Autophagy Inhibition by Fangji Huangqi Decoction Enhances Tumor Immunogenicity and Potentiates PD‐1 Blockade

ABSTRACT

Fangji Huangqi Decoction (FJHQ), a traditional Chinese medicine formula derived from “Jin Gui Yao Lue,” has been reported to exhibit immunomodulatory and anti‐inflammatory effects. This study investigates its potential anti‐tumor mechanism via the enhancement of CD8⁺ T cell‐mediated immunity. In murine B16 melanoma and Lewis lung carcinoma models, FJHQ significantly suppressed tumor growth and enhanced CD8⁺ T cell infiltration and cytotoxicity. CD8⁺ T cell depletion abrogated these effects, confirming their necessity. In vitro, FJHQ treatment augmented antigen presentation by tumor cells through upregulation of MHC‐I expression, leading to improved CD8⁺ T cell recognition and killing. Mechanistically, FJHQ inhibited autophagic flux in tumor cells, as evidenced by mCherry‐GFP‐LC3 reporter assays and electron microscopy, resulting in impaired lysosomal function and elevated MHC‐I levels. Furthermore, combining FJHQ with anti–PD‐1 therapy synergistically promoted CD8⁺ T cell activity, increased granzyme B and IFN‐γ expression, and induced substantial tumor regression. These findings demonstrate that FJHQ enhances antitumor immunity by suppressing autophagy and elevating MHC‐I expression, thereby sensitizing tumors to PD‐1 blockade. FJHQ represents a promising immunoadjuvant for cancer therapy.

In tumor cells, MHC‐I is typically degraded through autophagy, leading to reduced antigen presentation. However, Fangji Huangqi Decoction (FJHQ), acting as an autophagy inhibitor, suppresses lysosomal acidification, which in turn inhibits the degradation of MHC‐I. This results in increased surface expression of MHC‐I on tumor cells. In same time, FJHQ combined with PD‐1 blockade enhances CD8⁺ T cell–mediated anti‐tumor immunity in vivo. Consequently, CD8⁺ T cells are activated, enhancing their immune recognition of tumor cells and strengthening the anti‐tumor immune response.

Abbreviations

Baf

Bafilomycin A1

CatB

Cathepsin B

CatD

Cathepsin D

CTLs

Cytotoxic T Lymphocytes

DOX

Doxycycline

FJHQ

Fangji Huangqi Decoction

GZMB

Granzyme B

IFN‐γ

Interferon‐gamma

LC3

Microtubule‐Associated Protein 1 Light Chain 3

MHC‐I

Major Histocompatibility Complex Class I

OVA

Ovalbumin

p62

Sequestosome 1 (SQSTM1)

TOF‐MS

Time‐of‐Flight Mass Spectrometry

UPLC/Q‐TOF‐MS/MS

Ultra‐Performance Liquid Chromatography/Quadrupole‐Time of Flight‐Mass Spectrometry/Mass Spectrometry

1. Introduction

In cancer immunotherapy, tumor immune escape mediated by impaired antigen presentation and immunosuppressive microenvironments remains a major obstacle [1, 2]. MHC class I molecules are essential for presenting tumor antigens to CD8⁺ cytotoxic T lymphocytes (CTLs), and their downregulation allows tumors to evade immune surveillance [3]. This suppression, combined with inhibitory factors such as Tregs, MDSCs, and immune checkpoints, limits the efficacy of current immunotherapies [4, 5, 6, 7]. Enhancing antigen presentation and restoring CTL function are therefore critical for improving treatment outcomes.

As the principal effectors of antitumor immunity, CD8⁺ CTLs identify and eliminate malignant cells upon T cell receptor (TCR) recognition of tumor‐derived peptides presented by MHC‐I [8]. Upon activation, CTLs release cytotoxic mediators, including perforin and granzyme B (GZMB), to induce apoptosis in target cells [9]. However, within the tumor microenvironment, CTL function is often attenuated through multiple mechanisms, such as immune checkpoint activation, defects in antigen processing and presentation, and overall functional exhaustion [10, 11]. Thus, strategies aimed at enhancing antigen presentation and sustaining CD8⁺ T cell activation are vital for improving the efficacy of cancer immunotherapies.

Although immune checkpoint blockade (ICB), particularly anti‐PD‐1/PD‐L1 therapy, has shown clinical success, many patients exhibit resistance due to poor T cell infiltration, defective antigen presentation, or tumor‐intrinsic resistance mechanisms [12]. Recent evidence suggests that augmenting tumor immunogenicity and T cell priming may help overcome ICB resistance [13]. Thus, combination strategies that enhance MHC‐I expression and promote CTL activity are urgently needed.

Autophagy, a conserved cellular degradation process, plays a dual role in tumorigenesis and antitumor immunity [14, 15]. The role of autophagy in cancer is context‐dependent: while it exhibits a protective capacity against tumor initiation by preserving cellular homeostasis [16], it frequently promotes immune suppression in established tumors by interfering with antigen presentation via MHC‐I [17, 18]. Inhibiting autophagy has been shown to increase surface MHC‐I levels, suggesting a potential strategy to improve CTL recognition and killing of tumor cells.

Fangji Huangqi Decoction (FJHQ), a classical traditional Chinese medicine formula from “Jin Gui Yao Lue,” is composed of six herbal components and has reported immunomodulatory and anti‐inflammatory properties [19, 20, 21]. While FJHQ has shown potential in cancer therapy, its mechanisms—particularly concerning antigen presentation and adaptive immunity—remain unclear.

This study investigates whether FJHQ enhances CD8⁺ T cell–mediated antitumor immunity by inhibiting autophagy and upregulating MHC‐I expression. Using in vitro and in vivo models, we examined the effects of FJHQ on autophagic flux, antigen presentation, and CTL cytotoxicity, as well as its synergy with anti‐PD‐1 treatment. Our findings reveal FJHQ as a promising immunoadjuvant that can strengthen tumor‐specific immunity and improve response to checkpoint inhibition.

