| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
CRC Targeting and Imaging Group, Department of Oncology, Royal Free and University College Medical School, University College London, NW3 2PF, United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibody targeting of cytotoxic agents to tumor vasculature has the potential to destroy the vessels upon which hundreds of tumor cells rely for their continued growth. This approach has been shown to eradicate solid tumors in in vivo models (8 , 9) . Vascular endothelial cells are attractive targets because they are not subject to the limitations of penetration, antigen heterogeneity, and chemo- and radioresistance that are usually associated with delivery to solid tumors. The key to success is to develop an antibody to a target that is present on tumor vasculature endothelial cells that is absent, or present at very low levels, on the endothelium of normal tissues. The VEGF-A:receptor complex is a strong candidate target that was first described by Brekken et al. (10) , since VEGF R1 and R2, which bind VEGF-A are expressed predominantly on endothelial cells (11 , 12) and VEGF-occupied receptors are upregulated on tumor vasculature (6 , 13) .
To validate the VEGF:receptor as a target and to develop a suitable targeting system, we took the approach of making anti-VEGF antibodies as sFvs in the combinatorial phage-display system using VEGF-immunized mice. SFvs consist of the heavy and light chain variable regions of the antibody (which form the antigen-binding portion) tethered by a flexible linker (14) . As such, sFvs are the smallest fragment to retain full antigen binding capacity. Because they are expressed as a single recombinant protein, sFvs are ideal building blocks for fusion proteins with therapeutic as well as tumor-localizing properties (15, 16, 17, 18, 19, 20) . Phage display allows rapid isolation and bacterial production of desired sFvs from a wide repertoire because each phage displays an individual sFv on its surface and carries the gene for that sFv, allowing rapid selection of cloned antibodies to specific antigens (21 , 22) . Here we apply the recombinant system to the manufacture of sFvs that react with the VEGF:receptor complex, and with a selected sFv, we demonstrate that this target is valid for selective in vivo delivery of systemically administered antibodies to tumor vasculature.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Panning to Select VEGF-reactive Phage.
VEGF-reactive phage were captured from the library by panning using immunotubes (Nunc Maxisorb; Life Science Technologies, Ltd., Paisley, United Kingdom) coated with VEGF165 (10 µg/ml), followed by phage rescue as described by Marks et al. (26)
. Specificity was monitored by performing identical panning experiments with immunotubes coated in PBS alone. Four rounds of selection were carried out. The enrichment process was monitored by titration of phage colony-forming units eluted from the immunotubes at each stage of the process, as described for estimation of library size. Twenty individual colonies were picked at each round of selection and sFv-expressing phage produced using the KO7 helper phage as described previously (26)
. Cell-free supernatants containing the expressed phage were assessed for anti-VEGF activity by phage ELISA.
Phage ELISA.
ELISA wells were coated overnight with VEGF (10 µg/ml) or PBS before blocking with 3% DSM for 2 h at 37°C. Supernatants containing phage were preblocked with PBS/3% DSM for 1 h at room temperature before transfer of 50 µl to ELISA wells. After incubating at 37°C for 1 h, wells were washed three times each with PBS/0.02% Tween and PBS. Sheep anti-M13 antibody (100 µl; 5 Prime-3 Prime, Inc., Boulder, CO) diluted 1:1000 in PBS/3%. DSM was added to all wells for 1 h at 37°C. Plates were washed (three times) before addition of 100 µl of rabbit antigoat IgG conjugated to horseradish peroxidase (Sigma-Aldrich Chemical Co., Dorset, UK) diluted 1:500 for a further 1 h. Plates were washed (three times) in PBS/0.02% Tween, followed by saline solution (x1) before visualization by addition of 100 µl of 0.25 mg/ml o-phenylenediamine/H2O2 (Sigma-Aldrich Chemical Co., Dorset, UK) substrate in citrate phosphate buffer (pH 5.0). Color development was allowed to proceed for 20 min before quenching by addition of 4 N HCl (100 µl/well). Plates were read at 490 nm on an automated plate reader (Boots-Celltech Diagnostics, Slough, United Kingdom).
