Human subjects

Twenty-three patients who underwent surgery at the Third Affiliated Hospital of Sun Yat-Sen University (Guangzhou, China) with proven CRC were recruited for this study. Patients who received preoperative antitumor treatments (neoadjuvant radiotherapy or chemotherapy) were excluded. Tumor tissues and adjacent normal tissues were obtained as soon as possible after resection for the corresponding experiments. Patient consent was obtained from each patient, and the Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University approved the study. The demographic and clinical information and histopathological data for each patient are shown in Supplementary Table 1.

Public database

R2, UALCAN [34], and TIMER are interactive web portals that allow in-depth analysis of TCGA gene expression data. The R2: Genomics Analysis and Visualization Platform is publicly available at http://r2.amc.nl. UALCAN is publicly available at http://ualcan.path.uab.edu. TIMER is publicly available at https://cistrome.shinyapps.io/timer/.

Western blotting

Treated tumor cells were lysed with RIPA lysis buffer(Beyotime Biotech, China)supplemented with PMSF and phosphatase inhibitor at 4℃. For CRC samples, 20 mg of tissues was lysed with 200 µL of RIPA buffer in a grinder and then sonicated for 1 min. The lysate was subsequently centrifuged at 12,000 rpm for 10 min at 4℃. After the protein concentration was determined with a bicinchoninic acid protein assay kit (Beyotime Biotech, China), 30 µg of each sample was separated via 10% SDS-PAGE and then transferred to PVDF membranes (Millipore, USA). Then, the membrane was blocked with 5% BSA for 1 h and incubated with the corresponding primary antibodies (shown in the Supplementary Table 2) overnight at 4℃. The membrane was washed with TBST three times and incubated with horseradish peroxidase–conjugated secondary antibodies at room temperature for 2 h. Finally, the bands were visualized via an automatic chemiluminescence image system (Tanon 5200, China) and enhanced chemiluminescence detection kits (Beyotime Biotech, China).

Immunohistochemistry and immunofluorescence analyses of tissue samples

The paraffin-embedded tissues were sectioned into slices at a thickness of 4 μm, and the sections were deparaffinized and rehydrated after heated in 65 °C for 1 h. Tris-EDTA antigen retrieval solution was subsequently used for heat-induced epitope retrieval with a microwave oven. After incubation with 5% BSA at room temperature for 1 h, the sections were incubated with primary antibodies (shown in Supplementary Table 2) overnight at 4℃. For immunohistochemistry, the sections were incubated with peroxidase-conjugated antibodies and detected with a DAB system (Servicebio Technology, China). For immunofluorescence staining, the samples were incubated with fluorescently labelled secondary (Servicebio Technology, China) antibodies for 60 min at room temperature and the nuclei were stained with DAPI. Finally, the sections were imaged via a confocal microscope (Leica, Leica STELLARIS STED, Germany) or fluorescence microscope (Nikon, ECLIPSE Ti2-U, Japan).

Enzyme linked immunosorbent assay (ELISA)

To evaluate the release of kynurenic acid, ATP, and HMGB1 from CT26 tumor cells, the cells were seeded in 6 well plates at a density of 5 × 105 per well and incubated in RPMI-1640 medium supplemented with 10% FBS for 24 h. Then, the medium was removed and replaced with fresh medium containing phosphate-buffered saline (PBS), NP-I, NP-P, or NP-I/P. After incubation for 48 h, the culture medium was collected and kynurenic acid, ATP, and HMGB1 levels were measured by using ELISA kits according to the manufacturer’s instructions (Meimian, China). To evaluate the release of kynurenic acid and TGF-β from MDSCs, purified MDSCs were seeded in 6 well plates at a density of 5 × 105 per well and incubated in 1640 medium supplemented with 10% FBS for 24 h and then the same procedure as above was performed.

In vitro cytotoxicity experiments

Cell Counting Kit-8 (CCK8) assays were performed to evaluate the cytotoxicity of the nanodrugs in vitro. CT26 colon cancer cells were seeded into 96 well plate at a density of 5 × 103/100 µL and cultured for 24 h. Then, the culture media were replaced and nanodrugs were added to the media at a specific concentration gradient. After 48 h, CCK-8 reagent (Beyotime Biotech, China) was added, and cell viability was assessed according to the manufacturer’s instructions.

