Correlation of cytogenetic and fluorescence in situ hybridization (FISH) studies in normal and gliotic brain.

In vitro tissue culturing, for karyotype analysis, may introduce artifacts confounding the cytogenetic evaluation of tissues with low baseline proliferative activity. Utilizing a panel of fluorescence in situ hybridization (FISH) probes for chromosomes 7, 8, 9, 10, 12, 17, 18, X, and Y, we compared the results of FISH analysis of non-tumorous normal (11 patients) and gliotic (10 patients) brain tissue touch preparations with those of cytogenetic evaluation performed on short-term primary cultures of the same material. We found a significant rate of apparent monosomy of chromosomes 8 and 17 by FISH analysis, with no corresponding clonal chromosomal loss detected by karyotype evaluation. These monosomy rates were significantly lower in gliotic than in normal brain tissue, and image analysis suggested that this apparent monosomy was due to interphase pairing of homologous centromere signals. Two distinct Y-chromosome signals were seen in 9.4% of nuclei by FISH, with 3 of 15 males displaying disomy Y rates over 15%. Disomy Y rates correlated approximately with age and clonal disomy Y was seen in the karyotype of one of these specimens. Karyotype analysis demonstrated loss of a sex chromosome in 6 specimens, while no sex chromosome nullisomy was detected by FISH. FISH is a valuable adjunct to the cytogenetic evaluation of tissues with low baseline proliferative activity. The differences in relative monosomy rates between normal and gliotic brain suggest that alterations in nuclear architecture and/or DNA sequence accompany the transition from normal to reactive glia.

Transcript distribution of plasma membrane Ca2+ pump isoforms and splice variants in the human brain.

Plasma membrane calcium pumps (PMCAs) play a major role in the maintenance and fine regulation of the intracellular Ca2+ concentration. Fourteen subregions of the normal human brain were carefully dissected and analyzed by reverse transcriptase-polymerase chain reaction for the distribution of mRNAs corresponding to the four known PMCA genes as well as their alternative splicing products at two previously defined 'hotspots' A and C. All PMCA genes were found to be expressed in every brain subregion; however, consistent differences were found in the distribution of alternative splice options. The four cortical regions and hippocampus were characterized by the relative preference of variants that include an entire exon at site C and lead to the expression of isoforms of the a-type. Inferior olive and olfactory bulb showed a relative preponderance of the b-form 'default' types of alternative splicing at site C, and a decrease or even the lack of 'differentiated' forms such as variants 1a and 1c. At the N-terminal splice site A, the default x-type variants were predominant in all brain regions for PMCA 1, 3, and 4. By contrast, the pattern of PMCA2 variants was the most variable, ranging from the presence of the entire set of 2x, 2w, and 2z forms in inferior olive to the almost exclusive presence of form 2z (excluding all alternatively spliced sequences) in the four cortical regions, caudate, and hippocampus. Regional differences in the PMCA splice type distribution in normal human brain may correlate with different demands on the regulation of the set-point resting Ca2+ levels in these areas. Changes in these patterns may correlate with altered physiological states of the affected regions and/or reflect an (early) sign of Ca2+ dyshomeostasis characteristic of many neurodegenerative diseases.

Changes in proliferating cell nuclear antigen expression in glioblastoma multiforme cells along a stereotactic biopsy trajectory.

Proliferating cell nuclear antigen, an auxiliary protein of deoxyribonucleic acid polymerase-delta, has been shown to be a reliable marker of nuclear deoxyribonucleic acid synthetic activity. We applied a monoclonal antibody to proliferating cell nuclear antigen to a series of serial stereotactic biopsies from patients with glioblastoma multiforme and found the proliferative activity to vary relative to biopsy location within or surrounding the solid tissue component of the tumor. Twenty-seven trajectories in 26 patients were analyzed, each consisting of two to five sequential 10 x 1.5 mm core biopsies (mean = 3). The proliferative index (PI) was greatest in those cells located at the solid tumor-infiltrated parenchyma interface. PI values were significantly lower in those biopsy cores located proximal (within infiltrated parenchyma) and distal (within solid tumor tissue) to the solid tumor-infiltrated parenchyma interface (median PI values, proximal to distal: 0.38, 0.66, 5.45 solid tumor-infiltrated parenchyma interface], 0.39, 0.09%). The mean PI values were significantly lower in neoplastic cells samples from regions of peripheral hypodensity on computed tomographic scans compared with those sampled from contrast-enhancing regions (0.9 and 3.91%, respectively). There was no significant difference in the mean PI values of neoplastic cells sampled from regions of contrast enhancement or central hypodensity (3.91 and 4.31%, respectively).

