【作者】 何倩；

【导师】 苏纪平；

【作者基本信息】 广西医科大学 ， 耳鼻咽喉头颈外科， 2014， 博士

【摘要】 腺淋巴瘤（Warthin Tumor）也被称为乳头状淋巴性囊腺瘤或单形性腺瘤，是第二常见的唾液腺良性肿瘤。其结构主要包括两个部分：肿瘤上皮组织及淋巴间质。淋巴腺瘤组要影响60至70岁人群，吸烟被认为于该疾病密切相关。虽然以往的研究对于腺淋巴瘤的起病及进程进行了很多探讨，也设立了很多假说，但其病理机制至今仍不明确。大多数学者对腺淋巴瘤的研究多着重于基因组学及蛋白组学。而对于脂质组学这一门新兴学科，其在腺淋巴瘤中的研究甚少。尤其是对于脂质在腺淋巴瘤中的分布及组成，尚未发现任何报道。脂质是人体细胞中不可或缺的重要组成部分，其包括脂肪、水溶性维生素、甘油脂类、磷脂类等一系列生物活性物质。其不但参与生物膜构建，细胞增值、分化、代谢调节及免疫，而且涉及到储存能量、传递信号等生物活动。磷脂酰胆碱，又称为卵磷脂，为磷脂类的一种，其为细胞膜磷脂双分子层的主要成分，并且更多的存在于双分子层的外侧。磷脂酰胆碱成分在细胞及细胞膜内的改变被认为与细胞内生物信号的传送及各类细胞的新陈代谢有关。其对肿瘤的形成、增长及转归起到重要的作用。磷脂酰胆碱的特性是由其绑定的脂肪酸种类决定，因此，比较正常和病理组织中不同脂质的分布，尤其是不同脂肪酸的分布，能够获得代谢调控中关键的脂质标志物,最终揭示脂质在肿瘤活动中的作用机制。质谱成像（Imaging Mass Spetrometry, IMS）是一种最新的原位分析技术。其利用质谱直接扫描生物样品，分析分子在细胞或组织中的“结构、空间与时间分布”信息。此技术以质谱为基础，不局限分析特异的一种或者几种生物分子。质谱成像不但能够在短时间内一次性检测多种生物分子的空间分布，而且不需要烦杂的样本准备过程，更不需要任何免疫技术进行检测前的标记。作为唯一能够使绑定在脂质上的脂肪酸可视化的技术，质谱成像能够将绑定有不同脂肪酸种类的磷脂酰胆碱定位，以直观的方式观察各种磷脂酰胆碱在生物样本上的分布情况。为了更全面更深入地了解腺淋巴瘤的病理特征，为探究此疾病的发生发展及病理机制提供更多的理论依据。本实验利用人类腺淋巴瘤病理样本作为研究对象，采用质谱成像技术比较样本的肿瘤及非肿瘤ROI内磷脂酰胆碱的分布情况，并获得腺淋巴瘤特异性信号分子。进一步采用串联质谱法识别这些特异性分子，从而获得腺淋巴瘤内磷脂酰胆碱的组成情况并讨论其意义。方法：将术后获得的8个病理样本采用冰冻切片机进行组织样本的连续切片，样本厚度为10μm。将所获得的连续切片分别放置于ITO包埋及MAS包埋的载玻片上，用于IMS检测及HE染色。通过HE染色来确认腺淋巴瘤及非瘤组织的ROIs。通过对质谱成像仪参数的设定，将质荷比调整至m/z460-m/z1000范围用于脂质，特别是磷脂的检测。获得IMS检测数据后，通过flexImaging软件比较肿瘤及非肿瘤ROIs的质谱，质谱中每个信号峰即代表一种生物分子。通过SIMtools软件对获得的质谱进行信号峰进行挑选，同时观察相应信号峰对应的生物分子的离子成像图，判断各个生物分子的大概分布趋势。随后通过统计学分析比较各个信号峰在肿瘤及非肿瘤ROI的信号强度，以p <0.01为标准，获得相应区域的特异性信号即生物分子。最后，运用串联质谱法，通过观察离子碎片质谱以及各个离子碎片信号峰之间的质量差，即中性丢失，并根据此离子碎片质谱检索“人类代谢组数据库”(http://www.hmdb.ca/spectra/ms/search)，获得候选生物分子，从而识别上述步骤中获得的特异性生物分子。实验结果:(1) HE染色显示腺淋巴瘤的典型结构，包括肿瘤上皮及淋巴间质。(2)比较肿瘤及非肿瘤ROI的质谱，m/z740-m/z940区域内的信号峰有较大区别，此区域大部分信号峰在肿瘤ROI有较高强度。经过挑选后，此区域的15个信号峰被纳入随后的分析。(3)离子成像图显示的结果与质谱基本一致，15个信号中大部分在肿瘤ROI有较高的信号强度。其中m/z796.5, m/z895.5, m/z897.5and m/z923.5主要分布在非肿瘤ROI，m/z782.5and m/z798.5在肿瘤与非肿瘤ROI的分布未见明显区别。而剩下的九个信号则在肿瘤ROI有较高的信号强度。(4)统计学分析比较肿瘤及非肿瘤ROI中15个信号的强度，发现其中10个信号峰的信号强度在肿瘤及非肿瘤ROI有明显的差异，其差别具有统计学意义。其中，m/z796.5的信号强度在非肿瘤ROI中较高；其余九个信号: m/z741.5, m/z772.5, m/z820.5, m/z822.5, m/z824.5, m/z826.5,m/z844.5, m/z846.5and m/z848.5，在肿瘤ROI有较高信号强度。(5)在所有组织样本中按上述方法检测获得的九个信号的强弱，并确认各个信号在肿瘤ROI的部位。离子成像图结果显示，m/z741.5及m/z772.5重要分布于腺淋巴瘤的淋巴间质区域，m/z822.5, m/z824.5及m/z826.5则分布在肿瘤上皮和淋巴间质中，m/z820.5, m/z844.5, m/z846.5及m/z848.5则主要位于肿瘤上皮区域。最终，五个信号，即m/z772.5，m/z820.5，m/z822.5，m/z844.5及m/z846.5被发现始终于肿瘤ROI保持高强度，因此被确定为腺淋巴瘤特异性信号。(6)通过MS/MS识别这个五个腺淋巴瘤的特异性信号所代表的生物分子。m/z772.5为PC (diacyl-16:0/16:0)，m/z820.5为PC (diacyl-16:0/20:4)，m/z822.5为PC (diacyl-16:0/20:3)，m/z844.5为PC (diacyl-18:2/20:4)，and m/z846.5为PC (diacyl-18:0/20:5)。(7)对照HE染色以及m/z772.5即PC (diacyl-16:0/16:0)的离子成像图发现，此磷脂酰胆碱主要分布在腺淋巴瘤的淋巴基质区域，特别集中在淋巴滤泡的部位。结论：IMS技术能成功的检测磷脂酰胆碱在腺淋巴瘤的肿瘤及非肿瘤ROI的空间分布及信号强度。通过此技术，五种磷脂酰胆碱被发现在腺淋巴瘤内特异性的增高，即m/z772.5为PC (diacyl-16:0/16:0)，m/z820.5为PC(diacyl-16:0/20:4)，m/z822.5为PC (diacyl-16:0/20:3)，m/z844.5为PC(diacyl-18:2/20:4)，and m/z846.5为PC (diacyl-18:0/20:5)。其中PC (diacyl-16:0/20:3)主要位于肿瘤上皮区域， PC (diacyl-16:0/20:4)， PC (diacyl-18:2/20:4)及(diacyl-18:0/20:5)则位于肿瘤和淋巴间质区域，而PC (diacyl-16:0/16:0)则特异性在腺淋巴瘤淋巴间质，特别是淋巴滤泡内增加。更多还原

