Immunotherapy of tumors with xenogeneic endothelial cells as a vaccine

Endothelial cells in the angiogenic vessels in solid tumors express proteins on their surfaces that are absent or barely detectable in normal quiescent vascular endothelium, including v3 integrin and receptors for certain angiogenic growth factors^{2, 4, 8, 9}. The proteins on the endothelium of new vessels in the mouse are homologous to those in humans and in other species, to varying extents^{14, 15, 16}. The breaking of immune tolerance to autologous angiogenic endothelial cells should be a useful approach for cancer therapy. However, immunity to angiogenic vessels is presumably difficult to elicit by autologous or syngeneic endothelial cells or their proteins as vaccine because of the immune tolerance acquired during the development of the immune system. Here we explored the feasibility of immunotherapy of tumors with xenogeneic endothelial cells as a vaccine by breaking immune tolerance against autologous angiogenic cells in a cross-reaction between the xenogeneic homologs and self molecules.

We immunized mice intraperitoneally once weekly for 4 weeks with different doses of HUVECs, HDMVECs, HUV-EC-C, GEN-T cells, SVEC4-10 cells, T/G HA-VSMCs and RPMI 7666 cells or treated them with PBS alone (non-immunized mice), and then challenged with 1 10⁵−1 10 ⁷ live tumor cells after the fourth immunization. Tumor grew progressively in all non-immunized mice and mice immunized with SVEC4-10 cells, T/G HA-VSMCs or RPMI 7666 cells, but there was complete protection from tumor growth in mice immunized with HUVECs, HDMVECs, HUV-EC-C or GEN-T cells ( Fig. 1). The protective effect was long-lasting, as mice of different strains (BALB/c, C57Bl/6 and C3H) challenged up to 12 weeks after the last immunization resisted challenge with several solid tumor cells of different histological origin (data not shown).

Figure 1. Induction of protective anti-tumor immunity. BALB/C mice were immunized intraperitoneally with fixed endothelial cells or control cells, and then challenged with live Meth A cells. a, treatment with PBS alone (control). b−h, Vaccination with HUVECs (b), HUV-EC-Cs (c), HDMVECs (d), GEN-T cells (e), SVEC4-10 cells (f), T/G HA-VSMCs (g) and RPMI 7666 cells (h). Full Figure and legend (19K)

Figure 2. Induction of the therapeutic anti-tumor immunity. Meth A fibrosarcoma cells (a and e), H22 hepatoma cells (b and f) and MA782/5S mammary carcinoma cells (c, d, g and h) were introduced subcutaneously into mice, and then the mice were treated with fixed SVEC4-10 cells (), HUVECs or GEN-T cells () or PBS alone (). a, b and d, Tumor sizes in mice treated with HUVECs (). e, f and h, Survival of mice treated with HUVECs (). c and g, Tumor sizes and survival of mice treated with GEN-T cells (). Full Figure and legend (25K)

Figure 3. The inhibition of proliferation of endothelial cells in vitro with immunoglobulin. a, Exponentially growing HUVECs (),SVEC4-10 cells () or Meth A cells () were exposed to immunoglobulins isolated from mice immunized with HUVECs, and the percentage inhibition was calculated. b, HUVECs (),SVEC4-10 cells () or Meth A cells () were treated with immunoglobulins isolated from non-immunized mice. c, HDMVECs (), HUV-EC-Cs (),GEN-T cells () and LL/2 cells () were also treated with immunoglobulins isolated from mice immunized with HUVECs. d, As a control of the experiment in c, cells were treated with immunoglobulins isolated from non-immunized mice. Horizontal axes, immunoglobulin concentrations. Full Figure and legend (17K)

Figure 4. Adoptive transfer of immunoglobulins in vivo. a, Protective anti-tumor effect on H22 hepatoma was tested with purified immunoglobulins from mice immunized with HUVECs and control cells, the immunoglobulins were adsorbed with (+) HUVECs, SVEC4-10 cells and T/G HA-VSMCs before adoptive transfer. b, Protective effect on Lewis lung carcinoma (LL/2) with immunoglobulins from mice immunized with GEN-T cells and control cells, and immunoglobulins adsorbed with (+) GEN-T cells, SVEC4-10 cells and T/G HA-VSMCs. Full Figure and legend (25K)

