Materials and reagents

Anlotinib dihydrochloride was a gift from Jiangsu Chia-Tai Tianqing Pharmaceutical Co., Ltd. SAR131675 was purchased from TargetMol (USA). PBS, RPMI 1640 medium, penicillin-streptomycin, and fetal bovine serum (FBS) were purchased from Wisent (Canada). The BCA protein assay kit and 40 nm FluoSpheresTM Carboxylate-Modified Microspheres were obtained from Thermo Fisher Scientific (USA). HSPC, cholesterol, and DSPE-PEG2000 were purchased from Lipoid GmbH (Germany), J&K (China), and AVT (China), respectively. RIPA lysis buffer, phenylmethanesulfonyl fluoride (PMSF), Evans blue, and DAPI were obtained from Solarbio Life Science (China). Cy5.5 was purchased from Meryer (China). Rhodamine 6 G was obtained from Aladdin (China). FITC-dextran was purchased from Sigma-Aldrich (USA). Cell counting kit-8 (CCK-8) was purchased from Dojindo Molecular Technologies (Japan). Doxorubicin hydrochloride was purchased from MESO (China). Recombinant human VEGF-C (C546) was obtained from Novoprotein (China). Antibodies against LYVE-1 (ab14917), HIF-1α (ab16066), NG2 (ab129051), fibronectin (ab2413) were obtained from Abcam (UK). An antibody against LYVE-1 (103-PA50) was purchased from ReliaTech (Germany). Antibodies against VEGFR-3 (AF4201) and p-VEGFR-3 (AF3676) were purchased from Affinity Biosciences (USA). Antibodies against Akt (4691 T), p-Akt (4060 T), Erk (4695 T), p-Erk (4377 T) were obtained from cell signaling technology (USA). Antibodies against CD31 (GB123151), CD11c (GB11059), CD4 (GB11064) and CD8 (GB114196), and fluorescein (FITC) TUNEL cell apoptosis detection kits were purchased from Servicebio (China). Dylight@488-labeled lycopersicon esculentum (Tomato) lectin (100 μl in PBS) was purchased from Vector Laboratories (USA). Anti-PD-L1 antibody was obtained from BioXcell (USA). A Mouse Proinflammatory Chemokine Panel (13-plex) with a V-bottomed plate (740451) and a Mouse Inflammation Panel (13-plex) with a V-bottomed plate (740446) were purchased from BioLegend (USA).

Cell culture

The murine breast cancer cell line 4T1 and the murine colon cancer cell line CT26 were obtained from the American Type Culture Collection. The human lymphatic endothelial cell line (hLECs) was purchased from iCell, China. The luciferase-expressing 4T1 cell line (4T1-luc) was generated by our group using a lentivirus transfection system from Addgene. Mycoplasma contamination was routinely tested negative. CT26 and 4T1 cell lines were cultured in RPMI 1640 supplemented with 10% FBS, penicillin (100 units ml−1), and streptomycin (100 μg ml−1) at 37 °C in a humidified incubator with 5% CO2. hLECs was cultured in primary endothelial cell culture medium purchased from ScienCell (USA).

Animals

BALB/c mice (female, 6–8 weeks old, 16–18 g body weight) were purchased from Vital River Animal Laboratories Biotechnology Co., Ltd. (Beijing, China) and housed under specific sterile pathogen-free conditions. All animal protocols were approved by the Animal Care and Welfare Committee of the National Center for Nanoscience and Technology. The 4T1 and CT26 xenograft tumor models were established by subcutaneously injecting 4T1 or CT26 cells (1 × 106 cells) into the right flank of mice. The injected cells were suspended in a 100 μl mixture of PBS and Matrigel (1:1; BD, USA). Tumor size was measured every 2 days with a digital caliper. The volume of tumor was determined by the following formula: V = L × W2/2, in which L represents the maximum tumor diameter, and W represents the minimum tumor diameter. When the tumors reached ~100 mm,3 the mice underwent treatment with different drug formulations. Tumor-bearing mice received anlotinib treatment at a dose of 1.5 mg/kg daily via intraperitoneal injection. For SAR131675 treatment, tumor-bearing mice were given 100 mg/kg SAR131675 orally every day.

