Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Dec 1.
Published in final edited form as: Exp Mol Pathol. 2014 Jan 4;96(2):162–167. doi: 10.1016/j.yexmp.2013.12.011

VHL-deficient vasculogenesis in hemangioblastoma

Sven Gläsker a, Jonathan Smith c, Mark Raffeld d, Jie Li e, Edward H Oldfield b, Alexander O Vortmeyer e,*
PMCID: PMC7706334  NIHMSID: NIHMS1643711  PMID: 24394472

Abstract

Hemangioblasts are capable of differentiation into vascular structures and blood. Patients with von Hippel–Lindau (VHL) disease develop hemangioblastomas which are composed of VHL-deficient tumor cells with protracted hemangioblastic differentiation potential. In a subset of these tumors, hemangioblastic differentiation is characterized by different stages of red blood cell formation. It has remained controversial, however, whether VHL-deficient hemangioblastic cells are similarly capable of differentiating into vascular cells and functioning vascular structures in vivo.

By histologic, immunohistologic and microdissection-based genetic analysis of 60 VHL disease-associated hemangioblastomas, we re-examined the controversial question whether VHL-deficient neoplastic hemangioblastic cells are capable of vascular differentiation (vasculogenesis).

In most tumors (n = 47), there was no evidence of either vasculogenesis or hematopoiesis; tumor cells were either scattered between reactive angiogenetic vascular structures or arranged in solid clusters. A subset of tumors (n = 13), however, revealed vaculogenetic structures that were composed of cuboidal or flat cells and frequently contained red blood cell precursors or mature red blood cells. Microdissection-based deletion analysis of epithelial cells confirmed them to be VHL-deficient tumor cells. Immunohistochemistry for CD31 was consistently negative in these structures, and no evidence could be obtained for connectivity with reactive vasculature. We demonstrate that hemangioblastic differentiation capacity of VHL-deficient hemangioblastic cells includes not only erythropoiesis, but also differentiation into primitive vasculogenetic structures. Tumor cells, however, do not appear to have the potential of terminal differentiation into mature and functional vascular structures.

Keywords: von Hippel–Lindau disease, Hemangioblastoma, Vasculogenesis, Angiogenesis

Introduction

Hemangioblastomas are benign highly vascularized tumors which are typically found in the cerebellum, brain stem, spinal cord, and retina. They can occur as sporadic lesions or as component of von Hippel–Lindau disease, an autosomal dominant tumor syndrome caused by germline mutations of the VHL tumor suppressor gene.

Hemangioblastoma tissue is composed of neoplastic “stromal” cells and abundant reactive vascular cells (Lee et al., 1998; Vortmeyer et al., 1997a). The histogenesis of the neoplastic “stromal” cell has remained controversial, and a variety of cell types have been proposed to represent the cytologic equivalent of the “stromal” cell, because the neoplastic cells bear no resemblance with any cell type occurring in regular nervous system tissue (reviewed in (Aldape et al., 2007; Glasker et al., 2006; Lantos et al., 2002)). As originally proposed by Cushing and Lindau (Cushing and Bailey, 1928b; Lindau, 1931), evidence has been accumulating that the VHL-deficient tumor cell represents a cell with hemangioblastic differentiation capacity (Glasker et al., 2006; Shively et al., 2008; Vortmeyer et al., 2003; Vortmeyer et al., 2006). Hemangioblastomas have been further characterized as processes of protracted hemangioblastic differentiation (Shively et al., 2008; Stein et al., 1960; Vortmeyer et al., 2006). Although mechanisms are unclear, it can certainly be assumed that lack of VHL function is the cause of differentiation protraction of VHL-deficient hemangioblastic cells (Shively et al., 2008). Tumor precursor structures as well as small-sized symptomatic tumors never show evidence of blood island formation and hematopoiesis (Shively et al., 2008). In contrast, larger tumors frequently reveal epithelioid VHL-deficient blood islands with erythropoiesis (Shively et al., 2008; Vortmeyer et al., 2003).

