LL202 inhibits lipopolysaccharide-induced angiogenesis in vivo and in vitro

Abstract

Excessive or inappropriate angiogenesis occurs in the pathogenesis process of many diseases, such as cancer, wound healing and eye disease, and angiogenesis stimulated by inflammatory factors particularly contributes to the development of inflammation and cancer. In this study, we investigated the inhibitory effect of LL202, a newly synthesized flavonoid, on LPS-induced angiogenesis and further probed the potential molecular mechanisms by detecting MAPK and NF-κB pathways. We found that LL202 inhibited LPS-induced migration, tube formation of human umbilical vein endothelial cells (HUVECs), and microvessel sprouting from rat aortic ring in vivo. The result of the Matrigel plug assay and chicken chorioallantoic membrane (CAM) model also revealed that LL202 could inhibit LPS-induced angiogenesis in vivo. Western blot analysis indicated that LL202 could inhibit the expression of LPS acceptor toll-like receptor 4 (TLR4) and its downstream protein kinases, including the phosphorylation of JNK, p38, ERK, IKK and IκBα. Moreover, LL202 inhibited NF-κB nuclear translocation and its DNA-binding activity. Accordingly, the expression of vascular endothelial growth factor (VEGF), which is a crucial mediator in angiogenesis, was down-regulated at the level of gene transcription. Taken together, LL202 could suppress LPS-induced angiogenesis through the intervention of LPS/TLR4 signaling.

---

Introduction

Angiogenesis is a complicated process in which endothelial cells sprout from existing blood vessels and form new ones.¹ It is critical for many physiological events such as reproductive functions in adults, wound healing and tissue remodeling, while imbalanced angiogenesis will lead to tumor development, cardiovascular diseases, inflammation and so on.² On the other hand, in contrast to the protective mechanism of acute inflammation, chronic inflammation also has a detrimental relationship with angiogenesis, which causes substantial tissue damage.³ In this case, pathological angiogenesis continuously recruits inflammatory cells, and inflammatory cells which secrete proangiogenic factors in turn induce and promote angiogenesis.⁴ This crosstalk between endothelial and immune cells will facilitate carcinogenicity.⁵ Thus, targeting angiogenesis may be an important therapeutic approach for diseases caused by inflammation and come to be a hot spot in drug discovery.

Lipopolysaccharide (LPS) is one of the best-studied immunostimulatory components of bacteria and can induce inflammation.⁶ It has also been shown to have an angiogenic activity in vivo and vitro.⁷ LPS can induce angiogenesis, increase vascular permeability and promote metastatic growth.⁸ Toll-like receptors (TLR) are germline-encoded receptors and their interactions will trigger the expression of proinflammatory and proangiogenic cytokines in different cell types. TLR4 is an important sensor for LPS which recognizes LPS, undergoes oligomerization and then recruits its downstream adaptors. Transforming growth factor-β-activated kinase 1 (TAK1) is activated and TAK1 then stimulates downstream IKK (IκB kinase) and MAPK (mitogen-activated protein kinase). IKKα, IKKβ and IKKγ form a complex and phosphorylate IκB proteins, which results in the degradation of the IκB proteins and the subsequent translocation of the transcription factor NF-κB. NF-κB plays a pivotal role in the production of vascular endothelial growth factor (VEGF) and other proinflammatory cytokines, which promotes angiogenesis. In addition, activation of the downstream MAPK pathways leads to the induction of another transcription factor AP-1, which also controls the expression of VEGF and other factors.⁹ Therefore, LPS/TLR4 signaling bridges inflammation and angiogenesis.

Flavonoids are nearly ubiquitous in plants and act as pharmacologically active constituents in Chinese medicines. Meanwhile, various kinds of flavonoids have been reported to exert anti-tumor properties on an experimental basis, including fisetin, wogonin, luteolin and so on. Thus we reconstructed these compounds for better activity and discovered LL202, a newly-synthesized flavonoid, which bears the three-ring structure of the flavone backbone (Fig. 1A). In a previous study, we have demonstrated that LL202 possesses a strong anti-tumor potential in vitro and in vivo.¹⁰ However, other embedded anti-tumor mechanisms are not still fully understood and warrant further investigation.

Fig. 1 Effect of LL202 on LPS-induced angiogenesis in vitro. (A) The molecular structure of LL202. (B) The HUVECs were treated with various concentrations of LL202 for 4 h and cell viability was determined. (C) The HUVECs were treated with LL202 (4 μM) for 4 h or LPS (1 μg ml⁻¹) for 1 h. The anti-proliferative or apoptosis effects were investigated by an Annexin V/PI staining assay. (D) The HUVECs were treated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h before migration induced by LPS (1 μg ml⁻¹) for 4 h. The effect of LL202 on LPS-stimulated migration of HUVECs was tested. (E) The HUVECs were treated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h before tube formation induced by LPS (1 μg ml⁻¹) for 8 h. The effect of LL202 on LPS-stimulated tube formation of HUVECs was investigated. The comparisons were made relative to the LPS group and significant difference is indicated with *P value <0.05 and **P value <0.01.

