HIF-2α, also known as endothelial PAS domain protein 1 (EPAS1) or member of PAS superfamily 2 (MOP2)[12–15], was the second HIF family member to be identified and belonged to the basic helix-loop-helix (bHLH)/Per-ARNT-Sim (PAS) domain family of transcription factors . It activates gene expression via formation of a dimeric complex with HIF-1β (also called aryl hydrocarbon receptor nuclear translocator, ARNT) and subsequent binding to hypoxia response elements (HREs) within target genes. Among its transcription targets are genes involved in proliferation, metabolism, angiogenesis, differentiation, and metastasis . Numerous immunochemical analyses have demonstrated that HIF-2α was over-expressed in a number of primary and metastatic human cancers, and that the level of expression, either as a result of tumor hypoxia or genetic alterations, is correlated with tumor angiogenesis and patient mortality. High HIF-2α expression has been linked to poor patient outcome in several tumor types [17–21].
In the present study, we studied HIF-2α expression in paraffin-embedded tumor samples using the IHC method with the Abcam Mouse monoclonal HIF-2 alpha antibody [ep190b, ab8365], and the positive stain for HIF-2α is located mainly in the nucleus of cells as the product leaflet of the antibody indicates.
In our study, the expression HIF-2α of in most tumor cells (87.24%) of the breast invasive ductal carcinoma is positive, in which about 1/3 cases present relatively high HIF-2α expression in the breast cancer cells. Furthermore, the expression of HIF-2α is correlated with the histology-grade (p = 0.029) and the expression of Ki67 (p = 0.043). Those patients with high expression of HIF-2α were demonstrated to be more frequently showing high histology-grade and high expression of Ki67, which means that HIF-2α have some correlation with worse biological behavior and clinical aggressiveness.
To the best of our knowledge, this is the first report that the expression of HIF-2α is correlated with histology-grade and the expression of Ki67 in the cohort of breast invasive ductual carcinoma. We think it is of great importance for further research. In general, it is consistent with previous literature that high expression of HIF-2α have a worse prognosis than those with low expression. Because there was no case in clinical IV stage in the 196 patients which were used in the study, it is impossible to evaluate the correlation between HIF-2α expression and cases with distant metastasis.
In the study of Helczynska K et al. , the expression of HIF-2α is located both in nucleus and cytoplasm of the tumor cells by IHC detection and show a significant correlation to incidence of distant recurrence. We think the different antibodies and clinicopathologic variables used in the two studies may contribute the different results between the study of Helczynska K et al. and ours.
ABCG2 (or BCRP), is the second member of the G subfamily of the ATP-binding cassette (ABC) efflux transporter superfamily that has been the subject of intense study.
In our previous study , we found the expression of ABCG2 protein correlated with Her-2 expression, lymph node metastasis and clinical stage in breast invasive ductal carcinoma and ABCG2 could be a novel potential bio-marker which can predict biological behavior, clinical progression, prognosis and chemotherapy effectiveness.
In the present study, we found HIF-2α expression was correlated with ABCG2 expression significantly (p = 0.001) in paraffin-embedded tumor samples using the IHC method. It means that those patients with high expression of HIF-2α were demonstrated to be more frequently showing high immunoreactions with ABCG2, which suggests that HIF-2α may have some correlation with worse chemotherapy effectiveness. The finding is consistent with the study of Cindy M. Martin et al. , in whose report HIF-2α was found being a potent transcriptional regulator of the Abcg2 gene, and HIF-2α bond an evolutionary conserved HIF-2α response element (HRE) in the murine Abcg2 promoter. A hypothesis had been raised by Cindy M. Martin et al. that Abcg2 is a direct downstream target of HIF-2α in the side population (SP) progenitor cell population of the adult heart. But in the area of cancer research, there is no such kind of report. To the best of our knowledge, this is the first report that the expression of HIF-2α is correlated with ABCG2 expression in the cohort of breast invasive ductual carcinoma. Our findings could be another evidence for the above hypothesis. And in the area of the mechanism research of breast cancer MDR, our findings pointed out another possible genesis mechanism that hypoxia could induce the expression of ABCG2 by HIF-2α, which promoted the multidrug resistance of breast cancer cells.
Because of the evaluation of IHC staining and samples selecting of tissue microarray, some errors of our study could be possible. Whether HIF-2a is correlated with the prognosis of breast cancer and whether the above possible regulation pathways is correct in cancer cells need to be explored by further study.