It has remained controversial, however, whether hemangioblastic differentiation potential of VHL-deficient tumor cells includes differentiation potential into vascular structures. Numerous previous studies have concluded that neoplastic and vascular cells are separate cytologic constituents of hemangioblastoma (Becker et al., 1989; Böhling et al., 1996; Frank et al., 1989; Holt et al., 1986; Nemes, 1992; Tanimura et al., 1984). Other studies revealed cytologic evidence of vascular differentiation in some tumor cells (Cancilla and Zimmerman, 1965; Chaudhry et al., 1978; Hatva et al., 1996; Ho, 1984; Jellinger and Denk, 1974; Jurco et al., 1982; Kawamura et al., 1973; Lach et al., 1999; Omulecka et al., 1995). An in vitro study of a single tumor was suggestive of angiogenic lineage, but not mature vascular differentiation potential of tumor cells (Spence and Rubinstein, 1975). In contrast, permissive culture of VHL-deficient hemangioblastic cells revealed differentiation into blood and vascular cells (Park et al., 2007). It has therefore remained unknown, whether VHL-deficient tumor cells are capable of differentiating into mature vascular structures in vivo that functionally participate among the abundant reactive vessels known to exist within hemangioblastic tumors (Aldape et al., 2007).

To address this question, we performed a systematic structural analysis of 60 consecutive tumors that had been surgically resected from patients with VHL disease, combined with immunohistochemical and microdissection-based genetic studies. All tumors were histologically surveilled for blood island formation with or without erythropoiesis. Whenever blood island formation was detected, detailed structural, immunohistochemical and genetic studies were performed to pursue possible vasculogenesis.

Materials and methods

Tissue

Sixty consecutive hemangioblastomas of patients with VHL germline mutations, surgically resected from cerebellum, brain stem or spinal cord by the neurosurgical staff of Surgical Neurology Branch/NINDS, were analyzed for this study. Smaller tumors were bisected and entirely submitted and processed into a single paraffin block, larger tumors were serially sectioned and processed into two or more paraffin blocks. From these paraffin blocks consecutive serial sections were taken to allow for comparison of structural elements in different stains. Serial sections from all blocks were stained with H&E and immunostained for CD31 and NSE (see below) prior to careful histologic analysis. Based on morphological results, cases were elected to be stained for HIF1 and HIF2.

Immunohistochemistry

For detection of CD31 antigen, slides were digested with protease I for 20 min, then incubated with anti-DC31 (DAKO, Carpinteria, CA, 1/20) for 32 min and run on Ventana BenchMark XT with Ultra View kit. For detection of NSE, slides were exposed to antigen retrieval buffer (citrate buffer ph 6.0) for 20 min prior to incubation with anti-NSE (DAKO, Carpinteria, CA, 1/2000).

A modified protocol was used for detection of expression of HIF1 and HIF2. For antigen retrieval, sections were immersed in preheated Dako target retrieval solution in a pressure cooker for 90 s. Primary antibodies were: mouse monoclonal anti-human HIF1α H1α67 (Novus Biologicals, Littleton, CO) 1/4000; rabbit polyclonal anti-mouse HIF2α PM8 antiserum (Novus Biologicals, Littleton, CO), 1/100. Primary antibody was omitted for negative controls. Antigen/antibody complexes were revealed by means of the Catalyzed Signal Amplification system (DAKO, Carpinteria, CA, Dako) (HIF) according to the manufacturer’s instructions.

Microdissection and LOH analysis

Five micrometer tissue sections taken from formalin-fixed, paraffin-embedded tissue blocks were placed on glass slides and used for microdissection and LOH analysis. The sections were directly consecutive to those sections investigated by H&E staining. Microdissection was performed under direct light microscopic visualization using a 30 gauge needle as previously described (Lubensky et al., 1996; Vortmeyer et al., 1997b; Vortmeyer et al., 1997c; Zhuang and Vortmeyer, 1998). The endothelium of previously photodocumented vessels was microdissected. Control samples were obtained from adjacent cerebellum or spinal nerve tissue on the same histological slide. Procured cells were immediately resuspended in 10 to 20 μl of buffer containing Tris/HCl, pH 8.0, 10 mmol/l ethylenediamine tetra-acetic acid, pH 8.0, 1% Tween 20, and 0.1 mg/ml proteinase K, and were incubated at 37 °C for three days. The mixture was boiled for 10 min to inactivate proteinase K, and 1.5 μl of this solution was used for the polymerase chain reaction (PCR) amplification of DNA.

