Liposomal honokiol inhibits VEGF-D-induced lymphangiogenesis and metastasis in xenograft tumor model

Authors

  • Jing Wen,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
    • The first two authors contributed equally to this work.

  • A-fu Fu,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
    • The first two authors contributed equally to this work.

  • Li-Juan Chen,

    Corresponding author
    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    • State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Gaopeng Street, Keyuan Street Road 4, Chengdu 610041, Sichuan, China
    Search for more papers by this author
    • Fax: +86-28-85164060

  • Xing-Jiang Xie,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Guang-Li Yang,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Xian-Cheng Chen,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Yong-Sheng Wang,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Jiong Li,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Ping Chen,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Ming-Hai Tang,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Xi Ming Shao,

    1. Institute for Nanobiomedical Technology and Membrane Biology, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • You Lu,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Xia Zhao,

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    2. Department of Gynecology and Obstetrics, Second West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author
  • Yu-Quan Wei

    1. State Key Laboratory of Biotherapy, West China Hospital, West China Medical School, Sichuan University, Chengdu, China
    Search for more papers by this author

Abstract

Lymph nodes metastasis of tumor could be a crucial early step in the metastatic process. Induction of tumor lymphangiogenesis by vascular endothelial growth factor-D may play an important role in promoting tumor metastasis to regional lymph nodes and these processes can be inhibited by inactivation of the VEGFR-3 signaling pathway. Honokiol has been reported to possess potent antiangiogenesis and antitumor properties in several cell lines and xenograft tumor models. However, its role in tumor-associated lymphangiogenesis and lymphatic metastasis remains unclear. Here, we established lymph node metastasis models by injecting overexpressing VEGF-D Lewis lung carcinoma cells into C57BL/6 mice to explore the effect of honokiol on tumor-associated lymphangiogenesis and related lymph node metastasis. The underlying mechanisms were systematically investigated in vitro and in vivo. In in vivo study, liposomal honokiol significantly inhibited the tumor-associated lymphangiogenesis and metastasis in Lewis lung carcinoma model. A remarkable delay of tumor growth and prolonged life span were also observed. In in vitro study, honokiol inhibited VEGF-D-induced survival, proliferation and tube-formation of both human umbilical vein endothelial cells (HUVECs) and lymphatic vascular endothelial cells (HLECs). Western blotting analysis showed that liposomal honokiol-inhibited Akt and MAPK phosphorylation in 2 endothelial cells, and downregulated expressions of VEGFR-2 of human vascular endothelial cells and VEGFR-3 of lymphatic endothelial cells. Thus, we identified for the first time that honokiol provided therapeutic benefit not only by direct effects on tumor cells and antiangiogenesis but also by inhibiting lymphangiogenesis and metastasis via the VEGFR-3 pathway. The present findings may be of importance to investigate the molecular mechanisms underlying the spread of cancer via the lymphatics and explore the therapeutical strategy of honokiol on antilymphangiogenesis and antimetastasis. © 2008 Wiley-Liss, Inc.

The major cause of cancer mortality is the metastatic spread of tumor cells via multiple routes, including the vascular system (hematogenic metastasis) and the lymphatic system (lymphogenic metastasis).1–3 The lymphatic system is an important route contributing to metastasis of solid tumors.4 VEGFR-3/Flt-4 has been implicated in lymphangiogenesis and lymphatic metastasis. Clinic pathological data from research using animal models indicates that tumor-induced lymphangiogenesis driven by vascular endothelial growth factor (VEGF)-C- and/or VEGF-D-induced activation of VEGF receptor (VEGFR)-3 could promote metastasis to regional lymph nodes.5–8

Honokiol is a bioactive constituent isolated from the bark of Magnoliae. Previous reports have demonstrated that honokiol induces cell apoptosis in several cell lines, such as leukemia cell lines HL-60, colon cancer cell lines RKO, lung cancer cell lines A549 and CH27.9–14 It also has remarkable in vivo antitumor activities in tumor mouse models.15 Honokiol has demonstrated potent antiangiogenic and antitumor properties against aggressive angiosarcoma by blocking of VEGF-induced VEGF receptor 2 autophosphorylation.12, 13 The underlying molecular mechanisms may contribute to the inhibition of phosphorylation of AKT, p42/44 MAPK.15 Our previous studies have shown that honokiol possesses significant antitumor effect as well as an enhancement in tumor growth delay and improvement of survival by combination with cisplatin in ovarian carcinoma16 or with adriamycin in breast cancer models.17 The poor water solubility of honokiol could be improved by encapsulating honokiol with PEG modified liposome.17 However, it is still unclear whether honokiol effectively inhibits lymphangiogenesis and its related lymphatic metastasis.

In in vivo studies, we established tumor metastatic models by injecting high-expression VEGF-D Lewis lung cancer cells (VEGF-D-LL/2) both in muscle (n = 5) and subcutaneously in hind limb (n = 5) of C576BL/6 mice, respectively. Liposomal honokiol was injected i.p. into the immunocompetent C57BL/6 mice bearing VEGF-D high-expression Lewis lung cancer cells to explore honokiol's possible potential application in preventing tumor metastasis via lymphatic vessels. In in vitro studies, we explored the direct effects of honokiol on cultured lymphatic endothelial cells and high-expression VEGF-D Lewis lung cancer cells and examined the expression of AKT and p44/42 MAPK by western blotting to investigate the effect of honokiol on the VEGFR-3 signaling pathway. Tube formation of HLECs on Matrigel was used to investigate the possible antilymphogenic effect of honokiol. We demonstrated for the first time that honokiol not only has antiangiogenesis and antitumor activity but also inhibits lymphangiogenesis and metastasis via the VEGFR-3 pathway.

