Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Bone Marrow Suppression by c-Kit Blockade Enhances Tumor Growth of Colorectal Metastases through the Action of Stromal Cell-Derived Factor-1

Bone Marrow Suppression by c-Kit Blockade Enhances Tumor Growth of Colorectal Metastases through... Hindawi Publishing Corporation Journal of Oncology Volume 2012, Article ID 196957, 12 pages doi:10.1155/2012/196957 Research Article Bone Marrow Suppression by c-Kit Blockade Enhances Tumor Growth of Colorectal Metastases through the Action of Stromal Cell-Derived Factor-1 1, 2 1 3 3 Kathrin Rupertus, Gudrun C. Y. Haberl, Claudia Scheuer, Michael D. Menger, 1 1 Martin K. Schilling, and Otto Kollmar Department of General, Visceral, Vascular and Pediatric Surgery, University of Saarland, 66421 Homburg, Germany Department of Hematology, Oncology and Immunology, University of Tubinge ¨ n, 72074 Tubinge ¨ n, Germany Institute for Clinical and Experimental Surgery, University of Saarland, 66421 Homburg, Germany Correspondence should be addressed to Otto Kollmar, otto.kollmar@uniklinikum-saarland.de Received 1 June 2011; Revised 2 August 2011; Accepted 7 August 2011 Academic Editor: Debabrata Mukhopadhyay Copyright © 2012 Kathrin Rupertus et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Mobilization of c-Kit hematopoietic cells (HCs) contributes to tumor vascularization. Whereas survival and proliferation of HCs are regulated by binding of the stem cell factor to its receptor c-Kit, migration of HCs is directed by stromal cell-derived factor (SDF)-1. Therefore, targeting migration of HCs provides a promising new strategy of anti-tumor therapy. Methods. BALB/c mice (n = 16) were pretreated with an anti-c-Kit antibody followed by implantation of CT26.WT-GFP colorectal cancer cells into dorsal skinfold chambers. Animals (n = 8) additionally received a neutralizing anti-SDF-1 antibody. Animals (n = 8) treated with a control antibody served as controls. Investigations were performed using intravital fluorescence microscopy, immunohistochemistry, flow cytometry and western blot analysis. Results. Blockade of c-Kit significantly enhanced tumor cell engraftment compared to controls due to stimulation of tumor cell proliferation and invasion without markedly affecting tumor vascularization. C-Kit blockade significantly increased VEGF and CXCR4 expression within the growing tumors. Neutralization of SDF-1 completely antagonized this anti-c-Kit-associated tumor growth by suppression of tumor neovascularization, inhibition of tumor cell proliferation and reduction of muscular infiltration. Conclusion. Our study indicates that bone marrow suppression via anti-c-Kit pretreatment enhances tumor cell engraftment of colorectal metastases due to interaction with the SDF-1/CXCR4 pathway which is involved in HC-mediated tumor angiogenesis. 1. Introduction Although EPCs incorporate into tumors in only small numbers, targeting EPCs has been shown to be effective Angiogenesis is one of the crucial steps in tumor progression in reducing tumor angiogenesis and tumor growth in and metastasis [1, 2]. Due to the lack of oxygen supply experimental models [6]. One of the most important factors and the accumulation of toxic products, avascular tumors mediating survival and proliferation of HCs is the stem and tumor metastases cannot grow beyond a critical size cell factor (SCF) which binds to the c-Kit receptor on the of ∼1-2 mm and thus will stay clinically occult [1, 3, 4]. surface of HCs. Okamoto et al. have shown that bone marrow Therefore, only a small proportion of circulating cancer cells suppression by anti-c-Kit treatment induces a delay in tumor finally forms macroscopic tumors [5]. Tumor vessels can angiogenesis due to the inhibition of angiogenic sprouting grow by sprouting of preexisting host vessels, intussusception in colon tumors and that c-Kit blockade suppresses tumor or incorporation of bone marrow-derived endothelial pro- growth of subcutaneously implanted prostate carcinomas genitor cells (EPCs) which are a subset of hematopoietic cells (PC3) by inhibition of tumor angiogenesis regulated by HCs (HCs). This mechanism is called vasculogenesis and mimics embryonic angio-development [2]. [7]. 2 Journal of Oncology Recruitment of HCs and EPCs to avascular areas is The animals were housed in single cages at room temperature orchestrated by different angiogenic growth factors and of 22–24 C and at a relative humidity of 60–65% with a 12- cytokines [8], predominantly by the CXC-chemokine stro- hour light/dark cycle environment. The mice were allowed mal cell-derived factor (SDF)-1 and its receptor CXCR4 free access to drinking water and standard laboratory chow [9, 10]. HCs and EPCs fluctuate to peripheral organs in (Altromin; Lage, Germany). antiphase with the expression of SDF-1 within the bone marrow microenvironment [11] and migrate alongside a 2.3. Experimental Model—Dorsal Skinfold Chamber. To chemotactic gradient towards higher concentrations of SDF- allow repetitive analyses of the microcirculation of the grow- 1, for example, sites of injury and tumor tissue [12– ing tumors, the dorsal skinfold chamber model was used 14]. Furthermore, it has been demonstrated that SDF-1- for intravital microscopy as described previously in detail mediated recruitment of c-Kit-positive cells to the periphery [19]. For operative procedures, animals were anesthetized by is dependent on cofactors, including tumor necrosis factor-α intraperitoneal injection of 90 mg/kg BW ketamine (Ketavet, and the endothelial nitric oxide synthase (eNOS) [12]. Parke Davis; Freiburg, Germany) and 20 mg/kg BW xylazine However, the impact of circulating HCs and EPCs on (Rompun, Bayer; Leverkusen, Germany). The chamber, con- vasculogenesis and sprouting angiogenesis in tumor angio- sisting of two symmetrical titanium frames, was positioned development is still controversially discussed in the literature to sandwich the extended double layer of the dorsal skin. [15–17], and the role of the SDF-1/CXCR4 pathway is One layer of skin and subcutis was completely removed in not yet fully understood. Therefore, in the presented study a circular area of 15 mm diameter. The remaining layers, consisting of the epidermis, subcutaneous tissue, and striated we analyzed the influence of bone marrow suppression by skin muscle, were covered with a glass coverslip incorporated anti-c-Kit treatment combined with SDF-1 neutralization on tumor cell engraftment and neovascularization using a into one of the titanium frames. The animals tolerated the chambers well and showed no signs of discomfort or changes murine model of colorectal tumor metastasis. in sleeping and feeding habits. After a 48-hour recovery period, the animals were reanesthetized. For tumor cell 2. Materials and Methods implantation, the coverslip of the chamber was temporarily removed and 1 × 10 CT26.WT-GFP cells were implanted 2.1. Tumor Cell Line and Culture Conditions. The CT26 onto the surface of the striated muscle tissue within the cell line is an N-nitroso-N-methyl-urethane-induced undif- chamber. Immediately after cell implantation, the chamber ferentiated adenocarcinoma of the colon, syngeneic with tissue was covered again with the coverslip [20–22]. the BALB/c mouse. For our studies [18], the CT26.WT cells (ATCC CRL-2638, LGC Promochem GmbH, Wesel, Germany) were transfected with the enhanced GFP expres- 2.4. Experimental Protocol. Animals were assigned to three sion vector pEGFP-N1 (Clontech) with the use of CLON- different groups: the first group (cKit-Ab; n = 8) received fectin (Clontech, Palo Alto, CA, USA) according to the a pretreatment with a monoclonal anti-c-Kit receptor anti- manufacturer’s instructions. For the individual experiments, body (ACK45, BD Biosciences, Heidelberg, Germany) by CT26.WT-GFP cellsweregrown in cell cultureasmono- daily intraperitoneal injections starting 4 days before tumor layers in RPMI-1640 medium with 2 mM L-glutamine cell implantation. The ACK45-antibody was given four times (Sigma Aldrich Chemie GmbH, Taufkirchen, Germany) at a dose of 1 mg/kg BW daily as described by Okamoto supplemented with 10% fetal calf serum (FCS Gold, PAA et al. [7]. Animals of the second group (cKit/SDF1-Ab; Laboratories GmbH, Colbe, ¨ Germany), 100 U/mL penicillin, n = 8) were also pretreated with ACK45 as described and 100 μg/mL streptomycin (PAA Laboratories GmbH). above. Additionally, these animals received 1 mg/kg BW of a The cells were incubated at 37 C in a humidified atmosphere monoclonal mouse anti-mouse SDF-1 antibody (MAB310, R containing 5% CO , and only cells of the first three and D Systems, Wiesbaden, Germany). MAB310 application serial passages after cryostorage were used. At the day of was performed intraperitoneally, starting at the day of implantation, tumor cells were harvested from subconfluent tumor cell implantation (d0) and repeated every second day cultures (70 to 85%) by trypsinization (0.05% Trypsin and thereafter until day 12. The third group (control; n = 8) 0.02% EDTA, PAA Laboratories GmbH) and washed twice served as control and received pre- and posttreatment the in phosphate-buffered saline solution (PBS). same amount of the corresponding isotype-matched IgG control antibodies (A95-1, BD Biosciences, and MAB002 R&D Systems). All animals underwent repetitive intravital 2.2. Animals. Experiments were performed after approval by microscopic analyses directly (d0) as well as 5, 8, 11, and the local governmental ethic committee and conformed to 14 days after tumor cell implantation. At the end of the the United Kingdom Coordinating Committee on Cancer experiment (day 14), the chamber with the tumor tissue was Research (UKCCCR) Guidelines for the Welfare of Animals harvested for histology and immunohistochemistry. in Experimental Neoplasia (as described in 1998 in Br J Can- cer 77: 1–10) and the Guide for Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National 2.5. Intravital Fluorescence Microscopy. Intravital fluores- Research Council; NIH Guide, Vol. 25, No. 28, 1996). Female cence microscopy was performed in epi-illumination tech- BALB/c mice (Charles River Laboratories GmbH; Sulzfeld, nique using a modified Zeiss Axio-Tech microscope (Zeiss, Germany) with a body weight (BW) of 18–20 g were used. Oberkochen, Germany) with a 100-W HBO mercury lamp. Journal of Oncology 3 Microscopic images were monitored by a charge-coupled techniques. Therefore, deparaffinized sections were incu- device video camera (FK 6990, COHU, Prospective Mea- bated with 3% H O and 2% goat normal serum to 2 2 block endogenous peroxidases and unspecific binding sites. surements Inc., San Diego, CA, USA) and were transferred A monoclonal mouse anti-pan PCNA antibody (PC10, to a video system (VO-5800 PS, Sony, Munchen, ¨ Germany) DakoCytomation, Hamburg, Germany) and a polyclonal for subsequent off-line analysis. Tumor size, growth kinet- rabbit anti-mouse cleaved caspase-3 antibody (Asp175, Cell ics, migration of tumor cell, and neovascularization were Signaling Technology, Frankfurt, Germany) were used as analyzed using blue light epi-illumination (450 to 490 nm primary antibodies. Goat anti-mouse and goat anti-rabbit excitation wavelength and >520 nm emission wavelengths) POD-conjugated antibodies were used as secondary antibod- [20–22]. ies for streptavidin-biotin-complex peroxidase staining. 3,3 diaminobenzidine (DakoCytomation) served as chromogen. 