In our study, we performed several classic angiogenic experiments to evaluate the pharmacological effect of LL202 on angiogenesis with the stimulation of LPS. Moreover, we investigated LPS/TLR4 signaling and the downstream NF-κB and MAPK pathways, and found that LL202 inhibited the nuclear translocation of NF-κB and eventually the expression of VEGF. These results suggested that LL202 may serve as a candidate in the development of an angiogenesis inhibitor.

Materials and methods

Materials

LL202 was dissolved in 100% dimethyl sulfoxide (DMSO) as a stock solution, stored at −20 °C, and diluted with medium before experiments. The working solution was freshly prepared in the basal medium and the control group was treated with the same amount of DMSO as used in the corresponding experiments. Escherichia coli (055:B5) LPS was supplied by Sigma Chemical Co. (St. Louis, MO). Primary antibodies for TLR4 (H-80), p38, p-p38, p-ERK1/2, JNK, p-JNK, NF-κB, IκB and β-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Primary antibodies for ERK1/2, VEGF, p-IκB, p-IKK and Histone H3 were from Bioworld (St. Louis Park, MN). The primary antibody for IKK was from Cell Signaling (Danvers, MA). IRDye™800 conjugated secondary antibodies were obtained from Rockland Inc. (Philadelphia, PA).

Cell culture

Human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical cord veins by collagenase treatment as described previously.¹¹ The harvested cells were grown in Medium 199 (Gibco, Grand Island, NY) containing endothelial cell growth supplement (ECGS, 30 μg ml⁻¹; Sigma, St. Louis, MO), epidermal growth factor (EGF, 10 ng ml⁻¹; Sigma, St. Louis, MO), 20% fetal bovine serum (FBS, Gibco, Grand Island, NY), 100 U ml⁻¹ penicillin, and 100 μg ml⁻¹ streptomycin, pH 7.4. The cells were incubated in a humidified environment with 5% CO₂ at 37 °C. After 3–5 passages, the HUVECs were collected and incubated in the previous medium with 1% FBS and applied in each independent experiment.

MTT assay

The HUVECs were plated on 96-well plates with 1 × 10⁴ per well in 100 μL culture medium. The cells were incubated overnight and then exposed to LL202 at different concentrations for 4 h. Subsequently, 20 μL per well of MTT solution (5 mg ml⁻¹) was added and the plates were incubated for 4 h. Then the supernatants were removed and 100 μL per well DMSO was added, and the absorbance was recorded at 570 nm (EL800, BIO-TEK Instruments Inc.). Cell viability was determined based on mitochondrial conversion of MTT to formazan. The IC50 value was calculated as the concentration that caused 50% inhibition of cell viability.

Annexin V/PI staining

The HUVECs were harvested after treatment and stained with the Annexin V/PI Cell Apoptosis Detection Kit (KeyGen Biotech, Nanjing, China). Data acquisition and analysis were performed with a Becton–Dickinson FACSCalibur flow cytometer using CellQuest software at excitation/emission 488/530 nm. The cells in early stages of apoptosis were annexin V-positive and PI-negative, whereas the cells in the late stages of apoptosis were both annexin V and PI positive.

Endothelial cell migration assay

Migration of the HUVECs was evaluated by an endothelial cell migration assay with a Transwell chamber (Millicell, Billerico, MA) as described previously.¹² Briefly, cells were trypsinized and suspended at a final concentration of 5 × 10⁵ cells per ml after treatment. The cell suspension was loaded into the upper wells, and the fresh M199 (Medium 199) medium containing 1 μg ml⁻¹ LPS was placed in the lower wells. After 4 h, the cells were fixed with methanol, stained with crystal violet, the bound dye was released with 33% glacial acetic acid, and the optical density (OD) of the solution was measured at 570 nm using an enzyme immunosorbent assay reader.

Tube formation assay

The tube formation assay was performed as described previously.¹³ Briefly, a 96-well plate was coated with 90 μL Matrigel (Becton Dickinson, Bedford, MA), which was allowed to solidify and polymerize at 37 °C for 30 min. The pre-treated cells were trypsinized and suspended in medium containing LPS (1 μg ml⁻¹) or VEGF (10 ng ml⁻¹) before being planted onto the Matrigel. After 8 h of incubation, the plate was examined for capillary tube formation under an inverted microscope, photographed, tubular structures were quantified by manual counting of the tube numbers, and five randomly chosen fields were analyzed for each well.

