Search

Monometric DNA impulse cytophotometry of ploidism state (diploid or aneuploid tumors with stemmline shift, polyploidism ) and proportion of DNA-synthesized cells (S-phase proportion) were determined on over 500, mostly malignant, tumors of different sites. The results were compared with histopathological findings (TNM stage and degree of differentiation). Large diploid carcinomas of the oral cavity have lower S-phase activity than aneuploid tumors. Carcinomas of the breast, stomach and ovary are frequently diploid. Aneuploid carcinomas of these regions generally have high synthesizing activity. Mucoid signet-ring carcinomas of the stomach have lower S-phase activity than non-mucoid carcinomas. In the colorectal region aneuploid carcinoma predominates, in the colon more than in the rectum. Tumor metastases from colon and rectum into the liver predominantly are from aneuploid primary tumors. For tumors of the breast, corpus uteri and ovary with high and moderate differentiation there is a direct correlation between histological grade of differentiation and the S-phase proportions. Aneuploid tumors vary from low to high synthesizing activity. Diploid meningioma and glioma have a low S-phase proportion, aneuploidism correlates with an increase in growth. Supplementation of histopathological diagnosis by determining ploidism and S-phase activity makes an important contribution in the assessment of the degree of malignancy, as well as for therapeutic purposes.

Cadherins are a family of glycoproteins that are associated with cell adhesion mechanisms. They are divided into subclasses. The E- and P-cadherins are regarded as the epithelial subtype. Their expression has been demonstrated in many different carcinoma types. Using immunomorphological techniques, we studied the expression of E-cadherin in a series of 145 human brain tumours with the monoclonal antibody 5H9. Western blot analysis was used to confirm the immunohistochemical data. The tumour types represented were astrocytoma WHO I (n = 7), astrocytoma WHO II (n = 6), astrocytoma WHO III (n = 14), glioblastoma WHO IV (n = 8), oligodendroglioma WHO II (n = 5), ependymoma WHO II (n = 5), choroid plexus papilloma WHO I (n = 5), pineoblastoma WHO IV (n = 5), medulloblastoma WHO IV (n = 5), neurinoma WHO I (n = 5), meningioma WHO I and WHO III (n = 75) and pituitary adenoma WHO I (n = 5). Only choroid plexus papillomas (5/5) and meningiomas showed E-cadherin expression. In benign meningiomas (n = 45; 100%), positive E-cadherin immunoreactivity was found regardless of the histomorphological subtype. E-Cadherin was also expressed in 21 WHO I meningiomas (100%) invading dura, bone, brain, and muscle. In contrast, E-cadherin was absent from the majority of morphologically malignant meningiomas (6/9, 66.6%). In addition, in recurrent meningiomas (n = 9), E-cadherin expression in the recurrent tumours was identical to that in the primary neoplasm except in cases with malignant progression, where the malignant recurrent tumour was E-cadherin negative. In 2 cases of metastasizing meningiomas, no E-cadherin immunoreactivity was found in the primary tumours or their metastases.

We report on three patients who presented with a cerebellar medulloblastoma at age 48, 53, and 59 years. Histopathology showed typical features of medulloblastoma, in one case with marked neuronal differentiation. In addition, all neoplasms contained focal accumulations of mature fat cells. Immunoreactivity of adipocytes for S-100 protein, neuron-specific enolase, synaptophysin, microtubule-associated protein-2, and glial fibrillary acidic protein and the lack of immunoreactivity to type IV collagen suggest lipomatous differentiation of neoplastic primitive neuroectodermal cells rather than an admixture of mesenchymal elements. Mitotic activity was low and the growth fraction, as determined by the MIB-I labeling index, was less than 5%. All patients are alive with a recurrence-free interval ranging from 3.5 to 12 years. These three patients and five similar previously reported cases all fit into the concept of the lipomatous medulloblastoma as a new clinicopathological entity characterized by (a) typical features of a cerebellar medulloblastoma with advanced neuronal differentiation, (b) areas of lipomatous differentiation, (c) low proliferative potential, (d) manifestation in adults (mean age, 50 years), and (e) apparent favorable clinical prognosis.