2. Materials and Methods

2.1. Preparation and Composition Analysis of Fangji Huqi Decoction

FJHQ was prepared by decocting six authenticated herbs (4:5:3:2:3:1 ratio) in water, followed by concentration and lyophilization. Chemical characterization was conducted using UPLC/Q‐TOF‐MS/MS with a C18 column and a formic acid/acetonitrile‐water gradient system. Mass spectrometry was performed on an AB SCIEX Triple TOF^TM 5600+ instrument under optimized parameters, as previously reported [21].

2.2. Preparation of FJHQ Fractions

FJHQ lyophilized powder (25 g) was dissolved in 100 mL pure water and extracted with ethyl acetate. The ethyl acetate fraction was evaporated to yield the ER fraction (1.0 g, 4%). The aqueous phase was concentrated to 85 mL; 15 mL was lyophilized to obtain the degreased aqueous extract (4 g, 91% yield based on theoretical 4.4 g). The remaining 70 mL was subjected to ethanol precipitation, yielding the supernatant (13.6 g, 66.3%) and precipitate (6.4 g, 31.2%) fractions. Each fraction was dissolved in DMEM (or DMSO + DMEM for ER) to prepare stock solutions (ER: 8 mg/mL, supernatant: 130 mg/mL, precipitate: 62 mg/mL, degreased aqueous: 180 mg/mL), filtered (0.22 μm), and stored at −80°C. Working concentrations were adjusted to correspond to 800 μg/mL of crude FJHQ.

2.3. Reagents and Antibodies

Chemical inhibitors including bafilomycin A1 (Selleckchem S1413), chloroquine (Selleckchem C6628), Anti‐mouse PD‐1 (CD279)‐InVivo (Selleckchem A2122) and CD8α neutralizing antibody (Selleckchem A2102) were commercially obtained. Primary antibodies against β‐actin, LC3‐I/II, p62, Cathepsin B/D (Cell Signaling Technology), and MHC class I (ABclonal HLA‐A/B; Santa Cruz H‐2K^b) were used with species‐specific HRP‐secondary antibodies (ABclonal). The pSLIK_Hyg_mTurquoise2_Atg4b^C74A plasmid was provided by Dr. Alec Kimmelman (Addgene 161733), while hLAMP1‐mCherry (VectorBuilder) and pcDNA3‐OVA (Miao Ling Plasmid) were utilized. CD8⁺ T cells were isolated using Miltenyi Biotec kits (130‐117‐044).

2.4. Cell Culture

The following cell lines were utilized: B16‐F10 melanoma (ATCC), NCI‐H1299 and A549 lung adenocarcinoma (ATCC), A375 melanoma (Procell), and LLC lung carcinoma (GZUCM). The LLC‐OVA variant was generated via pcDNA3‐OVA transfection. Stable H2‐K1 knockdown B16, LLC, B16‐OVA and LLC‐OVA cells were generated by lentiviral transduction of pLV[shRNA]‐Puro‐U6 vectors expressing the following shRNA sequences (#1: ACAGCAGACCTGAAGATAAAG; #2: AGGGCTGATTTCTTGTTAATT; #3: CACTTCCACACTGAGAATAAT). Transduced cells were selected with puromycin (5 μg/mL) and maintained as stable pools or clones. Cells were cultured in complete media (RPMI‐1640/DMEM with 10% FBS and 1% penicillin–streptomycin) under standard conditions (37°C, 5% CO₂). Medium was refreshed every 2–3 days, and subculturing was performed at 70%–80% confluency using trypsin–EDTA. Prior to treatment, cells were seeded at appropriate densities and given 24 h for attachment.

2.5. Flow Cytometry Analysis

Following FJHQ treatment, cells were collected, washed with cold PBS, and stained with FITC‐conjugated H‐2K^b antibody (BD 553569) and OVA peptide–MHC complexes (BioLegend 141605) for 20 min at room temperature in darkness. After washing to remove unbound probes, cells were resuspended in FACS buffer (PBS with 2% serum) and analyzed using a BD Fortessa flow cytometer. Data processing was performed with FlowJo software.

2.6. Western Blot Analysis

Post‐treatment cellular lysates were prepared in RIPA buffer containing enzymatic inhibitors, with protein concentrations measured by the BCA method. Proteins were electrophoretically separated and transferred to PVDF membranes. Following 1‐h blocking with 5% non‐fat dried milk in TBST (room temperature), membranes were incubated with primary antibodies overnight at 4°C. After washing, HRP‐conjugated secondary antibodies were applied (1 h, RT). Immunoreactive bands were visualized by chemiluminescence (Tanon 5200CE) and quantified using GelPro analysis tools.

2.7. mCherry‐GFP‐LC3 Autophagy Flux Assay

The pBABE‐puro mCherry‐GFP‐LC3B plasmid (Addgene) was introduced into cancer cells to evaluate FJHQ's effects on autophagy progression. After pharmacological intervention, cells were processed through PBS rinsing and 4% paraformaldehyde fixation (RT, 15 min). Confocal imaging (Zeiss LSM 800) allowed visualization of autophagy intermediates: yellow vesicles (GFP and mCherry co‐positive) representing autophagosomes and red structures (mCherry single‐positive) indicating autolysosomes. Quantification of yellow puncta serves as an indicator of autophagic flux blockade.

2.8. Immunofluorescence Detection of Autophagosome‐Lysosome Fusion

A375 and NCI‐H1299 cell lines stably expressing mCherry‐tagged LAMP1 were established for autophagy studies. Cells were cultured on coverslips (12‐well plates) for 24 h before experimental treatments. Post‐incubation, samples were fixed (4% PFA, 15 min), permeabilized (0.3% Triton X‐100), and blocked (10% FBS). Immunostaining protocol included: LC3 primary antibody (overnight, 4°C), Alexa Fluor 488‐secondary antibody (1 h, dark), and DAPI. Confocal imaging revealed: red fluorescence (mCherry‐LAMP1, lysosomes), green fluorescence (Alexa Fluor 488‐LC3, autophagosomes), and yellow signals (colocalization) indicating autophagic flux progression.

2.9. Transmission Electron Microscopy (TEM) Analysis of Autophagosome‐Lysosome Fusion

For TEM analysis of autophagic structures, A375 cells treated with FJHQ were fixed with 2.5% glutaraldehyde and 1% OsO4, followed by ethanol dehydration and resin embedding. Ultrathin sections (70–90 nm) were examined to identify autophagic vacuoles: early autophagosomes (double‐membrane structures containing organelles) and late autolysosomes (single‐membrane compartments with electron‐dense contents).