Abbreviations:

DSMO, dimethyl-sulphoxide; HLEC, human lymphatic endothelial cells; HPLC, high performance liquid chromatography; HUVEC, human umbilical vein endothelial cells; LH, liposomal honokiol; MAPK, Mitogen-activated protein kinases; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PBS, phosphate buffered saline; PEG4000, polyethylene glycol 4000; RT-PCR, reverse-transcription polymerase chain reaction; TUNEL, terminal deoxynucleotidyl transferase-mediated nick-end labeling; VEGF-D, vascular endothelial growth factor D; VEGF-D-LL/2, Lewis lung carcinoma cells high expression VEGF-D; VEGFR-2(vascular endothelial growth factor receptor-2; VEGFR-3, vascular endothelial growth factor receptor-3.

Material and methods

Antibody and reagents

Anti-VEGF-D polyclonal rabbit antibody, anti-VEGFR-3 (Flt-4) polyclonal rabbit antibody, anti-β-actin monoclonal mouse antibody, goat anti-mouse LYVE-1 polyclonal antibody, goat anti-rabbit IgG/HRP and goat anti-mouse IgG/HRP were purchased from Santa Cruz Biotechnology (Santa Cruz, CA); anti-p44/42 MAP Kinase monoclonal rabbit antibody, anti-Phospho-p44/42 MAP Kinase monoclonal rabbit antibody, anti-AKT monoclonal rabbit antibody and anti-Phospho-AKT monoclonal mouse antibody were purchased from Cell Signaling Laboratories (Beverly, MA); anti-VEGFR-2 (KDR) monoclonal mouse antibody, polyclonal goat anti-mouse VEGF-D antibody and Rabbit anti-goat IgG/HRP was purchased from R&D systems (Minneapolis, MN); Rat anti-mouse CD-31(PECAM-1) monoclonal antibody was purchased from BD Pharmingen (La Jolla, CA).

PC, cholesterol and PEG4000 were purchased from Sigma Chemical Co, (St. Louis, MO); Matrigel was purchased from BD Pharmingen (La Jolla, CA); Gelatin was purchased from Sigma Chemical Co, (St. Louis, MO); Honokiol was separated and purified by our laboratory and its purity and structure were analyzed and identified by high performance liquid chromatography and nuclear magnetic resonance.18

Cell culture

Human lymphatic endothelial cells (HLECs) were purchased from the ScienCell™ Research Laboratories and cultured in endothelial cell medium (ECM) containing 5% fetal bovine serum (FBS), 1% endothelial cell growth supplement (ECGS), 100 IU/mL penicillin and 100 μg/mL streptomycin. Lewis lung carcinoma cells (LL/2) were obtained from the American Type Culture Collection (ATCC) and cultured in DMEM containing 10% fetal bovine serum (FBS), 100 IU/mL penicillin and 100 μg/mL streptomycin; Human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical cord with collagenase and used at passage 2–3.19 After dissociation, the cells were collected and cultured on gelatin-coated culture flasks in M199 medium with 20%FCS, 10 ng/mL bFGF, 2 ng/mL VEGF, 100 IU/mL penicillin and 100 μg/mL streptomycin. Subcultures were performed with trypsin-EDTA. Media were refreshed every second day. The identity of umbilical vein endothelial cells was confirmed by their cobblestone morphology and strong positive immunoreactivity to von Willebrand factor. All cells were incubated in an atmosphere of 5% CO2 at 37°C.

RNA isolation and RT-PCR

Total RNA extraction was performed as the introduction of TRIzol® (Invitrogen, USA) described. Total RNA (1 μg) was reverse transcribed into single-stranded cDNA using Superscript II reverse transcriptase (Life Technologies, USA) according to the manufacturer's instructions. Amplification of VEGF-D cDNA and β-actin cDNA as internal control in each reaction was carried out by PCR with the following primer pairs described previously.20 The primer sequences of VEGF-D were as follows: 5′-GC AAG CTT ATG TAT GGA GGA TGG GGA ATG-3′ (upstream). 5′-CG TCT AGA TCA AGG GTT CTC CTG GCT G-3′ (downstream) [amplifies nucleotides 1077 of mouse VEGF-D, Genbank accession number GI 6753873]. We designed the primers contained Hind III and Xba I sites for subsequent cloning. Reaction mixture was first denatured at 50°C for 30 min and 94°C for 2 min. The PCR condition was 94°C for 30 sec, 60°C for 30 sec and 72°C for 90 sec for 30 cycles, followed by 72°C for 10 min. The resulting PCR product was visualized by ethidium bromide staining after 1% agarose gel electrophoresis.