2.6. Microcirculation Analysis. Microcirculatory parameters Sections were counterstained with Hemalaun according to were assessed off line by frame-to-frame analysis of the Mayer and examined by light microscopy. videotaped images using a computer-assisted image analysis To assess the expression of CD31 as a marker for endothe- system (CapImage, Zeintl Software, Heidelberg, Germany). lial cells, zinc fixative fixed paraffin sections of tumor tissue The fluorescent labeling of the tumor cells allowed precise were used. After incubation with 3% H O and 2% goat nor- 2 2 delineation of the tumor from the surrounding host tissue. mal serum to block endogenous peroxidases and unspecific It also enabled for distinct identification of individual tumor binding sites, deparaffinized sections were incubated with a cells to study tumor cell migration. At each observation time rat anti-mouse CD31 antibody (MEC 13.3, BD Biosciences). point, the surface of the fluorescently labeled tumor mass A polyclonal goat anti-rat IgG antibody (BD Biosciences) within the chamber was scanned for determination of the 2 was used as secondary antibody. Colorimetric detection was tumor size (given as tumor area in mm ). Eight regions of performed using 3,3 diaminobenzidine (DakoCytomation) interest (ROIs) were randomly chosen next to the tumor substrates. Sections were counterstained with Hemalaun margin. The number of migrating cells was counted, and the according to Mayer and examined by light microscopy. distance to the tumor margin was measured (given in μm). Microcirculation was quantified as described before in detail [22]. First, eight representative ROIs within the tumor 2.8. Flow Cytometric Analysis of CT26.WT-GFP Cells. FAC- margin were chosen and analyzed for microcirculation Scan (Becton Dickinson, Mountain View, CA, USA) analysis parameters. In these ROIs, the onset of angiogenesis, that is, was performed to assess the expression of c-Kit on the the existence of angiogenic buds, sprouts, and newly formed CT26.WT-GFP cells in triplicate. Cells were fixed with 2% blood vessels, was documented and scored 0 to 8, with 0 formalin, washed twice with PBS, resuspended in FACS indicating existence of newly formed tumor microvessels in buffer and incubated with an FITC-conjugated rat anti- none of the ROIs and 8 indicating their existence in all of mouse IgG2b c-Kit antibody (1 : 50; 553354, BD Biosciences) the ROIs. Functional capillary density (given in cm/cm ) or an FITC-conjugated isotype-matched control antibody of the tumor microvessels was measured to quantify the (553988, BD Biosciences). Cells were washed again and then angiogenic activity. This parameter was defined as the length maintained in 2% paraformaldehyde in PBS. Tumor cells of red blood cell perfused microvessels per observation were selectively analyzed for their fluorescence properties area and was analyzed within the eight ROIs of the tumor using the CellQuest data handling program (BD Biosciences) margin and within four additional ROIs of the tumor center. with assessment of 5000 events per sample. The flow cytome- Diameters of the newly formed tumor microvessels were ter was calibrated with fluorescent standard microbeads measured perpendicularly to the vessel path (given in μm). (CaliBRITE Beads, BD Biosciences). To study vascular permeability of the newly formed tumor microvessels, petechial bleedings were documented in each 2.9. Western Blot Analysis. To study protein expression of the ROIs, and given as percentage of all the ROIs analyzed patterns, additional Western blot analyses were performed on [20, 21, 23]. tumor specimen gained from the dorsal skinfold chamber. Therefore, 12 additional animals received CT26.WT-GFP 2.7. Histology and Immunohistochemistry. At the end of the tumor cell implantation in the dorsal skinfold chamber experiment (day 14), the tumor and the adjacent host and were assigned to the three groups as described above. tissue were harvested and immediately fixed in formalin. For Eight animals were pretreated with the anti-c-Kit antibody light microscopy, formalin-fixed biopsies were embedded in (ACK45, BD Biosciences). Four of these pretreated animals paraffin. Sections of 5 μm were cut and stained with hema- received additional treatment with the neutralizing anti- toxylin and eosin (HE) according to standard procedures to SDF-1 antibody (MAB310, R and D Systems) starting at the analyze tumor growth characteristics. Tumor cell invasion day of tumor cell implantation. Four animals received the of the muscular layer on the surface of the dorsal skinfold same amount of the corresponding isotype-matched control chamber was quantified over the entire tumor basis and given antibody (A95-1, BD Biosciences). as percentage of the length of the tumor basis. For whole protein extracts and Western blot analysis of To study tumor cell proliferation and apoptotic cell the expression of vascular endothelial growth factor (VEGF), death, proliferating cell nuclear antigen (PCNA) and cleaved the chemokine receptor CXCR4 and eNOS, CT26.WT- caspase-3 were stained using indirect immunoperoxidase GFP-tumors were completely removed from the skinfold 4 Journal of Oncology at day 5. The tumor tissue samples were homogenized of the ROIs by day 5 after tumor cell implantation. At separately in lysis buffer and a protease inhibitor cock- day 8, angiogenesis was seen in almost 100% of the ROIs. tail (Sigma, Taufkirchen, Germany), incubated on ice and Pretreatment with the anti-c-Kit antibody did not influence centrifuged at 16.000×g. The supernatant was saved as the onset of the angiogenic switch compared to controls. whole protein extract fraction. Protein concentrations were Of interest, blockade of SDF-1 after pretreatment with anti- determined using the Lowry assay with bovine serum c-Kit significantly delayed the angiogenic switch, with only albumin as standard. Ten microgram protein per lane ∼50% of the ROIs showing newly formed microvessels by were separated discontinuously on sodium dodecyl sulfate day 8 and ∼70% by day 11 (P< 0.05). In this group, polyacrylamide gels (10% SDS-PAGE) and transferred to a angiogenesis was observed in all ROIs only at the end of the polyvinyldifluoride membrane (0.2 μm, BioRad, Munc ¨ hen, observation period (Figure 2). Germany). After blockade of nonspecific binding sites, The differential effects of c-Kit blockade with or without membranes were incubated with an anti-VEGF antibody additional SDF-1 blockade were also reflected by quantitative (A-20, Santa Cruz, Heidelberg, Germany), an anti-CXCR4 analysis of the microvascular densities of the newly formed antibody (ab2074, Abcam, Heidelberg, Germany), or an anti- tumor vessels. As characteristic for tumor vessels, the vas- eNOS antibody (BD Transduction Lab., Heidelberg, Ger- cular network of the tumors consisted of irregularly shaped many) followed by the corresponding secondary peroxidase- and chaotically arranged microvessels (Figures 3(a)–3(c)). conjugated antibodies (GE Healthcare, Freiburg, Germany Whereas no differences of microvascular density within and Santa Cruz). Protein expression was visualized by the tumor vasculature could be observed between controls means of luminol enhanced chemiluminescence (ECL, GE and anti-c-Kit pretreated animals, additional treatment Healthcare) and exposure of the membranes to a blue- with the anti-SDF-1 antibody significantly decreased the light-sensitive autoradiography film (Hyperfilm ECL, GE microvascular density within the tumor center compared Healthcare). Signals were densitometrically assessed (Bio- to the other groups from day 8 until the end of the Rad, Gel-Dokumentationssystem) and normalized to β-actin observation period (Figure 3(d)). Quantitative analysis of the signals (mouse monoclonal anti-β-actin, Sigma) to correct functional microvascular density within the tumor margin for unequal loading. showed similar results (data not shown). The functional microvascular density within the tumor margin and the tumor center was not significantly different within the 2.10. Statistical Analysis. Allvaluesare expressedasmeans individual treatment groups. ± SEM. After proving the assumption of normality and In contrast to the highly vascularized CT26.WT- homogeneity of variance across groups, differences between GFP tumors within the intravital fluorescence microscopy, groups were calculated by a one-way analysis of variance immunohistological staining for CD31 as a marker for (ANOVA) followed by the appropriate post hoc comparison, endothelial cells displayed positive staining of only a few cells including correction of the alpha error according to Bonfer- within tumor microvessels without significant differences roni probabilities to compensate for multiple comparisons. between the three groups (data not shown). Overall statistical significance was set at P< 0.05. Statistical As neovascularization is usually associated with vasodila- analysis was performed with the use of the software package tion due to the action of VEGF, microvessel diameters within SigmaStat (SPSS Inc, Chicago, Ill). the tumor were analyzed. However, no overall differences between the diameters of the microvessels within the tumor 3. Results margin and the tumor center could be observed (Table 1). During the 14 days observation period, diameters slightly 3.1. Tumor Growth. All animals had an uneventful post- increased in all groups until day 11, particularly in the operative recovery and tolerated well the dorsal skinfold control and anti-c-Kit-pretreated groups. Of interest, at day chamber implantation and the repetitive intravital micro- 14, microvessel diameters within tumors of mice treated with scopic analyses. The general conditions of the mice were not anti-c-Kit and anti-SDF-1 were markedly smaller compared affected and no changes in feeding or sleeping habits were to the other two groups (Table 1). observed. The take rate of the CT26.WT-GFP cells within Petechial bleedings within the areas of angiogenesis the dorsal skinfold chamber was 100%, and progressive of growing tumors are a characteristic indicator of the tumor growth was observed in all groups throughout the VEGF action on vascular permeability. To study this VEGF- entire observation period (Figures 1(a)–1(c)). Quantitative induced increase of vascular permeability, we analyzed the analysis of the increase of the tumor area revealed a petechial bleedings within the ROIs by intravital fluorescence significantly enhanced engraftment of tumor cells and a microscopy. Microbleedings were observed in all tumors. At significant stimulation of tumor growth after pretreatment day 5, signs of petechial bleedings were observed in ∼6% with the anti-c-Kit antibody compared to controls (P< of the tumor area of control tumors, whereas petechial 0.05). Interestingly, additional blockage of SDF-1 completely bleedings occurred in up to 30% of the tumor area in blunted this enhancement of tumor growth leading to similar animals pretreated with anti-c-Kit at that time. In all groups, tumor sizes as measured in controls (P< 0.05; Figure 1(d)). petechial bleedings were most pronounced 8 days after tumor cell implantation (control: 15–20%, anti-c-Kit: ∼30%, anti- c-Kit/anti-SDF-1: 35–40%). Comparing the three groups, 3.2. Angiogenesis and Neovascularization. In control animals, newly developed microvessels could be detected in ∼50% no statistical significance could be found due to the high Journal of Oncology 5 (a) (b) 058 11 14 (Days) (c) (d) Figure 1: Time course of tumor growth of CT26.WT-GFP tumors in the dorsal skinfold chamber. Stereomicroscopy photographs of representative 14 days old tumors after either treatment with an isotype-matched control antibody (a), pretreatment with anti-c-Kit (b) alone, or pretreatment with anti-c-Kit followed by anti-SDF-1 treatment (c). Quantitative analysis of the tumor area (d) displayed progressive tumor growth in all groups. After anti-c-Kit pretreatment (grey triangles), tumor growth was significantly accelerated compared to controls (white circles). Of interest, additional neutralization of SDF-1 (black squares) completely blunted this anti-c-Kit-associated enhancement of ∗ # tumor growth. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a)–(c) ×4. Table 1: Microvessel diameters within the tumor margin and the tumor center of control animals, animals pretreated with anti-c-Kit and animals pretreated with anti-c-Kit followed by anti-SDF-1 treatment. All values are given in μm. Diameters slightly increased during the observation period. No significant differences could be observed between the tumor margin and center. Fourteen days after tumor cell implantation, microvascular diameters within the tumor center were significantly smaller in anti-c-Kit/anti-SDF-1 treated animals compared to controls and to anti-c-Kit treated animals. Time Control Anti-c-Kit Anti-c-Kit/anti-SDF-1 d5 12.42 ± 0.53 11.88 ± 0.77 13.02 ± 0.52 d8 13.97 ± 0.48 14.60 ± 0.93 15.02 ± 0.57 Tumor margin d11 14.73 ± 1.36 15.39 ± 1.00 14.84 ± 1.06 d14 15.81 ± 1.20 15.24 ± 0.91 13.94 ± 0.75 d5 13.91 ± 0.67 12.85 ± 0.83 13.22 ± 1.00 d8 14.44 ± 0.52 13.93 ± 0.75 14.21 ± 0.47 Tumor centre d11 15.35 ± 1.40 15.26 ± 0.98 15.66 ± 1.83 ∗# d14 16.86 ± 1.10 16.84 ± 1.35 12.82 ± 0.67 ∗ # Data are given as mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Increase of tumor area (mm ) 6 Journal of Oncology infiltration compared to anti-c-Kit pretreated tumors (P< 0.05), resulting in invasive growth characteristics which were 8 comparable to controls (Figure 4(c)). PCNA as an indicator of cell proliferation displayed positive staining in 46.9 ± 3.1% of the tumor cells in control 6 # animals (Figure 5). After pretreatment with anti-c-Kit, the number of PCNA-positive tumor cells was significantly 4 higher compared to controls (P< 0.05, Figures 5(a) and 5(c)). In contrast, additional blockade of SDF-1 resulted in a significantly lower number of PCNA-positive tumor cells compared to controls and to anti-c-Kit pretreated animals (P< 0.05, Figures 5(b) and 5(c)). 0 To study apoptotic cell death, immunohistochemical 0 58 11 14 staining of cleaved caspase-3 products within the tumors was (Days) performed. Of interest, 14 days after tumor cell implantation, only a minor fraction of the total number of 351.3 ± 5.2 Figure 2: Semiquantitative analysis of the onset of angiogenesis within the CT 26.WT-GFP tumors. Animals were pretreated tumor cells within the high power fields (HPF) showed with anti-c-Kit (grey triangles) alone or anti-c-Kit and anti- positive staining for caspase-3. Whereas 0.73 ± 0.13 tumor SDF-1 neutralizing antibodies (black squares). Animals treated cells/HPF were positive for caspase-3 in controls, pretreat- with isotype-matched control antibodies served as controls (white ment with anti-c-Kit reduced apoptotic cell death within the circles). In controls and anti-c-Kit pretreated animals, ∼50% of the tumors (0.26 ± 0.06 tumor cells/HPF). In tumors of animals ROIs showed newly developed microvessels 5 days after tumor cell additionally treated with anti-SDF-1 antibodies, this effect implantation. By day 8, almost 100% of the ROIs were vascularized. was even more pronounced (0.15 ± 0.03 tumor cells/HPF, Treatment with anti-c-Kit and anti-SDF-1 resulted in a significant P< 0.05). delay of angiogenesis with only ∼50 and ∼70% of the ROIs showing newly developed microvessels by days 8 and 11. Within these tumors, angiogenesis could be observed in all ROIs only at day 14. 