Animals and cell culture

8-week-old male C57BL/6J mice and NOD.CB17-PrkdcscidIl2rgtm1/Bcgen (B-NDG) mice were obtained from Biocytogen (China) and housed under specific pathogen-free (SPF) conditions. All animal experiments in this study were conducted in accordance with the guidelines approved by the Ethics Committee for Animal Experimentation of Zhujiang Hospital (No. LAEC-2023-093). All the animals recruited in the study were anesthetized with 2% isoflurane delivered at a flow rate of 1 L/min prior to experimental procedures. For humane euthanasia, carbon dioxide was delivered at a controlled flow rate of 0.3 L/min in a 1-L capacity chamber. Euthanasia was performed both prior to primary cell isolation and at the end of the treatment monitoring period, in accordance with approved ethical standards.

For cell culture, Hepa1-6 cells and luciferase-transfected Hepa1-6 (Hepa1-6-luc) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10% fetal bovine serum (FBS; Procell, China) and 1% Penicillin–Streptomycin (Gibco, USA). L929 cells were cultured in Minimum Essential Medium (MEM) with the same supplements. All cells were maintained at 37 °C in a humidified incubator with 5% CO2. Cell lines were regularly tested for mycoplasma contamination.

Transcriptomic mRNA sequencing analysis

Hepa1-6 tumor bearing mice were randomly divided into 2 groups (n = 4). Mice in the KIF20A group received a subcutaneous injection of 5 μg KIF20A protein in sterile phosphate buffered solution (PBS), while the control group was injected with an equal volume of sterile PBS. After 3 days, the mice were sacrificed, and the tumor were harvested for transcriptomic mRNA sequencing.

Total RNA was extracted using Trizol reagent (Invitrogen, USA), and RNA quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA) as well as RNase-free agarose gel electrophoresis. Eukaryotic mRNA was enriched from the total RNA using Oligo (dT) beads (Epicentre, USA). The enriched mRNA was fragmented using a fragmentation buffer and reverse transcribed into cDNA with the NEBNext Ultra RNA Library Prep Kit for Illumina (NEB #7530, USA). Double-stranded cDNA was purified, followed by end repair, A-tailing, and ligation to Illumina sequencing adapters. The ligation products were purified with AMPure XP Beads (1.0×) and amplified by PCR to generate the cDNA library, which was sequenced using the NovaSeq6000 system (Illumina, USA).

Differential gene expression analysis was performed using R (version 4.10) with the ClusterProfiler package51. The genes/transcripts with false discovery rate (FDR) below 0.05 and absolute fold change (log2FC) ≥ mean ± 2 × SD were classified as DEGs. GO, KEGG enrichment analysis, and Gene Set Enrichment Analysis (GSEA) were conducted to identify functional and pathway enrichments of the DEGs, with gene sets obtained from the Molecular Signatures Database (MSigDB).

Preparation of KIF20A and R848 co-encapsulating liposome (K/RLip)

KIF20A and R848 co-encapsulating liposomes (K/RLip) were prepared using a thin film hydration method. Briefly, 28 mg (2,3-Dioleoyloxy-propyl)-trimethylammonium-chloride (DOTAP, Aladdin Scientific, China), 1.3 mg cholesterol (Aladdin Scientific, China)and 11.2 mg DSPE-PEG2000 (Aladdin Scientific, China) were dissolved in 40.5 mL chloroform to create a stock solution. Then, 2 mg R848 (Aladdin Scientific, China) was added into 10 mL stock solution. The mixture was subjected to rotary evaporation at 40°C until the solvent was completely removed to form a thin film. The evaporation step was repeated after the addition of another 10 mL chloroform. After the final evaporation, 0.5 mg KIF20A protein (Abmart, China) dissolved in 2 mL ultrapure water was used for hydration. The resulting suspension was then extruded through a mini extruder (Avanti, USA) equipped with 200 nm and 100 nm polycarbonate membranes (Millipore, USA), successively, for 11 passes each membrane to generate unilamellar liposomes. The liposomes were then dialyzed in a 200 kDa molecular weight cutoff (MWCO) dialysis tube for 12 h to remove any unencapsulated components, followed by lyophilization for another 12 h. The resulting K/RLip powder was stored at -20°C for further use. For the preparation of indocyanine green (ICG)-labeled Vac (ICG-K/RLip), 0.2 mg ICG was added in the hydration step.

Control formulations were also prepared, including R848-only liposomes (RLip) and KIF20A-only liposomes (KLip), using the same thin-film hydration method as described above.