2.10. Crystal Violet Cytotoxicity Assay

CD8⁺ T cell‐mediated cytotoxicity against FJHQ‐pretreated LLC‐OVA cells was assessed by crystal violet staining. After co‐culture, adherent cells were fixed with 4% PFA (15 min), stained with 0.5% crystal violet (20 min), washed, and solubilized in 1% acetic acid. Absorbance at 595 nm was measured, with higher values indicating reduced cytotoxicity.

2.11. ELISA Detection of IFN‐γ Release

IFN‐γ levels in co‐culture supernatants were quantified using an ELISA kit (ABclonal). Co‐culture supernatants from FJHQ‐treated groups were added to antibody‐coated wells. After incubation and washing, detection involved a biotinylated detection antibody and streptavidin‐HRP. Following substrate reaction termination, absorbance was measured. Reference samples included negative controls and culture media controls.

2.12. Flow Cytometry Analysis of CD8 ⁺ T Cell‐Mediated Cytotoxicity and T Cell Activation

OVA‐specific CD8⁺ T cells were obtained from OT‐I mice (GeneEasy Biotech). Splenocyte suspensions were prepared, and CD8⁺ populations were enriched via magnetic separation (Miltenyi Biotec microbeads) after incubation in MACS buffer (4°C, 10 min). Isolated cells were expanded in IL‐2‐containing medium.

To evaluate cytotoxic activity, CD8⁺ T cells were co‐cultured with FJHQ‐conditioned LLC‐OVA targets. Cell mixtures were subsequently stained with viability marker FVS620 and CD8‐specific antibody for flow cytometry, with dead tumor cells identified as CD8⁻/FVS620⁺ events.

For activation studies, co‐cultured cells were stained with CD8 and CD69 antibodies. T cell stimulation was assessed through CD69 expression levels (MFI) on CD8⁺ lymphocytes.

2.13. Detection of CD8 ⁺ T Cell‐Mediated Cytotoxicity by Propidium Iodide (PI) Staining

To investigate FJHQ‐enhanced T cell killing, LLC‐OVA cells were cultured in confocal dishes and treated with FJHQ (24 h) before OT‐I T cell addition (24 h co‐culture). After PBS washes to eliminate non‐adherent material, cells were sequentially stained with PI (10 min, dark) and DAPI. Confocal microscopy (Zeiss LSM 800) visualized apoptotic cells (red PI staining) and total nuclei (blue DAPI). Quantitative analysis of PI‐positive cells across treatment groups evaluated FJHQ's ability to potentiate T cell‐induced tumor cell death.

2.14. Assessment of Lysosomal Acidification Using LysoTracker

A375, NCI‐H1299 and A549 cell lines were cultured in confocal dishes and exposed to experimental treatments for 24 h. To monitor lysosomal pH, cells were subsequently labeled with LysoTracker Red (Thermo Fisher Scientific, L7528; 50 nM) for 20 min under physiological conditions (37°C, dark). This fluorescent probe specifically targets and accumulates in acidic organelles. Following PBS washing to remove excess dye, intracellular fluorescence patterns were analyzed by confocal microscopy to evaluate lysosomal acidification levels.

2.15. In Vivo Model

Eight‐week‐old male C57BL/6 mice (Guangdong Medical Laboratory Animal Center) were housed under SPF conditions. Following randomization (n = 8/group), tumor xenografts were established via subcutaneous inoculation of B16, B16‐shH2‐K1 (2 × 10⁵) LLC or LLC‐shH2‐K1 (2 × 10⁶) cells. Treatment groups received either pure water (control) or FJHQ (3.51 or 7.02 g/kg) via daily oral gavage. Tumor size was monitored by caliper measurements, with volumes computed as (length × width²)/2. CD8⁺ T cell depletion studies involved four experimental arms: control, αCD8 alone, FJHQ alone, and combination therapy. Anti‐CD8 antibody (200 μg) was administered intraperitoneally every 3 days. Combination therapy with PD‐1 blockade consisted of daily FJHQ (7.02 g/kg) plus intermittent anti‐PD‐1 mAb (200 μg, every 72 h). Treatment outcomes were evaluated through tumor growth curves, endpoint weights, and histological assessments. These animal experiments were performed in compliance with ethical standards and approved by the Guangzhou University of Chinese Medicine Animal Ethics Committee (20240419004).

2.16. Immunofluorescence Staining

Resected tumors underwent fixation (4% PFA), paraffin embedding, and sectioning (4 μm). After deparaffinization and graded alcohol rehydration, heat‐induced epitope retrieval was conducted in citrate buffer (pH 6.0). Endogenous peroxidase blockade was achieved with 3% hydrogen peroxide. Sections were blocked with 5% BSA before overnight incubation (4°C) with primary antibody combinations: CD8/Granzyme B or CD8/IFN‐γ (all 1:200). Following PBS washes, species‐matched fluorescent secondary antibodies (Alexa Fluor 594 and 488 conjugates) were added (1 h, dark). DAPI counterstaining preceded confocal microscopy (Zeiss LSM 800) imaging.

2.17. Flow Cytometry Analysis of Tumor‐Infiltrating Lymphocytes

Tumor tissues were harvested, minced, and digested in DMEM containing 5% FBS, 2 mg/mL collagenase D, 1 mg/mL collagenase IV, and 40 μg/mL DNase I at 37°C for 30 min with shaking. Cell suspensions were filtered and subjected to density gradient centrifugation using 40% and 70% Percoll. The lymphocyte layer at the interface was collected, washed, and stimulated with cell activation reagent (BioLegend 423303) for 6 h. Cells were stained with Fixable Viability Stain 620, followed by Fc block and surface staining with antibodies against CD45, CD3e, and CD8a. After fixation and permeabilization with Cyto‐Fast Fix/Perm solution, intracellular staining for GZMB and IFN‐γ was performed. Samples were acquired on a flow cytometer and analyzed using FMO controls. Antibody information is provided in Table S1, which lists the antibodies used for IHC and flow cytometry.

