Objective: To establish a collagen-induced arthritis (CIA) model of rheumatoid arthritis (RA) through plantar injection of bovine collagen Ⅱ and incomplete Freunds Adjuvant

in order to observe the effect of emodin from Polygoni Cuspidati Rhizoma et Radix(PCRR) on inflammation and neovascularization with corresponding indicators. Method: Wistar rats were randomly divided into 2 groups:the normal group and the model group. The model group was given the same amounts of bovine collagen Ⅱ and incomplete freunds adjuvant and then intradermally injected with 0.4 mL (containing 400 μg of collagen) of solution through plantar region. After 14 days

the same method (0.1 mL) was adopted for the plantar intradermal injection of additionally 0.1 mL of solution to induce inflammation and establish the model 20 days later. After the modeling

the model group was divided into the model subgroup

the emodin from PCRR subgroup and the dexamethasone acetate subgroup

with 10 in each group. Subsequently

the modin from PCRR subgroup was orally given 0.4 g·L-1emodin from PCRR (0.8 mg·kg-1

ig)

while the dexamethasone acetate subgroup was orally given 0.75 g·L-1 dexamethasone acetate (7.5 mg·kg-1

ig). The two drugs were continuously used separately for 20 d

and the plantar thickness

claw X-ray

claws gross pictures

synovial HE staining of knee joint and ELISA were used for detecting rat serum tumor necrosis factor alpha (TNF-α) in serum. The immunohistochemistry was used to detect TNF-α and CD34 expressions of knee joint synovium

and the changes in the two targets were observed in modeled rats. Result: Compared with the normal group

the model group show significant increases in plantar thickness of RA rats (P<0.01) and TNF-α content in rat serum (P<0.01)

after the administration

compared with the model subgroup

the emodin from PCRR subgroup and the dexamethasone acetate subgroup showed notable decreases in plantar thickness and TNF-α content in serum (P<0.05

P<0.01). According to X-ray result

the model subgroup showed discontinuous low-density shadows in plantar region

which were actually notable soft tissue swelling and suggested osteolysis phenomenon

and the normal group

the emodin from PCRR subgroup and the dexamethasone acetate subgroup showed no discontinuous low-density shadows

which suggested no notable soft tissue swelling and osteolysis phenomenon. According to the HE staining

the model subgroup showed plenty of inflammatory cell infiltration and pannus at knee joint synovium

the normal group showed no inflammatory cell infiltration and pannus

and the emodin from PCRR subgroup and the dexamethasone acetate subgroup showed only a little inflammatory cell infiltration and pannus. According to the immunohistochemical result

the model subgroup showed remarkable increases in TNF-α and CD34 expressions at knee joint synovium

whereas the emodin from PCRR subgroup and the dexamethasone acetate subgroup showed significant decreases in TNF-α and CD34 expressions. Conclusion: Emodin from PCRR can intervene RA by regulating two targets TNF-α and CD34 for the inflammation and neovascularization.