engineered with TMTP1 peptide and electroporated with salidroside. PD-1–resistant MA782/5s-8101-R cells and an orthotopic mouse model were used. Multi-omics, flow cytometry, enzyme-linked immunosorbent assay, immunofluorescence, in vivo imaging, and molecular assays examined immune and signaling outcomes.
Results
Salidroside@T-exo restored T-cell interferon γ (IFN-γ) and granzyme B secretion, suppressed CD8+ T-cell apoptosis, and inhibited p-PI3K/p-AKT/p-mTOR in T cells. CSC migration, invasion, and stemness (OCT4, NANOG, and SOX2) were markedly reduced. Tumor growth, Ki-67 index, and CSC frequency dropped while terminal deoxynucleotidyl transferase dUTP nick end labeling–positive cells rose.
Conclusion
Salidroside@T-exo reverses PD-1 blockade resistance by simultaneously inhibiting PI3K/AKT/mTOR signaling in T cells and eradicating breast CSCs, offering a clinically translatable strategy for refractory breast cancer immunotherapy.
Key words: Breast neoplasms, Salidroside-loaded, Oligopeptide-modified tumor-derived exosomes, PI3K/AKT/mTOR signaling pathway, PD-1 immunotherapy resistance, Immune microenvironment

Introduction

Breast cancer (BC) is a leading malignancy affecting women globally, characterized by high incidence and mortality [1–3]. Recent statistics indicate a continuous rise in the global incidence of BC, posing significant challenges to public health systems [4–6]. Molecular biology and genetics advances have shifted treatment strategies from conventional surgery, radiation, and chemotherapy to more personalized and targeted treatment strategies [7–9]. Immunotherapy has emerged as a promising treatment modality, demonstrating substantial potential in managing BC [10–12]. However, despite some clinical successes with programmed cell death-1 (PD-1) and programmed cell death-ligand 1 (PD-L1) inhibitors, acquired resistance remains a critical challenge in immunotherapy [13,14]. The immune evasion and resistance mechanisms are complex and varied, resulting in many patients not benefiting from current immunotherapeutic approaches [15]. Therefore, exploring the mechanisms of resistance to immunotherapy in BC and devising strategies to overcome it holds significant clinical importance for enhancing treatment outcomes.

The immune microenvironment (IME) of BC consists of various immune cells, cytokines, chemokines, and associated signaling pathways, which collectively determine tumor growth, metastasis, and patient prognosis [16–18]. The PD-1/PD-L1 axis within this microenvironment serves as a key regulator of tumor immune evasion [19,20]. PD-1 is a surface receptor that inhibits T cell activation upon binding to PD-L1, allowing tumor cells to evade immune surveillance [21–23]. Chronic immune suppression facilitates tumor growth and metastasis and leads to resistance to PD-1 inhibitor therapy [24–26]. Studies indicate that mechanisms of PD-1 resistance may include T cell exhaustion, increased suppressive cytokines in the tumor microenvironment (TME), and aberrant activation of related signaling pathways [27]. These findings highlight the need for deeper exploration of the IME and mechanisms underlying PD-1 resistance in BC.

The phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling axis is a key regulator of cell growth, differentiation, survival, and metabolism, and its dysregulation has been extensively implicated in multiple cancers, including BC [28–30]. Activation of this pathway promotes tumor cell proliferation and survival while inhibiting apoptosis, thus playing a vital role in cancer progression [31]. The PI3K/AKT/mTOR signaling axis profoundly shapes the TME and regulates immune cell function [32–34]. For example, abnormal activation of this pathway is closely associated with immune cell functional suppression and mechanisms of immune escape [35,36]. Therefore, targeting this pathway could not only directly inhibit tumor cells but also potentially enhance the effectiveness of immunotherapy by improving the IME.

In preliminary studies, cancer stem cell–derived exosomes (CSC-exo) and the natural compound salidroside have been shown to regulate tumor biological behaviors. CSC-exo is considered a promising therapeutic tool due to its ability to specifically carry and deliver signaling molecules to the TME [37]. In this study, we chemically modified the surface of CSC-exo with the oligopeptide TMTP1 and introduced salidroside using electroporation technology, creating novel salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo). This innovative structure leverages the targeting capabilities of CSC-exo and the biological activity of salidroside to precisely modulate the PI3K/AKT/mTOR axis, thereby reprogramming the IME of BC at the molecular level.

The study focused on exploring the modulation of the PI3K/AKT/mTOR signaling pathway by Salidroside@T-exo to reshape the IME of BC and reverse acquired resistance to PD-1 blockade therapy. We hypothesized that Salidroside@T-exo would suppress BC cell proliferation and survival at the molecular level while enhancing the tumor immune microenvironment (TIME) through the activation of T cells and other immune populations. This strategy holds scientific importance by offering a novel perspective to understand and address the resistance issues in immunotherapy while offering new molecular targets and therapeutic approaches for the treatment of BC. Clinically, this innovative therapeutic strategy is expected to improve BC patients’ survival rates and quality of life, particularly those who have developed resistance to traditional immune checkpoint inhibitors (ICIs). Thus, this research holds substantial scientific value and significant clinical application potential.

Materials and Methods

1. Multicellular tumor spheroids model

The Multicellular Tumor Spheroids (MCTS) model was used to enrich cancer stem cells (CSCs) from the mouse BC cell line MA782/5s-8101 (AC100914, American Type Culture Collection). Once adherent cultures attained approximately 85%–90% confluence, cells were dissociated into single-cell suspensions and cultured in serum-free Dulbecco’s modified Eagle’s medium (DMEM)/F-12 medium (11320033, Gibco) enriched with 2% B-27 (17504044, Gibco), 10 ng/mL epidermal growth factor (AF-315-09-500UG, PeproTech), and 10 ng/mL basic fibroblast growth factor (100-18C-1MG, PeproTech). The cells were seeded at 1×10⁵ cells/mL on poly(2-hydroxyethyl methacrylate)-coated plates, and the medium was refreshed every 3 days. After 7 days, mature spheroids were collected for subsequent analyses [38].

2. Spheroid formation assay

Cells enriched using the MCTS model were seeded into ultra-low attachment 96-well or 6-well plates (Corning Life Sciences) and cultured in DMEM/F12 medium supplemented with 1% methylcellulose (9004-67-5, Sigma-Aldrich), 1× B-27, and 20 ng/mL each of epidermal growth factor and essential fibroblast growth factor. After 2 weeks, cellular morphology was observed utilizing an optical microscope (CX43, Olympus), and diameters were quantified with LAS V4.9 software.

3. Exosome isolation

When CSCs reached 80%–90% confluence, the culture medium was removed, and cells were washed twice with phosphate-buffered saline (PBS). Serum-free Iscove’s modified Dulbecco’s medium (25 mL, 12440053, Gibco) was added, and cultures were maintained at 37°C with 5% CO₂ for 48 hours. The collected supernatant was centrifuged at 300 ×g for 10 minutes at 4°C to remove cell debris, followed by 2,000 ×g for 10 minutes to eliminate apoptotic bodies and 10,000 ×g for 30 minutes to remove larger vesicles. Exosomes were pelleted by ultracentrifugation at 110,000 ×g for 75 minutes using a Ti70 rotor (Optima XPN-100, Beckman Coulter), resuspended in 200 μL PBS, and stored at −80°C for further use.

4. Exosome identification

Exosome size distribution and concentration were determined by nanoparticle tracking analysis (NTA) using a NanoSight LM10 system (Malvern) equipped with a 640 nm laser and Viton O-ring (Viton). Exosome suspensions were diluted 1:500 in Milli-Q water, loaded into the sample chamber, and analyzed using NanoSight software version 2.3 (NanoSight Ltd.) (gain 6.0, threshold 11). Particle trajectories were recorded to calculate concentration and size distribution, and the final exosome concentration was adjusted according to the dilution factor.

For morphological assessment, exosome pellets were fixed with 2% paraformaldehyde (PFA) and 2.5% glutaraldehyde at 4°C for 1 hour, washed three times with PBS, post-fixed with 1% osmium tetroxide for 1.5 hours, dehydrated through graded ethanol, and embedded in epoxy resin polymerized at 35°C, 45°C, and 60°C (24 hours each). Ultrathin sections were stained with uranyl acetate and lead citrate and observed under a transmission electron microscope (JEM-1011, Jeol) at 80 kV using a Megaview III camera (Soft Imaging System).

Exosomal markers were verified by western blot (WB). Isolated exosomes were lysed in RIPA buffer (89901, Thermo Scientific), and proteins were probed for CD9, CD63, and CD81, with calnexin as a negative control. Antibodies were obtained from Invitrogen (CD9, 14-0091-82; CD63, 967-MSM2-P0; CD81, 975-MSM1-P0; calnexin, MA5-31501; all 1:1,000 dilution).

5. Synthesis of DSPE-PEG-TMTP1

DSPE-PEG-TMTP1 was synthesized via an amide coupling reaction between TMTP1 and DSPE-PEG-COOH incorporating a fluorescent tag. DSPE-PEG-COOH (18 mg) was dissolved in distilled water and activated with EDC (61.2 mg) and NHS (37.2 mg) for 30 minutes. TMTP1 peptide (7 mg) was added, and the reaction proceeded under dark conditions for 10 hours with gentle rotation. Unreacted components were removed by overnight dialysis (MWCO 1000 Da, YOBIOS, YD20DG28). The purified DSPE-PEG-TMTP1 was lyophilized and stored at −20°C until use.

6. Preparation of TMTP1-exo

DSPE-PEG-TMTP1 and CSC-exo were co-incubated at 37°C for 2 hours under light-protected conditions. The mixture was centrifuged at 16,000 ×g for 30 minutes to yield TMTP1-exo. Dual-fluorescence imaging verified conjugation efficiency, with TMTP1 labeled by rhodamine (Rhd) and CSC-exo stained with PKH67. Overlapping Rhd and PKH67 signals under a laser scanning confocal microscope confirmed successful surface coupling.

7. Drug loading and release test

Salidroside was loaded into TMTP1-exo (T-exo) by electroporation (Nepa Gene). Successful loading of FITC-labeled salidroside was verified by flow cytometry (FCM). For the release assay, Salidroside@T-exo (1.0 mg/mL) was suspended in PBS (pH 7.4) and sealed in dialysis bags containing 2 mL of the suspension, which were immersed in 10 mL PBS at 37°C with gentle agitation (100 rpm). At scheduled time points, 2 mL of medium was collected and replenished with fresh PBS. Salidroside levels were determined by high-performance liquid chromatography (HPLC), and cumulative release was plotted over time to generate the in vitro release curve.

8. Flow cytometry

CSCs were dissociated using 0.25% trypsin (25200072, Gibco) and suspended at 1×10⁶ cells in 200 μL of fluorescence-activated cell sorting (FACS) buffer (660585, BD Biosciences). Fluorescent antibodies (2 μL) or isotype controls were added, and samples were incubated on ice for 30 minutes, washed, fixed with 10% formalin (R04587, Merck), and analyzed on a BD FACSCalibur flow cytometer (BD Biosciences). A fluorescein-conjugated CD44 antibody (ab316123, 1:500, Abcam) was used to identify CSC markers.