The DNA samples were analyzed for LOH by PCR using the microsatellite markers D3S1038 and D3S1110 (Research Genetics, Huntsville, AL), flanking the VHL gene on chromosome 3p25. Each PCR sample contained 1.5 μl of template DNA as noted above, 10 pmol of each primer, 20 nmol each of dATP, dCTP, dGTP, and dTTP, 15 mmol/l MgCl2, 0.1 U of Taq DNA polymerase, 0.05 μl of [32P] dCTP (6000 Ci/mmol), and 1 μl of 10x buffer in a total volume of 10 μl. Labeled amplified DNA was mixed with an equal volume of formamide loading dye (95% formamide, 20 mmol/l EDTA, 0.05% bromophenol blue, 0.05% xylene cyanol) and analyzed on a polyacrylamide gel. The samples were denatured for 5 min at 95 °C and loaded onto a gel consisting of 6% acrylamide (49:1 acrylamide:bis), 5% glycerol, and 0.6x TBE. Samples were electrophoresed at 60 W at room temperature over 2.5 h. Gels were transferred to 3MM Whatman paper and dried, and autoradiography was performed with Kodak X-OMAT film (Eastman Kodak, Rochester, NY).

Results

Most hemangioblastomas show unequivocal separation between neoplastic “stromal” cells and reactive vascular cells

Sixty hemangioblastoma specimens were histologically assessed for evidence of advanced “stromal” cell differentiation using H&E stains and immunohistochemical preparations for NSE and CD31. Hemangioblastic tumor cells were consistently positive for NSE and negative for CD31, while reactive vascular cells were consistently positive for CD31 and negative for NSE. In forty-seven of these cases, NSE + hemangioblastic tumor cells were diffusely scattered and separated by abundant CD31+ vascular cells consistent with mesenchymal architecture (Fig. 1) (Shively et al., 2008; Vortmeyer et al., 2006). Within areas of mesenchymal architecture, clustering of NSE + hemangioblastic cells occurred in some of the cases consistent with epitheloid architecture and consistent with the morphologic sequence of early hemangioblastic differentiation into blood and vessels (Sabin, 1917; Sabin, 1920; Sabin, 2002) which is protracted in VHL-deficient tumor (Shively et al., 2008). While the extent of areas of epitheloid architecture was variable, in all 47 cases there was unequivocal morphologic separation between neoplastic tumor cells and reactive vascular cells. There was no evidence of any transitional morphology that would suggest differentiation capacity from tumor cells into vascular elements (Fig. 1).

Fig. 1. Proposed structural progression of hemangioblastoma.

Fig. 1.

Open in a new tab

Pictures of different areas within a hemangioblastoma show the proposed structural progression of hemangioblastoma from mesenchymal to vasculogenetic architecture and extramedullary erythropoiesis (A–D, immunohistochemistry for CD31; E–G, H&E). A, Mesenchymal architecture: Only few and small hemangioblastic tumor cells (negative for CD31) are diffusely scattered and separated by abundant CD31+ vascular cells.; B, early, and C, advanced epitheloid architecture: Clustering of tumor cells (CD31− negative) with distension of the CD31+ vascular network; D, Further distension of the CD31+ vascular network and formation of vasculogenetic lumina by CD31− tumor cells; (E) Focus of erythropoiesis; F, G, vasculogenetic structures, intraluminal red blood cells are seen.

