Cytochalasin H, an Active Anti-angiogenic Constituent of the Ethanol Extract of Gleditsia sinensis Thorns

Abstract

Angiogenesis, the process of new vessel formation from the pre-existing blood vasculature, is critical for continuous tumor growth and is considered to be a validated antitumor target. The results of our previous study demonstrate the anti-angiogenic potential of an extract of Gleditsia sinensis thorns, which has been traditionally used in Korean medicine to remedy diverse diseases, including tumors. In the present study, we attempted to identify the active anti-angiogenic constituents of the ethanol extract of G. sinensis thorns (EEGS). By virtue of in vitro activity-guided fractionation using human umbilical vein endothelial cells (HUVEC) primary endothelial cells, chromatographic separation, and NMR spectral analyses, we isolated and identified the potent active constituent, cytochalasin H, a biologically active secondary metabolite of fungi. This unexpected active constituent may have originated from the endophytic fungi, Chaetomium globosum, which naturally populate G. sinensis, the identity of which was determined by analysis of fungal community. Cytochalasin H isolated from the EEGS showed in vitro anti-angiogenic activities such as suppressed cell growth and mobility in HUVEC, and inhibited the pro-angiogenic protein-induced formation of new blood vessels in vivo. The anti-angiogenic effect of cytochalasin H was in part due to reduced expression of pro-angiogenic factor, such as endothelin-1. This is the first report regarding the isolation and identification of cytochalasin H, as an active anti-angiogenic constituent of G. sinensis thorns.

A continuous supply of oxygen and nutrients through the blood vessel system is critical for sustained tumor growth. When the size of the tumor is relatively small, tumor cells can take up oxygen and nutrients by diffusion from the environment surrounding the tumor. However, as tumors grow larger, the inner tumor cells are inevitably subjected to nutrient and oxygen limitation without a proper blood supply. In fact, it is known that tumor size cannot exceed >1 mm³ in an avascular state.¹⁾ Since first hypothesized by Folkman in 1971,²⁾ angiogenesis, the process of new vessels sprouting from the pre-existing vasculature, has been considered to be one of the potential targets of anticancer therapy. To date, only bevacizumab (trade name Avastin), a humanized monoclonal antibody targeting vascular endothelial growth factor (VEGF), has been approved by the U.S. Food and Drug Administration (FDA) for the treatment of certain types of cancer, including colon, lung, kidney, and brain cancer.^3,4) However, adverse effects of bevacizumab have been reported, such as hypertension, and the FDA removed the breast cancer indication from bevacizumab in 2010 due to its safety issues and ineffectiveness in breast cancer patients.

Gleditsia sinensis LAM. (Leguminosae) is a perennial shrub that is widely distributed in Eastern Asia, including the Chinese mainland and the Korean peninsula. Different parts of G. sinensis, such as the normal and anomalous fruits, seeds, radix cortex, and thorns, have been prescribed for the treatment of diverse diseases, including suppuration, carbuncle, obesity, thrombosis, and tumors in traditional oriental medicine in general and Korean Medicine (KM) specifically.^5–8) Previous studies have demonstrated that the extract of G. sinensis thorns exerts anticancer effects in vitro and in vivo in diverse cancer models, including gastric,⁵⁾ colon,^9,10) and cervical¹¹⁾ cancers, by suppressing cell cycle progression^5,9,10) as well as inhibiting proliferating cell nuclear antigen (PCNA) and mutant tumor protein 53 (p53) expression.¹¹⁾ The Tang research group demonstrated that the extract of the anomalous fruits of G. sinensis which is rich in saponins, exerts anticancer effects in a panel of solid and non-solid tumor cells by generating reactive oxygen species,¹²⁾ inhibiting proteasome activity,¹³⁾ inhibiting telomerase activity and oncogene expression,¹⁴⁾ and suppressing cyclooxygenase-2 (COX-2) expression.¹⁵⁾

The anti-angiogenic potential of G. sinensis has also been reported by various research groups, including us. Chow et al. demonstrated that the extract of the anomalous fruits of G. sinensis can reduce the expression of VEGF mRNA in a dose-dependent manner in MDA-MB-231 breast cancer and HepG2 hepatoma cell lines, and reduce fibroblast growth factor (FGF)-induced angiogenesis in the chick chorioallantoic membrane (CAM) assay.¹⁶⁾ Dai’s research group demonstrated that the saponin fraction and its active constituent, Gleditsioside B, which were isolated from the anomalous fruits of G. sinensis, inhibit FGF-induced mobility of human umbilical vein endothelial cells (HUVEC) by inhibiting matrix metallopeptidase 2 (MMP2) and focal adhesion kinase (FAK) signaling.^17,18) The results of our previous study demonstrated that the ethanol extract of G. sinensis thorns (EEGS) efficiently inhibits the angiogenic activity of HUVEC and reduces pro-angiogenic factor-induced new vessel formation in a matrigel plug assay.¹⁹⁾ To date, there have been no scientific reports identifying the active anti-angiogenic constituents of G. sinensis thorns. In the present study, we isolated and identified cytochalasin H, as an active anti-angiogenic constituent of the extract of G. sinensis thorns, using activity-guided fractionation.

MATERIALS AND METHODS

Plant Materials

G. sinensis thorns were purchased from the Kwangmyungdang Medicinal Herbs Co. (Batch no. K136201010, Ulsan, Republic of Korea) and identified by Dr. Go Ya Choi, Basic Herbal Medicine Research Group, Herbal Medicine Research Division, Korea Institute of Oriental Medicine, Republic of Korea. A voucher specimen (KIOM-CRC-1) was deposited at the Cancer Research Center, Herbal Medicine Research Division, Korea Institute of Oriental Medicine, Republic of Korea.