【Abstract】 Warthin tumor (War-T), also known as papillary cystadenoma or lymphaticmonomorphic adenoma, is the second common benign tumor of salivary gland.It consists of two components, neoplastical epithelium and lymphoid stroma.War-T is popular in the people between60to70years old, and smoking isconsidered strongly related to War-T. Although big efforts were made by otherresearches for investigating the occurrence and development of War-T, Thepathogenesis of this lesion still remains unclear. Previous researches werefocused on genomics and proteomics. However, lipidomic as an emergingdiscipline, there is little study for the role of it in War-T. Especially, the thedistribution of lipids in War-T, there is no study for it.Lipids is the important part in human body, it is including fat, Water-solublevitamins, triglycerides, phospholipids and so on. It is not only involved in thestructure of biomembrane, cell proliferation, differentiation, metabolism andimmune regulation but also energy storage, signal transmission and some otherbiological activities. Phosphatidylcholines (PCs) are one kind of phospholipids. It is main component of phospholipid bilayer of biomembrane, and more presentin the outside of bilayer. The alteration of PCs in the membrane reflects the thesignlling and cells metabolism. Therefore, PCs play an important role in theformation, growth and prognosis of tumors. The characristics of PCs are decidedby the types of binding fatty acids. Therefore, comaparing the distribution ofPCs, especially the distribution of fatty acids in normal and pathological samplewill provide the role of PCs in tumors.Imaging Mass Spectrometry (IMS) is one of the newest in-situ analysistechniques. It can directly scan the biological samples and analyze theinformations about the "structure, spatial and time distribution of molecules" ina cell or tissue. This techniques is based on the mass spectra, it is not limited inanalyzing only one or a few specifical biomolecules. IMS can detect a variety ofbiomolecules in the same time without complex sample preparation and label.IMS is the only technique which makes the binding fatty acids of lipid visual. Itcan locate and observe the distribution of the PCs containing different kinds offatty acids in an intuitive way on the biological samples. In order to deeplyunderstand the characristics of War-T and provide information for futherresearches of pathogenesis of War-T, we applied IMS analysis for detecting theprofile of PCs in War-T and Non-T of human samples, and five War-T specificPCs were found. These five PCs were identified by MS/MS analysis and themeaning of these finding was also discussed in