Identification of galectin-1 as a novel hypoxia-regulated protein via proteomic approaches

Abstract

Purpose/ObjectiveTumor hypoxia has been shown to affect tumor aggressiveness and treatment outcomes in solid tumors. In this study, we used SELDI-TOF (Surface Enhanced Laser Desorption/Ionization Time of Flight) and mass spectroscopy (LC/MS/MS) to identify Galectin-1, a carbohydrage binding protein, as a novel hypoxia-induced protein in both tumor cell lines and human head and neck squamous cell carcinomas (HNSCC).Materials/MethodsWe applied concentrated conditioned media from FaDu cells (HNSCC cell line) grown in either hypoxia or normoxia to strong anion exchange protein chips and used SELDI-TOF mass spectrometry to characterize hypoxia specific protein profiles from the media. We used LC/MS/MS to identify the most robustly induced protein peak as galectin-1. We used immublots and quantitative real-time PCR (QRTPCR) to confirm hypoxic-upregulation of the galectin-1 at the protein and message levels. We generated a double sandwich Elisa system to quantify the levels of secreted galectin-1 in both conditioned media and plasma samples. For xenograft studies, we exposed FaDu tumor bearing scid mice to either 72 hrs of 21% O2 (normoxia) or 10% O2 (hypoxia) and obtained plasma for analysis. We used the Eppendorf microelectrode to measure tumor oxygen tension in patients with previously untreated HNSCC. Plasma and tissue expression of galectin-1 was correlated to tumor pO2 and treatment outcomes (freedom from relapse [FFR] and disease-specific survival [DSS]).ResultsSELDI-TOF studies yielded detectable differences in the protein profile of normoxic and hypoxic cells at the 15 kDa position. LC/MS/MS analysis revealed the peak to be galectin-1. Immunoblot studies confirmed increased expression of galectin-1 in both cell lysate and conditioned media. QRTPCR also showed hypoxic induction of galectin-1 at the message level. ELISA studies confirmed that galectin-1 was secreted into the media of several cancer cells lines under hypoxia. Plasma levels of galectin-1 was higher in tumor bearing scid mice breathing 10% O2 compared to those breathing room air. Elisa studies showed a very low plasma level of galecin-1 for both head and neck cancer patients and healthy volunteers and no correlation between plasma galectin-1 levels and tumor pO2 or treatment outcomes in 81 patients with Eppendorf measurements. However, immunohistochemical staining of tumor tissues showed a significant correlation galectin-1 staining with tumor pO2 and treatment outcomes. Patients with strong staining had significantly lower tumor pO2 (p = 0.028), FFR (p = 0.01) and DSS (p = 0.04) than those with weak or no staining.ConclusionsGalectin-1 is a novel hypoxia regulated protein. Although secreted, it is only detected at a very low level in patient plasma samples. In contrast, Galectin-1 tumor tissue expression correlated well with tumor oxygenation and treatment outcomes, suggesting that it may serve as a novel intrinsic hypoxia marker Purpose/ObjectiveTumor hypoxia has been shown to affect tumor aggressiveness and treatment outcomes in solid tumors. In this study, we used SELDI-TOF (Surface Enhanced Laser Desorption/Ionization Time of Flight) and mass spectroscopy (LC/MS/MS) to identify Galectin-1, a carbohydrage binding protein, as a novel hypoxia-induced protein in both tumor cell lines and human head and neck squamous cell carcinomas (HNSCC). Tumor hypoxia has been shown to affect tumor aggressiveness and treatment outcomes in solid tumors. In this study, we used SELDI-TOF (Surface Enhanced Laser Desorption/Ionization Time of Flight) and mass spectroscopy (LC/MS/MS) to identify Galectin-1, a carbohydrage binding protein, as a novel hypoxia-induced protein in both tumor cell lines and human head and neck squamous cell carcinomas (HNSCC). Materials/MethodsWe applied concentrated conditioned media from FaDu cells (HNSCC cell line) grown in either hypoxia or normoxia to strong anion exchange protein chips and used SELDI-TOF mass spectrometry to characterize hypoxia specific protein profiles from the media. We used LC/MS/MS to identify the most robustly induced protein peak as galectin-1. We used immublots and quantitative real-time PCR (QRTPCR) to confirm hypoxic-upregulation of the galectin-1 at the protein and message levels. We generated a double sandwich Elisa system to quantify the levels of secreted