Isolation of the Active Constituent

Dried G. sinensis thorns (9.0 kg) were ground and extracted with 80% EtOH (40 L×2 d×3 times) by maceration at room temperature. The extracts were concentrated in vacuo to produce a dried extract (515.3 g). This EtOH extract (364.0 g) was suspended in distilled water and sequentially partitioned with n-hexane and EtOAc to generate the n-hexane- (32.3 g), EtOAc- (94.9 g), and water-soluble (236.0 g) fractions. Activity-guided fractionation was performed to isolate the active anti-angiogenic constituents from the subfractions using a HUVEC-mediated wound healing assay. Schematic depictions of the chromatographic separation procedure and the chemical structure of cytochalasin H are shown in Supplementary Fig. 1.

Chromatography, NMR, and Mass Spectroscopy

Flash chromatography was performed using the Isolera One flash purification system (Biotage, Uppsala, Sweden). Pre-packed SNAP HP-Sil (ϕ39×157 mm, Biotage) and SNAP KP-C18-HS (ϕ30×72 mm, Biotage) cartridges were used for flash chromatography. Dry load cartridges (ϕ71×168 mm or ϕ39×157 mm scales, Biotage) packed with Sephadex LH-20 (Sigma-Aldrich, St. Louis, MO, U.S.A.), Diaion HP-20 (Supelco, Bellefonte, PA, U.S.A.), and ODS-A gel (YMC, Kyoto, Japan) were also used for flash chromatography. The high resolution-electrospray ionization-mass spectrum (HR-ESI-MS) result was measured on a hybrid quadruple orthogonal time-of-flight (Q-TOF) mass spectrometer (SYNAPT G2, Waters, MS Technologies, Manchester, U.K.) coupled with an ESI source. The NMR experiments were conducted on an Advance 500 FT-NMR spectrometer (Bruker, Rheinstetten, Germany). Tetramethylsilane (TMS) was used as an internal standard. Thin layer chromatography (TLC) analysis was performed on Silica gel 60 F₂₅₄ (Merck, Darmstadt, Germany) and RP-18 F₂₅₄s (Merck) plates.

Ultra High Performance Liquid Chromatography (UHPLC) Analysis

UHPLC analysis was performed using an Agilent UHPLC system (1290 Infinity, Waldbronn, Germany) consisting of a diode array detector (G4212A, DAD), a thermostated column compartment (G1316A), a sampler (G4226A), a thermostat (G1330B), and a binary pump VL (G4220B). The system was operated by OpenLAB CDS (ChemStation Edition) software (Agilent). HPLC grade acetonitrile, methanol, and acetic acid (J.T. Baker, Center Valley, PA, U.S.A.) and ultrapure water (Millipore RiOs & Milli-Q-Gradient water purification system, Millipore, Bedford, MA, U.S.A.) were used for the analyses. A ZORBAX Eclipse XDB-C18 analytical column (50×3.0 mm, 1.8 µm, Agilent) was used with a mobile phase consisting of acetonitrile and 0.1% acetic acid in water. The mobile phase gradient elution was programmed as follows: acetonitrile: 5–30% (0–20 min), 30–70% (20–35 min), and 70–100% (35–40 min). The flow rate of the mobile phase was set at 0.3 mL/min. The sample injection volume was set at 2.0 µL. The column temperature was maintained at 40°C and the UV detector was set at 220 and 254 nm. All the sample solutions (2000 µg/mL, 100% methanol) were filtered (Millex-FG 0.2 µm, Millipore) prior to injection.

Cell Culture and Viability

HUVEC obtained from Lonza (Walkersvill, MD, U.S.A.) were grown in EGM-2 endothelial growth medium (Lonza). They were maintained at 37°C in a humidified atmosphere containing 5% CO₂. The cells were used for experiments between passage numbers 5 and 10. A total of 5×10⁴ cells were inoculated into each well of a 24-well tissue culture plate. After 24 h, the cells were exposed to increasing concentrations of the drug and were maintained for 24 h. The numbers of total (viable and dead) cells and dead cells were determined based on cell membrane integrity using an ADAM-MC automatic cell counter (NanoEnTek, Seoul, Republic of Korea), as described previously.¹⁹⁾ The total (viable and dead) and dead cell numbers were determined by mixing equal volumes of the cell suspension with AccuStain T cell lysis solution (total) or AccuStain N (dead) buffer solution, respectively. Cell viability was automatically calculated by ADAM-MC software.

Wound Healing Assay

When the HUVEC had reached approximately 90% confluence in a 24-well tissue culture plate, the cells were scratched with yellow tips. The wounds were photographed using an inverted microscope (Olympus IX71, Tokyo, Japan). After a brief wash with fresh EGM-2 medium, the cells were exposed to fresh EGM-2 containing increasing concentrations of the drug. After 12 h, the wound area was photographed and digitally quantified using MetaMorph image analysis software (Molecular Devices, Downingtown, PA, U.S.A.). The healed area (%) was calculated according to the following formula: healing area (%)=[1−wounded area (t=12 h)/wounded area (t=0 h)]×100.