High-expressing VEGF-D Lewis lung cell lines

The mouse VEGF-D cDNA was amplified from total RNA isolated from mouse embryo tissue, then 1077bp cDNA fragment was inserted into pDrive plasmid (Qiagen) and subcloned into pcDNA3.1 (+) (Invitrogen) eukaryotic expression vector. After the full-length sequence of expected was confirmed by emzymatic digestion and dideoxy sequence, we transfected mouse LL/2 Lewis lung carcinoma cells with the pcDNA3.1 (+) expression vector containing mouse VEGF-D or pcDNA3.1 (vector alone) as control.7, 21 The cell lines were maintained in an atmosphere of 5% CO2 at 37°C in cell culture medium DMEM supplemented with 10% fetal bovine serum. After 48 hr of transfection, cells were trypsinized and replated in cell culture medium containing 800 μg/mL of G418. G418-resistant clones were selected and then expanded for additional studies. Western blot analysis was performed to analysis the expression of recombinated VEGF-D of these cells.

Preparation of liposomal honokiol

The liposomal honokiol was prepared in our laboratory17 and described briefly as follows: PC, cholesterol, PEG4000 and honokiol in weight ratios of 1:0.15:0.24:0.22 were dissolved in 15 mL chloroform/methanol at a ratio of 3:1 (v/v) and evaporated under vacuum in a rotary evaporator until a thin lipid film was formed. The dried lipid films were left overnight and sonicated in 5% glucose solution followed by concentrated and lyophilized. The preparation of empty liposome was the same way as the liposomal honokiol without honokiol addition. Liposomal honokiol and empty liposome gave small multilamellar liposomes in a size range of 130 ± 20 nm and 80 ± 20 nm, respectively.

Cellular proliferation assay

MTT assay were performed to evaluate the drug activity.22 Cells were treated with various concentrations of LH in 96-well culture plates for 24 or 48 hr in a final volume of 200 μL. The control culture was treated with free liposome without addition of honokiol. Then 20 μL of MTT (5 mg/mL in PBS) was added to each well, incubated for an additional 4 hr, the plate was centrifuged at 1,000 rpm for 5 min and then the medium was removed. MTT formazan precipitate was dissolved in 150 μL of DMSO, shaken mechanically for 5 min and then absorbance readings at a wavelength of 570 nm were taken on a spectrophotometer (Molecular Devices, Sunnyvale).

Morphological assay and assessment of apoptosis

Cells (105/well) were plated in triplicate in 6-well plates and cultured for 24 hr. Fresh medium containing indicated concentrations of LH or free liposome were provided for an additional incubation of 12, 24 or 48 hr.

Endothelial cells were then examined under a light microscope and under fluorescence microscopy after staining with propidium dodide.23 Tumor cells were harvested for DNA fragment assessment. The pattern of DNA cleavage was analyzed by agarose gel electrophoresis as described.23 Briefly, cells (3 × 106) were lysed with 0.5 mL lysis buffer containing 5 mM Tris/HCL (pH 8.0), 0.25% Nonidet P-40 and 1 mM EDTA, followed by the addition of RNase A (Sigma) at a final concentration of 200 μg/mL and incubated for 1 hr at 37°C. Cells were then treated with 300 μg proteinase K/mL for an additional hour at 37°C. After addition of 4 μL loading buffer, 20 μL samples in each lane were subjected to electrophoresis on 1.8% agarose at 50 V for 3 hr. DNA was stained with ethidium bromide.

Tube formation assay

The antilymphogenic and antiangiogenic effects of honokiol were analyzed using an in vitro tube-formation assay as described previously.24, 25 HLECs or HUVECs were cultured in the presence or absence of 40 μM liposomal honokiol on polymerized Matrigel (B&D) at 37°C. Standard Matrigel was allowed to polymerize in a 96-well plate, and 1% gelatin (Sigma) was used as control to examine the morphology change. HLECs or HUVECs were seeded at a density of 3 × 104 per well in ECM or M199 medium. After 6 hr, tube formation by endothelial cells was evaluated and photographed.

Western blot analysis

Western blot analysis was performed as described previously.26 Briefly, cells were lysed with protein lysis buffer. Protein concentration was determined by the bradford protein assay. The sample were denatured in sample buffer, and proteins were separated according to molecular weight on a 6 to 12% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel and transferred onto a polyvinylidene difluoride (PVDF) membrane. Membranes were blocked for 1 hr in 5% dried milk (or in 0.5% bovine serum albumin) in TBST at room temperature and probed overnight at 4°C with rotation in primary antibody diluted in “blocking buffer.” Blots were washed thrice for about 15 min with TBST and incubated with a horseradish peroxidase-conjugated species-specific antibody diluted in blocking buffer for 1 hr at room temperature with rotation. After 3 additional washes, blots were developed by a 1-min incubation with enhanced chemiluminescent substrate and exposure to Kodak X-OMAT autoradiographic film (Kodak, Hemel Hempstead, United Kingdom), or in the case of the polyclonal goat anti-mouse VEGF-D antibody using a chemiluminescence's kit (Sigma) and autoradiography.

Establishment of lymph node metastasis models

VEGF-D-LL/2 cells (∼1 × 106) in culture medium were injected into muscle (n = 10) and subcutaneously in hind limb (n = 10) of female C57BL/6 mice. Their metastasis properties were compared with the control groups injected with LL/2 cells transduced only with null-vector (pcDNA-LL/2) or wild type LL/2 cells.