FCM was used to detect the fluorescence signals of CD44, CD81, and salidroside in Salidroside@T-exo samples to evaluate their surface marker characteristics and drug-loading status. Positive gates were established based on the negative control, and the proportion of positive events in each detection channel was calculated separately. Flow cytometric results were expressed as the percentage of positive events among the total detected events and analyzed using FlowJo software.

Apoptosis of MA782/5s-8101-R cells was assessed with the Annexin V-FITC/propidium iodide kit (556547, BD Biosciences). Following treatment, 2×10⁵ cells per well were harvested, rinsed, and suspended in 500 μL binding buffer before staining with 5 μL each of FITC and propidium iodide in darkness for 15 minutes. Samples were examined on a FACSCalibur cytometer, and apoptotic rates were calculated as the combined percentage of early and late apoptotic cells. All experiments were conducted in triplicate.

Single-cell suspensions from tumor tissues were obtained via collagenase/DNase digestion (1–2 hours) and density gradient centrifugation. Cells (1×10⁶/mL) were stimulated for 5 hours with phorbol 12-myristate 13-acetate (HY-18739, MedChemExpress), ionomycin (HY-13434, MedChemExpress), and GolgiPlug (555029, BD Biosciences) at 37°C, 5% CO₂. To block nonspecific binding, CD16/CD32 antibody (14-0161-82, 0.5 mg/mL, Thermo Fisher Scientific) was added for 10 minutes before surface staining. Cells were then stained with APC-CD19 (17-0193-82), PE-CD3 (12-0031-82), FITC-CD4 (11-0041-82), PE-Cy7-CD8 (A15385), APC–interferon-γ (IFN-γ) (17-7311-82), and PE–interleukin (IL)-4 (12-7041-82) (all from Thermo Fisher Scientific). Gating was determined using unstained and single-stained controls to ensure reliability. All procedures were performed under sterile conditions.

9. Immunofluorescence staining to observe cellular uptake of Exos

CSC-exo were labeled with DiI dye (C1036, Beyotime) by incubating 40 μg of exosomes with 25 μM DiI at ambient temperature for 30 minutes, followed by ultracentrifugation to remove excess dye. MA782/5s-8101 cells were co-cultured with DiO-labeled salidroside and TMTP1 for 12 hours, washed three times with PBS, and fixed with 4% PFA (AR1068, Boster) for 30 minutes. Nuclei were stained with DAPI (C1005, Beyotime) for 30 minutes. Images were captured using a BX53 fluorescence microscope (Olympus) at 400× magnification and analyzed with ImageJ Pro Plus 6.0 software (Media Cybernetics).

10. Cell Counting Kit-8 assay for cell viability

MA782/5s-8101 cells (1×10⁵ per well) were subjected to various treatments for 1–3 days. Cell Counting Kit-8 (CCK-8) solution (C0037, Beyotime) was added at 100 μL/mL and incubated for 2 hours. After incubation, 100 μL of supernatant was transferred to a 96-well plate, and absorbance at 450 nm was recorded using a Synergy Neo2 Hybrid microplate reader (Agilent Technologies).

11. Development of an acquired PD-1–resistant cell model and cell grouping

MA782/5s-8101 cells (0.5×10⁶ in 50 μL PBS) were inoculated subcutaneously into the hind legs of female C3H/HeNCrl mice (12–16 weeks old; Beijing Vital River Laboratory Animal Technology Co., Ltd.). Beginning on day 4 post-inoculation, mice received intraperitoneal injections of anti–PD-1 antibodies (10 mg/kg) twice weekly for four doses. Tumors exhibiting no response to treatment were excised, digested into single-cell suspensions, and cultured in vitro for 2–3 weeks. After four successive passages in syngeneic mice, cells exhibiting stable PD-1 blockade resistance were established as the MA782/5s-8101-R line. Cells were maintained at 37°C and 5% CO₂ in RPMI-1640 medium (R4130, Sigma-Aldrich) supplemented with 10% fetal bovine serum (FBS; F8318, Sigma-Aldrich) and 1% penicillin-streptomycin (V900929, Sigma-Aldrich). Mycoplasma contamination was excluded by routine testing. To identify breast cancer stem-like cells, cells maintained under routine adherent culture conditions were defined as the adherent group, whereas cells obtained after serial sphere culture and passaged to P3 were defined as the spheroid group. Cells from the spheroid group were further used for the isolation and characterization of CSC-exo.

T cells (Bio-73272, Biobw) and MA782/5s-8101-R cells were co-cultured for 24 or 48 hours in a Transwell invasion chamber (354480, Corning), with T cells in the upper chamber and MA782/5s-8101-R cells in the lower chamber.

Recilisib, a radioprotectant that activates AKT and PI3K activity within cells, was used at a concentration of 50 μM for 24 hours [39]. Specific cell groupings are detailed in S1 Table.

12. Establishment of a BC transplant mouse model

Twenty-four female C3H/HeNCrl mice (4–5 weeks old, Beijing Vital River Laboratory Animal Technology Co., Ltd.) were maintained under specific pathogen-free conditions at 26°C–28°C with 50%–65% humidity. All animal experiments complied with ethical standards and were approved by our institution’s Animal Ethics Committee.

For orthotopic tumor transplantation, MA782/5s-8101-R cell suspensions (2×10⁵ cells per site) were mixed with Matrigel and 50% FBS/PBS (1:1, v/v) and injected into the fourth mammary fat pad of 6-week-old female C57BL/6J mice. Once tumors reached approximately 10 mm³, mice were randomly divided into treatment cohorts. Doses of 200 μg of Salidroside@T-exo were administered intravenously, while recilisib was injected at a concentration of 500 mg/kg; the control group received equivalent volumes of PBS, all administered thrice weekly. After 4 weeks, tumor progression was evaluated by in vivo bioluminescence imaging (IVIS Lumina Series III, PerkinElmer). Mice were then euthanized via CO₂ inhalation, and tumors were collected for formalin fixation and paraffin embedding or frozen in liquid nitrogen at −80°C. Grouping details are listed in S2 Table [40,41].

13. Hematoxylin and eosin staining

Mouse tumor tissue specimens fixed in PFA underwent dehydration, paraffin embedding, and sectioning into slices 5 μm thick. Serial sections were collected from both central and peripheral tumor regions at 200 μm intervals. After dewaxing in xylene and rehydration with graded ethanol, sections were stained sequentially with Harris hematoxylin (5 minutes), differentiated in 0.5% hydrochloric acid–ethanol (10 seconds), and counterstained with eosin (40 seconds) (all reagents from Beyotime). Slides were then dehydrated, cleared, and mounted with neutral resin for microscopic observation.

14. Terminal deoxynucleotidyl transferase dUTP nick end labeling staining

Apoptosis was detected using the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) Kit (C1098, Beyotime). Cell slides were incubated with 50 μL of TUNEL reaction mixture for 60 minutes at 37°C in the dark, then treated with 0.2–0.5 mL of DAB substrate for 30 minutes and washed three times with PBS. Images were captured using an inverted microscope, and each condition was analyzed in triplicate.

Apoptosis in tumor sections was examined with a TUNEL kit. Sections were treated with 50 μL of TUNEL reaction mixture for 1 hour at 37°C in the dark, counterstained with DAPI (1 μg/mL, 30 minutes), rinsed thrice with PBS, and visualized under a fluorescence microscope (400×, BX53, Olympus). The apoptotic index was calculated as the ratio of TUNEL-positive to DAPI-stained areas using ImageJ (National Institutes of Health). Each experimental group included five mice, and five tissue sections per mouse were evaluated.

15. High-throughput transcriptome sequencing

Tumor tissues from the BC group (n=3) and the Salidroside@ T-exo group (n=3) were collected for sequencing. Library preparation and sequencing were conducted by CapitalBio Technology using 5 μg of total RNA per sample. Ribosomal RNA was depleted with the Ribo-Zero Magnetic Kit (MRZG12324, Epicentre), and libraries were constructed using the NEBNext Ultra RNA Library Prep Kit for Illumina (E7760S, NEB). RNA was fragmented into approximately 300 bp fragments in NEBNext First Strand Synthesis Reaction Buffer (5×), followed by first-strand cDNA synthesis using random hexamer primers and reverse transcriptase. Second-strand cDNA synthesis was performed with dUTP to ensure strand specificity. The resulting cDNA fragments were end-repaired, A-tailed, and ligated to Illumina sequencing adaptors. After USER enzyme (M5508, NEB) digestion, libraries were amplified and purified by polymerase chain reaction (PCR), validated with an Agilent 2100 Bioanalyzer, and quantified using the KAPA Library Quantification Kit (kk3605, Merck). Sequencing was conducted on an Illumina NextSeq CN500 platform with paired-end reads.

16. Transcriptome sequencing data analysis

Raw paired-end reads were assessed for quality using FastQC v0.11.8. Adapter sequences and poly(A) tails were removed with Cutadapt v1.18, and reads containing over 5% ambiguous bases (N) were excluded. Sequences with at least 70% of bases having Phred scores ≥ 20 were retained using the FASTX Toolkit v0.0.13 and subsequently realigned with BBMap. Clean reads were mapped to the mouse reference genome via HISAT2 v0.7.12.

Differential Analysis: Differential expression analysis of mRNA read counts was performed using the “edgeR” package in R, with selection criteria set at |log₂FC| > 1 and p < 0.05. Overlapping gene expression patterns were visualized with heatmaps generated in R using the heatmap package.

Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed with the clusterProfiler, org.Mm.eg.db (v3.16.0), enrichplot, DOSE, and ggplot2 packages in R (R Foundation for Statistical Computing). Gene annotation was based on the org. Mm.eg.db mouse database.

17. Comprehensive database analysis reveals expression and functional characteristics of the PI3K/AKT/mTOR axis in BC

The TIMER database (https://cistrome.shinyapps.io/timer/) was used to analyze the correlation between the PI3K/AKT/mTOR signaling axis and immune cell infiltration, including CD8⁺ T cells and macrophages, in BC tissues. The TISIDB database (http://cis.hku.hk/TISIDB/) also explored the associations between the PI3K/AKT/mTOR pathway and different T cell types in BC.