The development of human cancer is a multistep process that entails a progressively more malignant phenotype through the evolution of cellular subsets with increasing numbers of genetic alterations. Here we review the molecular genetics of human cancer predisposition and progression and describe paradigmatic cancer types and cancer syndromes. We also briefly consider the future impact of molecular biology on cancer diagnosis and treatment.

Synaptophysin is an acidic, integral membrane glycoprotein (Mr 38,000) of presynaptic vesicles in various neurons and neuroendocrine cells, and in tumours derived from such cells. By indirect immunofluorescence microscopy of cryostat sections, using the monoclonal antibody SY 38 to synaptophysin, a consistent positive immunoreactivity was observed in all medulloblastomas (n = 6) and neuroblastomas (n = 3) as well as a ganglioneuroma and a glioneuronal hamartoma. The presence of synaptophysin in medulloblastomas was confirmed biochemically by immunoblotting experiments. For purpose of comparison, the expression of intermediate-sized filament (IF) proteins was also examined. While neurofilament proteins were consistently expressed in the neuroblastomas (3/3), the ganglioneuroma and the glioneuronal hamartoma, IF distribution in medulloblastomas was variable. A neurofilament-positive type of tumour (1/6) could be distinguished from vimentin-expressing neoplasms (4/6) by immunocytochemistry. These data indicate that synaptophysin is a reliable marker for medulloblastomas as well as other differentiated and undifferentiated neuronal tumours and in this respect is superior to the more heterogeneous expression patterns of IF proteins in these tumours.

Using a lectin-peroxidase method, Concanavalin A binding was examined on formalin-fixed paraffin-embedded biopsy specimens (n = 143) of the most frequent central nervous system tumours. The brain tumours included oligodendrogliomas, astrocytomas, glioblastomas, ependymomas, neurinomas, meningiomas, medulloblastomas and plexus papillomas. In oligodendroglioma cells, only a weak granular intracytoplasmic staining was observed. The astrocytomas showed a strong reaction in fibrillary astrocytes and in tumour areas undergoing small cystic degeneration. Staining of protoplasmic astrocytes was weaker; pilocytic astrocytes demonstrated poor perinuclear staining. Intracytoplasmic Con A binding in gemistocytic astrocytes was distinct but inconstant and rather diffuse. In the glioblastomas the lymphocyte-like small astrocytes were negative. Giant multinucleated astrocytes stained strongly. In ependymomas no or at most a weak perinuclear reaction was observed, whereas the acceptor density was as high as in the normal ependymocytes in areas where the tumour was capable of producing organotypical structures. Plexus papillomas showed a strong intracytoplasmic staining comparable to the normal plexus epithelial cell. This feature was preserved in the malignant variants. In general, meningiomas and neurinomas were negative. Xanthomatous-degenerated meningioma cells, however, showed a distinct to strong intracytoplasmic staining. This feature was characteristic for the xanthomatous subtype of meningiomas. Granular cells with strong intracytoplasmic Con A staining often occurred at the border of fibrillary to reticular differentiated areas of neurinomas. Medulloblastomas were completely negative. Our results indicate that Con A binding to human brain tumours is specific and rather cytotypical than histotypical. The Con A acceptor density is probably related to the grade of differentiation. Lectin mapping of tumours leads to cytotypical binding patterns which may contribute to the differential diagnosis of neoplasias.

Human gliomas were grown as permanent tissue cultures and xenografts in nude mice. Growth kinetics, immunohistology and karyotypes were established. One tumor showed a distinct change in its karyotype after only two passages in nude mice; by contrast, no change of immunophenotype occurred during and after in vivo passaging. Treatment of this glioma as xenograft in nude mice revealed a high sensitivity towards single agent treatment with BCNU. In view of the cytologic aberrations, however, this result should be interpreted cautiously with regard to the clinical situation.

Peanut lectin (PNL) binding to a total of 13 granular cell tumours was examined by means of the peroxidase antiperoxidase technique. The tumours included six tumourettes of the neurohypophysis, one malignant granular cell tumour of the brain, and six peripheral tumours of distinct locations. Every tumour studied showed intracytoplasmic fine granular PNL binding; after pretreatment with neuraminidase, the weakly positive reaction was enhanced to a great extent. In all tumours simultaneous examination for the detection of lysozyme and glial fibrillary acidic protein (GFAP) was also carried out. Lysozyme was negative in all cases, whereas GFAP expression could be demonstrated at the periphery of the malignant granular cell tumour of the brain. The data presented clearly demonstrate that PNL can be used as a histochemical marker for granular cells regardless of their location. The fact that the presence of lysozyme could not be proved does not support the view of a histiocytic origin for granular cells, whereas the expression of GFAP in some immature granular cells of the brain tumour examined is considered to be an argument in favor of its glial origin.