Figure 5. In situ observation of microvessels and inhibition of angiogenesis. a−d, Immunohistochemical analysis of microvessels in tumor tissues, stained with immunoperoxidase. a and b, Microvessels in hepatoma tissues (a) and granulation tissues of healing wound ( b) from non-immunized mice, using serum from the mice immunized with HUVECs. c, Microvessels in kidney (control). d, Endothelial deposition of immunoglobulins in the hepatoma tissue from a mouse immunized with HUVECs. e, Sequential analysis of inhibition of angiogenesis in tumors. Mice were treated with PBS alone () or with fixed SVEC4-10 cells (), HUVECs () or GEN-T cells () after H22 hepatoma cells were introduced subcutaneously. Vessel density was determined by counting of the microvessels in each high-power field (hpf) in the sections. Full Figure and legend (69K)

We sequentially analyzed microvessel density as tumors regressed in response to the vaccine (Fig. 5e). Microvessel density gradually decreased as a result of prolongation of the vaccine treatment (Fig. 5e). Also, vessel length, 'clock-hours' (the proportion of the circumference that is vascularized if the eye is viewed as a clock) and area of neovascularization (assessed by Corneal micro-pocket assay) were inhibited by 68 6%, 72 5% and 81 7%, respectively, in mice treated by the adoptive transfer of immunoglobulins isolated from mice immunized with HUVECs, compared with control samples. Also, we obtained similar results by direct immunization with HUVECs.

Figure 6. Abrogation of the anti-tumor activity and immunoglobulin subclass response to the endothelial cells by the depletion of CD4+ T lymphocytes. a, Abrogation of anti-tumor activity. Mice were immunized and challenged with MA782/5S mammary carcinoma cells after depletion of immune cell subsets. b, Abrogation of immunoglobulin subclass response to endothelial cells. Sera obtained from mice immunized with HUVECs were tested against lysates of SVEC4-10 cells by ELISA, after depletion of immune cell subsets. Immunoglobulin detected: , IgG1; , IgG2a; , IgG2b; , IgM; , IgA. Data represent absorption at 560 nm. Anti-, antibody against; +, molecule used for subset depletion. Full Figure and legend (20K)

The endothelial cell extracts showed multiple positive bands by western blot analysis when probed with sera from mice immunized with HUVECs (Fig. 7a) but negative staining with those from non-immunized mice (Fig. 7b). At least two bands with molecular sizes of 220 and 130 kDa had sizes similar to those of the known angiogenesis-associated molecules vascular endothelial growth factor receptor (VEGFR) II and v integrin. These two molecules were also expressed on the endothelial cells we used, as shown by the use of commercially available antibodies against VEGFR II or v integrin in flow cytometric and western blot analysis, and immunoglobulins isolated from mice immunized with HUVECs showed positive reactions against the recombinant extracellular parts of VEGFR II and v integrin by ELISA (data not shown). Sequence comparison analysis using the SwissProt database in NCBI indicated that the primary sequences of VEGFR II and v integrin of mice and humans were homologs that were 82% and 89% identical, respectively, at the amino-acid level. Next, we selected pairs of peptides for synthesis from the regions that shared the most-identical amino-acid sequences between humans and mice. Each peptide synthesized was 35 amino acids long. We screened the possible peptides responsible for the cross-reaction in two steps. We probed these peptides with immunoglobulins from mice immunized with HUVECs, using ELISA. We immunized mice with pairs of homologous peptides that showed immunoglobulin-positive binding and determined their anti-tumor activity. Three pairs of the homologous peptides showed immunoglobulin-positive binding, and their human homologs showed anti-tumor activity. We identified two pairs within v integrin and one pair within VEGFR II ( Table 1). These immunoglobulin-binding regions were represented by amino-acid residues 330−364 and 545−579 within the extracellular fragment of v integrin in both human and mouse, and by residues 239−273 in human and the corresponding residues 243−277 in mouse within VEGFR II. We identified these immunoglobulin-positive binding peptides from 42 pairs of homologous peptides examined. Furthermore, adoptive transfer of immunoglobulins isolated from these mice immunized with xenogeneic peptide also showed inhibition of tumor growth, but immunoglobulins from mice immunized with mouse peptides or with Freund's adjuvant alone had no effect compared with those from mice treated with PBS alone (Fig. 7c).

