A Glycopeptide from Interstitial Cystitis Patients as a Novel Anticancer Lead

IC/PBS是一种慢性膀胱疾病，其特征是膀胱上皮变薄和溃疡，引起剧烈疼痛、尿频和尿急。巴尔的摩马里兰大学的开创性工作表明，IC/PBS中发生的许多特征性病理变化是由一个特定因素引起的。这种被称为APF的因子被证明在亚纳摩尔浓度下对膀胱上皮细胞具有抗增殖活性。APF导致细胞旁通透性增加，参与紧密连接形成的几种蛋白下调，肝素结合的表皮生长因子样生长因子（HB-EGF）水平降低。此外，在同样低的浓度下，APF也是一种有效的膀胱肿瘤细胞的抗增殖剂，随后被证明可以抑制其他肿瘤细胞系的增殖。APF的肽部分与Wnt信号受体frizzle8的第6跨膜结构域的一段氨基酸序列100%相同。而糖部分，Gal(β)1-3GalNAc(α)-O-Thr是众所周知的Thomsen Friedenreich双糖，一种用于疫苗设计和癌症免疫治疗的肿瘤相关碳水化合物抗原。2006年，人们开始合成一系列APF的asialo衍生物的类似物（as-APF，与天然唾液化化合物等效），以确定天然糖肽的结构-活性谱。我们已经确定，大多数分子都是完全抗增殖活性所必需的。糖是必需的，但是Thomsen Friedenreich双糖可以用连接苏氨酸的LacNAc (Gal(β)1-4GlcNAc) α代替。肽序列可以在c端截断（去除丙氨酸）至8聚体而不损害活性；进一步的截断消除了函数。活动的其他重要特征是1)在终点站保持电荷；2) n端氨基酸侧链上甲基的特殊排列，以及在c端尾部承担某些二级结构元件的能力。AXXXA基序是一个经常在螺旋基序中发现的基序，并参与与其他蛋白质螺旋的结合（蛋白质-蛋白质相互作用）。打乱这些氨基酸的正确排列对活性是有害的。一个非常重要的发现是，具有d -氨基酸取代的特异性衍生物是膀胱上皮细胞中APF抗增殖活性的抑制剂，这些衍生物现在可以作为IC/PBS患者的治疗先导。由于我们已经发表了分子的肽部分，并且现在正在完成一系列8-mer糖肽的研究，因此我们将重点放在分子的糖部分的元素上，这些元素是APF发挥作用所必需的。我们已经知道了一些关于糖的需求，但现在正在修改双糖周围的单个原子和立体化学，以绘制与APF糖部分有关的结合相互作用。2006年，我们的合作者在膀胱上皮中发现了一种APF细胞受体，这一研究取得了重大进展。细胞骨架相关蛋白4 （CKAP4）被鉴定为这种受体，其功能的下调使细胞对APF活性脱敏。CKAP4将细胞骨架连接到内质网，它结合表面活性剂蛋白A和组织纤溶酶原激活物，但对其实际功能或其抑制的后果知之甚少。我们对CKAP4与APF相互作用的细节非常感兴趣，我们获得了一笔小额资助，用于克隆和纯化细胞外结构域（474个残基），这项工作最近在NCI Frederick的先进技术项目的蛋白质表达实验室完成。我们在APF工作中的主要工作是致力于对溶液中APF及其类似物的首选构象进行结构表征。作为一个小的糖肽，它在水溶液中是相对非结构化的，但我们发现某些核磁共振观察结果表明，c端结构域有一个优先折叠，我们推测肽的这一部分需要被结构化才能使APF发挥作用。我们正在从几个方面探索溶液生物物理学：1)比较9-聚和8-聚系列中活性和非活性化合物的构象；2)研究APF可能的聚集特性，因为它具有亲水末端（糖）和非常疏水的肽序列；3)研究类似物与CKAP4的结合，观察活性和非活性化合物的变化趋势；4)与马里兰大学药学院的合作者一起开发糖肽的新力场，并进行模拟，可能为溶液中APF的行为提供其他线索。所有这些研究都将致力于确定APFs抗癌活性的药效团，以设计新的模拟物，可用于选择性地关闭肿瘤组织中的增殖。在过去的一年里，我们已经：1)定义了最简单的糖肽类似物，它是一种完全活跃的抗增殖剂；2)确定了APF活性的抑制剂；3)定义了将n -乙酰半乳糖胺转移到裸APF肽的特定糖基转移酶；4)编译了大量的核磁共振数据，用于马里兰大学的合作者进行的许多模拟。我们现在能够在结构上和原子尺度上定义活动所需的APF的特征。我们也获得了少量的CKAP4蛋白用于结合研究。这些延迟是由于蛋白质最初的低表达。我们的合作者在抗增殖实验中提取了CKAP4的一些细胞外结构域，并试图通过与APF共孵育来抑制APF的活性。这没有抑制，因此蛋白质需要在一个更天然的环境中活动。在这些成果的基础上，有许多新的途径可以寻求，我们将在来年积极参与寻求这些途径。

IC/PBS is a chronic disease of the bladder characterized by thinning and ulceration of the bladder epithelial layer causing severe pain, urinary frequency and urgency. Seminal work at the University of Maryland, Baltimore, showed that a specific factor was responsible for many of the characteristic pathological changes that occur in IC/PBS. This factor, called APF, was shown to have antiproliferative activity toward bladder epithelial cells at sub-nanomolar concentrations. APF caused an increase in paracellular permeability, the down regulation of several protein involved in tight junctions formation and reduced the levels of heparin-binding epidermal growth factor-like growth factor (HB-EGF). In addition, APF was also a potent antiproliferative agent against bladder tumor cells at equally low concentrations and has subsequently been shown to inhibit proliferation of other tumor cell lines. The peptide portion of APF has 100% sequence identity to a stretch of amino acids in the 6th transmebrane domain of Frizzled 8, a Wnt signaling receptor. Whereas the sugar portion, Gal(beta)1-3GalNAc(alpha)-O-Thr is the well-known Thomsen Friedenreich disaccharide, a tumor associated carbohydrate antigen used in vaccine design and in the immunotherapy of cancer. In 2006, synthesis began on a series of analogues of the asialo derivative of APF (as-APF, equipotent to the natural sialylated compound) to define the structure-activity profile of the natural glycopeptide. We have determined that most all of the molecule is necessary for full antiproliferative activity. The sugar is essential but the Thomsen Friedenreich disaccharide can be replaced with LacNAc (Gal(beta)1-4GlcNAc) alpha linked to the threonine. The peptide sequence may be truncated at the C-terminal end (removal of alanine) to an 8-mer without detriment to activity; further truncation abolishes function. Other important features for activity are 1) maintenance of charge at the termini; 2) A specific arrangement of methyl groups on the N-terminal amino acid sidechains and the ability to assume some secondary structural element in the C-terminal tail. The AXXXA motif is one that