Characterization of K/RLip and ICG-K/RLip

The morphology of K/RLip and ICG-K/RLip was observed using a TEM (FEI, USA). DLS and Zeta potential were measured with a Zetasizer (Malvern Panalytical, UK). The absorbance and fluorescence spectra were measured using a fluorescence spectrophotometer (Agilent, USA). The encapsulation efficiency and encapsulation capacity were determined with an ultramicro spectrophotometer (Thermo Scientific, USA) and calculated according to the following formulas:

$$\,}=\frac\; \; \,}\; \; \; }\times 100 \%$$

$$\,}=\frac}\,\; \; }}\; \; \; \; \; }\times 100 \%$$

Synthesis of PLGA-PEG-PLGA triblock copolymers

To synthesize PLGA-PEG-PLGA triblock copolymers, 60 g PEG1500 was heated to 120 °C for 30 min in the nitrogen atmosphere with magnetic stirring, followed by an additional 2 h under nitrogen flow to completely remove any moisture. Afterwards, 113.46 g DL-lactide and 30.48 g glycolide (molar ratio 3:1) were added to the PEG1500 and heated at 120°C under nitrogen flow. After 30 min, 0.04 g of stannous octoate was added and the reaction mixture was maintained at 150 °C for 8 h. Upon completion of the reaction, the system was cooled to room temperature. Then, 40 mL dichloromethane was added to dissolve the crude product, followed by precipitation with 500 mL petroleum ether to precipitate the product. The precipitate was collected by filtration. This dissolution-precipitation process was repeated three times to ensure a high-purity product. The successful synthesis of PLGA-PEG-PLGA triblock copolymers was confirmed using 1H-NMR (Bruker, Germany).

Preparation of thermosensitive hydrogel vaccine (K/RLip@Gel) and its characterization

PLGA-PEG-PLGA triblock copolymer were dissolved in ultrapure water to a final concentration of 15% (w/v). To generate K/RLip@Gel, 1 mg of K/RLip was added to 1 mL of the PLGA-PEG-PLGA solution and stirred at 4 °C for 1 h. Similarly, single-drug encapsulating liposomes containing either R848 or KIF20A protein were incorporated into the PLGA-PEG-PLGA solution to create RLip@Gel and KLip@Gel, respectively. The gelation properties of K/RLip@Gel were characterized using a rheometer (Anton Paar, Austria) at different temperature or under constant shear stress of 1% amplitude with 1 rad/s for 6 min.

Evaluation of in vivo sustained release

C57BL/6J mice were divided into three groups and subcutaneously injected 100 μL with small-molecule ICG, nanoscale K/RLip or hydrogel K/RLip@Gel. NIR-II fluorescence images were captured at 6, 12, 24, 48, 72 and 96 h post-injection using an 808-nm excitation laser and an 1100 nm long-pass filter to assess the sustained release properties of each formulation.

Isolation of BMDCs and spleen cells

BMDCs were isolated from 8-week-old C57BL/6J mice. Cells from the femur and tibia were flushed out with PBS and filtered through a 70 μm strainer (Corning, USA). The filtered suspension was centrifuged at 800×g for 5 min, and the supernatant was discarded. Red blood cells (RBCs) were lysed using ACK buffer (Elabscience, China), followed by a second centrifugation at 800×g for 5 min. The remaining cells were resuspended in Roswell Park Memorial Institute 1640 (RPMI1640) medium (Gibco, USA) supplemented with 10% FBS, 1% penicillin-streptomycin (Gibco, USA), 50 μM β-mercaptoethanol (Procell, China), 20 ng/mL murine granulocyte-macrophage colony-stimulating factor (GM-CSF, Peprotech, USA), and 10 ng/mL interleukin-4 (IL-4, MedChemExpress, USA). Cells were seeded into 6-well plates, and the cytokine-containing medium was replaced on days 3 and 5. After a 7-day incubation, the BMDCs were harvested for further experiments.

For the isolation of spleen cells, spleens from 8-week-old C57BL/6J mice were harvested, mechanically filtered against a 70 μm sterile strainer, and rinsed with PBS. The cell suspension was centrifuged at 800 × g for 5 minutes, and the supernatant was discarded. RBCs were lysed, and the remaining cells were resuspended in RPMI1640 medium supplemented with 10% FBS and 1% penicillin-streptomycin. The spleen cells were then cultured for further experiments.

Cytotoxicity and hemolysis test

L929 fibroblast cells were seeded into 96-well plates at a density of 1×104 cells per well and incubated with either Gel or K/RLip@Gel for 1, 3 and 5 days. The cytotoxicity of the formulations was assessed using the Cell Counting Kit-8 (CCK-8) assay according to the manufacturer’s instructions. Absorbance was measured at 450 nm using a microplate reader, and cell viability was calculated to evaluate the potential cytotoxic effects of Gel and K/RLip@Gel over time.