Screening Immunohistochemistry.
Thirty clones shown to be positive in phage ELISA against VEGF were induced to express soluble sFv as described previously (27)
. Bacterial supernatants were tested for their binding to vascular endothelium in cryostat sections of human term placenta and human colorectal carcinoma. Immunohistochemistry was performed using the 9E10 anti-myc tag antibody to detect sFv, as we have described previously (28)
.
Expression and Purification of LL4.
To facilitate purification by IMAC, LL4 gene was inserted via the NcoI-NotI restriction sites into a pUC119 vector incorporating a COOH-terminal His6 tag (27)
. His6-tagged LL4 protein was expressed from E. coli TG1 cells and subsequently purified from bacterial supernatant by IMAC as described previously (27)
. The samples were dialyzed against PBS, filtered, and concentrated before being stored at 
1 mg/ml at 4°C in PBS/0.02% sodium azide.
SDS PAGE and Western Blot.
Five µg of purified sFv were loaded into the wells of a 10% polyacrylamide gel under reducing conditions. Gels were stained using Coomassie Brilliant Blue and destained with methanol:acetic acid:H2O (3:1:6) until bands became visible. For Western blotting, samples were loaded and run as described for SDS-PAGE. After soaking the gels for 15 min in transfer buffer [12 mM Tris (pH 8.3), 96 mM glycine, and 20% methanol], proteins were transferred onto a polyvinylidene difluoride membrane (Millipore Waters, Watford, United Kingdom) for 1 h at 100 V. Membranes were blocked for 1 h with 4% DSM, followed by incubation with anti-His6 antibody (Dianova, Hamburg, Germany) diluted 1:500 in PBS/1% DSM. After washing three times in PBS/0.05% Tween 20 (PBS/T), antimouse IgG-horseradish peroxidase secondary conjugate diluted 1:500 in 1% DSM was added for 1 h. Membranes were washed (three times) in PBS/T, and bands were visualized by addition of 0.25 mg/ml diaminobenzidine/H2O2 substrate in PBS. The reaction was stopped by washing with tap water. Protein standards (range, Mr 6,000–250,000; Novex, Frankfurt, Germany) were loaded onto all gels for molecular weight estimation.
SELDI-AMS.
LL4 was immobilized on a preactivated SELDI-AMS chip surface, via primary amine and thiol groups, according to the manufacturer’s instructions (Ciphergen, Palo Alto, CA). Briefly, 5 µl of LL4 (175 µg/ml in PBS) were added for 1 h at room temperature, and after washing with PBS/0.5% Triton X-100, the surface was blocked by addition of 1 M ethanolamine for 15 min. After further washing, 5 µl of carrier free rhVEGF-A165 (R&D Systems, Oxford, United Kingdom) diluted to 100 µg/ml in PBS were then added for 1 h. The chip was washed again before the addition of rhVEGF R2/Fc chimera (R&D Systems) diluted to 100 µg/ml in PBS for a further 1 h. After a final wash in PBS/Triton X-100 followed by distilled water, a sinapinic acid matrix (Ciphergen, Palo Alto, CA) was added before the chip was placed in the chamber of a SELDI mass spectrometer (Ciphergen). Controls included immobilized LL4 incubated with rhVEGF165, or rhVEGF R2/Fc chimera, and immobilized rhVEGF R2/Fc chimera incubated with rhVEGF-A165.
Biotinylation.
sFvs were biotinylated at concentrations of 200 µg to 1 mg/ml using the ECL biotinylation kit (Amersham-Pharmacia, Buckinghamshire, United Kingdom) according to the manufacturer’s instructions. Briefly, antibody was diluted to the required concentration in carbonate buffer (pH 8.5) and incubated with the biotinylation reagent for 1 h at room temperature with constant agitation. sFvs were dialyzed against four changes of PBS overnight at 4°C before use.
