|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 4 (February 1), 2007: pp. 376-383 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.09.0696 Tumor Origin of Endothelial Cells in Human Neuroblastoma
From the Laboratories of Oncology and Pathology, G. Gaslini Institute, Genova, Italy Address reprint requests to Annalisa Pezzolo, PhD, Laboratorio di Oncologia, Istituto Giannina Gaslini 5, 16147 Genova-Quarto, Italy; e-mail: annalisapezzolo{at}ospedale-gaslini.ge.it
Purpose: Malignant cells are genetically unstable and prone to develop chemotherapy resistance, whereas tumor vasculature is usually of host origin and genetically stable. Tumor endothelial microvessels attract interest as therapeutic targets, but their genetic instability would curtail such approach. Here, we have investigated the tumor origin of endothelial microvessels in human neuroblastoma (NB). Materials and Methods: Paraffin-embedded tissue sections from 10 MYCN-amplified tumors (six stage 4, three stage 3, and one stage 1) were studied. Endothelial cells (ECs) were detected by immunofluorescent staining for CD31 or CD105, and MYCN amplification was detected using fluorescence in situ hybridization (FISH). In xenografts of the HTLA-230 human NB cell line, human ECs were detected by CD31 staining, mouse ECs were detected by CD34 staining, and MYCN amplification and murine DNA were detected using FISH. Results: MYCN-amplified ECs formed approximately 70% of tumor endothelial microvessels in two stage 4 tumors and 20% in one stage 3 tumor. Similar results were obtained after EC labeling with CD31 or CD105. Staining for alpha-smooth muscle actin in combination with MYCN FISH demonstrated that tumor-derived ECs were coated with pericytes. These vessels were functional because they contained RBCs. Approximately 70% of endothelial vessels from HTLA-230 xenografts stained for human CD31, but not murine CD34, and displayed MYCN amplification, thus proving their tumor origin. Conclusion: NB-associated endothelial microvessels can originate from tumor cells, and this finding challenges the tenet that tumor vasculature is genetically stable. The possibility that NB-derived ECs are chemotherapy resistant warrants further investigation.
Neuroblastoma (NB), a tumor originating from the sympathetic nervous system, is the most common extracranial malignancy in childhood.1 Although stage 1 or 2 tumors are localized and well differentiated and can usually be cured by surgical resection only, patients with stage 3 or 4 tumors present with regional or disseminated disease, respectively, that is often characterized by relapse after response to conventional treatments and poor prognosis.1 The growth of solid tumors is strictly dependent on the development of an adequate blood supply through neoangiogenesis.2,3 Numerous studies have demonstrated that the degree of tumor angiogenesis has prognostic impact in different malignancies.4 In NB, high tumor vascularity was found to correlate with MYCN amplification, advanced-stage disease, and poor outcome. Conversely, excellent survival was observed in the subset of patients who had tumors with low vascular density.5,6 These observations suggest that agents capable of interfering with the process of angiogenesis may inhibit NB growth7-11 and prove to provide effective treatment for patients with clinically aggressive disease.12 Tumors induce the sprouting of new blood microvessels from pre-existing capillaries in a process similar to physiologic angiogenesis.13 Moreover, tumor cells from some aggressive cancers can mimic the activities of endothelial cells (ECs) by forming vascular-like networks.14-18 Evidence for the direct involvement of tumor cells in tumor vessel formation has been provided in human xenograft models.19-21 An important concept in tumor angiogenesis is that tumor blood vessels contain genetically normal and stable ECs unlike tumors cells, which typically display genetic instability. The aforementioned studies challenge this concept because tumor-derived ECs share the same genetic abnormalities with the malignant cells from which they originate, including increased resistance to apoptosis and chemotherapy resistance.22,23 In view of the clinical relevance of this topic, we have investigated the origin of the microvascular ECs in 10 tumor samples of MYCN-amplified NB using a fluorescence-based technique that makes use of simultaneous interphase cytogenetics and immunophenotyping.24 In three tumors, a varying proportion of the microvascular ECs exhibited MYCN amplification, thus proving their tumor origin. Subsequent in vivo studies were performed by injecting the MYCN-amplified HTLA-230 NB cell line in NOD/scid mice and investigating the origin of tumor-associated endothelium.