Desmosomal proteins are co-expressed with intermediate-sized filaments (IF) of the cytokeratin type in epithelial cells, and these IF are firmly attached to the desmosomal plaque. In meningiomal and certain arachnoidal cells, however, vimentin IF are attached to desmosomal plaques. Meningiomas obtained after surgery, arachnoid "membranes", and arachnoid granulations at autopsy, as well as meningiomal cells grown in short-term culture have been examined by single and double immunofluorescence and immunoelectron microscopy using antibodies to desmoplakins, vimentin, cytokeratins, glial filament protein, neurofilament protein, and procollagen. In addition, two-dimensional gel electrophoresis of the cytoskeletal proteins has been performed. Using all of these techniques, vimentin was the only IF protein that was detected in significant amounts. The junctions morphologically resembling desmosomes of epithelial cells have been identified as true desmosomes by antibodies specific for desmoplakins and they provided the membrane attachment sites for the vimentin IF. These findings show that anchorage of IF to the cell surface at desmosomal plaques is not restricted to cytokeratin IF as in epithelial cells and desmin IF as in cardiac myocytes, suggesting that binding to desmosomes and hemidesmosomes is a more common feature of IF organization. The co-expression of desmosomal proteins and IF of the vimentin type only defines a new class of cell ("desmofibrocyte") and may also provide an important histodiagnostic criterion.

In order to determine the effect of routine fixation on the lectin affinity of tissue structures, we used unconjugated lectins and an indirect immunoalkaline-phosphatase method for frozen sections, and the peroxidase-anti-peroxidase method for paraffin-embedded, formalin-fixed tissue sections. Fourteen hyperplastic human tonsils were used, and the results of the binding spectra of each lectin were compared. In general, the binding spectrum detected in the paraffin sections was part of the broader range of affinity obtained in the frozen sections. The lectin receptors on the cell surface were especially affected by formalin fixation. On the other hand, the paraffin sections, because of their superior morphology, showed a better localization of the cytoplasmic reaction product and discriminated the cell types more accurately. Thus, the two tissue preparations are rather complementary. In the tonsil peanut agglutinin (PNA) and periodic acid/Concanavalin A (PA/Con A) proved to be suitable tools for distinguishing exactly between the crypt and the surface epithelium. Ulex europaeus agglutinin I (UEA) is a reliable endothelial marker with a strong affinity to the crypt epithelium. In the frozen sections, PNA regularly stained follicular-centre cells on their cell surface. PNA, Helix pomatia agglutinin (HPA), soybean agglutinin (SBA) and Con A stained the histiocytic population, especially PNA which additionally stained an "asteroid" histiocyte. This cell probably corresponds to the antigen-presenting histiocyte of the T-system.

The expression of intermediate-sized filaments (IF) was examined by immunocytochemical methods in 40 primary renal cell carcinomas and compared with the IF distribution in the normal adult human kidney. All tumours stained positively with cytokeratin IF antibodies. Co-expression of cytokeratins and vimentin was observed in 21/40 (52,5%) renal carcinomas. Double immunofluorescence labelling demonstrated that in most of these cases tumour cells contained both cytokeratin and vimentin type IF. In normal human kidneys, cells of the various tubular segments disclosed a positive reaction with cytokeratin antibodies in a different intensity and intracellular localization. Co-expression of cytokeratin and vimentin IF in normal adult human kidneys has never been observed. From a histogenetic point of view, co-expression of cytokeratins and vimentin in renal cell carcinoma obviously represents an atavistic phenomenon since vimentin is re-expressed by these tumour cells during neoplastic transformation. This finding indicates the metanephric origin of the renal parenchyma. In surgical pathology the possibility of very rare co-expression of cytokeratin and vimentin IF within tumour cells should be considered, particularly in the differential diagnosis of clear cell carcinomas.