IHC of Purified LL4.
Tissues were snap frozen in isopentane (cooled in liquid nitrogen), and 5-µm cryostat sections were cut and fixed in acetone. They were reacted with biotinylated LL4 (50 µg/ml) and visualized using avidin-biotin-peroxidase complexes (29)
. Biotinylated anti-CEA sFv (MFE-23) was included as control together with primary antibody omission controls and an anti-VEGF165 monoclonal antibody (MAB293; R&D Systems). Fifty tumors of various histological type were studied including: adenocarcinoma (7 large bowel, 5 ovary, 3 breast, and single examples of prostate, pancreas, stomach, lung, esophagus, and endometrium); carcinoma (examples of bladder, cervical, renal, and adrenal gland); 5 cases of lymphoma; and examples of sarcoma and teratoma. Forty-one samples of normal tissues from a broad range of histological types were also studied including gastrointestinal tract, pancreas, liver, kidney, lung, spleen, thyroid, adrenal, cervix, fallopian tube, testis, hypothalamus, and pituitary, as well as trophoblasts from both first trimester and term placenta. IHC-reacted sections were analyzed by bright-field microscopy. For all normal and neoplastic tissues, reactions of LL4 with endothelial cell surfaces and connective tissue stroma were assigned a score of negative (-), weak (+ to ++), or strong (+++). Reactivity with non-stromal cells of neoplastic and normal tissues was scored as a percentage LL4-positive cells, and individual cases were assigned to the appropriate group (negative, <10%, 11–24%, 24–49%, 50–75%, and 76–100%).
Whole Organ Biodistribution in LS174T Xenograft.
LL4 was labeled at 0.13 MBq µg-1 with Na125I by the Iodogen method (30)
. Incorporation of radiolabel was 98% as assessed by TLC. Anti-CEA sFv MFE-23 was labeled to similar specific activity (0.148 MBq µg-1) for comparison. Two groups of five nude mice bearing LS174T (31)
tumor xenografts were injected with radiolabeled LL4 or MFE-23 (2.11 MBq/mouse). Mice were sacrificed 1, 3, 6, and 24 h after injection, and tumor and normal tissues (blood, liver, lung, spleen, colon, muscle, and kidney) were removed for analysis by gamma counting. Briefly, samples were weighed, dissolved in 7 M KOH, and activity was assessed by gamma counting (Wizard; Pharmacia, Amersham, United Kingdom). Results were expressed as the percentage of injected activity/gram tissue, and tumor:normal tissue ratios were calculated.
Microautoradiography to Assess Blood Vessel Localization in LS174T Xenograft.
To assess the microdistribution of LL4, additional groups of four mice were injected with 1.85 MBq of either 125I-labeled LL4 or, for comparison, anti-CEA sFv 125I-labeled MFE-23 (0.15 MBq µg-1). Mice were sacrificed 1 and 3 h after injection, and LS174T tumor and normal tissues (liver, kidney, lung, spleen, and colon) were removed, fixed in 10% neutral formalin, and processed for histology as follows. Four-µm paraffin sections were cut and mounted on 3-aminopropyl triethoxysilane (2%) coated slides. Sections were dewaxed in Clearene (Surgipath, United Kingdom), dehydrated in graded alcohols, and transferred to warm (45°C) distilled water. Under darkroom conditions, slides were covered with a nuclear emulsion, K5 (Ilford, United Kingdom), and diluted (1:3) with 2% warm glycerol by dipping for 15 s. After air drying on the bench for 30 min, slides were transferred to a light-proof cabinet containing silica gel for 24 h and then placed in light-proof boxes for 8 weeks at 4°C. Slides were then developed in a darkroom using a sequence of Kodak D-19 developer (2.5 min), 1% acetic acid (4 min), and fixer (Ilford) diluted 1:10 w/v for 10 min. Sections were washed in running tap water for 5 min and then stained by H&E, dehydrated, cleared, and mounted in DPX (Merck Ltd., Dorset, United Kingdom). Autoradiographs were examined using bright-field microscopy. At least 50 randomly selected high power fields (x1000 oil immersion) were examined per tissue (four mice sampled/group). The number of grains present, overlying endothelial cells on the lumenal surface of