Angiogenesis and Tumor Growth

我们重点关注了4个相关领域。1)我们之前的研究已经确定了Notch1和Notch4的内皮特异性膜结合配体Delta4 （Dll4）作为内皮细胞功能调节剂的关键作用。Dll4在发育中的内皮细胞中选择性表达，是正常血管发育所必需的。出生后，Dll4在血管生成内皮中表达，特别是在肿瘤血管中。我们发现Dll4是血管生成的负调节因子，因为它通过下调VEGF-A的主要信号受体VEGFR-2和共受体neuropilin-1 （Npn-1）来作为VEGF-A的选择性抑制剂。在临床前癌症模型中，我们已经证明Dll4可以通过降低肿瘤血管内皮中VEGF-A的反应来显著减少肿瘤血管生成和淋巴源性肿瘤的生长。在相关实验中，我们已经开始探索Notch配体JAG2在血管生成中的作用。为此，我们开发并继续完善了一种新的jag2缺失小鼠模型，并探索了notch依赖性和notch非依赖性通路在内皮细胞功能和血管生成中的潜在贡献。初步结果为Jag2在维持肠内皮干细胞生态位中的作用提供了强有力的证据。2)我们继续研究ephrinB配体及其EphB受体如何在血管系统中协调内皮/内皮/周细胞组装。EphrinB配体是表面结合的；b型Eph/Ephrin相互作用中的受体-配体相互作用涉及相邻细胞（反式）或可以发生在同一细胞（顺式）。b型ephrin除了激活其同源EphB受体外，当受体通过“反向信号传导”参与时，b型ephrin还可以作为信号分子发挥作用。Eph受体是酪氨酸激酶与其膜锚定的ephrin配体相互作用。我们研究了Eph/ephrin信号在内皮细胞存活调控中的潜在作用。我们发现，沉默EphrinB的表达或酪氨酸磷酸化缺陷突变体EphrinB的表达（包含所有酪氨酸残基的替代，阻止尾部磷酸化，并作为内源性WT ephrin的显性阴性抑制剂）会导致内皮细胞死亡。这种结果不能通过添加外源性VEGFA或FGF2来阻止。生化和遗传学实验表明，这种死亡是由JNK3/MAPK10信号介导的，EphrinB2酪氨酸磷酸化依赖性信号是MAPK10/JNK3表达的调节剂。因此，JNK3的沉默可以防止EphrinB信号缺陷的内皮细胞的细胞死亡。与这些结果一致的是，在基因缺乏EphrinB2的小鼠中，细胞死亡增加与JNK3激活有关，JNK3缺乏的小鼠表现出与EphrinB2信号缺乏相对应的眼部血管缺陷。这些结果提供了证据支持EphrinB信号作为内皮促生存途径和抑制血管生成的治疗靶点的作用。在此基础上，我们进一步探索了靶向肿瘤血管中EphrinB2信号通路诱导血管消退、促进肿瘤细胞饥饿坍缩的可能性。3)根据这一观察结果，我们探索了不同的方法来阻断ephrinb2在血管系统中衍生的促生存信号。我们已经确定磷酸酶SHP2是内皮细胞中EphrinB2促存活功能的重要介质。此外，我们已经确定了SHP2变抗抑制剂SHP099作为磷酸化- ephrinb2 - stat信号的有效抑制剂和内皮细胞死亡的选择性诱导剂在体外和体内。我们已经描述了内皮细胞中SHP2抑制的信号后果，并利用这一信息选择性地靶向肿瘤血管而不是肿瘤细胞。此外，我们已经确定TIE2（血管生成素1和2的受体）的内皮磷酸化是顺式内皮EphrinB2磷酸化的来源。在此基础上，我们分析了SHP099与肽体AMG386联合阻断SHP2和TIE2的联合抗血管生成活性；4)在早期的观察中，我们已经将肿瘤抑制蛋白DLC1的缺失与应激条件下原代内皮细胞存活率的增加联系起来。我们现在发现dcl1是原代人内皮细胞细胞接触抑制增殖的关键调节因子，当细胞达到高密度时促进细胞死亡。由于NF-kB激活增加与肿瘤坏死因子α诱导蛋白3 （TNFAIP3/A20）信号传导增加相关，DLC1缺失赋予融合内皮细胞促生存表型，而非稀疏内皮细胞。与DLC1缺失在内皮细胞肿瘤发生中的作用一致，我们发现在人血管肉瘤中DLC1异常低而TNFAIP3/A20异常高。NF-kB抑制剂Tanespimycin/17-AAG的实验治疗可显著降低小鼠血管肉瘤肿瘤的生长。这些结果表明，dcl1是血管肉瘤中内皮细胞接触抑制增殖的一种先前未被识别的调节因子，并为靶向NF-kB治疗dcl1缺失的血管肉瘤提供了证据。5)基于肿瘤抑制因子DLC1蛋白（我们的结果）和转录共激活因子YAP（文献）调控细胞接触抑制生长的观察，我们探索了DLC1和YAP之间的生化相互作用。我们发现DLC1是YAP的调节因子，并且YAP的转录共激活因子功能是缺失DLC1的内皮细胞所表现的细胞接触抑制丧失所必需的。如果内皮细胞中YAP缺失，dlc1缺失的内皮细胞在融合时停止生长，不能堆积。这些结果在体外得到了证实，血管肉瘤组织中含有相当比例的dlc1阴性恶性内皮细胞，其中YAP是核的和活跃的。在肿瘤附近的正常皮肤血管中则不是这种情况。维替波芬是一种YAP抑制剂，在小鼠血管肉瘤实验模型中显示出明显的抗肿瘤作用。6)关于dcl1在内皮细胞中的作用的其他正在进行的实验已经促进了内皮特异性诱导dcl1缺陷小鼠细胞系的产生。我们目前正在评估这种缺陷在不同情况下的作用，包括小鼠发育、造血中的生理性内皮细胞功能、伤口愈合和癌症。

We have focused on 4 related areas. 1) Our previous studies have identified a critical role of Delta4 (Dll4), an endothelial-specific membrane-bound ligand for Notch1 and Notch4, as a regulator of endothelial cell function. Dll4 is selectively expressed in the developing endothelium and is required for normal vascular development. Post-natally, Dll4 is expressed in the angiogenic endothelium, particularly in the tumor vasculature. We have found that Dll4 is a negative regulator of angiogenesis, as it functions as a selective inhibitor of VEGF-A by down-regulating the principal VEGF-A signaling receptor, VEGFR-2 and co-receptor neuropilin-1 (Npn-1). In pre-clinical cancer models, we have documented that Dll4 can markedly reduce tumor angiogenesis and the growth of tumors of lymphoid origin by reducing VEGF-A