RESUMO
OBJECTIVE: Owing to variations in the exterior appearances of noncancerous diseases in the oral cavity, clinicians may have difficulty diagnosing oral squamous cell carcinoma (OSCC). Tissue biopsy is confirmatory, but invasive. Therefore, reliable tumor markers for OSCC are required. Here, exosomal Alix (exoAlix) levels were measured in serum/salivary samples from patients with OSCC and healthy controls (HCs). METHODS: Fifty-seven patients admitted to Nagoya University Hospital from 2017 through 2019 were enrolled, and serum samples (OSCC, n = 29; HC, n = 21) and/or saliva samples (OSCC, n = 23; HC, n = 20) were collected. Exosomal fractions were isolated using ultracentrifugation. ExoAlix levels were measured using enzyme-linked immunosorbent assay. RESULTS: Serum/salivary exoAlix levels were significantly higher in patients with OSCC than in HCs. Receiver operating characteristic analyses revealed that sensitivity, specificity, positive predictive value, and area under the curve were 0.345, 1.000, 1.000, and 0.685, respectively, for serum exoAlix and 0.348, 1.000, 1.000, and 0.712, respectively, for salivary exoAlix at optimal cut-off values (serum, 0.205; saliva, 0.193). All tested OSCC tissue sections (n = 21) were immuno-reactive for Alix. CONCLUSION: Serum and salivary exoAlix were identified as potential diagnostic OSCC biomarkers. Serum exoAlix was suitable for prediction of therapeutic responses.
Assuntos
Carcinoma de Células Escamosas , Neoplasias de Cabeça e Pescoço , Neoplasias Bucais , Biomarcadores Tumorais , Carcinoma de Células Escamosas/diagnóstico , Humanos , Neoplasias Bucais/diagnóstico , Saliva , Carcinoma de Células Escamosas de Cabeça e PescoçoRESUMO
Exosomes are 50-100-nm-diameter membrane vesicles released from various types of cells. Exosomes retain proteins, mRNAs and miRNAs, which can be transported to surrounding cells. CD109 is a glycosylphosphatidylinositol-anchored glycoprotein, and is released from the cell surface to the culture medium in vitro. Recently, it was reported that secreted CD109 from the cell surface downregulates transforming growth factor-ß signaling in human keratinocytes. In this study, we revealed that CD109 is a component of the exosome in conditioned medium. FLAG-tagged human CD109 (FLAG-CD109) in conditioned medium secreted from HEK293 cells expressing FLAG-CD109 (293/FLAG-CD109) was immunoprecipitated with anti-FLAG affinity gel, and the co-precipitated proteins were analyzed by mass spectrometry and western blotting. Exosomal proteins were associated with CD109. We revealed the presence of CD109 in exosome fractions from conditioned medium of 293/FLAG-CD109. Moreover, the localization of CD109 in the exosome was demonstrated using immuno-electron microscopy. When we used HEK293 cells expressing FLAG-tagged truncated CD109, which does not contain the C-terminal region, the association of truncated CD109 with exosomes was not detected in conditioned medium. These findings indicate that CD109 is an exosomal protein and that the C-terminal region of CD109 is required for its presence in the exosome.
Assuntos
Antígenos CD/química , Meios de Cultura/metabolismo , Exossomos/metabolismo , Proteínas de Neoplasias/química , Proteínas de Neoplasias/metabolismo , Proteínas Ligadas por GPI/química , Proteínas Ligadas por GPI/metabolismo , Células HEK293 , Humanos , Relação Estrutura-AtividadeRESUMO
CD109, a glycosylphosphatidylinositol-anchored glycoprotein, is expressed at high levels in some human tumors including squamous cell carcinomas. As CD109 is reportedly cleaved by furin and its soluble form is secreted into culture medium in vitro, we hypothesized that CD109 could serve as a tumor marker in vivo. In this study, we investigated CD109 as a novel serum tumor marker using transgenic mice that overexpress mouse CD109 (mCD109-TG mice) and tumor xenografted mice inoculated with human CD109 (hCD109)-overexpressing HEK293 cells. In sera and urine of mCD109-TG mice, mCD109 was detected using western blotting. In xenografted mice, hCD109 secreted from inoculated tumors was detected in sera, using western blotting and CD109 ELISA. Concentrations of tumor-secreted CD109 increased proportionally as tumors enlarged. Concentrations of secreted CD109 decreased notably by 17 h after tumor resection, and became undetectable 48 h after resection. The half-life of tumor-secreted CD109 was about 5.86±0.17 h. These results indicate that CD109 is present in serum as a soluble form, and suggest its potential as a novel tumor marker in patients with cancers that express CD109.