Interference of VEGF-C-mediated activation of VEGFR-3 and downstream molecules in hLECs

To assess the impact of anlotinib and SAR131675 on VEGFR-3 phosphorylation, hLECs were pretreated with the two drugs at different concentrations (0, 0.1, 1 μg/ml), followed by stimulation with VEGF-C (200 ng/ml). The whole protein of hLECs was extracted using RIPA buffer containing 1.0 mM PMSF within 15 min. The whole protein concentration was quantified utilizing a BCA protein assay kit. Cell lysates were first separated by SDS-PAGE and then transferred onto PVDF membranes for further analysis. The membranes were initially blocked in 5% milk or BSA, and then incubated with primary antibody at 4 °C overnight. Subsequently, the membrane was washed and incubated with an HRP-conjugated secondary antibody (Santa Cruz Biotechnology, USA) for 1 h at room temperature. Detection of bands was achieved using the enhanced chemiluminescence reagents. GAPDH was used as an internal control.

Cell viability assay

hLECs or 4T1 cells were seeded at a density of 5 × 103 cells per well in 96-well cell culture plates and incubated for 24 h. To evaluate the effects of anlotinib and SAR131675 on the proliferation of hLECs, the cells were exposed to various concentrations of anlotinib, SAR131675, or PBS. For the cytotoxicity assays with Lip-Dox, 4T1 cells were incubated with various concentrations of Lip-Dox or the corresponding concentration of Dox. After 24 h incubation, the CCK-8 assay kit was utilized to determine cell viability.

Wound healing assay

hLECs were placed in 12-well cell culture plates at a density of 1.5 × 105 cells per well until fully confluent. Scratch wounds were made perpendicular to the cell layer with a sterile 200 μl pipet tip. The fixed observation position in each well was marked on the bottom of the dish with a marker. After being washed by PBS for three times, the cells were then incubated with various concentrations of anlotinib or SAR131675 for 15 h. The wounds were imaged by bright-field microscopy (Leica, Germany). The recovered area was measured at three repeated wells using ImageJ software.

Tubule formation assay

hLECs were seeded into 48-well plates pre-coated with 100 μL Matrigel per well and exposed to various concentrations of anlotinib or SAR131675 for 20 h. For each experimental condition, three parallel groups were set up. Tubule formation was imaged under an inverted light microscope. Tube length was quantified using ImageJ software.

In vivo anti-tumor lymphangiogenesis assay

When the size of tumors in 4T1 tumor-bearing mice reached approximately 100 mm,3 mice were treated daily with saline, anlotinib or SAR131675 for 10 consecutive days. Then tumors were dissected and immersed in 4% paraformaldehyde (PFA) for fixation, and immunohistochemical staining for LYVE-1 (103-PA50) was performed to show the distribution of LVs in tumors. Sections were observed using an inverted light microscope, and the LVs density was analyzed by ImageJ software. In addition, 4T1 or CT26 tumors from mice treated with saline and anlotinib for 10 consecutive days were dissected, embedded in OCT, and stored at −80 °C for subsequent immunofluorescence assays. Immunofluorescence staining of LYVE-1 (ab14917) was employed to analyze the density of LVs. All sections were photographed with a fluorescence microscope and quantitatively analyzed utilizing ImageJ software with the same threshold settings.

The expression levels of LYVE-1 were detected through Western blot analysis. Tumor tissues from mice treated daily with anlotinib, SAR131675 or saline for 10 consecutive days were surgically excised and mechanically disrupted in RIPA buffer with 1.0 mM PMSF. The following steps are the same with the above-mentioned method. Anti-LYVE-1 antibody (103-PA50) was used as the primary antibody. GAPDH was chosen as an internal control.

Evaluation of lymphatic drainage

To assess the drainage function of tumor-associated LVs, 4T1 tumor-bearing mice, 15 days after inoculation, were intravenously injected with 100 μl Lip-Rhodamine solution. 4 h later, the mice were anesthetized and intratumorally administered with 10 μl FITC-dextran 2,000,000 (25 mg/ml) to label tumor-associated LVs. Then, the inguinal LN adjacent to the tumor was surgically exposed, and the drainage of Lip-Rhodamine from the tumor to draining LN was observed by multiphoton laser scanning microscopy (Olympus, Japan). For the observation of the drainage of intratumoral material from the tumor to draining LN, 30 μl 0.2% Evans blue was locally injected into tumor tissue. 2 h later, the mice were euthanized, and the drainage of Evans blue from tumor tissue into the draining LN was imaged under an inverted light microscope.