Tissues Formalin-fixed, paraffin-embedded tissue sections (2 to 4 µm) from 10 NB tumors carrying MYCN amplification (one stage 1, three stage 3, and six stage 4 tumors according to the International Neuroblastoma Staging System classification25; Table 1) and from HTLA tumor xenografts (see HTLA-230 Tumor Xenografts in NOD/scid Mice) were incubated at 60°C overnight and deparaffinized twice using xylene. The slides were subsequently hydrated in a series of ethanol solution and washed with phosphate-buffered saline. Studies with human tissues were performed after approval by a local human investigations committee.
Antibodies Mouse antihuman CD31 (Clone JC70A; Dako Cytomation, Hamburg, Germany), antihuman endoglin (CD105; Dako), antihuman CD45 (Clone PD7/26 e 2B11; Dako), antihuman alpha-smooth muscle actin (  -SMA; Clone 1A4; Dako), and antihuman vascular endothelial growth factor receptor 2 (VEGF-R2; R&D Systems, Minneapolis, MN) were used. A rat antimouse CD34 (Clone MEC14.7; Dako) was also used. R-phycoerythrin –conjugated rabbit antimouse immunoglobulin (Dako) and fluorescein isothiocyanate–conjugated goat antirat immunoglobulin (Southern Biotechnology Inc, Birmingham, AL) were used as second reagents.
Probes
Immunofluorescence and Fluorescent In Situ Hybridization
HTLA-230 Tumor Xenografts in NOD/scid Mice
Cytogenetic Analysis
Tumor Endothelial Vessels Can Originate From Primary NB Cells To identify microvascular ECs on paraffin sections from primary NB tumors, we selected CD31, CD105,27,28 and CD309 (anti–VEGF-R2)29 mAbs. Amplification of the MYCN oncogene, which was present in at least 50% of tumor cells in all specimens studied (Table 1), was chosen as the tumor-specific genetic marker. As shown in Figures 1A and 1B, both CD31 and CD105 mAbs stained ECs but not tumor cells. In contrast, a large proportion of cells present in the sections reacted with the CD309 mAb (Fig 1C). Because the latter observation suggested that tumor cells expressed VEGF-R2, tissue sections were stained with CD309 mAb and then subjected to MYCN FISH. Indeed, as shown in Figure 1D, most MYCN-amplified NB cells stained for CD309, providing the first demonstration that VEGF-R2 is expressed at the protein level by primary NB cells. Because of the lack of specificity for EC, the CD309 mAb was omitted from further studies.
ECs carrying MYCN amplification were identified in three of 10 NB tumors (Table 1, samples 2, 3, and 5). The results listed in Table 1 refer to ECs staining with CD31 mAb, but superimposable results were obtained after staining with CD105 mAb (data not shown). Tumor-derived ECs formed 70% and 78% of blood microvessels in two stage 4 tumors (Table 1, samples 2 and 3) and 20% in one stage 3 tumor (Table 1, sample 5). In these cases, the following two types of green staining (CD31+ or CD105+) ECs were detected: cells with multiple red hybridization signals indicative of MYCN amplification as double minutes (Figs 2A and 2B) and cells with two red signals and hence MYCN gene single copy (Fig 2C, arrows). Thus, the former ECs were of tumor origin because they showed the same genetic abnormality as malignant cells, whereas the latter ECs were derived from normal progenitors.
Endothelial microvessels formed by tumor cells displayed an open lumen and consistently contained RBCs, indicating that these vessels were functional. A representative experiment is shown in Figure 2B, where an endothelial microvessel lined by MYCN-amplified CD105+ ECs contains three clearly visible erythrocytes.