18. Quantitative proteomics analysis (DDA-iTRAQ)

Tumor samples from the BC group (n=3) and Salidroside@T-exo group (n=3) were used for proteomic profiling. Total proteins were extracted with RIPA buffer containing protease inhibitors and quantified using the bicinchoninic acid (BCA) assay. Equal protein amounts from each sample were digested with trypsin (enzyme-to-protein ratio 1:50, w/w) at 37°C for 16 hours after pH adjustment to 8.0. The resulting peptides were desalted on a C18 column and labeled with an iTRAQ 8-plex reagent kit (AB Sciex). Peptides were fractionated by high-pH reversed-phase HPLC and analyzed on an Eksigent NanoLC 425 system coupled to a QSTAR Elite Hybrid LC-MS/MS (AB Sciex) in data-dependent acquisition mode. Mass spectrometry data were acquired across 350–1,500 m/z with 15,000 resolution and auto-optimized collision energy. Protein identification and quantification were performed using ProteinPilot 5.0 (AB Sciex) against the Mus musculus UniProt database (release 2023_02), with peptide and protein false discovery rates set at ≤ 1% (Q ≤ 0.01). Quantification relied on iTRAQ reporter ion intensities normalized by the global median method. Differentially expressed proteins (DEPs) were defined as those with |log₂FC| > 1.0 and p < 0.05 (Welch’s t test). Protein functional annotation and GO/KEGG enrichment analyses were performed in R using clusterProfiler and the org.Mm.eg.db database (v3.16.0).

19. Metabolomics analysis

Metabolomic analyses were conducted on tumor tissues from both the BC group (n=6) and the Salidroside@T-exo group (n=6). Each sample (300 μL homogenate) was mixed with 900 μL of 80% methanol containing 0.1% formic acid (FA), vortexed for 2 min, and centrifuged at 12,000 ×g for 10 minutes. The resulting supernatants were transferred to autosampler vials for liquid chromatography–mass spectrometry analysis.

Chromatographic separation was carried out on a Waters ACQUITY UPLC HSS T3 C18 column (100×2.1 mm, 1.8 μm) using a Shimadzu LC20 UPLC system coupled with a Triple TOF 6600 mass spectrometer (AB Sciex). The column temperature was maintained at 40°C with a 0.4 mL/min flow rate. The mobile phase consisted of an acetonitrile-water solution with 0.1% FA. The gradient for phase B was programmed as follows: 5% (0.0–11.0 minutes), 90% (11.0–12.0 minutes), and 5% (12.1–14.0 minutes). Eluents were directly introduced into the mass spectrometer without splitting.

Mass spectrometry parameters included an ionization voltage of 5500 V, capillary temperature of 550°C, and gas flows of 50 psi (nebulizer) and 60 psi (auxiliary). Data were processed by orthogonal partial least squares–discriminant analysis (OPLS-DA), and model validity was confirmed by 100 permutation tests. Metabolites with variable importance in projection > 1 and p < 0.05 were considered differentially expressed. Univariate validation identified significant metabolites with fold change ≥ 1.0 or ≤ 0.5 and p < 0.05. Pathway enrichment analysis of differential metabolites was conducted using MetaboAnalyst 5.0.

20. Immunofluorescence staining to assess protein expression

MA782/5s-8101-R cells and tumor tissues were fixed with 4% PFA at ambient temperature for 15 minutes and washed twice with PBS. Samples were permeabilized with 0.5% Triton X-100 (P0096, Beyotime) for 10 minutes and incubated overnight at 4°C with the following primary antibodies: p-PI3K (1:200, PA5-104853), p-AKT (1:200, MA1-20325), p-mTOR (1:250, 44-1125G), granzyme B (GZMB; 1:200, 701395), CD8 (1:100, PA5-141191). After three PBS washes, FITC- or rhodamine-conjugated phalloidin (F432 or R415, Invitrogen) was applied to label the cytoskeleton, followed by Alexa Fluor 647- or 488-conjugated goat anti-rabbit IgG (ab150083, ab150077, 1:200, Abcam) for 1 hour. Nuclei were counterstained with DAPI (D3571, 10 μg/mL) for 10 minutes, and slides were preserved at 4°C. Fluorescence signals were visualized using an Olympus IMT-2 fluorescence microscope and quantified with ImageJ software by measuring mean intensity normalized to nuclear fluorescence. For each group, five sections and 6–10 randomly chosen fields were analyzed, with all experiments independently repeated three times.

21. Enzyme-linked immunosorbent assay

Cell culture supernatants were analyzed for IFN-γ content using the IFN-γ enzyme-linked immunosorbent assay (ELISA) Kit (ab252363, Abcam). Antigens were prepared in coating buffer at the specified dilution, and wells were subsequently blocked with 5% calf serum (F8318, MSK) for 40 minutes at 37°C. After adding appropriately diluted samples, enzyme-linked secondary antibodies and chromogenic substrate solution were sequentially applied. The reaction was terminated with 50 μL of stop solution per well, and absorbance was recorded at 450 nm using a microplate reader (Bio-Rad) within 20 minutes. A standard curve was generated for quantitative analysis.

22. Immunofluorescence assay for cell proliferation

Cell proliferation was assessed using the EdU Cell Proliferation Kit (C0071S, Beyotime). After staining with Hoechst 33342, cells were visualized under a fluorescence microscope (IMT-2, Olympus). Quantitative analysis was performed with ImageJ software by determining the proportion of EdU-positive cells relative to total Hoechst-labeled nuclei in each microscopic field.

23. Wound healing assay

MA782/5s-8101-R cells were cultured to full confluence in 6-well plates. Linear wounds were created using a sterile 200 μL pipette tip, and wells were washed three times with PBS to remove detached cells. To inhibit cell proliferation before wounding, cultures were pretreated with mitomycin C (1 μg/mL, M5353, Sigma-Aldrich) for 1 hour. Phase-contrast images of wound closure were captured at 0, 24, and 48 hours. The migration rate was determined by comparing the wound widths at each time point and expressed as a percentage of closure. For each condition, five sections and 6–10 random fields were analyzed, with all experiments performed in triplicate.

24. Transwell assay

Cell migration and invasion were evaluated using Transwell inserts with 8.0 μm polycarbonate membranes (CLS3422, Corning) and BioCoat Matrigel chambers (354480, Corning).

For invasion analysis, MA782/5s-8101-R cells (1×10⁴) were plated into Matrigel-coated chambers (354234, BD Biosciences) preincubated with 50 μL Matrigel and polymerized at 37°C for 30 minutes. Migration assays were performed in uncoated inserts under identical conditions. The lower chambers were filled with DMEM containing 10% FBS to provide a chemoattractant gradient. After incubation for 16–24 hours, non-migrated cells were gently removed from the upper membrane surface. Cells that penetrated the membrane were fixed in 4% PFA, stained with crystal violet (C0121, Beyotime), and visualized using an inverted microscope (XDS-900, Caikon). Quantification was performed with ImageJ software (“Analyze Particles” function) based on five sections and 6–10 random fields per well. Each condition was examined in triplicate.

25. Immunohistochemistry experiment

Mouse tumor tissues were fixed in 4% PFA, dehydrated, cleared, and embedded in paraffin before sectioning. Sections were deparaffinized, rehydrated, and incubated with 1% hydrogen peroxide to quench endogenous peroxidase activity. Antigen retrieval was performed by heating the slides in PBS until boiling. Immunostaining was performed using the Universal Two-Step Detection Kit (PV-9000, Proteintech) with the primary antibody Ki-67 (ab15580, 1:500, Abcam). Stained sections were visualized under an optical microscope (CX43, Olympus), where positive signals appeared light brown. For each group (n=6), five sections per mouse and 6–10 random fields per section were analyzed. Quantitative assessment of immunohistochemistry staining intensity was conducted using ImageJ software.

26. WB analysis

Cells and tumor tissues were lysed in RIPA buffer, and protein concentrations were determined using the BCA Protein Assay Kit (A53226, Thermo Fisher Scientific). Equal protein amounts were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (PVH85R, Millipore) via wet electroblotting. Membranes were blocked with 10% bovine serum albumin (37520, Thermo Fisher Scientific) for 1 hour and incubated overnight at 4°C with primary antibodies against CD44 (14-0441-82), OCT4 (MA1-104), NANOG (14-5761-80), SOX2 (740013T), PI3K (MA5-32917), p-PI3K (PA5-104853), AKT (MA5-45046), p-AKT (MA1-20325), mTOR (AHO1232), p-mTOR (44-1125G), GZMB (701395), IFN-γ (AMC4739), IL-12 (16-7123-81), IL-6 (M620), IL-13 (500-P178-BT-1MG), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; MA5-15738) (all 1:1,000, Invitrogen). After washing, membranes were incubated with horseradish peroxidase–conjugated goat anti-rabbit IgG (1:5,000, AN-170-250UG) for 2 hours. Protein visualization was carried out using a Syngene G:BOX F3 imaging system (Antpedia), and band intensities were quantified with ImageJ software. Expression levels were normalized against GAPDH. All in vitro experiments were repeated three times, and in vivo analyses were performed with six mice per group.

27. Reverse transcription quantitative polymerase chain reaction analysis

Total RNA was extracted from cells and tumor tissues using TRIzol (15596026, Thermo Fisher Scientific), and its concentration and purity were assessed with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Complementary DNA was synthesized using the PrimeScript RT Reagent Kit (RR047A, Takara). Quantitative PCR was conducted with the Fast SYBR Green PCR Kit (11736059, Thermo Fisher Scientific). All reactions were performed in triplicate, with GAPDH serving as an internal reference. Relative mRNA expression levels were determined using the 2^−ΔΔCt method, and each assay was independently repeated three times. Primer sequences (Origene) are provided in S3 Table.

28. Statistical analysis

Data for this study were analyzed using SPSS software ver. 21.0 (IBM Corp.). Quantitative data are presented as mean±standard deviation. Initially, tests for normality and homogeneity of variance were conducted. Data meeting these criteria were compared using an unpaired t test between two groups and one-way ANOVA or repeated measures ANOVA for multiple groups. A p-value of less than 0.05 was considered statistically significant.

Results

1. Preparation and characterization of Salidroside@T-exo

BC remains a major global health burden for women [42]. Despite advances in diagnostics and treatments, challenges such as recurrence and resistance to therapy persist, particularly resistance to ICIs like PD-1/PD-L1, which are crucial factors in treatment failure [43,44]. Consequently, developing new strategies to overcome this resistance and enhance therapeutic efficacy is crucial.

In tumor therapy, exosomes, as naturally targeted nanocarriers, have been extensively studied for drug delivery [45]. Utilizing exos can enhance the bioavailability and targeting of drugs while reducing side effects [46]. Salidroside, a natural bioactive compound derived from Rhodiola species, has attracted considerable attention for its potent antioxidant, anti-inflammatory, and antitumor activities [47,48].

In this study, an innovative nanodrug delivery system, Salidroside@T-exo, was constructed. By integrating the natural biocompatibility and targeting ability of exosomes with the anti-inflammatory, antioxidant, and antitumor effects of salidroside, a novel strategy was proposed to overcome PD-1 therapy resistance in BC. Specifically, TMTP1-modified CSC-exo were loaded with salidroside to evaluate their therapeutic potential in BC treatment.

To identify breast cancer stem-like cell populations, cells cultured under standard adherent conditions were defined as the adherent group, whereas cells generated through serial sphere formation and passaged to P3 were defined as the spheroid group. The expression levels of stemness-related markers were then compared between the two groups.