3.5. C-Kit Expression and Western Blot Analysis. FACScan ∗ # Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. analysis demonstrated that 8.2 ± 0.8% of the CT26-GFP cells transfected with the enhanced GFP expression vector pEGFP-N1 was c-Kit receptor positive. At day 5 after tumor cell implantation, the tumors standard deviation within the individual groups (data not of animals pretreated with anti-c-Kit antibodies showed shown). a significantly higher expression of VEGF and CXCR4 compared to controls (P< 0.05; Figures 6(a) and 6(b)). 3.3. Tumor Cell Migration. As SDF-1 has been demonstrated Additional SDF-1 blockade did not further increase VEGF to exert chemotactic effects on CT26.WT-GFP tumor cells, and CXCR4 expression. Furthermore, eNOS expression we additionally focused on migrating tumor cells next to within the tumors was decreased by anti-c-Kit pretreatment the tumor margin. The migrating tumor cells were easily with and without additional anti-SDF-1 treatment compared detectable due to their GFP labeling (Table 2). Starting at to controls (data not shown). day 5 after tumor cell implantation, tumor cell migration was detectable in all tumors until the end of the observation 4. Discussion period with slightly increasing distances from the tumor margin. The number of migrating tumor cells ranged The major finding of the present study is that bone between 20 and 25 per representative field during the marrow suppression by anti-c-Kit treatment significantly whole observation period. Treatment with anti-c-Kit alone enhances tumor cell engraftment of colorectal tumors due or additional blockade of SDF-1 did not impair tumor cell to an increase of tumor cell proliferation and invasion. migration compared to controls (Table 2). Additional anti-SDF-1 treatment neutralizes this increased tumor outgrowth by inhibition of tumor cell proliferation 3.4. Tumor Cell Morphology, Proliferation, and Apoptotic and tumor neovascularization. These findings suggest that Cell Death. Histological examinations on hematoxylin-eosin the enhanced tumor growth under the conditions of bone stained sections revealed solid tumor growth within the marrow suppression induced by anti-c-Kit treatment is dorsal skinfold chamber. Signs of malignant tumor growth related to the SDF-1/CXCR4 pathway. such as invasion of the adjacent host tissue by the tumor Bone-marrow-derived hematopoetic cells (HCs) con- cells were detectable in each group (Figure 4). Quantitative tribute to physiological and pathological vessel forma- analysis of tumor cell invasion showed a significant increase tion. During tumor growth, the release of cytokines and of tumor cell infiltration through the muscular layer of chemokines mediates the recruitment of EPCs and HCs the dorsal skinfold chamber in tumors of animals which contributing to the early initiation and stabilization of were pretreated with anti-c-Kit compared to controls (P< newly formed blood vessels, so-called vasculogenesis [24– 0.05). Of interest, tumors of animals additionally treated with anti-SDF-1 showed a significant reduction of muscular 26]. However, their precise role has not been fully elucidated Angiogenesis (score 0–8) Journal of Oncology 7 (a) (b) # 200 0 5 8 11 14 (Days) (c) (d) Figure 3: Time course of the functional capillary density within CT-26.WT-GFP tumors in dorsal skinfold chambers analyzed by intravital fluorescence microscopy. Representative fluorescence microscopic images show the network of chaotically arranged microvessels within the tumor center of control animals (a), animals pretreated with anti-c-Kit (b), and animals additionally treated with anti-SDF-1 (c) at day 11 after tumor cell implantation. Quantitative analysis of the functional capillary density (d) revealed a significant inhibition of tumor neovascularization within the tumor center after combined anti-c-Kit and anti-SDF-1 treatment (black squares) compared to controls (white circles). Anti-c-Kit pretreatment alone (grey triangles) did not significantly influence the extent of neovascularization. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a)–(c) ×40. Table 2: Tumor cell migration in dorsal skinfold chambers of control animals, animals pretreated with anti-c-Kit, and animals pretreated with anti-c-Kit followed by anti-SDF-1 treatment during a 14-day observation period. Migrated tumor cells next to the tumor margin were observed from day 5 until the end of the observation time. Comparing all three groups, no significant differences could be observed concerning the total number of migrated cells as well as their distance [μm] from the tumor margin. Of interest, at the later time points, tumor cells were found at greater distances from the tumor margin. Time Control Anti-c-Kit Anti-c-Kit/anti-SDF-1 d5 22.08 ± 1.33 23.00 ± 1.33 19.91 ± 1.14 d8 22.67 ± 1.33 23.14 ± 1.18 20.67 ± 1.07 Number d11 22.42 ± 1.53 23.48 ± 1.05 20.80 ± 1.34 d14 20.84 ± 1.71 23.96 ± 1.32 21.17 ± 1.30 d5 276.24 ± 16.08 291.82 ± 15.95 309.45 ± 50.93 d8 337.94 ± 9.38 351.11 ± 23.57 323.38 ± 20.45 Distance d11 383.85 ± 15.80 423.32 ± 25.65 360.83 ± 18.58 d14 421.50 ± 22.40 438.49 ± 32.01 458.51 ± 59.01 Data are given as mean ± SEM. Capillary density tumor center (cm/cm ) 8 Journal of Oncology (a) (b) Control cKit-Ab cKit/SDF1-Ab (c) Figure 4: Hematoxylin-eosin staining of CT26.WT-GFP tumors shows solid tumor growth 14 days after tumor cell implantation within the dorsal skinfold chamber. Sections display tumor infiltration through the underlying muscular layer (marked by arrows) after pretreatment with anti-c-Kit (a) and lack of muscular infiltration after pretreatment with anti-c-Kit and additional neutralization of SDF-1 (b). Quantitative analysis of tumor cell invasion of the muscular layer is given as percentage of the length of the tumor. Tumors of animals pretreated with anti-c-Kit showed a significantly pronounced infiltration of the muscular layer compared to controls. Of interest, blockade of SDF-1 after anti-c-Kit pretreatment abrogated this anti-c-Kit-associated increase of tumor cell infiltration. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a, b) ×88. yet. HCs and EPCs express stem cell markers such as c-Kit. growth, they do not contribute to tumor angiogenesis [29]. Whereas many studies indicate that bone-marrow-derived In a model of subcutaneously implanted prostate carcinoma, EPCs incorporate into tumor neovessels [25, 27], HCs may Okamoto et al. observed an accumulation of HCs around promote vasculogenesis via paracrine release of angiogenic newly formed blood vessels, but did not find these cells to be factors enhancing the recruitment and incorporation of part of the tumor microvessels themselves. Because of these EPCs into neovessels [28]. findings, the authors postulated a stabilizing and supportive In the early phases of tumor growth, 50–90% of the role of HCs for the developing vascular network [7]. neovessels within the tumor mass are derived from the bone In the present study, intravital fluorescence microscopy marrow dependent on the tumor type [25]. However, recent showed highly vascularized CT26.WT-GFP tumors 14 days reports have already doubted the impact of vasculogenesis after tumor cell implantation. However, immunohistological from bone-marrow-derived cells for tumor neovasculariza- staining using the endothelial cell marker CD31 displayed tion and claim an exclusive role for sprouting angiogenesis positive staining of only a few endothelial cells within these in tumor blood vessel development [15, 16]. In a model of tumors. This finding reflects the diverging structure between syngeneic bone marrow transplantation, Patil et al. demon- normal microvessels and tumor neovessels which consist of strated that GPF-expressing c-Kit bone-marrow-derived only cancer cells or a mosaic of cancer and endothelial cells progenitor cells are recruited to subcutaneously implanted [2]. Lewis lung carcinoma but do not directly contribute to Previous experimental studies in mice have shown that a microvascular structure. They, therefore, concluded that depletion of the myeloid and erythroid cell lineages including even if circulating HCs and EPCs home to sites of tumor EPCs and HCs from the bone marrow could be performed Muscular infiltration (tumor %) Journal of Oncology 9 (a) (b) Control cKit-Ab cKit/SDF1-Ab (c) Figure 5: PCNA immunohistochemistry in CT26.WT-GFP tumors at day 14 after tumor cell implantation. Animals were pretreated with anti-c-Kit (cKit-Ab, (a)) alone or anti-c-Kit and anti-SDF-1 neutralizing antibodies (cKit/SDF1-Ab, (b)). Animals treated with an isotype-matched control antibody served as controls (control). Quantitative analysis demonstrated that anti-c-Kit pretreatment significantly increased the rate of proliferating tumor cells compared to controls (c). Of interest, additional neutralization of SDF-1 significantly decreased the amount of proliferating tumor cells compared to controls and anti-c-Kit-treated animals. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a, b) ×175. by daily injection of 1 mg/kg ACK2, an antagonistic anti- of tumor vessels were slightly reduced in these studies of c-Kit antibody, which blocks the function of c-Kit without Okamoto et al. during the first 5–7 days, tumor growth cytotoxic side effects on c-Kit cells [30, 31]. Of interest, was rapidly reinitiated as the number of circulating HCs whereas no polymorphonuclear cells or erythroblasts were in the peripheral blood increased again [7]. Therefore, present in the bone marrow of these ACK2-treated animals, the authors concluded that HCs migrating into the tumor B lineage cells continued to grow to fill the space from which mass promote the initiation of tumor neovascularization. myeloid and erythroid progenitor cells were purged [30, 31]. In our present study, although neoangiogenesis was not Moreover, the reduction rate of the colony forming cells influenced after administration of anti-c-Kit, tumor growth within the bone marrow was dependent on the amount of was stimulated as a result of an increased tumor cell prolif- anti-c-Kit antibodies [30]. In the present study, we, therefore, eration and invasion. This observation might be the result used the anti-c-Kit antibody in a dosage of 1 mg/kg BW of immunological mechanisms that suppress tumor growth for the induction of bone marrow suppression over a time in animals with an intact immune system. C-Kit blockade period of 4 days to avoid surgical complications and death for the induction of bone marrow suppression concurrently of the animals resulting from the immune incompetence by stimulates B lymphopoiesis [30, 31]. As the number of B bone marrow depletion. cell precursors normally decreases in tumor bearing mice As shown by Okamoto et al., bone marrow suppression [32, 33], it must be speculated that in our experiment of bone by anti-c-Kit pretreatment over a time period of 4 days before marrow suppression, c-Kit blockade stimulates B cell genesis subcutaneous implantation of colon tumor cells induced and thus increases tumor cell engraftment. Furthermore, B leucopenia which was still detectable 10 days after the lymphopoiesis is associated with increased angiogenesis and last injection [7]. Although tumor growth and sprouting cellular proliferation [34–36]. In our study, neoangiogenesis PCNA-positive tumor cells (%) 10 Journal of Oncology VEGF- precursor β-actin CXCR4 β-actin VEGF Control cKit-Ab cKit/SDF1-Ab Control cKit-Ab cKit/SDF1-Ab 7 7 5 5 3 3 2 2 0 0 Control cKit-Ab cKit/SDF1-Ab Control cKit-Ab cKit/SDF1-Ab (a) (b) Figure 6: Western blot analysis of VEGF (a) and CXCR4 (b) expression within CT 26.WT tumors five days after tumor cell implantation. Animals were pretreated with anti-c-Kit (cKit-Ab) alone or additionally treated with a SDF-1 neutralizing antibody (cKit/SDF1-Ab). Animals treated with isotype-matched control antibodies served as controls (control). Quantitative analysis showed that VEGF expression was significantly higher after anti-c-Kit pretreatment than in control animals (a). Additional neutralization of SDF-1 had no further effect on the increased VEGF expression. CXCR4 was also significantly higher expressed after anti-c-Kit pretreatment compared to controls and was not further influenced by additional SDF-1 neutralization (b). Mean ± SEM; P< 0.05 versus control. within the colorectal tumors of animals pretreated with significantly improved the long-term survival of MCA26 anti-c-Kit was comparable to controls despite the lack tumor bearing mice [41]. In our study, only 8% of the of functional HCs which are necessary for angiogenesis CT26.WT-GFP cells were c-Kit receptor positive. Therefore, and vasculogenesis. Thus, we