Protracted haemangioblastic proliferation and differentiation in von Hippel-Lindau disease.

Von Hippel-Lindau (VHL) disease is caused by germline mutation of the VHL tumour suppressor gene. Patients frequently develop multiple nervous system tumours, denominated haemangioblastomas. Analysis of affected autopsy tissues suggests that tumourigenesis propagates from developmentally arrested, embryonic cells and progresses with consistent architectural, cytological, and molecular sequences similar to haemangioblastic formation and differentiation in the embryo. In this study, we analysed 156 nervous system tumours, 139 of which had been surgically resected from 83 VHL patients. We demonstrate that large tumours consistently contain epithelioid components characteristic of haemangioblastic differentiation in comparison to small tumours that solely display a poorly differentiated, mesenchymal structure. We further show exclusive activation of HIF2alpha in both small mesenchymal tumours and the mesenchymal component of large tumours, whereas activation of HIF1alpha is associated with epithelioid structure. We also show that the MIB1 proliferative index is variably increased in the epithelioid component of large tumours, with extramedullary haematopoiesis foci within the epithelioid component at 100%. These data provide compelling evidence that nervous system tumourigenesis in VHL disease represents a protracted process of haemangioblastic proliferation and differentiation that parallels haemangioblastic formation and differentiation in the embryo.

Vascular endothelial growth factor (vascular permeability factor) expression in injured rat brain.

We investigated the expression of vascular endothelial growth factor (VEGF)/vascular permeability factor (VPF) in stab and freeze brain injury models in rats. Immunohistochemical staining with anti-VEGF antibodies demonstrated an increase in VEGF-positive cells in and around both lesions. Morphologically, the injury-induced VEGF-positive cells resembled astrocytes. Double immunofluorescent staining for the astrocytic marker glial fibrillary acidic protein (GFAP) and VEGF demonstrated directly that VEGF-positive cells which appeared in response to these injuries were astrocytes. VEGF expression in astrocytes was maximal on days 3 and 4 after injury in terms of both cell number and affected area. The increase in VEGF-positive cells was more widespread in the freeze lesion than in the stab wound, and occurred in both the lesioned and nonlesioned hemispheres. VEGF-positive cells were still present 3 weeks after both injuries, but their numbers were reduced and their distribution became limited to the immediate vicinity of the lesions. These observations indicate that astrocytes react to injury by increasing VEGF expression, suggesting that VEGF might participate in the central nervous system response to injury.

Mechanism of dexamethasone suppression of brain tumor-associated vascular permeability in rats. Involvement of the glucocorticoid receptor and vascular permeability factor.

Brain tumor-associated cerebral edema arises because tumor capillaries lack normal blood-brain barrier function; vascular permeability factor (VPF, also known as vascular endothelial growth factor, VEGF) is a likely mediator of this phenomenon. Clinically, dexamethasone reduces brain tumor-associated vascular permeability through poorly understood mechanisms. Our goals were to determine if suppression of permeability by dexamethasone might involve inhibition of VPF action or expression, and if dexamethasone effects in this setting are mediated by the glucocorticoid receptor (GR). In two rat models of permeability (peripheral vascular permeability induced by intradermal injection of 9L glioma cell-conditioned medium or purified VPF, and intracerebral vascular permeability induced by implanted 9L glioma), dexamethasone suppressed permeability in a dose-dependent manner. Since 80% of the permeability-inducing activity in 9L-conditioned medium was removed by anti-VPF antibodies, we examined dexamethasone effects of VPF expression in 9L cells. Dexamethasone inhibited FCS- and PDGF-dependent induction of VPF expression. At all levels (intradermal, intracranial, and cell culture), dexamethasone effects were reversed by the GR antagonist mifepristone (RU486). Dexamethasone may decrease brain tumor-associated vascular permeability by two GR-dependent mechanisms: reduction of the response of the vasculature to tumor-derived permeability factors (including VPF), and reduction of VPF expression by tumor cells.

VEGF165 expressed by a replication-deficient recombinant adenovirus vector induces angiogenesis in vivo.

To evaluate the concept that localized delivery of angiogenic factors via virus-mediated gene transfer may be useful in the treatment of ischemic disorders, the replication-deficient adenovirus (Ad) vector AdCMV.VEGF165 (where CMV is cytomegalovirus and VEGF is vascular endothelial growth factor) containing the cDNA for human VEGF165, a secreted endothelial cell-specific angiogenic growth factor, was constructed. Human umbilical vein endothelial cells (HUVECs) and rat aorta smooth muscle cells (RASMCs) infected with AdCMV.VEGF165 (5 and 20 plaque-forming units [pfu] per cell) demonstrated VEGF mRNA expression and protein secretion into the supernatant. Furthermore, the conditioned medium from these cells enhanced vascular permeability in vivo. In contrast, neither VEGF mRNA nor secreted protein was found in uninfected HUVECs or RASMCs or in cells infected with the control vector AdCMV.beta gal (where beta gal is beta-galactosidase). Assessment of starved HUVECs at 14 days demonstrated sixfold more cells for AdCMV.VEGF165-infected HUVECs (20 pfu per cell) than for either infected or uninfected control cells. RASMC proliferation was unaffected by infection with AdCMV.VEGF165. When plated in 2% serum on dishes precoated with reconstituted basement membrane (Matrigel), HUVECs infected with AdCMV.VEGF165 (20 pfu per cell) differentiated into capillary-like structures. Under similar conditions, both uninfected HUVECs and HUVECs infected with AdCMV.beta gal did not differentiate. To evaluate the ability of AdCMV.VEGF165 to function in vivo, either AdCMV. VEGF165 or AdCMV.beta gal (2 x 10(10) pfu) was resuspended in 0.5 mL Matrigel and injected subcutaneously into mice. Immunohistochemical staining demonstrated VEGF in the tissues surrounding the Matrigel plugs containing AdCMV.VEGF165 up to 3 weeks after injection, whereas no VEGF was found in the control plugs with AdCMV.beta gal. Two weeks after injection, there was histological evidence of neovascularization in the tissues surrounding the Matrigel containing AdCMV.VEGF165, whereas no significant angiogenesis was observed in response to AdCMV.beta gal. Furthermore, the Matrigel plugs with AdCMV.VEGF165 demonstrated hemoglobin content fourfold higher than the plugs with AdCMV.beta gal. Together, these in vitro and in vivo studies are consistent with the concept that Ad vectors may provide a useful strategy for efficient local delivery of VEGF165 in the treatment of ischemic diseases.