vessels, was counted using a 1-cm2 graticule (Graticules Ltd., United Kingdom), i.e., number of grains per 10-µm length of endothelium. In addition, numbers of grains overlying tumor cells only were counted per 100 µm2 area (total, 120 areas) to compare density of LL4 localization with that of MFE-23. Results were expressed as median grains per 10-µm length of endothelium or median per 100-µm2 area for tumor cells. Differences between grain counts measured for the radiolabeled sFvs in tumor endothelium and in endothelium from normal tissues were analyzed statistically using a nonparametric test (Mann-Whitney U test). The Mann-Whitney U test was also used to compare results obtained with the LL4 (anti-VEGF) sFv and the MFE-23 (anti-CEA) sFv.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows examples of each of the four patterns of reactivity. From the 12 clones with the most selective binding to blood vessels, LL4 was chosen for further analysis on the basis of its relatively strong reactivity with blood vessel endothelium coupled with relatively weak binding to other elements.
|
, which shows SDS-PAGE analysis of the eluted fractions; the majority of purified LL4 was eluted by 200 mM imidizole (Lanes 4–6). Western blot using an anti-His6 antibody confirmed the presence of the His6 tag on LL4 in these fractions (Fig. 2B
, Lanes 4–6) and showed that LL4 was nondegraded and ran as a single band of the expected molecular weight of Mr 28,000. LL4-containing fractions were pooled, and after concentration, the final product was shown to be monomeric by fast protein liquid chromatography of 125I-labeled LL4 (data not shown). The final yield of LL4 after IMAC was 29.178 mg (equivalent to 17 mg/l bacterial supernatant).
|
, which demonstrates that when the VEGF receptor is presented to LL4 that has been loaded previously with VEGF, the receptor binds to form a LL4-VEGF:receptor complex, resulting in distinct peaks of molecular mass 40 kDa (rhVEGF-A165) and 155 kDa (rhVEGF R2) being desorped from covalently attached LL4 on the chip (Fig. 3A)
. Fig. 3B
shows that coupled LL4 is able to bind rhVEGF-A165 alone, as shown by a laser-desorped peak of molecular mass 40 kDa but does not bind to the soluble VEGF receptor (Fig. 3C)
when presented separately.
|
. In contrast, in the same tissue, anti-VEGF monoclonal antibody MAB293 appears to react with material within the lumen of tumor vessels (Fig. 4B)
. Some reactivity of LL4 with the tumor cells was also observed. More detailed analysis of LL4 reactivity with the panel of neoplastic and normal tissues showed that LL4 bound to vascular endothelium in 44 of 50 tumor sections. Most notable was strong (+++) reactivity with primary adenocarcinomas of the gastrointestinal tract, ovary, pancreas, lung, and breast. Bladder and renal carcinomas, lymphomas, and trophoblastic tumors were also positive (+ to ++) but more weakly. Squamous carcinomas of the lung and cervix were also weakly reactive (+). Negative (-) tissues included, three cases of hepatic metastasis of colorectal adenocarcinoma, and a single case of prostate carcinoma, one Conn’s tumor, and one alveolar soft part sarcoma of the lung. Fig. 5A
illustrates the total number of cases of endothelial cell reactivity for both neoplastic and normal tissues. Fifty % (25 of 50) of tumor cases showed strong binding of LL4 to vascular endothelium compared with only 16% (8 of 50) of normal tissues. The remainder of cases for both tumor and normal tissues were classified as being only weakly reactive or negative. Binding of LL4 to connective tissue stroma is shown in Fig. 5B
. All cases of normal tissues were scored as either negative 57.5% (23 of 40) or weak 42.5% (18 of 41). In contrast, strong reactions with stroma were only observed in tumor cases (8 of 40). Thirty-six cases (72%) of tumor showed weak reactivity, but only 12% (6 of 50) were negative. In general, reactivity of LL4 with endothelial cells and connective tissue stroma in the tumor was more widespread and more intense compared with that in normal tissues.
