We report, for the first time to our knowledge, non-invasive fluorescent monitoring of HIF-active microenvironments in cancer in living animals using a oxygen-dependent degradation protein probe, POH. Many hypoxia-specific probes have been reported, including positron emission tomography (PET) probes such as 18
. These label hypoxic regions with reductase activity and have been used to image hypoxic tissues in tumors, mice and humans. In contrast, POH is more specific for HIF (+) regions, which are closely associated with the relatively mild hypoxic status of cells and do not overlap with regions labeled by pimonidazole ( and ref. 
), a nitroimidazole compound similar to FMISO, that targets reductases to detect severely hypoxic regions (pO2
≤10 mmHg) 
. Furthermore, in cancers, HIF activity does not always reflect intratumoral hypoxic status. HIFα expression is increased by genetic changes in the components of the PI3K-mTOR growth signal pathways, in oncogenes products such as ras and raf or in the von Hippel-Lindau and p53 tumor suppressor proteins 
. Therefore, there is an urgent need for an imaging probe to detecting HIF (+) cells. Many luciferase reporter genes, which specifically image HIF (+) cells, have been developed 
. However, the use of luciferase reporter genes is limited because endogenous expression of the reporter genes in target cells is required. The current study is the first to demonstrate an exogenous optical probe specific to HIF (+) cells. We present three prerequisites for an optical HIF (+)-specific exogenous protein probe () and demonstrate that POH-N fits these criteria (, , , , ).
Whole-body imaging using bioluminescence and fluorescence can greatly refine our understanding of small animal models of human physiology and disease 
. We took advantage of the multiple optical imaging techniques and evaluated the target specificity of fluorescently labeled POH by using HIF-specific bioluminescence imaging in cancer mouse models (, , , ).
We failed to detect pancreatic cancer after intravenous injection of POH-N probes (Figure S4
), possibly due to preferential uptake of the probe by the liver before reaching the pancreas. However, it is also possible that the pancreas showed poor uptake of the probes or less blood flow to the xenografts.
In our experiments, OH-I was delivered to the tumors in detectable amounts and imaged tumors () probably due to high blood supply to tumors 
. However, its retention time was significantly shorter than POH-I and it did not give clear images of the tumors (), indicating that PTD increases the retention of POH-I by transducing it into tumor cells. POH-I was also delivered to regions with less blood flow, including the hypoxic regions of tumors (), most probably by diffusion. These results further support the idea that PTD allows PTD-ODD fusion proteins to be delivered to regions far from blood vessels in tumors and support previous studies in which hypoxic/HIF (+) cancer cells were eradicated by PTD-ODD-procaspase-3 
POH-N was successfully detect ischemic lesions in mouse models for ischemic diseases such as focal cerebral ischemia (Fujita et al. in preparation) and myocardial infarction (Watanabe et al. in preparation), suggesting that POH could be used to deliver any drug selectively to hypoxic tissue, including the penumbra, which has potential for recovery, and therefore, is a target for medical interventions.
POH was significantly accumulated in tumors (), and had been specifically detected in HIF (+) cells () 3–6 h after administration, although some strong fluorescent signals were also detected in clearance organs, especially in the bladder; this is also observed in clinical PET imaging such as 18F-Fluorodeoxyglucose-PET. Currently, clinical application of optical imaging probes is limited to endoscopic and intraoperative uses, and thus clinical application of POH-N would be limited. However, POH-N would be a useful tool for translational studies using small animals. Moreover, as the HaloTag ligand can be conjugated to a wide range of biomaterials, including radioisotopes and anti-apoptotic agents through an interchangeable labeling system, POH offers sensitive imaging and targeting bioprobes with specificity for HIF (+) cells; POH can be used in a wide range of applications, including as imaging probes for PET, single photon emission computed tomography and MRI, to deliver drugs for treatment of cancer and ischemic diseases, and to analyze the biology of HIF. Further studies are needed to evaluate the tissue specificity and pharmacokinetic characteristics of POH to determine its most appropriate clinical uses, as well as identify the most appropriate sites of administration to avoid unwanted accumulation or rapid excretion, based on the target tissue.