A subset of hemangioblastomas reveals vasculogenetic structures

Thirteen of sixty cases, however, revealed additional features, predominantly caused by structural changes in hemangioblastic cell clusters. Consistent with their interpretation of neoplastic “blood islands”, some NSE + hemangioblastic tumor cell clusters underwent additional structural changes. One type of structural change occurring in tumor cell clusters was characterized by transformation of hemangioblastic tumor cells into red blood cell precursor elements (Fig. 1). Another type of structural change, however, was characterized by the transformation of solid hemangioblast cell clusters into cystic, possibly vasculogenetic structures (Figs. 1, 2). Smaller cystic spaces were characteristically lined by cuboidal tumor cells resembling neoplastic hemangioblasts (Fig. 1) while larger spaces were lined by flat cells with increasing resemblance to endothelial cells (Fig. 2). “Cystic” spaces were free of cells or contained scattered red blood cells and red blood cell precursors, but no lymphoid or other white blood cells (Figs. 1, 2). In spite of the increasing resemblance to endothelial cells, immunohistochemical analysis revealed cuboidal or flat cyst-lining cells to be consistently immunoreactive with anti-NSE (Fig. 2C, D); immunoreactivity with anti-CD31 was never observed in cyst-lining cells (Fig. 2 E, F). Degeneration of cyst-lining cells could be observed in a few cases, but a transformation or transition into mature CD31+ vascular cells was not observed. Therefore, both vasculogenesis and angiogenesis occurred in this group of tumors; there was no apparent transition or connection between the two vascular networks. We conclude that cystic structures represent an early stage of vasculogenesis and result from differentiation of VHL-deficient hemangioblastic cells; however, immunohistochemistry for CD31 failed to identify cyst-lining cells as mature vascular cells, and structural analysis failed to provide evidence for connectivity with the reactive vascular network.

Fig. 2. Vasculogenetic structures versus angiogenesis.

Fig. 2.

Open in a new tab

A, B, Vasculogenetic structures, lined by cuboidal or flat cells, H&E stains. C,D, Similar to VHL-deficient tumor cells, vasculogenetic structures are uniformly positive for NSE; E,F, in contrast to CD31+ reactive angiogenetic vessels, vasculogenetic structures are negative for CD31; G, Low power and H, high power illustration of strong activation of HIF2 in vasculogenetic structures.

Immunohistochemical and genetic analysis confirms primitive vascular structures – but not mature vessels – to be neoplastic

In VHL disease-associated hemangioblastomas, tumor cells are known to activate HIF1alpha and HIF2alpha as a direct consequence of VHL loss of function (Mandriota et al., 2002; Maxwell et al., 1999). In addition, neoplastic cells are known to have inactivation of the VHL wild-type allele, which can be demonstrated by allelic deletion analysis (Crossey et al., 1994).

Immunohistochemistry for HIF1 and HIF2 confirmed the vasculogenetic structures to be HIF + neoplastic cells (Fig. 2 G, H). To prove that vasculogenetic structures are in fact part of the neoplastic differentiation profile, microdissection was performed on representative structures. Microdissection of vasculogenetic structures and subsequent genetic analysis revealed LOH of the VHL gene locus, while microdissected angiogenetic structures retained the VHL wildtype allele (Fig. 3). CD31 immunoreactivity consistently demonstrated endothelium of reactive vessels. On the other hand, vascular lining cells that exhibited a loss of heterozygosity (LOH) at the VHL locus were consistently immunonegative for the CD31 antigen.

Fig. 3. Vasculogenetic, but not angiogenetic cells reveal LOH of the VHL gene locus.

Fig. 3.

Open in a new tab

The endothelium of previously photodocumented vessels was microdissected. Normal control samples (N) were obtained from adjacent cerebellum or spinal nerve tissue on the same histological slide. On top the H&E stains of an angiogenetic and a vasculogenetic (right, arrows) structure are shown before microdissection. On bottom the results of LOH analysis for the marker D3S1038 at the VHL locus are shown (Lanes 1, 4, 7 and 9 normal control; lanes 2 and 3 angiogenetic cells; lanes 5, 6, 8 and 10 vasculogenetic cells). Angiogenetic cells retain heterozygosity at the VHL locus, whereas vasculogenetic cells reveal loss of heterozygosity demonstrated by loss of the upper allele.

Discussion

The cardiovascular system develops in two steps, vasculogenesis and angiogenesis (Risau, 1997; Risau, 1998). Vasculogenesis is defined as de-novo formation of vascular structures from hemangioblastic precursor cells, while angiogenesis is defined as vascular proliferation from preexistent vascular cells (Risau, 1997; Risau, 1998).