Blood was collected from a C57BL/6J mouse into a heparinized anticoagulant tube and washed three times with PBS to isolate RBCs. The RBCs were then incubated with varying concentrations of Gel or K/RLip@Gel for 4 hours at 37 °C. After incubation, the suspensions were centrifuged at 800×g for 5 min, and the supernatant was collected. Absorbance was measured at 532 nm to assess hemoglobin release, and the hemolysis rate was calculated using PBS as a negative control and distilled water as a positive control.

Flow Cytometry of BMDCs

BMDCs were divided into different groups and incubated with 100 μL solidified KLip@Gel, RLip@Gel or K/RLip@Gel in 6-well plates for 24 h. Then the BMDCs were collected, and the Fc receptors (FcR) were blocked using anti-mouse CD16/32 antibody (Biolegend, USA). Cells were subsequently stained with APC-conjugated anti-mouse CD11c, FITC-conjugated anti-mouse CD80, and PE-conjugated anti-mouse CD86 antibodies (Elabscience, China) for flow cytometry analysis (Beckman, USA). The CD80+CD86+ cells within the CD11c+ dendritic cell population were gated and quantified to assess BMDC maturation and activation.

T cell proliferation rate

BMDCs were pulsed with KLip@Gel, RLip@Gel or K/RLip@Gel for 24 h. Afterwards, Spleen cells labeled with CFDA-SE (MedChemExpress, USA) were co-cultured with the pulsed BMDCs for 3 days. After co-culture, the cells were harvested, and FcR was blocked using anti-mouse CD16/32 antibody. The cells were then stained with APC-conjugated anti-mouse CD3 antibody (Elabscience, China), and the CFDA-SE fluorescence intensity of CD3+ cells was analyzed by flow cytometry to assess T cell proliferation rate.

T cell infiltration

BMDCs were pulsed with KLip@Gel, RLip@Gel or K/RLip@Gel for 24 h. Subsequently, spleen cells were labeled with CellTrackerTM deep red (MKBio, China) and added to the BMDC-containing 6-well plates. After 3 days of co-culture, the cells were collected, and CD8+ T cells were isolated using the miniMACSTM system and CD8a+ T cell isolation kit (Miltenyi Biotec, Germany). The isolated CD8a+ T cells were counted and diluted to a concentration of 1 × 105 cells/mL. Then, the cell suspension was added into a round-bottom ultra-low attachment 96-well plate (Corning, USA) that contains Hepa1-6 cell spheroids to incubate for 24 h. Afterwards, the spheroids were fixed with 4% paraformaldehyde for 15 min and stained with DAPI for 10 min. CLSM (Nikon, Japan) was used to observe the CD8a+ T cells infiltrated into the Hepa1-6 spheroids.

Establishment of Hepa1-6 tumor models

To establish the subcutaneous CDX models, 1 × 106 Hepa1-6 cells suspended in 100 μL PBS were subcutaneously injected into the right hind limb or axillary region of 8-week-old male C57BL/6J mice. Mice with tumors that exhibited rapid growth without signs of necrosis or scab formation were selected for subsequent experimental studies.

For the establishment of orthotopic CDX models, male C57BL/6J mice (8 weeks old) were anesthetized using isoflurane, and an incision (roughly 8 mm) for each mouse was made along the abdominal midline. Subsequently, 1 × 106 Hepa1-6-luc cells suspended in a mixture of 10 μL PBS and 10 uL Matrigel (BD Biosciences, USA) were injected into the left lobe of the liver. The incision was then sutured using 6-0 threads. The successful establishment of orthotropic CDX models was comfirmed using bioluminescence imaging following intraperitoneal administration of D-luciferin potassium salt (150 mg/kg, GlpBio, USA).

In vivo tumor prevention effect of K/RLip@Gel

C57BL/6J mice were subcutaneously vaccinated with K/RLip@Gel (1 mg/mL, 100 μL) or sterile PBS and divided into 2 groups. 3 days post-vaccination, Hepa1-6-luc cells were injected orthotopically (n = 3) or subcutaneously (n = 3). Tumor growth rate was subsequently monitored with an IVIS Imaging Spectrum System (PerkinElmer, USA) using bioluminescence mode.

In vivo K/RLip@Gel and PD-L1 blockade therapy on Hepa1-6 tumor model

Hepa1-6 subcutaneous tumor-bearing mice were randomly divided into 4 groups (n = 5). Each group received different treatments: subcutaneous vaccination with K/RLip@Gel (5 mg/kg), intraperitoneal injection of anti-mouse PD-L1 antibody (10 mg/kg) (Bio X Cell, USA, Clone: 10F.9G2™)or a combination of both. Body weight and tumor volume were recorded every 3 days after drug administration. The tumor volume was calculated using the formula: tumor volume = length × width2/2. At the end of the treatment period, all mice were humanely euthanized. Organs and tumor tissues were harvested for histological analysis and the blood samples were collected for biochemical index testing.

