Antibody Therapy of Cancer

具体目标1：制备人源性单抗。抗体工程通常是通过在微生物(如噬菌体、细菌和酵母)表面展示人类抗体或抗体片段来进行的。利用人HEK-293T细胞高效表达蛋白质的特点和流式细胞术的高通量筛选特性，我们发展了一种新的方法，称为哺乳动物细胞展示。利用这种方法，在人HEK-293细胞中表达了功能性单链抗体可变片段(ScFv)，并随后通过流式细胞术从组合文库中鉴定和分离了高亲和力的抗原结合。HEK 293细胞比微生物更有可能正确表达抗体。我们和其他人已经证明，哺乳动物细胞展示也可以通过靶向内在热点或共表达激活诱导胞苷脱氨酶(AID)来模拟体细胞超突变，以达到抗体亲和力的成熟。AID是B细胞体细胞高突变的关键酶。在2009-2010财年，我们在哺乳动物细胞展示方面的工作产生了两篇发表在《分子生物学方法》上的方法论文，并在国内和国际治疗性抗体会议上发表了几次演讲。由于其对患者的免疫原性较低，全人单抗(MAb)是临床上最理想的抗体形式。间皮蛋白在包括间皮瘤和卵巢癌在内的多种癌症中都有高表达，因此是一种很有前途的新的治疗靶点。Ira Pastan和他的同事之前的研究导致了一种抗间硫蛋白免疫毒素(SS1P)的开发。它含有一种与细菌毒素融合的小鼠SS1抗体。含有相同鼠Fv的嵌合抗体(MORAb-009)也被开发出来，目前正在进行间皮瘤和胰腺癌的II期临床试验。在08-09财年，我们(1)确定HN1是一种高亲和力抗体，对间皮瘤和卵巢癌细胞具有高亲和力；(2)基于HN1抗体生产了一个完全人的免疫球蛋白分子；(3)发现HN1能够高亲和力地与间皮瘤和卵巢癌细胞上的细胞表面相关间硫蛋白结合。这项工作获得了专利申请。在2010财年，我们继续表征HN1，并证明它可以特异性地杀死具有强烈抗体依赖细胞介导的细胞毒性(ADCC)的癌细胞。在2010财年，我们关于HN1的研究发表在《国际癌症杂志》上，并在几个国际癌症和抗体会议上发表。这项工作得到了间皮瘤应用研究基金会和卵巢癌研究基金的部分支持。2009财年，我们对GPC3作为一种新的肝癌特异性抗原产生了兴趣。GPC3在正常组织中不表达，在肝细胞癌中高表达，是一种很有前途的肝癌治疗候选基因。GPC3功能缺失突变导致Simpson-Golabi-Behmel综合征，一种罕见的X连锁过度生长障碍。GPC3通过其N端与IGF-2和IGF-1R结合，通过其C端与Wnt和Hedgehog蛋白结合。GPC3还通过其硫酸乙酰肝素糖链与成纤维细胞生长因子-2结合。在2010财年，我们在HEK-293细胞中制备了重组可溶性GPC3(SGPC3)蛋白，发现sGPC3蛋白对肝癌细胞的增殖有抑制作用。这项工作发表在《国际癌症杂志》上，并在2010财年申请了专利。我们正在进行的研究集中在针对GPC3的新型单抗的开发上，以抑制肝癌生长。具体目标2：设计新型免疫结合物。卵巢癌和间皮瘤通常同时表达间皮素和粘蛋白CA125/MUC16。间充质蛋白与CA125/MUC16的相互作用可能通过细胞黏附促进肿瘤的种植和腹膜扩散。然而，这种分子相互作用的详细性质却鲜为人知。在2009财年，我们在Mesothelin中发现了一个新的CA125/MUC16不同的功能结合域(称为IAB)。我们进一步证明IAb可以通过一种新的机制有效地阻断间充质蛋白与CA125的结合，并抑制癌细胞的黏附。IAb是治疗CA125表达肿瘤的良好候选药物。我们针对CA125的工作在09财年获得了一份出版物和一项国际专利申请。分离出人类单抗并不总是可能的，特别是对于那些在人类中免疫原性低的抗原。CA125就是这样一种抗原。自1981年OC125产生以来，针对CA125的小鼠单抗相继问世。一种名为B43.13(也称为Oregoomab)的鼠单抗最初是作为一种肿瘤显像剂开发出来的，用于检测复发的卵巢癌，并已在卵巢癌治疗的第二阶段临床研究中进行了评估。然而，由于CA125在免疫系统下调中的作用，目前还没有描述一种完全人源性的抗CA125单抗。在2010财年，我们制作了IAB-人Fc杂化分子免疫粘附素HN125，并表明它可以高亲和力结合表达CA125的癌细胞，并在功能上阻断CA125-间皮蛋白的相互作用。我们目前正在检测其抗肿瘤活性。

Specific Aim 1: Creating human monoclonal antibodies. Antibody engineering is typically carried out by displaying human antibodies or antibody fragments on the surface of microorganisms (e.g. phage, bacteria and yeast). By taking advantage of highly efficient protein expression in human HEK-293T cells and high throughput screening features of flow cytometry, we have developed a new method called mammalian cell display. Using this approach, functional single-chain antibody variable fragments (scFvs) are expressed in human HEK-293 cells and high affinity antigen binders are subsequently identified and isolated from a combinatory library via flow cytometry. HEK 293 cells are more likely than microorganisms to correctly express antibodies. We and others have shown that mammalian cell display can also be used to mimic somatic hypermutation for antibody affinity maturation by targeting intrinsic hotspots or co-expressing activation-induced cytidine deaminase (AID), the key enzyme for somatic hypermutation in B cells. In FY09-10, our work on mammalian cell display resulted in two method papers published in Methods in Molecular Biology and in several presentations at national and international therapeutic antibody conferences. Due to their lower immunogenicity in patients, a fully human monoclonal antibody (mAb) is the most desirable antibody format for clinical application. Mesothelin is a promising new therapeutic target given its high expression in a variety of cancers including mesothelioma and ovarian cancer. Previous research by Ira Pastan and colleagues led to the development of an anti-mesothelin immunotoxin (SS1P). It contains a murine SS1 Fv fused to a bacterial toxin. A