Microscopic examination revealed that at 50,000× magnification, round membrane-bound vesicles characteristic of exosomes were clearly visible (S4E Fig.). NTA showed that the average diameter of CSC-exo was 85.1±0.7 nm (S4F Fig.). WB analysis showed strong expression of exosomal markers CD9, CD63, and CD81, along with reduced calnexin levels compared with CSCs (S4G Fig.), validating the successful purification of CSC-exo.

Salidroside@T-exo was prepared as illustrated in Fig. 1A. TMTP1 was conjugated to DSPE-PEG-COOH via an amide reaction and anchored onto CSC-exo. Salidroside was subsequently encapsulated into the TMTP1-modified CSC-exo through electroporation, yielding Salidroside@T-exo. Transmission electron microscopy and NTA revealed that Salidroside@T-exo retained the spherical morphology of CSC-exo, with a slightly larger mean diameter of 87.1 nm (Fig. 1B). FCM confirmed the successful encapsulation of salidroside, detecting CSC marker CD44, exo marker CD81, and salidroside fluorescence (Fig. 1C). WB analysis showed that the peptide TMTP1 and the loaded salidroside did not significantly alter the expression levels of markers such as CD9, CD63, and CD81 (Fig. 1D). Drug release tests indicated that Salidroside@T-exo released salidroside more slowly compared to free salidroside (Fig. 1E).

Dil staining indicated enhanced exosome internalization by MA782/5s-8101 cells in the T-exo group compared with the blank control, with maximal uptake detected in the Salidroside@T-exo group (S5A Fig.). The CCK-8 assay revealed a decline in cell viability following treatment with T-exo, which became more pronounced upon exposure to Salidroside@T-exo (S5B Fig.). Immunofluorescence confirmed increased intracellular localization of TMTP1 and salidroside in the T-exo and Salidroside@T-exo cohorts, respectively (S5C Fig.). Collectively, these findings suggest that Salidroside@T-exo preserves the structural integrity of CSC-exo, enables sustained drug release, and enhances cellular uptake and cytotoxicity toward BC cells, offering a promising targeted strategy for precision BC therapy.

2. Target identification for BC therapy through high-throughput transcriptomic analysis

In the field of tumor therapy, salidroside has demonstrated regulatory potential on tumor cell growth and survival by modulating multiple signaling pathways, particularly the PI3K/AKT/mTOR pathway [47,49]. This versatility makes salidroside a promising candidate for developing novel anti-tumor therapies. Overcoming resistance to PD-1 therapy remains a significant challenge in BC treatment. Sequencing analysis revealed that tumor tissues from the Salidroside@T-exo group showed 104 upregulated and 52 downregulated genes compared with the BC group (Fig. 2A and B). A protein-protein interaction (PPI) network constructed using STRING and Cytoscape identified AKT1 as the central hub among the top 15 highly connected genes (Fig. 2C and D). Functional enrichment analysis revealed that the identified genes were predominantly associated with glial cell differentiation, cell-cell adhesion, protein kinase B (AKT) signaling, and phosphatidylinositol-mediated signaling, indicating activation of the PI3K/AKT/mTOR pathway.

For cellular component (CC) enrichment, the genes were mainly clustered in categories associated with coagulation regulation, membrane rafts, membrane microdomains, and the phosphatidylinositol 3-kinase complex. In terms of molecular function (MF), enrichment was primarily observed in activities related to 1-phosphatidylinositol-3-kinase, phosphatidylinositol bisphosphate kinase, phosphatidylinositol 3-kinase, and phosphatidylinositol phosphate kinase (Fig. 2E). KEGG pathway analysis revealed significant enrichment in choline metabolism in cancer, AMPK signaling, cellular senescence, and proteoglycans in cancer (Fig. 2F). Gene set enrichment analysis revealed that in the Salidroside@T-exo group, genes were significantly associated with the PI3K/AKT/mTOR axis, suggesting potential therapeutic targets (Fig. 2G and H).

Using the TIMER database, the correlation between the PI3K/AKT/mTOR signaling axis and immune cell infiltration, including CD8⁺ T cells and macrophages, was systematically evaluated. We found significant associations between PI3K-related genes (PIK3CA, PIK3CB, and PIK3CD) and T cells, as well as between AKT-related genes (AKT1, AKT2, and AKT3) and macrophages (S6 Fig.). Further analysis using the TISIDB database explored correlations with different T cell subtypes, revealing that PI3K-related genes were significantly associated with Act_CD4 T cells, AKT-related genes with Tem_CD4 T cells, and mTOR with Act_CD8 cells (S7 Fig.).

These findings underscore the substantial effect of Salidroside@ T-exo on BC cells, particularly in modulating the PI3K/AKT/mTOR axis and its associated immune responses. Through detailed gene and protein network analyses, this study identifies therapeutic targets and elucidates the potential mechanisms by which Salidroside@T-exo modulates the IME to combat drug-resistant BC. Collectively, these findings provide a solid theoretical framework and experimental evidence supporting its potential application in precision oncology.

3. Proteomic insights into the modulation of PI3K/AKT/mTOR signaling axis by Salidroside@T-exo in mice with acquired resistance to PD-1 immunotherapy

Resistance to the PD-1 pathway has become a major obstacle in BC immunotherapy. To investigate how Salidroside@T-exo alleviates this resistance, proteomic profiling was performed on tumor tissues from BC and Salidroside@T-exo-treated mice. Comparative analysis revealed distinct molecular signatures between the two cohorts. Principal component analysis (PCA) and loading plots demonstrated clear separation, indicating that Salidroside@T-exo significantly altered the protein expression landscape associated with PI3K/AKT/mTOR signaling (S8A-C Fig.). OPLS-DA confirmed this distinction, with permutation testing yielding an R2Y=0.999, indicating a robust model (S8D-F Fig.). Differential analysis identified 278 proteins with altered expression levels, including 24 upregulated and 20 downregulated proteins (Fig. 3A and B). Construction of a PPI network using STRING and Cytoscape revealed AKT1 as the top hub protein (Fig. 3C and D). GO enrichment analysis revealed that these proteins were predominantly associated with peptidylserine modification, serine phosphorylation, and the insulin receptor–mediated signaling pathway.

Regarding CC, proteins were primarily enriched in the phosphatidylinositol 3-kinase complex, extrinsic components of the membrane, Golgi transport complex, and U5 snRNP. MF enrichment included protein serine/threonine/tyrosine kinase activity, 1-phosphatidylinositol-3-kinase activity, insulin receptor substrate binding, and phosphatidylinositol bisphosphate kinase activity (Fig. 3E). KEGG pathway analysis indicated significant enrichment in pathways related to non-small cell lung cancer, estrogen signaling, prostate cancer, and endometrial cancer (Fig. 3F).

These findings align with the transcriptomic results, confirming that the DEPs were primarily enriched in the PI3K/AKT/mTOR signaling. Comprehensive proteomic profiling and bioinformatic analyses revealed that Salidroside@T-exo plays a crucial role in mitigating PD-1 immunotherapy resistance by modulating this pathway. Collectively, these results elucidate the molecular basis through which Salidroside@T-exo counteracts tumor resistance, offering solid experimental evidence and a theoretical framework for developing novel therapeutic strategies against drug-resistant BC.

4. Metabolomic insights into key metabolic pathway alterations in BC pathology

This study comprehensively explored how Salidroside@T-exo modulates the tumor metabolic environment and growth dynamics by influencing amino acid metabolic pathways through a comparative analysis of tumor tissues from the BC and Salidroside@T-exo groups. Six samples from each group were analyzed, with the methodological flow depicted in Fig. 4A. PCA and loading plots clearly distinguished the two groups (Fig. 4B-D). OPLS-DA analysis further validated the distinct separation between the groups, yielding a model validation R²Y = 0.859 (> 0.8), indicating robustness (Fig. 4E-G). KEGG pathway analysis revealed significant enrichment in pathways such as phenylalanine, tyrosine, and tryptophan biosynthesis, starch and sucrose metabolism, phenylalanine metabolism, and alanine, aspartate, and glutamate metabolism (Fig. 4H, S9 Table).

Amino acids also act as upstream regulators of mTOR activity [50], facilitating the conversion of branched-chain α-keto acids into their corresponding branched-chain keto acids and subsequently into α-ketoglutarate, further promoting the growth of BC cells [51]. The metabolomic analysis of this study shows that Salidroside@T-exo significantly affects amino acid metabolism in BC cells, particularly in the biosynthesis of phenylalanine, tyrosine, and tryptophan, as well as in starch and sucrose metabolism. These metabolic changes, potentially via modulation of the mTOR signaling pathway, impact BC cell growth and survival, offering new therapeutic targets for metabolic intervention in BC.

5. Salidroside@T-exo inhibits PI3K/AKT/mTOR axis to modulate the BC IME, reversing PD-1 resistance and cell proliferation

Resistance to the PD-1 pathway poses a major barrier to effective BC immunotherapy. Tumor growth curve analysis showed that the resistant+anti–PD-1 group developed significantly larger tumors than the parental+anti-PD-1 group (S10A Fig.). Hematoxylin and eosin (H&E) staining further revealed denser tumor cell infiltration and more disordered tissue architecture in the resistant tumors, indicating enhanced malignancy and immune evasion compared with the parental group (S10B Fig.). Examination under an inverted microscope showed that both MA782/5s-8101 and MA782/5s-8101-R cells grew in monolayers and were round in shape, with no significant morphological differences between them. However, MA782/5s-8101 cells were more tightly clustered, whereas MA782/5s-8101-R cells exhibited less aggregation (S10C Fig.). TUNEL assays demonstrated significantly reduced apoptosis in the MA782/5s-8101-R cells compared to the MA782/5s-8101 cells (S10D Fig.), confirming the successful establishment of the MA782/5s-8101-R cell line.

This study explored the function of Salidroside@T-exo in remodeling the TIME and overcoming PD-1 resistance in BC through modulation of the PI3K/AKT/mTOR axis. WB analysis demonstrated that, relative to the control group, the phosphorylation ratios of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR were markedly decreased in the Salidroside@ T-exo group but significantly increased in the recilisib group. Furthermore, co-treatment with Salidro-side@ T-exo and recilisib further increased these phosphorylation ratios compared to Salidroside@T-exo alone (Fig. 5A). Immunofluorescence staining confirmed these results, showing decreased fluorescence intensities of p-PI3K, p-AKT, and p-mTOR in the Salidroside@T-exo group, with progressive enhancement in the recilisib and Salidroside@T-exo+ recilisib groups (Fig. 5B-D).

This study investigated the enhanced cytotoxicity of T cells through the secretion of GZMB and IFN-γ, which sensitizes T cells to kill tumor cells more effectively [52] (S11A Fig.). In co-culture assays of T cells with MA782/5s-8101-R cells, ELISA analysis revealed a significant increase in IFN-γ secretion in the Salidroside@T-exo group relative to the control. In contrast, IFN-γ levels were markedly decreased in both the recilisib and Salidroside@T-exo+recilisib groups (S11B Fig.). WB analysis revealed that GZMB protein expression in T cells was significantly upregulated following Salidroside@T-exo treatment, while recilisib and combination treatments significantly downregulated its expression (S11C Fig.). FCM confirmed these findings, demonstrating increased levels of both IFN-γ and GZMB in the Salidroside@T-exo group, with progressive reductions in the recilisib and combination groups (S11D and S11E Fig.). Furthermore, immunofluorescence analysis revealed a pronounced increase in CD8⁺/GZMB⁺ double-positive T cells in the Salidroside@T-exo group relative to the control, whereas their abundance was significantly reduced in both the recilisib and Salidroside@ T-exo+recilisib groups (S11F Fig.).

