杨西川

杨西川

论文题目：抗人血管内皮生长因子单链抗体的构建、表达及活性鉴定

作者简介：杨西川，男，1964年04月出生，1995年09月师从于四川大学王大章教授，于1998年06月获博士学位。

摘要

恶性肿瘤对人类健康危害极大。其发病率呈逐年上升的趋势，死亡率亦在各种疾病前列。目前，临床上主要依靠手术切除，放射治疗和化学治疗为主的常规疗法以治疗恶性肿瘤，并取得了一定的疗效。但其远期疗效很不理想，五年生存率仍不高。这与恶性肿瘤的局部扩散和远位转移的生物学特性密切相关。因此，医务人员和研究人员在肿瘤治疗的工作中不断创造总结，提出和转变抗癌研究新思路。

1971年，Folkman指出“肿瘤生长依赖血管生成”。并由此把恶性肿瘤的生长分为血管前期和血管化期。血管前期中，肿瘤组织的营养及代谢产物的传送依赖与宿主组织间的相互弥散，从而进行肿瘤细胞的克隆性增殖，其直径可达2～3mm。此时如果肿瘤组织内无新生的毛细血管和小血管形成而建立新的血供，则肿瘤将停止生长或退化。血管化期对肿瘤组织的持续生长非常重要。在此期间，将有大量的肿瘤血管生成，进而形成肿瘤组织的血管系统，完善了肿瘤组织的血液供应，提供了肿瘤组织生存和发展的必要条件。血管生成（angiogenesis）是指从已存在的毛细血管和毛细血管后微静脉的血管内皮细胞分化而形成新的血管。正常生理状态下，除女性月经期以外，VEC基本处于不增殖状态，更新周期达1000天以上。这是由于与血管生成的有关刺激因素和抑制因素常处于动态平衡状态。肿瘤组织通过使血管生成刺激因素增加和/或使抑制因素减少而打破两者之间的平衡，启动血管生成表型。此时，VEC处于快速增殖和迁移状态(其更新周期仅为5天)，并伴有血管基底膜溶解，蛋白溶解的表达及其对细胞外基质的降解，新VEC外基质的重排和具有VEC的血管壁形成等一系列过程。从而导致肿瘤血管生成。

研究表明，血管生成的刺激因素包括有一类称为“血管生成因子”(angiogenic growth factors)的细胞因子，起着刺激新血管形成的作用。目前已经发现了几十种肿瘤血管生成刺激因子和十几种血管生成抑制因子。这些因子相互作用而形成网络，其中在肿瘤血管生成中起着关键作用的是血管内皮生长因子（Vascular Endothelial Growth Factor, VEGF）。VEGF是肿瘤细胞分泌的具有多种功能的细胞因子, 它由两条N端相同而其它区域存某些差异的多肽链通过二硫键连接而成的高度保守糖蛋白二聚体，分子量34-42KD。目前人体共发现VEGF121，165，188，206等四肿同功异构体。 VEGF选择性作用于血管内皮细胞上的两种Ⅲ型酪氨酸激酶受体––KDR/Flk-1和Flt-1，使后者自身的酪氨酸残基磷酸化，激活磷脂酶C及GTP酶活化蛋白（GAP），生物作用为增加血管通透性，使血管内皮细胞迁移、增殖，促进新生血管形成。

VEGF有以下生物学作用(1)特异性促进血管内皮细胞增殖和迁移。(2)是目前所知最强的引起血管通透性增加的物质，比组织胺的效力大50，000倍(3)改变血管内皮细胞外基质，促使外周基质和血管基底膜降解；(4)促进VEC能量转运，有利VEC增殖和迁移；(5)抑制抗原呈递细胞成熟，产生肿瘤免疫逃避。VEGF的生物学特性不仅证实了它是血管生成不可缺少的促进因子，亦说明其对肿瘤生长、扩散与转移起着重要作用。常见的恶性实体肿瘤均有VEGF的过量表达。大量研究表明，抑制VEGF的活性能显著抑制肿瘤的生长。1993年，Kim首次报道抗VEGF单克隆抗体抑制荷瘤鼠的肿瘤生长及其转移瘤的发生。目前，抗人VEGF单克隆抗体已成功地用于多种恶性肿瘤的荷瘤动物实验，并取得了很好的疗效。