Hepa1-6 orthotopic tumor-bearing mice were randomly assigned to the same 4 treatment groups as in the subcutaneous tumor model (n = 4). Body weight was recorded after injection. Additionally, the bioluminescent signal from orthotopic tumors was measured using bioluminescence imaging. At the end of the monitoring period, mice were humanely euthanized, and tumor tissues, inguinal lymph nodes, spleen, and blood samples were collected for flow cytometry analysis.

Establishment of immune-humanized PDX model

To establish the immune-humanized PDX model, primary tumor samples were obtained from a patient undergoing HCC resection surgery at Zhujiang Hospital and engrafted into B-NDG mice for 3 passages. The protocol involving human HCC samples was approved by the Medical Ethics Committee of Zhujiang Hospital (No. 2024-KY-219). PBMCs were isolated from the peripheral blood of a healthy volunteer with PBMC isolation tubes (Stemcell Technology, USA) and Histopaque®-1077 (Sigma-Aldrich, USA) and intravenously injected into the B-NDG mice (5 × 108 cells per mouse). Then, tumor tissues were planted into the PBMC injection mice as the 4th passages. The percentage of hCD45+ cells in the mouse peripheral blood was monitored weekly using flow cytometry. After 4 weeks, the successful establishment of the immune-humanized PDX model was confirmed when the tumor volume reached approximately 100 mm3 and the proportion of hCD45+ cell in mouse peripheral blood exceeded 25%.

In vivo K/RLip@Gel and PD-L1 blockade therapy on PDX model

Immune-humanized PDX models were divided into 4 groups (n = 3) and received treatments with K/RLip@Gel (5 mg/kg), anti-human PD-L1 antibody (10 mg/kg) (Bio X Cell, USA, Clone: Atezolizumab), or a combination of both. For the PDX models, the antigen used in K/RLip@Gel was human KIF20A protein instead of the mouse protein. Body weight and tumor size were monitored regularly after drug administration. At the end of the monitoring period, blood samples and tumor tissues were collected for flow cytometry and histological analysis to evaluate the therapeutic effects.

Flow cytometry analysis of tissue and blood

Single-cell suspensions were prepared from tissues prior to flow cytometry analysis. Tumor tissue fragments (1–2 mm³) were digested with collagenase I (200 U/mL), hyaluronidase (100 U/mL), and DNase I (50 U/mL) at 37 °C for 45 min with shaking at 90 rpm. Spleen tissues were lysed using collagenase IV (150 U/mL) and DNase I (50 U/mL) under the same conditions. Inguinal lymph nodes were digested with collagenase I (200 U/mL), hyaluronidase (100 U/mL), and DNase I (50 U/mL). Blood samples were washed with PBS at 800 rpm for 5 minutes with 3-time repeat. All samples underwent red blood cell (RBC) lysis before antibody labeling.

For the Hepa1-6 tumor model (CDX model), cells were first blocked with anti-mouse CD16/32 antibody. Intratumoral, splenic, and peripheral blood T cells were labeled with PerCP-conjugated anti-mouse CD45, APC-conjugated anti-mouse CD3, PE-conjugated anti-mouse CD4, and FITC-conjugated anti-mouse CD8a antibodies. Lymphatic dendritic cells were labeled with APC-conjugated anti-mouse CD11c, FITC-conjugated anti-mouse CD80, and PE-conjugated anti-mouse CD86 antibodies. All antibodies used for the CDX model were purchased from Elabscience (China).

For the PDX model, cells were blocked with FcR Blocking Solution (Biolegend, USA). cDCs were labeled with PerCP-conjugated anti-human CD45, FITC-conjugated anti-human Lineage Cocktail-1 (Lin-1, containing anti-CD3, CD14, CD16, CD19, CD20, and CD56 antibodies) (Biolegend, USA), APC-conjugated anti-human HLA-DR (Elabscience, China), and PE-conjugated anti-human CD11c (Elabscience, China). T cells were labeled with PerCP-conjugated anti-human CD45 (Elabscience, China), APC-conjugated anti-human CD3 (Elabscience, China), FITC-conjugated anti-human CD8a (Elabscience, China), PE/Cyanine7-conjugated anti-human IFN-γ (Biolegend, USA), and PE-conjugated anti-human Granzyme B (AntibodySystem, USA). The FIX & PERM™ Cell Permeabilization Kit (Invitrogen, USA) was used before labeling with PE/Cyanine7 anti-human IFN-γ and PE anti-human Granzyme B.

statistical analysis

All statistical analyzes were performed using GraphPad Prism 10.0 (GraphPad Software, USA). Differences among 2 groups were analyzed using the student’s T test unless otherwise indicated. Values of P < 0.05 were considered statistically significant. The significant differences were noted in the figures: ns (no significant difference), *(P < 0.05), **(P < 0.01), ***(P < 0.001), ****(P < 0.0001).