Preparation and characterization of Au nanoparticles (AuNPs)

AuNPs in size of 100 nm were prepared based on the previously described protocols.54 Specifically, 600 μl fresh NaBH4 solution (10 × 10−3 M) was injected into 10 ml deionized water with HAuCl4 (0.25 × 10−3 M) and cetyltrimethylammonium bromide (CTAB, 100 × 10−3 M) under vigorous stirring. After incubation at 27 °C for 3 h, 50 μl mixture was mixed with 2 ml cetyltrimethylammonium chloride (CTAC, 200 × 10−3 M) and 1.5 ml ascorbic acid (AA, 100 × 10−3 M) followed by one-shot injection of 2 ml HAuCl4 solution (0. 5 × 10−3 M), obtaining 10 nm AuNPs. The nanoparticles were obtained by centrifugation at 14,500 rpm for 25 min, followed by a single rinse with deionized water, and then stored in 1 ml CTAC solution (20 × 10−3 M). 10 μl 10 nm seed solution was mixed with 2 ml CTAC (100 × 10−3 M) and 130 μl AA (10 × 10−3 M), then 2 ml HAuCl4 (0.5 × 10−3 M) was dropwise added into the solution at a speed of 2 mg h−1. The reaction was kept for 10 min under 27 °C, obtaining 46 nm AuNPs. The product was collected by centrifugation at 14,500 rpm for 10 min. After being washed by deionized water, 46 nm AuNPs were suspended in 0.86 ml of deionized water. Lastly, 0.5 ml 46 nm seed solution was mixed 2 ml CTAC (100 × 10−3 M) and 130 μl AA (10 × 10−3 M), and 2 ml HAuCl4 (0.5 × 10−3 M) was dropwise added into the solution at a speed of 2 mg h−1. The reaction was kept for 10 min under 27 °C. After centrifugation at 14,500 rpm for 10 min, 100 nm AuNPs were collected. Washed the product once with deionized and stored it in 0.86 ml deionized water.

The morphology of 100 nm AuNPs was observed by TEM (HT7700, HITACHI, Japan) at an 80 kV acceleration voltage. To prepare the samples for TEM, 10 μl of diluted AuNPs were deposited onto a carbon-coated copper grid. The surface plasmon absorption of AuNPs was detected by UV-Vis absorbance spectrophotometry (PerkinElmer, USA). The hydrated size, polydispersity index (PDI), and zeta potential of 100 nm AuNPs were determined utilizing a Malvern Zetasizer Nano ZS90 (Malvern, UK). Three measurements for each sample were performed.

Evaluation of the effects of anti-tumor lymphangiogenesis on lymphatic drainage of intratumoral nanoparticles

4T1 tumor-bearing mice were treated with anlotinib or saline for 10 consecutive days. 30 μl Cy5.5-labeled liposomes were locally injected into tumor tissues. 8 h later, tumors and the inguinal LNs adjacent to the tumors were dissected, and the fluorescence intensities of the nanoparticles in the LNs and tumors were measured using an IVIS. For TEM observation, mice bearing 4T1 tumors treated with anlotinib, SAR131675 or saline for 10 consecutive days were intratumorally injected with 100 nm Au nanoparticles. At 4 h post-injection, the inguinal LNs adjacent to the tumors were dissected and sectioned into fragments approximately 1 mm.3 After fixation with 2.5% glutaraldehyde in H2O (pH 7.2–7.4), rinse with buffer, and post-fixation with 1% OsO4 in PB (0.1 M, pH 7.4), the pieces were dehydrated through a graded ethanol series gradient. Then pieces were encased in resin and subjected to polymerization. 60–80 nm slices were prepared using ultramicrotome and stained with uranyl acetate and lead citrate. The cuprum grids were observed using TEM.