Figure 7. Identification of the possible antigens responsible for the cross-reaction. a and b, Western blot analysis for possible antigens of both HUVECs and SVEC4-10 cells recognized by the immune sera and control sera respectively. a, HUVECs, SVEC4-10 cells and Meth A cells, stained with sera isolated from mice immunized with HUVECs. b, The cells in a, stained with sera from non-immunized mice (control). Left margin, molecular sizes. c, Inhibition of tumor growth by adoptive transfer of immunoglobulins. There is a protective anti-tumor effect with immunoglobulins isolated from the mice immunized with xenogeneic peptides, but not from mouse peptides, Freund's adjuvant or PBS alone. Data represent day 20 after tumor cell injection. Full Figure and legend (63K)

Table 1. The amino-acid sequence of cross-reactive peptides in v integrin and VEGFR II Full Table

Here we obtained many results regarding xenogeneic endothelial cell vaccines, anti-tumor immunity and angiogenesis. Fixed xenogeneic endothelial cells as vaccine induced both protective and therapeutic anti-tumor immunity. The autoreactive immune response against the microvessels in solid tumors may be provoked in a cross-reaction by immunization of xenogeneic endothelial cells, and the autoreactive immunity targeting to microvessels in solid tumor was probably responsible for the anti-tumor activity. These suggestions were supported by our results. Endothelial cell proliferation was inhibited in vitro by purified immunoglobulins from xenogeneic endothelial cell- immunized mice. Anti-tumor activity and inhibition of angiogenesis was acquired by adoptive transfer of purified immunoglobulins. Immunoglobulins present in sera positively stained microvessels in the tumor and endothelial cell lines of both human and mouse. We identified cross-reactive peptides within v integrin and VEGFR II on endothelial cells, and similar molecules in both human and mouse endothelial cell lines were recognized by western blot analysis. IgG1, IgG2a and IgG2b were substantially increased in response to the endothelial cells. There was endothelial deposition of immunoglobulins in tumor. There was also anti-tumor activity and production of immunoglobulins against the endothelial cells that was abrogated by the depletion of CD4⁺ T lymphocytes. Angiogenesis was apparently inhibited in tumor ( Fig. 5e), and corneal angiogenesis induced by basic fibroblast growth factor was inhibited. In addition, the anti-tumor activity with xenogeneic endothelial cells may not result from the nonspecifically augmented immune response against tumor growth in host mice, as we found no increase in natural killer cell activity of spleen cells or in levels of cytokines such as interferons , and , tumor necrosis factor and chemokines in sera from immunized mice (data not shown), and found no anti-tumor activity after immunization with a xenogeneic smooth muscle cell line or B-lymphoblastoid cells.

Here we prepared vaccines using proliferative endothelial cells cultured in vitro, like new vessels with proliferative activity in solid tumors. Endothelial cells in culture may be heterogenous and express genes that may not be expressed in the original tissue, whereby they may lose the expression of a number of antigens. However, we found that immunoglobulin or serum isolated from mice immunized with cultured xenogeneic endothelial cells showed positive staining not only for cultured endothelial cells but also for microvessels in tumor tissues or in a healing wound. These findings indicate that there may still be some common antigens or cross-reactive epitopes between cultured endothelial cells and those in tumor tissues, which are responsible for the autoimmune response against the tumor endothelium in a cross-reaction. This is also supported by the findings that the cultured endothelial cells can still express some genes that are expressed in the microvessels in tumor tissues. For example, molecules such as VEGFR II, v3 integrin and endoglin, which are associated with angiogenesis in tumors, can be found in primary culture of HUVECs, HDMVECs and bovine endothelial cells as well as HUV-EC-Cs and some bovine endothelial cell lines^{25, 26, 27, 28, 29, 30, 31, 32}. Some of these molecules were also present on the endothelial cells used here.