Tube Formation Assay

The effects of the test drugs on 3-dimensional tube formation by the HUVEC were tested using a Cultrex in vitro angiogenesis assay kit (Trevigen, Gaithersburg, MD, U.S.A.), as described in the manufacturer’s user guide. A total of 1.5×10⁴ cells were inoculated into a 96-well tissue culture plate that had been pre-coated with basement membrane extracts (BME). After 12 h drug treatments, tubes were observed and photographed under the microscope. The degree of tube formation by the HUVEC was performed by quantifying the tube length and branch numbers using image analysis software (MetaMorph).

Endothelin 1 (EDN1) Immunoassay

The levels of EDN1 that were released by the HUVEC into the culture media were quantified using the human EDN1 immunoassay kit (R&D Systems, Minneapolis, MN, U.S.A.). After 24 h of drug treatment, the culture media were cleared by centrifugation and loaded into an immune-microplate that had been pre-coated with EDN1 monoclonal antibodies. Standards consisting of serially diluted recombinant EDN1 were loaded in parallel with the samples. The EDN1 in the supernatants was captured for 1 h at room temperature (rt). After washing, enzyme-linked EDN1 monoclonal antibody was added and incubated for 3 h at rt. After washing, color development was initiated by adding a substrate solution. The absorption was determined at 450 nm using a microplate reader (E_max, Molecular Devices). The concentration of EDN1 in the culture medium was determined by interpolating it from the standard curve.

Semi-quantitative Reverse Transcription Polymerase Chain Reaction (RT-PCR)

Gene expression in HUVEC following cytochalasin H treatment was determined by semi-quantitative RT-PCR. Total RNA was isolated from the HUVEC treated with cytochalasin H for 24 h using the Easy-spin^TM total RNA extraction kit (iNtRON Biotechnology, Seoul, Republic of Korea. Single-stranded cDNAs were synthesized from 1 µg of total RNA using the SuperScript^TM III first-strand synthesis system (Invitrogen, Carlsbad, CA, U.S.A.). PCR was carried out using 100 ng of cDNA as a template and gene specific primer sets. The PCR primers were as follows: for EDN1, 5′-TAT TTG CTC ATG ATT TTC TCT CTG C-3′, 5′-TTG ACC CAA ATG ATG TCC AG-3′; for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5′-AAG GCT GAG AAC GGG AAG-3′, 5′-GGA CTC CAC GAC GTA CTC-3′.

In Vivo Angiogenesis Assay

A directed in vivo angiogenesis assay kit (Trevigen) was used to investigate the effects of the drugs on in vivo angiogenesis induced by pro-angiogenic factors (VEGF and FGF2), as previously described.¹⁹⁾ In brief, angioreactors that had been pre-filled with growth factor-reduced BME with combinations of VEGF, FGF2, and increasing concentrations of the drug were subcutaneously implanted into the dorsal flanks of 6 week old C57BL/6 nude mice. After 12 d, the implanted angioreactors were harvested, and the degree of new vessel formation was determined by FITC-lectin mediated quantification of the vascular endothelial cells that had infiltrated into the angioreactors. The intensities of fluorescence were determined at an excitation 485 nm and an emission 510 nm (SPECTRA MAX GEMINI EM, Molecular Devices). The animal studies were approved by the Institutional Animal Care and Use Committee at the Korea Institute of Oriental Medicine (Protocol #12-058).

Statistics

The results are represented as mean±standard deviation (S.D.). The differences between the control and treated groups were determined using Student’s t-test. Values were considered significant at p<0.05.

RESULTS

Anti-angiogenic Potential of EEGS

We previously revealed that EEGS has potential anti-angiogenic activity both in vitro and in vivo.¹⁹⁾ EEGS efficiently inhibited cellular mobility and 3-dimensional vascular tube formation by HUVEC primary cells, which led to the suppression of vessel formation in the matrigels that had been premixed with pro-angiogenic proteins and implanted in nude mice. In the present study, we tested the in vitro anti-angiogenic potential of G. sinensis thorns which had been purchased from 5 different herbal plant suppliers. EEGS of 5 different suppliers were prepared, and the HUVEC was subjected to increasing concentrations of the extracts. As shown in Fig. 1, all of the extracts inhibited the cellular mobility of HUVEC in a dose-dependent manner. However, the efficacies of the extracts from different suppliers differed slightly. EEGS that had been obtained from suppliers #1 and #2 showed higher anti-angiogenic potential than those obtained from the other suppliers. We chose to use the G. sinensis thorns from supplier #2 (Kwangmyungdang Medicinal Herbs Co.) for the large-scale extraction and activity-guided fractionation to identify the active constituents.

Fig. 1. Inhibitory Effects of the EEGS on HUVEC Mobility Based on the Sources of G. Sinensis Thorns

The areas that were healed by the migration of HUVEC were measured after 12 h of EEGS treatment (0–200 µg/mL). The data are presented as the mean±S.D. of at least three independent experiments. * p<0.05, ** p<0.01, and *** p<0.001 compared with vehicle treatment.