|
|
. Thirteen of 50 cases (25.5%) of tumor showed >50% positive malignant cells (predominantly adenocarcinomas), which was in marked contrast to the normal tissues tested in which only 1 of 41 (2.4%) cases showed >50% positive cells. Approximately half (20 of 41; 49%) of the normal cases were either negative or showed <10% LL4-positive cells.
|
, which shows the whole organ tumor:tissue ratios, and Table 1
, which shows the microdistribution of 125I-labeled LL4 to endothelial cells in those tissues. Although whole organ tumor:tissue ratios indicated selective tumor retention of 125I-labeled LL4 in comparison with spleen colon and muscle, there was no apparent tumor-selective retention of 125I-labeled LL4 in relation to blood and liver until 24 h after injection (Fig. 6)
.
|
, although still obvious at 3 h after injection.
|
. The difference in endothelial-associated grain counts for 125I-labeled LL4 and 125I-labeled MFE-23 was highly significant (P = <0.001; Table 2
), with a median of three grains/10-µm length of tumor endothelium for LL4 compared with only one for MFE-23. Grain counting from LL4 autoradiographs also demonstrated that LS174T tumor endothelium was significantly labeled by 125I-labeled LL4 compared with the endothelial cells lining normal vessels of liver, lung, and colon (P = <0.001; median ratio, 3:1; Table 2
). Kidney was the only normal tissue studied that showed high uptake of 125I-labeled LL4 in endothelium (Table 2)
, a result that was consistent with the whole organ counts, although there was little obvious accumulation of grains in glomerular endothelium. MFE-23 did not associate with endothelium in any of the normal tissues studied (median grain count, 0). Autoradiographs prepared from tissues 6 and 24 h after injection could not be interpreted reliably because grain densities were too low.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The in vitro experiments were supported by IHC studies that show strong reactivity of LL4 with blood vessel endothelium in human placenta and a wide range of human tumors. LL4 also reacted with connective tissue stroma in these tumors and with the tumor cells, but to a lesser extent. This was not unexpected because LL4 recognizes VEGF in both bound and complexed form. The tumor cells secrete VEGF, which then either binds to receptor on blood vessels or becomes trapped in "sumps" within the connective tissue stroma. The pattern of blood vessel selectivity observed with LL4 was marked in comparison with the pattern obtained with MAB293, a commercial monoclonal antibody to VEGF. In the same tissues, MAB293 reacted more strongly with stromal elements and showed little, if any, concentration on endothelial cell surfaces. One explanation for these findings is that MAB293 is a "neutralizing" antibody (R&D Systems product information) and interferes with VEGF binding to its cognate receptor. In contrast, we have shown by SELDI-AMS that LL4 binding to VEGF does not interfere with the VEGF:receptor interaction, and consistent with this, LL4 did not inhibit the proliferation of human umbilical vein endothelial cells, although MAB293 did (data not shown). Thus, although LL4 can target VEGF when it is concentrated on its receptor, other antibodies may be limited to reaction with uncomplexed forms of VEGF. Our findings in this area are consistent with those of Brekken et al. (10) , who were the first to report that monoclonal antibodies generated from animals immunized with NH2-terminal peptides of human VEGF165 [which is not part of the receptor binding region (32) ] react with VEGF:receptor complex and have specificity for VEGF associated with tumor blood vessel endothelial cells.
The data from our immunohistochemical analysis showed that LL4 binding was stronger and more widely distributed in neoplastic tissues than in normal tissues. Generally, in nonneoplastic tissues, the distribution of LL4 was as reported by other workers for anti-VEGF antibodies, with reactivity in tissues such as heart and lung (33) .