responses in the tumor vascular endothelium. In related experiments, we have begun to explore the role of the Notch ligand JAG2 in angiogenesis. To this end, we have developed and continue to perfect a new mouse model of JAG2-deficiency and explored the potential contribution of Notch-dependent and Notch-independent pathways in endothelial cell function and angiogenesis. Preliminary results provide strong evidence for a role of Jag2 in the maintenance of the endothelial intestinal stem cell niche. 2) We have continued investigations on how ephrinB ligands and their EphB receptors orchestrate endothelial/endothelial/pericyte assembly in the vasculature. EphrinB ligands are surface-bound; receptor-ligand interactions in the B-type Eph/Ephrin interactions involve adjacent cells (trans) or can occur in the same cell (cis). In addition to activating their cognate EphB receptors, B-type Ephrins can function as signaling molecules when engaged by the receptor through "reverse signaling". Eph receptors are tyrosine kinases interacting with their membrane-anchored ephrin ligands. We have investigated the potential role of Eph/ephrin signaling in the regulation of endothelial cells survival. We have found that silencing EphrinB expression or expression of a tyrosine-phosphorylation-deficient mutant EphrinB (contains substitutions of all tyrosine residues that prevent tail phosphorylation and acts as a dominant-negative inhibitor of endogenous WT ephrin) causes endothelial cell death. This outcome cannot be prevented by the addition of exogenous VEGFA or FGF2. Biochemical and genetic experiments have revealed that such death is mediated by JNK3/MAPK10 signaling, and that EphrinB2 tyrosine phosphorylation-dependent signaling serves as a modulator of MAPK10/JNK3 expression. Thus, the silencing of JNK3 prevents cell death in endothelial cells that are EphrinB signaling-deficient. Consistent with these results, the hyaloid vasculature in mice genetically-deficient of EphrinB2 undergoes increased cell death in association with JNK3 activation, and JNK3-deficient mice display ocular vascular defects that mirror those of EphrinB2 signaling deficiency. These results provide evidence supporting a role for EphrinB signaling as an endothelial pro-survival pathway and a therapeutic target for inhibition of angiogenesis. Based on this evidence, we have further explored the possibility of targeting EphrinB2 signaling in the tumor vasculature to induce vessel regression and promote tumor cell starvation of collapse. 3) Pursuing this observation, we have explored different approaches to block EphrinB2-derived pro-survival signals in the vasculature. We have identified the phosphatase SHP2 as an essential mediator of EphrinB2 prosurvival functions in endothelial cells. Further, we have identified the SHP2 allosteric inhibitor, SHP099 as a potent inhibitor of phospho-EphrinB2-STAT signaling and a selective inducer of endothelial cell death in vitro and in vivo. We