The molecules v integrin and VEGFR II are important during angiogenesis^{2, 3, 4, 10, 12, 29, 33}. Blockade of the ligand binding domain of these molecules results in the inhibition of angiogenesis in vivo or of endothelial cell proliferation in vitro and of anti-tumor activity^{2, 3, 4, 10, 12, 29, 33}. Here, at least two bands by western blot analysis showed sizes similar to those of VEGFR II and v integrin. We also identified these two molecules on the endothelial cells using commercially available antibodies against VEGFR II and v integrin, by flow cytometric and western blot analysis, and immunoglobulins isolated from mice immunized with HUVECs showed positive reactions against the recombinant extracellular parts of VEGFR II and v integrin by ELISA (data not shown). We identified three pairs of possible peptides responsible for cross-reaction in the extracellular parts of these molecules. Furthermore, two of three pairs of immunoglobulin-binding sites were located within the regions encompassing the ligand-binding domain (residues 247−261) of VEGFR II (ref. 29) and partially encompassing the ligand-binding domain (residues 139−349) of v integrin³⁴. The other immunoglobulin-binding site was located outside the ligand-binding domain of v integrin. However, some antibodies against the non-binding domain can block the function of the integrin allosterically as well³⁵. Immunoglobulins isolated from the mice immunized with xenogeneic peptides of v integrin and VEGFR II identified v integrin and VEGFR II, respectively, on the endothelial cells, and we demonstrated inhibition of endothelial cell proliferation in vitro (data not shown). Also, the adoptive transfer of immunoglobulins isolated from these mice immunized with xenogeneic peptides showed inhibition of tumor growth. The findings described above indicate that the cross-reaction seen here may involve in part the epitopes within VEGFR II and v integrin on endothelial cells. The other four bands on the western blot were difficult to match to sizes of known angiogenesis-associated molecules. Whether they belong to new angiogenesis-associated molecules for cross-reaction must be explored further. The findings described above also indicate that the high potency of the polyclonal serum isolated from xenogeneic whole endothelial cells found here may result from the blockade of some important angiogenesis-associated molecules such as v integrin and VEGFR II, and may involve targeting to multiple sites, as angiogenesis is a complex process involving many molecules on new vessels.

	Top

Mice were immunized intraperitoneally or subcutaneously once weekly for 4 continuous weeks with different doses (1 10²−1 10⁷ cells/mouse) of a variety of endothelial cell vaccines or T/G HA-VSMCs as well as RPMI 7666 cells (as a control) or were treated with PBS alone (non-immunized). Mice were then challenged with 1 = 10⁵−1 10 ⁷ live tumor cells after the fourth immunization. For investigation of the therapeutic effect against the established tumors, ten mice in each group were treated with intraperitoneal injection of the vaccines or control cells or PBS alone twice weekly for 4 weeks starting a day 7 after subcutaneous introduction of 1 10⁶ live tumor cells. The Meth A fibrosarcoma and H22 hepatoma models were established in BALB/c mice. MA782/5S and FM3A mammary carcinoma models were established in BALB/c and C3H mice, respectively. The Lewis lung carcinoma model was in C57Bl/6 mice. All studies involving mice were approved by the institute's Animal Care and Use Committee.

Immunoglobulins were purified from the pooled sera derived from the mice on day 7 after the fourth immunization or from control mice, by affinity chromatography (CM affi-gel blue gel kit; BioRad, Richmond, California). For determination of the effects of purified immunoglobulins on cell proliferation, exponentially growing endothelial cells or tumor cells, at a concentration of 2 10⁵ cels/ml were exposed to various concentrations (1−150g/ml) of the imunoglobulin for 72 h of culture. The number of viable cells was determined by a trypan blue dye exclusion test, and the percentage inhibition was calculated³⁷.

For assessment of the efficacy of imunoglobulin in anti-tumor in vivo , purified imunoglobulins (10−300 mg/kg) were adoptively transferred intravenously 1 d before mice were challenged with 1 10⁵−1 10⁷ tumor cells, and then mice were treated twice per week for 3 weeks. As a control, immunoglobulin was adsorbed four times by incubation for 1 h at 4 °C, with rocking, with fixed xenogeneic endothelial cells or VSMCs.

Immunoglobulin sublass was determined by ELISA as described³⁸. Endothelial cells and tumor cells were washed and lysed by three cycles of freezing and thawing, and then 1 10⁴ cell equivalents (50 l) were plated, dried and blocked. Experimental mouse sera were serially diluted and added to the wells. Plates were incubated for 2 h at 37 °C, washed, and then incubated with serially diluted alkaline phosphatase-conjugated antibody against mouse IgG subclass, IgM or IgA. Enzyme activity was measured with an ELISA reader (BioRad, Richmond, California). For the identification of cross-reactive peptides, the peptides were coupled to 96-well assay plates at a concentration of 1 g/well, as described³⁹, then probed with immunoglobulins from mice immunized with HUVECs, by ELISA.