is frequently found within a helical motif and involved in binding to other protein helices (protein-protein interactions). Disruption of proper arrangement of these amino acids is detrimental to activity. A very important discovery was one that showed that specific derivatives with D-amino acid substitutions are inhibitors of APF antiproliferative activity in bladder epithelial cells, and these can now be developed as therapeutic leads for IC/PBS patients. Since we have already published on the peptide portion of the molecule and are now completing a study for a series of 8-mer glycopeptides, we are concentrating on the elements of the sugar portion of the molecule that are necessary for APF to function. We know several things about the sugar requirements already, but are now modifying individual atoms and stereochemistries around the disaccharide to map the binding interactions that are involved with the sugar portion of APF. A significant advance in the research was made by our collaborators when in 2006 they published on the discovery of a cellular receptor for APF in bladder epithelium. Cytoskeletal-associated protein 4 (CKAP4) was identified and characterized as this receptor, and knock down of its function desensitized cells to APF activity. CKAP4 links the cytoskeleton to the endoplasmic reticulum, it binds surfactant protein A and tissue plasminogen activator, but little is known of its actual function or the consequences of its inhibition. We are very interested in the details of the putative interactions of CKAP4 with APF, and a small grant was awarded to us to clone and purify the extracellular domain (474 residues) which has recently been accomplished in the Protein Expression Lab of he Advanced Technology Program here at NCI Frederick. A major effort in our APF work is dedicated to the structural characterization of the preferred conformation of APF and its analogues in solution. Being a small glycopeptide, it is relatively unstructured in water solution, but we have found certain NMR observables that suggest a preferred fold in the C-terminal domain, the portion of the peptide that we have surmised needs to be structured for APF to function. We are exploring the solution biophysics on several fronts: 1) Compare the conformations of active and inactive compounds in both the 9-mer and 8-mer series; 2) Investigate the possible aggregative properties of APF since it has a hydrophilic end (sugar) and a very hydrophobic peptide sequence; 3) Study the binding of analogues to CKAP4 and observe trends with actives and inactive compounds and 4) Develop new force fields for glycopeptide with our collaborators at the School of Pharmacy at the U. of Maryland and perform simulations that may offer other clues to APF behavior in solution. All of these studies will be geared toward defining a pharmacophore for APFs anticancer activity to design novel mimetics that may be used to selectively shut down proliferation in tumor tissue. In the past year we have: 1) Defined the simplest glycopeptide analogue that is fully active as an antiproliferative agent, 2) Identified inhibitors of APF activity, 3)Defined the specific glycosyl transferase that transfers N-acetyl galactosamine to the naked APF peptide and 4) Compiled mass amounts of NMR data to use with teh many simulations that have been performed by our collaborators at the University of Maryland. We are now in a position to define structurally and on an atomic scale the features of APF needed for activity. W e have also obtained small amounts of CKAP4 protein for binding studies. These were delayed due to the low expression initially of the protein. Our collaborators have taken some of the extracellular domain of CKAP4 and atempted to inhibit APF activity by co-incubation with APF in an antiproliferative assay. This did not inhibit, thus the protein needs to be in a more native environment for activity. THere are a host of new avenues to pursue based on these results and we are actively involved in pursuing htese in the coming year.