Primitive and early hemangioblastic differentiation has been discovered in the yolk sac of chicken embryos, and the structural sequence of primitive hemangioblastic differentiation has been defined (Sabin, 1917; Sabin, 1920). The key event of this structural sequence is increase of size of progenitor cells and aggregation of these cells into blood islands (Sabin, 1917; Sabin, 1920; Sabin, 2002). Blood island formation is the prerequisite for further differentiation into red blood cells of center cells, and for blood vessel differentiation of peripheral cells (Sabin, 1917; Sabin, 1920; Sabin, 2002). Notably, the center portion of a blood island exclusively differentiates into red, but not white blood cells (Sabin, 2002), while the peripheral portion of the blood island differentiates into vascular structures.

Subsequently, primitive pre-liver hematopoiesis has been discovered in the aorto-gonadal mesonephros of the embryo proper (Medvinsky and Dzierzak, 1996), and early hemangioblastic activity has been found to co-develop with neuroectodermal cells to provide the vascularization of the developing central nervous system and nerve (Risau, 1998). This has been experimentally supported by the detection of hemangioblastic activity in developing nervous system roots, midbrain and hindbrain (Gering et al., 1998; Green et al., 1992; Sinclair et al., 1999). Of interest, this anatomic distribution closely recapitulates the distribution of hemangioblastomas in VHL disease (Glasker et al., 2006).

In VHL disease, VHL-deficient hemangioblast progenitor cells are found multifocally in CNS and nerve roots (Vortmeyer et al., 2004) and cerebellum (Shively et al., 2011), and these VHL-deficient foci have the capacity to proliferate and differentiate into VHL-deficient tumor, hemangioblastoma (Vortmeyer et al., 2006). The progression of VHL-deficient progenitor cells into hemangioblastic tumor cells is associated with increase of cytoplasmic and nuclear size, cytoplasmic lipidization and activation of HIF1 (Vortmeyer et al., 2006).

Although differentiation capacity of VHL-deficient hemangioblasts into red blood cells has been demonstrated in hemangioblastomas (Vortmeyer et al., 2003; Zec et al., 1991), differentiation capacity into vascular cells has remained controversial. In contrast to previous studies that performed detailed studies on small numbers of tumors only, we surveilled 60 consecutively sampled tumors for evidence of vascular differentiation. With this study we obtained evidence for vasculogenetic differentiation potential of tumor cells in 13 of 60 hemangioblastomas by detailed histologic and immunohistologic analysis and by tissue microdissection and genetic analysis.

Thirteen hemangioblastomas with blood island formation revealed features of primitive vasculogenesis with formation of cystic lumina with red blood cells in the center. As evidence for a “stromal” cell derivation of the epithelial cells lining the lumina, the cells stained positive for NSE, but negative for vascular markers. In addition, these cells revealed inactivation of VHL and up-regulation of HIF as a further proof of their “stromal” cell derivation. Cystic structures in these tumors have therefore been identified as primitive vasculogenesis. Previous studies have identified Weibel–Palade bodies – hallmarks of endothelial differentiation – in the “stromal” cells suggesting angiogenic mesenchymal origin of the tumor cells (Ho, 1984; Omulecka et al., 1995). VHL-deficient hemangioblastic cells, however, express neuron specific enolase (Becker et al., 1989), neural cell adhesion molecule (Ishizawa et al., 2005), and frequently GFAP (Cruz-Sanchez et al., 1990; Jurco et al., 1982; Tanimura et al., 1984) which is more suggestive of neuroectodermal origin to others.