Immune cell subsets were depleted as described⁴⁰. Mice were injected intraperitoneally with 500 g monoclonal antibodies against CD4 (clone GK 1.5, rat IgG), CD8 (clone 2.43, rat IgG) or natural killer cells (clone PK136) or isotype controls 1 d before the immunization, and then twice per week for 3 weeks. Tumor cells (1 10⁶−1 10⁷) were challenged after the fourth immunization. These hybridomas were obtained from American Type Culture Collection (Rockville, Maryland). The depletion of CD4⁺, CD8⁺ and natural killer cells was consistently greater than 98%, as determined by flow cytometry (Coulter Elite ESP; Coulter, Hialeah, Florida)⁴⁰.

Western blot analysis was done as described⁴³. Cells (2 10⁷) were lysed in 1 ml lysis buffer. The membrane blots were blocked by incubation at 4 °C in 5% non-fat dry milk, then were washed and probed with mouse sera at a dilution of 1:100. Blots were then washed and incubated with a biotinylated secondary antibody (Biotinylated horse antibody against mouse IgG or IgM, followed by transfer to Vectastain ABC solution (Vector Laboratories, Burlingame, California).

Peptides (100 g antigen per injection) were emulsified 1:1 (volume/volume) with complete Freund's adjuvant for the first immunization, followed by a boost in incomplete Freund's adjuvant at 2 weeks and weekly thereafter, as described⁴⁶, then challenged with 1 10⁵−1 10⁷ live tumor cells after the fourth immunization. Purified immunoglobulins were prepared from the pooled sera derived from the mice at day 7 after the fourth immunization without injection of tumor cells.