Effect of liposomal honokiol on lymph metastasis

In the antimetastasis treatment, VEGE-D-LL/2 cells (1 × 106) were injected subcutaneously in the hind limb of mice. The treatment initiated the 7th day after implantation when tumors became visible, the mice were randomized into 5 groups and treated with 12.5, 25 and 50 mg/kg/day liposomal honokiol based on the content of honokiol, free liposome and saline alone, respectively. Each mouse received treatment intraperitoneally injection once a day for 28 days. Tumor growth was monitored with calipers every 3 days, and tumor volume was calculated using the formula: volume = 0.52 × length × width2. Animals were sacrificed when the diameter of tumors reached 2.5 cm. All experiments performed on animals were in accordance with the guidelines set by the institute's Animal Care and Use Committee.

Histologic analysis

Sections of paraffin embedded from each group were stained with H&E. The superficial inguinal nodes, post peritoneal lymph node and the lateral axillary lymph node were isolated for histology analysis. In addition, the organs such as lungs, livers, kidneys, etc. were also analyzed by histology. The immunohistochemistry staining was performed according to other report27 and described as follows: tissues of each group were either frozen or fixed in 4% paraformaldehyde (PFA), dehydrated and embedded in paraffin. Goat anti-mouse LYVE-1 and monoclonal rat anti-mouse CD-31 antibodies used for immunohistochemistry were applied to paraffin-embedded material following antigen retrieval (high-pressure). Immunoreactivity was visualized using peroxidase-DAB. Quantification was done as described by Vermeulen et al.28 Microvessel counting was done at 200×. The results regarding angiogenesis and lymphangiogenesis were expressed as the absolute number of the microvessel per high-power field of 5 sections in each tumor.

Quantitative assessment of apoptosis

Terminal deoxynucleotidyl transferase-mediated nick-end labeling staining was done in above tumor species using an in situ cell death detection kit (DeadEnd™ Fluorometric TUNEL System, promega, Madison) following the manufacturer's introduction. It is based on the enzymatic addition of digoxigenin nucleotide to the nicked DNA by terminal deoxynucleotidyl transferase. In tumor sections, 5 equal-sized fields were randomly chosen and analyzed. Density was evaluated in each field, yielding the density of apoptotic cells (apoptosis index).

Evaluation of potential side effects

Potential toxicity treated with liposomal honokiol has been investigated as described.29 Gross measures such as weight loss, life span, behavior and feeding were investigated. Tissues of heart, liver, spleen, lung, kidney, brain and bone marrow were also fixed in 10% neutral buffered formalin solution and embedded in paraffin. Sections of 3–5 μm were stained with H&E, according to the standard procedures.

Statistical analysis

Data was assayed by ANOVA and unpaired student's t-test. Survival curves were constructed according to the Kaplan-Meier method, and the survivals were compared by means of the log-rank test. Differences between means or ranks as appropriate were considered significant when yielding a p < 0.05.

Results

Honokiol inhibits production of VEGF-D-LL/2 and induces apoptosis of tumor cells

High-expressing VEGF-D Lewis lung carcinoma cells (VEGF-D-LL/2) were established by transfected with the pcDNA3.1(+) expression vector containing mouse VEGF-D as mentioned in “Material and methods.” The VEGF-D-LL/2 cells were obtained from 4 independent single clones after G418 selection. Comparably, the level of VEGF-D mRNA was clearly elevated at transfected clones (Fig. 1a). Western blot analysis showed that the positive bands of VEGF-D could be recognized with the medium isolated from the cells transfected VEGF-D, but negative staining from cells transfected null-vectors (pcDNA-LL/2) or the control group (Fig. 1b). This result indicated that a mature and functionally active VEGF-D protein was simultaneously generated in VEGF-D over-expressed cells.

Figure 1.

Honokiol inhibits proliferation and induces apoptosis of both tumor cells and endothelial cells in vitro. (a, b) The expression of protein VEGF-D in the medium of VEGF-D-LL/2, pc-LL/2 and LL/2 cells was analyzed by RT-PCR and western blotting. The positive bands exhibit the mature form of VEGF-D (21 kDa). (c) Growth inhibition in VEGF-D-LL/2, pcDNA-LL/2, HLEC and HUVEC cells by LH was assessed by MTT reduction assay after 48-hr culture. Data represent mean plus or minus the standard deviation (SD) of 3 independent experiments. (d) Agarose gel electrophoretic patterns of DNA isolated from LH (75 μmol/L; 48 hr)-treated and free liposome-treated tumor cells. Lane 1, Marker. Lane 2, free liposome-treated pcDNA-LL/2. Lane 3, LH-treated pcDNA-LL/2. Lane 4, free liposome-treated VEGF-D-LL/2. Lane E, LH-treated VEGF-D-LL/2. (e) Endothelial cells were cultured with LH (50 μmol/L) for 24 or 48 hr (×400). (a) untreated HLEC. (b) HLEC-treated for 24 hr. (c) HLEC-treated for 48 hr. (d) untreated HUVEC. (e) HUVEC-treated for 24 hr. (f) HUVEC-treated for 48 hr. (f) Fluorescence microscopic appearance of PI-stained nuclei of LH (50 μmol/L; 48 hr)-treated or free liposome-treated endothelial cells (×400). (a), free liposome-treated HLEC. (b) LH-treated HLEC. (c) Free liposome-treated HUVEC. (d) LH-treated HUVEC.