Upon our investigation of 13 hemangioblastomas with primitive vasculogenesis, we did not find any hints for a connection of the primitive network to the mature capillary system of the tumor. No transitions or morphologic continuum were seen and there were only red blood cells and no lymphocytes or other white blood cells within the primitive lumina. Instead, cystic structures had a tendency to degenerate. At the same time we were unable to detect any evidence for hemangioblastic differentiation into mature and functional vascular structures. In contrast to the capacity of neoplastic hemangioblasts to differentiate into red blood cells, tumor cells appear only able to differentiate into immature vascular channels and unable to complete differentiation into mature vessels. Vice versa, we investigated numerous mature vascular structures and failed to obtain evidence for a neoplastic nature. Mature vascular structures stained negative for NSE, but positive for vascular markers and therefore appear not to derive from “stromal” cells. In contrast to the “stromal” cell derivates they did also consistently reveal a normal-functioning VHL allele as shown by retained heterozygosity in microdissection-based LOH analyses and a negative immunoreactivity with anti-HIF. We further conclude that these observations provide further support for the hypothesis that hemangioblastomas are processes of protracted hemangioblastic differentiation (Cushing and Bailey, 1928a; Glasker et al., 2006; Lindau, 1931; Park et al., 2007; Shively et al., 2008; Stein et al., 1960; Vortmeyer et al., 2006).

Conclusions

In addition to previous reports of a hematopoietic differentiation capacity of the “stromal” cells in hemangioblastomas we here report that “stromal” cells are also capable to differentiate into primitive vascular structures. Hemangioblastomas are therefore unique neoplasms that may show two types of vascularization. First, they show an intense mature capillary network as result of VEGF-driven reactive angiogenesis. Second, they additionally show vasculogenesis by the neoplastic “stromal” cells in a subset of tumors. In vivo, both processes occur next to each other and there appears to be no transition and no connection between the two vascular networks. We conclude that the complex biology of hemangioblastomas is not only characterized by intense reactive angiogenesis secondary to VHL deficiency, but also by erythropoiesis and vasculogenesis secondary to the differentiation potential of the neoplastic cells.