Differential expression of vascular endothelial growth factor (vascular permeability factor) forms in rat tissues.

Vascular endothelial growth factor (VEGF)/vascular permeability factor (VPF), exists as multiple forms due to alternative splicing of mRNA. VEGF165/164 (human/rodent homologue) is often assumed to be the predominant form, although truly quantitative assessments are lacking. We have used the RNase protection assay to directly quantitate the relative abundance of VEGF mRNA forms in five rat tissues (brain, kidney, lung, spleen, and heart) and two rat glioma cell lines (C6 and 9L). The three major forms, which code for proteins of 188, 164, and 120 amino acids, were observed in all of the tissues and cells examined. However, the relative abundance differed among the samples. VEGF188 was the predominant form (> 50% of total VEGF mRNA) in heart and lung, but was the least abundant form (6-15%) in all other samples. VEGF164 was lower (approximately 25%) in heart and lung, but was predominant (> 50%) in brain and kidney. VEGF164 and VEGF120 were present in equimolar amounts (each form approximately 46% of total) in the spleen, C6, and 9L. VEGF120 was also present in kidney (38%) and lung (27%) and was least abundant (approximately 15%) in brain and heart. A rat homologue of VEGF206 was not observed. VEGF mRNA splicing occurs in a tissue-specific manner. The assumption that the predominant physiologic form of VEGF is a VEGF165/164 homodimer should be viewed with caution.

Glial shape and cytoskeletal protein synthesis.

We investigated whether the shape of astroglial derived cells influences the expression of cytoskeletal proteins. In reaggregating cultures GFAP, vimentin and actin synthesis was approximately 52%, 50% and 37% the level found in monolayer cultures, respectively. Monolayer cultures consisted of polygonal shaped cells adhering to plastic, while reaggregating cultures were comprised of round cells growing in a suspension like culture. Additionally, human glioma cells induced to grow as round cells on poly-2-hydroxyethyl methacrylate (polyhema) coated plastic exhibited a level of GFAP synthesis that was approximately 20% the level displayed by polygonal shaped cells grown on uncoated plastic. Glioma cells initially grown on a polyhema surface and replated onto uncoated plastic were capable of reinitiating GFAP synthesis. Thus, alterations in the synthesis of GFAP and other cytoskeletal proteins can occur when astrocytes change their shape.

Insulin-like growth factor-I receptors in human glial tumors.

Insulin and insulin-like growth factors (IGFs) are anabolic effectors in many tissues and cultured cells, including astrocytes and neurons. Receptors for insulin and IGFs are found throughout the human brain. We examined the level of insulin and IGF receptors on membranes prepared from surgical specimens of tumor (astrocytomas and glioblastomas) and normal human brain. Specific binding (per 100 micrograms membrane protein) of insulin was less than 5% in all normal and tumor samples. Specific binding of IGF-I to 12 normal brain specimens ranged from 1-8%. IGF-I binding to 18 glioma specimens ranged from 2-25%. Scatchard analyses of IGF-I binding confirmed increased IGF-I-binding sites in some glial tumors vs. normal brain, but detected no difference in affinity characteristics. Cross-linking of [125I]IGF-I demonstrated that glioma tissue expressed the same lower mol wt (approximately 118 kDa) alpha-subunit as the normal brain confirming the neural origin of the cells expressing the IGF-I receptor. IGF-binding proteins (approximately 40 kDa) were also found in the membranes of some of the glioma but none of the normal brain specimens. In cell lines derived from glioma specimens, IGF binding was readily detectable (4-10% specific binding), but insulin binding was barely detectable (0-03%) in every line examined. The size of the IGF-I alpha-subunit in the cultured cells was larger (approximately 133 kDa) than that in the original tissue. Most glioma cell lines exhibited an IGF-I dose-dependent stimulation of thymidine incorporation into DNA, and partially purified IGF-I receptors from these cells exhibited a dose-dependent stimulation of the autophosphorylation of the beta-subunit. We conclude that human glioma cells have functional IGF-I receptors and suggest a role for this receptor in glioma cell growth.

The secretion of glucose into the egg of Gallus domesticus and observations on the uterine secretion of cations.

1. The concentrations of glucose, K+, Na+ and Ca2+ were measured in the uterine fluid and albumen of eggs removed from the shell gland at intervals of from 6 to 20 h after the previous oviposition. 2. A steady increase in the concentrations of glucose and K+ occurred in the uterine fluid during the initial 13 h of the egg's sojourn in the gland, resulting in their diffusion into the albumen. Meanwhile Na+ concentrations of the fluid and albumen decreased. 3. After 14 h in the gland and until oviposition there was little change in the concentrations. 4. Uterine fluid drained artificially from the gland using a Perspex egg had lower glucose and K+, but higher Na+ concentrations than the normal fluid. 5. The rates of secretion of glucose and K+ measured with the artificial egg resembled their entry rates into the normal egg, but there was an apparent deficit in Ca2+ secretion.