Evaluation of the influence of anlotinib/SAR131675 treatment on drug accumulation in tumor tissue

Mice bearing 4T1 tumors were treated daily with anlotinib or saline for 10 consecutive days and then intravenously injected with 100 μl of diluted 40 nm FluospheresTM Carboxylate-Modified Microspheres (1:133), Cy5.5-labeled anti-PD-L1 antibody (0.5 mg for each mouse), or 0.2% Evans blue (10 mg/kg). Then the mice were imaged through an IVIS at various time intervals, and fluorescence intensities of the tumor tissues were quantified. In order to evaluate the effect of anlotinib treatment on Dox accumulation in tumor tissue, 4T1 tumor-bearing mice received saline or anlotinib pre-treatment were injected with Dox (8 mg/kg) through the tail vein. At 12 h and 24 h post-injection, tumors were surgically isolated. Pieces of tumors (~200 mg) were mechanically disrupted in 500 μl RIPA lysis buffer. Next, 500 μl acetonitrile was added into the lysate, followed by incubation overnight at 4 °C. The mixture was then centrifuged at 14,000 × g at 4 °C for 30 min. The supernatant was obtained, and the concentration of Dox was determined by HPLC with a fluorescence detector at an excitation wavelength of 480 nm and emission wavelength of 590 nm. Samples obtained from untreated tumors were used as blank controls. For evaluation of the influence of SAR131675 treatment on intratumoral accumulation of small molecular drugs and nanoparticles, 4T1 tumor-bearing mice pretreated with saline or SAR131675 were injected with Dox (8 mg/kg) or Lip-Cy5.5 through the tail vein. For mice administered with Dox, Dox concentration in each tumor was measured utilizing HPLC 12 h post-injection. For mice administered with Lip-Cy5.5, the tumors were excised 8 h post-injection and imaged using an IVIS.

Evaluation of tumor blood vessels and tumor stroma

To evaluate the influence of anlotinib on the structure and function of tumor blood vessels, tumor tissues from mice treated daily with saline or anlotinib for 10 consecutive days were collected for further analysis. The density of tumor blood vessels was evaluated by immunohistochemistry (IHC) staining for the endothelial cell-specific marker CD31 (GB123151). Anti-NG2 (ab129051) and anti-CD31 (GB123151) antibodies were used to analyze the pericyte coverage of tumor vasculature by immunofluorescence staining. The percentage of pericyte-covered blood vessels was defined as the percentage of NG2+CD31+ area to the total CD31+ area. Immunofluorescence staining of cryosections was performed to investigate the perfused tumor blood vessels. Mice were injected with Dylight@488-labeled lectin through the tail vein. 10 min post-injection, tumors were harvested and cut into 15 µm sections after being embedded in OCT. An anti-CD31 (GB123151) antibody was used to label tumor blood vessels, and vessel perfusion was observed through a Zeiss LSM 710 confocal microscope. The proportion of perfused vessels was defined as the percentage of lectin+CD31+ area to the CD31+ area. Tumor hypoxia was analyzed by IHC staining using an anti-HIF-1α antibody (ab16066). To evaluate the impact of SAR131675 on tumor vasculature, tumors were excised from mice treated daily with saline or SAR131675 for 10 consecutive days. The density of tumor blood vessels and tissue hypoxia were evaluated using IHC staining for the CD31 (ab182981) and HIF-1α (ab16066), respectively. To assess changes in tumor stroma after anlotinib/SAR131675 treatment, Masson’s trichrome staining for collagen and IHC staining for fibronectin (ab2413) were performed on paraffin sections of tumors from mice treated with saline, anlotinib, or SAR131675 for consecutive 10 days. The results were analyzed using ImageJ software.

Photoacoustic imaging

Mice bearing 4T1 tumors were pretreated with saline or anlotinib for 10 consecutive days. Following anesthesia with isoflurane, the mice were placed on a heat pad. Then, transparent and bubble-free ultrasound gel was applied to the tumor area. The ultrasound and photoacoustic imaging system (Vevo LAZR-X, Japan) was employed for image acquisition and quantification. Photoacoustic signals were acquired under excitation at wavelengths of 750 nm and 850 nm, respectively.