EdU fluorescence assays revealed a pronounced decrease in cell proliferation in the Salidroside@T-exo group relative to the control, whereas proliferation markedly increased in the recilisib group. Notably, co-treatment with Salidroside@T-exo and recilisib further enhanced proliferation relative to Salidroside@T-exo alone (Fig. 6A). Consistently, scratch wound assays showed a significantly reduced rate of wound closure in the Salidroside@T-exo group, while recilisib treatment accelerated migration. The Salidroside@T-exo+Recilisib group exhibited an even faster closure rate than Salidroside@T-exo alone (Fig. 6B). Transwell assays further demonstrated that both cell migration and invasion were suppressed in the Salidroside@T-exo group, whereas these processes were markedly enhanced following recilisib treatment.

Transwell assays further demonstrated that cell migration and invasion were markedly increased in the Salidroside@T-exo+ recilisib group compared with Salidroside@T-exo alone (Fig. 6C and D). FCM analysis revealed that apoptosis was significantly elevated in the Salidroside@T-exo group relative to the control, whereas it was decreased in the recilisib group. Notably, co-treatment with Salidroside@T-exo and recilisib further decreased apoptosis compared to Salidroside@T-exo alone (Fig. 6E).

These findings demonstrate that Salidroside@T-exo effectively enhances T cell function and improves the IME in BC by significantly inhibiting the activation of the PI3K/AKT/mTOR axis. This increases tumor cell sensitivity and reduces cellular proliferation and migration capabilities. These results provide scientific evidence and experimental support for using Salidroside@T-exo as a potential therapeutic strategy to reverse PD-1 resistance.

6. Salidroside@T-exo inhibits tumor growth in PD-1–resistant mice by targeting the PI3K/AKT/mTOR axis and reprogramming the BC IME

Overcoming PD-1 resistance in BC remains a major therapeutic challenge, particularly when the PI3K/AKT/mTOR axis is aberrantly activated [53]. This study employed Salidroside@ T-exo to inhibit this pathway, aiming to explore its effects on tumor proliferation and the IME in a mouse model of PD-1-resistant BC. WB analysis showed that the p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR ratios were markedly decreased in the Salidroside@T-exo group compared with the control, but significantly increased in the recilisib group. Co-treatment with Salidroside@T-exo and recilisib further enhanced pathway activation relative to Salidroside@T-exo alone (Fig. 7A). Immunofluorescence analysis confirmed consistent results, showing significantly decreased fluorescence intensities of p-PI3K, p-AKT, and p-mTOR in the Salidroside@T-exo group relative to the control, pronounced enhancement in the recilisib group, and further elevation in the Salidroside@T-exo+recilisib group (Fig. 7B-D).

FCM analysis revealed marked alterations in immune cell composition among treatment groups. Compared with the control group, Salidroside@T-exo treatment significantly increased the proportions of T cells and B cells within tumor tissues, while these populations were notably decreased in the recilisib group. Moreover, co-administration of Salidroside@ T-exo and recilisib resulted in a significant reduction in T and B cell numbers compared with Salidroside@ T-exo alone (S12A Fig.).

Further analysis demonstrated that the numbers of CD4⁺ and CD8⁺ T cells were markedly elevated in tumors from the Salidroside@T-exo group relative to controls but were significantly reduced following recilisib treatment. Similarly, combined administration of Salidroside@T-exo and recilisib significantly decreased both CD4⁺ and CD8⁺ T cell counts compared with Salidroside@T-exo alone (S12B Fig.).

CD4⁺ T cells were more abundant than CD8⁺ T cells within the tumor microenvironment. Th1 cells primarily secrete pro-inflammatory cytokines such as IFN-γ and IL-12 to mediate cellular immunity, whereas Th2 cells release anti-inflammatory cytokines including IL-6 and IL-13 to promote humoral responses [54]. To further define the immune profile, Th1 and Th2 subsets within CD4⁺ T cells were analyzed. Salidroside@T-exo treatment markedly increased Th1 and reduced Th2 cell proportions compared with the control group. In contrast, recilisib treatment decreased Th1 and elevated Th2 populations, while co-treatment with Salidroside@ T-exo and recilisib further enhanced the Th2 shift relative to Salidroside@T-exo alone (S12C and S12D Fig.).

WB analysis revealed that Salidroside@T-exo treatment markedly increased the expression of pro-inflammatory cytokines IFN-γ and IL-12, while reducing anti-inflammatory cytokines IL-6 and IL-13 compared with the control group. Conversely, recilisib treatment significantly suppressed IFN-γ and IL-12 expression but elevated IL-6 and IL-13 levels. Moreover, co-administration of Salidroside@T-exo and recilisib led to a pronounced reduction in IFN-γ and IL-12 and a concomitant rise in IL-6 and IL-13 levels relative to Salidroside@T-exo alone (S12E Fig.).

In vivo imaging experiments demonstrated significant differences in tumor growth rates among treatment groups. Compared with the control group, Salidroside@T-exo treatment significantly slowed tumor progression, whereas tumors in the recilisib group grew more rapidly. Notably, co-administration of Salidroside@T-exo and recilisib further accelerated tumor growth compared with Salidroside@T-exo alone (Fig. 8A). Further analysis involving visual assessment and tumor volume and weight measurements confirmed these findings. Tumors in the Salidroside@T-exo group were markedly smaller and lighter than those in the control, whereas recilisib treatment led to larger and heavier tumors. The Salidroside@T-exo+recilisib group resulted in even more significant and heavier tumors compared to the Salidroside@T-exo treatment alone (Fig. 8B). Histopathological analysis by H&E staining showed reduced tumor cell infiltration in the Salidroside@T-exo group but increased infiltration in the recilisib group, with further aggravation in the combination group (Fig. 8C). TUNEL assays demonstrated enhanced apoptosis in tumors treated with Salidroside@T-exo and diminished apoptosis following recilisib treatment.

Moreover, Salidroside@T-exo+recilisib decreased tumor cell apoptosis compared to Salidroside@T-exo alone (Fig. 8D). Immunohistochemical analysis of Ki67⁺ expression, a cell proliferation marker, corroborated these results. Ki67⁺ expression was significantly reduced in the Salidroside@ T-exo cohort but elevated in the Salidroside@T-exo+recilisib group relative to both Salidroside@T-exo alone and the control (Fig. 8E).

Collectively, these results indicate that Salidroside@T-exo reshapes the IME in BC by attenuating the PI3K/AKT/mTOR axis, thereby inhibiting tumor growth and promoting apoptosis. The findings underscore the therapeutic potential of Salidroside@T-exo in restoring immune responsiveness and overcoming PD-1 resistance, offering a promising direction for future clinical translation.

Discussion

Immunotherapy has become a transformative strategy in BC treatment, especially for advanced or treatment-resistant cancers [55]. While PD-1 inhibitors have significantly improved outcomes for some patients, the challenge of acquired resistance to immunotherapy remains a critical issue to address [10,56,57]. Researchers have begun exploring novel strategies to modulate the TME and immune evasion mechanisms to enhance therapeutic efficacy. Salidroside@T-exo has garnered scientific interest in this context, particularly its potential to regulate the critical PI3K/AKT/mTOR signaling axis [58,59]. A deeper understanding of this pathway’s role in BC can lead to more effective treatments to reverse immunotherapy resistance.

This study demonstrates that Salidroside@T-exo modulates the PI3K/AKT/mTOR axis, thereby reshaping the IME in BC and reversing acquired resistance to PD-1 immunotherapy. Previous studies have highlighted the critical involvement of the PI3K/AKT/mTOR axis in diverse cancers, particularly in regulating tumor growth, metabolism, and immune escape [60,61]. For instance, Manning and Toker (2017) [62] highlighted this pathway as a crucial regulatory axis in tumor development, affecting cell survival and proliferation. Our findings further delineate the role of Salidroside@T-exo within this pathway and its regulatory mechanisms. We discovered that it impacts the levels of phosphorylated proteins within the pathway and enhances T cell functionality, thereby boosting their cytotoxicity against tumor cells (Fig. 9).

Compared to existing research, this study innovates by using CSC-exo as a drug carrier, capitalizing on the self-renewal and multidirectional differentiation potential of CSCs. In the field of cancer therapy, exosomes have emerged as naturally targeted nanocarriers that can enhance the bioavailability and specificity of therapeutic agents while minimizing side effects [45,46]. In this study, Salidroside@T-exo not only successfully retained the original structure and size of CSC-exo but also exhibited a significantly improved ability for the sustained release of salidroside. Further validation using Dil staining and CCK-8 assays demonstrated the effective targeting and inhibitory activity of Salidroside@T-exo against BC cells. Moreover, immunofluorescence analysis confirmed that TMTP1 modification enhanced the uptake of Salidroside@T-exo by BC cells, offering a new strategy and approach for precision treatment of BC. The intrinsic biocompatibility and targeting capability of the exosomes, combined with the anti-inflammatory, antioxidant, and antitumor properties of salidroside, provide a novel avenue to overcome PD-1 therapy resistance, particularly in BC patients unresponsive to conventional immunotherapies. Previous studies have primarily focused on directly targeting the PI3K/AKT/mTOR signaling pathway with small molecule inhibitors or traditional chemotherapy drugs [63–65]. For instance, one previous study employed direct inhibition of AKT activity in liver cancer treatment [66], contrasting with our approach of modulating this pathway through exo-based delivery. Our method enhances targeting precision and may reduce systemic side effects, offering dual modulation of the TME, especially the IME.

Research indicates that the IME’s state directly impacts immunotherapy’s efficacy. By altering the IME with Salidroside@ T-exo, particularly by enhancing T-cell functionality and reducing the activity of immunosuppressive cells, we offer a new perspective for overcoming PD-1 resistance. This aligns with Duan et al.’s study [67], which modulated tumor-associated macrophages by improving the IME. However, our strategy leverages the pharmacological effects of salidroside and the natural properties of CSC-exo, providing a relatively natural and potentially safer approach to regulate the complex TME.

Resistance to PD-1/PD-L1 pathway inhibitors presents a significant challenge in current cancer therapy. In this study, Salidroside@T-exo reversed PD-1 resistance by reactivating suppressed immune cells, offering a novel strategy for treating BC. This complements existing research, such as Pardoll’s (2012) work [68], which explored resistance mechanisms to ICIs. By combining the natural delivery capabilities of exos with the immunomodulatory functions of salidroside, our approach offers a comprehensive solution to overcome resistance issues.