References

  1. Aldape KD, et al. , 2007. Haemangioblastoma In: Louis DN, et al. (Eds.), WHO Classification of Tumours of the Nervous System. IARC Press, Lyon. [Google Scholar]
  2. Becker I, et al. , 1989. Histogenesis of stromal cells in cerebellar hemangioblastomas. An immunohistochemical study. Am. J. Pathol 134, 271–275. [PMC free article] [PubMed] [Google Scholar]
  3. Böhling T, et al. , 1996. Different expression of adhesion molecules on stromal cells and endothelial cells of capillary hemangioblastoma. Acta Neuropathol. (Berl) 92, 461–466. [DOI] [PubMed] [Google Scholar]
  4. Cancilla PA, Zimmerman HM, 1965. The fine structure of a cerebellar hemangioblastoma. J. Neuropathol. Exp. Neurol 24, 621–628. [DOI] [PubMed] [Google Scholar]
  5. Chaudhry AP, et al. , 1978. Ultrastructure of cerebellar hemangioblastoma. Cancer 42, 1834–1850. [DOI] [PubMed] [Google Scholar]
  6. Crossey PA, et al. , 1994. Molecular genetic investigations of the mechanism of tumourigenesis in von Hippel-Lindau disease: analysis of allele loss in VHL tumours. Hum. Genet 93, 53–58. [DOI] [PubMed] [Google Scholar]
  7. Cruz-Sanchez FF, et al. , 1990. Haemangioblastoma: histological and immunohistological study of an enigmatic cerebellar tumour. Histol. Histopathol 5, 407–413. [PubMed] [Google Scholar]
  8. Cushing H, Bailey P, 1928a. Hemangiomas of cerebellum and retina (Lindau’s disease). Arch. Ophthalmol 57, 447–463. [PMC free article] [PubMed] [Google Scholar]
  9. Cushing H, Bailey P, 1928b. Hemangiomas of cerebellum and retina (Lindau’s disease): with the report of a case. Trans. Am. Ophthalmol. Soc 26, 182–202. [PMC free article] [PubMed] [Google Scholar]
  10. Frank TS, et al. , 1989. A detailed immunohistochemical analysis of cerebellar hemangioblastoma: an undifferentiated mesenchymal tumor. Mod. Pathol 2, 638–651. [PubMed] [Google Scholar]
  11. Gering M, et al. , 1998. The SCL gene specifies haemangioblast development from early mesoderm. EMBO J. 17, 4029–4045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Glasker S, et al. , 2006. Hemangioblastomas share protein expression with embryonal hemangioblast progenitor cell. Cancer Res. 66, 4167–4172. [DOI] [PubMed] [Google Scholar]
  13. Green AR, et al. , 1992. SCL is coexpressed with GATA-1 in hemopoietic cells but is also expressed in developing brain. Oncogene 7, 653–660. [PubMed] [Google Scholar]
  14. Hatva E, et al. , 1996. Vascular growth factors and receptors in capillary hemangioblastomas and hemangiopericytomas. Am. J. Pathol 148, 763–775. [PMC free article] [PubMed] [Google Scholar]
  15. Ho K, 1984. Ultrastructure of cerebellar capillary hemangioblastoma. I. Weibel-Palade bodies and stroma cell histogenesis. J. Neuropathol. Exp. Neurol 43, 592–608. [DOI] [PubMed] [Google Scholar]
  16. Holt SC, et al. , 1986. Capillary hemangioblastoma. An immunohistochemical study. Am. J. Clin. Pathol 86, 423–429. [DOI] [PubMed] [Google Scholar]
  17. Ishizawa K, et al. , 2005. Stromal cells in hemangioblastoma: neuroectodermal differentiation and morphological similarities to ependymoma. Pathol. Int 55, 377–385. [DOI] [PubMed] [Google Scholar]
  18. Jellinger K, Denk H, 1974. Blood group isoantigens in angioblastic meningiomas and hemangioblastomas of the central nervous system. Virchows Arch. A Pathol. Anat. Histol 364, 137–144. [DOI] [PubMed] [Google Scholar]
  19. Jurco S 3rd, et al. , 1982. Hemangioblastomas: histogenesis of the stromal cell studied by immunocytochemistry. Hum. Pathol 13, 13–18. [DOI] [PubMed] [Google Scholar]
  20. Kawamura J, et al. , 1973. Cerebellar hemangioblastoma: histogenesis of stroma cells. Cancer 31, 1528–1540. [DOI] [PubMed] [Google Scholar]
  21. Lach B, et al. , 1999. Angiogenic histogenesis of stromal cells in hemangioblastoma: ultrastructural and immunohistochemical study. Ultrastruct. Pathol 23, 299–310. [DOI] [PubMed] [Google Scholar]
  22. Lantos PL, et al. , 2002. Tumors of the nervous system In: Graham DI, Lantos PL (Eds.), Greenfields Pathology. Arnold, London, pp. 767–1052. [Google Scholar]
  23. Lee J-Y, et al. , 1998. Loss of heterozygosity and somatic mutations of the VHL tumor suppressor gene in sporadic cerebellar hemangioblastomas. Cancer Res. 58, 504–508. [PubMed] [Google Scholar]
  24. Lindau A, 1931. Discussion on vascular tumors of the brain and spinal cord. Proc. R. Soc. Med 24, 363–370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lubensky IA, et al. , 1996. Allelic deletions of the VHL gene detected in multiple microscopic clear cell renal lesions in von Hippel-Lindau disease patients. Am. J. Pathol 149, 2089–2094. [PMC free article] [PubMed] [Google Scholar]