However, because IHC is only semiquantitative, a more rigorous measurement of LL4 binding to vascular endothelium was made using 125I-labeled LL4 in vivo. The LS174T colorectal xenograft model used secretes human VEGF that binds to mouse VEGF receptor to induce angiogenesis in the tumor. Our autoradiographic data showed that systemically administered 125I-labeled LL4 localized selectively to tumor vascular endothelium, although this was not obvious from whole organ counting. Interpretation of data obtained from gamma counting of whole organs and tumor is limited because differences in microdistribution of antibody cannot be readily ascertained. Therefore, we used microautoradiography to determine the sites of localization of 125I-labeled LL4 within tumor and normal tissues. For comparison, we used 125I-labeled MFE-23, an anti-CEA sFv that localizes in LS174T colorectal xenografts. VEGF and CEA are both secreted by the LS174T xenograft, but anchorage of CEA is on the tumor cells themselves because of GPI linkage (34) , although we hypothesized that retention of VEGF would be on endothelial cells because of the VEGF:receptor interaction.
The extent of binding of LL4 to vascular endothelium, in comparison with MFE-23, was quantified by grain counting to support this hypothesis. Our experiments demonstrated discrete localization of grains associated with LL4 either directly adjacent to or overlying vascular endothelial cells in tumor blood vessels. The anti-CEA sFv showed no significant association of grains within the vascular spaces, even at 1 h, but did show an increased density of grains associated with tumor cells directly underlying the vascular channels. This is consistent with the anti-CEA sFv binding to a tumor-secreted product that, unlike VEGF, is not captured by endothelial receptors. The observed accumulation of grains of LL4 within vessels may explain why no positive tumor:blood ratio was seen until 24 h after injection in the whole organ study. The relatively long circulatory half-life (t1/2) of LL4 (LL4: t1/2
= 1.54 h and t1/2ß = 5.25 h; MFE-23: t1/2 = 0.23 h and t1/2ß = 1.63 h) is unlikely to be explained by LL4 binding to circulating VEGF, because no VEGF (murine or human) was detectable in the mouse model by ELISA with a detection limit of 7.5 pg/ml VEGF. The treated mice were each given 16 µg of LL4 (around a million-fold molar excess of antibody); therefore, binding to circulating VEGF would be minimal. However, the longer half-life of LL4 may be related to its association with receptor-bound VEGF, resulting in retarded extravasation. The longer circulatory half-life observed with LL4 may be an advantage in terms of tumor targeting because tumor uptake of sFv molecules has been shown to be directly related to half-life in the blood (35)
.
Of the normal tissues assessed by microautoradiography, only kidney showed significant in vivo localization of 125I-labeled LL4 to blood vessels. This result is not readily explained by the generic mechanism of kidney retention reported with sFv molecules (36 , 37) because the finding was specific for LL4 in comparison with the anti-CEA sFv. A more likely explanation is that LL4 was localizing to VEGF in the mouse kidney because VEGF is normally expressed in the kidney of adult mice (38) . Because our immunohistochemical studies on normal tissues indicate that this not the case in humans, retention of LL4 by renal endothelium is unlikely to a problem for antibody-directed targeting strategies in humans. LL4 did, however, show a pattern of reactivity in normal mouse tissues comparable with that shown by polyclonal antihuman VEGF antibodies in human tissues described in the literature (33) . This cross-reactivity with murine VEGF demonstrates that the mouse model used in this study exhibits some of the complexity (i.e., possible binding to endogenous VEGF) that would be encountered in the human system regarding specific localization of LL4 to tumor endothelium.