Rat aortic ring assay

The rat aortic ring assay was performed as described previously.¹⁴ The thoracic aorta was dissected from male Sprague Dawley rats (6 weeks old), cut into 1 mm long rings and set in 24-well plates. The clotting media contained M199⁺ (M199 with 100 U ml⁻¹ penicillin and 100 μg ml⁻¹ streptomycin) plus 0.3% fibrinogen and 0.5% ε-amino-n-caproic acid (ACA; Sigma, St. Louis, MO). The growth media consisted of M199⁺ with 20% FBS and 0.5% ACA. The growth media was added to the wells with various concentrations of LPS (0 and 1 μg ml⁻¹), VEGF (0 and 10 ng ml⁻¹) and LL202 (0, 1, 2 and 4 μM). The plates were then stored in an incubator at 37 °C and 5% CO₂. After 7 days, the sprouting microvessels in five randomly chosen fields were counted and photographed under a microscope for each group.

Chicken chorioallantoic membrane (CAM) assay

The CAM assay was carried out according to the reported method.¹⁵ Briefly, fertilized chicken eggs were incubated at 37 °C for 9 days. On day 10, a small hole was punched on the broad side of the egg, and a window was carefully created through the egg shell. Sterilized filter paper disks (5 × 5 mm) saturated with various concentrations of LPS (0 and 1 μg ml⁻¹), VEGF (0 and 20 ng ml⁻¹) and LL202 (3, 6 and 12 ng/CAM) were placed on the CAMs. They were then incubated at 37 °C for another 2 days. After 48 h of incubation, the shells of the fertilized chicken eggs were opened. Then, an appropriate volume of 10% fat emulsion (Intralipose, 10%) was injected into the embryo chorioallantois to observe the density and length of vessels toward the CAM face. Neovascular zones under the filter paper disks were observed and photographed with a digital camera at ×5 magnification.

Matrigel plug assay

The Matrigel plug assay was performed as described previously.¹⁶ Briefly, C57BL/6 mice were injected subcutaneous with 600 μL of Matrigel. The injected Matrigel rapidly formed a single, solid gel plug. The mice were administered subcutaneous with LPS (0 and 1 μg kg⁻¹) and intragastric with LL202 (0 and 10 mg kg⁻¹) on day 2, 4 and 6. On day 7 the skin of the mouse was easily pulled back to expose the Matrigel plug which remained intact and photographed.

Western blot analysis

The pre-treated cells were collected and lysed in lysis buffer (50 mM Tris-Cl, pH 7.6, 150 mM NaCl, 1 mM EDTA, 1% (m/v) NP-40, 0.2 mM PMSF, 0.1 mM NaF and 1 mM DTT), and the lysates were clarified by centrifugation at 4 °C for 15 min at 13 000×g. The concentration of protein in the supernatants was detected using bicinchoninic acid (BCA) assay with a Varioskan multimode microplate spectrophotometer (Thermo, Marietta, OH). Then, equal total amounts of proteins were separated by SDS-PAGE and transferred onto the NC membranes (Millipore, Billerica, MA). The blots were incubated with specific antibodies against the indicated primary antibodies overnight at 4 °C followed by IRDye™800 conjugated secondary antibody for 1 h at 37 °C. Detection was performed by the Odyssey Infrared Imaging System (LI-COR Inc., Lincoln, NE).

Immunofluorescence

The cells were seeded on coverslips in 6-well plates, treated with LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h and then stimulated with LPS (1 μg ml⁻¹) for 1 h. Immunofluorescence was performed according to the method described with modification.¹⁷ The cells were fixed with 4% paraformaldehyde in PBS for 20 min, permeabilized with 0.5% Triton X-100 for 20 min, and blocked with 3% bovine serum albumin (BSA) for 1 h. Samples were incubated overnight at 4 °C with primary antibodies (diluted 1 : 50) directed against NF-κB. After the cells were washed, they were exposed to FITC-conjugated secondary antibodies (1 : 1000; Invitrogen, Carlsbad, CA, M30101, L42001), and then stained with DAPI for 20 min. The samples were observed and images were captured with confocal microscopy (Fluoview FV1000, Olympus, Tokyo, Japan).

Preparation of cytosolic and nuclear extracts

The cells were incubated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h and then stimulated with LPS (1 μg ml⁻¹) for 1 h. Nuclear and cytosolic protein extracts were prepared according to the modified method as described previously.^18,19

Electrophoretic mobility shift assays (EMSA)

The cells were incubated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h and then stimulated with LPS (1 μg ml⁻¹) for 1 h. Nuclear extract preparation was conducted as described as above and an electrophoretic mobility shift assay (EMSA) was performed with a Chemiluminescent EMSA Kit (Beyotime, Nantong, China) following the manufacturer’s protocol. Briefly, nuclear extracts were incubated with biotin-labeled oligonucleotides, 5′-AGT TGA GGG GAC TTT CCC AGG C-3′ and 3′-TCA ACT CCC CTG AAA GGG TCC G-5′ in reaction buffers, for 30 min at room temperature. Samples were run on a 6% polyacrylamide gel, which was transferred into a Nylon membrane and then blocked and washed. Bands were detected by chemiluminescence.