this manuscript.Methods:CLINICAL SAMPLES Eight human samples (five from men and three fromwomen) were analyzed. The ages of the patients ranged from30to78years(Table1). None of the patients had received medical treatment for War-T, and there was no recurrent patient.IMAGING MASS SPECTROMETRY (IMS) The tissue blocks were sliced to athickness of10μm at20°C with a cryostat (CM1950; Leica, Wetzler,Germany). The consecutive tissue sections were mounted on an indium-tin-oxide (ITO)-coated glass slide (Bruker Daltonics) and a MAS-coated glass slide(Matsunami, Osaka, Japan) for the IMS analysis and HE staining, respectively.DHB matrix solution was sprayed uniformly by a0.2-mm nozzle caliberairbrush (Procon Boy FWA Platinum; Mr. Hobby, Tokyo, Japan) MALDI-TOF-TOF-type instrument (ultraflex II TOF/TOF, Bruker Daltonics) was applied toanalyze the tissue sections. The setting parameters for the laser energy, detectorgain and random walk function were optimized for each measurement manuallyto obtain the highest sensitivity for the m/z range of460–1000for detectingphospholipids. All data were acquired and visualized by flexImaging2.1software (Bruker Daltonics). The War-T and Non-T regions of interest (ROIs)were determined by comparison with the HE staining results of the consecutivetissue section. After the exclusion of the matrix-derived peaks and isotopic peaksand the peaks were picked up within m/z460–1000according to their intensitieswith the SIMtool software from the War-T and Non-T ROIs, respectively. Thedistributions of these peaks were visualized on a tissue section. The signalintensities of these peaks in each ROI were statistically compared by Welch t-test. Differences with p <0.01were considered significant. All datasets wereanalyzed by this way.MS/MS ANALYSIS The specific peaks in the War-T and Non-T regions wereidentified by MS/MS analyses performed on consecutive tissue sections in thepositive-ion mode by QSTAR Elite (Applied Biosystems/MDS Sciex, FosterCity, CA). The PL species and its FA component were determined by ion- fragment pattern referring to the candidate molecule hits in an MS Search on theHuman Metabolome Database (http://www.hmdb.ca/spectra/ms/search)Results:(1) HE staining was performed to reveal the morphological characteristics of theWar-T and Non-T regions. According to the HE staining, the War-T and Non-Tregions can be distinguished by their histological features. In addition, theneoplastic epithelium and lymphoid stroma can be distinguished in the War-Tregion.(2) Comparison of the mass spectra from the War-T and Non-T regions of case1,a clear difference in signal intensity was observed in m/z740–940between theWar-T and Non-T regions.15peaks were picked up. In comparison with thesemass spectra, most of the peaks were higher in War-T than Non-T.