Article pubs.acs.org/jpr
Analysis of Glycan-Related Genes Expression and Glycan Profiles in Mice with Liver Fibrosis HanJie Yu,† MinZhi Zhu,† YanNan Qin,† YaoGang Zhong,† Hua Yan,† Qi Wang,† HuiJie Bian,*,§ and Zheng Li*,† †
Laboratory for Functional Glycomics, College of Life Sciences, Northwest University, Xi’an 710069, China Cell Engineering Research Centre and Department of Cell Biology, Fourth Military Medical University, Xi’an, China
§
S Supporting Information *
ABSTRACT: Protein glycosylation plays an important role in the pathogenesis and progression of various liver diseases. However, little is known about the precise alterations in protein glycosylation or the potential correlation between glycan-related genes expression and glycan profiles in liver fibrosis. The aim of the study was to investigate potential associations between glycan-related genes expression and glycan profiles to evaluate liver fibrosis in a mouse model. Analyses of glycan-related genes expression and glycan profiles were performed using oligonucleotide microarrays and lectin microarrays, respectively. Real-time PCR and Western blot were used to confirm any altered glycan-related genes expression levels and protein levels. Moreover, altered glycan patterns on the surface of hepatocytes were verified by lectin histochemistry. These results revealed that the mRNA levels of 10 glycan-related genes were significantly altered in fibrotic liver. Furthermore, we observed an increase in multivalent sialic acid, poly-LacNAc, sialyl-T-antigen, Fucoseα-1,3/6GlcNAc, and GalNAcα1−3Gal in fibrotic liver specimens, whereas GlcNAc oligomers was decreased in fibrotic liver. Our findings indicated that the synthetic pathway of “Tn antigen → T antigen (core-1) → sialyl-T antigen” was activated for O-glycan during the process of liver fibrosis. KEYWORDS: hepatic fibrosis, gene expression microarray, lectin microarray, glycan-related genes, glycan structure, lectin histochemistry, tumorigenesis
■
INTRODUCTION Liver fibrosis represents a wound healing response to virtually all forms of chronic liver injury, which is a reversible scarring response. However, uncontrolled liver fibrosis may lead to cirrhosis, which represents a major healthcare issue worldwide. Patients with liver cirrhosis are predisposed and at a higher risk for the development of hepatic carcinoma, and this process occurs in almost all patients with chronic liver injury.1 Damaged hepatocytes produce several mediators including reactive oxygen species and fibrogenic cytokines, which induce the activation and proliferation of hepatic stellate cells (HSCs) and other fibrogenic cells.2 Activated HSCs lose the potential to store intracellular vitamin A, up-regulate the expression of cytoskeletal protein alpha-smooth muscle actin (α-SMA), increase the activity of DNA synthesis, and contain abundant rough endoplasmic reticulum (ER). Excess extracellular matrix (ECM) is produced in activated HSCs (in particular collagen types 1 and 3) in fibrotic liver and is coupled with the overexpression of other glycoproteins such as fibronectin, laminin, merosin, tenascin, nidogen, and hyaluronic acid.3 Protein glycosylation plays an important role in the pathogenesis and progression of various liver diseases.4,5 © 2012 American Chemical Society
Carbohydrate moieties on glycoproteins also play roles in intercellular contact and communication, which are important aspects of host immunity and are related to the development of cancer.6−8 The liver contains various receptors on sinusoidal and hepatocyte surfaces, and many proteins that bind to these receptors rely on carbohydrate moieties during the development of various chronic liver diseases.9 Changes in glycosylation of proteins associated with liver fibrosis received an increased attention. It demonstrated that increases in serum hyaluronan levels might be a marker of fibrosis in chronic viral hepatitis C.10 The result of Mehta et al. showed that the increased fucose binding lectin AAL (Aleuria Aurantia Lectin) reacted with fucosylated agalacto IgG was observed in patients with stage III or greater fibrosis and appeared to be correlated with the degree of fibrosis.11 Using DSA-FACE, Vanderschaeghe et al. have observed the ratio of two N-glycans: bisecting GlcNAc-modified agalacto glycan and triantennary glycan correlated with the histological fibrosis stage equally well as FibroTest which is used in the clinic Received: May 30, 2012 Published: October 8, 2012 5277
dx.doi.org/10.1021/pr300484j | J. Proteome Res. 2012, 11, 5277−5285
Journal of Proteome Research