但是，随着研究的进一步深入，发现了其应用的局限性。由于VEGF单克隆抗体为鼠源性抗体，免疫源性大，用于人体将会产生人抗鼠抗体免疫反应(HAMA)，使临床疗效减弱或消失，并带来副作用。人源化抗人VEGF MAb尽管取得了一定进展，部分消除了鼠源单抗的免疫原性，但由于技术复杂不易推广。另外，据放射性核素示踪研究表明，注入人体内的单克隆抗体仅有0.01%～0.1%在肿瘤靶部位获得摄取，这主要是由于肿瘤在增殖生长过程中，破坏了宿主的淋巴引流系统；肿瘤自身的血管和“肿瘤化”的宿主血管结构紊乱，缺乏平滑肌，有时出现内皮不完整、膨大纽曲及动静脉交通，使血液内的大分子物质极易渗漏而致间质压的升高。人类肿瘤组织的中心压力较周边高30～40mmHg。加之单克隆抗体分子量约150～200KD，转运能力差，最终使单抗在靶部位难以达到有效浓度。所以抗VEGF MAb难以大量地均匀分布于人体的肿瘤组织而发挥阻断VEGF的作用。

要克服抗人VEGF单抗以上临床应用障碍，必须采用基因工程抗体技术对抗VEGF单克隆抗体进行改造，使抗体“小型化”以增强其穿透能力。目前研究最多的量单链抗体，它是由重链可变区(VH)和轻链可变区(VL)通过一段连接肽连接而成的基因重组蛋白。由于抗体的特异性和高亲合性由Fv决定，因此ScFv是具有完整抗原结合位点的最小功能片段。其分子量仅及完整抗体的1/6。单链抗体与单克隆抗体相比，由于不存在Fc(约占分子量90%)，因此ScFv穿透力增强，容易进入实体瘤微循环且能均匀的分布于肿瘤组织。并且ScFv的免疫源性也大为降低。目前已有不少ScFv的进入临床试验。

本研究的目的为构建并表达抗VEGF scFv，并对其活性进行初步鉴定，为在临床的有效应用，提供理论和实验依据。本研究根据抗原指数等抗原表位预测方法，设计VEGF氨基端1-50残基内一段多肽为合成序列，在ABI 431固相自动肽合成仪上合成。利用合成肽及杂交瘤技术制备抗人VEGF单克隆抗体（MAb）杂交瘤细胞株E11。并对其分泌的MAb之反应性、特异性及体外阻断内皮细胞增殖的活性进行了鉴定。继用该MAb结合免疫组化方法，对20例颊癌组织标本中VEGF的表达及分布进行了研究。结果表明：合成肽单抗可与VEGF原蛋白特异结合，有体外阻断内皮细胞增殖的活力。免疫组化显著：20例标本中阳性率为80%，其中12例（高、中分化）呈阳性，4例（低分化）呈强阳性。VEGF主要分布在肿瘤细胞浆及邻近肿瘤组织的小血管内皮细胞中。肿瘤周围正常组织未见VEGF表达。说明VEGF表达与颊癌的血管生成具有显著的相关性。

体内抑瘤实验证实，该MAb能显著抑制荷瘤鼠实体瘤的生长（资料见另文）

由于鼠源单抗在人体内易诱发人抗鼠抗体（HAMA）免疫反应，其临床应用严重受限。基因工程技术对抗体的改造（即抗体基因重组及人源化）可降低其免疫源性，抗体工程也因此成为国内外研究热点。单链抗体（scFv）以其分子量小，穿透力强，免疫源性低，易于大量生产等优点，得到了较为广泛的应用。为有效地提供临床应用，本研究从杂交瘤细胞株E11中，通过逆转录及多聚酶连式反应，扩增并克隆了E11的可变区基因片断。用一编码亲水性多肽接头的DNA片断将E11单抗轻、重链可变区基因连接，构建表达载体PET-15YV，并导入大肠杆菌BL21（DE3）中进行表达。表达产物经变性及复性处理后，用免疫组化方法检测其活性。经测序表明，VL基因全长333bp，编码111个氨基酸，归属小鼠重链可变区基因Ⅱ（A）亚组。表达产物主要以包涵体形式存在，表达量约40%，用免疫组化方法对颊癌组织进行检测的结果表明，基因工程抗体scFv与原代抗体一样，具有相同的抗原结合活性及较高的组织特异性。本实验结果提示：抗人VEGF单链抗体具有潜在的临床应用前景，可作为生物导向药物，为肿瘤放射免疫疫显象及以血管为靶目标的抗血管生成治疗奠定了良好基础。

最后，本研究又利用Biosym公司开发的计算机辅助分子设计系统模建了抗人VEGF scFv的三维结构，并对其重要的残基进行了分析。结构的合理性验证显示模建结构是合理的。

近十年的抗体人源化研究已经指出人源化的基础在于抗体一级结构的获得和残基相互影响规律的明晰。本研究为深入理解抗原体相互识别机制，为抗体的人源化改造打下了基础。将开辟以癌瘤血管为靶的抗体导向治疗的新领域。并对抗肿瘤治疗策略有极大的理论和实践价值。

Abstract

Malignant tumors cause great harm to human beings. Its incidence increases year after year and the mortality is high. Surgical dissection, chemotherapy and radiotherapy are still the three currently reliable treatments to malignant tumors. Those traditional cancer treatments based on attacking the tumor cells directly (either by removing or destroying them) have produced a certain extent of curative effect. However, the late result is still unsatisfactory with the dismal five year prognosis, which is tightly associated with cancer’s biological properties--- local invasion and metastasis. The idea for anticancer therapy has been changing.