Cytochalasin H, an Active Anti-angiogenic Constituent of G. sinensis Thorns

Activity-guided fractionation to identify the active anti-angiogenic constituents was performed using a HUVEC-mediated wound healing assay. Supplementary Fig. 1 summarizes the chromatographic separation procedures and yields of the active subfractions from each step. The inhibition of cell migration after 12 h of treatment with the active subfractions is also presented at each step. To identify cytochalasin H, the 80% EtOH extract, its solvent fractions (n-hexane-, EtOAc-, and water-soluble), and the activity-guided chromatographic subfractions (F06, F0604, and F060405) from the active EtOAc-soluble fraction were analyzed and monitored at UV 254 and 220 nm using a UHPLC system (Supplementary Fig. 2). No cytochalasin H peak was observed in the 80% EtOH extract and its solvent fractions at both UV 254 and 220 nm (Supplementary Figs. 2A, B). Although the EtOAc-soluble fraction was active, it showed no peak of cytochalasin H at UV 220 nm (Supplementary Fig. 2B), which may be attributable to the fact that its concentrations in the crude extracts were low (i.e., its concentrations were below the limit of detection). The chromatograms of F0604 and F060405, the chromatographic subfractions, showed smaller and larger target peaks (t_R 24.5 min), respectively, at UV 220 nm (Supplementary Fig. 2D), as expected. The UV spectrum of cytochalasin H showed an absorption maximum at 220 nm. Supplementary Fig. 2C showed that the sensitivity of cytochalasin H at UV 254 nm was lower than it was at UV 220 nm. Consequently, these UHPLC data demonstrated that cytochalasin H was successfully isolated and purified from the 80% EtOH extract of the thorns of G. sinensis using activity-guided chromatographic separation. The structure of cytochalasin H was unambiguously determined by NMR (1D- and 2D-NMR) and HRESIMS data²⁰⁾ (Supplementary Fig. 1).

A white powder; ¹H-NMR (500 MHz, CDCl₃) δ: 7.33 (2H, t, J=7.0 Hz, H-28/H-30), 7.25 (1H, t, J=7.5 Hz, H-29), 7.15 (2H, d, J=7.0 Hz, H-27/H-31), 5.87 (1H, dd, J=2.5, 16.5 Hz, H-20), 5.74 (1H, dd, J=9.5, 15.5 Hz, H-13), 5.56 (1H, m, H-21), 5.55 (1H, s, NH), 5.54 (1H, br d, J=16.5 Hz, H-19), 5.41 (1H, m, H-14), 5.36 (1H, s, H-12a), 5.12 (1H, s, H-12b), 3.83 (1H, d, J=10.5 Hz, H-7), 3.26 (1H, m, H-3), 2.95 (1H, t, J=10.5 Hz, H-8), 2.86 (1H, dd, J=4.5, 13.5 Hz, H-10a), 2.79 (1H, m, H-5), 2.65 (1H, dd, J=9.5, 13.5 Hz, H-10b), 2.25 (3H, s, H-25), 2.13 (1H, t, J=4.0 Hz, H-4), 2.05 (1H, m, H-15a), 1.88 (1H, brd, J=14.0 Hz, H-17a), 1.82 (1H, m, H-15b), 1.81 (1H, m, H-16), 1.57 (1H, brd, J=14.5 Hz, H-17b), 1.35 (3H, s, H-23), 1.05 (3H, d, J=6.0 Hz, H-22), 1.00 (3H, d, J=6.5 Hz, H-11). ¹³C-NMR (125 MHz, CDCl₃) δ: 174.4 (C-1, s), 170.3 (C-24, s), 148.1 (C-6, s), 138.9 (C-14, d), 138.3 (C-19, d), 137.6 (C-26, s), 129.2 (C-27/C-31, d), 129.2 (C-28/C-30, d), 127.31 (C-13, d), 127.29 (C-29, d), 126.2 (C-20, d), 114.3 (C-12, t), 77.7 (C-21, d), 74.5 (C-18, s), 69.9 (C-7, d), 54.0 (C-3, d), 53.9 (C-17, t), 52.0 (C-9, s), 50.6 (C-4, d), 47.4 (C-8, d), 45.8 (C-10, t), 42.9 (C-15, t), 33.1 (C-5, d), 31.4 (C-23, q), 28.6 (C-16, d), 26.7 (C-22, q), 21.1 (C-25, q), 14.3 (C-11, q). HR-ESI-MS m/z: 516.2719 [M+Na]⁺ (Calcd for C₃₀H₃₉NO₅Na: 516.2726).

In Vitro Anti-angiogenic Potential of Cytochalasin H

First, we determined the viability of HUVEC in the presence of increasing concentrations of cytochalasin H (0–2000 nmol/L). Cell viabilities were assessed by measuring cytoplasmic membrane integrity as described in Materials and Methods. As shown in Fig. 2A, cytochalasin H efficiently inhibited HUVEC proliferation in a dose-dependent manner. A significant inhibitory effect of cytochalasin H on HUVEC proliferation was observed at concentrations ≥125 nmol/L. However, the anti-proliferative effect of cytochalasin H was not related to cytotoxicity during the 24 h drug treatment. Although a slight decrease in cell viability was observed in the cytochalasin H-treated HUVEC, greater than 80% cell viability was maintained at concentrations up to 2000 nmol/L.

Fig. 2. Anti-angiogenic Activity of Cytochalasin H on HUVEC in Vitro

(A) HUVEC was exposed to increasing concentrations (0–2000 nmol/L) of cytochalasin H, and total cell number and viability were determined after 24 h of drug treatment. (B) The areas healed by migrating HUVEC that had been exposed to increasing concentrations (0–2000 nmol/L) of cytochalasin H were quantified after 12 h of drug treatment. Sulforaphane (Sulfo, 5 µM) was included in the assay as a positive control. (C) Vessel formation indices, such as tube length and branch number, were calculated after 12 h of cytochalasin H treatment (0–2000 nmol/L). Sulforaphane (Sulfo, 5 µM) was included in the assay as a positive control. (D) HUVEC was exposed to increasing concentrations (0–2000 nmol/L) of cytochalasin H and extracellular levels of EDN1 were quantified by immunoassay after 24 h of treatment. The changes in intracellular mRNA following cytochalasin H treatment were also determined by semi-quantitative RT-PCR (inlet). The data are presented as the relative mean±S.D. of at least three independent experiments compared with vehicle treatment. * p<0.05, ** p<0.01, and *** p<0.001.