Our research employed an integrated approach using high-throughput sequencing, proteomics, and metabolomics, enabling a multidimensional understanding of the mechanisms behind Salidroside@T-exo. This multi-omics strategy provided a more comprehensive perspective on how modulation of the PI3K/AKT/mTOR axis can influence tumor biological behavior. Compared to single-technology approaches, this integrated method better reveals the interactions and regulatory networks within complex BP, providing a scientific basis for future research directions and the development of therapeutic strategies.

Although our study yielded promising results, there are several limitations. First, despite encouraging outcomes in vitro and mouse models, the complex human body environment might yield different therapeutic effects. This study is the first to demonstrate that Salidroside@T-exo overcomes PD-1 immunotherapy resistance in BC by suppressing the PI3K/AKT/mTOR axis and remodeling the TIME in preclinical models. However, it should be clearly noted that all experimental data were derived from murine models, whose TME and immune system differ from those of humans, representing a major limitation of this study. Future investigations will aim to validate the therapeutic efficacy and biosafety of Salidroside@T-exo in large-animal models and patient-derived samples. Analyses using clinical specimens and publicly available datasets such as The Cancer Genome Atlas and METABRIC will be critical to corroborate these preclinical findings. Moreover, the potential synergistic benefits of combining Salidroside@T-exo with current immunotherapies will be examined to accelerate its clinical translation. Long-term safety and efficacy assessments through large-scale clinical trials remain essential to establish its translational feasibility. Also, the PD-1 resistance model used in our study may not fully replicate all resistance mechanisms present in clinical scenarios. Most of the omics analyses in this study were based on biological triplicates (n=3 per group). This sample size meets the basic standards for omics research, ensuring experimental consistency and data quality while allowing for the identification of reliable differential results. However, the relatively small sample size may limit the detection of subtle molecular changes and reduce statistical power. To address this limitation, we performed multidimensional experimental validation of key findings and plan to increase the sample size in future studies to enhance statistical robustness and the generalizability of our conclusions.

Future research should address these limitations, mainly through preclinical and clinical trials, to validate and optimize the therapeutic potential of Salidroside@T-exo. Further investigations should also explore the combinatory effects of other immunotherapy drugs and their applicability to different types and genetic backgrounds of tumor models. Our study offers new insights and approaches for treating BC, particularly demonstrating new possibilities in reversing immunotherapy resistance. By continuing in-depth research and development, Salidroside@T-exo promises to become an effective treatment tool, offering hope to patients with BC.

Electronic Supplementary Material

Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).

crt-2025-849_S1_Table.pdf

crt-2025-849_S2_Table.pdf

crt-2025-849_S3_Table.pdf

crt-2025-849_S4_Fig.pdf

crt-2025-849_S5_Fig.pdf

crt-2025-849_S6_Fig.pdf

crt-2025-849_S7_Fig.pdf

crt-2025-849_S8_Fig.pdf

crt-2025-849_S9_Table.pdf

crt-2025-849_S10_Fig.pdf

crt-2025-849_S11_Fig.pdf

crt-2025-849_S12_Fig.pdf

NOTES

Ethical Statement

This study does not involve any clinical ethics approval. All animal experiments were approved by the Animal Ethics Committee of First Affiliated Hospital of Bengbu Medical University (No. 2023-481).

Author Contributions

Conceived and designed the analysis: Yin F, Jin X, Zhang L, Qian J.

Collected the data: Yin F, Jin X, Zhang L, Xie Q, Qian J.

Contributed data or analysis tools: Yin F, Jin X, Zhang L, Xie Q, Qian J.

Performed the analysis: Yin F, Zhang L, Xie Q, Qian J.

Wrote the paper: Yin F, Jin X, Qian J.

Conflicts of Interest

Conflict of interest relevant to this article was not reported.

Funding

This study was supported by a grant from the Anhui Provincial Department of Education (No. 2022AH051479).

Fig. 1

Preparation and characterization of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo). (A) Schematic diagram of the preparation process for Salidroside@T-exo. (B) Transmission electron microscopy and nanoparticle tracking analysis showing the morphological characteristics and size distribution of cancer stem cell–derived exosomes (CSC-exo), TMTP1-exo (T-exo), and Salidroside@T-exo. Scale bars=200 nm. (C) Flow cytometry detection of the fluorescence signals of the CSC marker CD44, the exosomal marker CD81, and salidroside. Percentages indicate the proportion of fluorescence-positive events among the total detected events within the defined gating region. (D) Western blot analysis of CD9, CD63, CD81, and calnexin expression in CSC-exo, T-exo, and Salidroside@T-exo. (E) Drug release profiles for free salidroside and Salidroside@T-exo. Each experiment was repeated three times, and values are presented as mean±standard deviation.

Fig. 2

Target identification for breast cancer (BC) intervention based on high-throughput sequencing data. (A) Volcano plot of differential genes between BC group and salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) group in mouse tumor tissues derived from high-throughput sequencing data. (B) Heatmap of differential genes between Salidroside@T-exo group and BC group in mouse tumor tissues from high-throughput sequencing data. (C) Display of degree values for the top 15 most connected genes in the network. (D) Construction of a protein-protein interaction network for differential genes. (E) Bubble chart of gene ontology enrichment analysis for differential genes between BC group and Salidroside@T-exo group in mouse tumor tissues from high-throughput sequencing data. (F) Bubble chart of Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differential genes between BC group and Salidroside@T-exo group in mouse tumor tissues from high-throughput sequencing data. (G) Gene set enrichment analysis (GSEA) enrichment analysis of gene expression profiles in tumor groups of BC and Salidroside@T-exo mice. (H) Mountain plot of GSEA enrichment analysis of gene expression profiles in tumor groups of BC and Salidroside@T-exo mice. Each group consisted of three mice; values are presented as mean±standard deviation. BP, biological process; CC, cellular component; FDR, false discovery rate; MF, molecular function; NES, normalized enrichment score.

Fig. 3

Selection of potential targets in breast cancer (BC) based on proteomics data. (A) Volcano plot of differential proteins between BC group and salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) group in mouse tumor tissues from proteomics data. (B) Heatmap of differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues from proteomics data. (C) Construction of the differential protein-protein interaction network. (D) Display of Degree values for the top 15 proteins with the highest connectivity in the network. (E) Bubble chart of gene ontology enrichment analysis for differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues. (F) Bubble chart of Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues. Each group consisted of three mice, and values are presented as mean±standard deviation. BP, biological process; CC, cellular component; MF, molecular function.

Fig. 4

Key metabolic pathways in breast cancer (BC) pathology revealed by metabolomics data. (A) Technical roadmap for revealing critical metabolic pathways in BC pathology through metabolomics. (B) Principal component analysis (PCA) plot of metabolomics data from six BC and six salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) mouse tumor tissues. (C) 3D-PCA plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (D) Loading plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. eQTL, expression quantitative trait locus; GWAS, genome-wide association study; mQTL, metabolite quantitative trait locus. (E) Orthogonal partial least squares–discriminant analysis plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (F) S-plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (G) Permutation plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (H) Kyoto Encyclopedia of Genes and Genomes pathway analysis of 23 differential metabolites.

Fig. 5

Impact of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) on the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway. (A) Western blot analysis of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR expression in MA782/5s-8101-R cells. (B) Immunofluorescence staining of p-PI3K expression in MA782/5s-8101-R cells. (C) Immunofluorescence staining of p-AKT expression in MA782/5s-8101-R cells. (D) Immunofluorescence staining of p-mTOR expression in BC MA782/5s-8101-R cells. Scale bars=25 μm (B-D). Each experiment was repeated three times; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 6

Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) inhibit cell proliferation by modulating the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin pathway and reshaping the immune microenvironment in breast cancer programmed cell death-1–resistant cells. (A) EdU immunofluorescence staining of proliferation in MA782/5s-8101-R cells. (B) Wound healing assay measuring the wound closure speed in MA782/5s-8101-R cells. (C) Transwell assay of migration in MA782/5s-8101-R cells. (D) Transwell assay of invasion in MA782/5s-8101-R cells. Scale bars=50 μm (A, C, D), 100 μm (B). (E) Flow cytometry analysis of apoptosis in MA782/5s-8101-R cells. Each experiment was repeated three times; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 7

Effects of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) on the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway. (A) Western blot analysis of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR expression in mouse tumor tissues. (B) Immunofluorescence staining showing p-PI3K expression in mouse tumor tissues. GAPDH, glyceraldehyde 3-phosphate dehydrogenase. (C) Immunofluorescence staining showing p-AKT expression in mouse tumor tissues. (D) Immunofluorescence staining showing p-mTOR expression in breast cancer mouse tumor tissues. Scale bars=50 μm (B-D). Each group consisted of six mice; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 8

Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) inhibit tumor proliferation by modulating the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin pathway and reshaping the immune microenvironment in programmed cell death-1–resistant mice. (A) In vivo imaging of tumor growth in different groups. (B) Images of tumors and measurements of tumor volume and weight in different groups. (C) H&E staining of tumor tissue morphology and structure in different groups. (D) TUNEL staining of cell apoptosis in tumor tissues of different groups. (E) Immunohistochemistry of Ki67 expression in tumor tissues of different groups. Scale bars=50 μm (C-E). Each group consisted of six mice; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 9

Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) induce reversal of acquired resistance in immunotherapy for breast cancer through modulation of the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathway and reshaping the immune microenvironment. PD-1, programmed cell death-1.