hypothesize that the lack of we do not expect a direct inhibitory effect of anti-c-Kit HCs was compensated by local angiogenic factors or B treatment on tumor cell proliferation. lymphopoiesis-associated angiogenesis, resulting in a tumor Recruitment of HCs and EPC is predominantly mediated neovascularization comparable to that observed in controls. by SDF-1 and its receptor CXCR4 [9, 10] because HCs and The SCF receptor c-Kit has been shown to be essential EPCs migrate along a chemotactic gradient towards higher for the development of blood cells, melanocytes, germ cells, concentrations of SDF-1 [12–14]. Kaminski et al. have shown interstitial cells of Cajal in the gastrointestinal tract, and that almost 100% of bone marrow and peripheral blood c- mast cells [37]. Furthermore, the majority of tumor cells, Kit cells are positive for CXCR4 [12]. In the present study, especially those of the neural axis, breast, lung, prostate, we could demonstrate that neutralization of SDF-1 after anti- and colon show an aberrant c-Kit expression [38]. Especially c-Kit pretreatment significantly reduces neovascularization gastrointestinal stromal tumors (GISTs) express c-Kit on the most probably by the inhibition of HC and EPC recruitment cell-surface, and mutations of Kit in these tumors results in within the tumors. Additionally, in combination with anti-c- an activation of Kit signaling, which leads to uncontrolled Kit pretreatment, SDF-1 neutralization is capable of decreas- cell proliferation and resistance to apoptosis [39, 40]. Today, ing tumor cell proliferation and invasion. Taken together, patients with GIST are treated with Imatinib, an inhibitor of neutralization of SDF-1 counteracts the stimulating effects certain protein tyrosine kinases including KIT, depending on of bone marrow suppression by anti-c-Kit treatment on the mitotic index of the GIST. Imatinib induces an arrest of tumor cell engraftment and inhibits compensatory local and tumor cell proliferation and causes apoptotic cell death. In B lymphopoiesis-associated angiogenesis. established MCA26 tumors, Pan et al. showed that injection As described by Kaminski et al., the presence of SDF- of anti-c-Kit antibodies markedly reduces tumor-induced 1 chemoattractant activity and inflammatory endothelial immune tolerance exhibited by myeloid-derived suppressors activation by TNF-α is required for c-Kit cells to form in mice [41]. Furthermore, their experiments demonstrate functionally relevant interactions with the endothelium in that anti-c-Kit treatment can prevent tumor-specific T-cell postcapillary venules [12]. Blockade of ICAM-1 and CXCR4 anergy and development of T regulatory cells (Treg). In abolishes adhesion of c-Kit cells to the vascular endothe- combination with an immune modulatory therapy of IL-12 lium despite application of SDF-1 and TNF-α.Moreover, plus 4-1BB activation, treatment with anti-c-Kit antibodies in their cremaster muscle microcirculation model, stem cell VEGF expression (OD/mm tumor) CXCR4 expression (OD/mm tumor) Journal of Oncology 11 adhesion was significantly reduced when eNOS was not [12] A. Kaminski, N. Ma, P. Donndorf et al., “Endothelial NOS is required for SDF-1alpha/CXCR4-mediated peripheral present or systemic NOS inhibited [12]. As SDF-1 and TNF- endothelial adhesion of c-kit+ bone marrow stem cells,” α stimulation activates the endothelium by an increase of the Laboratory Investigation, vol. 88, no. 1, pp. 58–69, 2008. CXCR4 expression leading to relevant stem cell attraction [13] J. A. Burger and T. J. Kipps, “CXCR4: a key receptor in the [12], our results indicate that the increase of CXCR4 and crosstalk between tumor cells and their microenvironment,” VEGF expression within the tumors represents a compen- Blood, vol. 107, no. 5, pp. 1761–1767, 2006. satory pathway after anti-c-Kit pretreatment. Neutralization [14] T. Papayannopoulou, “Current mechanistic scenarios in of SDF-1 inhibits interactions of c-Kit positive cells with hematopoietic stem/progenitor cell mobilization,” Blood, vol. tumor vessels and as a consequence, leads to inhibition of 103, no. 5, pp. 1580–1585, 2004. tumor neovascularization. [15] M. De Palma, M. A. Venneri, C. Roca, and L. Naldini, “Target- In conclusion, bone marrow suppression by anti-c-Kit ing exogenous genes to tumor angiogenesis by transplantation pretreatment significantly enhances tumor cell engraftment of genetically modified hematopoietic stem cells,” Nature of colorectal tumors. As anti-SDF-1 treatment counteracts Medicine, vol. 9, no. 6, pp. 789–795, 2003. this increased tumor outgrowth by inhibition of neovascu- [16] J. R. Gother ¨ t, S. E. Gustin, J. A. M. van Eekelen et al., larization, the SDF-1/CXCR4 pathway seems to be crucial for “Genetically tagging endothelial cells in vivo: bone marrow- tumor angiogenesis mediated by HCs and EPCs. derived cells do not contribute to tumor endothelium,” Blood, vol. 104, no. 6, pp. 1769–1777, 2004. [17] S. Purhonen, J. Palm, D. Rossi et al., “Bone marrow- Acknowledgment derived circulating endothelial precursors do not contribute to vascular endothelium and are not needed for tumor growth,” The authors appreciate the excellent technical assistance of Proceedings of the National Academy of Sciences of the United Janine Becker and Ruth Nickels. States of America, vol. 105, no. 18, pp. 6620–6625, 2008. [18] O. Kollmar, M. K. Schilling, and M. D. Menger, “Experimental liver metastasis: standards for local cell implantation to study References isolated tumor growth in mice,” Clinical and Experimental Metastasis, vol. 21, no. 5, pp. 453–460, 2004. [1] G. Bergers and L. E. Benjamin, “Tumorigenesis and the [19] M. D. Menger, M. W. Laschke, and B. Vollmar, “Viewing angiogenic switch,” Nature Reviews Cancer,vol. 3, no.6,pp. the microcirculation through the window: some twenty 401–410, 2003. years experience with the hamster dorsal skinfold chamber,” [2] P. Carmeliet and R. K. Jain, “Angiogenesis in cancer and other European Surgical Research, vol. 34, no. 1-2, pp. 83–91, 2002. diseases,” Nature, vol. 407, no. 6801, pp. 249–257, 2000. [20] O. Kollmar, K. Rupertus, C. Scheuery et al., “Stromal cell- [3] M.A.GimbroneJr.,S.B.Leapman, R. S. Cotran,and derived factor-1 promotes cell migration and tumor growth of J. Folkman, “Tumor dormancy in vivo by prevention of colorectal metastasis,” Neoplasia, vol. 9, no. 10, pp. 862–870, neovascularization,” Journal of Experimental Medicine, vol. 136, no. 2, pp. 261–276, 1972. [21] O. Kollmar, B. Junker, K. Rupertus, M. D. Menger, and M. [4] R. S. Kerbel, “Tumor angiogenesis: past, present and the near K. Schilling, “Studies on MIP-2 and CXCR2 expression in a future,” Carcinogenesis, vol. 21, no. 3, pp. 505–515, 2000. mouse model of extrahepatic colorectal metastasis,” European [5] K. J. Luzzi, I. C. MacDonald, E. E. Schmidt et al., “Multistep Journal of Surgical Oncology, vol. 33, no. 6, pp. 803–811, 2007. nature of metastatic inefficiency,” American Journal of Pathol- [22] K. Rupertus,O.Kollmar,C.Scheuer,B.Junker,M.D.Menger, ogy, vol. 153, no. 3, pp. 865–873, 1998. and M. K. Schilling, “Major but not minor hepatectomy [6] D. Gao, D. J. Nolan, A. S. Mellick, K. Bambino, K. McDonnell, accelerates engraftment of extrahepatic tumor cells,” Clinical and V. Mittal, “Endothelial progenitor cells control the and Experimental Metastasis, vol. 24, no. 1, pp. 39–48, 2007. angiogenic switch in mouse lung metastasis,” Science, vol. 319, [23] O. Kollmar, B. Junker, K. Rupertus, C. Scheuer, M. D. no. 5860, pp. 195–198, 2008. Menger, and M. K. Schilling, “Liver resection-associated [7] R. Okamoto, M. Ueno, Y. Yamada et al., “Hematopoietic cells macrophage inflammatory protein-2 stimulates engraftment regulate the angiogenic switch during tumorigenesis,” Blood, but not growth of colorectal metastasis at extrahepatic sites,” vol. 105, no. 7, pp. 2757–2763, 2005. Journal of Surgical Research, vol. 145, no. 2, pp. 295–302, 2008. [8] D.C.Rafii, B. Psaila,J.Butler, D. K. Jin, andD.Lyden,“Reg- [24] T. Asahara, H. Masuda, T. Takahashi et al., “Bone marrow ulation of vasculogenesis by platelet-mediated recruitment of origin of endothelial progenitor cells responsible for postnatal bone marrow-derived cells,” Arteriosclerosis, Thrombosis, and vasculogenesis in physiological and pathological neovascular- Vascular Biology, vol. 28, no. 2, pp. 217–222, 2008. ization,” Circulation Research, vol. 85, no. 3, pp. 221–228, 1999. [9] K. Hattori, B. Heissig, K. Tashiro et al., “Plasma elevation of [25] D. Lyden, K. Hattori, S. Dias et al., “Impaired recruitment stromal cell-derived factor-1 induces mobilization of mature of bone-marrow-derived endothelial and hematopoietic pre- and immature hematopoietic progenitor and stem cells,” cursor cells blocks tumor angiogenesis and growth,” Nature Blood, vol. 97, no. 11, pp. 3354–3360, 2001. Medicine, vol. 7, no. 11, pp. 1194–1201, 2001. [10] J. Deshane, S. Chen, S. Caballero et al., “Stromal cell-derived [26] A. M. Davidoff,C.Y.C.Ng, P. Brownetal., “Bonemarrow- factor 1 promotes angiogenesis via a heme oxygenase 1- derived cells contribute to tumor neovasculature and, when dependent mechanism,” Journal of Experimental Medicine, vol. modified to express an angiogenesis inhibitor, can restrict 204, no. 3, pp. 605–618, 2007. tumorgrowthinmice,” Clinical Cancer Research, vol. 7, no. 9, pp. 2870–2879, 2001. [11] S. Mendez-F ´ errer, D. Lucas, M. Battista, and P. S. Frenette, “Haematopoietic stem cell release is regulated by circadian [27] D. J. Nolan, A. Ciarrocchi, A. S. Mellick et al., “Bone marrow- oscillations,” Nature, vol. 452, no. 7186, pp. 442–447, 2008. derived endothelial progenitor cells are a major determinant 12 Journal of Oncology of nascent tumor neovascularization,” Genes and Development, vol. 21, no. 12, pp. 1546–1558, 2007. [28] M. Grunewald, I. Avraham, Y. Dor et al., “VEGF-induced adult neovascularization: recruitment, retention, and role of accessory cells,” Cell, vol. 124, no. 1, pp. 175–189, 2006. [29] V. R. S. Patil, E. B. Friedrich, A. E. Wolley, R. E. Gerszten, J. R. Allport, and R. Weissleder, “Bone marrow-derived lin-c- kit+Sca-1+ stem cells do not contribute to vasculogenesis in Lewis lung carcinoma,” Neoplasia, vol. 7, no. 3, pp. 234–240, [30] M. Ogawa, Y. Matsuzaki, S. Nishikawa et al., “Expression and function of c-kit in hemopoietic progenitor cells,” Journal of Experimental Medicine, vol. 174, no. 1, pp. 63–71, 1991. [31] S. A. Rico-Vargas, B. Weiskopf, S. I. Nishikawa, and D. G. Osmond, “c-kit Expression by B cell precursors in mouse bone marrow: stimulation of B cell genesis by in vivo treatment with anti-c-kit antibody,” Journal of Immunology, vol. 152, no. 6, pp. 2845–2852, 1994. [32] J. Richards, B. McNally, X. Fang, M. A. Caligiuri, P. Zheng, and Y. Liu, “Tumor growth decreases NK and B cells as well as common lymphoid progenitor,” PLoS One, vol. 3, no. 9, Article ID e3180, 2008. [33] I. Kirman, E. H. Huang, and R. L. Whelan, “B cell response to tumor antigens is associated with depletion of B progenitors in murine colocarcinoma,” Surgery, vol. 135, no. 3, pp. 313–318, [34] F. Pessler, L. Dai, C. Diaz-Torne et al., “Increased angiogenesis and cellular proliferation as hallmarks of the synovium in chronic septic arthritis,” Arthritis Care and Research, vol. 59, no. 8, pp. 1137–1146, 2008. [35] A. Tzankov, S. Heiss, S. Ebner et al., “Angiogenesis in nodal B cell lymphomas: a high throughput study,” Journal of Clinical Pathology, vol. 60, no. 5, pp. 476–482, 2007. [36] D. Gratzinger, S. Zhao, R. J. Tibshirani et al., “Prognostic significance of VEGF, VEGF receptors, and microvessel density in diffuse large B cell lymphoma treated with anthracycline- based chemotherapy,” Laboratory Investigation,vol. 88, no.1, pp. 38–47, 2008. [37] B. Heissig, Z. Werb, S. Rafii, and K. Hattori, “Role of c-kit/Kit ligand signaling in regulating vasculogenesis,” Thrombosis and Haemostasis, vol. 90, no. 4, pp. 570–576, 2003. [38] A. M. Turner,K.M.Zsebo,F.Martin, F. W. Jacobsen,L.G. Bennett, and V. C. Broudy, “Nonhematopoietic tumor cell lines express stem cell factor and display c- kit receptors,” Blood, vol. 80, no. 2, pp. 374–381, 1992. [39] M. C. Heinrich, K. Owzar, C. L. Corless et al., “Correlation of kinase genotype and clinical outcome in the North American intergroup phase III trial of imatinib mesylate for treatment of advanced gastrointestinal stromal tumor: CALGB 150105 study by cancer and leukemia group B and southwest oncology group,” Journal of Clinical Oncology, vol. 26, no. 33, pp. 5360– 5367, 2008. [40] G. D. Demetri, M. von Mehren, C. D. Blanke et al., “Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors,” The New England Journal of Medicine, vol. 347, no. 7, pp. 472–480, 2002. [41] P. Y. Pan, G. X. Wang, B. Yin et al., “Reversion of immune tolerance in advanced malignancy: modulation of myeloid- derived suppressor cell development by blockade of stem-cell factor function,” Blood, vol. 111, no. 1, pp. 219–228, 2008. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Hindawi Publishing Corporation

Bone Marrow Suppression by c-Kit Blockade Enhances Tumor Growth of Colorectal Metastases through the Action of Stromal Cell-Derived Factor-1

Loading next page...
