PurposeExosomal microRNAs (miRNAs) have been attracting major interest as potential diagnostic biomarkers of cancer. The aim of this study was to characterize the miRNA profiles of serum exosomes and to identify those that are altered in colorectal cancer (CRC). To evaluate their use as diagnostic biomarkers, the relationship between specific exosomal miRNA levels and pathological changes of patients, including disease stage and tumor resection, was examined.Experimental DesignMicroarray analyses of miRNAs in exosome-enriched fractions of serum samples from 88 primary CRC patients and 11 healthy controls were performed. The expression levels of miRNAs in the culture medium of five colon cancer cell lines were also compared with those in the culture medium of a normal colon-derived cell line. The expression profiles of miRNAs that were differentially expressed between CRC and control sample sets were verified using 29 paired samples from post-tumor resection patients. The sensitivities of selected miRNAs as biomarkers of CRC were evaluated and compared with those of known tumor markers (CA19-9 and CEA) using a receiver operating characteristic analysis. The expression levels of selected miRNAs were also validated by quantitative real-time RT-PCR analyses of an independent set of 13 CRC patients.ResultsThe serum exosomal levels of seven miRNAs (let-7a, miR-1229, miR-1246, miR-150, miR-21, miR-223, and miR-23a) were significantly higher in primary CRC patients, even those with early stage disease, than in healthy controls, and were significantly down-regulated after surgical resection of tumors. These miRNAs were also secreted at significantly higher levels by colon cancer cell lines than by a normal colon-derived cell line. The high sensitivities of the seven selected exosomal miRNAs were confirmed by a receiver operating characteristic analysis.ConclusionExosomal miRNA signatures appear to mirror pathological changes of CRC patients and several miRNAs are promising biomarkers for non-invasive diagnosis of the disease.

Cancer cells secrete small membranous extracellular vesicles (EVs) into their microenvironment and circulation. Although their potential as cancer biomarkers has been promising, the identification and quantification of EVs in clinical samples remains challenging. Here we describe a sensitive and rapid analytical technique for profiling circulating EVs directly from blood samples of patients with colorectal cancer. EVs are captured by two types of antibodies and are detected by photosensitizer-beads, which enables us to detect cancer-derived EVs without a purification step. We also show that circulating EVs can be used for detection of colorectal cancer using the antigen CD147, which is embedded in cancer-linked EVs. This work describes a new liquid biopsy technique to sensitively detect disease-specific circulating EVs and provides perspectives in translational medicine from the standpoint of diagnosis and therapy.

Negative feedback is a crucial physiological regulatory mechanism, but no such regulator of angiogenesis has been established. Here we report a novel angiogenesis inhibitor that is induced in endothelial cells (ECs) by angiogenic factors and inhibits angiogenesis in an autocrine manner. We have performed cDNA microarray analysis to survey VEGF-inducible genes in human ECs. We characterized one such gene, KIAA1036, whose function had been uncharacterized. The recombinant protein inhibited migration, proliferation, and network formation by ECs as well as angiogenesis in vivo. This inhibitory effect was selective to ECs, as the protein did not affect the migration of smooth muscle cells or fibroblasts. Specific elimination of the expression of KIAA1036 in ECs restored their responsiveness to a higher concentration of VEGF. The expression of KIAA1036 was selective to ECs, and hypoxia or TNF-alpha abrogated its inducible expression. As this molecule is preferentially expressed in ECs, we designated it "vasohibin." Transfection of Lewis lung carcinoma cells with the vasohibin gene did not affect the proliferation of cancer cells in vitro, but did inhibit tumor growth and tumor angiogenesis in vivo. We propose vasohibin to be an endothelium-derived negative feedback regulator of angiogenesis.

Negative feedback is a crucial physiological regulatory mechanism, but no such regulator of angiogenesis has been established. Here we report a novel angiogenesis inhibitor that is induced in endothelial cells (ECs) by angiogenic factors and inhibits angiogenesis in an autocrine manner. We have performed cDNA microarray analysis to survey VEGF-inducible genes in human ECs. We characterized one such gene, KIAA1036, whose function had been uncharacterized. The recombinant protein inhibited migration, proliferation, and network formation by ECs as well as angiogenesis in vivo. This inhibitory effect was selective to ECs, as the protein did not affect the migration of smooth muscle cells or fibroblasts. Specific elimination of the expression of KIAA1036 in ECs restored their responsiveness to a higher concentration of VEGF. The expression of KIAA1036 was selective to ECs, and hypoxia or TNF-α abrogated its inducible expression. As this molecule is preferentially expressed in ECs, we designated it "vasohibin." Transfection of Lewis lung carcinoma cells with the vasohibin gene did not affect the proliferation of cancer cells in vitro, but did inhibit tumor growth and tumor angiogenesis in vivo. We propose vasohibin to be an endothelium-derived negative feedback regulator of angiogenesis.