The anti-angiogenic potential of cytochalasin H in HUVEC was evaluated using an in vitro wound healing assay. In the absence of cytochalasin H, HUVEC residing at the borders of the scratches migrated toward the wounded area and covered it within 12 h (Supplementary Fig. 3). A significant, and concentration-dependent decrease in cell mobility was observed at concentrations ≥125 nmol/L and only 6.3% of the wound was recovered at 2000 nmol/L (Fig. 2B). In addition, we evaluated the effects of cytochalasin H on 3-dimensional vessel formation by HUVEC on matrix-coated surfaces. In the absence of cytochalasin H, the HUVEC constructed vessel tubes by connecting to neighboring cells (Supplementary Fig. 4). However, in the presence of cytochalasin H, the intercellular connections were limited and the HUVEC cells failed to form tubes. To quantify the degree of tube formation, photographs were taken at the end of the experiments under the microscope and the tube length and branch numbers were calculated using image analysis software (Fig. 2C). Both parameters decreased by cytochalasin H in a dose-dependent manner. Statistical analyses revealed that the tube length and branch number were significantly reduced by the treatment with 125 nmol/L of cytochalasin H. In our previous study, proteomic and immunoassay studies revealed that both the intracellular and extracellular expression of the pro-angiogenic protein EDN1 was reduced by treatment with EEGS in a dose-dependent manner.¹⁹⁾ In an effort to verify whether cytochalasin H, an active anti-angiogenic constituent isolated from the EEGS, could also down-regulate EDN1 expression, the extracellular levels of EDN1 that had been released by HUVEC were quantified using an ELISA. As shown in Fig. 2D, EDN1 expression by HUVEC was decreased by cytochalasin H treatment in a dose-dependent manner, and 2000 nmol/L of cytochalasin H inhibited EDN1 expression by 83.8% as compared to vehicle treatment. Down-regulation of EDN1 expression by cytochalasin H was due to decrease in intracellular EDN1 mRNA level by cytochalasin H, which was demonstrated by semi-quantitative RT-PCR (Fig. 2D, inlet).

In Vivo Anti-angiogenic Potential of Cytochalasin H

Because cytochalasin H inhibited in vitro angiogenic properties, such as cell migration and tube formation, we next investigated its effects on the formation of new blood vessels in vivo using a commercially available direct in vivo angiogenesis kit. Vessel formation was not observed in the absence of the pro-angiogenic inducers VEGF/FGF2 (negative control). Massive vessel ingrowth from the open ends of the angioreactors was induced by VEGF/FGF2. Vessel ingrowth was suppressed by cytochalasin H treatment in a dose-dependent manner (Fig. 3, top panel). To quantify vessel ingrowth, the endothelial cells constructing the blood vessels were labeled using FITC-lectin. Anti-angiogenic effects of cytochalasin H were observed at concentrations ≥250 nmol/L, and vessel formation was completely inhibited at concentrations ≥1000 nmol/L (Fig. 3, bottom panel).

Fig. 3. Anti-angiogenic Activity of Cytochalasin H in Vivo

Angioreactors filled with BME that had been premixed with combinations of the angiogenic factors VEGF/FGF2 (inducers) and various concentrations of cytochalasin H (0–1000 nmol/L) were implanted into the dorsal flanks of nude mice. After 12 d, the implanted cylinders were harvested (upper panel) and vascular endothelial cells that had migrated into the BME gels of the bioreactors were quantified using FITC-lectin detection (lower panel). The negative control represents the bioreactors without VEGF/FGF2 inducers and cytochalasin H. Relative angiogenesis was normalized to the mean of the VEGF/FGF2-treated positive control. The data are presented as the mean±S.D. of four cylinders per group. * p<0.05, ** p<0.01, and *** p<0.001 compared with the control groups.

DISCUSSION

In our previous study, we demonstrated that EEGS has a cytostatic and anti-angiogenic potential in HUVEC primary cells in vitro and pro-angiogenic proteins-induced angiogenesis in vivo by suppressing pro-angiogenic proteins such as EDN1 and MMP2.¹⁹⁾ In the present study, we attempted to identify the active constituents of EEGS that display these cytostatic and anti-angiogenic properties. Unexpectedly, activity-guided fractionation and spectral studies identified cytochalasin H, a fungal secondary metabolite, as an active constituent of G. sinensis thorns. To our knowledge, the present study is the first to report cytochalasin H is an active constituent of EEGS.