2.18. RNA Extraction and Quantitative Real‐Time PCR

Total RNA was isolated from tumor tissues using TRIzol reagent according to the manufacturer's instructions. RNA concentration and purity were measured with a NanoDrop spectrophotometer. Reverse transcription was performed using AG reverse transcription kit in a 10 μL system (AG AG11706). Quantitative real‐time PCR was conducted using SYBR Green Premix (ROX) on a system. The following genes were analyzed: Actb, H2‐K1, Ifng, Ccl5, Cxcl9, Cxcl10, Cxcl11, Tnf, Il12a, and Il12b. Relative expression was calculated by the 2^−ΔΔCt method. All samples were run in quadruplicate. Primer information is provided in Table S2, which lists the primer sequences.

2.19. Statistical Analysis

Data were analyzed using GraphPad Prism and are presented as mean ± SD from ≥ 3 independent experiments. Group comparisons used Student's t‐test (two groups) or one‐way ANOVA with Tukey's test (multiple groups). Statistical significance was defined as p < 0.05. Specific tests are indicated in the figure legends.

3. Results

3.1. Analysis of the Main Components in Fangji Huangqi Decoction Using UPLC/Q‐TOF‐MS/MS

The major constituents of FJHQ have been previously identified in our laboratory using UPLC‐Q‐TOF‐MS/MS, including Laurolitsine, Nicotiflorin, Scopoletin, Liquiritigenin, Fangchinoline, Tetrandrine, Licochalcone B, Formononetin, Ononin, Glycyrrhizic acid, 18β‐Glycyrrhetinic acid, Atractylenolide I, Atractylenolide III, Gancaonin T, and Gancaonin R [21]. Activity‐guided fractionation was subsequently performed to pinpoint the bioactive fractions, as detailed in Materials and Methods, and the corresponding validation data are presented in later sections (Figures S4F,G and S6C).

3.2. FJHQ Inhibits Tumor Growth Through CD8 ⁺ T Cell–Mediated Immunity

FJHQ's antitumor efficacy was evaluated in B16 melanoma and Lewis lung carcinoma (LLC) murine models (Figure 1A,E). Mice were subcutaneously inoculated with 2 × 10⁵ (B16) or 2 × 10⁶ (LLC) tumor cells and received daily oral FJHQ starting on day 2. Treatment continued for 14 days (B16) or 17 days (LLC), with tumor growth monitored regularly. Representative tumor images visually demonstrated reduced tumor sizes in FJHQ‐treated groups (Figure 1B,F). FJHQ significantly suppressed tumor progression in both models, as evidenced by reduced tumor volumes (Figure 1C,G) and final weights (Figure 1D,H).

FIGURE 1.

FJHQ inhibits tumor growth through CD8⁺ T cell‐mediated immunity. (A) Schematic of the B16 melanoma model and FJHQ dosing regimen. (B–D) Representative tumor images (B), growth curves (C), and weights (D) in B16‐bearing mice. (E) Schematic of the LLC model. (F–H) Tumor images (F), growth curves (G), and weights (H) in LLC‐bearing mice. (I–L) Flow cytometric analysis of CD8⁺ T cells isolated from B16 (I, K) and LLC (J, L) tumors showing the frequencies of GZMB⁺ and IFN‐γ⁺ populations. (M–P) Relative mRNA levels of Ccl5, Cxcl9, Cxcl10, Cxcl11, Ifng, Tnf, Il12a, and Il12b in tumors from FJHQ‐treated mice compared to controls, as determined by qPCR. (Q) Schematic of CD8⁺ T cell depletion protocol. (R–T) Tumor images (R), growth curves (S), and weights (T) following CD8 depletion. Data are mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001.

Flow cytometric analysis of tumor‐infiltrating lymphocytes (TILs) demonstrated that FJHQ treatment significantly increased the percentages of IFN‐γ⁺ and Granzyme B⁺ CD8⁺ T cells in both LLC and B16 models (Figures 1I–L and S1A), indicating enhanced effector function. Immunofluorescence staining further confirmed increased CD8⁺ T cell infiltration and co‐localization with GZMB and IFN‐γ (Figure S1D–M). Furthermore, FJHQ significantly increased the mRNA expression levels of CCL5, CXCL9, CXCL10, CXCL11, IFNG, TNF, IL12a, and IL12b (Figure 1M–P). To confirm the functional contribution of CD8⁺ T cells, depletion experiments were performed using neutralizing antibodies (Figure 1Q). Depletion efficiency was verified by flow cytometric analysis of CD8⁺ T cells in the spleen, which confirmed > 95% reduction in CD8⁺ populations compared to isotype control‐treated mice (Figure S1B,C). CD8⁺ T cell depletion abrogated FJHQ's antitumor effects, resulting in increased tumor burden (Figure 1R), accelerated growth (Figure 1S), and higher tumor weights (Figure 1T).

These results demonstrate that FJHQ inhibits tumor growth in both models through a CD8⁺ T cell‐dependent mechanism, highlighting its potential as an immunomodulatory agent for cancer therapy.

3.3. FJHQ Enhances CD8 ⁺ T Cell‐Mediated Tumor Cell Killing In Vitro

To evaluate FJHQ's potential in enhancing CD8⁺ T cell‐mediated cytotoxicity, we established an in vitro co‐culture system using OT‐I mouse‐derived CD8⁺ T cells and FJHQ‐pretreated LLC‐OVA tumor cells (Figure 2A). Prior control experiments confirmed that FJHQ exerted no direct cytotoxic or pro‐apoptotic effects on A375, H1299, B16, or LLC cells after 24 or 48 h of treatment, as determined by CCK‐8 and annexin V/PI assays (Figure S2). Flow cytometric analysis revealed that FJHQ significantly enhanced CD8⁺ T cell activation, as demonstrated by upregulated CD69 expression (Figure 2B). This activation was accompanied by increased IFN‐γ secretion into the co‐culture medium, measured by ELISA (Figure 2C). Crystal violet staining showed substantially reduced viability of FJHQ‐pretreated tumor cells after co‐culture (Figure 2D). Apoptosis analysis via Annexin V/PI staining confirmed enhanced tumor cell death following FJHQ pretreatment (Figure 2E). Immunofluorescence microscopy further supported these findings, showing increased PI‐positive tumor cells in FJHQ‐pretreated groups (Figure 2F). Collectively, these results demonstrate that FJHQ sensitizes tumor cells to CD8⁺ T cell‐mediated killing by enhancing their susceptibility to T cell‐induced apoptosis.

