Ardisia species are rich in polyphenols, triterpenoid saponins, isocoumarins, quinones and alkylphenols. Ardisia species and their constituents exhibit a wide range of biological activities, indicating that some of these plants could be exploited for the development of novel phytopharmaceuticals . Ardisia japonica and Ardisia sieboldii, for example, were reported to inhibit 5-lipoxygenase [43, 44], and therefore suppress inflammatory activities. Some Ardisia species were also reported to possess cytotoxic activities in vitro[45, 46]. Because inhibition of angiogenesis has the potential to suppress tumour growth and metastases, its inhibition is one of the most promising strategies in the development of novel anti-cancer therapies, and in the treatment of other human diseases associated with angiogenesis. Thus far, however, no-one has reported on the anti-angiogenic properties of any Ardisia species, including Ardisia crispa.
Our preliminary phytochemical analysis of ACRH revealed that it contained significant amounts of triterpenes, flavonoids, and tannins. Although, the phytochemical screening did not reveal the presence of a benzoquinone, it was detected in the HPLC profiles of QRF when compared with the reference AC-2 compound, 2-methoxy-6-undecyl-1,4-benzoquinone. Benzoquinonoid isolated from Ardisia crispa was previously reported to possess anti-tumour and anti-metastatic properties in vitro.
The LD50 of ACRH was 617 mg/kg in the acute toxicity test. The doses of ACRH and QRF used in the experiments to determine their biological activities elicited no apparent behavioural side effects or signs of toxicity, such as convulsions. Nevertheless, a complete toxicity assessment of ACRH is needed to determine the margin of safety between the efficacy and toxicity of Ardisia crispa roots.
Several models are used to quantify angiogenesis and to evaluate the activities of candidate anti-angiogenic agents. These include in vivo models such as chorioallantoic membrane , rabbit cornea  and rat air pouch . In the Miles vascular permeability assay, indomethacin was used as a positive control because it is an established anti-angiogenic agent, based on data obtained in the cornea and tumour . Additionally, indomethacin was reported to strongly suppress VEGF-induced permeability in mice .
VEGF is a key factor in angiogenesis and vascular permeability, being involved in many pathological processes. VEGF-induced permeability in this assay was quantified by prior intravenous injection of Evans blue dye into the mice. Evans blue dye binds to plasma proteins and is a marker for vascular hyperpermeability to macromolecules. Increased permeability is measured by spectrophotometric quantification of the blue dye . In the Miles assay, both ACRH and QRF significantly reduced VEGF-induced permeability. Interestingly, they had a therapeutic effect at the lowest dose tested (10 mg/kg), and that dose was more potent than the higher doses (30 and 100 mg/kg). The results obtained are consistent with previous reports, in which the anti-angiogenic therapies were more effective when administered at a more frequent, or metronomic, low-dose schedule . It is postulated that the benzoquinonoid present in QRF might be responsible for the suppression of VEGF-induced permeability. Interestingly, a similar benzoquinonoid compound present in A. crispa was previously reported to possess anti-metastatic activity .
In this assay, vascular leakiness of the Evans blue dye was observed at sites of intradermal injection of VEGF, indicating increased vascular permeability. VEGF is a potent permeabilizing mediator that is essential for neovascularisation . Suppression of VEGF reduces vascular leakiness and ultimately prevents the migration of plasma proteins, endothelial cells and fibrin rich matrix that are essential for neovascularisation . Because it was previously reported that ACRH possesses anti-inflammatory properties by blocking COX-2 , the mechanism involved in the attenuation of vascular permeability might involve reduced synthesis of COX-2-derived prostanoids, including PGs. Increased vascular permeability was reported to be elicited by PGs through synergistic effects with other mediators . PGs also contribute to VEGF-induced hyperpermeability by targeting vascular endothelial cells . The mechanism by which PGE2 induces angiogenesis in this study is not fully understood. However, we can speculate that ACRH and QRF can halt neovascularisation by suppressing VEGF-induced vascular permeability.
Murine air pouch granuloma is an animal model of inflammation that is characterised by intense angiogenesis during the chronic inflammatory phase with elevated levels of cytokines, such as interleukin (IL)-1β, tumour necrosis factor (TNF) and IL-6, the induction of COX-2, and enhanced PGE2 production . We used this model to assess the regulation of angiogenesis in chronic inflammatory disease. Our results indicate that ACRH and QRF significantly and dose-dependently reduced the VI associated with FCA-induced inflammation, without affecting granuloma size (Tables 2 and 3). The significant suppression of VI elicited by ACRH and QRF indicates there was a significant reduction in the carmine content in blood vessels surrounding the granulomatous tissue. The lowest dose of QRF (10 mg/kg) reduced angiogenesis to a level similar to that achieved by 10 mg/kg indomethacin. Therefore, we consider that lower doses of QRF might show better potency once the major bioactive compound has been isolated.
ACRH and QRF blocked angiogenesis without markedly reducing granuloma growth. This differs from the effects of glucocorticoids in this model, which severely inhibit granuloma formation . However, our results are consistent with those reported by Gilroy et al. , who used the same model to differentiate between COX-1 and COX-2 inhibitors. This explained the reduction in granuloma weight achieved by indomethacin in our study, consistent with the results of aspirin, a selective COX-1 inhibitor, in a prior study . We found that QRF had a greater anti-angiogenic effect in our study, as the effects of 10 mg/kg QRF and 100 mg/kg ACRH were similar. Although the mechanism is not fully understood, we think that ACRH and QRF contain a selective COX-2 inhibitor, considering that our results are consistent with those reported by Gilroy et al. .
It is well established that inflammation promotes angiogenesis through several ways. Inflammatory cells such as macrophages, lymphocytes, mast cells and fibroblasts are capable of stimulating vessel growth . Inflammatory mediators, including PGE1, PGE2, TNF, IL-1, IL-6 and IL-8, in addition to having pro-inflammatory activities, are capable of directly and/or indirectly inducing angiogenesis in vivo, which in turn may stimulate tumourigenesis .
As inflammation plays a crucial role in inducing rapid angiogenesis in the air pouch model, suppressing inflammation may help to reduce angiogenesis. Our present results may support those of a previous report describing the anti-inflammatory activity of Ardisia crispa roots . The primary mode of action of ACRH and QRF might involve anti-inflammatory activities, with their anti-angiogenic potential being secondary. The ability of ACRH and QRF to suppress angiogenesis might be due to their ability to inhibit COX activity. Inflammatory mediators (e.g., PGs) and enzymes (e.g., COX) are involved in rheumatoid arthritis and cancer-induced angiogenesis . Inhibition of COX downregulates the production of inflammatory mediators, and thus reduces the levels of angiogenic mediators . Further studies are needed to confirm the anti-angiogenic activities of ACRH and QRF using a series of models displaying prominent features of angiogenesis, particularly endothelial proliferation, migration and tube formation  to further elucidate the mechanisms underlying the anti-angiogenic effects of ACRH and QRF.