: IDO is expressed more strongly in both primary and secondary glioblastoma tissue than low-grade glioma and may affect clinical outcome. If IDO promotes glioma cells to escape from the immune system, IDO may be a crucial therapeutic target for malignant gliomas.

Pituitary adenomas are usually soft, but 5-13.5% are fibrous adenomas which are difficult to remove. Magnetic resonance (MR) imaging and operative findings were evaluated in eight patients (12.1%) with fibrous pituitary adenoma among 66 patients. Tumor specimens were examined histologically and immunohistochemically for collagen content and subtypes. Seven patients had clinically inactive nonfunctioning pituitary adenomas and one patient growth hormone-secreting adenoma. All patients underwent transsphenoidal surgery. Four cases were giant adenomas with suprasellar extension of more than 2 cm. T 1 -and T 2 -weighted MR imaging showed the tumors as nearly isointense to the surrounding brain, except in one case where the tumor was high intense on T 2 -weighted MR imaging. All tumors required piecemeal resection using a micro-dissector and tumor forceps. Four tumors of maximum size more than 3 cm needed a second operation. The interface between the thinned normal pituitary gland and fibrous adenoma was intended to identify at the anterior-superior portion in recent four cases, which was helpful to remove the tumors and preserve pituitary functions. Histological examination revealed prominent deposition of collagen in the perivascular area. The percentage of collagen content in fibrous adenomas was more than 5% and significantly higher than that in soft adenomas and normal pituitary glands. Immunohistochemical examination showed positive staining for collagen types I and III in the fibrous adenomas, but only for type V collagen in the normal pituitary glands. Large fibrous adenomas can be resected by transsphenoidal surgery which may require two-stage operations. Identification of the interface between the normal pituitary gland and adenoma is helpful to remove fibrous adenomas and to preserve pituitary functions. We propose that firm adenomas containing more than 5% collagen are``fibrous'' adenomas.

Invasion of tumor cells into the surrounding normal brain tissues is a prominent feature of malignant gliomas. Malignant glioma cells secrete thrombospondin-1 which participates in the motility of glioma cells and binds cell surface heparan sulfate proteoglycan. To clarify the invasion mechanism of tumor cells, expression of the syndecans (syndecan-1, -2, -3, and -4), a major cell surface heparan sulfate proteoglycan family, was analyzed in malignant gliomas. Involvement of nuclear factor-kappaB (NF-kappaB) on syndecan-1 expression was also investigated. Using reverse transcription-PCR, the authors analyzed the expression of syndecan-1, -2, -3, and -4 in 10 malignant glioma cell lines, 2 glioblastoma specimens, and 2 normal brain specimens. All malignant glioma cell lines and glioblastoma specimens expressed all types of syndecan mRNA, except in one glioma cell line that lacked syndecan-3 expression. On the other hand, normal brain specimens expressed syndecan-2, -3, and -4 mRNA, but did not syndecan-1 mRNA. Syndecan-1 protein was localized in the cell surface of all malignant glioma cell lines by flow cytometry. Various levels of active nuclear factor-kappa B (NF-kappaB) was detected in all malignant glioma cell lines using immunoblotting. The expression of active NF-kappaB and syndecan-1 increased in U251 glioma cells after tumor necrosis factor-alpha or interleukin-1beta treatment, which can activate NF-kappaB. The amplification of active NF-kappaB and syndecan-1 by tumor necrosis factor-alpha or interleukin-1beta was suppressed by an inhibitor of NF-kappaB activation (emodin). Emodin also downregulated the expression of syndecan-1 mRNA in U251 cells. These results indicate that malignant glioma cells express all types of syndecans and suggest that NF-kappaB participates in the upregulation of the syndecan-1 expression at the transcriptional level, and increased expression of syndecan-1 could associate with extracellular matrices including thrombospondin-1.

Malignant glioma cells secrete thrombospondin-1 (TSP-1) which participates in the motility of glioma cells, and binds to cell surface alphavbeta3 and alpha3beta1 integrins, and syndecan-1. This study evaluated the amount of TSP-1 secretion from malignant glioma cells, and the expression of alphavbeta3 and alpha3beta1 integrins, and syndecan-1. The amounts of TSP-1 in the supernatants from 10 malignant glioma cell lines and eight non-glioma malignant tumor cell lines were measured by enzyme-linked immunosorbent assay. Expression of alphavbeta3 and alpha3beta1 integrins, and syndecan-1 were examined by flow cytometry. The amounts of TSP-1 secreted by malignant glioma cells were 43 to 2431 ng/l x 10(6) cells/24 h (mean +/- SD = 626 +/- 792). Seven of 10 glioma cell lines secreted more than 100 ng of TSP-1 and three of these cell lines secreted more than 1 microg. Seven of eight non-glioma cell lines secreted less than 100 ng of TSP-1. All glioma cell lines expressed alpha3beta1 integrin and syndecan-1, and seven of 10 glioma cell lines expressed alphavbeta3 integrin. Treatment of the glioma cell lines with TGF-beta2 did not change the expression of alphavbeta3 integrin. These results suggest that malignant glioma cells secrete high levels of TSP-1, which may be important in the migration of glioma cells via interactions with alphavbeta3 and alpha3beta1 integrins, and syndecan-1.

