RESUMO
It is crucial to classify cervical lesions into high-grade squamous intraepithelial lesions (HSILs) and low-grade SILs (LSILs), as LSILs are conservatively treated by observation, based on an expectation of natural regression, whereas HSILs usually require electrosurgical excision. In the present study, peripheral blood gene expression profiles were analyzed to identify transcriptomic biomarkers distinguishing HSILs from LSILs. A total of 102 blood samples were collected from women with cervical SILs (66 HSIL and 36 LSIL) for microarray hybridization. Candidate gene signatures were identified using AdaBoost algorithms, and a predictive model was constructed using logistic regression to differentiate HSILs from LSILs. To correct for possible bias as a result of the limited sample size and to verify the stability of the predictive model, a two-fold cross validation and null set analysis was conducted over 1,000 iterations. The functions of the transcriptomic biomarkers were then analyzed to elucidate the pathogenesis of cervical SIL. A total of 10 transcriptomic genes (STMN3, TRPC4AP, DYRK2, AGK, KIAA0319L, GRPEL1, ZFC3H1, LYL1, ITGB1 and ARHGAP18) were identified. The predictive model based on the 10-gene panel exhibited well-discriminated power. A cross validation process using known disease status exhibited almost the same performance as that of the predictive model, whereas null-set analysis with randomly reassigned disease status exhibited much lower predictive performance for distinguishing HSILs from LSILs. These biomarkers were involved in the 'Rho GTPase cycle', 'mitochondrial protein import', 'oncogenic MAPK signaling', 'integrin cell surface interaction' and 'signaling by BRAF and RAF fusions'. In conclusion, peripheral blood gene expression analysis is a promising method for distinguishing HSILs from LSILs. The present study proposes 10 candidate genes that could be used in the future as diagnostic biomarkers and potential therapeutic targets for cervical SILs. A simple, non-invasive blood test would be clinically useful in the diagnosis and classification of patients with cervical SILs.
RESUMO
BACKGROUND: Acemannan is a bioactive polysaccharides promoting tissue repair. However, the roles of acemannan in skin wound healing and the underlying molecular mechanisms are largely unclear. OBJECTIVE: The goal of this study is to investigate the positive role of acemannan in cutaneous wound healing and its mechanism. METHODS: Mouse skin wound model and skin primary fibroblasts were used to demonstrate the positive effect of acemannan on cutaneous wound healing. The expressions of cell proliferation nuclear antigen ki-67, cyclin D1 and activity of AKT/mTOR signaling were analyzed in acemannan-treated fibroblasts and mice. Rapamycin and AKT inhibitor VIII were used to determine the key role of AKT/mTOR signaling in acemannan-promoting cutaneous wound healing. RESULTS: We found that acemannan significantly accelerated skin wound closure and cell proliferation. Acemannan promoted the expression of cyclin D1 in cultured fibroblasts, which was mediated by AKT/mTOR signal pathway leading to enhanced activity of the eukaryotic translation initiation factor-4F (eIF4F) and increased translation of cyclin D1. In contrast, pharmaceutical blockade of AKT/mTOR signaling by mTOR inhibitor rapamycin or AKT inhibitor VIII abolished acemannan-induced cyclin D1 translation and cell proliferation. In vivo studies confirmed that the activation of AKT/mTOR by acemannan played a key role in wound healing, which could be reversed by rapamycin. CONCLUSION: Acemannan promoted skin wound healing partly through activating AKT/mTOR-mediated protein translation mechanism, which may represent an alternative therapy approach for cutaneous wound.