Cytochalasin H is a member of the cytochalasin family, which are biologically active fungal secondary metabolites. Cytochalasin H was first isolated from crude extracts of endophytic Phomopsis paspali, which were isolated from weevil-damaged pecans.²¹⁾ Since then, cytochalasin H has been isolated from solid cultures of other endophytic fungi, such as Endothia gyrosa,²²⁾ mangrove endophytic fungus Phomopsis sp.,²⁰⁾ and culture fluids of the ascomycete Hymenoscyphus epiphyllus.²³⁾ A group of cytochalasins has also been isolated from Phomopsis, Chaetomium, Zygosporium spp., and Hypoxylon sp., but cytochalasins have not been found in bacteria and plants.^24,25) Therefore, the cytochalasin H that was isolated as an active anti-proliferative and anti-angiogenic constituent of EEGS may have originated from naturally populating endophytic fungi. To demonstrate our hypothesis, we investigated fungal populations in the G. sinensis thorn using high-throughput pyrosequencing (ChunLab, Inc., Seoul, Republic of Korea, http://www.chunlab.com). The dried G. sinensis thorns supplied by two different suppliers (supplier #2 and #3 in Fig. 1) were used for analysis. As a result, we could identify cytochalasin-producing endophytic fungus, Chaetomium globosum, from both samples (Supplementary Fig. 5). The C. globosum has been isolated from diverse medicinal plants, such as Terminalia arjuna, Crataeva magna, Azadirachta indica, and Holarrhena antidysentrica.²⁶⁾ Based on this observation, we here suggest that cytochalasin H isolated from the G. sinensis thorn may originate from the C. globosum which was endophytic associated with G. sinensis thorn. Other research group has suggested that Gleditsioside B, which was isolated from the saponin fractions of the anomalous fruits of G. sinensis, is an active anti-angiogenic constituent in vitro,¹⁷⁾ which differed from the results of the present study. Our unpublished data showed that the in vitro anti-angiogenic potential of G. sinensis thorns was replicated by the extracts of seeds and twigs, but was replicated by the extracts of leafs, barks, and pods of G. sinensis (data not shown). However, the anti-angiogenic potential of G. sinensis thorns was more potent than the extracts of any other parts. We have not yet confirmed whether the active constituents of the seeds and twigs are cytochalasin H.

Until now, cytochalasins had not been under investigation in clinical trials due to their broad spectrum of cytotoxicity. However, computer-assisted phase-contrast microscopy performed by Van Goietsenoven et al. revealed that cytochalasins can be grouped into two categories, i.e., cytotoxic versus cytostatic, indicating that it may be possible to develop novel, non-toxic cytostatic anticancer agents by tuning cytochalasin targets.²⁷⁾ Our results showed that cytochalasin H isolated from G. sinensis thorns can successfully inhibit the angiogenic properties, such as cell growth, mobility, and 3 dimensional-tube formation (≥125 nmol/L), of HUVEC in vitro, and blood vessel formation by mouse endothelial cells in vivo (≥250 nmol/L) without seriously altering cell viability. The effective doses in our study are relatively low doses when compared with the LD₅₀ of 12.5 mg/kg that was observed in day-old cockerels.²¹⁾

The cytostatic mechanism of cytochalasin including cytochalasin H is well known. After permeating cell membrane, cytochalasins bind to actin and inhibit its polymerization.²⁸⁾ Bundle formation of actin filaments is required for vessel tube formation by endothelial cells induced by interacting with extracellular matrix.²⁹⁾ Therefore, cytochalasin can exert anti-angiogenic effect by directly inhibiting cytoskeleton formation. However, our present as well as previous studies¹⁹⁾ suggested another anti-angiogenic mechanism of EEGS and its active constituent, cytochalasin H. We demonstrated that the expression of autocrine pro-angiogenic effector, EDN1, can be significantly inhibited by EEGS¹⁹⁾ or cytochalasin H (Fig. 2D) treatment in HUVEC primary endothelial cells. Angiogenesis by endothelial cells depends on diverse cellular activities, such as cell migration, cell proliferation, and capillary-like tube formation.³⁰⁾ Previous studies reported that EDN1 can stimulate proliferation and migration of endothelial cells^31,32) and modulate the formation of cord-like vascular structures on matrigel.³³⁾ In addition, EDN1 plays a role as an active mitogen for vascular smooth muscle³⁴⁾ and tumor cells.³⁵⁾ In signaling pathways EDN1 induces the expression of potent pro-angiogenic factor VEGF,³³⁾ and activates hypoxia-responsive pathway by stabilizing hypoxia inducible factor-1 alpha under normoxic condition.³⁶⁾ Therefore, EDN1 is a potent pro-angiogenic factor and can be considered as a new therapeutic molecular target for developing anticancer agents. In the present study, we demonstrated that cytochalasin H can significantly reduce the expression of EDN1 from HUVEC. Cytochalasin H down-regulates the expression of EDN1 by decreasing its intracellular mRNA levels (Fig. 2D, inlet), which is consistent with our previous report showing that the EEGS, a source of cytochalasin H used in the present study, can down-regulate the expression of EDN1 at its transcription level.¹⁹⁾ These results suggest that cytochalasin H can be a good candidate of anti-angiogenic anticancer drug targeting EDN1.

A couple of studies reported the application of cytochalasin H as an anticancer drug. Stoeger and Cowan demonstrated that the drug sensitivity of cytochalasin H was correlated in an NCI60 panel of cancer cells with the expression of the kinase suppressor of Ras-1 (KSR1), which modulates the ERK1/2 MAP kinase signaling pathway. They also showed that mouse embryonic fibroblast cells lacking KSR1 (MEF^−/−) are more sensitive to cytochalasin H than wild-type MEFs. Therefore, the EEGS and its active constituent, cytochalasin H, may be useful anticancer drug candidates for use in KSR1 down-regulated tumor types. Furthermore, cytochalasin H efficiently inhibited angiogenesis in both the mouse cornea and Lewis lung tumor in vivo models at the maximum tolerable doses (2.5 mg/kg) without apparent evidence of toxicity.³⁷⁾ Therefore, suitable dose-finding studies evaluating effective drug windows and apparent side effects in vivo, may indicate that the anti-proliferative and anti-angiogenic potential of cytochalasin H and its medicinal plant source, G. sinensis thorns, are good candidates for anticancer drug development.

