Researches On Collagen Induced Arthritis Rats Of Compound Recipe Based On Text Mining Technology

Aim:To summarize these traditional medicines used for treating rheumatoid arthritis(RA) in traditional Chinese medicine(TCM) by text mining technology, and compose a potential compound recipe for RA, to primarily research pharmacological effects of the compound recipe on rheumatoid arthritis model rats, and to explore its effect on collagen induced arthritis rats'immune-regulate system, and based on these results, the potential for text mining technology on forming TCM compound recipe to treat RA chould be disscussed.Method:Take "rheumatoid arthritis(leifeishi-guanjieyan in Chinese)" as keyword to search SinoMed database (hyperlink:http://sinomed.imicams.ac.cn/index.jsp), get these literature about medicine used for treating RA, then retrieve the information of species and frequency of traditional medicine, after statistics and analysis, we summarized a potential TCM compound recipe for RA. The compound recipe was composed by7herbal medicines, which were Rumulus Ginnamomi(Guizhi), Rhizoma Anemarrhenae(Zhimu), Radix Paeoniae Alba(Baishao), Radix Astragali Mongolici(Huangqi), Radix Angelicae Sinensis(Danggui), Radix Angelicae Biserratae(Duhuo), and Rhizoma et Radix Notopterygii(Qianghuo), after aqueous extration, concentration, drying, there came out dry powder of compound recipe as medicine for animal experiment.60male SD rats were randomly grouped in two teams:normal and Modelling, modelling rats were injected type II collagen in the root of tail to make Collagen Induced Arthritis (CIA) model, after the model was built, model rats then randomly divided into five teams:Model, Dexamethasone, Compound recipe in low dosage, middle dosage and high dosage. The rats in dexamethasone group was given dexamethasone sodium phosphate i.p., rats in compound recipe based on text-mining in low, middle or high groups were oral administrated different concentration of water solution of dry powder, respectively; rats in normal and model were oral administrated with equal volume saline. These drugs were administrated in continuous3weeks and rats'arthritis index in both hind feet were recorded once3days.24hours after last administration, rats'blood were collected, then the concentrations of IL-1β IFN-γ TNF-a and IL-17in serum were detected, and the content of type II antibody IgG, contents of CD4+T and CD8+T cells in spleen were detected by flow cytometry. The ankle joints were collected simultaneously, after paraformaldehyde fixation, decalcification, paraffin-embedded, microtomation, then colored by eosin-hematoxylin, pathological changes were observed in microscope.Results:After administration of the compound recipe based on text mining, CIA rats' arthritis index were not changed significantly when compared with model rats, organ index of spleen and thymus both took the tendency of decline, but the difference was not obvious. The concentrations of IL-1B, TNF-a and IL-17in serum were down-regulated while IFN-y was up-regulated, the content of IgG was also up-regulated, the ratio of CD4+T/CD8+T in spleen cells was down-regulated. For pathological observation:After administration of the compound recipe, pathological changes of model rats were released, these manifests included:proliferation of synovial cells was inhibited, formation of pannus was reduced, infiltration of inflammatory cells was significantly reduced, damage to articular cartilage was slightly reduced as cartilage surface became smoother.Conclusion:Though this compound recipe based on text mining didn't have significant effect on CIA rats'joint swollen, but it could delay speed of pathologic changes, ameliorate pathological extent, and regulate functions of immune system. These results suggested that text mining technology takes some roles and meanings in exploring principles of traditional medicines for RA, but its operation method needs more improvements.