FIGURE 2.

FJHQ enhances CD8⁺ T cell‐mediated tumor cell killing in vitro. (A) Experimental scheme: CD8⁺ T cells isolated from OT‐I mice were co‐cultured with FJHQ‐pretreated LLC‐OVA cells at 10:1 E:T ratio. (B) CD69 expression on CD8⁺ T cells by flow cytometry. (C) IFN‐γ secretion measured by ELISA. (D) Tumor cell viability assessed by crystal violet staining. (E) Tumor cell apoptosis analyzed by Annexin V/PI staining. (F) Cell death detection by PI staining. Scale bar: 100 μm. Data are mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001.

3.4. FJHQ Combined With PD‐1 Blockade Enhances CD8 ⁺ T Cell–Mediated Anti‐Tumor Immunity

As a key immune checkpoint inhibitor, programmed death‐1 (PD‐1) monoclonal antibody enhances antitumor immunity by blocking PD‐1/PD‐L1/PD‐L2 signaling. Based on FJHQ's ability to enhance cytotoxic T lymphocyte activity, we investigated its cooperative potential with PD‐1 blockade. Using an in vitro co‐culture approach, IL‐2‐stimulated CD8⁺ T cells were incubated with FJHQ‐pretreated B16‐OVA tumor cells in the presence or absence of anti‐PD‐1 antibody (Figure 3A). The combination treatment significantly enhanced IFN‐γ production (Figure 3B) and increased tumor cell apoptosis (Figure 3C) compared to monotherapies.

FIGURE 3.

FJHQ combined with PD‐1 blockade enhances CD8⁺ T cell‐mediated anti‐tumor immunity. (A) Schematic of the in vitro cytotoxicity assay: CD8⁺ T cells were isolated and purified from the spleens of OT‐I mice using magnetic bead sorting and cultured in the presence of IL‐2 (15 ng/mL). CD8⁺ T cells were pretreated with anti–PD‐1 antibody (clone [RMP1‐14], 2 μg/mL) for 24 h and co‐cultured with B16‐OVA tumor cells that had been pretreated with FJHQ (800 μg/mL) for 24 h at an effector‐to‐target (E:T) ratio of 10:1. (B) IFN‐γ secretion (ELISA). (C) Tumor cell apoptosis (Annexin V/PI staining). (D, H) Treatment schedules for B16 and LLC models combining oral FJHQ with intraperitoneal anti‐PD‐1. (E–G, I–K) Tumor images, growth curves, and weights from B16 (E–G) and LLC (I–K) models. (L, M) Immunofluorescence of CD8⁺ T cell activation (GZMB/IFN‐γ) in B16 tumors. Scale bar: 20 μm. Data are mean ± SD, **p < 0.01, ***p < 0.001.

These findings were validated in B16 melanoma and LLC models, where the FJHQ plus anti‐PD‐1 combination synergistically suppressed tumor progression, reducing tumor volumes and weights (Figure 3D–K). Immunofluorescence analysis revealed enhanced CD8⁺ T cell infiltration and increased co‐localization with GZMB and IFN‐γ in tumors receiving combination therapy (Figures 3L,M and S3). Collectively, FJHQ synergizes with PD‐1 blockade to enhance CD8⁺ T cell‐mediated antitumor immunity, representing a promising combination strategy for cancer immunotherapy.

3.5. FJHQ Enhances MHC‐I Antigen Presentation in Tumor Cells

To investigate whether FJHQ enhances tumor antigen presentation, we first measured major histocompatibility complex class I (MHC‐I, H‐2K^b) levels in B16 melanoma and Lewis lung carcinoma (LLC) tissues from FJHQ‐treated mice. Western blot analysis revealed significantly elevated MHC‐I protein in treated tumors (Figure 4A). In vitro, FJHQ dose‐dependently increased MHC‐I expression across A375, B16, NCI‐H1299 LLC, and A549 cell lines, with maximal effect at higher concentrations (Figures 4B and S4A). Sustained FJHQ exposure induced a time‐dependent upregulation of MHC‐I surface expression in A375, A549, and NCI‐H1299 cells (Figures 4C and S4B).

FIGURE 4.

FJHQ enhances MHC‐I antigen presentation in tumor cells. (A) MHC‐I (H‐2K^b) expression in B16 and LLC tumor tissues by Western blot. (B, C) Dose‐ and time‐dependent increase of MHC‐I in A375, B16, NCI‐H1299, and LLC cells after FJHQ treatment (200–800 μg/mL, 0–24 h); bafilomycin A1 (Baf, 15 nM) served as positive control. (D) Immunofluorescence analysis of MHC‐I expression in tumor cells. Scale bar: 5 μm. (E–G) Flow cytometry analysis of MHC‐I surface expression in tumor cells following FJHQ treatment. (H–J) Increased H‐2K^b/SIINFEKL complex expression in B16 LLC, and LLC‐OVA cells measured by flow cytometry. Data are mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001.

Immunofluorescence analysis confirmed that FJHQ increased MHC molecule expression in both in vitro and in vivo settings (Figures 4D and S4D,E). Flow cytometry confirmed these results, demonstrating a dose‐dependent increase in surface MHC‐I in A375, NCI‐H1299 (Figures 4E and S4C), B16, and LLC cells (Figure 4F,G). Critically, FJHQ enhanced antigen presentation via MHC‐I. Quantitative analysis showed increased surface density of H‐2K^b/SIINFEKL complexes on peptide‐pulsed B16 and LLC cells after FJHQ treatment (Figure 4H,I). Similarly, in LLC‐OVA cells expressing endogenous ovalbumin, FJHQ significantly boosted the presentation of H‐2K^b‐bound OVA epitopes (Figure 4J). To identify the active fraction responsible for MHC‐I upregulation LLC cells treated with equimolar amounts of various extracts revealed that ethyl acetate fraction (EA) and defatted water‐soluble extract (DW), like FJHQ crude extract, significantly increased surface MHC‐I and H‐2K^b/SIINFEKL complexes, whereas ethanol precipitation precipitate (EP, macromolecules) showed modest or no effect and EA was inactive, indicating that the immunomodulatory activity of FJHQ is predominantly enriched in the ethanol precipitation supernatant (ES, small molecules) (Figure S4F,G).