chimeric antibody (MORAb-009) containing the same murine Fv was also developed and is currently being examined in a Phase II clinical trial for mesothelioma and pancreatic cancer. In FY08-09, we (1) identified HN1 as a high affinity Fv specific for mesothelin, (2) produced a fully human IgG molecule based on the HN1 Fv, and (3) showed that HN1 was able to bind specifically to cell surface-associated mesothelin on mesothelioma and ovarian cancer cells with high affinity. This work yielded a patent application. In FY10, we continued to characterize HN1 and demonstrated it could specifically kill cancer cells with strong antibody-dependent cell-mediated cytotoxicity (ADCC). In FY10, our work on HN1 was published in the International Journal of Cancer and presented at several international cancer and antibody meetings. This work was in part supported by the Mesothelioma Applied Research Foundation and the Ovarian Cancer Research Fund. In FY09, we became interested in glypican-3 (GPC3) as a new liver cancer-specific antigen. GPC3 is a promising candidate for liver cancer therapy given that it is not expressed in normal tissues and shows high expression in hepatocellular carcinoma (HCC). Loss-of-function mutations of GPC3 cause Simpson-Golabi-Behmel syndrome, a rare X-linked overgrowth disorder. GPC3 binds, through its N-terminus, to IGF-2 and IGF-1R, and through its C-terminus to Wnt and Hedgehog proteins. GPC3 also binds FGF-2 through its heparan sulfate glycan chains. In FY10, we made recombinant soluble GPC3 (sGPC3) protein in HEK-293 cells and found that sGPC3 protein could inhibit the proliferation of HCC cells. This work yielded a publication in the International Journal of Cancer and a patent application in FY10. Our ongoing studies are focused on the development of novel mAbs targeting GPC3 to inhibit HCC growth. Specific Aim 2: Designing novel immunoconjugates. Ovarian cancer and mesothelioma frequently express both mesothelin and mucin CA125/MUC16. The interaction between mesothelin and CA125/MUC16 may facilitate the implantation and peritoneal spread of tumors by cell adhesion. However, the detailed nature of this molecular interaction is poorly understood. In FY09, we identified a novel distinct functional binding domain (called IAB) in mesothelin for CA125/MUC16. We further showed that IAB could effectively block the binding of mesothelin to CA125 and inhibit cancer cell adhesion by a novel mechanism. IAB is a good candidate for use as an antagonist to treat CA125-expressing tumors. Our work targeting CA125 yielded one publication and an international patent application in FY09. Isolation of a human mAb is not always possible, especially for those antigens with low immunogenicity in humans. One such antigen is CA125. Since the generation of OC125 in 1981, numerous mouse mAbs directed against CA125 have been developed. A mouse mAb called B43.13 (also called Oregovomab) was first developed as a tumor-imaging agent to detect recurrent ovarian cancer and has been evaluated in Phase II clinical studies for ovarian cancer treatment. However, a fully human mAb against CA125 has not been described, possibly due to the role of CA125 in down-regulation of the immune system. In FY10, we made the IAB-human Fc hybrid molecule called immunoadhesin HN125 and showed it could bind CA125-expressing cancer cells with high affinity and functionally block the CA125-mesothelin interaction. We are currently examining its anti-tumor activity.