To investigate the possible direct effects of honokiol on VEGF-D-induced tumor lymphangiogenesis, VEGF-D-LL/2, HLECs and HUVECs were treated with LH or free liposome at different concentrations for 24 (data not shown) or 48 hr. As shown in Figure 1c, LH potently inhibited the proliferation of HLECs and HUVECs over 48 hr, with 50% inhibition (IC50) of 40 μmol/L and 55 μmol/L, respectively, there was no significant contrast between inhibition of honokiol on HLECs and tumor cells (Supp. Fig. 1). Honokiol at 75 μmol/L obviously reduced VEGF-D-LL/2 production of VEGF-D over 12 hr and also induced obvious apoptosis characteristic as ladder-like pattern of DNA fragments of VEGF-D-LL/2 cells on a more advanced stage of 48 hr (Fig. 1d).

Honokiol interrupts neogenesis of lymphatic and vascular endothelial cells

Neogenesis of HLECs and HUVECs were effectively interrupted after incubated with 50 μmol/L honokiol for 24 and 48 hr (Fig. 1e). Treatment over 48 hr of these 2 primary cultured cells also resulted in some morphological changes characteristic for apoptosis: a brightly red-fluorescent condensed (intact or fragmented) by fluorescence microscopy of PI-stained nuclei, blebbing, reduction of cell volume, condensation of nuclear chromatin, nuclear fragmentation and apoptotic bodies (Fig. 1f).

Honokiol inhibits VEGFR-3-mediated lymphangiogenesis

The binding of VEGF-D and VEGFR-3 promotes lymphatic endothelial cell proliferation and migration via the phosphatidylinositol 3-kinase/AKT and p42/44 MAPK pathways. Therefore, blockade of the VEGFR-3 signaling pathways could efficiently inhibit tumor lymphangiogenesis and metastasis. As shown in Figures 2a and 2b, both Akt and p44/42 MAPK were constitutively activated in tumor cells and endothelial cells.

Figure 2.

Honokol interacts neogenesis signaling pathways and inbibits tube formation of endothelial cells via modulating the growth and survival signaling pathways in tumor cells. Tumor cells (a) and endothelial cells (b) were treated with LH (20 μg/mL) or free liposome without addition of honokiol as control for 12 hr. Whole-cell lysates were then subjected to western blotting to assess phosphorylation and protein expression of AKT (68 kDa), P44/42 MAPK, and β-actin. VEGFR-3 on HLECs and VEGFR-2 on HUVECs were examined as other proteins. The mature form of VEGF-D (21 kDa) expressed in the cell lysates of VEGF-D-LL/2 and pcDNA-LL/2 were also investigated as proteins mentioned above. (c) HLECs were cultured with (b) or without (a) 40 μM LH on Matrigel for 6 hr, and tube formation was assessed. Gelatin was used as control to show the morphology of HLECs cultured with (d) or without (c) 40 μM LH for 6 hr. Bars, 500 μm. (d) HUVEC were cultured with (b) or without (a) 40 μM LH on Matrigel for 6 hr. Bars, 500 μm.

To determine whether honokiol could modulate these activated pathways, both HUVECs and HELCs were treated with increasing amounts (5–25 μg/mL) of LH or free liposome for 12 hr and analyzed for changes in activated AKT and p44/42MAPK, VEGFR-2 and VEGFR-3 by western blotting. As shown in Figures 2a and 2b, honokiol of 20 μg/mL (75 μmol/L) obviously inhibited Akt and P44/42 MAPK phosphorylation both in tumor cells and endothelial cells. Meanwhile, there were significant downregulations of VEGFR-2 expression in HUVECs and VEGFR-3 in cultured HLECs, which were concordant with downregulated Phospho-AKT and Phospho-MAPK (Fig. 2b). However, not as the same effect of honokiol on SVR cells reported by Bai et al.,15 phosphorylation of AKT of HLEC was minor affected compared with MAPK at the same dose (Fig. 2b and Supp. Fig. 2). The reason why the small change in Akt could be purely due to the change of cell surface expression levels of VEGFR-3 is not clear; it may be the fact that MAPK but not AKT changes the expression levels in lymphatic endothelial cells. Moreover, this result may indicate that MAPK signaling possesses the same importance as PI3-kinase signaling pathway in tumor metastasis. Our results suggest for the first time that honokiol could inhibit lymphangiogenesis via blockade of the VEGFR-3 signaling pathways.

Liposomal honokiol inhibits tube formation of HLEC in vitro

Tube formation assay has been reported to analysis angiogenesis24 or lymphangiogenesis25in vitro. Ishitsuka et al.24 used tube formation assay of HUVECs to show that honokiol can inhibit vascular formation in the bone marrow microenvironments. We have observed the same results both on HUVECs (Fig. 2d) and HLECs (Fig. 2c). The antilymphogenic effect of honokiol was analyzed with cultured HLEC cells in vitro. After seeding on Matrigel, HLECs were incubation with LH (40 μM) for 6 hr, the resulting tube-like structures were poorly organized (Fig. 2c,-b). By contrast, in the absence of LH, HLECs spreaded and aligned with each other, formed a rich meshwork of branching anastomosing capillary-like tubules with multicentric junctions within 6 hr (Fig. 2c-a). However, this effect did not result from cell death or cell proliferative suppression as shown on gelatin (Figs. 2c-c and 2c-d).