Pericytes are fibroblastic/smooth muscle–like cells that establish close contact with ECs in small blood vessels and capillaries, serving as regulators of their development and function.29 Therefore, in subsequent experiments, we investigated pericyte coating of endothelial microvessels formed by NB cells using immunofluorescence with an anti–
Next, we addressed the question of whether pericytes were of tumor or host origin using immunofluorescence with an anti– The possibility that the reported observations with primary tumor samples represented artifacts as a result of overlapping of tumor cells and ECs is contradicted by the following: only thin tissue sections were used throughout this study, and focusing during microscopic examination allowed for selection of fields in which tumor cells and ECs were consistently on the same plane; the present results were obtained on analysis of both cross and longitudinal tumor sections; and when serial sections of a microscopic area where MYCN-amplified ECs had been detected were stained with a CD45 mAb (panhematopoietic cell marker), the scanty CD45+ infiltrating lymphoid cells always displayed MYCN single-copy gene (data not shown).
Human NB Xenograft Contains Tumor-Derived Endothelial Microvessels Preliminarily to these experiments, we performed FISH analysis of in vitro cultured HTLA-230 cells using the MYCN probe to identify the hybridization pattern characteristic of these cells (ie, double minutes or homogeneously staining regions [HSR]). An HSR on the short arm of chromosome 4 was detected in HTLA-230 cells by FISH (Figs 3A and 3B) and classical cytogenetic analysis (Figs 3C and 3D, white arrows). In addition, the latter assay demonstrated a derivative chromosome der(11)t(11;Y) (yellow arrow), the balanced translocation t(1;17)(p36;q21) (red arrows), and dup(11p) (blue arrows).
In accordance with these results, tumors formed by HTLA-230 cells in NOD/scid mice displayed MYCN amplification in the form of HSR. Thus, the FISH pattern was characterized by clustering of red hybridization signals representing the HSR region (Fig 4A, long arrow) and a normal signal corresponding to the nonamplified allele (Fig 4A, short arrow).
The following two distinct populations of ECs contributing to tumor vasculature formation were detected: green staining human CD31+ ECs, which showed the same pattern of MYCN amplification as the HTLA-230 cell line (Fig 4A, arrowheads); and green staining murine CD34+ cells, which showed several red signals generated by hybridization of the mouse Cot-1 probe to repetitive sequences in murine DNA (Figs 4B and 4C, arrows). Mouse Cot-1 probe never hybridized to human tumor DNA (Figs 4B and 4C). In these experiments, mouse Cot-1 probe was preferred to the pancentromeric mouse probe because of the better performance of the former probe. Control experiments showed the following. First, green staining CD34+ murine ECs did not hybridize to the MYCN probe. Thus, in Figure 4D, MYCN amplification is observed in CD34– HTLA-230 cells but not in CD34+ murine ECs. Second, green staining CD31+ human ECs did not hybridize to mouse Cot-1 DNA (Fig 4E). The inset in Figure 4E shows a CD31– murine non-EC hybridizing to mouse Cot-1 DNA, together with a human tumor cell not reactive with the same probe. Double staining with human CD31 (green) and murine CD34 (red) mAbs demonstrated that, in three different experiments, CD31+ human ECs represented approximately 80% of HTLA-230 xenograft microvessels, with CD34+ murine ECs accounting for the residual 20%. One representative experiment is shown in Figure 4F. No chimeric human-murine endothelial microvessels were detected, suggesting that differentiation of human and murine ECs occurred through unrelated pathways. Taken together, these results demonstrated unambiguously that the majority of tumor-associated endothelial vessels in the HTLA-230 xenograft model originated directly from human malignant cells.