Peanut lectin (PNL), Concanavalin A (Con A) and Ulex europaeus lectin I (Ulex) were chosen to map their binding sites in different regions of formalin fixed and paraffin embedded human central nervous system tissue and pituitary gland tissues. An extended PaP method was used for PNL and Ulex, whereas a direct peroxidase technique was employed for Con A. In astrocytes, the cytoplasm as well as the delicate processes were stained by PNL and Con A; the most conspicuous binding of PNL was seen in the ependymal cells and on the surface of plexus epithelial cells; in the anterior part of the pituitary gland a selective population was PNL positive. Intracytoplasmic Con A acceptors could be demonstrated in neurons, in ependymal cells, and in plexus epithelial cells. Intracytoplasmic Con A receptors were finely granular in astrocytes, oligodendrocytes, and in some cells in the pituitary gland. Ulex binding was restricted to the vascular endothelial cells and a selective population of cells in the pituitary gland. Our results suggest that lectins may be good tools for the evaluation of their respective target cells in the central nervous system and in the pituitary gland.

Ten tumours of tonsillar or epipharyngeal localization showing the histological picture of “lymphoepithelial carcinoma” (Schmincke 1921) were examined immunohistochemically using Peanut lectin, Ulex europaeus lectin-I and an antiserum to S-100 protein. The findings suggest a close relationship of this type of carcinoma to the normal tonsillar crypt epithelium. The majority of tumour cells are UEA-I-positive and PNL-negative, as is the crypt epithelium, while oral mucosa is both PNL- and to a lesser extent UEA-I-reactive. Tumour areas expressing this pattern contain a large number of asteroid-shaped PNL-positive histiocytes and arachnoid-shaped histiocytes reacting with anti-S-100 protein; both cell types being probably identical and representing typical elements of the normal tonsillar crypt epithelium. Consequently, the WHO-term “nasopharyngeal carcinoma, undifferentiated type” seems to be inadequate for this type of tumour.

The cytoskeleton comprises the three-dimensional network of cellular structures including nuclei, organelles, fibrils, and membranes. Among these, the intracytoplasmic filaments form the framework of the cytoskeleton in a narrower sense. The ubiquitous intermediate filaments (7–11 nm) whose diameters lie approximately between those of microfilaments (6 nm) and microtubules (25 nm) can be separated biochemically and immunologically into five classes. Each kind of intermediate filament polypeptide is associated with a well defined cell and tissue type. Cytokeratin polypeptides can be identified in epithelial cells, vimentin predominantly in mesenchymal cells, desmin in muscle cells, glial fibrillary acidic protein (GFAP) typically in astrocytes, and neurofilament polypeptides (NF) mainly in neurons (Table 1; Osborn and Weber 1983; Traub 1985).

Peroxidase-antiperoxidase (PAP) technique was selected for visualizing the binding of peanut lectin (PNL) to the most frequent human brain tumors. The randomly selected material included neoplasms of neuroectodermal and mesenchymal origin. We employed 1--5 micrometers of routinely processed and paraffin-embedded tissues. PNL receptors were detected to a variable extent on the cell surface of astrocytoma, glioblastoma multiforme, oligodendroglioma, ependymoma, meningotheliomatous meningeoma, and plexus papilloma. In gliomas as increase in malignancy seems to be associated with a decrease in PNL binding. Except for the plexus papillomas, neuraminidase pretreatment had neither a qualitative nor a quantitative influence on the binding behavior of PNL. Intracellular PNL receptors could be detected in "granular cells" and in the perinuclear region of malignant gliomas.

A sensitive immunoenzymatic double-staining technique is presented for the simultaneous visualization of lectin-binding sites and antigenic structures detected by monoclonal antibodies. The lectin is demonstrated by an extended unlabeled peroxidase-antiperoxidase (PAP) technique and the monoclonal antibody by an alkaline phosphatase-antialkaline phosphatase (APAAP) method, which corresponds to the standard PAP technique. 3-amino-9-ethylcarbazole (AEC) and fast blue BB salt serve as substrates for the peroxidase and the alkaline phosphatase, respectively. The antisera and the enzyme complexes raised in different animals enable the performance of three parallel incubation steps. The staining procedure requires three and a half hours altogether. This method proved to be highly discriminative and rather insensitive to interference by various artifacts.