Preparation and characterization of nanodrugs

mPEG-PCL was purchased from Tansh Technology (Guangzhou China) and its structure was confirmed by nuclear magnetic resonance (NMR, Supplementary Fig. 1). Panobinostat and epacadostat were obtained from MedChemExpress Co., Ltd (Shanghai, China). A self-assembly method was utilized to prepare the nanodrugs as described previously [35]. For the synthesis of a nanodrug loaded with epacadostat and panobinostat (NP-I/P), 20 mg of mPEG-PCL was first dissolved in tetrahydrofuran (THF) solution. Then, 0.3 mg of panobinostat and 0.12 mg of epacadostat were added to the above mPEG-PCL solution. The mixture was added dropwise to ultrapure water (6 ml) while stirring at 1000 rpm, after which evaporation of THF at room temperature was performed for 48 h. Finally, the formed NP-I/P was collected and filtered through a 0.22 μm filter. For the synthesis of nanodrugs loaded with epacadostat (NP-I), panobinostat (NP-P), 1,1’-dioctadecyl-3,3,3’,3’-tetramethylindotricarbocyanine iodide (DiR), or coumarin, only epacadostat, panobinostat, DiR, or coumarin was added to the mPEG-PCL solution, respectively. The size distribution and morphology of the nanodrugs were measured via dynamic light scattering (DLS) and transmission electron microscopy (TEM). To confirm the stability of NP-I/P, 0.1 mL of NP-I/P solution was diluted in 0.9 mL of PBS containing 10% FBS at 37℃, and nanoparticle size was measured via DLS at predetermined time points.

Apoptosis analysis

A total of 5 × 105 CT26 cells or MDSCs were seeded in six well plate and treated with PBS, NP-I, NP-P, or NP-I/P for 48 h. The treated cells were collected and washed with PBS. The cells were subsequently resuspended and stained with the Annexin V-EGFP/PI Apoptosis Detection Kit (KeyGEN Biotech, China) at room temperature and analyzed via flow cytometry.

Cell cycle analysis

A total of 5 × 105 CT26 cells were seeded in six-well plates and treated with PBS, NP-I, NP-P, or NP-I/P for 48 h. The treated cells were collected and fixed in cold ethanol overnight at 4 °C. The cells were subsequently resuspended and stained with Cell Cycle Detection Kit (KeyGEN Biotech, China) reagents at room temperature for 30 min and analyzed via flow cytometry.

Live/dead viability assay

CT26 cells were seeded in 35 mm dishes at a density of 2 × 105 cells per dish and incubated in RPMI-1640 medium for 24 h. Then, the medium was removed and replaced with fresh medium containing PBS, NP-I, NP-P, and NP-I/P. After incubation for 48 h, the cells were stained with calcein acetoxymethyl ester (calcein AM) and PI (KeyGEN Biotech, China) and observed via fluorescence microscopy.

Migration, invasion, and colony formation assays

Cell migration and invasion assays were performed in 24-well Transwell chambers with 8-µm pore size polycarbonate. Matrigel was coated on the chamber for the invasion assay. A total of 5 × 104 cells were counted, suspended in serum-free culture medium, and then plated in the upper chamber. A total of 600 µL of cell culture medium containing 10% foetal bovine serum was added to the bottom chamber. After incubation for 24 h at 37 °C, the cells that migrated or invaded through the pores of the membrane were fixed with 4% paraformaldehyde and then stained with 0.1% crystal violet staining solution. The stained cells were visualized and counted under a microscope. For cell colony formation, the collected cells were counted, seeded into six-well cell culture plates (1000 cells per well) and incubated at 37 °C for 1 week. Then, the colonies were fixed with 4% paraformaldehyde for 15–20 min and stained with 0.1% crystal violet staining solution for 20–30 min. The cell colonies were photographed and counted.

Flow cytometry

Cells from different experiments were harvested and stained with fluorochrome-conjugated Abs according to the manufacturer’s instructions. For surface staining, the cells were collected and resuspended in the corresponding staining buffer. For intracellular staining, the cells were first stained with surface markers, fixed and permeabilized with Cytofix/Cytoperm Soln Kit (Becton Dickinson, USA) reagents, and finally stained with intracellular markers. To detect foxp3 in the nucleus, mononuclear cells were fixed and permeabilized with specific reagents (eBioscience, USA) after surface staining. For some experiments, T cells were stimulated with Leukocyte Activation Cocktail (Becton Dickinson, USA) at 37 °C for 4 h before staining. Finally, the data were acquired with a Calibur flow cytometer (Becton Dickinson, USA) and analysed with FlowJo software. The antibodies used in this study are summarized in Supplementary Table S2.