This study shows for the first time that a subset of human MYCN-amplified NB tumors contains endothelial microvessels formed by the malignant cells themselves, as demonstrated by detection of MYCN amplification in CD31+ or CD105+ ECs. In three of 10 NB tumors investigated, the proportion of tumor-derived microvessels ranged from 20% to 78%. A detailed characterization of such vessels showed that they were completely covered with pericytes, similarly to endothelial microvessels bearing single-copy MYCN gene; that pericytes were of host and not tumor origin because they never showed MYCN amplification; and that endothelial microvessels lined by MYCN-amplified ECs displayed open lumen and contained erythrocytes, indicating that they were functional. VEGF is a specific EC growth factor that stimulates angiogenesis and can directly promote proliferation and survival of tumor cells.29 The functional activities of VEGF are mediated by three different receptors. VEGF-R1 is required for the recruitment of hematopoietic precursors and migration of monocytes and macrophages, whereas VEGF-R2 and VEGF-R3 are essential for the functions of vascular ECs and lymphoendothelial cells, respectively.29 In this study, we tested VEGF-R2/CD309 as human vascular EC marker in addition to CD31 and CD105. In contrast to the latter markers that were expressed selectively by ECs, VEGF-R2/CD309 was detected also on most NB cells. This finding provides the first demonstration that the VEGF-R2/CD309 protein is expressed in MYCN-amplified primary NB and supports previous studies showing that such a receptor is involved in paracrine/autocrine regulation of NB cell line growth.10,30 Consistent with the results with primary NB, the MYCN-amplified HTLA-230 NB cell line formed in NOD/scid mice tumors and contained a major proportion of human, tumor-derived endothelial vessels. In this study, the latter cells were detected by expression of human CD31 in combination with MYCN amplification, but we have recently obtained similar results on injection of the MYCN-nonamplified ACN NB cell line in NOD/scid mice.31 MYCN amplification has been associated with enhanced angiogenic activity of primary NB tumors.32 Because in this study only MYCN-amplified NBs were tested, the possibility that formation of tumor-derived ECs is a distinctive feature of this NB subset cannot be ruled out. Nonetheless, the aforementioned results with NB cell line xenografts are in support of the concept that the ability of tumor cells to generate tumor vessels is independent of MYCN amplification. Tumor cells can give rise to tumor-associated endothelial microvessels in malignancies other than NB, such as human B-cell lymphomas, multiple myeloma, and malignant melanoma.14-18 In the case of tumors of hematopoietic origin, it has been postulated that a common progenitor targeted by neoplastic transformation can differentiate in tumor cells or ECs sharing the same genetic abnormalities.15 This fascinating hypothesis is not applicable to solid tumors of nonhematopoietic origin. An alternative hypothesis deals with formation of hybrid vessels by fusion of malignant elements and normal ECs.15 The results of our experiments in the HTLA-230 xenograft model mitigate this hypothesis because murine ECs hybridizing to the MYCN probe or human ECs hybridizing to the mouse Cot-1 DNA probe were never detected. Another hypothesis entails uptake of apoptotic tumor cell bodies by normal ECs as the potential mechanism for formation of apparently tumor-derived endothelial microvessels.15 This hypothesis seems unlikely because the clear visualization of MYCN amplification in tumor-derived ECs obtained in our study would have been conceivably hampered by apoptotic DNA degradation. A final possibility is that neoplastic cells or tumor stem cells receive signals in the tumor microenvironment triggering their transdifferentiation into ECs. This phenomenon would occur in some but not all NB tumors through mechanisms that are as yet unknown and warrant further investigation. Pericytes are solitary cells with characteristic morphology (Fig 2F) embedded in the basement membrane of microvessels.33 Pericytes extend along and wrap the endothelial tube through long cytoplasmic processes that allow establishment of close interactions with ECs. A major function of pericytes is the stabilization of endothelial microvessels.33 Endothelial microvessels formed by NB cells were covered with a layer of host-derived pericytes, as demonstrated by the consistent lack of MYCN amplification in the latter cells. These findings allow us to propose a model summarizing the sequence of events involved in tumor-derived endothelial microvessel formation, which is depicted in Figure 5.
Whatever the mechanism underlying their formation, MYCN-amplified ECs recruit pericytes with single-copy MYCN gene along growing vessels, likely through release of platelet-derived growth factor beta.33 Pericytes stabilize these endothelial microvessels that are chimeric in the sense that their EC component is of tumor origin, whereas their pericyte component is of host origin. Human breast tumor–derived ECs exhibited increased resistance to vincristine and doxorubicin–induced apoptosis compared with normal micro-ECs,23 suggesting by analogy that NB-derived ECs may be chemotherapy resistant. This hypothesis, which is being addressed in ongoing studies, points to the need for a careful characterization of tumor-associated endothelium in high-risk NB patients. Finally, although the number of NB tumors investigated was too limited to allow any statistical analysis, our results raise the question of whether the presence of tumor-derived microvessels represents an unfavorable prognostic parameter. A final answer will come from the investigation of a large series of MYCN-amplified NB tumors.