Concanavalin A (Con A) acceptors have been demonstrated in large differentiated neurons in a previous paper. In order to elucidate the correlation between Con A binding in normal and neoplastic neurons and lectin binding dependence upon the differentiation grade, 26 tumours of the neuronal series were examined using formalin fixed and paraffin embedded biopsy specimen. The neoplasms included 3 gangliocytomas, 7 gangliogliomas, 1 central neuroblastoma, 11 medulloblastomas, 2 retinoblastomas, and 2 sympathicoblastomas. Well differentiated neurons in gangliocytomas and gangliogliomas expressed a high intracytoplasmic Con A acceptor density comparable to the feature in large non-neoplastic neurons. Less differentiated neurons and neuroblasts showed a weak perinuclear fine granular binding or an absolute lack of binding molecules, respectively. Our results suggest that in a variety of tumours, Concanavalin A receptor density in neurons depends upon the degree of differentiation of the cell. Well differentiated cells have a higher density than poorly differentiated neoplastic neurons.

Twenty non-neoplastic parotid glands (removed during neck dissection for regional tumours) were examined for cellular and subcellular binding sites of Ulex europaeus-lectin I (UEA-I), a lectin reported to be specific for alpha-L-fucose. For light microscopy, an extended peroxidase-antiperoxidase method was applied; for the evaluation of the subcellular localization of bound lectin, three of these glands were examined following immunocryoultramicrotomy and staining by the protein A-gold technique. In addition to the known cytoplasmic affinity of UEA-I for capillary endothelium, acinar cells bound the lectin within the cytoplasmic compartment; the number and distribution of stained acinar cells varied among individuals. Furthermore, cytomembrane-bound labelling that occurred most markedly at the luminar surface was observed in striated-duct epithelium. Using the electron microscope, protein A-gold particles were seen in zymogen granules and in Golgi cisternae of serous acinar cells; primary saliva secreted in the lumina exhibited strong labelling; serous acinar cells had binding sites on their cell membranes, striated-duct epithelium had binding sites on its surface membrane and in the vicinity of apical vesicles. Our results show that UEA-I is a useful tool for the study of the structure and functional states of the parotid gland epithelium and its associated pathological alterations.

Regulatory subunit free of cAMP and catalytic subunit has been prepared from type I and type II cAMP-dependent protein kinase. The purified regulatory subunit bound cAMP with high affinity. Activity of the catalytic subunit was inhibited by stoichiometric concentrations of each regulatory subunit. Kinetic experiments with each regulatory subunit indicated that type I and type II protein kinase may be activated by a different mechanism. Gel filtration of purified cGMP-dependent protein kinase from bovine cardiac muscle in the presence of cGMP did not result in dissociation of this enzyme into smaller subunits.

Monometric DNA impulse cytophotometry (ICP) is a valuable tool in obtaining data on tumor kinetics, such as the ploidy state (diploid versus aneuploid stemline, polyploidy) and the proportion of DNA synthesizing cells (S-phase rate), by analyzing about 104–105 tumor cells (for review see GOERTTLER and STOHR 1982). This method has been applied in neuro-oncology, too, in order to obtain information on various types of brain tumor regarding diagnosis, degree of malignancy, and prognosis (1, 5, 6).

Tumor localization with monoclonal antibodies (MAbs) is fairly successful when MAbs are recognizing glycoconjugates, epitopes being variably associated with different types of surface molecules and with high molecular weight secretory products, especially mucins. It appears that extracellular pools of antigen in tumor tissue with remnants of epithelial organization offer particularly favourable conditions for MAb access. The problem of access has been adressed by using a new monoclonal anti-mucin antibody (RA-96), which shows excellent targeting properties to tissular deposits, although the target epitope is not detectable at the surface of tumor cells.