Our work provides the first quantitative evidence that receptor-bound VEGF is a suitable target for selective in vivo delivery to tumor vasculature. The system we have described to validate this target is widely applicable to other antigens and is not labor intensive. The use of phage-display with immunized mice provided a diverse source of antibody specificities that were expressed in high yield and readily analyzed to identify sFvs with desired VEGF:receptor-binding characteristics. Bacterially expressed recombinant sFvs are highly practical building blocks for antibody-directed therapies and can be easily manipulated to modify their structure and/or to fuse with effector molecules such as cytokines or toxins; furthermore, they are economical to produce to clinical grade. In addition to its clear tumor targeting potential, the LL4 sFv described in this study has potential to image areas of angiogenesis in humans without interfering with the biological activity of VEGF.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by the Cancer Research Campaign. 
2 To whom requests for reprints should be addressed, at Royal Free and University College, Medical School, Royal Free Campus, Rowland Hill Street, London NW3 2PF, United Kingdom.
3 The abbreviations used are: VEGF, vascular endothelial growth factor; rhVEGF, recombinant human VEGF; sFv, single chain Fv; R1 and R2, receptors 1 and 2; DSM, dried skimmed milk; IMAC, immobilized metal affinity chromatography; IHC, immunohistochemistry; SELDI-AMS, surface enhanced laser desorption ionization-affinity mass spectrometry; CEA, carcinoembryonic antigen.
Received 9/28/00. Accepted 2/27/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Cancer Res., 56: 4758-4765, 1996.,6-diphenylglycoluril. Biochem. Biophys. Res. Commun., 80: 849-857, 1978.[Medline]
This article has been cited by other articles:
|
|
 |
|
  H.-F. Duan, X.-W. Hu, J.-L. Chen, L.-H. Gao, Y.-Y. Xi, Y. Lu, J.-F. Li, S.-R. Zhao, J.-J. Xu, H.-P. Chen, et al. Antitumor Activities of TEM8-Fc: An Engineered Antibody-like Molecule Targeting Tumor Endothelial Marker 8 J Natl Cancer Inst, October 17, 2007; 99(20): 1551 - 1555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Abdelrahim, C. H. Baker, J. L. Abbruzzese, D. Sheikh-Hamad, S. Liu, S. D. Cho, K. Yoon, and S. Safe Regulation of Vascular Endothelial Growth Factor Receptor-1 Expression by Specificity Proteins 1, 3, and 4 in Pancreatic Cancer Cells Cancer Res., April 1, 2007; 67(7): 3286 - 3294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. D. D'Andrea, G. Iaccarino, R. Fattorusso, D. Sorriento, C. Carannante, D. Capasso, B. Trimarco, and C. Pedone Targeting angiogenesis: Structural characterization and biological properties of a de novo engineered VEGF mimicking peptide PNAS, October 4, 2005; 102(40): 14215 - 14220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Smith, R. E. Kontermann, J. Embleton, and S. Kumar Antibody phage display technologies with special reference to angiogenesis FASEB J, March 1, 2005; 19(3): 331 - 341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. F. Dvorak Vascular Permeability Factor/Vascular Endothelial Growth Factor: A Critical Cytokine in Tumor Angiogenesis and a Potential Target for Diagnosis and Therapy J. Clin. Oncol., November 1, 2002; 20(21): 4368 - 4380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Molica, A. Vacca, D. Ribatti, A. Cuneo, F. Cavazzini, D. Levato, G. Vitelli, L. Tucci, A. M. Roccaro, and F. Dammacco Prognostic value of enhanced bone marrow angiogenesis in early B-cell chronic lymphocytic leukemia Blood, October 16, 2002; 100(9): 3344 - 3351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Goto, S. Yano, H. Zhang, Y. Matsumori, H. Ogawa, D. C. Blakey, and S. Sone Activity of a New Vascular Targeting Agent, ZD6126, in Pulmonary Metastases by Human Lung Adenocarcinoma in Nude Mice Cancer Res., July 1, 2002; 62(13): 3711 - 3715. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and normal tissues (
) (A) and connective tissue stroma in tumor (
), 6 h (
), and 24 h (
) after injection.