(3) The ion images from these signals for case1showed four signals (m/z796.5,m/z895.5, m/z897.5and m/z923.5) located mainly in the Non-T region. Thedistributions of m/z782.5and m/z798.5appeared almost uniform. The othernine signals (m/z741.5, m/z772.5, m/z820.5, m/z822.5, m/z824.5, m/z826.5m/z844.5, m/z846.5and m/z848.5) were accumulated in the War-T region.(4) Statistical analysis of signal intensities from case1showed that10of the15peaks had a significant difference in signal intensity (p <0.01) between the War-T and Non-T regions. One signal, m/z796.5, was higher in Non-T. The othernine signals, m/z741.5, m/z772.5, m/z820.5, m/z822.5, m/z824.5, m/z826.5,m/z844.5, m/z846.5and m/z848.5, had higher signal intensity in War-T.(5) These nine signals in War-T were visualized for all cases, and the ion imagesshowed that the signals m/z822.5, m/z824.5and m/z826.5were located at theregion which seemed to be neoplastic epithelium in the War-T region. The signals at m/z820.5, m/z844.5, m/z846.5and m/z848.5were located at a regionwhich was considered to include neoplastic epithelium and lymphoid stroma.However, the signals at m/z741.5and m/z772.5appeared partially in the War-Tregion and seemed to be involved in the lymphoid stroma structure. Finally, afterwe applied the War-T signals to all samples, those at m/z772.5, m/z820.5, m/z822.5, m/z844.5and m/z846.5were found to be common to all War-T regions.(6) MS/MS analysis identified these five signals (m/z772.5, m/z820.5, m/z822.5, m/z844.5and m/z846.5) from tissue sections. According to the ionfragmentation and the candidate from The MS Search of the HumanMetabolome Database, m/z772.5were identified as [PC (diacyl-16:0/16:0)+K]+. The molecules at m/z820.5and m/z822.5were identified as [PC (diacyl-16:0/20:4)+K]+and [PC (diacyl-16:0/20:3)+K]+. The molecules at m/z844.5and m/z846.5were identified as [PC (diacyl-18:2/20:4)+K]+and [PC (diacyl-18:0/20:5)+K]+.(7) By comparing the hot spots in the ion image and the enlarged HE-stainingimages, we found that the strong signal of PC (diacyl-16:0/16:0) was from thefolliculus lymphaticus structure.Summary and Conclusions: An IMS analysis revealed that10signals havesignificant differences in signal intensities between the War-T and Non-Tregions from case1. There were nine signals specifically detected in the War-Tregion. After applied these War-T positive signals to all dataset, five War-Tspecific signals were confirmed. One signal, identified as were PC (diacyl-16:0/20:3), located mainly in neoplastic epithelium; three signals identified asPC (diacyl-16:0/20:4), PC (diacyl-18:2/20:4) and PC (diacyl-18:0/20:5), locatedmainly in neoplastic epithelium and lymphoid stroma; one signal was identified as PC (diacyl-16:0/16:0) and was specifically distributed in the folliculuslymphaticus of lymphoid stroma in War-T.更多还原