In conclusion, we report for the first time that cytochalasin H, a biologically active fungal secondary metabolite, is an active anti-angiogenic constituent of the extract of G. sinensis thorns. Although cytochalasin H was isolated as an active constituent of the G. sinensis thorns, further studies should be conducted to understand the pharmaceutical mechanism action of G. sinensis thorns because we could not completely rule out the existence of other active constituents of the G. sinensis thorns. The toxic side effects of cytochalasin H may be reduced by finding a pharmaceutically effective dose window without apparent toxicity, or by developing cytochalasin H analogues that lack anti-actin activity.

Acknowledgment

The authors would like to thank the Korea Basic Science Institute (KBSI) for acquiring the NMR and MS data, and the ChunLab for analyzing fungal community. This research was supported by a Grant from the Korea Institute of Oriental Medicine (KIOM, K13061).

REFERENCES

• 1)  Tosetti F, Ferrari N, De Flora S, Albini A. Angioprevention: angiogenesis is a common and key target for cancer chemopreventive agents. FASEB J., 16, 2–14 (2002).
• 2)  Folkman J. Tumor angiogenesis: therapeutic implications. N. Engl. J. Med., 285, 1182–1186 (1971).
• 3)  Los M, Roodhart JM, Voest EE. Target practice: lessons from phase III trials with bevacizumab and vatalanib in the treatment of advanced colorectal cancer. Oncologist, 12, 443–450 (2007).
• 4)  Lord S, Harris AL. Angiogenesis-still a worthwhile target for breast cancer therapy? Breast Cancer Res., 12 (Suppl. 4), S19 (2010).
• 5)  Lee SJ, Ryu DH, Jang LC, Cho SC, Kim WJ, Moon SK. Suppressive effects of an ethanol extract of Gleditsia sinensis thorns on human SNU-5 gastric cancer cells. Oncol. Rep., 29, 1609–1616 (2013).
• 6)  Dai Y, Chan YP, Chu LM, But PP. Antiallergic and anti-inflammatory properties of the ethanolic extract from Gleditsia sinensis. Biol. Pharm. Bull., 25, 1179–1182 (2002).
• 7)  Lai P, Du JR, Zhang MX, Kuang X, Li YJ, Chen YS, He Y. Aqueous extract of Gleditsia sinensis LAM. fruits improves serum and liver lipid profiles and attenuates atherosclerosis in rabbits fed a high-fat diet. J. Ethnopharmacol., 137, 1061–1066 (2011).
• 8)  Lian XY, Zhang Z. Quantitive analysis of gleditsia saponins in the fruits of Gleditsia sinensis LAM. by high performance liquid chromatography. J. Pharm. Biomed. Anal., 75, 41–46 (2013).
• 9)  Lee SJ, Park K, Ha SD, Kim WJ, Moon SK. Gleditsia sinensis thorn extract inhibits human colon cancer cells: the role of ERK1/2, G2/M-phase cell cycle arrest and p53 expression. Phytother. Res., 24, 1870–1876 (2010).
• 10)  Lee SJ, Cho YH, Kim H, Park K, Park SK, Ha SD, Kim WJ, Moon SK. Inhibitory effects of the ethanol extract of Gleditsia sinensis thorns on human colon cancer HCT116 cells in vitro and in vivo. Oncol. Rep., 22, 1505–1512 (2009).
• 11)  Long L, Geng GX, Li QW. Effect of Gleditsia sinensis stings on growth inhibition and expression of PCNA and p53 in mice bearing uterine cervical carcinoma (U14). Zhongguo Zhong Yao Za Zhi, 31, 150–153 (2006).
• 12)  Chui CH, Lau FY, Chan AS, Cheng GY, Wong RS, Lai KB, Kok SH, Yeung TT, Teo IT, Yau MY, Cheung F, Cheng CH, Tang JC. Gleditsia sinensis fruit extract-induced apoptosis involves changes of reactive oxygen species level, mitochondrial membrane depolarization and caspase 3 activation. Int. J. Mol. Med., 15, 539–543 (2005).
• 13)  Cheung F, Chui CH, Chan AS, Lau FY, Cheng GY, Wong RS, Kok SH, Teo IT, Cheng CH, Tang JC. Inhibition of proteasome activity in Gleditsia sinensis fruit extract-mediated apoptosis on human carcinoma cells. Int. J. Mol. Med., 16, 925–929 (2005).
• 14)  Tang WK, Chui CH, Fatima S, Kok SH, Pak KC, Ou TM, Hui KS, Wong MM, Wong J, Law S, Tsao SW, Lam KY, Beh PS, Srivastava G, Ho KP, Chan AS, Tang JC. Inhibitory effects of Gleditsia sinensis fruit extract on telomerase activity and oncogenic expression in human esophageal squamous cell carcinoma. Int. J. Mol. Med., 19, 953–960 (2007).
• 15)  Pak KC, Lam KY, Law S, Tang JC. The inhibitory effect of Gleditsia sinensis on cyclooxygenase-2 expression in human esophageal squamous cell carcinoma. Int. J. Mol. Med., 23, 121–129 (2009).
• 16)  Chow LM, Chui CH, Tang JC, Lau FY, Yau MY, Cheng GY, Wong RS, Lai PB, Leung TW, Teo IT, Cheung F, Guo D, Chan AS. Anti-angiogenic potential of Gleditsia sinensis fruit extract. Int. J. Mol. Med., 12, 269–273 (2003).