The authors indicated no potential conflicts of interest.
Conception and design: Annalisa Pezzolo, Vito Pistoia Financial support: Vito Pistoia Provision of study materials or patients: Maria Valeria Corrias, Roberta Cinti, Claudio Gambini Collection and assembly of data: Annalisa Pezzolo, Federica Parodi, Maria Valeria Corrias, Roberta Cinti Data analysis and interpretation: Annalisa Pezzolo, Federica Parodi, Maria Valeria Corrias, Roberta Cinti, Vito Pistoia Manuscript writing: Annalisa Pezzolo, Vito Pistoia Final approval of manuscript: Annalisa Pezzolo, Federica Parodi, Maria Valeria Corrias, Roberta Cinti, Claudio Gambini, Vito Pistoia
We thank John Maris, Children's Hospital, Philadelphia, PA; Kate Matthay, University of California San Francisco, San Francisco, CA; and Bob Seeger, University of California Los Angeles, Los Angeles, CA for critically reviewing the manuscript and providing criticism and helpful suggestions. We also thank Chiara Bernardini for excellent secretarial assistance.
Supported by grants from the Italian Ministry of Health and Compagnia di San Paolo, Torino, Italy. Presented in part at the Advances in Neuroblastoma Research Meeting, May 17-20, 2006, Los Angeles, CA. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Brodeur GM: Neuroblastoma: Biological insights into a clinical enigma. Nat Rev Cancer 3:203-216, 2003[CrossRef][Medline] 2. Folkman J: What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 82:4-6, 1990 3. Folkman J: Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1:27-31, 1995[CrossRef][Medline] 4. Weidner N, Folkman J, Pozza F: Tumor angiogenesis: A new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 84:1875-1887, 1992 5. Meitar D, Crawford SE, Rademaker AW, et al: Tumor angiogenesis correlates with metastatic disease, MYCN amplification, and poor outcome in human neuroblastoma. J Clin Oncol 14:405-414, 1996 6. Eggert A, Ikegaki N, Kwiatkowski J, et al: High-level expression of angiogenic factors is associated with advanced tumor stage in human neuroblastomas. Clin Cancer Res 6:1900-1908, 2000 7. Kim ES, Serur A, Huang J, et al: Potent VEGF blockade causes regression of coopted vessels in a model of neuroblastoma. Proc Natl Acad Sci U S A 99:11399-11404, 2002 8. Pastorino F, Brignole C, Marimpietri D, et al: Vascular damage and anti-angiogenic effects of tumor vessel-targeted liposomal chemotherapy. Cancer Res 63:7400-7409, 2003 9. Shusterman S, Grupp SA, Barr R, et al: The angiogenesis inhibitor tup-470 effectively inhibits human neuroblastoma xenograft growth, especially in the setting of subclinical disease. Clin Cancer Res 7:977-984, 2001 10. Beppu K, Nakamura K, Linehan WM, et al: Topotecan blocks hypoxia-inducible factor-1-alpha and vascular endothelial growth factor expression induced by insulin-like growth factor-I in neuroblastoma cells. Cancer Res 65:4775-4781, 2005 11. Marimpietri D, Nico B, Vacca A, et al: Synergistic inhibition of human neuroblastoma-related angiogenesis by vinblastine and rapamycin. Oncogene 24:6785-6795, 2005[CrossRef][Medline] 12. Shusterman S, Maris JM: Prospects for therapeutic inhibition of neuroblastoma angiogenesis. Cancer Lett 228:171-179, 2005[CrossRef][Medline] 13. Carmeliet P: Mechanisms of angiogenesis and arteriogenesis. Nat Med 6:389-395, 2000[CrossRef][Medline] 14. Maniotis AJ, Folberg R, Hess A, et al: Vascular channel formation by human melanoma cells in vivo and in vitro: Vasculogenic mimicry. Am J Pathol 155:739-752, 1999 15. Streubel B, Chott A, Huber