Center for Proteomics and Metabolomics, Leiden University Medical Center (LUMC), Leiden, The Netherlands.
Genos Glycoscience Research Laboratory, Zagreb, Croatia.
Gastroenterology. 2018 Sep;155(3):829-843. doi: 10.1053/j.gastro.2018.05.030. Epub 2018 May 21.
BACKGROUND & AIMS: Biomarkers are needed for early detection of Crohn's disease (CD) and ulcerative colitis (UC) or to predict patient outcomes. Glycosylation is a common and complex posttranslational modification of proteins that affects their structure and activity. We compared plasma N-glycosylation profiles between patients with CD or UC and healthy individuals (controls).
We analyzed the total plasma N-glycomes of 2635 patients with inflammatory bowel diseases and 996 controls by mass spectrometry with a linkage-specific sialic acid derivatization technique. Plasma samples were acquired from 2 hospitals in Italy (discovery cohort, 1989 patients with inflammatory bowel disease [IBD] and 570 controls) and 1 medical center in the United States (validation cohort, 646 cases of IBD and 426 controls). Sixty-three glycoforms met our criteria for relative quantification and were extracted from the raw data with the software MassyTools. Common features shared by the glycan compositions were combined in 78 derived traits, including the number of antennae of complex-type glycans and levels of fucosylation, bisection, galactosylation, and sialylation. Associations of plasma N-glycomes with age, sex, CD, UC, and IBD-related parameters such as disease location, surgery and medication, level of C-reactive protein, and sedimentation rate were tested by linear and logistic regression.
Plasma samples from patients with IBD had a higher abundance of large-size glycans compared with controls, a decreased relative abundance of hybrid and high-mannose structures, lower fucosylation, lower galactosylation, and higher sialylation (α2,3- and α2,6-linked). We could discriminate plasma from patients with CD from that of patients with UC based on higher bisection, lower galactosylation, and higher sialylation (α2,3-linked). Glycosylation patterns were associated with disease location and progression, the need for a more potent medication, and surgery. These results were replicated in a large independent cohort.
We performed high-throughput analysis to compare total plasma N-glycomes of individuals with vs without IBD and to identify patterns associated with disease features and the need for treatment. These profiles might be used in diagnosis and for predicting patients' responses to treatment.
需要生物标志物来早期检测克罗恩病（CD）和溃疡性结肠炎（UC），或预测患者的预后。糖基化是一种常见且复杂的蛋白质翻译后修饰，会影响其结构和活性。我们比较了 CD 或 UC 患者与健康个体（对照）之间的血浆 N-糖基化谱。
我们通过质谱法，采用特定于连接的唾液酸衍生技术，分析了 2635 名炎症性肠病患者和 996 名对照者的总血浆 N-糖组。血浆样本来自意大利的 2 家医院（发现队列，1989 名炎症性肠病患者[IBD]和 570 名对照者）和美国的 1 家医疗中心（验证队列，646 例 IBD 和 426 名对照者）。从原始数据中使用 MassyTools 软件提取了 63 种符合相对定量标准的糖型，这些糖型的组成特征具有共性，被组合成 78 个衍生特征，包括复杂型聚糖的天线数量和岩藻糖基化、二分枝、半乳糖基化和唾液酸化的水平。通过线性和逻辑回归，检测血浆 N-糖组与年龄、性别、CD、UC 和 IBD 相关参数（疾病部位、手术和药物治疗、C 反应蛋白水平和沉降率）之间的关系。
与对照者相比，IBD 患者的血浆样本中存在大量大尺寸聚糖，杂交和高甘露糖结构的相对丰度降低，岩藻糖基化、半乳糖基化和唾液酸化（α2,3-和α2,6-连接）水平降低。我们可以根据二分枝、半乳糖基化和唾液酸化（α2,3-连接）水平的差异，区分 CD 患者与 UC 患者的血浆样本。糖基化模式与疾病部位和进展、需要更有效的药物治疗和手术有关。这些结果在一个大型独立队列中得到了验证。
我们进行了高通量分析，比较了 IBD 患者与非 IBD 患者的总血浆 N-糖组，并鉴定了与疾病特征和治疗需求相关的模式。这些图谱可能用于诊断，并预测患者对治疗的反应。