Article
today.12 Recently, our lab has used lectin arrays and lectin histochemistry to investigate the alteration of protein glycosylation in the activated HSCs (LX-2) compared with the quiescent HSCs. Our findings revealed that 14 lectins showed increased signal while 7 lectins showed decreased signal in the activated LX-2, which was induced by transforming growth factor-β1 (TGF-β1) compared with the quiescent LX2.13 Mouse models of hepatic fibrosis and high-throughput techniques are needed to perform a thorough analysis of this complex phenomenon, including techniques such as gene microarrays and lectin microarrays. The lectin microarrays contain a collection of mostly plant-derived carbohydrate binding proteins that are immobilized onto a solid support at a high spatial density. Interrogation of these arrays with fluorescently labeled samples creates a pattern of binding that is dependent upon the carbohydrate structures present in the system and provides a method for the rapid characterization of carbohydrates on glycoproteins, bacteria, or mammalian cells.14 The lectin microarrays do not require the separation of N or Olinked glycans before analysis or the removal of sialic acids or other substructural motifs that might impede analysis by mass spectrometry. In this study, hepatic fibrosis animal models were constructed and used to investigate the differential expression profiles of glycan-related genes and the glycan structure profile by gene expression and lectin microarrays, in an effort to understand a mechanism of control of the glycan code in animal models of hepatic fibrosis.
■
amount of RNA was measured using a Nano Photometer (IMPLEN; München, Germany). Expression Profile Microarrays and Data Analysis
Amplification of cRNA, hybridization and scanning of the microarray were performed according to the manufacturer’s protocol (Agilent Technologies; Santa Clara, CA). Briefly, cRNAs were produced from the total RNA of F1, F2, F3 and the pooled normal reference (N) by a cRNA Linear Amplification Kit (Agilent Technologies) and purified by RNeasy Mini Kit (QIAGEN; Chasworth, CA). Subsequently, the cRNAs from F1, F2, and F3 were labeled with Cy5 (GE Healthcare; Biosciences, Piscataway, NJ, USA), and the N cRNA pool was labeled with Cy3 (GE Healthcare). Labeled cRNA was then repurified using RNeasy Mini Kit (QIAGEN). Finally, 1.5 μg of labeled cRNA from each of the reference and fibrotic samples was combined and mixed with hybridization buffer before being applied to the microarray. Hybridization of each pair was carried out using a whole mouse genome oligo microarray (Agilent Technologies). Hybridization was performed in the Agilent Microarray hybridization chamber (Agilent Technologies) at 60 °C for 17 h in a rotisserie oven set at 4 rpm. Conditions of hybridization and washing were based on the Agilent Oligonucleotide Microarray Hybridization protocol (Agilent Technologies). The microarrays were scanned using settings of 70% photomultiplier tube and 100% laser power on a GenePix 4000B confocal scanner (version 3.0, Axon Instruments Inc.; Sunnyvale, CA). The fluorescence intensity of the microarray data was extracted using Feature Extraction software (version 9.3; Agilent Technologies), and the microarray data were analyzed by Spotfire software (version 8.0; Spotfire Inc.; Cambridge, MA). Normalization methods (linear and lowess normalization) were applied to eliminate dye-related bias in the microarray results. Generated files were imported into spreadsheets for downstream data analysis and statistical evaluation. A confidence level of P < 0.05 was used in all data sets for the identification of changes in gene expression profiles. The “fold change” value represents the ratio of the individual signal intensity in the fibrotic group to the individual signal intensity in the reference group from 3 paired samples. Genes were classified based upon fold-changes according to the following criteria: fold changes >1.5-fold in the pairs indicated up-regulation of the gene, whereas a fold change of 1.5-fold expression difference compared to the reference group (Tables 1). Among these genes, eight glycan-related genes (C1GALT1C1, ALG5, DPM1, UGT2B5, SULT1D1, SULT1B1, LAMP1, and LAMP2) were up-regulated in liver fibrosis compared to reference livers. Conversely, two glycanrelated genes (HS6ST1 and HYAL2) were expressed at a higher level in the normal reference liver compared to fibrotic liver (criteria of fold change >1.5, p value 2, p value 1.5 or 2.0) and 2 down-regulated genes (fold change 1.5 or