D, et al: Lymphoma-specific genetic aberrations in microvascular endothelial cells in B-cell lymphomas. N Engl J Med 351:250-259, 2004 16. Gunsilius E, Duba HC, Petzer AL, et al: Evidence from a leukaemia model for maintenance of vascular endothelium by bone-marrow-derived endothelial cells. Lancet 355:1688-1691, 2000[CrossRef][Medline] 17. Rigolin GM, Fraulini C, Ciccone M, et al: Neoplastic circulating endothelial cells in multiple myeloma with 13q14 deletion. Blood 107:2531-2535, 2006 18. Fonsato V, Buttiglieri S, Deregibus MC, et al: Expression of Pax2 in human renal tumor-derived endothelial cells sustains apoptosis resistance and angiogenesis. Am J Pathol 168:706-713, 2006 19. Hammersen F, Endrich B, Messmer K: The fine structure of tumor blood vessels: I. Participation of non-endothelial cells in tumor angiogenesis. Int J Microcirc Clin Exp 4:31-43, 1985[Medline] 20. Chang YS, di Tomaso E, McDonald DM, et al: Mosaic blood vessels in tumors: Frequency of cancer cells in contact with flowing blood. Proc Natl Acad Sci U S A 97:14608-14613, 2000 21. di Tomaso E, Capen D, Haskell A, et al: Mosaic tumor vessels: Cellular basis and ultrastructure of focal regions lacking endothelial cell markers. Cancer Res 65:5740-5749, 2005 22. Bussolati B, Deambrosis I, Rosso S, et al: Altered angiogenesis and survival in human tumor-derived endothelial cells. FASEB J 17:1159-1161, 2003 23. Grange C, Bussolati B, Bruno S, et al: Isolation and characterization of human breast tumor-derived endothelial cells. Oncol Rep 15:381-386, 2006[Medline] 24. Martinez-Ramirez A, Cigudosa JC, Maestre L, et al: Simultaneous detection of the immunophenotypic markers and genetic aberrations on routinely processed paraffin section of lymphoma samples by means of the FICTION technique. Leukemia 18:348-353, 2004[CrossRef][Medline] 25. Brodeur GM, Pritchard J, Berthold F, et al: Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment. J Clin Oncol 11:1466-1477, 1993 26. Bogenmann E: A metastatic neuroblastoma model in SCID mice. Int J Cancer 67:379-385, 1996[CrossRef][Medline] 27. Burrows FJ, Derbyshire EJ, Tazzari PL, et al: Up-regulation of endoglin on vascular endothelial cells in human solid tumors: Implications for diagnosis and therapy. Clin Cancer Res 1:1623-1634, 1995[Abstract] 28. Kumar S, Ghellal A, Li C, et al: Breast carcinoma: Vascular density determined using CD105 antibody correlates with tumor prognosis. Cancer Res 59:856-861, 1999 29. Olsson AK, Dimberg A, Kreuger J, et al: VEGF receptor signalling in control vascular function. Nat Rev Mol Cell Biol 7:359-371, 2006[CrossRef][Medline] 30. Das B, Yeger H, Tsuchida R, et al: A hypoxia-driven vascular endothelial growth factor/Flt1 autocrine loop interacts with hypoxia-inducible factor-1alpha through mitogen-activated protein kinase/extracellular signal-regulated kinase [1/2] pathway in neuroblastoma. Cancer Res 65:7267-7275, 2005 31. Ribatti D, Nico B, Pezzolo A, et al: Angiogenesis in human neuroblastoma xenograft model: Mechanism and inhibition by tumor-derived interferon- 32. Ribatti D, Raffaghello L, Pastorino F, et al: In vivo angiogenic activity of neuroblastoma correlates with MYCN oncogene overexpression. Int J Cancer 102:351-354, 2002[CrossRef][Medline] 33. von Tell D, Armulik A, Betsholtz C: Pericytes and vascular stability. Exp Cell Res 312:623-629, 2006[CrossRef][Medline] Submitted September 6, 2006; accepted November 14, 2006.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
. Br J Cancer 94:1845-1852, 2006