These findings demonstrate that FJHQ upregulates MHC‐I molecule expression and promotes the presentation of tumor‐specific antigens, thereby facilitating more efficient recognition of malignant cells by cytotoxic T lymphocytes.

3.6. MHC‐I Deficiency Abolishes FJHQ‐Enhanced CD8 ⁺ T Cell Cytotoxicity and Antitumor Efficacy

To directly test whether MHC‐I upregulation is required for FJHQ‐enhanced CD8⁺ T cell killing, we stably knocked down H2‐K1, the gene encoding the mouse MHC‐I heavy chain, in B16 LLC, B16‐OVA, and LLC‐OVA cells using shRNA. Among three independent shRNA sequences, the shH2‐K1 #2 achieved the highest knockdown efficiency as confirmed by qPCR and was selected for subsequent experiments (Figure 5A,B).

FIGURE 5.

MHC‐I deficiency abolishes FJHQ‐enhanced CD8⁺ T cell cytotoxicity and antitumor efficacy. (A, B) qPCR validation of H2‐K1 knockdown in B16 (A) and LLC (B) cells. shH2‐K1 #2 was used for subsequent studies. (C, D) IFN‐γ secretion (ELISA) from WT or shH2‐K1 B16‐OVA (C) and LLC‐OVA (D) cells pretreated with FJHQ and co‐cultured with OT‐1 T cells. (E, F) Apoptosis rates of the same co‐cultures by annexin V staining. (G, J) Representative tumor images. (H, K) Tumor growth curves. (I, L) Tumor weights. Data are mean ± SD. ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

In vitro, wild‐type (WT) or H2‐K1‐knockdown (shH2‐K1) tumor cells were pretreated with FJHQ and co‐cultured with OT‐1 CD8⁺ T cells. FJHQ significantly increased IFN‐γ secretion and tumor cell apoptosis in WT cells. In contrast, these enhancing effects were completely abrogated in MHC‐I‐deficient cells, where IFN‐γ levels and apoptosis rates remained comparable to those of untreated controls (Figure 5C–F).

To corroborate these findings in vivo, WT or H2‐K1‐knockdown B16 and LLC cells were implanted subcutaneously into C57BL/6 mice, followed by FJHQ treatment. FJHQ potently suppressed the growth of WT tumors and sustained this inhibition throughout the observation period. However, its antitumor efficacy was largely lost in MHC‐I‐knockdown tumors; the growth curves of FJHQ‐treated shH2‐K1 tumors nearly overlapped with those of vehicle controls, with only minimal, non‐specific suppression (Figure 5G–L).

Collectively, these results demonstrate that MHC‐I expression on tumor cells is indispensable for FJHQ‐mediated enhancement of CD8⁺ T cell cytotoxicity and antitumor activity.

3.7. FJHQ Increases MHC‐I Levels on Tumor Cells by Inhibiting Autophagy

To determine if autophagy inhibition underlies MHC‐I upregulation, we utilized B16‐Atg4b^C74A and LLC‐Atg4b^C74A cells expressing a dominant‐negative Atg4b mutant. Successful transgene induction was confirmed by doxycycline‐dependent increase in mTurquoise2 fluorescence using flow cytometry (Figure S5). Immunoblot analysis further demonstrated that doxycycline treatment led to LC3‐II accumulation and increased p62 levels, confirming effective autophagy inhibition in both cell lines (Figure 6A,E). Doxycycline (DOX)‐induced Atg4b^C74A expression elevated MHC‐I (H‐2K^b) levels in a concentration‐dependent manner (Figure 6A). Flow cytometry confirmed increased surface H‐2K^b in DOX‐induced B16‐Atg4b^C74A cells (Figure 6B). In LLC‐Atg4b^C74A cells, DOX enhanced H‐2K^b/SIINFEKL complex presentation (Figure 6C), a finding replicated in LLC‐OVA‐Atg4b^C74A cells with endogenous OVA expression (Figure 6D).

FIGURE 6.

FJHQ increases surface MHC‐I expression in tumor cells by inhibiting autophagy. (A) Western blot analysis of p62, LC3‐II, and MHC‐I (H‐2K^b) expression in DOX‐induced LLC‐Atg4b^C74A and B16‐Atg4b^C74A cells expressing dominant‐negative Atg4b mutant. (B) Flow cytometry of surface H‐2K^b expression in B16‐Atg4b^C74A cells after DOX treatment. (C, D) H‐2K^b/SIINFEKL complex expression in LLC‐Atg4b^C74A and LLC‐OVA‐Atg4b^C74A cells measured by flow cytometry after DOX induction. (E, F) MHC‐I expression and antigen presentation in LLC‐Atg4b^C74A and LLC‐OVA‐Atg4b^C74A cells treated with FJHQ (800 μg/mL), DOX (8 μg/mL), or combination for 24 h, analyzed by Western blot and flow cytometry. Data are mean ± SD. ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

To assess whether FJHQ operates via analogous autophagy‐related pathways, we compared its effects with DOX induction. Both FJHQ and DOX increased MHC‐I expression, but their combination yielded no synergistic enhancement (Figure 6E). Similarly, flow cytometry revealed that FJHQ and DOX individually enhanced H‐2K^b/SIINFEKL surface display in LLC‐OVA‐Atg4b^C74A cells without additive effects in combination (Figure 6F).

These results demonstrate that genetic inhibition of autophagy elevates MHC‐I expression and antigen presentation, and that FJHQ likely acts through a similar mechanism to enhance tumor antigen presentation capacity.