Lymph node metastasis models can be established by injecting VEGF-D-LL/2 into C576 mice

VEGF-D-LL/2 cells produced tumors at an increased growth rate compared with the control groups injected with LL/2 cells transduced only with null-vector and wild type LL/2 cells both in s.c tumor models (Fig. 3a) and in intramuscular tumor models (Fig. 3b). Tumor weights demonstrated significant differences between the VEGF-D-LL/2 tumors (6.93 ± 1.007 g; n = 10; mean ± s.d.), the pcDNA-LL/2 tumors (3.218 ± 0.437 g; n = 10; p < 0.05) and the wide-type LL/2 tumors (3.18 ± 0.703 g n = 10; p < 0.05). VEGF-D-LL/2 tumors were highly vascularized with extensive edema, consistent with VEGF-D being a potent tumor angiogenesis factor and an inducer of vascular permeability ((Figs. 3c3e). VEGF-D enhanced the abundance of vessels in tumors as assessed by immunohistochemistry for the endothelial cell marker CD31 (Fig. 3f) compared with pcDNA-LL/2 (Fig. 3g) and control LL/2 tumors (Fig. 3h). No statistically significant differences were obtained between pcDNA and control. Quantification of micro vessels had shown statistical differences between VEGF-D groups and control in 2 tumor models (Fig. 3i)

Figure 3.

High-expression of VEGF-D enhances tumor growth and angiogenesis Measurement of intramuscular tumors (a), and s.c. tumors (b) in C576/BL/6 mice (n = 10) were described in Material and methods. Tumor volume in VDGF-D-LL/2 group(•) exhibited significant increase versus the wide-type LL/2 group(▴) and null-vector-LL/2 group(▴), *′p < 0.05 relative to control groups. (ch) Tumor angiogenesis was assessed by immunohistochemical staining with anti-CD31antibody (brown) on frozen sections. (c–e) tumor sections from s.c. tumor models of mice. (f–h) tumor sections from intramuscular tumor models. VEGF-D-LL/2 cells (c, f) displayed increased microvessels compared with pcDNA-LL/2 (d, g) and control LL/2 (e, h), p < 0.05. (i) microvessel counting was done at ×200. Columns, mean of microvessel in s.c. tumor model (black columns) or intramuscular tumor model (white columns) per high-power field; bars, ±SD. Asterisks denote a statistically significant increase compared with control values (**, p < 0.01).

The effect of VEGF-D on tumor-associated lymphatic vessels was analyzed by immumostaining for LYVE-1, which was expressed exclusively on lymphatic endothelia.5 The lymphatic specific marker LYVE-1 revealed highly hyperplastic lymphatic vessels in the VEGF-D-LL/2 tumors (Fig. 4a–4f) both in s.c.tumor and intramuscular tumor models. Quantification of the LYVE-1 positive vessels in the tumors indicated that high expression VEGF-D-LL/2 tumors had a higher density of tumor lymphatic vessels both in s.c tumor models and intramuscular tumor models (125.44 ± 10.24, N = 10, p < 0.01, and 134.33 ± 7.55, n = 10, p < 0.01) than in pcDNA-LL/2 tumors (17.22 ± 4.55, n = 10, p < 0.01) and control LL/2 tumors (14.38 ± 5.34,n = 10, p < 0.01) (Fig. 4g). The density in high expressing VEGF-D-LL/2 tumors was almost 6-fold higher than in control groups. These results are consistent with that VEGF-D is capable of driving lymphangiogenesis.

Figure 4.

Overexpression of VEGF-D induces tumor-associated lymphangiogenesis and lymphatic metastasis. Immunohistochemical staining for LYVE-1 to identify the lymphangiogenesis (af). (ac) Tumor sections from s.c. tumor models of mice. (df) Tumor sections from intramuscular tumor models. (a) LYVE-1 staining of sections from VEGF-D-LL/2 tumors (a, d); sections from the pcDNA- LL/2 tumors (b, e); sections from the control LL/2 tumors (c, f). The numbers of LYVE-1 positive vessels were determined from 5 microscopic fields of the hightest vessel density (g). There is a significant increase of LYVE-1-stained vessels (**, p < 0.01, unpaired t-test) in VEGF-D-LL/2 tumors compared with those of control LL/2 tumors in both s.c. tumor models (black column) and intramuscular tumor models (white column). Tumor-invaded lymph nodes from inguinal of mouse with a VEGF-D-LL/2 tumor (e). In the center of the lymph node, there were plenty of tumor cells with big karyon and abnormal modality. T indicates tumor mass, LF indicates remaining lymphoid follicle. Section stained with H&E. High-powered photograph of metastatic deposits in (i). (j) Percentage of mice with lymph node metastasis in the VEGF-D-LL/2 and control groups. Black column indicates mice beared with s.c. tumors, and white column represents mice with intramuscular tumors. In the VEGF-D-LL/2 group, 8 of the 10 mice burdening s.c. tumors exhibited lymphatic spread, and 10 of the 10 mice beared with intramuscular tumors developed lymph node metastasis. No lymph node metastasis was detected both in the s.c. tumor models and in the intramuscular tumor models resulting by pcDNA-LL/2 cells. Similar to the control group, no lymph node metastasis was found in the s.c. tumor model and only 1 of the 10 mice in the intramuscular tumor model exhibited lymphatic spread.