• 17)  Tong B, Lu D, Wei Z, Wang T, Xia Y, Dai Y. Gleditsioside B, a triterpene saponin isolated from the anomalous fruits of Gleditsia sinensis LAM., abrogates bFGF-induced endothelial cell migration through preventing the activation of MMP-2 and FAK via inhibiting ERK and PI3K/AKT signaling pathways. Vascul. Pharmacol., 58, 118–126 (2013).
• 18)  Lu D, Xia Y, Tong B, Zhang C, Pan R, Xu H, Yang X, Dai Y. In vitro anti-angiogenesis effects and active constituents of the saponin fraction from Gleditsia sinensis. Integr. Cancer Ther., 2012, in press.
• 19)  Yi JM, Park JS, Oh SM, Lee J, Kim J, Oh DS, Bang OS, Kim NS. Ethanol extract of Gleditsia sinensis thorn suppresses angiogenesis in vitro and in vivo. BMC Complement. Altern. Med., 12, 243–252 (2012).
• 20)  Tao Y, Zeng X, Mou C, Li J, Cai X, She Z, Zhou S, Lin Y. ¹H and ¹³C NMR assignments of three nitrogen containing compounds from the mangrove endophytic fungus (ZZF08). Magn. Reson. Chem., 46, 501–505 (2008).
• 21)  Wells JM, Cutler HG, Cole RJ. Toxicity and plant growth regulator effects of cytochalasin H isolated from Phomopsis sp. Can. J. Microbiol., 22, 1137–1143 (1976).
• 22)  Xu S, Ge HM, Song YC, Shen Y, Ding H, Tan RX. Cytotoxic cytochalasin metabolites of endophytic Endothia gyrosa. Chem. Biodivers., 6, 739–745 (2009).
• 23)  Thines E, Anke H, Steglich W, Sterner O. New botrydial sesquiterpenoids from Hymenoscyphus epiphyllus. Z. Naturforsch. C, 52, 413–420 (1997).
• 24)  Dombrowski AW, Bills GF, Sabnis G, Koupal LR, Meyer R, Ondeyka JG, Giacobbe RA, Monaghan RL, Lingham RB. L-696,474, a novel cytochalasin as an inhibitor of HIV-1 protease. I. The producing organism and its fermentation. J. Antibiot., 45, 671–678 (1992).
• 25)  Ondeyka J, Hensens OD, Zink D, Ball R, Lingham RB, Bills G, Dombrowski A, Goetz M. L-696,474, a novel cytochalasin as an inhibitor of HIV-1 protease. II. Isolation and structure. J. Antibiot., 45, 679–685 (1992).
• 26)  Kumar S, Kaushik N. Endophytic fungi isolated from oil-seed crop Jatropha curcas produces oil and exhibit antifungal activity. PLoS ONE, 8, e56202 (2013).
• 27)  Van Goietsenoven G, Mathieu V, Andolfi A, Cimmino A, Lefranc F, Kiss R, Evidente A. In vitro growth inhibitory effects of cytochalasins and derivatives in cancer cells. Planta Med., 77, 711–717 (2011).
• 28)  Huang FY, Li YN, Mei WL, Dai HF, Zhou P, Tan GH. Cytochalasin D, a tropical fungal metabolite, inhibits CT26 tumor growth and angiogenesis. Asian Pac. J. Trop. Med., 5, 169–174 (2012).
• 29)  Grant DS, Lelkes PI, Fukuda K, Kleinman HK. Intracellular mechanisms involved in basement membrane induced blood vessel differentiation in vitro. In Vitro Cell. Dev. Biol., 27A, 327–336 (1991).
• 30)  Mabeta P, Pepper MS. A comparative study on the anti-angiogenic effects of DNA-damaging and cytoskeletal-disrupting agents. Angiogenesis, 12, 81–90 (2009).
• 31)  Morbidelli L, Orlando C, Maggi CA, Ledda F, Ziche M. Proliferation and migration of endothelial cells is promoted by endothelins via activation of ETB receptors. Am. J. Physiol., 269, H686–H695 (1995).
• 32)  Goligorsky MS, Budzikowski AS, Tsukahara H, Noiri E. Co-operation between endothelin and nitric oxide in promoting endothelial cell migration and angiogenesis. Clin. Exp. Pharmacol. Physiol., 26, 269–271 (1999).
• 33)  Salani D, Taraboletti G, Rosano L, Di Castro V, Borsotti P, Giavazzi R, Bagnato A. Endothelin-1 induces an angiogenic phenotype in cultured endothelial cells and stimulates neovascularization in vivo. Am. J. Pathol., 157, 1703–1711 (2000).
• 34)  Alberts GF, Peifley KA, Johns A, Kleha JF, Winkles JA. Constitutive endothelin-1 overexpression promotes smooth muscle cell proliferation via an external autocrine loop. J. Biol. Chem., 269, 10112–10118 (1994).
• 35)  Bagnato A, Catt KJ. Endothelins as autocrine regulators of tumor cell growth. Trends Endocrinol. Metab., 9, 378–383 (1998).
• 36)  Spinella F, Rosano L, Del Duca M, Di Castro V, Nicotra MR, Natali PG, Bagnato A. Endothelin-1 inhibits prolyl hydroxylase domain 2 to activate hypoxia-inducible factor-1alpha in melanoma cells. PLoS ONE, 5, e11241 (2010).
• 37)  Udagawa T, Yuan J, Panigrahy D, Chang YH, Shah J, D’Amato RJ. Cytochalasin E, an epoxide containing Aspergillus-derived fungal metabolite, inhibits angiogenesis and tumor growth. J. Pharmacol. Exp. Ther., 294, 421–427 (2000).