We describe two novel monoclonal antibodies specific for glial filament protein (GFP), i.e., GF12.23 and GF12.24 (both IgG2a]. These cross-react over a broad range of species with epitopes located in the alpha-helical rod domain typical of all intermediate filament (IF) proteins. These monoclonal antibodies were used, in conjunction with other monoclonal GFP antibodies, rabbit antiserum to GFP, and various antibodies to other cytoskeletal proteins, to examine the occurrence of GFP in cells outside of the central nervous system of rodents, cows, and humans. We detected some scattered GFP-containing cells in the neural sheaths in some species but not in others, and we obtained different results when comparing the rabbit antisera with the monoclonal GFP antibodies. In the enteric glia of rats, we observed GFP-positive cells with all of the antibodies used, whereas in human intestine, the various monoclonal antibodies showed no reaction with any intestinal cells. Similarly, no GFP was detected in surface cells of the lens of cows and rats using any of the GFP antibodies, whereas some reaction was seen in murine lens tissue. We were also unable to detect GFP-positive cells in human, bovine, or rat liver with any of the monoclonal antibodies, which is in contrast to the reactivity of the rabbit GFP antisera with some stellate perisinusoidal cells of rat but not bovine or human liver. The possible reasons for the discrepancies between the different species and the different antibody preparations used are discussed. In addition, using double-label immunofluorescence microscopy, we showed that normal human parotid glands contain a certain type of epithelial cell that co-expresses cytokeratins and desmosomal proteins with GFP. The histological distribution of these GFP-positive cells suggests that they represent a subset of the myoepithelial cells present in this tissue. Cells co-expressing cytokeratins and GFP - in some cases, apparently together with vimentin as the third IF protein present - were also identified in tumors derived from this salivary-gland epithelium, i.e., pleomorphic adenomas, in which GFP-positive cells were relatively frequent in the myxoid and chondroid components, thus confirming the work of other investigators. Possible implications for the concept of histogenesis of these tumor cells are discussed, as are possible mechanisms resulting in the co-expression of IF proteins.

We have correlated the light-microscopic features in the unmyelinated hypothalamus with gonadal atrophy in 15 women of 30-111 years of age and in 7 men between 29 and 82 years. In the postmenstrual cases there is a distinct concordance of gonadal atrophy and the manifestation of nucleolar changes (augmentation, multiplication and vacuolization) in many nerve cells of the arcuate and subventricular nuclei. In younger, still fertile women this nucleolar finding was seen only rarely and sporadically, was limited to the arcuate nucleus and was absent in the subventricular nucleus. We interpret this nucleolar finding as a feedback effect. In man, too, this coincidence is obvious with age and in gonadal atrophy, with fewer nucleolar changes in old age than are seen in women. This difference is probably caused by a more rapid drop of the estradiol than of the testosterone level. - The hypertrophy of the subventricular nucleus (Sheehan and Kovacs) was also observed in our postmenstrual cases.

The distribution of selective cytokeratin polypeptides, vimentin, and glial fibrillary protein (GFP) in 5 human cystadenolymphomas of the parotid gland was compared with normal human parotid (n = 5) and submandibular (n = 4) glands using a panel of monoclonal antibodies against diverse and selective cytokeratin polypeptides, vimentin and glial fibrillary protein (GFP). A biotinstreptavidin method was used on cryostat sections. The immunocytochemical finding of identical cytokeratin polypeptides Nos. 7, 8, 18 and 19 and basal cells selectively labeled by the monoclonal antibody KS 8.58, in both the epithelial part of the cystadenolymphomas and in the duct epithelium of the parotid gland, confirms the hypothesis that the epithelial compartment of cystadenolymphomas is derived from the duct system. The triple expression of cytokeratin, vimentin and GFP in myoepithelial cells of the parotid gland is discussed.

The differential diagnosis of sarcomas is difficult. Different sarcomas can exhibit almost identical histological features when examined with light microscopic techniques. The visualization of structural and functional proteins in the tumor cells, using immunohistochemical methods, is an important factor in establishing a precise diagnosis. However, it is imperative to be familiar with the reaction range of the antibodies of the different tumors. The literature does not provide uniform findings with regard to the reaction range of “histiocytic” markers and thus far, there are no systematic investigations for ferritin and transferrin when applied to soft tissue tumors. This is why, in addition to the literature, we rely on our investigations carried out on 73 soft tissue sarcomas and 16 benign soft tissue lesions. Antisera have been used against the following proteins: desmin (Euro Diagnostics), myoglobin, lysozyme, ax-antitrypsin, ax-antichymotrypsin, ferritin, transferrin, factor VHI-related antigen, and S-100 protein (all from Dako). Myosin was identified on frozen sections of selected tumors with monoclonal antibodies.