Evaluation of the tumor interstitial fluid pressure (IFP)

Mice bearing 4T1 tumors were pretreated for 10 consecutive days with saline, anlotinib, or SAR131675. A saline-filled tube connected a needle to the pressure measuring system. After intraperitoneal anesthesia, the needle was carefully inserted into the center of tumor, and this step would cause a transient fluctuation in pressure. When the pressure stabilized, the data in the computer were recorded. The pressure signals within the tumors were converted into electric signals via a pressure transducer (PowerLab, ADInstruments) and recorded by LabChart 8 software.

Synthesis of Lip-Dox and fluorescent dyes-labeled liposomes

Lip-Dox was fabricated by the lipid thin-film hydration method. HSPC, Chol, and DSPE-PEG2000 were dissolved in ethanol at a molar ratio of 11.3:7.7:1. The ethanol was then evaporated under vacuum, resulting in a transparent film at the bottom of the flask. The dried lipid film was rehydrated with 2 ml PBS containing 2 mg Dox, followed by sonication to form a clear suspension. Then the suspension was extruded through a liposome extruder with 200 nm, 100 nm, and 50 nm pore size filters. The excess free Dox was removed by centrifugation at 5000 × g for 30 min through a 10 kDa molecular weight cut-off ultrafiltration device (Millipore, USA). For the preparation of fluorescent dyes-labeled liposomes, Rhodamine, lipophilic Cy5.5 was dissolved in ethyl alcohol together with HSPC, Chol, DSPE-PEG2000. Then ethyl alcohol was evaporated under vacuum, and the dried lipid film was rehydrated with 2 ml PBS, followed by successive extrusion through a liposome extruder with 200 nm, 100 nm, and 50 nm pore size filters. The liposome solutions were concentrated by centrifugation at 5000 × g for 30 min through a 10 kDa molecular weight cut-off ultrafiltration device (Millipore, USA).

Characterization of Lip-Dox and fluorescent dyes-labeled liposomes

The morphologies of Lip-Dox and fluorescent-dyes-labeled liposomes were characterized by TEM operating at an 80 kV acceleration voltage. To prepare the TEM samples, 10 μl of diluted suspended nanoparticles (1:100) was deposited onto a carbon-coated copper grid and subjected to negative staining with 1% uranyl acetate. Hydrated size, PDI, and zeta potential of Lip-Dox and fluorescent dyes-labeled liposomes were determined by dynamic light scattering utilizing a Malvern Zetasizer Nano ZS90 (Malvern, UK). Three measurements for each sample were performed.

For the determination of the drug encapsulation and loading efficiency of Lip-Dox, an equivalent volume of DMSO was mixed with freshly prepared Lip-Dox suspension followed by sonication. The supernatant containing Dox was collected after centrifugation at 18,516 × g for 10 min, and the content of Dox encapsulated in Lip-Dox was calculated by measuring the absorbance of the supernatant at 485 nm utilizing a UV–vis spectrophotometer (Lambda650, PerkinElmer, USA). To determine the drug loading efficiency, Lip-Dox suspension was freeze-dried, and the total mass of the nanoparticles was measured. The following formulas were employed to determine the encapsulation efficiency (EE) and loading efficiency (LE): EE = (weight of loaded drug) / (weight of initially added drug) × 100%; LE = (weight of loaded drug in the nanoparticle)/(weight of the nanoparticle) × 100%.

To evaluate Dox release profile of Lip-Dox, 1 ml freshly prepared Lip-Dox suspension was placed in a dialysis bag (molecular weight cutoff: 3.5 kDa) and dialyzed against 30 mL PBS (pH 4.4 or 7.4). This process was conducted at 37 °C with the system being orbital shaken at 200 rpm. At specific time intervals, 1 ml of the dialysis buffer was taken out for the purpose of Dox quantification, and concurrently, another 1 mL PBS adjusted to the corresponding pH was added to the system. The amount of released Dox was determined by measuring the absorbance at 485 nm via UV–vis spectrometry.

Flow cytometry and confocal microscopy were utilized for assessment of the cellular uptake of Lip-Dox. For flow cytometry experiments, 4T1 tumor cells were plated in 6-well plates with a seeding density of 5 × 103 cells per well and incubated overnight. The cultured cells were exposed to Lip-Dox or the corresponding concentration of Dox for 2 h. The cells were then collected for flow cytometry analysis using a BD Accuri C6 (BD, USA). Gating strategies are provided in Supplementary Fig. S21. For confocal microscopy, 4T1 tumor cells were placed in 8-well confocal microscopy dishes at a seeding density of 5 × 103 cells per well. After 24 h of incubation, Lip-Dox or the corresponding concentration of Dox was added, and samples were incubated for another 2 h. The cells were fixed by 4% paraformaldehyde for 15 min at room temperature, and the nuclei were stained with DAPI prior to imaging with a Zeiss LSM 710 confocal microscope.