Cell immunofluorescence analysis

For cell immunofluorescence analysis, CT26 cells (1 × 105 cells per dish) were seeded into 30 mm dishes with 2 ml of RPMI-1640 and incubated at 37 °C for 24 h. PBS, NP-I, NP-P, and NP-I/P were added to different dishes at the same concentration of panobinostat and epacadostat and cultured for 48 h. After fixation with 4% paraformaldehyde for 20 min and antigen blocking with 5% BSA, the cells were stained with anti-CRT (Abcam, England) or anti-HSP70 (Abcam, England) antibody overnight at 4 °C and subsequently labelled with a FITC-conjugated secondary antibody for 1 h at room temperature. The cells were subsequently stained with DAPI and observed via laser scanning confocal microscopy (LSCM).

Isolation and purification of MDSCs and CD8+ T cells

For the isolation of MDSCs, a CT26-bearing mouse model was first constructed as described above. When the tumor size reached 500 mm3, the mice were sacrificed. The spleens were harvested and mechanically digested, followed by filtration through a 70 mm filter to obtain a single-cell suspension. Mononuclear cells were subsequently isolated via a standard Ficoll procedure as described previously [36]. MDSCs were purified via an anti-mouse Gr-1-biotin antibody and anti-biotin microbeads (Miltenyi Biotech, Germany) according to the manufacturer’s instructions.

For CD8+ T cell purification, BALB/c mice (male, 4–6 weeks of age) were sacrificed, and the spleens were subjected to isolation of mononuclear cells. Then, the CD8+ T cells were purified with anti-CD8a (Ly-2) MicroBeads (Miltenyi Biotech, Germany). Finally, the purity of the MDSCs and CD8+ T cells was determined via flow cytometry.

Isolation and culture of bone marrow-derived mononuclear cells (BMMCs)

BALB/c mice (aged 4–6 weeks) were sacrificed, and the femurs and tibias were collected after immersion in alcohol for 30 min. The ethanol on the surface of the tibia and femur was washed off with PBS. After the two ends of the femur and tibia were cut with scissors, a 1 mL syringe of cold PBS was used to aspirate the bone marrow from the femur and tibia. The bone marrow was subsequently filtered through a 70 μm mesh, and the blood cells were removed with red blood cell lysate to obtain BMMCs. The BMMCs were washed with PBS and resuspended in 1640 medium for the next experiment.

For macrophage induction, 20 ng/ml M-CSF was added to the culture medium of BMMCs, which were subsequently incubated at 37 °C with 5% CO2 for 7 days to obtain bone marrow-derived macrophages (BMDMs).

For DC induction, 15 ng/ml GM-CSF and 10 ng/ml IL-4 were added to the culture medium of BMMCs and incubated at 37 °C with 5% CO2 for 7 days. In accordance with routine culture, half of the culture medium was replaced every other day, and GM-CSF/IL-4 was added. On day 7, the culture medium was discarded, and loosely adhered BMDCs were collected to obtain immature DCs (iDCs).

Coculture of immune cells with CT26 tumor cells

For CD8+ T cells cocultured with CT26 tumor cells, purified CD8+ T cells were seeded in 6-well plates precoated with a-CD3 and a-CD28 was added to the culture medium. CT26 cells were placed into the plate at a ratio of 10:1 (CD8+ T cells:CT26 cells ) after treatment with different nanodrugs for 48 h. Then, the cells were cocultured for another 48 h, and the CD8+ T cells were harvested for IFN-γ detection. In some experiments, purified CD8+ T cells were first labelled with CFSE and then cocultured with CT26 cells to investigate the impact on cell proliferation.

For coculture of macrophages and DCs with CT26 tumor cells, macrophages and DCs isolated from the bone marrow were seeded in a 6-well plate, and CT26 cells were placed into the plate at a ratio of 1:2 (macrophage/DC: CT26) after treatment with different nanodrugs for 48 h. Then, the cells were cocultured for another 48 h and harvested for flow cytometry.

Animals and cell lines

BALB/c male and BALB/c-nude mice (4–6 weeks of age) were purchased from JinWei Biotechnology Co., Ltd. (Guangzhou, China). All animals were maintained under specific pathogen-free conditions. The animal experimental procedures were approved by the Institutional Animal Care and Use Committee of Guangdong Pharmaceutical University and were performed in accordance with the guidelines. CT26 cells were purchased from ATCC via iCell Bioscience, Inc. (Shanghai. China).

CT26-bearing mouse model and treatment

To construct the CT26-bearing mouse model, CT26 cells (5 × 105 cells) were subcutaneously inoculated into the left flanks of male BALB/c mice. The diameters of the tumors were measured with caliper, and the tumor volume was calculated as 1/2a×b2.

For in vivo fluorescence imaging, DiR-labelled NP-I/P was injected into CT26-bearing mice at a DiR dose of 0.75 mg∙kg− 1 body weight via the tail vein when the tumor volume reached 200 mm3. A fluorescence imaging system (Carestream IS 4000, USA) was subsequently used to capture in vivo images at 1, 3, 6, 12, 24, and 48 h after DiR injection. Forty-eight hours later, the tumors and major organs (heart, liver, spleen, lung, and kidney) were harvested for in vitro imaging.