	Top

1. Folkman, J. What is the evidence that tumors are angiogenesis dependent? J. Natl. Cancer Inst. 82, 4−6 (1990). | Article | PubMed  | ChemPort |
2. Ferrara, N. & Alitalo, K. Clinical application of angiogenic growth factors and their inhibitors. Nature Med. 5, 1359−1364 (1999) | Article | PubMed  | ISI | ChemPort |
3. Risau, W. Mechanism of angiogenesis. Nature 386, 671−674 (1997) | Article | PubMed  | ISI | ChemPort |
4. Zetter, B.R. Angiogenesis and tumor metastasis. Annu. Rev. Med. 49, 407−424 (1998) | Article | PubMed  | ISI | ChemPort |
5. Cao, R., et al. Suppression of angiogenesis and tumor growth by the inhibitor K1-5 generated by plasmin-mediated proteolysis. Proc. Natl. Acad. Sci. USA 96, 5728−5733 (1999) | Article | PubMed  | ChemPort |
6. Bicknell, R. in Tumor Angiogenesis. (eds. Bicknell, R., Lewis, C.E. & Ferrara. N.) 19−28 (Oxford University Press, Oxford, 1997). | ChemPort |
7. O'Reilly, M.S. et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 88, 277−285 (1997) | Article | PubMed  | ChemPort |
8. Kerbel. R. S. A cancer therapy resistant to resistant. Nature 390, 335 (1997) | Article | PubMed  | ISI | ChemPort |
9. Fidler. I.J. & Ellis, L. M. The implications of angiogenesis for the biology and therapy of cancer metastasis. Cell 79, 185−188 (1994). | Article | PubMed  | ISI | ChemPort |
10. Arap, W., Pasqualini, R. & Ruoslahti, E. Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science 279, 377−379 (1998); | Article | PubMed  | ISI | ChemPort |
11. Fan, T.P., Jaggar, R. & Bicknell, R. Controlling the vasculature: angiogenesis, anti-angiogenesis and vascular targeting of gene therapy. Trends Pharmacol. Sci. 16, 57−66 (1995) | Article | PubMed  | ISI | ChemPort |
12. Kim, K.J. et al. Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumor growth in vivo. Nature 362, 841−844 (1993) | Article | PubMed  | ISI | ChemPort |
13. Hanahan D. & Folkman, J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. J. Natl. Cancer Inst. 88, 1091−1092 (1996). | PubMed  |
14. Quinn, T.P., Peters, K.G., De Vries, C., Ferrara, N. & Williams, L.T. Fetal liver kinase 1 is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium. Proc. Natl. Acad. Sci. USA 90, 7533−7537 (1993) | PubMed  | ChemPort |
15. Djaffar, I. et al. A new alternative transcript encodes a 60 kDa truncated form of integrin beta 3. Biochem. J. 15, 69−74 (1994)
16. Eichmann, A., Marcelle C., Breant, C. Le Douarin, N.M. Molecular cloning of Quek 1 and 2, two quail vascular endothelial growth factor (VEGF) receptor-like molecules. Gene 174, 3−8 (1996) | Article | PubMed  | ISI | ChemPort |
17. Barth, R.J.J., Bock, S.N., Mule, J.J. & Rosenberg, S.A. Unique murine tumor-associated antigens identified by tumor infiltrating lymphocytes. J. Immunol. 144, 1531−1537 (1990). | PubMed  | ISI | ChemPort |
18. Boon, T., Coulie, P.G. & Van den Eynde, B. Tumor antigens recognized by T cells. Immunol. Today 18, 267−268 (1997). | Article | PubMed  | ISI | ChemPort |
19. Rosenberg, S.A. Cancer vaccines based on the identification of genes encoding cancer regression antigens. Immunol. Today 18, 175−178 (1997). | Article | PubMed  | ISI | ChemPort |
20. Murray, J.S. How the MHC selects Th1/Th2 immunity. Immunol. Today 19, 157−163 (1998). | Article | PubMed  | ISI | ChemPort |
21. Romagnani, S. The Th1/Th2 paradigm. Immunol. Today 18, 263−266 (1997). | Article | PubMed  | ISI | ChemPort |
22. Ohashi, P.S. T cell selection and autoimmunity: flexibility and tuning. Curr. Opin. Immunol. 8, 808−814 (1996). | Article | PubMed  | ISI | ChemPort |
23. Kumar, V., Stellrecht, K. & Sercarz, E. Inactivation of T cell receptor peptide-specific CD4 regulatory T cells induces chronic experimental autoimmune encephalomyelitis (EAE). J. Exp. Med. 184, 1609−1617 (1996). | Article | PubMed  | ISI | ChemPort |
24. De Silva, H.D., Van Driel, I.R., La Gruta, N., Toh, B.H. & Gleeson, P.A. CD4+ T cells, but not CD8+ T cells, are required for the development of experimental autoimmune gastritis. Immunology 93, 405−408 (1998). | Article | PubMed  | ISI | ChemPort |
25. Marconcini, L. et al. C-fos-induced growth factor/vascular endothelial growth factor D induces angiogenesis in vivo and in vitro. Proc. Natl. Acad. Sci. USA 96, 9671−9676 (1999). | Article | PubMed  | ChemPort |
26. Mandriota, S.J. et al. Transforming growth factor 1 down-regulates vascular endothelial growth factor receptor 2/fik-1 expression in vascular endothelial cells. J. Biol. Chem. 271, 11500−11505 (1996). | Article | PubMed  | ISI | ChemPort |
27. Yeh, C.H., Peng H.C. & Huang T.F. Cytokines modulate integrin v3-mediated human endothelial cell adhesion and calcium signaling. Exp. Cell Res. 251, 57−66 (1999). | Article | PubMed  | ISI | ChemPort |
28. Soker, S. et al. Characterization of novel vascular endothelial growth factor (VEGF) receptors on tumor cells that bind VEGF165 via its exon 7-endoded domain. J. Biol. Chem. 271, 5761−5767 (1996). | Article | PubMed  | ISI | ChemPort |
29. Piossek, C. et al. Vascular endothelial growth factor (VEGF) receptor II-derived peptides inhibit VEGF. J. Biol. Chem. 274, 5612−5619 (1999). | Article | PubMed  | ISI | ChemPort |
30. Cheifetz, S. et al. Endoglin is a component of the transforming growth factor-b receptor system in human endothelial cells. J. Biol. Chem. 267, 19027−19030 (1992). | PubMed  | ISI | ChemPort |
31. Jones, N. et al. Identification of Tek/Tie 2 binding partners. Binding to a multifunctional docking site mediates cell survival and migration. J. Biol. Chem. 274, 30896−30905 (1999). | Article | PubMed  | ISI | ChemPort |
32. Witzenbichler, B. et al. Chemotactic properties of angiopoietin-1 and −2, ligands for the endothelial-specific receptor tyrosine kinase Tie-2. J. Biol. Chem. 273, 18514−18521 (1998). | Article | PubMed  | ISI | ChemPort |
33. Eliceiri, B.P & Cheresh, D.A. The role v integrins during angiogenesis: insights into potential mechanisms of action and clinical development. J. Clin. Invest. 103, 1227−1230 (1999). | PubMed  | ISI | ChemPort |
34. Smith, J.W. et al. Integrin avb3-ligand interaction. Identification of a heterodimeric RGD binding Site on the vitronectin receptor. J. Biol. Chem. 265, 2168−2172 (1990). | PubMed  | ISI | ChemPort