Real-time PCR analysis

The cells were incubated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h and then stimulated with LPS (1 μg ml⁻¹) for 1 h. The mRNA levels of VEGF were determined using the method described previously.²⁰ The primer sets used in the PCR amplification were as follows: VEGF (forward, 5′-GGT GGA CAT CTT CCA GAG TA-3′, reverse, 5′-GGC TTG TCA CAT CTG CAA GTA-3′) and β-actin (forward, 5′-CTG TCC CTG TAT GCC TCT G-3′, reverse, 5′-ATG TCA CGC ACG ATT TCC-3′).

Small interfering RNA transfection

For transfection, the HUVECs were seeded in 6-well plates at approximately 50–70% confluence. Either TLR4 siRNA duplexes (30 nM) or scrambled siRNA was introduced into the cells using the siPORT NeoFX Transfection Agent (AbmionInc., Austin, TX) according to the manufacturer’s recommendations. The cells were exposed to a M199 medium with LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h and stimulated with LPS (1 μg ml⁻¹) for 1 h, then harvested for further experiments.

Statistical analysis

All data in different experimental groups were expressed as the mean ± SEM (standard error of the mean). The data shown in the study were obtained in at least three independent experiments. Statistical analyses were performed using an unpaired, two-tailed Student’s t-test. All comparisons were made relative to LPS-treated groups; p value less than 0.05 was considered as statistically significant.

Results

LL202 inhibition of LPS-induced tumor angiogenesis in vitro and in vivo

MTT assays were performed to ascertain the anti-proliferative effect of LL202 on the HUVECs. As shown in Fig. 1B, after 4 h treatment, LL202 obviously reduced the number of viable cells with an IC50 value of 7.5 μM. It indicated that the concentration we used in the experiment was less toxic. Additionally, it was confirmed by an Annexin V/PI staining assay, which showed that neither 1 h treatment of LPS (1 μg ml⁻¹) nor 4 h of LL202 (4 μM) induced apoptosis of the HUVECs (Fig. 1C). Therefore, these concentrations were used in the following experiments.

Migration of endothelial cells is one of the multiple stages of angiogenesis, thus a Transwell assay was used to evaluate the effect of LL202 on the migration of the HUVECs. As shown in Fig. 1D, when the cells were treated with 0.2, 0.4 and 0.8 μM LL202, we found that LL202 inhibited LPS-stimulated migration of the HUVECs in a concentration-dependent manner. The inhibition rate of the migrated cells was reduced by 11%, 34% and 81%, respectively, compared with the LPS group.

The late stage of angiogenesis requires tube formation of endothelial cells, so we next evaluated the effects of LL202 on LPS-induced tube formation of the HUVECs on a Matrigel substratum. As shown in Fig.1E, LL202 suppressed formation of capillary-like tubes in a concentration-dependent manner, and the number of tubes was much less than seen in the LPS group. Quantitative analysis revealed that treatment with LL202 resulted in 20%, 59% and 80% inhibition of LPS-induced tube formation at 0.2, 0.4 and 0.8 μM, respectively.

Several stages in angiogenesis, including endothelial cell proliferation, migration, and tube formation, can be mimicked in the rat aortic ring model. We utilized the model to evaluate the effect of LL202 on LPS-induced angiogenesis in vitro. The rat aortic ring is an isolated organ and LL202 presented little toxicity on the pathological section in the preliminary histochemical analysis. As shown in Fig. 2A, LL202 (1, 2 and 4 μM) inhibited microvessel growth by 55%, 81% and 90%, respectively. In addition, LL202 suppressed the formation of microvessel outgrowth from explants of rat aorta in a concentration-dependent manner.

Fig. 2 Effect of LL202 on LPS-induced angiogenesis in vitro and in vivo. (A) Rat aortic rings were treated with various concentrations of LL202 (0, 1, 2, and 4 μM) with or without LPS (1 μg ml⁻¹) as indicated. The effect of LL202 on rat aortic ring microvessel sprouting induced by LPS was investigated. (B) LL202 was placed on the exposed CAM with or without LPS. The effect of LL202 on LPS-induced angiogenesis was detected by CAM model in vivo. (C) The effect of LL202 on LPS-induced angiogenesis was tested by the Matrigel plug assay in vivo. The comparisons were made relative to the LPS group and significant difference is indicated with *P value <0.05 and **P value <0.01.

The CAM assay, which provides a unique model for investigating the effect of antiangiogenic agents on the process of new blood vessel formation, is frequently used to determine the extent of angiogenesis in vivo. As shown in Fig. 2B, when LL202 (3, 6, and 12 ng/CAM) was added, fewer and thinner new blood vessels were formed compared with the LPS group. The quantity of vessels was reduced by 8%, 24% and 42%, respectively. It demonstrated that LL202 inhibited the in vivo angiogenesis in a dose-dependent manner.

Matrigel serves as a vehicle for vascularization in mice, so a Matrigel plug assay was used to further investigate whether LL202 directly inhibited LPS-induced angiogenesis in vivo. Matrigel plugs with LPS appeared a dark-red color (Fig. 2C). In contrast, plugs with Matrigel treated with LL202 (10 mg kg⁻¹) and LPS (1 μg kg⁻¹) were pale in their color, indicating less blood vessel formation. Consequently, the result also suggested that LL202 could inhibit LPS-induced angiogenesis in vivo.

