Medulloblastomas exhibit an array of diverse cytogenetic abnormalities. To evaluate the signi®cance of epigenetic rather than genetic lesions in medulloblastomas and other primitive neuroectodermal tumors (PNETs) of the childhood CNS we performed a systematic analysis of gene speci®c and global methylation. Methylationspeci®c PCR detected no methylation for p15 INK4B, von Hippel Lindau and TP53 and only limited methylation for E-Cadherin and p16INK4A in tumors. The cell lines Daoy and MHH-PNET-5 in which the p16 INK4A promoter was methylated did not express the gene, but demonstrated abnormalities by SSCP. Immunohistochemistry for p16 was negative in all examined normal cerebella and medulloblastomas. Using the technique of Restriction Landmark Genomic Scanning we detected methylation aecting up to 1% of all CpG islands in primary MB/PNETs and 6% in MB cell lines. Methylation patterns diered between medulloblastomas and PNETs. Examination of several methylated sequences revealed homologies to known genes and expressed sequences. Analysis of survival data identi®ed seven of 30 hypermethylated sequences signi®cantly correlating with poor prognosis. We suggest that DNA hypermethylation has an outstanding potential for the identi®cation of novel tumor suppressors as well as diagnostic and therapeutic targets in MBs and other PNETs of the CNS. Oncogene (2001) 20, 5033 ± 5042.

No abstract

SummaryWe hypothesized that non-proliferating (quiescent) human vascular endothelial cells would not express somatostatin receptor subtype 2 (sst 2) and that this receptor would be expressed when the endothelial cells begin to grow. To test this hypothesis, placental veins were harvested from 6 human placentas and 2 mm vein disks were cultured in 0.3% fibrin gels. Morphometric analysis confirmed that 50-75% of cultured vein disks developed radial capillary growth within 15 days. Sst 2 gene expression was determined by reverse transcription-polymerase chain reaction (RT-PCR) analysis of the RNA from veins before culture and from tissue-matched vein disks that exhibited an angiogenic response. The sst 2 gene was expressed in the proliferating angiogenic sprouts of human vascular endothelium. The presence of sst 2 receptors on proliferating angiogenic vessels was confirmed by immunohistochemical staining and in vivo scintigraphy. These results suggest that sst 2 may be a unique target for antiangiogenic therapy with sst 2 preferring somatostatin analogues conjugated to radioisotopes or cytotoxic agents.

Somatostatin is known to modulate mesangial and tubular cell function and growth, but the somatostatin receptor (sst) subtypes responsible for these effects have not been defined. There are at least five different sst receptor subtypes (sst(1)-sst(5)). We used RT-PCR to demonstrate that normal human kidney consistently expresses mRNA for sst(1) and sst(2) (9 of 9 donors). Some donors expressed sst(4) or sst(5) mRNA, but none expressed sst(3) mRNA. Expression of sst(1) and sst(2) was further assessed by staining serial sections of normal human kidney with sst(1) and sst(2) antisera, Arachis hypogaea (AH) lectin (to define distal tubule/collecting duct cells), Phaseolus vulgaris lectin (proximal tubules), and Tamm-Horsfall protein (THP) antiserum (thick ascending limb of the loop of Henle). Specificity of antisera was demonstrated by transfection and absorption studies. Sst(2), but not sst(1), was expressed in glomeruli. Intense sst(1) and sst(2) staining localized exclusively to AH+ and THP+ tubules. Thus sst(1) and sst(2) subtype-selective analogs may be useful to beneficially modulate renal cell function in pathological conditions.