For the antitumor response study, the mice were randomly divided into five groups (n = 6 mice per group) when the tumor volume reached 100 mm3 and treated with 200 µL of saline, I/P, NP-I, NP-P, or NP-I/P containing panobinostat and/or epacadostat at doses of 5 mg∙kg and 2 mg∙kg− 1 body weight every three days. Tumor size and body weight were recorded every two days. After five cycles of treatment, the mice were sacrificed, and the tumors, major organs, and sera were collected for further experiments.

To evaluate the in vivo toxicity of the nanodrugs, the sera of CT26 tumor-bearing mice were utilized to assess liver function and renal function, and the tissues of major organs were stained with haematoxylin and eosin (H&E).

H&E staining

Formalin-fixed paraffin-embedded tissues were sectioned into slices at a thickness of 4 μm. The sections were deparaffinized and rehydrated after heated in 65 °C for 1 h. Then, the sections were stained with haematoxylin solution for 5 min and rinsed with water for 5 min. Next, the tissues were stained with eosin solution. After being sealed with neutral gum, the sections were observed under Nikon light microscope (Nikon, ECLIPSE Ti2-U, Japan).

Sirius red staining

Formalin-fixed paraffin-embedded tissues were sectioned into slices at a thickness of 4 μm. The sections were deparaffinized and rehydrated after heating in 65 °C for 1 h. Then, the sections were stained with Sirius red for 8 min and dehydrated with anhydrous ethanol. After being sealed with neutral gum, the sections were observed under a Nikon light microscope (Nikon, ECLIPSE Ti2-U, Japan).

PDX model and treatment

To establish the PDX mouse model, fresh tumor tissues from CRC patients were sliced into 3–5 mm-diameter pieces, and then, the tissues (defined as G0) were subcutaneously implanted in the left flank of BALB/c-nude mice (defined as G1). When the tumors in G1 mice reached 500 mm3, the tumors were harvested, sliced into 3–5 mm pieces and subcutaneously implanted in the left flank of another BALB/c-nude mouse to generate G2 mice, which were suitable for in vivo experiments.

For the antitumor efficacy study in PDX model, the mice were randomly divided into five groups (n = 6 mice per group) when the tumor volumes reached 100 mm3 and treated with 200 µL of saline, I/P, NP-I, NP-P, or NP-I/P containing panobinostat and/or epacadostat at doses 5 mg∙kg− 1 and 2 mg∙kg− 1 body weight every four days. Tumor size and body weight were recorded every two days. After five cycles of treatment, the mice were sacrificed, and the tumors were harvested for further experiments.

Human PDO culture

Human PDOs were cultured as described previously. Type I collagen gel solution was prepared according to the manufacturer’s instructions, added to 30 mm Transwells inserts containing membranes with 0.4 μm pores and solidified at 37 °C for 30 min. Fresh tumor tissues were minced into 125–500 mm3 fragments, washed twice with ADMEM/F12 (Invitrogen) containing 1X Normocin (InvivoGen), resuspended in 1 mL of type I collagen gel and placed into inner Transwells precoated with gel. The Transwells inserts containing the tumor tissues were subsequently placed into other culture dishes containing 1 mL of ADMEM/F12 culture medium supplemented with Wnt3a, RSPO1, or Noggin-conditioned media (L-WRN, ATCC) supplemented with HEPES (1 mM, Invitrogen), glutamine (1X, Invitrogen), gastrin (10 mM, Sigma), N-acetylcysteine (1 mM, Sigma), epidermal growth factor (50 ng/mL, PeproTech), IL-2 (6000 IU/mL, PeproTech), G-CSF (50 ng/mL, PeproTech) or GM-CSF (50 ng/mL, PeproTech). Nanodrugs were added to the culture medium and incubated for 5 days, after which the samples were collected for further experiments.

Drug phagocytosis in PDOs

For the drug phagocytosis experiment with the PDO model, NP-encapsulated coumarin was added to the PDO culture medium as described above. After incubation for 2, 4, 8, or 24 h, the samples were collected and made into frozen slices. The frozen slices were stained with DAPI and observed under a fluorescence microscope to assess whether the nanodrugs could be phagocytosed by PDO samples.

Statistical analysis

The GraphPad Prism 7 software program was used for statistical analysis of the experimental data, and the results are presented as the means ± standard deviations (SDs). Significant differences between two unpaired groups were determined by the log-rank test or Student’s t test. P < 0.05 was considered to indicate statistical significance. The significance level was indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001.