LL202 inhibition of the TLR4 and MAPKs signaling pathways

In this part of the study, we performed Western blot analysis to evaluate the molecular mechanism by which LL202 inhibited LPS-induced angiogenesis in HUVECs. As shown in Fig. 3, the protein expression level of TLR4 following treatment with LL202 (0.2, 0.4 and 0.8 μM) was apparently down-regulated (13%, 28% and 48%, respectively) compared with the level of TLR4 protein induced by LPS. Afterwards, MAPK pathways, downstream of LPS/TLR4 signaling, were investigated. We observed that the total steady state protein levels of ERK1/2, JNK and p38 MAPKs remained unchanged, while LL202 decreased LPS-stimulated phosphorylated protein levels of ERK1/2 (5%, 12% and 37%, respectively), JNK (8%, 4% and 68%, respectively) and p38 MAPKs (11%, 20% and 45%, respectively) in a concentration-dependent manner.

Fig. 3 Effect of LL202 on LPS-stimulated activation of TLR4 and MAPKs signaling pathways. The HUVECs were treated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h, and then treated with or without LPS (1 μg ml⁻¹) for 1 h as specified. Western blotting assays were used to examine the expression of TLR4 and the phosphorylation of ERK1/2, JNK and p38 MAPKs. The comparisons were made relative to the LPS group and significant difference is indicated with *P value <0.05 and **P value <0.01.

LL202 inhibition of the IKK/NF-κB pathway

As an important transcription factor, NF-κB regulates the expression of various target genes and the related cell activities such as proliferation, apoptosis, invasion and angiogenesis. The NF-κB pathway is also downstream of LPS/TLR4 signaling, so the inhibitory effect of LL202 on the LPS-activated NF-κB pathway was investigated. As shown in Fig. 4A, when the HUVECs were treated with LL202 (0.2, 0.4 and 0.8 μM), the cytosolic IκB degradation (15%, 67% and 78%, respectively) and IKK phosphorylation (18%, 34% and 60%, respectively) induced by LPS were inhibited in a concentration-dependent manner. Activated IKK phosphorylates IκBα, and NF-κB is released. The NF-κB heterodimer p65 is then translocated to the nucleus and binds to its target promoter. With the treatment of LL202, the nuclear NF-κB protein levels decreased (15%, 54% and 78%, respectively) while the cytoplasmic NF-κB correspondingly increased (150%, 280% and 560%, respectively) compared with the LPS group in a concentration-dependent manner (Fig. 4B). The results of immunofluorescence (Fig. 4C) also confirmed that LL202 (0.8 μM) could inhibit LPS-induced nuclear translocation of NF-κB.

Fig. 4 Effect of LL202 on the LPS-stimulated activation of the NF-κB signaling pathway. The HUVECs were treated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h, and then treated with or without LPS (1 μg ml⁻¹) for 1 h as specified. (A) The effect of LL202 on the protein levels of p-IKK, IKK, p-IκB and IκB was analyzed by Western blot assay. (B) The effect of LL202 on the LPS-induced nuclear translocation of NF-κB was tested by Western blot assay. (C) The effect of LL202 on the nuclear translocation of NF-κB was examined by immunofluorescence assay. The comparisons were made relative to the LPS group and significant difference is indicated with *P value <0.05 and **P value <0.01.

To investigate whether LL202 affected the binding of NF-κB to DNA, an EMSA assay was performed. As shown in Fig. 5A, the EMSA results indicated that LPS could increase the binding activity of exogenous consensus DNA oligonucleotide with NF-κB in nuclear extracts. However, LL202 (0.2, 0.4 and 0.8 μM) decreased the binding in a concentration-dependent manner.

Fig. 5 Effect of LL202 on LPS-induced NF-κB and VEGF signaling. The HUVECs were treated with various concentrations of LL202 (0, 0.2, 0.4, and 0.8 μM) for 4 h, and then treated with or without LPS (1 μg ml⁻¹) for 1 h as specified. (A) EMSA was conducted to investigate the binding activity of exogenous consensus DNA oligonucleotide with NF-κB. (B) The effect of LL202 on the protein expression of VEGF was tested by Western blot assay. (C) The effect of LL202 on the mRNA level of VEGF in the HUVECs was detected by a real-time PCR assay. The comparisons were made relative to the LPS group and significant difference is indicated with *P value <0.05 and **P value <0.01.