 
/lp/hindawi-publishing-corporation/bone-marrow-suppression-by-c-kit-blockade-enhances-tumor-growth-of-y7n8B5kn9J

References (43)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2012 Kathrin Rupertus et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1687-8450
eISSN
1687-8469
DOI
10.1155/2012/196957
Publisher site
See Article on Publisher Site

Abstract

Hindawi Publishing Corporation Journal of Oncology Volume 2012, Article ID 196957, 12 pages doi:10.1155/2012/196957 Research Article Bone Marrow Suppression by c-Kit Blockade Enhances Tumor Growth of Colorectal Metastases through the Action of Stromal Cell-Derived Factor-1 1, 2 1 3 3 Kathrin Rupertus, Gudrun C. Y. Haberl, Claudia Scheuer, Michael D. Menger, 1 1 Martin K. Schilling, and Otto Kollmar Department of General, Visceral, Vascular and Pediatric Surgery, University of Saarland, 66421 Homburg, Germany Department of Hematology, Oncology and Immunology, University of Tubinge ¨ n, 72074 Tubinge ¨ n, Germany Institute for Clinical and Experimental Surgery, University of Saarland, 66421 Homburg, Germany Correspondence should be addressed to Otto Kollmar, otto.kollmar@uniklinikum-saarland.de Received 1 June 2011; Revised 2 August 2011; Accepted 7 August 2011 Academic Editor: Debabrata Mukhopadhyay Copyright © 2012 Kathrin Rupertus et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Mobilization of c-Kit hematopoietic cells (HCs) contributes to tumor vascularization. Whereas survival and proliferation of HCs are regulated by binding of the stem cell factor to its receptor c-Kit, migration of HCs is directed by stromal cell-derived factor (SDF)-1. Therefore, targeting migration of HCs provides a promising new strategy of anti-tumor therapy. Methods. BALB/c mice (n = 16) were pretreated with an anti-c-Kit antibody followed by implantation of CT26.WT-GFP colorectal cancer cells into dorsal skinfold chambers. Animals (n = 8) additionally received a neutralizing anti-SDF-1 antibody. Animals (n = 8) treated with a control antibody served as controls. Investigations were performed using intravital fluorescence microscopy, immunohistochemistry, flow cytometry and western blot analysis. Results. Blockade of c-Kit significantly enhanced tumor cell engraftment compared to controls due to stimulation of tumor cell proliferation and invasion without markedly affecting tumor vascularization. C-Kit blockade significantly increased VEGF and CXCR4 expression within the growing tumors. Neutralization of SDF-1 completely antagonized this anti-c-Kit-associated tumor growth by suppression of tumor neovascularization, inhibition of tumor cell proliferation and reduction of muscular infiltration. Conclusion. Our study indicates that bone marrow suppression via anti-c-Kit pretreatment enhances tumor cell engraftment of colorectal metastases due to interaction with the SDF-1/CXCR4 pathway which is involved in HC-mediated tumor angiogenesis. 1. Introduction Although EPCs incorporate into tumors in only small numbers, targeting EPCs has been shown to be effective Angiogenesis is one of the crucial steps in tumor progression in reducing tumor angiogenesis and tumor growth in and metastasis [1, 2]. Due to the lack of oxygen supply experimental models [6]. One of the most important factors and the accumulation of toxic products, avascular tumors mediating survival and proliferation of HCs is the stem and tumor metastases cannot grow beyond a critical size cell factor (SCF) which binds to the c-Kit receptor on the of ∼1-2 mm and thus will stay clinically occult [1, 3, 4]. surface of HCs. Okamoto et al. have shown that bone marrow Therefore, only a small proportion of circulating cancer cells suppression by anti-c-Kit treatment induces a delay in tumor finally forms macroscopic tumors [5]. Tumor vessels can angiogenesis due to the inhibition of angiogenic sprouting grow by sprouting of preexisting host vessels, intussusception in colon tumors and that c-Kit blockade suppresses tumor or incorporation of bone marrow-derived endothelial pro- growth of subcutaneously implanted prostate carcinomas genitor cells (EPCs) which are a subset of hematopoietic cells (PC3) by inhibition of tumor angiogenesis regulated by HCs (HCs). This mechanism is called vasculogenesis and mimics embryonic angio-development [2]. [7]. 2 Journal of Oncology Recruitment of HCs and EPCs to avascular areas is The animals were housed in single cages at room temperature orchestrated by different angiogenic growth factors and of 22–24 C and at a relative humidity of 60–65% with a 12- cytokines [8], predominantly by the CXC-chemokine stro- hour light/dark cycle environment. The mice were allowed mal cell-derived factor (SDF)-1 and its receptor CXCR4 free access to drinking water and standard laboratory chow [9, 10]. HCs and EPCs fluctuate to peripheral organs in (Altromin; Lage, Germany). antiphase with the expression of SDF-1 within the bone marrow microenvironment [11] and migrate alongside a 2.3. Experimental Model—Dorsal Skinfold Chamber. To chemotactic gradient towards higher concentrations of SDF- allow repetitive analyses of the microcirculation of the grow- 1, for example, sites of injury and tumor tissue [12– ing tumors, the dorsal skinfold chamber model was used 14]. Furthermore, it has been demonstrated that SDF-1- for intravital microscopy as described previously in detail mediated recruitment of c-Kit-positive cells to the periphery [19]. For operative procedures, animals were anesthetized by is dependent on cofactors, including tumor necrosis factor-α intraperitoneal injection of 90 mg/kg BW ketamine (Ketavet, and the endothelial nitric oxide synthase (eNOS) [12]. Parke Davis; Freiburg, Germany) and 20 mg/kg BW xylazine However, the impact of circulating HCs and EPCs on (Rompun, Bayer; Leverkusen, Germany). The chamber, con- vasculogenesis and sprouting angiogenesis in tumor angio- sisting of two symmetrical titanium frames, was positioned development is still controversially discussed in the literature to sandwich the extended double layer of the dorsal skin. [15–17], and the role of the SDF-1/CXCR4 pathway is One layer of skin and subcutis was completely removed in not yet fully understood. Therefore, in the presented study a circular area of 15 mm diameter. The remaining layers, consisting of the epidermis, subcutaneous tissue, and striated we analyzed the influence of bone marrow suppression by skin muscle, were covered with a glass coverslip incorporated anti-c-Kit treatment combined with SDF-1 neutralization on tumor cell engraftment and neovascularization using a into one of the titanium frames. The animals tolerated the chambers well and showed no signs of discomfort or changes murine model of colorectal tumor metastasis. in sleeping and feeding habits. After a 48-hour recovery period, the animals were reanesthetized. For tumor cell 2. Materials and Methods implantation, the coverslip of the chamber was temporarily removed and 1 × 10 CT26.WT-GFP cells were implanted 2.1. Tumor Cell Line and Culture Conditions. The CT26 onto the surface of the striated muscle tissue within the cell line is an N-nitroso-N-methyl-urethane-induced undif- chamber. Immediately after cell implantation, the chamber ferentiated adenocarcinoma of the colon, syngeneic with tissue was covered again with the coverslip [20–22]. the BALB/c mouse. For our studies [18], the CT26.WT cells (ATCC CRL-2638, LGC Promochem GmbH, Wesel, Germany) were transfected with the enhanced GFP expres- 2.4. Experimental Protocol. Animals were assigned to three sion vector pEGFP-N1 (Clontech) with the use of CLON- different groups: the first group (cKit-Ab; n = 8) received fectin (Clontech, Palo Alto, CA, USA) according to the a pretreatment with a monoclonal anti-c-Kit receptor anti- manufacturer’s instructions. For the individual experiments, body (ACK45, BD Biosciences, Heidelberg, Germany) by CT26.WT-GFP cellsweregrown in cell cultureasmono- daily intraperitoneal injections starting 4 days before tumor layers in RPMI-1640 medium with 2 mM L-glutamine cell implantation. The ACK45-antibody was given four times (Sigma Aldrich Chemie GmbH, Taufkirchen, Germany) at a dose of 1 mg/kg BW daily as described by Okamoto supplemented with 10% fetal calf serum (FCS Gold, PAA et al. [7]. Animals of the second group (cKit/SDF1-Ab; Laboratories GmbH, Colbe, ¨ Germany), 100 U/mL penicillin, n = 8) were also pretreated with ACK45 as described and 100 μg/mL streptomycin (PAA Laboratories GmbH). above. Additionally, these animals received 1 mg/kg BW of a The cells were incubated at 37 C in a humidified atmosphere monoclonal mouse anti-mouse SDF-1 antibody (MAB310, R containing 5% CO , and only cells of the first three and D Systems, Wiesbaden, Germany). MAB310 application serial passages after cryostorage were used. At the day of was performed intraperitoneally, starting at the day of implantation, tumor cells were harvested from subconfluent tumor cell implantation (d0) and repeated every second day cultures (70 to 85%) by trypsinization (0.05% Trypsin and thereafter until day 12. The third group (control; n = 8) 0.02% EDTA, PAA Laboratories GmbH) and washed twice served as control and received pre- and posttreatment the in phosphate-buffered saline solution (PBS). same amount of the corresponding isotype-matched IgG control antibodies (A95-1, BD Biosciences, and MAB002 R&D Systems). All animals underwent repetitive intravital 2.2. Animals. Experiments were performed after approval by microscopic analyses directly (d0) as well as 5, 8, 11, and the local governmental ethic committee and conformed to 14 days after tumor cell implantation. At the end of the the United Kingdom Coordinating Committee on Cancer experiment (day 14), the chamber with the tumor tissue was Research (UKCCCR) Guidelines for the Welfare of Animals harvested for histology and immunohistochemistry. in Experimental Neoplasia (as described in 1998 in Br J Can- cer 77: 1–10) and the Guide for Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National 2.5. Intravital Fluorescence Microscopy. Intravital fluores- Research Council; NIH Guide, Vol. 25, No. 28, 1996). Female cence microscopy was performed in epi-illumination tech- BALB/c mice (Charles River Laboratories GmbH; Sulzfeld, nique using a modified Zeiss Axio-Tech microscope (Zeiss, Germany) with a body weight (BW) of 18–20 g were used. Oberkochen, Germany) with a 100-W HBO mercury lamp. Journal of Oncology 3 Microscopic images were monitored by a charge-coupled techniques. Therefore, deparaffinized sections were incu- device video camera (FK 6990, COHU, Prospective Mea- bated with 3% H O and 2% goat normal serum to 2 2 block endogenous peroxidases and unspecific binding sites. surements Inc., San Diego, CA, USA) and were transferred A monoclonal mouse anti-pan PCNA antibody (PC10, to a video system (VO-5800 PS, Sony, Munchen, ¨ Germany) DakoCytomation, Hamburg, Germany) and a polyclonal for subsequent off-line analysis. Tumor size, growth kinet- rabbit anti-mouse cleaved caspase-3 antibody (Asp175, Cell ics, migration of tumor cell, and neovascularization were Signaling Technology, Frankfurt, Germany) were used as analyzed using blue light epi-illumination (450 to 490 nm primary antibodies. Goat anti-mouse and goat anti-rabbit excitation wavelength and >520 nm emission wavelengths) POD-conjugated antibodies were used as secondary antibod- [20–22]. ies for streptavidin-biotin-complex peroxidase staining. 3,3 diaminobenzidine (DakoCytomation) served as chromogen. 