REFERENCES

1. Wu D, Zhou R, Chen H, Pan Y, Tang Y, Zhou D. The role of traditional Chinese medicine in the management of cervical cancer. Am J Chin Med. 2024;52:1013–25. Article PubMed
2. Nagy Z, Jeselsohn R. ESR1 fusions and therapeutic resistance in metastatic breast cancer. Front Oncol. 2022;12:1037531.Article PubMed PMC
3. Ye L, Zhong F, Sun S, Ou X, Yuan J, Zhu J, et al. Tamoxifen induces ferroptosis in MCF-7 organoid. J Cancer Res Ther. 2023;19:1627–35. Article PubMed
4. Wilkinson L, Gathani T. Understanding breast cancer as a global health concern. Br J Radiol. 2022;95:20211033.Article PubMed PMC PDF
5. Boutas I, Kontogeorgi A, Dimitrakakis C, Kalantaridou SN. Soy isoflavones and breast cancer risk: a meta-analysis. In Vivo. 2022;36:556–62. Article PubMed PMC
6. Bodewes FT, van Asselt AA, Dorrius MD, Greuter MJ, de Bock GH. Mammographic breast density and the risk of breast cancer: a systematic review and meta-analysis. Breast. 2022;66:62–8. Article PubMed PMC
7. Derakhshan F, Reis-Filho JS. Pathogenesis of triple-negative breast cancer. Annu Rev Pathol. 2022;17:181–204. Article PubMed PMC
8. Nolan E, Lindeman GJ, Visvader JE. Deciphering breast cancer: from biology to the clinic. Cell. 2023;186:1708–28. Article PubMed
9. Kumar T, Nee K, Wei R, He S, Nguyen QH, Bai S, et al. A spatially resolved single-cell genomic atlas of the adult human breast. Nature. 2023;620:181–91. Article PubMed PMC PDF
10. Liu Y, Hu Y, Xue J, Li J, Yi J, Bu J, et al. Advances in immunotherapy for triple-negative breast cancer. Mol Cancer. 2023;22:145.Article PubMed PMC PDF
11. Ye F, Dewanjee S, Li Y, Jha NK, Chen ZS, Kumar A, et al. Advancements in clinical aspects of targeted therapy and immunotherapy in breast cancer. Mol Cancer. 2023;22:105.Article PubMed PMC PDF
12. Friedlaender A, Subbiah V, Russo A, Banna GL, Malapelle U, Rolfo C, et al. EGFR and HER2 exon 20 insertions in solid tumours: from biology to treatment. Nat Rev Clin Oncol. 2022;19:51–69. Article PubMed PDF
13. Weng J, Li S, Zhu Z, Liu Q, Zhang R, Yang Y, et al. Exploring immunotherapy in colorectal cancer. J Hematol Oncol. 2022;15:95.Article PubMed PMC PDF
14. Kumar S, Sarthi P, Mani I, Ashraf MU, Kang MH, Kumar V, et al. Epitranscriptomic approach: to improve the efficacy of ICB therapy by co-targeting intracellular checkpoint CISH. Cells. 2021;10:2250.Article PubMed PMC
15. Bagati A, Kumar S, Jiang P, Pyrdol J, Zou AE, Godicelj A, et al. Integrin αvβ6-TGFβ-SOX4 pathway drives immune evasion in triple-negative breast cancer. Cancer Cell. 2021;39:54–67. Article PubMed PMC
16. Tan K, Naylor MJ. Tumour microenvironment-immune cell interactions influencing breast cancer heterogeneity and disease progression. Front Oncol. 2022;12:876451.Article PubMed PMC
17. Cheng X, Wang Z. Immune modulation of metastatic niche formation in the bone. Front Immunol. 2021;12:765994.Article PubMed PMC
18. Mathias C, Muzzi JC, Antunes BB, Gradia DF, Castro MA, Carvalho de Oliveira J. Unraveling immune-related lncRNAs in breast cancer molecular subtypes. Front Oncol. 2021;11:692170.Article PubMed PMC
19. Yi M, Li T, Niu M, Wu Y, Zhao Z, Wu K. TGF-β: a novel predictor and target for anti-PD-1/PD-L1 therapy. Front Immunol. 2022;13:1061394.Article PubMed PMC
20. Liu M, Wei F, Wang J, Yu W, Shen M, Liu T, et al. Myeloid-derived suppressor cells regulate the immunosuppressive functions of PD-1(-)PD-L1(+) Bregs through PD-L1/PI3K/AKT/NF-κB axis in breast cancer. Cell Death Dis. 2021;12:465.Article PubMed PMC PDF
21. Yu ZZ, Liu YY, Zhu W, Xiao D, Huang W, Lu SS, et al. ANXA1-derived peptide for targeting PD-L1 degradation inhibits tumor immune evasion in multiple cancers. J Immunother Cancer. 2023;11:e006345.Article PubMed PMC
22. Lopez de Rodas M, Nagineni V, Ravi A, Datar IJ, Mino-Kenudson M, Corredor G, et al. Role of tumor infiltrating lymphocytes and spatial immune heterogeneity in sensitivity to PD-1 axis blockers in non-small cell lung cancer. J Immunother Cancer. 2022;10:e004440.Article PubMed PMC
23. Rodrigues M, Vanoni G, Loap P, Dubot C, Timperi E, Minsat M, et al. Nivolumab plus chemoradiotherapy in locally-advanced cervical cancer: the NICOL phase 1 trial. Nat Commun. 2023;14:3698.Article PubMed PMC PDF
24. Que H, Fu Q, Lan T, Tian X, Wei X. Tumor-associated neutrophils and neutrophil-targeted cancer therapies. Biochim Biophys Acta Rev Cancer. 2022;1877:188762.Article PubMed
25. Zhai J, Chen H, Wong CC, Peng Y, Gou H, Zhang J, et al. ALKBH5 drives immune suppression via targeting AXIN2 to promote colorectal cancer and is a target for boosting immunotherapy. Gastroenterology. 2023;165:445–62. Article PubMed
26. Saginala K, Barsouk A, Aluru JS, Rawla P, Barsouk A. Epidemiology of melanoma. Med Sci (Basel). 2021;9:63.Article PubMed PMC
27. Liu M, Li Z, Yao W, Zeng X, Wang L, Cheng J, et al. IDO inhibitor synergized with radiotherapy to delay tumor growth by reversing T cell exhaustion. Mol Med Rep. 2020;21:445–53. Article PubMed
28. Garcia-Lezana T, Lopez-Canovas JL, Villanueva A. Signaling pathways in hepatocellular carcinoma. Adv Cancer Res. 2021;149:63–101. PubMed
29. Luo Q, Du R, Liu W, Huang G, Dong Z, Li X. PI3K/Akt/mTOR signaling pathway: role in esophageal squamous cell carcinoma, regulatory mechanisms and opportunities for targeted therapy. Front Oncol. 2022;12:852383.Article PubMed PMC
30. Huang J, Chen L, Wu J, Ai D, Zhang JQ, Chen TG, et al. Targeting the PI3K/AKT/mTOR signaling pathway in the treatment of human diseases: current status, trends, and solutions. J Med Chem. 2022;65:16033–61. Article PubMed PDF
31. Zhang L, Liu J, Deng M, Chen X, Jiang L, Zhang J, et al. Enterococcus faecalis promotes the progression of colorectal cancer via its metabolite: biliverdin. J Transl Med. 2023;21:72.Article PubMed PMC PDF
32. Glaviano A, Foo AS, Lam HY, Yap KC, Jacot W, Jones RH, et al. PI3K/AKT/mTOR signaling transduction pathway and targeted therapies in cancer. Mol Cancer. 2023;22:138.Article PubMed PMC PDF
33. He J, Huang Z, Han L, Gong Y, Xie C. Mechanisms and management of 3rd-generation EGFR-TKI resistance in advanced non-small cell lung cancer (review). Int J Oncol. 2021;59:90.Article PubMed PMC
34. Barzegar Behrooz A, Talaie Z, Jusheghani F, Los MJ, Klonisch T, Ghavami S. Wnt and PI3K/Akt/mTOR survival pathways as therapeutic targets in glioblastoma. Int J Mol Sci. 2022;23:1353.Article PubMed PMC
35. Oyewole-Said D, Konduri V, Vazquez-Perez J, Weldon SA, Levitt JM, Decker WK. Beyond T-cells: functional characterization of CTLA-4 expression in immune and non-immune cell types. Front Immunol. 2020;11:608024.Article PubMed PMC
36. Du Y, Hou S, Chen Z, Li W, Li X, Zhou W. Comprehensive analysis identifies PKP3 overexpression in pancreatic cancer related to unfavorable prognosis. Biomedicines. 2023;11:2472.Article PubMed PMC
37. Zhang S, Jiang L, Hu H, Wang H, Wang X, Jiang J, et al. Pretreatment of exosomes derived from hUCMSCs with TNF-α ameliorates acute liver failure by inhibiting the activation of NLRP3 in macrophage. Life Sci. 2020;246:117401.Article PubMed PMC
38. Miao Y, Wang P, Huang J, Qi X, Liang Y, Zhao W, et al. Metabolomics, transcriptome and single-cell RNA sequencing analysis of the metabolic heterogeneity between oral cancer stem cells and differentiated cancer cells. Cancers (Basel). 2024;16:237.Article PubMed PMC
39. Kang AD, Cosenza SC, Bonagura M, Manair M, Reddy MV, Reddy EP. ON01210.Na (Ex-RAD(R)) mitigates radiation damage through activation of the AKT pathway. PLoS One. 2013;8:e58355.Article PubMed PMC
40. Lee KM, Lin CC, Servetto A, Bae J, Kandagatla V, Ye D, et al. Epigenetic repression of STING by MYC promotes immune evasion and resistance to immune checkpoint inhibitors in triple-negative breast cancer. Cancer Immunol Res. 2022;10:829–43. Article PDF
41. Perez-Nunez I, Rozalen C, Palomeque JA, Sangrador I, Dalmau M, Comerma L, et al. LCOR mediates interferon-independent tumor immunogenicity and responsiveness to immune-checkpoint blockade in triple-negative breast cancer. Nat Cancer. 2022;3:355–70. Article PubMed PDF
42. Veronesi U, Boyle P, Goldhirsch A, Orecchia R, Viale G. Breast cancer. Lancet. 2005;365:1727–41. Article PubMed
43. Vranic S, Cyprian FS, Gatalica Z, Palazzo J. PD-L1 status in breast cancer: current view and perspectives. Semin Cancer Biol. 2021;72:146–54. Article PubMed PMC
44. Kim M, Lee JS, Kim W, Lee JH, Jun BH, Kim KS, et al. Aptamer-conjugated nano-liposome for immunogenic chemotherapy with reversal of immunosuppression. J Control Release. 2022;348:893–910. Article PubMed
45. Gao X, Ran N, Dong X, Zuo B, Yang R, Zhou Q, et al. Anchor peptide captures, targets, and loads exosomes of diverse origins for diagnostics and therapy. Sci Transl Med. 2018;10:eaat0195.Article PubMed
46. Mondal J, Pillarisetti S, Junnuthula V, Saha M, Hwang SR, Park IK, et al. Hybrid exosomes, exosome-like nanovesicles and engineered exosomes for therapeutic applications. J Control Release. 2023;353:1127–49. Article PubMed
47. Rong L, Li Z, Leng X, Li H, Ma Y, Chen Y, et al. Salidroside induces apoptosis and protective autophagy in human gastric cancer AGS cells through the PI3K/Akt/mTOR pathway. Biomed Pharmacother. 2020;122:109726.Article PubMed
48. Pan Y, He Y, Zhao X, Pan Y, Meng X, Lv Z, et al. Engineered red blood cell membrane-coating salidroside/indocyanine green nanovesicles for high-efficiency hypoxic targeting phototherapy of triple-negative breast cancer. Adv Healthc Mater. 2022;11:e2200962.Article PubMed PDF
49. Yan R, Xu H, Fu X. Salidroside protects hypoxia-induced injury by up-regulation of miR-210 in rat neural stem cells. Biomed Pharmacother. 2018;103:1490–7. Article PubMed
50. Jung MK, Okekunle AP, Lee JE, Sung MK, Lim YJ. Role of branched-chain amino acid metabolism in tumor development and progression. J Cancer Prev. 2021;26:237–43. Article PubMed PMC
51. Zhang L, Han J. Branched-chain amino acid transaminase 1 (BCAT1) promotes the growth of breast cancer cells through improving mTOR-mediated mitochondrial biogenesis and function. Biochem Biophys Res Commun. 2017;486:224–31. Article PubMed
52. McKenzie MD, Dudek NL, Mariana L, Chong MM, Trapani JA, Kay TW, et al. Perforin and Fas induced by IFNγ and TNFα mediate beta cell death by OT-I CTL. Int Immunol. 2006;18:837–46. Article PubMed
53. Exposito F, Redrado M, Houry M, Hastings K, Molero-Abraham M, Lozano T, et al. PTEN loss confers resistance to anti-PD-1 therapy in non-small cell lung cancer by increasing tumor infiltration of regulatory T cells. Cancer Res. 2023;83:2513–26. Article PubMed PMC PDF
54. Yao Y, Xue H, Chen X, Cao Y, Yu J, Jiang X, et al. Polarization of helper T lymphocytes maybe involved in the pathogenesis of lumbar disc herniation. Iran J Allergy Asthma Immunol. 2017;16:347–57. PubMed
55. Mungra N, Biteghe FA, Malindi Z, Huysamen AM, Karaan M, Hardcastle NS, et al. CSPG4 as a target for the specific killing of triple-negative breast cancer cells by a recombinant SNAP-tag-based antibody-auristatin F drug conjugate. J Cancer Res Clin Oncol. 2023;149:12203–25. Article PubMed PMC PDF
56. Hong R, Xu B. Breast cancer: an up-to-date review and future perspectives. Cancer Commun (Lond). 2022;42:913–36. Article PubMed PMC PDF
57. Salas-Benito D, Perez-Gracia JL, Ponz-Sarvise M, Rodriguez-Ruiz ME, Martinez-Forero I, Castanon E, et al. Paradigms on immunotherapy combinations with chemotherapy. Cancer Discov. 2021;11:1353–67. Article PubMed PDF
58. Zhao J, Liu GW, Tao C. Hotspots and future trends of autophagy in traditional Chinese medicine: a bibliometric analysis. Heliyon. 2023;9:e20142.Article PubMed PMC
59. Postwala H, Shah Y, Parekh PS, Chorawala MR. Unveiling the genetic and epigenetic landscape of colorectal cancer: new insights into pathogenic pathways. Med Oncol. 2023;40:334.Article PubMed PDF
60. Lv S, Jiang H, Yu L, Zhang Y, Sun L, Xu J. SNX14 inhibits autophagy via the PI3K/AKT/mTOR signaling cascade in breast cancer cells. J Mol Histol. 2024;55:391–401. Article PubMed PDF
61. Han EJ, Choi EY, Jeon SJ, Lee SW, Moon JM, Jung SH, et al. Piperine induces apoptosis and autophagy in HSC-3 human oral cancer cells by regulating PI3K signaling pathway. Int J Mol Sci. 2023;24:13949.Article PubMed PMC
62. Manning BD, Toker A. AKT/PKB signaling: navigating the network. Cell. 2017;169:381–405. Article PubMed PMC
63. Yu L, Wei J, Liu P. Attacking the PI3K/Akt/mTOR signaling pathway for targeted therapeutic treatment in human cancer. Semin Cancer Biol. 2022;85:69–94. Article PubMed
64. Zhu Z, McGray AJ, Jiang W, Lu B, Kalinski P, Guo ZS. Improving cancer immunotherapy by rationally combining oncolytic virus with modulators targeting key signaling pathways. Mol Cancer. 2022;21:196.Article PubMed PMC PDF
65. Deng H, Chen Y, Wang L, Zhang Y, Hang Q, Li P, et al. PI3K/mTOR inhibitors promote G6PD autophagic degradation and exacerbate oxidative stress damage to radiosensitize small cell lung cancer. Cell Death Dis. 2023;14:652.Article PubMed PMC PDF
66. Zhai B, Hu F, Jiang X, Xu J, Zhao D, Liu B, et al. Inhibition of Akt reverses the acquired resistance to sorafenib by switching protective autophagy to autophagic cell death in hepatocellular carcinoma. Mol Cancer Ther. 2014;13:1589–98. Article PubMed PDF
67. Duan Z, Luo Y. Targeting macrophages in cancer immunotherapy. Signal Transduct Target Ther. 2021;6:127.Article PubMed PMC PDF
68. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12:252–64. Article PubMed PMC PDF