LL202 inhibition of NF-κB target protein VEGF induced by LPS in HUVECs

VEGF, a proangiogenic factor that plays an important role in inflammation-induced angiogenesis, is a target gene of NF-κB. Therefore, we next investigated the effect of LL202 on the expression of VEGF induced by LPS in HUVECs. Western blot analysis indicated that LL202 (0.2, 0.4 and 0.8 μM) could apparently reverse the increasing protein level of VEGF (28%, 72% and 84%, respectively) induced by LPS in a concentration-dependent manner (Fig. 5B). Futhermore, real-time PCR was used to investigate the mRNA level of VEGF in the HUVECs. As shown in Fig. 5C, compared with the LPS group, LL202 inhibited mRNA expression of VEGF (19%, 48% and 72%, respectively) in a concentration-dependent manner. The results above suggested that LL202 inhibited the expression of VEGF at the transcriptional level.

The specific effect of LL202 on LPS-induced angiogenesis

In order to investigate whether TLR4 was required in LL202-mediated antiangiogenesis, TLR4 siRNA was further used in this study. Western blot analysis showed that LPS could not stimulate the downstream signaling of TLR4 and LL202 had little effect on either LPS-induced MAPK or NF-κB pathways (Fig. 6A). In the tube formation assay, LPS could not promote the tube formation of the HUVECs and LL202 did not have any effect either (Fig. 6B).

Fig. 6 Effect of LL202 on LPS-induced angiogenesis requiring TLR4 signaling. (A) The HUVECs were transfected with TLR4 siRNA and then treated with LL202 (0.8 μM) for 4 h and LPS (0 and 1 μg ml⁻¹) for 1 h as specified. The protein level of the TLR4 downstream pathways was analyzed by Western blot assay. (B) The HUVECs were transfected with TLR4 siRNA and then treated with LL202 (0.8 μM) for 4 h before tube formation was induced by LPS for 8 h. The effect of LL202 on LPS-stimulated tube formation of the HUVECs was investigated. The comparisons were made relative to the scrambled siRNA group and significant difference is indicated with *P value <0.05 and **P value <0.01.

VEGF is the most important proangiogenic cytokine. We carried out a preliminary investigation on the effects of LL202 on VEGF-induced angiogenesis. The concentration or dosage of VEGF referred to our previous research.²¹ The results of the CAM assay (Fig. 7A), rat aortic ring assay (Fig. 7B) and tube formation assay (Fig. 7C) showed that LL202 had inhibitory effects on VEGF-induced angiogenesis. When TLR4 was knocked down, LL202 had almost no effect on VEGF-induced angiogenesis by the tube formation assay (Fig. 7D).

Fig. 7 Effect of LL202 on VEGF-induced angiogenesis. (A) LL202 was placed on the exposed CAM with or without VEGF (20 ng/CAM). The effect of LL202 on VEGF-induced angiogenesis was detected by CAM model in vivo. (B) Rat aortic rings were treated with LL202 (4 μM) with or without VEGF (10 ng ml⁻¹) as indicated. The effect of LL202 on rat aortic ring microvessel sprouting induced by VEGF was investigated. (C) The HUVECs were treated with LL202 (0.8 μM) for 4 h before tube formation was induced by VEGF (10 ng ml⁻¹) for 8 h. The effect of LL202 on VEGF-stimulated tube formation of the HUVECs was investigated. (D) The HUVECs were transfected with TLR4 siRNA and then treated with LL202 (0.8 μM) for 4 h before tube formation was induced by VEGF for 8 h. The effect of tube formation was investigated. The comparisons were made relative to the VEGF group and significant difference is indicated with *P value <0.05 and **P value <0.01.

Discussion

It seems that angiogenesis and inflammation are independent of each other, because angiogenesis is the outgrowth of new vessels from pre-existing capillaries while inflammation is a host defense mechanism when injuries and pathogenic challenges take place. Recently, it is reported that the processes of inflammation and angiogenesis are interconnected, especially in human pathologies.²² In the process of chronic inflammation, angiogenesis and inflammation promote each other. Newly formed blood vessels enable the continuous recruitment of inflammatory cells, which release a variety of pro-angiogenic cytokines that further promote angiogenesis. Therefore, anti-inflammatory agents such as inhibitors of cyclo-oxygenase can prevent tumorigenesis and tumor development, which involve angiogenesis and inflammation.²³ LPS is usually used as a potent inflammatory stimulus, and it can induce angiogenesis. In this study, we investigate the antiangiogenic effect of LL202 and the model of LPS-induced angiogenesis was adopted.

Angiogenesis is a multistep process. At the initial stage, the pro-angiogenic factors activate the endothelial cells by enhancing their permeability and proliferation. The following stage involves the degradation of the membrane matrix components to promote the migration of endothelial cells. Afterwards, the buildup of lumen is confirmed with the formation of multicellular vessel sprouts. The final stage of angiogenesis is the stabilization of newly formed capillaries.²⁴ In our study, we used several in vivo and in vitro models mimicking the process of angiogenesis to evaluate the antiangiogenic effects. The Transwell assay and tube formation assay were performed to illuminate the effects of LL202 on LPS-induced migration and tube formation of HUVECs in vitro. The results showed that LL202 remarkably suppressed LPS-stimulated migration and tube formation of the HUVECs. The rat aortic ring assay was used to investigate whether LL202 could inhibit microvessel sprouting from rat aorta induced by LPS. Subsequently, we used the CAM model and Matrigel plug assay to illustrate that LL202 inhibited new vessel formation and vascular remodeling in vivo. Our results demonstrated that LL202 could inhibit microvessel outgrowth from rat aorta in vitro and it also had antiangiogenic activity in vivo. These results implied that LL202 could act as a probable antiangiogenic compound in the process of inflammation.