galectin-1 in both conditioned media and plasma samples. For xenograft studies, we exposed FaDu tumor bearing scid mice to either 72 hrs of 21% O2 (normoxia) or 10% O2 (hypoxia) and obtained plasma for analysis. We used the Eppendorf microelectrode to measure tumor oxygen tension in patients with previously untreated HNSCC. Plasma and tissue expression of galectin-1 was correlated to tumor pO2 and treatment outcomes (freedom from relapse [FFR] and disease-specific survival [DSS]). We applied concentrated conditioned media from FaDu cells (HNSCC cell line) grown in either hypoxia or normoxia to strong anion exchange protein chips and used SELDI-TOF mass spectrometry to characterize hypoxia specific protein profiles from the media. We used LC/MS/MS to identify the most robustly induced protein peak as galectin-1. We used immublots and quantitative real-time PCR (QRTPCR) to confirm hypoxic-upregulation of the galectin-1 at the protein and message levels. We generated a double sandwich Elisa system to quantify the levels of secreted galectin-1 in both conditioned media and plasma samples. For xenograft studies, we exposed FaDu tumor bearing scid mice to either 72 hrs of 21% O2 (normoxia) or 10% O2 (hypoxia) and obtained plasma for analysis. We used the Eppendorf microelectrode to measure tumor oxygen tension in patients with previously untreated HNSCC. Plasma and tissue expression of galectin-1 was correlated to tumor pO2 and treatment outcomes (freedom from relapse [FFR] and disease-specific survival [DSS]). ResultsSELDI-TOF studies yielded detectable differences in the protein profile of normoxic and hypoxic cells at the 15 kDa position. LC/MS/MS analysis revealed the peak to be galectin-1. Immunoblot studies confirmed increased expression of galectin-1 in both cell lysate and conditioned media. QRTPCR also showed hypoxic induction of galectin-1 at the message level. ELISA studies confirmed that galectin-1 was secreted into the media of several cancer cells lines under hypoxia. Plasma levels of galectin-1 was higher in tumor bearing scid mice breathing 10% O2 compared to those breathing room air. Elisa studies showed a very low plasma level of galecin-1 for both head and neck cancer patients and healthy volunteers and no correlation between plasma galectin-1 levels and tumor pO2 or treatment outcomes in 81 patients with Eppendorf measurements. However, immunohistochemical staining of tumor tissues showed a significant correlation galectin-1 staining with tumor pO2 and treatment outcomes. Patients with strong staining had significantly lower tumor pO2 (p = 0.028), FFR (p = 0.01) and DSS (p = 0.04) than those with weak or no staining. SELDI-TOF studies yielded detectable differences in the protein profile of normoxic and hypoxic cells at the 15 kDa position. LC/MS/MS analysis revealed the peak to be galectin-1. Immunoblot studies confirmed increased expression of galectin-1 in both cell lysate and conditioned media. QRTPCR also showed hypoxic induction of galectin-1 at the message level. ELISA studies confirmed that galectin-1 was secreted into the media of several cancer cells lines under hypoxia. Plasma levels of galectin-1 was higher in tumor bearing scid mice breathing 10% O2 compared to those breathing room air. Elisa studies showed a very low plasma level of galecin-1 for both head and neck cancer patients and healthy volunteers and no correlation between plasma galectin-1 levels and tumor pO2 or treatment outcomes in 81 patients with Eppendorf measurements. However, immunohistochemical staining of tumor tissues showed a significant correlation galectin-1 staining with tumor pO2 and treatment outcomes. Patients with strong staining had significantly lower tumor pO2 (p = 0.028), FFR (p = 0.01) and DSS (p = 0.04) than those with weak or no staining. ConclusionsGalectin-1 is a novel hypoxia regulated protein. Although secreted, it is only detected at a very low level in patient plasma samples. In contrast, Galectin-1 tumor tissue expression correlated well with tumor oxygenation and treatment outcomes, suggesting that it may serve as a novel intrinsic hypoxia marker Galectin-1 is a novel hypoxia regulated protein. Although secreted, it is only detected at a very low level in patient plasma samples. In contrast, Galectin-1 tumor tissue expression correlated well with tumor oxygenation and treatment outcomes, suggesting that it may serve as a novel intrinsic hypoxia marker