We next examined the lymph nodes in VEGF-D-LL/2 tumors-bearing animals by histological analysis. The lymph nodes from VEGF-D-LL/2 tumors-bearing animals showed a massive tumor cells infiltrated within the lymph nodes (Figs. 4h, and 4i). Postmortem analysis and serial sections of dubious lymph nodes revealed that VEGF-D-LL/2 had developed metastatic lesions in the lateral axillary or superficial inguinal nodes in 10 of 10 mice compared with 0 of 10 mice for pcDNA-LL/2 tumors and 0 of 10 mice for control LL/2 tumors. Accelerated formation of lymph node metastasis may further induce distant organ metastasis, so we also examined lung metastasis as well as other organs; obvious metastatic tubercles were observed on the lungs from 2 VEGF-D-LL/2 tumor-bearing animals (Supp. Figs. 3a and 3b), no obvious lung or other organ metastatic evidence was found in other animals. The metastatic ratios of mice bearing VEGF-D-LL/2 tumors were 100% (Fig. 4j). These results support the report of Stacker et al.5 that VEGF-D promotes the metastatic spread of tumor cells via the lymphatics.

Liposomal honokiol effectively inhibits the growth of tumors and induces apoptosis

As can be seen in Figure 5a, tumor growth was significantly delayed in both 25 mg/kg/day and 50 mg/kg/day liposomal honokiol-treated groups versus controls (p < 0.05). Animals treated by PBS and empty liposome groups showed a progressive increase in tumor volume and survived 33 days on average. In contrast, 50, 25 and 12.5 mg/(kg/day) liposomal honokiol-treated groups resulted in a significant increase in life span (p < 0.01, by log-rank test; Fig. 5b) and 80%, 60% and 40% mice still survived after 46 day treatment, respectively. The results demonstrated that honokiol was effective to inhibit growth of tumors and prolong lifespan of tumor-bearing mice.

Figure 5.

Induction of the therapeutic antitumor effects of liposomal honokiol. Mice (n = 10 mice/group) were treated with 50 mg/kg/day liposomal honokiol (▴), 25 mg/kg/day liposomal honokiol (▵), 12.5 mg/kg/day liposomal honokiol (▪), free liposome (•) or saline alone (□) intraperitoneally for 4 weeks starting at day 7 after 1 × 106 VEGF-D-LL/2 cells were introduced into mice. Disease progression and outcome were determined as described in materials and methods. (a) Indicate a significant difference in tumor volume between liposomal honokiol-treated groups and control groups. Points, average tumor volume; bars±SD. (b) Remarkable increase of survival time in LH-treated mice was exhibited compared with the control groups (p < 0.01, by log-rank test). The effects of liposomal honokiol on survival time were in a dose-dependent manner. (c, d) TUNEL staining of tumor tissues. Representative sections were from 50 mg/kg/day LH-treated tumors (c), 25 mg/kg/day LH-treated tumors (d), 12.5 mg/kg/day LH-treated tumors (e), blank-liposomes group (f) and control(g). Sequential analysis showed that treatment with liposomal honokiol resulted in significant increment of apoptotic index versus controls. (h) Columns, mean apoptotic index of cancer cells; bars, ±SD; **, p < 0.01 *, p < 0.05 relative to control.

To investigate the effect of liposomal honokiol on apoptosis of tumor cells, tumors resected 4 days after the completion of treatment and subjected to terminal deoxynucleotidyl transferase-mediated nick-end labeling assays for the respective determination of apoptotic index. Liposomal honokiol treatment significantly increased the apoptosis rate of tumor cells and the density of apoptotic cancer cells (Fig. 5c–5e). Data represent the mean apoptotic index ± SDs of cancer cells as percent normalized to apoptotic index of cancer cells (Fig. 5f).

Liposomal honokiol effectively inhibits lymphangiogenesis and angiogenesis

To determine the effect of liposomal honokiol on lymphangiogenesis and angiogenesis in tumors, tumor angiogenesis and lymphangiogenesis were assessed by immunolabeling of CD31 and LYVE-1 in tissue sections. The most highly vascularized area of each tumor was identified on low power and 5 high-powered fields in this area of greatest vessel density. Immunohistochemistry of primary tumors for LYVE-1 indicated that LYVE-1+ vessels were less abundant and smaller in the 50 mg/kg/day liposomal honokiol (Fig. 6a) and 25mg/kg/day liposomal honokiol-treated groups (Fig. 6b) compared with control group (Fig. 6c). The same results were observed in the analysis of angiogenesis. The liposomal honokiol (Figs. 6d and 6e) apparently reduced the number of blood vessels compared with control group (Fig. 6f). Liposomal honokiol (Fig. 6g) treated tumors revealed only occasional and isolated microvessels and apparently reduced the number of lymphatic vessels compared with control groups, including free liposome and saline alone.

Figure 6.

Assessment of lymphatic and blood vessel density in control and liposomal honokiol-treated mice. Tumor lymphangiogenesis and angiogenesis were assessed by immunohistochemical staining with anti-LYVE-1 and anti-CD31 antibody (brown) on paraffin-embedded sections. Staining of tumor sections for LYVE-1 indicated that LYVE-1+ vessels were less abundant and smaller in 50 mg/kg/day LH-treated tumors (a) compared with control group(c). Liposomal honokiol (25 mg/kg/day) also exhibited inhibiting effect (b). (df) immunohistochemical staining with CD31. 50 mg/kg/day LH (d) and 25 mg/kg/day LH (e)-treated tumor revealed only occasional, isolated microvessels, while the section of representative images with well-formed capillaries surrounding nests of tumor cells in control (f). The number of LYVE-1-positive lympatic vessels (black column) and the CD31-positive microvessel (white column) was significantly different between the control group and LH-treated groups in sequential analysis (g). Microvessel counting was done at ×200. Bars, ±SD. **, p < 0.01;* p < 0.05 relative to control.