Folkman (1971) put forward a immunotherapeutic hypothesis that tumor growth is dependent crucially on the neovascularization. Most tumors exist early as a small population of cells whose growth relied on nutrients which diffuse from the extravascular space. With proliferation of tumor cells, such a colony eventually expands until it reaches a certain size (2-3mm in diameter) where simple diffusion of nutrients is insufficient. Thus the tumor would stop growing or degenerate, if there did not exist the tumorous angiogenesis to form a new blood-supply network. Angiogenesis is a fundamental process by which new blood vessels are formed. The pathophysiological basis of angiogenesis is that neovascularization i.e. the capillary sprout comes from the differentiation of the vascular endothelial cells (VEC) of the pre-existing capillary and postcapillary venule. In the physiologic condition, except the menses,VEC maintains inactive with a thousands- of-day cell turnover, which attributes to the dynamic balance of the angiogenic stimulators and the inhibitors . Tumor cells can break this dynamic balance in virtue of switching to the angiogenic phenotype, then starts the neovascularization which is of extreme importance to the tumor tissue’ continual growth. At this stage, VEC proliferates rapidly and its turnover is less that 5 days. Relating closely to the formation of its own blood vessel system, the tumor expands and the malignant one shows the properties of invasion and metastasis.

It is reported that angiogenic stimulator includes the kind of more than ten cytokines called angiogenic growth factors (AGF) which promote the neovascularization. It was found tens of those stimulators and more than 10 inhibitors. Those factors interact to form a network, in which the most pivotal one is VEGF which is validated the tumor-derived cytokine with multifunction and the most important tumor angiogenic stimulator. VEGF exists prevalently in the body of most mammal including human being, pig, rabbit and rodent et al.. Linked by disulfide bonds, VEGF is a glucoprotein dimer composing two hyperconservative polypeptide chains with same N-terminals. Its molecular weight is about 34-42KD. It has been identified in human body at least four analogues: VEGF₁₂₁VEGF₁₆₅VEGF₁₈₉VEGF₂₀₆, VEGF is a multifunctional cytokine expressed and secreted at high levels by many tumor cells. As secreted by tumor cells, VEGF is a 34-45 kDa heparinbinding, dimeric, disulfide-bonded blycoprotein that acts directly on endothelial cells (EC) by way of specific receptors to activate phospholipase C and GTP. Two high affinity VEGF receptors, both tyrosine kinases, have thus far been described. VEGF is likely to have a number of important roles in tumor biology. VEGF has the following biofunctions: 1. Specifically promote the proliferation and migration of VEC; 2. It is the most powerful substance known by far to increase blood vessel permeability, about 50,000 times more than histidine; 3. By inducing extracellular proteolysis, promote the degradation of the investing basement membrane of the parent vessel, invasion of the interstitial extracellular matrix by migrating VEC and capillary lumen formation; 4. Activate the glucose transportation of VEC favoring the migration; 5. Inhibit the functional maturation of dendritic cells, which probably brings on the tumor immune evasion. The biological properties of VEGF not only endow it the absolutely necessary stimulator of tumor angiogenisis, but bring out the fact that VEGF is of great importance for the growth, even invasion and metastasis of cancer.The murine anti-human VEGF monoclonal antibody (VEGF MAb) was reported to potently suppress angiogenesis and growth in a variety of human tumor cells lines transplanted in nude mice.