Figure & Data

REFERENCES

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Salidroside-Loaded, TMTP1-Modified Cancer Stem Cell–Derived Exosomes Reprogram the PI3K/AKT/mTOR Axis to Overcome PD-1 Resistance in Breast Cancer

DOI: https://doi.org/10.4143/crt.2025.849

XML Download

Abstract
Introduction
Materials and Methods
Results
Discussion
Electronic Supplementary Material
NOTES
REFERENCES

Salidroside-Loaded, TMTP1-Modified Cancer Stem Cell–Derived Exosomes Reprogram the PI3K/AKT/mTOR Axis to Overcome PD-1 Resistance in Breast Cancer

Fig. 1 Preparation and characterization of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo). (A) Schematic diagram of the preparation process for Salidroside@T-exo. (B) Transmission electron microscopy and nanoparticle tracking analysis showing the morphological characteristics and size distribution of cancer stem cell–derived exosomes (CSC-exo), TMTP1-exo (T-exo), and Salidroside@T-exo. Scale bars=200 nm. (C) Flow cytometry detection of the fluorescence signals of the CSC marker CD44, the exosomal marker CD81, and salidroside. Percentages indicate the proportion of fluorescence-positive events among the total detected events within the defined gating region. (D) Western blot analysis of CD9, CD63, CD81, and calnexin expression in CSC-exo, T-exo, and Salidroside@T-exo. (E) Drug release profiles for free salidroside and Salidroside@T-exo. Each experiment was repeated three times, and values are presented as mean±standard deviation.

Fig. 2 Target identification for breast cancer (BC) intervention based on high-throughput sequencing data. (A) Volcano plot of differential genes between BC group and salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) group in mouse tumor tissues derived from high-throughput sequencing data. (B) Heatmap of differential genes between Salidroside@T-exo group and BC group in mouse tumor tissues from high-throughput sequencing data. (C) Display of degree values for the top 15 most connected genes in the network. (D) Construction of a protein-protein interaction network for differential genes. (E) Bubble chart of gene ontology enrichment analysis for differential genes between BC group and Salidroside@T-exo group in mouse tumor tissues from high-throughput sequencing data. (F) Bubble chart of Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differential genes between BC group and Salidroside@T-exo group in mouse tumor tissues from high-throughput sequencing data. (G) Gene set enrichment analysis (GSEA) enrichment analysis of gene expression profiles in tumor groups of BC and Salidroside@T-exo mice. (H) Mountain plot of GSEA enrichment analysis of gene expression profiles in tumor groups of BC and Salidroside@T-exo mice. Each group consisted of three mice; values are presented as mean±standard deviation. BP, biological process; CC, cellular component; FDR, false discovery rate; MF, molecular function; NES, normalized enrichment score.

Fig. 3 Selection of potential targets in breast cancer (BC) based on proteomics data. (A) Volcano plot of differential proteins between BC group and salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) group in mouse tumor tissues from proteomics data. (B) Heatmap of differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues from proteomics data. (C) Construction of the differential protein-protein interaction network. (D) Display of Degree values for the top 15 proteins with the highest connectivity in the network. (E) Bubble chart of gene ontology enrichment analysis for differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues. (F) Bubble chart of Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for differential proteins between BC group and Salidroside@T-exo group in mouse tumor tissues. Each group consisted of three mice, and values are presented as mean±standard deviation. BP, biological process; CC, cellular component; MF, molecular function.

Fig. 4 Key metabolic pathways in breast cancer (BC) pathology revealed by metabolomics data. (A) Technical roadmap for revealing critical metabolic pathways in BC pathology through metabolomics. (B) Principal component analysis (PCA) plot of metabolomics data from six BC and six salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) mouse tumor tissues. (C) 3D-PCA plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (D) Loading plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. eQTL, expression quantitative trait locus; GWAS, genome-wide association study; mQTL, metabolite quantitative trait locus. (E) Orthogonal partial least squares–discriminant analysis plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (F) S-plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (G) Permutation plot of metabolomics data from six BC and six Salidroside@T-exo mouse tumor tissues. (H) Kyoto Encyclopedia of Genes and Genomes pathway analysis of 23 differential metabolites.

Fig. 5 Impact of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) on the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway. (A) Western blot analysis of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR expression in MA782/5s-8101-R cells. (B) Immunofluorescence staining of p-PI3K expression in MA782/5s-8101-R cells. (C) Immunofluorescence staining of p-AKT expression in MA782/5s-8101-R cells. (D) Immunofluorescence staining of p-mTOR expression in BC MA782/5s-8101-R cells. Scale bars=25 μm (B-D). Each experiment was repeated three times; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 6 Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) inhibit cell proliferation by modulating the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin pathway and reshaping the immune microenvironment in breast cancer programmed cell death-1–resistant cells. (A) EdU immunofluorescence staining of proliferation in MA782/5s-8101-R cells. (B) Wound healing assay measuring the wound closure speed in MA782/5s-8101-R cells. (C) Transwell assay of migration in MA782/5s-8101-R cells. (D) Transwell assay of invasion in MA782/5s-8101-R cells. Scale bars=50 μm (A, C, D), 100 μm (B). (E) Flow cytometry analysis of apoptosis in MA782/5s-8101-R cells. Each experiment was repeated three times; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 7 Effects of salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) on the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling pathway. (A) Western blot analysis of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR expression in mouse tumor tissues. (B) Immunofluorescence staining showing p-PI3K expression in mouse tumor tissues. GAPDH, glyceraldehyde 3-phosphate dehydrogenase. (C) Immunofluorescence staining showing p-AKT expression in mouse tumor tissues. (D) Immunofluorescence staining showing p-mTOR expression in breast cancer mouse tumor tissues. Scale bars=50 μm (B-D). Each group consisted of six mice; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 8 Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) inhibit tumor proliferation by modulating the phosphoinositide 3-kinase/AKT/mammalian target of rapamycin pathway and reshaping the immune microenvironment in programmed cell death-1–resistant mice. (A) In vivo imaging of tumor growth in different groups. (B) Images of tumors and measurements of tumor volume and weight in different groups. (C) H&E staining of tumor tissue morphology and structure in different groups. (D) TUNEL staining of cell apoptosis in tumor tissues of different groups. (E) Immunohistochemistry of Ki67 expression in tumor tissues of different groups. Scale bars=50 μm (C-E). Each group consisted of six mice; values are presented as mean±standard deviation, ***p < 0.001.

Fig. 9 Salidroside-loaded, oligopeptide-modified tumor exosomes (Salidroside@T-exo) induce reversal of acquired resistance in immunotherapy for breast cancer through modulation of the phosphoinositide 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathway and reshaping the immune microenvironment. PD-1, programmed cell death-1.

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Salidroside-Loaded, TMTP1-Modified Cancer Stem Cell–Derived Exosomes Reprogram the PI3K/AKT/mTOR Axis to Overcome PD-1 Resistance in Breast Cancer