3.8. FJHQ Significantly Inhibits Late‐Stage Autophagic Flux in Tumor Cells

To evaluate FJHQ's impact on autophagic flux, we first examined GFP‐LC3‐expressing A375 and NCI‐H1299 cells. Fluorescence imaging revealed significantly increased GFP‐LC3 puncta after FJHQ treatment, indicating autophagosome accumulation (Figure 7A). Consistent with this, Western blot analysis of LLC tissues from FJHQ‐treated mice showed elevated p62 and LC3‐II levels, confirming impaired autophagic clearance in vivo (Figure 7B). Dose–response studies in multiple cell lines (A375, B16, A549, NCI‐H1299 LLC) demonstrated concentration‐dependent accumulation of p62 and LC3‐II (Figures 7C and S6A,C). Time‐course experiments further revealed progressive increases in both markers with prolonged FJHQ exposure (Figures 7D and S6B). These results indicate that FJHQ inhibits the terminal phase of autophagy, causing autophagic vesicle accumulation and substrate retention. This impairment of lysosomal degradation may contribute to enhanced tumor immunogenicity.

FIGURE 7.

FJHQ significantly inhibits late‐stage autophagic flux in tumor cells. (A) GFP‐LC3 puncta formation in A375 and NCI‐H1299 cells after FJHQ (800 μg/mL, 24 h) treatment. HBSS and Baf (15 nM) were used as controls. Scale bar: 5 μm. (B) p62 and LC3‐II levels in LLC tumor tissues from FJHQ‐treated mice by Western blot. (C, D) Dose‐ and time‐dependent increases in p62 and LC3‐II in A375, B16, NCI‐H1299, and LLC cells after FJHQ treatment, analyzed by Western blot. Baf (15 nM) served as positive control. Data are mean ± SD, *p < 0.05, ***p < 0.001.

3.9. FJHQ Inhibits Late‐Stage Autophagic Flux by Suppressing Lysosomal Acidification

To mechanistically characterize FJHQ's impact on autophagy, we employed A375, NCI‐H1299 and A549 cells expressing the mCherry‐GFP‐LC3 reporter. FJHQ treatment markedly increased yellow puncta (mCherry⁺GFP⁺), indicating impaired autolysosomal acidification and blocked autophagic flux (Figures 8A and S7A). To determine if this resulted from disrupted autophagosome‐lysosome fusion, we examined LC3‐LAMP1 colocalization in mCherry‐LAMP1‐expressing cells. FJHQ did not reduce their co‐localization (Figure 8B), and transmission electron microscopy revealed accumulated autolysosomes in treated cells (Figure 8C), confirming intact fusion but impaired degradation. We next assessed lysosomal function. LysoTracker staining showed significantly reduced lysosomal acidity following FJHQ treatment (Figures 8D and S7B). Western blot analysis demonstrated dose‐ and time‐dependent decreases in Cathepsin B and D maturation (Figures 8E,F and S7C,D), indicating disrupted hydrolase activation.

FIGURE 8.

FJHQ inhibits late‐stage autophagic flux by suppressing lysosomal acidification in tumor cells. (A) Increased yellow puncta (mCherry⁺GFP⁺) in mCherry‐GFP‐LC3‐expressing A375 and NCI‐H1299 cells after FJHQ (800 μg/mL, 24 h), indicating impaired autophagosome‐lysosome fusion. HBSS and Baf (15 nM) were controls. Scale bar: 10 μm. The Y‐axis value represents the proportion of yellow puncta (i.e., mRFP⁺GFP⁺, representing autolysosomes) to the total puncta (yellow + red + green) in the mRFP‐GFP‐LC3 tandem labeling system. (B) Colocalization analysis of mCherry‐LAMP1 (lysosomes) and GFP‐LC3 (autophagosomes) showing increased yellow signals after FJHQ; CQ (20 μM) was used as control. Fluorescence intensity profiles are shown. Scale bar: 5 μm. (C) Electron micrographs revealing accumulated autolysosomes in FJHQ‐treated A375 cells. Scale bars: 1 μm (upper), 200 nm (lower). (D) Reduced LysoTracker Red staining in FJHQ‐treated cells, indicating decreased lysosomal acidification. Scale bar: 10 μm. (E, F) Dose‐ and time‐dependent suppression of cathepsin B/D maturation by Western blot. Baf (15 nM) served as positive control. Data are mean ± SD: ***p < 0.001.

Collectively, these data establish that FJHQ blocks late‐stage autophagy by impairing lysosomal acidification and subsequent protease activation, ultimately preventing autophagic substrate clearance.

4. Discussion

This study provides compelling evidence that FJHQ enhances antitumor immunity by modulating autophagy and promoting CD8⁺ T cell‐mediated responses. Specifically, FJHQ‐mediated inhibition of autophagic flux in cancer cells increases MHC‐I membrane expression and antigen presentation, thereby augmenting cytotoxic T cell‐mediated immune surveillance (Figure 9). These findings align with emerging research indicating that impaired autophagy strengthens antitumor immunity via MHC‐I upregulation, facilitating more efficient tumor cell clearance [18]. Importantly, this study demonstrates for the first time that the crude extract of FJHQ functions as an immunomodulatory adjuvant capable of sensitizing tumors to immune checkpoint blockade, positioning FJHQ as a promising candidate for overcoming resistance to PD‐1/PD‐L1 inhibitors.

FIGURE 9.

Mechanism by which FJHQ enhances T cell‐mediated anti‐tumor immune responses through autophagy inhibition. In tumor cells, autophagy mediates the degradation of MHC‐I molecules, thereby limiting antigen presentation and facilitating immune evasion. FJHQ counteracts this process by inhibiting autophagic flux, primarily through impairment of lysosomal acidification and cathepsin maturation. This suppression of autophagy prevents MHC‐I degradation, leading to its accumulation on the tumor cell surface and enhancing antigen presentation to CD8⁺ T cells. Furthermore, when combined with PD‐1 blockade, FJHQ synergistically promotes CD8⁺ T cell activation and cytotoxic function, resulting in strengthened antitumor immunity in vivo.