As we know, a major limitation in the use of murine antibodies in human therapy is the production of human antimouse antibodies (HAMAs) which may combine with the administered Mab to diminish or banish the curative effect and bring serious side effects. Scientists have been able to partially eliminate the immunogenicity by humanizing the murine antibodies (including VEGF MAb). But a humanized integral antibody still have the limitation of slow tumor penetration due to its comparatively high molecular weight. Using isotope tracing method, it was found that only 0.01%-0.1% of the MAb could reach the tumor tissue. In the course of transplantation and the following growth of tumor, the host’s lymph drainage is destroyed. The deranged tumor’s blood vessels and the “tumorized” host’s blood vessels are lack of smooth muscle. And sometimes endothelial membrane shows discontinuous, swollen and tortile appearance. Arteriovenous shunt also can be found. All of those disturbances contribute to the leakage of macromolecular substance and the formation of high interstitial fluid pressure within the tumor tissue which is the major barrier for the transport of MAb . It was reported that the central interstitial fluid pressure of tumor is 30-40mmHg higher than that of the periphery. In addition, the high molecular weight (150-200KD) also restricts MAb’s transportation. That is why MAb can not reach its effective concentration in situ. Protein engineering offers a way to rebuilt or “minimize” the anti-human VEGF MAb in order to increase its tumor penetration ability. Sing-chain Fv (ScFv) is becoming the most attractive project for such purpose. ScFv is the recombinant antibody fragments consisting of only the variable light chain (VL) and variable heavy chain (VH) domains covalently connected to one another by a polypeptide linker. Since the specificity and affinity of antibody are determined by its variable region, ScFv is the smallest fragment with the whole antigen binding site, its molecular weight is only 1/6 of the integral antibody. For a lack of conservative region (Fc), ScFv not only has an increased tumor penetration ability but also has a decreased immunogenicity ).

The aim of this study is to construct a gene ingeneering anti-human VEGF ScFv and detect parts of its bioactivity. In this paper, we generated a hybridoma cell line (E11) secreting Mab against VEGF by routine methods using synthetic peptide derived from the predicted human VEGF amino acid sequence as antigen. The Mab was proved to be the ability to specifically bind native VEGF and block the VEGF-mediated stimulation of endothelial cell growth. Then, the Mab was used to investigate VEGF expression in 20 specimens from buccal mucosal cancer patients by Immunohistochemical methods. The results showed: VEGF expression was detected in 16 (80%) out of total 20 samples. 12 specimens (high and moderate differentiation) were immunoreactive to VEGF. 4 specimens (low differentiation) were strongly stained. VEGF was mainly localized in the cytoplasm of carcinoma cell and endothelial cells of small vessels adjacent to malignant tumor cells. The intense expression of VEGf by buccal nucosal cancer provides strong evidence linking VEGF expression to the angiogenesis associated with buccal mucosal cancer. The Mab also significantly supressed the growth of cultured buccal mucosal cancer cell in BALB/c nude mice with s.c.inoculation. (data not included here).

Since the early description of monoclonal antibodies by Kohler and Milstein in 1975, there has been considerable interest in their adoption for therapy application. But when the Mab administered to patients, the mouse monoclonal antibodies stimulate the production of a human antigody response-Human Anti-Mouse Antibody (HAMA). HAMA diminishes or obliterates the efficacy of the mouse antibody. One solution to the problem is the preparation of recombinant antibodies. Now, It is possible to raise small antibody fragments, such as single-chain Fvs (scFvs) against virtually any desired antigen. scFv can maintain specific binding ability of antigen as its parent Mab. These binding molecules are being used to target a wide variety of antigens for numerous diagnostic and therapeutic applications. For above reasons, so we intend to build scFv from the hybridoma cell line secreting Mab against VEGF, total RNA was prepared and used as a template for cDNA synthesis and cloning. Reversetranscription and polymerize chain reaction (RT-PCR) was used to clone the immunoglobulin variable region gene (VH, VL). The cloned VH and VL were linked by well documented flexible peptide bridging sequence (GGGGS)₃ in the VH-linker-VL orientation. The constructed scFv gene was inserted in pET-15S vector with a short sequence encoding a 12-amino-acidpeptide sequence (E-tag) at the 3’-end. The E-tag attachment at the C-end made it easy to test the avidity of scFv, and expressed in E.coli. Biding activity of scFv was evaluated by immunohistochemical methods after its denature and renature treatment. The sequence analysis revealed that the full length of VH is 369bp, and VL, 333bp. According to Kabat classification, VH and VL are member of mouse Ig variable gene heavy chain subgroupⅡ(A) and light chain subgroupⅢ, respectively scFv was highly produced in the from of inclusion bodies which activity was recovered after being renatured. Immunohistochemical results showed that scFv retained almost the same antigen affinity and specificity as its parent monoclonal antibody. Our results showed that recombinant antibody scFv is potentially useful in both diagnostic and therapeutic application.

Lastly, a three-demensional graphic model of scFv was constructed by homology modeling on workstation. Then a systematisc analysis was performed to determine the important residues in Fv. This study provides a method for comprehending the relationship between antibody structure and function, and benefit humanization of scFv in further research.

The scientific research on the antibody humanization in this decade suggested that the foundation of antibody humanization is the acquirement of the primary structure and the clarification of the regularity of the residues’ interaction. It is certain that these investigations will lead to a new field of tumor vasculature targeted therapy by antibody mediation and many valuable reagents for additional studies in the anti-angiogenesis cancer treatment

The futher study of our team is to purify the ScFv and detect its properties of tumor penetration and inhibition.