Liposomal honokiol effectively inhibits the lymphatic metastasis to lymph nodes

Treatment of mice-bearing VEGF-D-LL/2 tumors with liposomal honokiol blocked the metastatic spread to lymph nodes (Table I). Postmortem analysis revealed that control groups had developed metastatic lesions in either the lateral axillary lymph node and/or superficial inguinal nodes compared with the liposomal honokiol-treated groups. One of the 10 mice honokiol-treated 50 mg/(kg/day) exhibited lymphatic spread, whereas 7 of the 10 mice with free liposome treated and 8 of 10 mice saline-treated exhibited lymphatic spread. These results indicate that honokiol is an effective inhibitor to block lymphatic metastasis to lymph nodes.

Table I. Metastatic Spread of Tumors in C576BL/N Mice
TreatmentNumber of mice with primary tumorsNumber of mice with spread to local lymph nodes
50 mg/kg/day LH101 (10%)
25 mg/kg/day LH103 (30%)
12.5 mg/kg/day LH103 (30%)
Free liposome107 (70%)
Saline108 (80%)

Discussion

Metastasis, particularly via the lymphatic system, is a common prognostic factor in the spread of cancer. Some animal studies have shown that the expression of VEGF-D promotes lymph-angiogenesis and metastatic spread of tumor cells via lymphatics in papillary thyroid carcinoma,30 lung cancer,31 gastric cancer32 and pancreatic beta-cell carcinogenesis.33 However, most studies are based on cultured tumor cells, little shows a correlation between high VEGF-D levels and lymph node metastasis in xenograft lung carcinoma. It remains unclear whether expression of VEGF-D in culture accurately represents the in vivo situation.

In this study, we demonstrated that VEGF-D played a remarkable role in upregulating lymphangiogenesis and regional lymph node metastasis. More than 80% of mice harboring VEGF-D-LL/2 Lewis carcinoma had lymph node metastasis both in the s.c. tumor models and in the intramuscular tumor models. Our results indicate that the mouse lymph node metastasis model could be established by transfecting high expression VEGF-D into LL/2 Lewis lung carcinoma cells. The expression of VEGF-D in culture can accurately represent the in vivo situation.

Honokiol has a variety of pharmacological effects such as inducing apoptosis, antiangiogenesis and antisynthesizing of DNA and RNA. However, whether honokiol plays an inhibitive role in lymphangiogenesis remains unclear. Our study has shown for the first time that honokiol possessed the capability of suppressing tumor metastasis at several stages in the process of tumor lymphangiogenesis: (i) honokiol reduces VEGF-D production of tumor cells before cells apoptosis; (ii) honokiol inhibits proliferation and induces apoptosis of tumor cells; (iii) honokiol downregulates the expression of VEGFR-3 on LECs and blockades the VEGFR-3 signaling pathways by interfering with the activation of AKT and p42/44 MAPK; (iv) honokiol interrupts HLECs proliferation and even induces apoptosis; (v) honokiol interrupts and inhibits the tube formation of HLECs. These results are further supported by the tumor-associated lymphangiogenesis and metastasis in Lewis lung carcinoma model induced by VEGF-D in vivo. As mentioned above, VEGFR-3 expression is closely associated with lymph node metastasis in several tumor models. The binding of VEGF-D and VEGFR-3 promotes lymphatic endothelial cell proliferation and migration via the phosphatidylinositol 3-kinase/AKT and p42/44 MAPK pathways. Therefore, blockade of the VEGFR-3 signaling pathways could efficiently inhibit tumor lymphangiogenesis and metastasis. Our results support the hypothesis that the therapeutic effect of honokiol on lung cancer can be attributed to the direct inhibition of angiogenesis and lymphangiogenesis through downregulation of VEGFR-2 and VEGFR-3.

Liposome is a widely used vehicle for administering therapeutic agents including drugs and genes.16, 34–36 As a carrier for antitumor drugs, liposome has been shown to reduce side effects by accumulating preferentially at tumor sites. Our current studies indicate that honokiol-encapsulated liposome have shown improvement of water solubility. The effective dose administration of honokiol was only 50 mg/kg in vivo compared with 250 mg/kg of oral administration of free honokiol.

In conclusion, our results indicate that liposomal honokiol significantly inhibits the tumor-associated lymphangiogenesis and metastasis in Lewis lung carcinoma model induced by VEGF-D. Tumor lymph node metastasis model could be established by injecting overexpressing VEGF-D Lewis lung carcinoma cells into C57BL/6 mice. Tumor growth was significantly delayed and prolong life span was observed after treated by liposomal honokiol in tumor-bearing mice. We have demonstrated for the first time that liposomal honokiol suppresses tumor lymphangiogenesis and lymphatic metastasis by downregulation of VEGF-D in VEGF-D-LL/2 cancer cells and interacting neogenesis and tube formation of lymphatic endothelial cells directly through the VEGFR-3 pathway. Our findings suggest that liposomal honokiol may be a promising agent against lymphangiogenesis and metastasis. In addition, liposomal honokiol is well tolerated by the host animal in therapeutically beneficial doses, making it an attractive candidate for further preclinical testing as an antilymphatic metastasis agent.

Ancillary