In both B16 melanoma and Lewis lung carcinoma (LLC) models, FJHQ exerted significant antitumor effects that were largely CD8⁺ T cell‐dependent (Figure 1). FJHQ promoted substantial CD8⁺ T cell infiltration into tumors, and flow cytometry confirmed their activated phenotype by detecting GZMB and IFN‐γ (Figure 1I–L). This enhanced T cell functionality aligns with evidence that therapies boosting CD8⁺ T cell activity can significantly improve tumor control and treatment outcomes in immunotherapy [22, 23]. Furthermore, FJHQ significantly increased the mRNA expression levels of CCL5, CXCL9, CXCL10, CXCL11, IFNG, TNF, IL12a, and IL12b. These findings indicate that FJHQ may enhance CD8⁺ T cell infiltration through multiple pathways, including the induction of key chemotactic and pro‐inflammatory mediators (Figure 1M–P). CD8⁺ T cell depletion reversed these effects, confirming their essential role (Figure 1Q–T). In vitro, tumor cells pretreated with FJHQ enhanced CD69 expression and IFN‐γ secretion in co‐cultured CD8⁺ T cells (Figure 2B,C). As a key immunoregulatory cytokine, IFN‐γ augments CTL effector function and tumor immunogenicity [24], indicating that FJHQ enhances both T cell activation and cytotoxicity.

A central finding is that FJHQ significantly increased MHC‐I surface expression on tumor cells (Figure 4). Since malignant cells often downregulate MHC‐I to evade immunity [18], FJHQ‐induced upregulation represents a key immunorestorative mechanism. This was further supported by enhanced H‐2K^b/SIINFEKL presentation across tumor lines (Figure 4B–J), suggesting improved antigen‐specific T cell recognition. These properties support the potential of FJHQ to complement immunotherapies targeting antigen presentation [25, 26].

Mechanistically, MHC‐I deficiency abolishes FJHQ‐enhanced CD8⁺ T cell cytotoxicity and antitumor efficacy (Figure 5). FJHQ elevated MHC‐I levels by inhibiting autophagy (Figure 6). This was evidenced by dose‐ and time‐dependent accumulation of p62 and LC3‐II (Figure 7), indicating blocked autophagosome degradation. FJHQ specifically impaired late‐stage autophagy by disrupting lysosomal acidification and cathepsin maturation (Figure 8). Unlike chloroquine (CQ), which disrupts both lysosomal acidification and autophagosome‐lysosome fusion, FJHQ acts as a more specific “fusion‐compatible” acidification inhibitor, potentially through direct targeting of V‐ATPase activity. This distinct mechanism preserves fusion capability while effectively blocking degradation. Lysosomal dysfunction likely prevents MHC‐I degradation, increasing surface expression [27], consistent with reports that autophagy inhibition stabilizes MHC‐I and enhances immunogenicity [18].

Bioactivity‐guided fractionation revealed that the immunopotentiating effects of FJHQ reside predominantly in the FS and DW fractions (Figures S4F,G and S6C), while comprehensive UPLC/Q‐TOF‐MS/MS profiling has established a phytochemical reference library now being used to isolate the specific active principle(s) from these fractions [21]. Nevertheless, the constituent(s) responsible for the formulation‐level effects on autophagy and MHC‐I upregulation have not yet been conclusively assigned. Because FJHQ is a multi‐herb formulation, the observed activity may reflect contributions from multiple components, which limits mechanistic interpretation at the single‐compound level. We are therefore pursuing systematic fractionation coupled with bioactivity‐guided identification to pinpoint the responsible compound(s). Establishing these active constituents will enable a more precise understanding of the underlying mechanism and support future translation by facilitating development of a well‐characterized lead candidate.

Autophagy exhibits stage‐dependent roles in cancer: suppressing tumor initiation but promoting progression in established tumors [16]. Immunologically, autophagy in cancer cells can reduce MHC‐I presentation, impairing CD8⁺ T cell activation [17, 28]. By inhibiting autophagy, FJHQ counteracts this immune‐evasion mechanism, potentially restoring tumor immunosurveillance.

Furthermore, FJHQ synergized with anti–PD‐1 therapy, enhancing IFN‐γ production, CD8⁺ T cell infiltration, and tumor control in vivo (Figure 3), supporting its use as an immunoadjuvant to checkpoint blockade.

In summary, FJHQ enhances CD8⁺ T cell‐dependent antitumor immunity by suppressing autophagy and increasing MHC‐I‐mediated antigen presentation, thereby synergizing with anti PD‐1 therapy. These findings not only provide the first proof‐of‐concept that FJHQ functions as an immune checkpoint sensitizer but also establish a clear roadmap for deconvoluting its bioactive constituents through integrated fractionation and phytochemical profiling. Given the context‐dependent role of autophagy [16], further studies are warranted to clarify FJHQ's mechanism and therapeutic applicability across tumor types.

Author Contributions

Yuxiao Jiang: investigation, methodology, visualization, writing – original draft. Shiji Zeng: investigation, methodology, visualization. Dingye Wang: investigation. Junbo Jiang: investigation, software, writing – review and editing. Zhenzhen Li: formal analysis. Jingbi Li: formal analysis. Dan Chen: validation. Xu Zhao: validation. Zheng Ying: data curation. Yuehua Deng: data curation. Qianwei Cai: validation. Silin Mai: formal analysis. Jiayi Zheng: software. Qiaohong Yang: supervision. Fang Liu: resources, supervision. Biaoyan Du: resources, supervision. Jianyong Xiao: conceptualization, writing – review and editing. Xiaoyi Liu: conceptualization, funding acquisition. Kun Wang: conceptualization, funding acquisition, project administration, writing – original draft, writing – review and editing.

Funding

This work was supported by the National Natural Science Foundation of China (Grant nos. 82304938, 82574775, and 82104571), the Natural Science Foundation of Guangdong Province (Grant nos. 2025A1515012841, 2025A1515012216, and 2022A1515110282), the Science and Technology Plan Project of Guangzhou, China (Grant nos. 2023A04J2477 and 2024A04J10028), and the Research Special Fund for the High‐Level Hospital Co‐Construction Project between Guangzhou University of Chinese Medicine and Zhongshan Traditional Chinese Medicine Hospital (Grant no. GZYZS2024G01).

Ethics Statement

All animal procedures were performed in strict accordance with ethical standards and were approved by the Animal Ethics Committee of Guangzhou University of Chinese Medicine (Approval No. 20240419004).

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: cas70399‐sup‐0001‐FigureS1‐S7.docx.

Table S1: Antibodies used for IHC and flow cytometry.

Table S2: Primer sequences.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.