have characterized the signaling consequences of SHP2 inhibition in endothelial cells and exploited this information to selectively target the tumor vasculature rather than the tumor cells. Furthermore, we have identified endothelial phosphorylation of TIE2 (receptor for Angiopoietins 1 and 2) as a source of endothelial EphrinB2 phosphorylation in cis. Based on this information we have analyzed the combined anti-angiogenic activity of SHP2 and TIE2 blockade using SHP099 in combination with the peptibody AMG386; 4) In earlier observations we have linked the loss of the tumor-suppressor protein DLC1 with increased survival in primary endothelial cells under conditions of stress. We now discovered that DLC1 is a critical regulator of cell contact inhibition of proliferation in primary human endothelial cells, promoting cell death when the cells reach high density. DLC1 depletion confers a pro-survival phenotype to confluent, but not sparse endothelial cells, attributable to increased NF-kB activation associated with increased tumor necrosis factor alpha-induced protein 3 (TNFAIP3/A20) signaling. Consistent with a role of DLC1 depletion in endothelial cell tumorigenesis, we found that DLC1 is abnormally low and TNFAIP3/A20 is abnormally high in human angiosarcoma. Experimental treatment with the NF-kB inhibitor Tanespimycin/17-AAG significantly reduced angiosarcoma tumor growth in mouse. These results identify DLC1 as a previously unrecognized regulator of endothelial cell contact inhibition of proliferation that is depleted in angiosarcoma, and provide evidence supporting the targeting of NF-kB for the treatment of angiosarcoma where DLC1 is lost. 5) Based on the observation that the tumor suppressor DLC1 protein (our results) and the transcriptional co-activator YAP (literature) regulate cell-contact inhibition of growth, we have explored biochemical interactions between DLC1 and YAP. We found that DLC1 is a regulator of YAP and that the transcriptional co-activator function of YAP are required for the loss of cell-contact inhibition manifested by DLC1-deficient endothelial cells. If YAP is depleted from endothelial cells, DLC1-depleted endothelial cells stop growing when confluent and fail to pile-up. These results in vitro were corroborated by the observation that angiosarcoma tissues contain a significant proportion of DLC1-negative malignant endothelial cells where YAP is nuclear and active. This is not the case in the normal skin vasculature adjacent to the tumor. Verteporfin, an inhibitor of YAP, displayed a clear anti-tumor effect in an experimental model of angiosarcoma in mice. 6) Additional ongoing experiments on the role of DLC1 in endothelial cells have prompted the generation of an endothelial-specific inducible DLC1-deficient mouse cell line. We are currently evaluating the role of such deficiency in different contexts, including mouse development, physiologic endothelial cell functions in hematopoiesis, wound healing and cancer.