2.6. Microcirculation Analysis. Microcirculatory parameters Sections were counterstained with Hemalaun according to were assessed off line by frame-to-frame analysis of the Mayer and examined by light microscopy. videotaped images using a computer-assisted image analysis To assess the expression of CD31 as a marker for endothe- system (CapImage, Zeintl Software, Heidelberg, Germany). lial cells, zinc fixative fixed paraffin sections of tumor tissue The fluorescent labeling of the tumor cells allowed precise were used. After incubation with 3% H O and 2% goat nor- 2 2 delineation of the tumor from the surrounding host tissue. mal serum to block endogenous peroxidases and unspecific It also enabled for distinct identification of individual tumor binding sites, deparaffinized sections were incubated with a cells to study tumor cell migration. At each observation time rat anti-mouse CD31 antibody (MEC 13.3, BD Biosciences). point, the surface of the fluorescently labeled tumor mass A polyclonal goat anti-rat IgG antibody (BD Biosciences) within the chamber was scanned for determination of the 2 was used as secondary antibody. Colorimetric detection was tumor size (given as tumor area in mm ). Eight regions of performed using 3,3 diaminobenzidine (DakoCytomation) interest (ROIs) were randomly chosen next to the tumor substrates. Sections were counterstained with Hemalaun margin. The number of migrating cells was counted, and the according to Mayer and examined by light microscopy. distance to the tumor margin was measured (given in μm). Microcirculation was quantified as described before in detail [22]. First, eight representative ROIs within the tumor 2.8. Flow Cytometric Analysis of CT26.WT-GFP Cells. FAC- margin were chosen and analyzed for microcirculation Scan (Becton Dickinson, Mountain View, CA, USA) analysis parameters. In these ROIs, the onset of angiogenesis, that is, was performed to assess the expression of c-Kit on the the existence of angiogenic buds, sprouts, and newly formed CT26.WT-GFP cells in triplicate. Cells were fixed with 2% blood vessels, was documented and scored 0 to 8, with 0 formalin, washed twice with PBS, resuspended in FACS indicating existence of newly formed tumor microvessels in buffer and incubated with an FITC-conjugated rat anti- none of the ROIs and 8 indicating their existence in all of mouse IgG2b c-Kit antibody (1 : 50; 553354, BD Biosciences) the ROIs. Functional capillary density (given in cm/cm ) or an FITC-conjugated isotype-matched control antibody of the tumor microvessels was measured to quantify the (553988, BD Biosciences). Cells were washed again and then angiogenic activity. This parameter was defined as the length maintained in 2% paraformaldehyde in PBS. Tumor cells of red blood cell perfused microvessels per observation were selectively analyzed for their fluorescence properties area and was analyzed within the eight ROIs of the tumor using the CellQuest data handling program (BD Biosciences) margin and within four additional ROIs of the tumor center. with assessment of 5000 events per sample. The flow cytome- Diameters of the newly formed tumor microvessels were ter was calibrated with fluorescent standard microbeads measured perpendicularly to the vessel path (given in μm). (CaliBRITE Beads, BD Biosciences). To study vascular permeability of the newly formed tumor microvessels, petechial bleedings were documented in each 2.9. Western Blot Analysis. To study protein expression of the ROIs, and given as percentage of all the ROIs analyzed patterns, additional Western blot analyses were performed on [20, 21, 23]. tumor specimen gained from the dorsal skinfold chamber. Therefore, 12 additional animals received CT26.WT-GFP 2.7. Histology and Immunohistochemistry. At the end of the tumor cell implantation in the dorsal skinfold chamber experiment (day 14), the tumor and the adjacent host and were assigned to the three groups as described above. tissue were harvested and immediately fixed in formalin. For Eight animals were pretreated with the anti-c-Kit antibody light microscopy, formalin-fixed biopsies were embedded in (ACK45, BD Biosciences). Four of these pretreated animals paraffin. Sections of 5 μm were cut and stained with hema- received additional treatment with the neutralizing anti- toxylin and eosin (HE) according to standard procedures to SDF-1 antibody (MAB310, R and D Systems) starting at the analyze tumor growth characteristics. Tumor cell invasion day of tumor cell implantation. Four animals received the of the muscular layer on the surface of the dorsal skinfold same amount of the corresponding isotype-matched control chamber was quantified over the entire tumor basis and given antibody (A95-1, BD Biosciences). as percentage of the length of the tumor basis. For whole protein extracts and Western blot analysis of To study tumor cell proliferation and apoptotic cell the expression of vascular endothelial growth factor (VEGF), death, proliferating cell nuclear antigen (PCNA) and cleaved the chemokine receptor CXCR4 and eNOS, CT26.WT- caspase-3 were stained using indirect immunoperoxidase GFP-tumors were completely removed from the skinfold 4 Journal of Oncology at day 5. The tumor tissue samples were homogenized of the ROIs by day 5 after tumor cell implantation. At separately in lysis buffer and a protease inhibitor cock- day 8, angiogenesis was seen in almost 100% of the ROIs. tail (Sigma, Taufkirchen, Germany), incubated on ice and Pretreatment with the anti-c-Kit antibody did not influence centrifuged at 16.000×g. The supernatant was saved as the onset of the angiogenic switch compared to controls. whole protein extract fraction. Protein concentrations were Of interest, blockade of SDF-1 after pretreatment with anti- determined using the Lowry assay with bovine serum c-Kit significantly delayed the angiogenic switch, with only albumin as standard. Ten microgram protein per lane ∼50% of the ROIs showing newly formed microvessels by were separated discontinuously on sodium dodecyl sulfate day 8 and ∼70% by day 11 (P< 0.05). In this group, polyacrylamide gels (10% SDS-PAGE) and transferred to a angiogenesis was observed in all ROIs only at the end of the polyvinyldifluoride membrane (0.2 μm, BioRad, Munc ¨ hen, observation period (Figure 2). Germany). After blockade of nonspecific binding sites, The differential effects of c-Kit blockade with or without membranes were incubated with an anti-VEGF antibody additional SDF-1 blockade were also reflected by quantitative (A-20, Santa Cruz, Heidelberg, Germany), an anti-CXCR4 analysis of the microvascular densities of the newly formed antibody (ab2074, Abcam, Heidelberg, Germany), or an anti- tumor vessels. As characteristic for tumor vessels, the vas- eNOS antibody (BD Transduction Lab., Heidelberg, Ger- cular network of the tumors consisted of irregularly shaped many) followed by the corresponding secondary peroxidase- and chaotically arranged microvessels (Figures 3(a)–3(c)). conjugated antibodies (GE Healthcare, Freiburg, Germany Whereas no differences of microvascular density within and Santa Cruz). Protein expression was visualized by the tumor vasculature could be observed between controls means of luminol enhanced chemiluminescence (ECL, GE and anti-c-Kit pretreated animals, additional treatment Healthcare) and exposure of the membranes to a blue- with the anti-SDF-1 antibody significantly decreased the light-sensitive autoradiography film (Hyperfilm ECL, GE microvascular density within the tumor center compared Healthcare). Signals were densitometrically assessed (Bio- to the other groups from day 8 until the end of the Rad, Gel-Dokumentationssystem) and normalized to β-actin observation period (Figure 3(d)). Quantitative analysis of the signals (mouse monoclonal anti-β-actin, Sigma) to correct functional microvascular density within the tumor margin for unequal loading. showed similar results (data not shown). The functional microvascular density within the tumor margin and the tumor center was not significantly different within the 2.10. Statistical Analysis. Allvaluesare expressedasmeans individual treatment groups. ± SEM. After proving the assumption of normality and In contrast to the highly vascularized CT26.WT- homogeneity of variance across groups, differences between GFP tumors within the intravital fluorescence microscopy, groups were calculated by a one-way analysis of variance immunohistological staining for CD31 as a marker for (ANOVA) followed by the appropriate post hoc comparison, endothelial cells displayed positive staining of only a few cells including correction of the alpha error according to Bonfer- within tumor microvessels without significant differences roni probabilities to compensate for multiple comparisons. between the three groups (data not shown). Overall statistical significance was set at P< 0.05. Statistical As neovascularization is usually associated with vasodila- analysis was performed with the use of the software package tion due to the action of VEGF, microvessel diameters within SigmaStat (SPSS Inc, Chicago, Ill). the tumor were analyzed. However, no overall differences between the diameters of the microvessels within the tumor 3. Results margin and the tumor center could be observed (Table 1). During the 14 days observation period, diameters slightly 3.1. Tumor Growth. All animals had an uneventful post- increased in all groups until day 11, particularly in the operative recovery and tolerated well the dorsal skinfold control and anti-c-Kit-pretreated groups. Of interest, at day chamber implantation and the repetitive intravital micro- 14, microvessel diameters within tumors of mice treated with scopic analyses. The general conditions of the mice were not anti-c-Kit and anti-SDF-1 were markedly smaller compared affected and no changes in feeding or sleeping habits were to the other two groups (Table 1). observed. The take rate of the CT26.WT-GFP cells within Petechial bleedings within the areas of angiogenesis the dorsal skinfold chamber was 100%, and progressive of growing tumors are a characteristic indicator of the tumor growth was observed in all groups throughout the VEGF action on vascular permeability. To study this VEGF- entire observation period (Figures 1(a)–1(c)). Quantitative induced increase of vascular permeability, we analyzed the analysis of the increase of the tumor area revealed a petechial bleedings within the ROIs by intravital fluorescence significantly enhanced engraftment of tumor cells and a microscopy. Microbleedings were observed in all tumors. At significant stimulation of tumor growth after pretreatment day 5, signs of petechial bleedings were observed in ∼6% with the anti-c-Kit antibody compared to controls (P< of the tumor area of control tumors, whereas petechial 0.05). Interestingly, additional blockage of SDF-1 completely bleedings occurred in up to 30% of the tumor area in blunted this enhancement of tumor growth leading to similar animals pretreated with anti-c-Kit at that time. In all groups, tumor sizes as measured in controls (P< 0.05; Figure 1(d)). petechial bleedings were most pronounced 8 days after tumor cell implantation (control: 15–20%, anti-c-Kit: ∼30%, anti- c-Kit/anti-SDF-1: 35–40%). Comparing the three groups, 3.2. Angiogenesis and Neovascularization. In control animals, newly developed microvessels could be detected in ∼50% no statistical significance could be found due to the high Journal of Oncology 5 (a) (b) 058 11 14 (Days) (c) (d) Figure 1: Time course of tumor growth of CT26.WT-GFP tumors in the dorsal skinfold chamber. Stereomicroscopy photographs of representative 14 days old tumors after either treatment with an isotype-matched control antibody (a), pretreatment with anti-c-Kit (b) alone, or pretreatment with anti-c-Kit followed by anti-SDF-1 treatment (c). Quantitative analysis of the tumor area (d) displayed progressive tumor growth in all groups. After anti-c-Kit pretreatment (grey triangles), tumor growth was significantly accelerated compared to controls (white circles). Of interest, additional neutralization of SDF-1 (black squares) completely blunted this anti-c-Kit-associated enhancement of ∗ # tumor growth. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a)–(c) ×4. Table 1: Microvessel diameters within the tumor margin and the tumor center of control animals, animals pretreated with anti-c-Kit and animals pretreated with anti-c-Kit followed by anti-SDF-1 treatment. All values are given in μm. Diameters slightly increased during the observation period. No significant differences could be observed between the tumor margin and center. Fourteen days after tumor cell implantation, microvascular diameters within the tumor center were significantly smaller in anti-c-Kit/anti-SDF-1 treated animals compared to controls and to anti-c-Kit treated animals. Time Control Anti-c-Kit Anti-c-Kit/anti-SDF-1 d5 12.42 ± 0.53 11.88 ± 0.77 13.02 ± 0.52 d8 13.97 ± 0.48 14.60 ± 0.93 15.02 ± 0.57 Tumor margin d11 14.73 ± 1.36 15.39 ± 1.00 14.84 ± 1.06 d14 15.81 ± 1.20 15.24 ± 0.91 13.94 ± 0.75 d5 13.91 ± 0.67 12.85 ± 0.83 13.22 ± 1.00 d8 14.44 ± 0.52 13.93 ± 0.75 14.21 ± 0.47 Tumor centre d11 15.35 ± 1.40 15.26 ± 0.98 15.66 ± 1.83 ∗# d14 16.86 ± 1.10 16.84 ± 1.35 12.82 ± 0.67 ∗ # Data are given as mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Increase of tumor area (mm ) 6 Journal of Oncology infiltration compared to anti-c-Kit pretreated tumors (P< 0.05), resulting in invasive growth characteristics which were 8 comparable to controls (Figure 4(c)). PCNA as an indicator of cell proliferation displayed positive staining in 46.9 ± 3.1% of the tumor cells in control 6 # animals (Figure 5). After pretreatment with anti-c-Kit, the number of PCNA-positive tumor cells was significantly 4 higher compared to controls (P< 0.05, Figures 5(a) and 5(c)). In contrast, additional blockade of SDF-1 resulted in a significantly lower number of PCNA-positive tumor cells compared to controls and to anti-c-Kit pretreated animals (P< 0.05, Figures 5(b) and 5(c)). 0 To study apoptotic cell death, immunohistochemical 0 58 11 14 staining of cleaved caspase-3 products within the tumors was (Days) performed. Of interest, 14 days after tumor cell implantation, only a minor fraction of the total number of 351.3 ± 5.2 Figure 2: Semiquantitative analysis of the onset of angiogenesis within the CT 26.WT-GFP tumors. Animals were pretreated tumor cells within the high power fields (HPF) showed with anti-c-Kit (grey triangles) alone or anti-c-Kit and anti- positive staining for caspase-3. Whereas 0.73 ± 0.13 tumor SDF-1 neutralizing antibodies (black squares). Animals treated cells/HPF were positive for caspase-3 in controls, pretreat- with isotype-matched control antibodies served as controls (white ment with anti-c-Kit reduced apoptotic cell death within the circles). In controls and anti-c-Kit pretreated animals, ∼50% of the tumors (0.26 ± 0.06 tumor cells/HPF). In tumors of animals ROIs showed newly developed microvessels 5 days after tumor cell additionally treated with anti-SDF-1 antibodies, this effect implantation. By day 8, almost 100% of the ROIs were vascularized. was even more pronounced (0.15 ± 0.03 tumor cells/HPF, Treatment with anti-c-Kit and anti-SDF-1 resulted in a significant P< 0.05). delay of angiogenesis with only ∼50 and ∼70% of the ROIs showing newly developed microvessels by days 8 and 11. Within these tumors, angiogenesis could be observed in all ROIs only at day 14. 