The initial event of angiogenesis begins when angiogenic cytokines are increased to break the balance with antiangiogenic cytokines. Several potent angiogenic factors may influence angiogenesis during the development of chronic inflammation. VEGF is a pro-angiogenic factor secreted by hypoxic and nutrient-deprived tumor cells within the developing solid tumor. It promotes the formation of new blood vessels from preexisting ones, which is important in tumor growth and metastasis.²⁵ As an important pro-inflammatory factor, LPS promotes angiogenesis not only by promoting inflammatory responses, but also by stimulating VEGF production from endothelial cells. Our results suggested that the down-regulation of VEGF expression by LL202 was regulated primarily at the level of gene transcription. It indicated that the regulation of LL202 on LPS-induced angiogenesis was dependent on transcriptional factors. LPS/TLR4-activated NF-κB signaling plays a pivotal role in LPS-induced angiogenesis due to its ability to induce transcription of VEGF genes.²⁶ MAPKs are also important mediators of LPS signaling leading to AP-1 activation in endothelial cells. Therefore, the NF-κB and MAPK pathways play a pivotal role in LPS-induced angiogenesis due to their ability to induce transcription of VEGF genes and inhibition of these two pathways may be the underlying molecular mechanisms.

It has been proposed that TLR4 contributes to inflammation-induced angiogenesis.²⁷ When the HUVECs were pretreated with LL202, the upregulated expressions of TLR4 and phosphorylated JNK, ERK and p38 MAPKs induced by LPS were blocked. This indicated that the inhibitory effects of LL202 on the angiogenesis of HUVECs might derive partly from inhibiting the transduction of MAPK signaling. Moreover, it is also known that LPS initially promotes the production of VEGF, which elicits a wide spectrum of angiogenic responses by activating the transcription factor NF-κB. In our study, we found that LL202 could prevent NF-κB nuclear translocation due to the inhibition of IKK phosphorylation and the subsequent IκB degradation. As a master transcription factor, NF-κB translocates to the nucleus and interacts with DNA to promote the transcription of genes. We studied the effect of LL202 on the nuclear translocation of NF-κB and observed that LL202 inhibited the binding of NF-κB to DNA induced by LPS. This may prevent the NF-κB downstream target gene of VEGF from being activated in angiogenesis. Taken together, we supposed that LL202 can modulate LPS-induced angiogenesis by inhibiting the expression of TLR4, phosphorylation of MAPKs and activation of NF-κB, finally leading to inhibition of LPS-induced VEGF production.

Although LL202 showed some effects in VEGF-induced angiogenesis, it is less effective than that in LPS-induced angiogenesis, which may be due to the strong proangiogenic effect of VEGF. However, when TLR4 was knocked down, the antiangiogenic effects of LL202 were blocked, which suggested that the inhibition of TLR4 signaling was involved in the underlying mechanism of LL202 on angiogenesis. Accordingly, the specific effect of LL202 on TLR4 signaling will be studied in future research.

Although LL202 shows some effects in VEGF-induced angiogenesis, it is less effective than that in LPS-induced angiogenesis. It may be due to the strong proangiogenic effect of VEGF. However, when TLR4 was knocked down, the antiangiogenic effects of LL202 disappeared, which suggested the blockage of TLR4 signaling was involved in the action of LL202 on VEGF-induced angiogenesis. The specific effect of LL202 will be studied in the future research.

In summary, we demonstrated that LL202 inhibited LPS-induced angiogenesis in vitro and in vivo through suppressing MAPK and NF-κB signaling pathways. LL202 may be an inhibitor of angiogenesis as it could suppress LPS-induced VEGF production. Our findings provided new insights into the molecular mechanisms of LL202-induced antiangiogenesis and suggested promising applications of LL202 as a potent tumor angiogenesis inhibitor and that it can be developed into a chemotherapeutic agent in therapeutics for inflammation diseases in the future.

Conflict of interest

None declared.

Abbreviations

DMSO	Dimethyl sulfoxide
HUVECs	Human umbilical vein endothelial cells
LPS	Lipopolysaccharide
NF-κB	Nuclear factor-kappa B
IKK	IκB kinase
VEGF	Vascular endothelial growth factor
TLR4	Toll-like receptor 4
ERK	Extracellular signal-regulated kinases
JNK	c-Jun N-terminal kinase
p38	p38 mitogen-activated protein kinases
MAPK	Mitogen-activated protein kinase
IL-8	Interleukin-8
CAM	Chicken chorioallantoic membrane
EMSA	Electrophoretic mobility shift assay

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 30973556, 91029744 and 81001452), the Program for Changjiang Scholars and Innovative Research Team in University (IRT1193), the Project Program of State Key Laboratory of Natural Medicines, China Pharmaceutical University (no. JKGZ20110, G140042), the National Science & Technology Major Project (no. 2012ZX09304-001) and the Natural Science Foundation of Jiangsu province (no. BK2009297 and no. BK2010432).