The effect of transforming growth factor-beta (TGF-beta) secreted by glioblastoma (T98G) cells on the secretion of interferon-gamma (IFN-gamma) by lymphokine-activated killer (LAK) cells stimulated with tumor cells was investigated in cocultures of LAK and Daudi cells supplemented with T98G culture supernatant, T98G culture supernatant preincubated with anti-TGF-beta 1 and anti-TGF-beta 2 neutralizing antibodies, anti-TGF-beta 1 and anti-TGF-beta 2 antibodies, or natural human TGF-beta 1 or recombinant human TGF-beta 2. LAK cells were incubated with anti-TGF-beta 1 and anti-TGF-beta 2 antibodies, and with T98G cells of which the supernatant contained both active and latent forms of TGF-beta 1 and TGF-beta 2, with or without neutralizing antibodies. Addition of the supernatant from T98G cells to LAK/Daudi culture caused inhibition of IFN-gamma secretion by LAK cells. The inhibition was abolished by pretreatment of the supernatants with anti-TGF-beta antibodies. Addition of TGF-beta 1 and TGF-beta 2 to the LAK/Daudi culture inhibited IFN-gamma secretion by LAK cells in a dose-dependent manner. Addition of anti-TGF-beta antibodies to the LAK culture resulted in increased IFN-gamma secretion. T98G cells failed to stimulate LAK cells to secrete more IFN-gamma. Addition of anti-TGF-beta antibodies to the LAK-T98G culture resulted in increased IFN-gamma secretion by LAK cells. These results suggest that most malignant glioma cells which secrete high levels of TGF-beta can inhibit IFN-gamma secretion by LAK cells even after tumor cell stimulation.

The secretion of transforming growth factor-beta (TGF-beta), a growth inhibitory factor with immunosuppressive properties, was investigated in one glioblastoma cell line and seven surgically resected malignant glioma cells. Cultured cells from surgically resected tumors were examined immunohistochemically for glial fibrillary acidic protein (GFAP) and S-100 protein. The levels of TGF-beta 1 and TGF-beta 2 in culture supernatants from malignant glioma cells were determined by a specific bioassay using anti-TGF-beta 1 and anti-TGF-beta 2 antibodies. Two glioblastoma cell lines were cultured in the presence of TGF-beta 1 or TGF-beta 2 to assess the effect of TGF-beta on the growth of glioblastoma cells. Cultured cells from surgically resected tumors were positive for both GFAP and S-100 protein. Both active and latent forms of TGF-beta 1 and TGF-beta 2 were detected in the culture supernatants from malignant gliomas, except in one patient with anaplastic astrocytoma which secreted only latent forms of TGF-beta 1 and TGF-beta 2. There was no statistical difference in the levels of TGF-beta 1 and TGF-beta 2 in glioblastomas and anaplastic astrocytomas. Neither TGF-beta 1 nor TGF-beta 2 affected the growth of glioblastoma cells. These findings suggest that most malignant glioma cells secrete both TGF-beta 1 and TGF-beta 2, can convert TGF-beta from a latent to active form, and may suppress cytokine secretion by activated lymphocytes in vivo as well as in vitro.

Orbital cavernous hemangiomas (CHs) manifest as slowly developing symptoms indicative of slow growth. The present study investigated the involvement of angiogenic factors and their receptors in the growth of orbital CHs. Surgical specimens of orbital CHs were obtained from nine patients. Formalinfixed, paraffin-embedded specimens were stained immunohistochemically using antibodies against Ki-67, CD31, a-smooth muscle actin (a-SMA), vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and VEGF receptors (flt-1 and flk-1). CD31 was expressed in the single layer of endothelial cells lining the vascular cavity. The thick vascular walls were positive for a-SMA, indicating that the vascular walls were smooth muscle cells. Ki-67 antigen immunostaining was mostly positive in the vascular walls and the staining index ranged from 0% to 6.8% (mean ± standard deviation, 2.7 ± 1.9%). VEGF and bFGF immunostaining were positive in all specimens. Flt-1 immunostaining was negative in all specimens, but flk-1 immunostaining was positive in both endothelial cells and smooth muscle cells. These results suggest that both VEGF and its receptor flk-1 are important in the growth of orbital CH.

Malignant glioma cells secrete transforming growth factor-b (TGF-b) and can activate latent TGF-b. However, the mechanism of the latent TGF-b activation has not yet been determined. This study examined whether thrombospondin-1 (TSP-1) secreted by malignant glioma cell lines participates in the activation of latent TGF-b secreted by the glioma cells. Western blot analysis revealed that TSP-1 was present in both the cell lysates and the culture supernatants of all three malignant glioma cell lines (T98G, A172, and U251). A bioassay for TGF-b activity revealed that all malignant glioma cell lines used in this study could activate latent TGF-b by themselves. Latent TGF-b1 activation, evaluated by enzyme-linked immunosorbent assay, was inhibited by more than 50% by the addition of neutralizing anti-TSP-1 monoclonal antibody or anti-TSP-1 polyclonal antibody. These results indicate that TSP-1 has a predominant role in the activation of latent TGF-b in malignant glioma cells.