3.5. C-Kit Expression and Western Blot Analysis. FACScan ∗ # Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. analysis demonstrated that 8.2 ± 0.8% of the CT26-GFP cells transfected with the enhanced GFP expression vector pEGFP-N1 was c-Kit receptor positive. At day 5 after tumor cell implantation, the tumors standard deviation within the individual groups (data not of animals pretreated with anti-c-Kit antibodies showed shown). a significantly higher expression of VEGF and CXCR4 compared to controls (P< 0.05; Figures 6(a) and 6(b)). 3.3. Tumor Cell Migration. As SDF-1 has been demonstrated Additional SDF-1 blockade did not further increase VEGF to exert chemotactic effects on CT26.WT-GFP tumor cells, and CXCR4 expression. Furthermore, eNOS expression we additionally focused on migrating tumor cells next to within the tumors was decreased by anti-c-Kit pretreatment the tumor margin. The migrating tumor cells were easily with and without additional anti-SDF-1 treatment compared detectable due to their GFP labeling (Table 2). Starting at to controls (data not shown). day 5 after tumor cell implantation, tumor cell migration was detectable in all tumors until the end of the observation 4. Discussion period with slightly increasing distances from the tumor margin. The number of migrating tumor cells ranged The major finding of the present study is that bone between 20 and 25 per representative field during the marrow suppression by anti-c-Kit treatment significantly whole observation period. Treatment with anti-c-Kit alone enhances tumor cell engraftment of colorectal tumors due or additional blockade of SDF-1 did not impair tumor cell to an increase of tumor cell proliferation and invasion. migration compared to controls (Table 2). Additional anti-SDF-1 treatment neutralizes this increased tumor outgrowth by inhibition of tumor cell proliferation 3.4. Tumor Cell Morphology, Proliferation, and Apoptotic and tumor neovascularization. These findings suggest that Cell Death. Histological examinations on hematoxylin-eosin the enhanced tumor growth under the conditions of bone stained sections revealed solid tumor growth within the marrow suppression induced by anti-c-Kit treatment is dorsal skinfold chamber. Signs of malignant tumor growth related to the SDF-1/CXCR4 pathway. such as invasion of the adjacent host tissue by the tumor Bone-marrow-derived hematopoetic cells (HCs) con- cells were detectable in each group (Figure 4). Quantitative tribute to physiological and pathological vessel forma- analysis of tumor cell invasion showed a significant increase tion. During tumor growth, the release of cytokines and of tumor cell infiltration through the muscular layer of chemokines mediates the recruitment of EPCs and HCs the dorsal skinfold chamber in tumors of animals which contributing to the early initiation and stabilization of were pretreated with anti-c-Kit compared to controls (P< newly formed blood vessels, so-called vasculogenesis [24– 0.05). Of interest, tumors of animals additionally treated with anti-SDF-1 showed a significant reduction of muscular 26]. However, their precise role has not been fully elucidated Angiogenesis (score 0–8) Journal of Oncology 7 (a) (b) # 200 0 5 8 11 14 (Days) (c) (d) Figure 3: Time course of the functional capillary density within CT-26.WT-GFP tumors in dorsal skinfold chambers analyzed by intravital fluorescence microscopy. Representative fluorescence microscopic images show the network of chaotically arranged microvessels within the tumor center of control animals (a), animals pretreated with anti-c-Kit (b), and animals additionally treated with anti-SDF-1 (c) at day 11 after tumor cell implantation. Quantitative analysis of the functional capillary density (d) revealed a significant inhibition of tumor neovascularization within the tumor center after combined anti-c-Kit and anti-SDF-1 treatment (black squares) compared to controls (white circles). Anti-c-Kit pretreatment alone (grey triangles) did not significantly influence the extent of neovascularization. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a)–(c) ×40. Table 2: Tumor cell migration in dorsal skinfold chambers of control animals, animals pretreated with anti-c-Kit, and animals pretreated with anti-c-Kit followed by anti-SDF-1 treatment during a 14-day observation period. Migrated tumor cells next to the tumor margin were observed from day 5 until the end of the observation time. Comparing all three groups, no significant differences could be observed concerning the total number of migrated cells as well as their distance [μm] from the tumor margin. Of interest, at the later time points, tumor cells were found at greater distances from the tumor margin. Time Control Anti-c-Kit Anti-c-Kit/anti-SDF-1 d5 22.08 ± 1.33 23.00 ± 1.33 19.91 ± 1.14 d8 22.67 ± 1.33 23.14 ± 1.18 20.67 ± 1.07 Number d11 22.42 ± 1.53 23.48 ± 1.05 20.80 ± 1.34 d14 20.84 ± 1.71 23.96 ± 1.32 21.17 ± 1.30 d5 276.24 ± 16.08 291.82 ± 15.95 309.45 ± 50.93 d8 337.94 ± 9.38 351.11 ± 23.57 323.38 ± 20.45 Distance d11 383.85 ± 15.80 423.32 ± 25.65 360.83 ± 18.58 d14 421.50 ± 22.40 438.49 ± 32.01 458.51 ± 59.01 Data are given as mean ± SEM. Capillary density tumor center (cm/cm ) 8 Journal of Oncology (a) (b) Control cKit-Ab cKit/SDF1-Ab (c) Figure 4: Hematoxylin-eosin staining of CT26.WT-GFP tumors shows solid tumor growth 14 days after tumor cell implantation within the dorsal skinfold chamber. Sections display tumor infiltration through the underlying muscular layer (marked by arrows) after pretreatment with anti-c-Kit (a) and lack of muscular infiltration after pretreatment with anti-c-Kit and additional neutralization of SDF-1 (b). Quantitative analysis of tumor cell invasion of the muscular layer is given as percentage of the length of the tumor. Tumors of animals pretreated with anti-c-Kit showed a significantly pronounced infiltration of the muscular layer compared to controls. Of interest, blockade of SDF-1 after anti-c-Kit pretreatment abrogated this anti-c-Kit-associated increase of tumor cell infiltration. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a, b) ×88. yet. HCs and EPCs express stem cell markers such as c-Kit. growth, they do not contribute to tumor angiogenesis [29]. Whereas many studies indicate that bone-marrow-derived In a model of subcutaneously implanted prostate carcinoma, EPCs incorporate into tumor neovessels [25, 27], HCs may Okamoto et al. observed an accumulation of HCs around promote vasculogenesis via paracrine release of angiogenic newly formed blood vessels, but did not find these cells to be factors enhancing the recruitment and incorporation of part of the tumor microvessels themselves. Because of these EPCs into neovessels [28]. findings, the authors postulated a stabilizing and supportive In the early phases of tumor growth, 50–90% of the role of HCs for the developing vascular network [7]. neovessels within the tumor mass are derived from the bone In the present study, intravital fluorescence microscopy marrow dependent on the tumor type [25]. However, recent showed highly vascularized CT26.WT-GFP tumors 14 days reports have already doubted the impact of vasculogenesis after tumor cell implantation. However, immunohistological from bone-marrow-derived cells for tumor neovasculariza- staining using the endothelial cell marker CD31 displayed tion and claim an exclusive role for sprouting angiogenesis positive staining of only a few endothelial cells within these in tumor blood vessel development [15, 16]. In a model of tumors. This finding reflects the diverging structure between syngeneic bone marrow transplantation, Patil et al. demon- normal microvessels and tumor neovessels which consist of strated that GPF-expressing c-Kit bone-marrow-derived only cancer cells or a mosaic of cancer and endothelial cells progenitor cells are recruited to subcutaneously implanted [2]. Lewis lung carcinoma but do not directly contribute to Previous experimental studies in mice have shown that a microvascular structure. They, therefore, concluded that depletion of the myeloid and erythroid cell lineages including even if circulating HCs and EPCs home to sites of tumor EPCs and HCs from the bone marrow could be performed Muscular infiltration (tumor %) Journal of Oncology 9 (a) (b) Control cKit-Ab cKit/SDF1-Ab (c) Figure 5: PCNA immunohistochemistry in CT26.WT-GFP tumors at day 14 after tumor cell implantation. Animals were pretreated with anti-c-Kit (cKit-Ab, (a)) alone or anti-c-Kit and anti-SDF-1 neutralizing antibodies (cKit/SDF1-Ab, (b)). Animals treated with an isotype-matched control antibody served as controls (control). Quantitative analysis demonstrated that anti-c-Kit pretreatment significantly increased the rate of proliferating tumor cells compared to controls (c). Of interest, additional neutralization of SDF-1 significantly decreased the amount of proliferating tumor cells compared to controls and anti-c-Kit-treated animals. Mean ± SEM; P< 0.05 versus control; P< 0.05 versus anti-c-Kit. Original magnification (a, b) ×175. by daily injection of 1 mg/kg ACK2, an antagonistic anti- of tumor vessels were slightly reduced in these studies of c-Kit antibody, which blocks the function of c-Kit without Okamoto et al. during the first 5–7 days, tumor growth cytotoxic side effects on c-Kit cells [30, 31]. Of interest, was rapidly reinitiated as the number of circulating HCs whereas no polymorphonuclear cells or erythroblasts were in the peripheral blood increased again [7]. Therefore, present in the bone marrow of these ACK2-treated animals, the authors concluded that HCs migrating into the tumor B lineage cells continued to grow to fill the space from which mass promote the initiation of tumor neovascularization. myeloid and erythroid progenitor cells were purged [30, 31]. In our present study, although neoangiogenesis was not Moreover, the reduction rate of the colony forming cells influenced after administration of anti-c-Kit, tumor growth within the bone marrow was dependent on the amount of was stimulated as a result of an increased tumor cell prolif- anti-c-Kit antibodies [30]. In the present study, we, therefore, eration and invasion. This observation might be the result used the anti-c-Kit antibody in a dosage of 1 mg/kg BW of immunological mechanisms that suppress tumor growth for the induction of bone marrow suppression over a time in animals with an intact immune system. C-Kit blockade period of 4 days to avoid surgical complications and death for the induction of bone marrow suppression concurrently of the animals resulting from the immune incompetence by stimulates B lymphopoiesis [30, 31]. As the number of B bone marrow depletion. cell precursors normally decreases in tumor bearing mice As shown by Okamoto et al., bone marrow suppression [32, 33], it must be speculated that in our experiment of bone by anti-c-Kit pretreatment over a time period of 4 days before marrow suppression, c-Kit blockade stimulates B cell genesis subcutaneous implantation of colon tumor cells induced and thus increases tumor cell engraftment. Furthermore, B leucopenia which was still detectable 10 days after the lymphopoiesis is associated with increased angiogenesis and last injection [7]. Although tumor growth and sprouting cellular proliferation [34–36]. In our study, neoangiogenesis PCNA-positive tumor cells (%) 10 Journal of Oncology VEGF- precursor β-actin CXCR4 β-actin VEGF Control cKit-Ab cKit/SDF1-Ab Control cKit-Ab cKit/SDF1-Ab 7 7 5 5 3 3 2 2 0 0 Control cKit-Ab cKit/SDF1-Ab Control cKit-Ab cKit/SDF1-Ab (a) (b) Figure 6: Western blot analysis of VEGF (a) and CXCR4 (b) expression within CT 26.WT tumors five days after tumor cell implantation. Animals were pretreated with anti-c-Kit (cKit-Ab) alone or additionally treated with a SDF-1 neutralizing antibody (cKit/SDF1-Ab). Animals treated with isotype-matched control antibodies served as controls (control). Quantitative analysis showed that VEGF expression was significantly higher after anti-c-Kit pretreatment than in control animals (a). Additional neutralization of SDF-1 had no further effect on the increased VEGF expression. CXCR4 was also significantly higher expressed after anti-c-Kit pretreatment compared to controls and was not further influenced by additional SDF-1 neutralization (b). Mean ± SEM; P< 0.05 versus control. within the colorectal tumors of animals pretreated with significantly improved the long-term survival of MCA26 anti-c-Kit was comparable to controls despite the lack tumor bearing mice [41]. In our study, only 8% of the of functional HCs which are necessary for angiogenesis CT26.WT-GFP cells were c-Kit receptor positive. Therefore, and vasculogenesis. Thus, we hypothesize that the lack of we do not expect a direct inhibitory effect of anti-c-Kit HCs was compensated by local angiogenic factors or B treatment on