Schwannomas are benign tumors of cranial, spinal, and other nerve sheaths that develop sporadically or are inherited as part of neurofibromatosis type 2 (NF2). The NF2 gene (SCH) on chromosome 22 has recently been identified and shown to be inactivated by mutation and allele loss in some schwannomas. However, only limited regions in the SCH coding region were examined for mutations. We have extended these studies by screening virtually all coding sequences of the SCH gene (95% coverage) and adjacent splice site sequences for the presence of mutations in 48 schwannomas. All tumors (34 vestibular schwannomas and 14 schwannomas of other locations) were additionally characterized for allele loss on chromosome 22. By PCR-DGGE screening of the 16 known exons of the SCH gene, 22 mutations were found. Most of these give rise to a premature stop codon and are expected to result in the synthesis of a truncated gene product (schwannomin). Although there was no apparent hotspot for mutations, 16 of the 22 mutations occurred in the first eight exons or adjacent splice site sequences of the SCH gene. In several vestibular as well as other schwannomas loss of one SCH allele and mutational inactivation of the second allele were identified in the same tumor. Our data indicate that the SCH gene is implicated in the development of schwannomas of all locations in the nervous system.

Schwannomas are tumors of the cranial, spinal, and peripheral nerve sheaths that originate from Schwann cells. Acoustic neurinomas are the most frequent cranial schwannomas. They might develop sporadically or in the context of neurofibromatosis type 2 (NF2). Loss of part or all of chromosome 22 is frequently found in acoustic schwannomas, suggesting that the NF2 gene is a tumor suppressor gene involved in the genesis of these tumors. Only a few spinal schwannomas have been molecularly characterized so far, showing that chromosome 22 loss might also occur in these tumors. Here we present the molecular analysis of chromosome 22 in 23 acoustic schwannomas and nine schwannomas of other locations (including other cranial nerves and spinal and peripheral nerves). Most of these tumors were from sporadic cases. Multiple schwannomas of various locations were analyzed in two patients with NF2. We found partial or complete monosomy for chromosome 22 in 22% of the acoustic schwannomas and 55% of the non-acoustic schwannomas. The tumors with partial monosomy included four with terminal deletions and one with a deletion of the centromeric part of the long arm of chromosome 22. The region between the beta B2-1 crystallin locus (CRYB2A) and the myoglobin locus (MB) was commonly deleted in these tumors. Our studies suggest that a schwannoma-related tumor suppressor gene within this region, which might be the NF2 gene, is involved in the development of schwannomas of various locations in the nervous system. Our studies indicate that the second hit in the genesis of different schwannomas within one (predisposed) NF2 patient occurs independently and via different mechanisms.

Background. The endothelial cell marker PAL‐E is not reactive to vessels in the normal brain. The present study concerns the PAL‐E reactivity in brain tumors in contrast to normal brain and nonneoplastic brain disease. Methods. A total of 122 specimens were examined: brain tumors (n = 94), nonneoplastic brain disease (n = 19), normal brain (n = 8), and fetal brain (n = 1). Standard immunohistochemical procedures using a panel of endothelial cell markers were applied to detect vessels reactive to PAL‐E. Results. PAL‐E reactivity to endothelial cells was found in all cases of glioblastoma multiforme, in 75% of the cases of anaplastic astrocytoma, and in 46% of the cases of astrocytoma. Furthermore, PAL‐E reactivity was present in diseases with a developmental etiology, such as primitive tumors and congenital vascular malformations. The developing human brain (6‐weeks' gestation age) and special sites of the mature brain, sites without blood‐brain barrier, showed a strong reactivity, which indicates a relation with the status of blood‐brain barrier development. Conclusions. PAL‐E is the only marker out of a panel of endothelial cell markers that shows no reactivity to endothelial cells in the normal brain with an intact blood‐brain barrier. In primary and metastatic brain tumors, PAL‐E is reactive to endothelial cells, except for 25% of anaplastic astrocytoma and 54% of astrocytoma. PAL‐E reactivity in brain tumors most likely is related to angiogenesis and to blood‐tumor barrier properties not present in the normal blood‐brain barrier.

Loss of heterozygosity (LOH) for chromosome arms 9p, 10p, 10q, and 17p and amplification of the epidermal growth factor receptor (EGFR) gene have been identified as frequent genetic changes in malignant astrocytomas. We have found amplification of the anonymous marker D17S67 on chromosome arm 17p in 10% (3 of 30 cases) of astrocytomas of the highest malignancy grade. The tumors with D17S67 amplification displayed other genetic changes on chromosome 17, including additional amplifications and deletions. All three patients with D17S67 amplification developed severe brain edema and died within 1 month after operation.

The cytotoxic action of various single doses of MNU and ENU on developing neural and extraneural tissues was studied at different stages of development. Examination revealed lethal damage. (L.I.) and mitotic inhibition (M.I.), confined to proliferating cells only, and caused by the number of alkyl groups administered. In studying the duration of M.I. a difference was found in duration of the cell cycle arrest after MNU or ENU. The arrest lasted longer for MNU than for ENU, and the neural tissues turned out to be more sensitive than the extraneural ones. Moreover, among the reappearing mitotic figures abnormal ones were noticed frequently. After pulse-labeling with thymidine this arrest could be traced to take place in or before entering the S-phase. During the period of this arrest a low, but specific, activity was found that might point to the existence of repair-processes in vivo. Finally, we directly demonstrated alkylations in tissue-sections by the use of (14C-methyl)-MNU. High radioactivity was found with a random distribution over the various tissues, cell types and even cellular compartments. Therefore--in contrast with the cytotoxic effects--alkylation seems to occur in all cell types. In conclusion, it seems justified to consider the matrices of proliferating cells in the central nervous system as the target tissue-areas for the carcinogenic action of both MNU and ENU. Re-entrance of these damaged cells into their cycle prior to the elimination of altered bases from DNA might be of great importance for the problem of oncogenesis.