以蛋白質組學探索人類角質細胞核心岩藻醣化相關之生物功能

Abstract

背景:核心岩藻醣化藉由第八型岩藻醣轉移酶催化，此為體內蛋白質維持正常生理功能的重要步驟。我們過去已知第八型岩藻醣轉移酶在乾癬患者病灶處與非病灶處相比的表現量較高，並且在人類角質細胞第八型岩藻醣轉移酶減量的細胞株中，發現細胞生長受到抑制，另一方面，第八型岩藻醣轉移酶對於細胞凋亡則沒有顯著影響。在此研究中，我們希望能透過蛋白質組學，來探索人類角質細胞中與核心岩藻醣化蛋白相關的生物功能，並進一步探索在人類角質細胞中重要的核心岩藻醣化標的蛋白。研究方法:在人類角質細胞細胞株HaCaT的培養過程中，以穩定同位素標記胺基酸，以便在後續的質譜儀分析區分野生型HaCaT、及第八型岩藻醣轉移酶減量的HaCaT，透過小扁豆凝集素將核心岩藻醣化蛋白富集化，接著以胰蛋白酶及PNGase F切割成胜肽，並以液相層析串聯質譜儀分析，透過Mascot搜尋引擎比對鑑別蛋白質，再以Proteome Discoverer計算出各別蛋白質在兩組HaCaT之間的表現量比值。研究結果:在野生型HaCaT及第八型岩藻醣轉移酶減量的HaCaT之間，總計有233個蛋白質有核心岩藻醣化程度的差異，其中45個蛋白質在第八型岩藻醣轉移酶減量的HaCaT表現量上升兩倍以上，46個蛋白質在第八型岩藻醣轉移酶減量的HaCaT表現量則下降兩倍以。針對這些蛋白質以Ingenuity Pathway Analysis進行下游效應分析，與這些蛋白最顯著相關的生物功能中，「細胞間的訊息傳遞與交互作用」、「細胞發育」、「細胞生長與增殖」預期活化狀態將會減少，相對地，「細胞死亡與存活」預期活化的狀態則會增加。此外，我們以PANTHER進行Gene Ontology富集分析，再透過REVIGO的演算法將語意相近而冗餘的Gene Ontology剔除，當simRel設定為預設值0.7時，861個顯著相關的Gene Ontology可進一步縮減為216個，我們將這些Gene Ontology再以Cytoscape做視覺化的網絡呈現，得到顯著相關的生物功能與Ingenuity Pathway Analysis分析結果相似。最後利用Ingenuity Pathway Analysis，探索在野生型HaCaT及第八型岩藻醣轉移酶減量的HaCaT表現量有差異的核心岩藻醣化蛋白，並以圖形呈現這些蛋白彼此之間的關聯性，我們發現，表皮生長因子受體及整合素beta-1皆為可能的重要核心岩藻醣化標的蛋白。結論:在本研究中，我們使用兩套不同的蛋白質組學分析方法，均得到類似的分析結果，在人類角質細胞中核心岩藻醣化與細胞生長及增殖顯著相關，且此生物功能的活化狀態預期會隨著第八型岩藻醣轉移酶減量而下降，這與我們先前的研究結論一致，而表皮生長因子受體為細胞生長及增殖相關的重要核心岩藻醣化標的蛋白。在此研究中，我們以人類角質細胞的細胞株作為疾病模型，預期能更佳地反映人體內的真實狀況，而透過系統生物學，能以更宏觀而有更效率的方式，探討與岩藻醣化蛋白相關的生物功能及標的蛋白。

Introduction: Core-fucosylation is catalyzed by fucosyltransferase 8 (FUT8) and is essential for protein function. Upregulation of FUT8 expression has been observed in the epidermis of lesion sites compared with non-lesion sites in patients with psoriasis. Our previous studies have demonstrated FUT8-knockdown in human keratinocytes cell line HaCaT inhibited cell proliferation while FUT8 expression level had no influence on cells apoptosis. In this study, we aimed to investigate the biological functions of core-fucosylated glycoproteins and potential significant targets of core-fucosylation in human keratinocytes using proteomic approach. Materials and Methods: Wild type and FUT8-knockdown HaCaT clones were cultured with stable isotope labeling by amino acids (SILAC) to differentiate the mass spectrometry peaks from each other. The core-fucosylated glycoproteins were enriched by lens culinaris agglutinin (LCA) and digested by trypsin and Peptide:N-Glycosidase F (PNGase F). These enriched peptides were analyzed by liquid chromatography tandem mass spectrometry (LC/MS/MS). Protein identification and relative abundance quantification were performed using Proteome Discoverer v 1.4.1.14 (Thermo Fisher Scientific, Germany). The MS/MS spectra were searched with the Mascot engine (Matrix Science, USA) against the UniProtKB/Swiss-Prot human database to generate the peak lists. Results: A total of 233 proteins showed different levels of fucosylation between FUT8-knockdown and wild type HaCaT clones. Among them, 45 up-regulated proteins and 46 down-regulated proteins showed fold changes greater than two. We performed downstream effects analysis for these differentially expressed proteins which had fold changes greater than two using Ingenuity Pathway Analysis (IPA) (Qiagen, Germany). P-values were calculated using Fisher’s exact test to determine whether the overrepresentation was significant. The activation z-score was used to infer the presumptive activation state of implicated biological functions. The “Cell-to-cell signaling and interaction”, “Cellular development” and “Cellular growth and proliferation” were predicted to be inactivated. In contrast, the “Cell death and survival” was predicted to be activated. We also conducted gene ontology (GO) enrichment analysis for the differentially expressed proteins with fold changes greater than two using PANTHER (Thomas lab at the University of Southern California, USA). The significance of overrepresentation was determined by binomial test. The differentially expressed proteins were significantly enriched in 861 GO terms. REVIGO (Rudjer Boskovic Institute, Croatia) was further utilized to summarize the list of GO terms by finding representative subsets of GO terms using semantic similarity measure simRel and clustering algorithm. The cut-off value of simRel was set to be 0.7. The 216 non-redundant GO terms were visualized in Cytoscape (National Institute of General Medical Science, USA) in semantic similarity-based network graphs. The clusters of these GO terms were in line with those enriched biological functions identified from IPA. The relationships among core-fucosylated glycoproteins were visualized and investigated using IPA. According to the Ingenuity Knowledge Base, the epidermal growth factor receptor (EGFR) and integrin beta-1 (ITGB1) were the potential targets of core fucosylation in HaCaT. Conclusions: The two analytical methods showed similar enriched biological functions. Using proteomic approach, cellular growth and proliferation was significantly enriched with decreased activation in FUT8-knockdown HaCaT clones, which was consistent with our previous findings. EGFR was the significant core-fucosylated target involved in keratinocytes growth and proliferation. The study of system biology with LCA-enriched glycoproteins in disease model may better reflect the real condition in human bodies.