The mechanisms linking ER stress to inflammation are only partly understood

d as repeated cycles of hypoxia and reoxygenation, seems more effective than prolonged hypoxia in enhancing spontaneous metastasis. Recent studies indicate intermittent hypoxia is a key regulator of the interplay between cancer cell and endothelial cell for tumor angiogenesis and growth and resistance to chemo- and radiotherapy. Furthermore, HIF-1 levels seem well maintained during the cycling. To reproduce intermittent hypoxia in cancer, we elected to treat the transduced cells with tetracycline for 3 days and without for 4 days every week for 8 weeks. Interestingly, at the end of the treatment most of these treated cells no longer or barely responded to tetracycline. Specifically, HIF1 expression was essentially lost at protein levels in all cell lines, while various degrees of HIF2 attenuation were observed. However, -gal expression remained inducible. The loss at protein levels correlated with that at transcript levels. To exclude the possibility of slow recovery, we continued to culture these cells for additional weeks in the absence of tetracycline and found no recovery of HIF1 expression. Although the underlying mechanism requires further Birinapant investigation, we reasoned that the loss of 5 / 15 Lasting Effect of HIF-1 on Malignant Progression Fig 2. Intermittent induction resulted in loss of of HIF1 expression. Intermittent induction involves the administration of tetracycline into cell culture each week on day 1 and removal on day 4 each week for a total of 8 weeks. Afterwards, cells were allowed to expand for further analyses and injections. After intermittent induction, different types of cells as indicated were induced again with tetracycline for 2 days and analyzed by Western blotting in reference to those without intermittent induction. Cell proliferation was determined by cell counting after intermittent induction. , p-value < 0.001. doi:10.1371/journal.pone.0125125.g002 HIF1 induction would help us determine the consequence of intermittent induction, rather than de novo expression, of HIF1 on malignant progression. U-2 OS cells acquire tumorigenicity in the absence of continued expression of PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19777456 HIF-1 variant We first observed that the HIF1-induced U-2 OS cells grew twofold faster than the -gal control without further induction, whereas only a modest increase was detected in the HIF2-induced cells. Importantly, the former but not the latter became tumorigenic when injected subcutaneously into the flanks of NSG mice; all 6 injections with the HIF1 cells resulted in tumor formation, whereas none of the HIF2 cells did at the contralateral sites. Of note, none of the -gal controls became tumorigenic. Moreover, the HIF1 tumors grew at an exponential pace. Histological examination confirmed the malignant growth of these tumors, as indicated by hypercellularity and necrosis, increased mitosis, and invasion into the dermis and the striated muscle layers. Similar data were obtained when CD1 nude mice were used. Thus, we conclude that intermittent induction of HIF1, but not HIF2, programs the non-tumorigenic U-2 OS cells for malignant progression independent of its continued expression. 6 / 15 Lasting PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19776277 Effect of HIF-1 on Malignant Progression Fig 3. U-2 OS cells acquired tumorigenicity after intermittent induction of HIF1. Tumor incidence is shown in 6 NSG mice after bilateral, subcutaneous injections of the 8-week HIF1 and HIF2 cells. Only injections of the HIF1 cells produced tumors, as indicated by arrowheads. Scale bar, 1 cm. Tumor volume