In vivo combinatorial therapy of anlotinib with Dox or Lip-Dox

The anti-tumor efficacy of anlotinib in combination with nanomedicine or free drug was investigated in 4T1-luc and CT26 subcutaneous xenograft tumor models. Mice bearing 4T1-luc tumors with a size of ~100 mm3 were divided at random into six groups (n = 5), and received anlotinib or saline daily by intraperitoneal injection for 19 consecutive days. The mice were also treated with Dox or Lip-Dox (the administered Dox dosage was equivalent to 3 mg/kg) or saline through intravenous injection every 3 days. The tumor volume and body weight were monitored every other day. On day 26, three animals were euthanized and treated with D-luciferin potassium salt by intraperitoneal injection. At 10 min post-injection, the lungs of these mice were surgically isolated to observe the metastatic foci via an IVIS. Additionally, the excised lungs and tumors were preserved in 4% paraformaldehyde. Next, the tissues were encased in paraffin and sectioned into 5 μm slices for H&E and TUNEL staining. For the CT26 xenograft tumor model, the tumor tissues were surgically isolated on day 14 (n = 7), weighed, and fixed for H&E and TUNEL staining.

Anti-metastatic effects of anlotinib

Mice bearing 4T1 tumors with a size of ~100 mm3 were treated by anlotinib or saline by daily intraperitoneal injection for 10 consecutive days. Once the tumors reached 2000 mm3 in volume, the experiment ended, and the mice were sacrificed. Tumor-draining LNs as well as lungs were isolated for H&E staining and imaging.

Hematologic examination

Whole blood was harvested from CT26 tumor-bearing mice treated with different drug regimens through retro-orbital bleeding. The serum was separated for blood biochemistry analysis through centrifugation at 3000 rpm for 15 min. Liver function was evaluated via the measurement of alanine aminotransferase (ALT) and aspartate transaminase (AST) levels. Heart function was evaluated via measuring plasma lactate dehydrogenase (LDH), lactate dehydrogenase isoenzyme 1 (LDH1), creatine kinase (CK), and creatinine kinase isoenzyme MB (CK-MB).

Evaluation of anti-tumor immunity

Tumors and tumor-draining LNs of 4T1 tumor-bearing mice treated daily with saline or anlotinib for 10 consecutive days were excised for further analysis. IHC staining for CD11c was performed to evaluate the number of DCs in tumor-draining LN paraffin sections. IHC staining for CD4 or CD8 was performed to evaluate CD4+ and CD8+ T cell infiltration in tumors. The results were analyzed by ImageJ software.

In order to evaluate the expression levels of proinflammatory chemokines and inflammatory cytokines, tumors from mice treated daily with saline or anlotinib for 10 consecutive days were harvested and homogenized in PBS. Next, the tumor homogenates were analyzed using pro-inflammatory chemokine and inflammatory cytokine detection kits, according to the instructions.

For investigation of the effects of anlotinib on the therapeutic efficacy of the anti-PD-L1 antibody, mice bearing 4T1 tumors of approximately 100 mm3 were divided into four groups randomly (n = 6) and administered daily with anlotinib or saline by intraperitoneal injection for 10 consecutive days. At the same time, the mice received anti-PD-L1 antibody (0.75 mg/kg) or saline administration through intravenous injection every 2 days starting on day 3. The tumor volumes and body weights were monitored every 2 days. On day 14, the mice were euthanized, and the tumor tissues were surgically isolated and weighed.

Statistical analysis

The data were analyzed using GraphPad Prism 8. The data are presented as the mean ± standard deviation (mean ± s.d.). Significant differences between two groups were identified using unpaired student’s t-test. Comparisons among multiple groups employed one-way analysis of variance (ANOVA) with Tukey’s post-hoc test. Statistical significance was acknowledged when the two-sided p < 0.05.