References

1. J. Folkman and Y. Shing, J. Biol. Chem., 1992, 267, 10931–10934.
2. A. W. Griffioen and G. Molema, Pharmacol. Rev., 2000, 52, 237–268.
3. Y. W. Kim, X. Z. West and T. V. Byzova, J. Mol. Med., 2013, 91, 323–328.
4. D. M. Noonan, A. De Lerma Barbaro, N. Vannini, L. Mortara and A. Albini, Cancer Metastasis Rev., 2008, 27, 31–40.
5. J. K. Kundu and Y. J. Surh, Free Radical Biol. Med., 2012, 52, 2013–2037.
6. Y. C. Lu, W. C. Yeh and P. S. Ohashi, Cytokine, 2008, 42, 145–151.
7. I. Mattsby-Baltzer, A. Jakobsson, J. Sorbo and K. Norrby, Int. J. Exp. Pathol., 1994, 75, 191–196.
8. J. H. Harmey, C. D. Bucana, W. Lu, A. M. Byrne, S. McDonnell, C. Lynch, D. Bouchier-Hayes and Z. Dong, Int. J. Cancer, 2002, 101, 415–422.
9. K. A. Fitzgerald, E. M. Palsson-McDermott, A. G. Bowie, C. A. Jefferies, A. S. Mansell, G. Brady, E. Brint, A. Dunne, P. Gray, M. T. Harte, D. McMurray, D. E. Smith, J. E. Sims, T. A. Bird and L. A. O’Neill, Nature, 2001, 413, 78–83.
10. L. Tao, R. Fu, X. Wang, J. Yao, Y. Zhou, Q. Dai, Z. Li, N. Lu and W. Wang, Toxicol. Lett., 2014, 228, 1–12.
11. E. A. Jaffe, R. L. Nachman, C. G. Becker and C. R. Minick, J. Clin. Invest., 1973, 52, 2745–2756.
12. Y. Xu, L. Feng, S. Wang, Q. Zhu, Z. Zheng, P. Xiang, B. He and D. Tang, J. Ethnopharmacol., 2011, 137, 359–370.
13. A. W. Ashton, R. Yokota, G. John, S. Zhao, S. O. Suadicani, D. C. Spray and J. A. Ware, J. Biol. Chem., 1999, 274, 35562–35570.
14. A. C. Berger, X. Q. Wang, A. Zalatoris, J. Cenna and J. C. Watson, Microvasc. Res., 2004, 68, 179–187.
15. R. Auerbach, L. Kubai, D. Knighton and J. Folkman, Dev. Biol., 1974, 41, 391–394.
16. E. J. Cui, J. Hwang-Bo, J. H. Park, N. I. Baek, J. Kim, S. G. Hong and I. S. Chung, Biotechnol. Lett., 2013, 35, 1807–1815.
17. C. Li, N. Lu, Q. Qi, F. Li, Y. Ling, Y. Chen, Y. Qin, Z. Li, H. Zhang, Q. You and Q. Guo, Biochem. Pharmacol., 2011, 82, 1873–1883.
18. R. Steel, J. P. Doherty, K. Buzzard, N. Clemons, C. J. Hawkins and R. L. Anderson, J. Biol. Chem., 2004, 279, 51490–51499.
19. M. Rehbein, S. Kindler, S. Horke and D. Richter, Brain Res. Mol. Brain Res., 2000, 79, 192–201.
20. Z. Lu, N. Lu, C. Li, F. Li, K. Zhao, B. Lin and Q. Guo, Toxicol. Lett., 2012, 209, 211–220.
21. X. Song, J. Yao, F. Wang, M. Zhou, Y. Zhou, H. Wang, L. Wei, L. Zhao, Z. Li, N. Lu and Q. Guo, Toxicol. Appl. Pharmacol., 2013, 271, 144–155.
22. S. Konisti, S. Kiriakidis and E. M. Paleolog, Nat. Rev. Rheumatol., 2012, 8, 153–162.
23. S. I. Grivennikov, F. R. Greten and M. Karin, Cell, 2010, 140, 883–899.
24. M. S. Pepper, Arterioscler., Thromb., Vasc. Biol., 2001, 21, 1104–1117.
25. N. Ferrara, Endocr. Rev., 2004, 25, 581–611.
26. M. W. Covert, T. H. Leung, J. E. Gaston and D. Baltimore, Science, 2005, 309, 1854–1857.
27. S. Murad, Front. Immunol., 2014, 5, 313.

---

Footnote

† These authors contributed equally to this work.

---