tumor cell proliferation. lymphopoiesis-associated angiogenesis, resulting in a tumor Recruitment of HCs and EPC is predominantly mediated neovascularization comparable to that observed in controls. by SDF-1 and its receptor CXCR4 [9, 10] because HCs and The SCF receptor c-Kit has been shown to be essential EPCs migrate along a chemotactic gradient towards higher for the development of blood cells, melanocytes, germ cells, concentrations of SDF-1 [12–14]. Kaminski et al. have shown interstitial cells of Cajal in the gastrointestinal tract, and that almost 100% of bone marrow and peripheral blood c- mast cells [37]. Furthermore, the majority of tumor cells, Kit cells are positive for CXCR4 [12]. In the present study, especially those of the neural axis, breast, lung, prostate, we could demonstrate that neutralization of SDF-1 after anti- and colon show an aberrant c-Kit expression [38]. Especially c-Kit pretreatment significantly reduces neovascularization gastrointestinal stromal tumors (GISTs) express c-Kit on the most probably by the inhibition of HC and EPC recruitment cell-surface, and mutations of Kit in these tumors results in within the tumors. Additionally, in combination with anti-c- an activation of Kit signaling, which leads to uncontrolled Kit pretreatment, SDF-1 neutralization is capable of decreas- cell proliferation and resistance to apoptosis [39, 40]. Today, ing tumor cell proliferation and invasion. Taken together, patients with GIST are treated with Imatinib, an inhibitor of neutralization of SDF-1 counteracts the stimulating effects certain protein tyrosine kinases including KIT, depending on of bone marrow suppression by anti-c-Kit treatment on the mitotic index of the GIST. Imatinib induces an arrest of tumor cell engraftment and inhibits compensatory local and tumor cell proliferation and causes apoptotic cell death. In B lymphopoiesis-associated angiogenesis. established MCA26 tumors, Pan et al. showed that injection As described by Kaminski et al., the presence of SDF- of anti-c-Kit antibodies markedly reduces tumor-induced 1 chemoattractant activity and inflammatory endothelial immune tolerance exhibited by myeloid-derived suppressors activation by TNF-α is required for c-Kit cells to form in mice [41]. Furthermore, their experiments demonstrate functionally relevant interactions with the endothelium in that anti-c-Kit treatment can prevent tumor-specific T-cell postcapillary venules [12]. Blockade of ICAM-1 and CXCR4 anergy and development of T regulatory cells (Treg). In abolishes adhesion of c-Kit cells to the vascular endothe- combination with an immune modulatory therapy of IL-12 lium despite application of SDF-1 and TNF-α.Moreover, plus 4-1BB activation, treatment with anti-c-Kit antibodies in their cremaster muscle microcirculation model, stem cell VEGF expression (OD/mm tumor) CXCR4 expression (OD/mm tumor) Journal of Oncology 11 adhesion was significantly reduced when eNOS was not [12] A. Kaminski, N. Ma, P. Donndorf et al., “Endothelial NOS is required for SDF-1alpha/CXCR4-mediated peripheral present or systemic NOS inhibited [12]. As SDF-1 and TNF- endothelial adhesion of c-kit+ bone marrow stem cells,” α stimulation activates the endothelium by an increase of the Laboratory Investigation, vol. 88, no. 1, pp. 58–69, 2008. CXCR4 expression leading to relevant stem cell attraction [13] J. A. Burger and T. J. Kipps, “CXCR4: a key receptor in the [12], our results indicate that the increase of CXCR4 and crosstalk between tumor cells and their microenvironment,” VEGF expression within the tumors represents a compen- Blood, vol. 107, no. 5, pp. 1761–1767, 2006. satory pathway after anti-c-Kit pretreatment. Neutralization [14] T. Papayannopoulou, “Current mechanistic scenarios in of SDF-1 inhibits interactions of c-Kit positive cells with hematopoietic stem/progenitor cell mobilization,” Blood, vol. tumor vessels and as a consequence, leads to inhibition of 103, no. 5, pp. 1580–1585, 2004. tumor neovascularization. [15] M. De Palma, M. A. Venneri, C. Roca, and L. Naldini, “Target- In conclusion, bone marrow suppression by anti-c-Kit ing exogenous genes to tumor angiogenesis by transplantation pretreatment significantly enhances tumor cell engraftment of genetically modified hematopoietic stem cells,” Nature of colorectal tumors. As anti-SDF-1 treatment counteracts Medicine, vol. 9, no. 6, pp. 789–795, 2003. this increased tumor outgrowth by inhibition of neovascu- [16] J. R. Gother ¨ t, S. E. Gustin, J. A. M. van Eekelen et al., larization, the SDF-1/CXCR4 pathway seems to be crucial for “Genetically tagging endothelial cells in vivo: bone marrow- tumor angiogenesis mediated by HCs and EPCs. derived cells do not contribute to tumor endothelium,” Blood, vol. 104, no. 6, pp. 1769–1777, 2004. [17] S. Purhonen, J. Palm, D. Rossi et al., “Bone marrow- Acknowledgment derived circulating endothelial precursors do not contribute to vascular endothelium and are not needed for tumor growth,” The authors appreciate the excellent technical assistance of Proceedings of the National Academy of Sciences of the United Janine Becker and Ruth Nickels. States of America, vol. 105, no. 18, pp. 6620–6625, 2008. [18] O. Kollmar, M. K. Schilling, and M. D. Menger, “Experimental liver metastasis: standards for local cell implantation to study References isolated tumor growth in mice,” Clinical and Experimental Metastasis, vol. 21, no. 5, pp. 453–460, 2004. [1] G. Bergers and L. E. Benjamin, “Tumorigenesis and the [19] M. D. Menger, M. W. Laschke, and B. Vollmar, “Viewing angiogenic switch,” Nature Reviews Cancer,vol. 3, no.6,pp. the microcirculation through the window: some twenty 401–410, 2003. years experience with the hamster dorsal skinfold chamber,” [2] P. Carmeliet and R. K. Jain, “Angiogenesis in cancer and other European Surgical Research, vol. 34, no. 1-2, pp. 83–91, 2002. diseases,” Nature, vol. 407, no. 6801, pp. 249–257, 2000. [20] O. Kollmar, K. Rupertus, C. Scheuery et al., “Stromal cell- [3] M.A.GimbroneJr.,S.B.Leapman, R. S. Cotran,and derived factor-1 promotes cell migration and tumor growth of J. Folkman, “Tumor dormancy in vivo by prevention of colorectal metastasis,” Neoplasia, vol. 9, no. 10, pp. 862–870, neovascularization,” Journal of Experimental Medicine, vol. 136, no. 2, pp. 261–276, 1972. [21] O. Kollmar, B. Junker, K. Rupertus, M. D. Menger, and M. [4] R. S. Kerbel, “Tumor angiogenesis: past, present and the near K. Schilling, “Studies on MIP-2 and CXCR2 expression in a future,” Carcinogenesis, vol. 21, no. 3, pp. 505–515, 2000. mouse model of extrahepatic colorectal metastasis,” European [5] K. J. Luzzi, I. C. MacDonald, E. E. Schmidt et al., “Multistep Journal of Surgical Oncology, vol. 33, no. 6, pp. 803–811, 2007. nature of metastatic inefficiency,” American Journal of Pathol- [22] K. Rupertus,O.Kollmar,C.Scheuer,B.Junker,M.D.Menger, ogy, vol. 153, no. 3, pp. 865–873, 1998. and M. K. Schilling, “Major but not minor hepatectomy [6] D. Gao, D. J. Nolan, A. S. Mellick, K. Bambino, K. McDonnell, accelerates engraftment of extrahepatic tumor cells,” Clinical and V. Mittal, “Endothelial progenitor cells control the and Experimental Metastasis, vol. 24, no. 1, pp. 39–48, 2007. angiogenic switch in mouse lung metastasis,” Science, vol. 319, [23] O. Kollmar, B. Junker, K. Rupertus, C. Scheuer, M. D. no. 5860, pp. 195–198, 2008. Menger, and M. K. Schilling, “Liver resection-associated [7] R. Okamoto, M. Ueno, Y. Yamada et al., “Hematopoietic cells macrophage inflammatory protein-2 stimulates engraftment regulate the angiogenic switch during tumorigenesis,” Blood, but not growth of colorectal metastasis at extrahepatic sites,” vol. 105, no. 7, pp. 2757–2763, 2005. Journal of Surgical Research, vol. 145, no. 2, pp. 295–302, 2008. [8] D.C.Rafii, B. Psaila,J.Butler, D. K. Jin, andD.Lyden,“Reg- [24] T. Asahara, H. Masuda, T. Takahashi et al., “Bone marrow ulation of vasculogenesis by platelet-mediated recruitment of origin of endothelial progenitor cells responsible for postnatal bone marrow-derived cells,” Arteriosclerosis, Thrombosis, and vasculogenesis in physiological and pathological neovascular- Vascular Biology, vol. 28, no. 2, pp. 217–222, 2008. ization,” Circulation Research, vol. 85, no. 3, pp. 221–228, 1999. [9] K. Hattori, B. Heissig, K. Tashiro et al., “Plasma elevation of [25] D. Lyden, K. Hattori, S. Dias et al., “Impaired recruitment stromal cell-derived factor-1 induces mobilization of mature of bone-marrow-derived endothelial and hematopoietic pre- and immature hematopoietic progenitor and stem cells,” cursor cells blocks tumor angiogenesis and growth,” Nature Blood, vol. 97, no. 11, pp. 3354–3360, 2001. Medicine, vol. 7, no. 11, pp. 1194–1201, 2001. [10] J. Deshane, S. Chen, S. Caballero et al., “Stromal cell-derived [26] A. M. Davidoff,C.Y.C.Ng, P. Brownetal., “Bonemarrow- factor 1 promotes angiogenesis via a heme oxygenase 1- derived cells contribute to tumor neovasculature and, when dependent mechanism,” Journal of Experimental Medicine, vol. modified to express an angiogenesis inhibitor, can restrict 204, no. 3, pp. 605–618, 2007. tumorgrowthinmice,” Clinical Cancer Research, vol. 7, no. 9, pp. 2870–2879, 2001. [11] S. Mendez-F ´ errer, D. Lucas, M. Battista, and P. S. Frenette, “Haematopoietic stem cell release is regulated by circadian [27] D. J. Nolan, A. Ciarrocchi, A. S. Mellick et al., “Bone marrow- oscillations,” Nature, vol. 452, no. 7186, pp. 442–447, 2008. derived endothelial progenitor cells are a major determinant 12 Journal of Oncology of nascent tumor neovascularization,” Genes and Development, vol. 21, no. 12, pp. 1546–1558, 2007. [28] M. Grunewald, I. Avraham, Y. Dor et al., “VEGF-induced adult neovascularization: recruitment, retention, and role of accessory cells,” Cell, vol. 124, no. 1, pp. 175–189, 2006. [29] V. R. S. Patil, E. B. Friedrich, A. E. Wolley, R. E. Gerszten, J. R. Allport, and R. Weissleder, “Bone marrow-derived lin-c- kit+Sca-1+ stem cells do not contribute to vasculogenesis in Lewis lung carcinoma,” Neoplasia, vol. 7, no. 3, pp. 234–240, [30] M. Ogawa, Y. Matsuzaki, S. Nishikawa et al., “Expression and function of c-kit in hemopoietic progenitor cells,” Journal of Experimental Medicine, vol. 174, no. 1, pp. 63–71, 1991. [31] S. A. Rico-Vargas, B. Weiskopf, S. I. Nishikawa, and D. G. Osmond, “c-kit Expression by B cell precursors in mouse bone marrow: stimulation of B cell genesis by in vivo treatment with anti-c-kit antibody,” Journal of Immunology, vol. 152, no. 6, pp. 2845–2852, 1994. [32] J. Richards, B. McNally, X. Fang, M. A. Caligiuri, P. Zheng, and Y. Liu, “Tumor growth decreases NK and B cells as well as common lymphoid progenitor,” PLoS One, vol. 3, no. 9, Article ID e3180, 2008. [33] I. Kirman, E. H. Huang, and R. L. Whelan, “B cell response to tumor antigens is associated with depletion of B progenitors in murine colocarcinoma,” Surgery, vol. 135, no. 3, pp. 313–318, [34] F. Pessler, L. Dai, C. Diaz-Torne et al., “Increased angiogenesis and cellular proliferation as hallmarks of the synovium in chronic septic arthritis,” Arthritis Care and Research, vol. 59, no. 8, pp. 1137–1146, 2008. [35] A. Tzankov, S. Heiss, S. Ebner et al., “Angiogenesis in nodal B cell lymphomas: a high throughput study,” Journal of Clinical Pathology, vol. 60, no. 5, pp. 476–482, 2007. [36] D. Gratzinger, S. Zhao, R. J. Tibshirani et al., “Prognostic significance of VEGF, VEGF receptors, and microvessel density in diffuse large B cell lymphoma treated with anthracycline- based chemotherapy,” Laboratory Investigation,vol. 88, no.1, pp. 38–47, 2008. [37] B. Heissig, Z. Werb, S. Rafii, and K. Hattori, “Role of c-kit/Kit ligand signaling in regulating vasculogenesis,” Thrombosis and Haemostasis, vol. 90, no. 4, pp. 570–576, 2003. [38] A. M. Turner,K.M.Zsebo,F.Martin, F. W. Jacobsen,L.G. Bennett, and V. C. Broudy, “Nonhematopoietic tumor cell lines express stem cell factor and display c- kit receptors,” Blood, vol. 80, no. 2, pp. 374–381, 1992. [39] M. C. Heinrich, K. Owzar, C. L. Corless et al., “Correlation of kinase genotype and clinical outcome in the North American intergroup phase III trial of imatinib mesylate for treatment of advanced gastrointestinal stromal tumor: CALGB 150105 study by cancer and leukemia group B and southwest oncology group,” Journal of Clinical Oncology, vol. 26, no. 33, pp. 5360– 5367, 2008. [40] G. D. Demetri, M. von Mehren, C. D. Blanke et al., “Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors,” The New England Journal of Medicine, vol. 347, no. 7, pp. 472–480, 2002. [41] P. Y. Pan, G. X. Wang, B. Yin et al., “Reversion of immune tolerance in advanced malignancy: modulation of myeloid- derived suppressor cell development by blockade of stem-cell factor function,” Blood, vol. 111, no. 1, pp. 219–228, 2008. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 International Journal of Journal of Immunology Research Endocrinology Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com BioMed PPAR Research Research International Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Research and Treatment Cellular Longevity Neurology Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal

Journal of OncologyHindawi Publishing Corporation

Published: Oct 2, 2011

There are no references for this article.