Use of fluorescence in situ hybridization to detect loss of chromosome 10 in astrocytomas.

Models describing progression in the genetic derangement of glial tumors have shown loss of chromosome 10 to occur most frequently in high-grade lesions, suggesting that identification of this loss may be prognostically significant. Fluorescence in situ hybridization (FISH) analysis may be a valuable adjunct to histological grading if it can accurately detect this loss. In this paper, the authors correlate results obtained from FISH, cytogenetic, molecular genetic, and flow cytometric analyses of a series of 39 brain specimens, including seven normal, two gliotic, and 30 neoplastic (one Grade II, one Grade III, and 28 Grade IV astrocytoma) specimens. Contiguous section of freshly resected surgical tissue were submitted for tissue culturing (karyotype) and touch preparation (FISH), snap-frozen (molecular genetic), or paraffin-embedded (histology and flow cytometry). Centromere-specific probes for chromosomes 10 and 12 were used for FISH analysis, and 19 restriction fragment length polymorphisms (two p-arm and 17 q-arm) and four microsatellite sequence polymorphisms (three p-arm and one q-arm) were used for molecular genetic analysis of chromosome 10. Findings showed FISH and loss of heterozygosity (LOH) analyses to be concordant in 33 of 38 specimens (sensitivity 94%, specificity 81%), with one specimen indeterminate on LOH analysis. Both FISH and LOH analyses were more sensitive at detecting chromosome 10 loss than conventional cytogenetic (karyotype) analysis. The authors conclude that FISH is a sensitive test for detecting chromosome 10 loss and ploidy in astrocytic tumors.

Computer-assisted stereotactic third ventriculostomy in the management of noncommunicating hydrocephalus.

Between January, 1984, and December, 1990, a total of 85 patients have undergone stereotactic third ventriculostomy (TV) at the Mayo Clinic. Sixty-one patients (74%) presented with mass lesions in the region of the posterior third ventricle, aqueduct of Sylvius or fourth ventricle. Twenty-four patients (26%) presented with non-tumoral, adult/adolescent-onset aqueductal stenosis. Follow-up was available for all 85 patients and ranged from 1 to 66 months (mean 25 months). Follow-up revealed initial patency in 84 patients. Eleven patients (13%) ultimately required extracranial shunting for persistent symptomatic hydrocephalus. Two patients underwent revision of their TV. Twenty-seven patients had been previously shunted. Of these, 23 (85%) have remained asymptomatic after TV without the need for further shunting. Stereotactic TV is a safe and effective way of re-establishing normal cerebrospinal fluid flow dynamics in selected cases of obstructive hydrocephalus.

Molecular genetics of astrocytomas and meningiomas.

Although both spatial and temporal heterogeneity confound the description of the genetic events underlying glioma tumorigenesis, it is becoming evident that chromosome 17 loss and p53 inactivating mutations are probably involved early in the pathway of tumorigenesis of some, but not all, astrocytomas. Chromosome 10 loss and epidermal growth factor receptor amplification are seen predominantly in high-grade lesions, although they have not been shown to be independent prognostic indicators. Data is accumulating on the presence of a tumor suppressor gene on chromosome 9p, although the gene remains to be identified. The roles of chromosome 22 and the NF-2 tumor suppressor gene in the tumorigenesis of sporadic and familial meningiomas are discussed here, along with other nonrandom chromosomal alterations that are seen in both astrocytomas and meningiomas.