Mucopolysaccharidosis II (MPS II) is an X-linked recessive lysosomal storage disease caused by mutations in the iduronate-2-sulfatase (IDS) gene. Since IDS catalyzes the degradation of glycosaminoglycans (GAGs), deficiency in this enzyme leads to accumulation of GAGs in most cells in all tissues and organs, resulting in severe somatic and neurological disorders. Although enzyme replacement therapy with human IDS (hIDS) has been used for the treatment of MPS II, this therapy is not effective for defects in the CNS mainly because the enzyme cannot cross the blood-brain barrier (BBB). Here, we developed a BBB-penetrating fusion protein, JR-141, which consists of an anti-human transferrin receptor (hTfR) antibody and intact hIDS. The TfR-mediated incorporation of JR-141 was confirmed by using human fibroblasts in vitro. When administrated intravenously to hTfR knockin mice or monkeys, JR-141, but not naked hIDS, was detected in the brain. In addition, the intravenous administration of JR-141 reduced the accumulation of GAGs both in the peripheral tissues and in the brain of hTfR knockin mice lacking Ids, an animal model of MPS II. These data provide a proof of concept for the translation of JR-141 to clinical study for the treatment of patients with MPS II with CNS disorders.

Motivated by the problem of background independence of closed string field theory we study geometry on the infinite vector bundle of local fields over the space of conformal field theories (CFT's). With any connection we can associate an excluded domain D for the integral of marginal operators, and an operator one-form ω µ . The pair (D, ω µ ) determines the covariant derivative of any correlator of local fields. We obtain interesting classes of connections in which ω µ 's can be written in terms of CFT data. For these connections we compute their curvatures in terms of four-point correlators, D, and ω µ . Among these connections three are of particular interest. A flat, metric compatible connection Γ, and connections c andc with non-vanishing curvature, with the latter metric compatible. The flat connection cannot be used to do parallel transport over a finite distance. Parallel transport with either c orc, however, allows us to construct a CFT in the state space of another CFT a finite distance away. The construction is given in the form of perturbation theory manifestly free of divergences.

We recently isolated a novel angiogenesis inhibitor, vasohibin-1, and its homologue, vasohibin-2. In this study we characterize the role of these 2 molecules in the regulation of angiogenesis. In a mouse model of subcutaneous angiogenesis, the expression of endogenous vasohibin-1 was low in proliferating ECs at the sprouting front but high in nonproliferating endothelial cells (ECs) in the termination zone. In contrast, endogenous vasohibin-2 was preferentially expressed in mononuclear cells mobilized from bone marrow that infiltrated the sprouting front. When applied exogenously, vasohibin-1 inhibited angiogenesis at the sprouting front where endogenous vasohibin-1 was scarce but did not influence vascularity in the termination zone where endogenous vasohibin-1 was enriched. Exogenous vasohibin-2 prevented the termination of angiogenesis in the termination zone and increased vascularity in this region. Angiogenesis was persistent in the termination zone in the vasohibin-1 knockout mice, whereas angiogenesis was deficient at the sprouting front in the vasohibin-2 knockout mice. Supplementation of deficient proteins normalized the abnormal patterns of angiogenesis in the vasohibin knockout mice. These results indicate that vasohibin-1 is expressed in ECs in the termination zone to halt angiogenesis, whereas vasohibin-2 is expressed in infiltrating mononuclear cells in the sprouting front to promote angiogenesis. (Blood. 2009; 113:4810-4818)

Metastasis is the main cause of cancer mortality for many types of cancer; however, difficulties remain in effectively preventing metastasis. It has been recently and widely reported that cancer-derived extracellular vesicles (EVs) contribute to cancer metastasis. Thus, therapeutic strategies targeting cancerderived EVs hold great promise because of the possibility of EVs driving the cancer microenvironment toward metastasis. Here, we provide a novel strategy for therapeutic antibody treatment to target cancer-derived EVs and inhibit the metastasis of breast cancer in a mouse model, establishing a rationale for further clinical investigation. Treatment with human-specific anti-CD9 or anti-CD63 antibodies significantly decreased metastasis to the lungs, lymph nodes, and thoracic cavity, although no obvious effects on primary xenograft tumor growths were observed. In in vitro and in vivo experiments, the EVs incubated with the targeted antibodies were preferentially internalized by macrophages, suggesting that antibody-tagged cancer-derived EVs would be eliminated by macrophages. Our results suggested that therapeutic antibody administration effectively suppresses EV-triggered metastasis in cancer and that the removal of EVs could be a novel strategy for cancer therapy.