Introduction
Cervical cancer is a common malignant tumor (1) with an incidence of up to 1.2–4.5 per 10,000 delivery women. The incidence of cervical cancer has shown an increasing trend in Chinese women (2,3).
Cervical intraepithelial neoplasia (CIN) is a type of precancerous lesion closely related to cervical cancer. With a high incidence of up to 6.5% (4), CIN significantly impacts the development of cervical cancer. Therefore, accurate diagnosis of CIN and better understanding of the pathogenesis of this disease will definitely improve the prevention of cervical cancer (5). However, most studies mainly focused on the treatment of this disease, and studies on its pathogenesis are relatively rare (6,7). In previous years, with the explosion of gene expression data, bioinformatics-based data digging for gene expression profile analysis has become a hot research field (8,9).
In the present study, the bioinformatics method was applied to analyze gene expression data to identify differentially expressed genes in CIN tissue. Our study provide references for further studies on the molecular pathogenesis of CIN.
Discussion
Cervical cancer is a common malignant tumor in women (10). In China, approximately 30,000 deaths and 100,000 new cases are reported annually (11). CIN, the precancerous lesion closely related to cervical cancer, is considered to be of great significance in studies of cervical cancer (12,13). Transition from CIN to cervical cancer may take as long as 10 years (14). Therefore, early diagnosis, on-time follow-up and early treatment of CIN may effectively inhibit the development of cervical cancer (15). The latest report of American Society of Clinical Oncology (ASCO) showed that the patients with CIN were becoming younger, especially for urban residents (16).
In order to investigate the molecular pathogenesis of CIN, we analyzed the differential expression between CIN patients and healthy controls using gene expression profiling data with multiple bioinformatic methods including enrichment analysis and PPI analysis. In the present study, strict inclusion criteria were followed to select the most reliable chips from microarray candidates. The reliability of the results was secured by the use of microarray data from multiple samples, which can reduce the error rate.
Based on GEO public database, we analyzed and integrated the chip data using software package, and the resulting 287 differential genes were further treated for PPI analysis with STRING. PPI analysis with differential genes showed that EGR1 and ESR1 genes are important factors affecting CIN. EGR1 protein is regarded as the protein with the most significant impact on CIN. Van den Brandt et al (17) suggested that EGR1 was closely associated with the development of myopia and non-small cell lung cancer in human. EGR1 is also an important factor affecting the cell dysplasia, dedifferentiation, and synthesis of nucleoli and ribosomes (18). The activation of EGR1 is related to the normal growth and differentiation of cells. However, cell dysplasia and dedifferentiation play an important role in the development of CIN. Therefore, EGR1 can serve as a marker for the diagnosis of CIN. Schiavon et al (19) suggested that ESR1 is a potential risk factor for breast cancer and can be used as a tumor marker for targeted therapy of breast cancer. The biological effects of ESR1 mainly affect estrogen-targeted organs. ESR1 is mainly expressed in cytoplasm of cervical cancer cells, but not in the nucleus, possibly due to the blocked protein translation and modification. However, with the progression of CIN, the expression of ESR1 in epithelial cells is gradually declining, indicating that ESR1 can be used as marker for the early diagnosis of CIN. But further studies are needed to confirm these conclusions. GO enrichment analysis found that differential expression between CIN cancer cells and normal cells was mainly observed in epithelial cell differentiation, epithelial cell development, and epidermal development. Nagaoka et al (20) concluded that the biological processes of ‘epithelial cell differentiation’ and ‘development’ played an equally dominant role in the pathogenesis of breast and lung cancer, making the two the focus of study on the biological process of lung cancer. The specific maintenance of differentiation ability in squamous epithelial cells is an important feature of CIN, indicating the important impact of EGR1 and ESR1 on CIN. KEGG pathway analysis revealed the dominant role of glutathione metabolism and arachidonic acid metabolism pathways in CIN. A previous study carried out by Liu et al (21) suggested that glutathione metabolism was involved in varying aspects of the development of cancers by affecting the rate of cancer progression. Glutathione, which can be found in every cell in the body, plays importance roles in the maintenance of normal immune system function. Glutathione has been used widely as basic ingredient in functional foods due to its function in improving resistance and inhibiting tumorigenesis. Arachidonic acid in the human body can be synthesized by linoleic acid. The metabolism of arachidonic acid can affect cell proliferation rate, which is related to cell dysplasia of CIN. These four signal pathways also play pivotal roles in the progression of other tumors, but the correlation with CIN still has not been reported. The specific mechanism remains to be further explored.
Chip data used in this study are relatively old and sample size was relatively small. Considering that CIN-relevant genes may change with contributing factors or demographic reason (region, and ethnicity), occult genetic difference may exist. We have reduced the avoidable human error to the lowest possible level. Tumor development is difficult to predict. Further studies should focus on the gene and pathway candidates to elucidate the mechanism.
The long-term analysis led to identification that, CIN was closely related to EGR1 and ESR1 genes, epithelial cell differentiation and glutathione metabolism. However, the internal connection of the three factors remains to be explored with further studies. Considering the fact that only few studies on CIN-relevant genes have been reported, better understanding of CIN at the genetic level may significantly benefit the diagnosis, treatment and prognosis of CIN.