  26. Mandriota SJ, et al. , 2002. HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in the nephron. Cancer Cell 1, 459–468. [DOI] [PubMed] [Google Scholar]
  27. Maxwell PH, et al. , 1999. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis [see comments]. Nature 399, 271–275. [DOI] [PubMed] [Google Scholar]
  28. Medvinsky A, Dzierzak E, 1996. Definitive hematopoiesis is autonomously initiated by the AGM region. Cell 86, 897–906. [DOI] [PubMed] [Google Scholar]
  29. Nemes Z, 1992. Fibrohistiocytic differentiation in capillary hemangioblastoma. Hum. Pathol 23, 805–810. [DOI] [PubMed] [Google Scholar]
  30. Omulecka A, et al. , 1995. Immunohistochemical and ultrastructural studies of stromal cells in hemangioblastoma. Folia Neuropathol. 33, 41–50. [PubMed] [Google Scholar]
  31. Park DM, et al. , 2007. von Hippel-Lindau disease-associated hemangioblastomas are derived from embryologic multipotent cells. PLoS Med. 4, e60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Risau W, 1997. Mechanisms of angiogenesis. Nature 386, 671–674. [DOI] [PubMed] [Google Scholar]
  33. Risau W, 1998. Development and differentiation of endothelium. Kidney Int. Suppl 67, S3–S6. [DOI] [PubMed] [Google Scholar]
  34. Sabin FR, 1917. Preliminary note on the differentiation of angioblasts and the method by which they produce blood-vessels, blood-plasma and red blood-cells as seen in the living chick. Anat. Rec 13, 199–204. [DOI] [PubMed] [Google Scholar]
  35. Sabin F, 1920. Studies on the origin of red blood corpuscles as seen in the living blastoderm of chicks during the second day of incubation. Contrib. Embryol. Carnegie Inst. Wash 9, 214–262. [Google Scholar]
  36. Sabin FR, 2002. Preliminary note on the differentiation of angioblasts and the method by which they produce blood-vessels, blood-plasma and red blood-cells as seen in the living chick. 1917. J. Hematother. Stem Cell Res 11, 5–7. [DOI] [PubMed] [Google Scholar]
  37. Shively SB, et al. , 2008. Protracted haemangioblastic proliferation and differentiation in von Hippel-Lindau disease. J. Pathol 216, 514–520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shively SB, et al. , 2011. Developmentally arrested structures preceding cerebellar tumors in von Hippel-Lindau disease. Mod. Pathol 24, 1023–1030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sinclair AM, et al. , 1999. Distinct 5’ SCL enhancers direct transcription to developing brain, spinal cord, and endothelium: neural expression is mediated by GATA factor binding sites. Dev. Biol 209, 128–142. [DOI] [PubMed] [Google Scholar]
  40. Spence AM, Rubinstein LJ, 1975. Cerebellar capillary hemangioblastoma — its histogenesis studied by organ culture and electron microscopy. Cancer 35, 326–341. [DOI] [PubMed] [Google Scholar]
  41. Stein AA, et al. , 1960. The histogenesis of hemangioblastoma of the brain. J. Neurosurg 17, 751. [DOI] [PubMed] [Google Scholar]
  42. Tanimura A, et al. , 1984. Hemangioblastoma of the central nervous system: nature of the stromal cells as studied by the immunoperoxidase technique. Hum. Pathol 15, 866–869. [DOI] [PubMed] [Google Scholar]
  43. Vortmeyer AO, et al. , 1997a. von Hippel-Lindau gene deletion detected in the stromal cell component of a cerebellar hemangioblastoma associated with von Hippel-Lindau disease. Hum. Pathol 28, 540–543. [DOI] [PubMed] [Google Scholar]
  44. Vortmeyer AO, et al. , 1997b. Von Hippel-Lindau gene deletion detected in the stromal cell component of a cerebellar hemangioblastoma associated with von Hippel-Lindau disease. Hum. Pathol 28, 540–543. [DOI] [PubMed] [Google Scholar]
  45. Vortmeyer AO, et al. , 1997c. Allelic deletion and mutation of the von Hippel-Lindau (VHL) tumor suppressor gene in pancreatic microcystic adenomas. Am. J. Pathol 151, 951–956. [PMC free article] [PubMed] [Google Scholar]
  46. Vortmeyer AO, et al. , 2003. Developmental arrest of angioblastic lineage initiates tumorigenesis in von Hippel-Lindau disease. Cancer Res. 63, 7051–7055. [PubMed] [Google Scholar]
  47. Vortmeyer AO, et al. , 2004. Developmental effects of von Hippel-Lindau gene deficiency. Ann. Neurol 55, 721–728. [DOI] [PubMed] [Google Scholar]
  48. Vortmeyer AO, et al. , 2006. Evolution of VHL tumorigenesis in nerve root tissue. J. Pathol 210, 374–382. [DOI] [PubMed] [Google Scholar]
  49. Zec N, et al. , 1991. Extramedullary hematopoiesis in cerebellar hemangioblastoma. Neurosurgery 29, 34–37. [DOI] [PubMed] [Google Scholar]
  50. Zhuang Z, Vortmeyer AO, 1998. Applications of tissue microdissection in cancer genetics. Cell Vis 5, 43–48. [PubMed] [Google Scholar]

RESOURCES