Collagenase in the immunohistochemical demonstration of laminin, fibronectin and factor VIII/RAg in nervous tissue after fixation

Disappearance of cerebrovascular laminin immunoreactivity as related to the maturation of astroglia in rat brain

The present paper provides novel findings on the temporo-spatial correlation of perivascular laminin immunoreactivity with the early postnatal astrocyte development. The cerebrovascular laminin immunoreactivity gradually disappears during development. The fusion of the glial and vascular basal laminae during development makes the laminin epitopes inaccessible for antibody molecules (Krum et al., 1991, Exp Neurol 111:151). The fusion is supposed to correlate with the maturation of the glio-vascular connections. Glial development was followed by immunostaining for GFAP (glial fibrillary acidic protein), S100 protein, glutamine synthetase as glial markers and for nestin to visualize the immature glial structures. Our investigation focused on the period from postnatal day (P)2 to P16, on the dorso-parietal pallium. In the wall of the telencephalon the laminin immunoreactivity disappeared between P5 and P10; in subcortical structures it persisted to P12 or even to P16. Its disappearance overlapped the period when GFAP-immunopositive astrocytes were taking the place of radial glia. Despite the parallel time courses, however, the spatial patterns of the two processes were just the opposite: disappearance of the laminin immunoreactivity progressed from the middle zone whereas the appearance of GFAP from the pial surface and the corpus callosum. Rather, the regression of the vascular laminin immunoreactivity followed the progression of the immunoreactivities of glutamine synthetase and S100 protein. Therefore, the regression really correlates with a 'maturation' of astrocytes which, however, affects other astrocyte functions rather than cytoskeleton.

Postmortem immunohistochemical alterations following cerebral lesions: A possible pathohistological importance of the β-dystroglycan immunoreactivity

The frequency of cerebrovascular injuries raises the importance of their immunohistological investigation in postmortem materials. Most injuries involve the impairment of the blood-brain barrier. The barrier is maintained by the glio-vascular connections which break up following injuries. Some immunohistochemical alterations may refer to the impairment of the gliovascular connections. Laminin and the components of the dystroglycan complex show characteristic immunohistochemical alterations following various experimental injuries (stab wound, cryogenic lesion, arterial occlusions): immunoreactivity of β-dystroglycan, α-dystrobrevin and aquaporin 4 disappeared while that of utrophin and laminin appeared along the vessels, whereas α-syntrophin visualized the reactive astrocytes but not the resting ones. The aims of the present study were to investigate whether these post-lesion alterations: (i) are reproducible with immersive fixation, which is used in postmortem histology; (ii) are resistant to a postmortem delay before fixation; and (iii) are to be attributed to a direct effect of the lesion, or are mediated by processes occurring only in the living brain. Three models were investigated: (i) following lesions, some brains were fixed by transcardial perfusion, others by immersion; (ii) following lesions, the animals were decapitated and stored at room temperature for 8 or 16 h before fixation; and (iii) the lesions were performed after decapitation. Cryogenic lesions were performed by applying a dry ice cooled copper rod to the brain surface of ketamine-xylazine anesthetized rats. The immunohistochemical reactions were performed on free-floating sections cut with vibratome. Both immunoperoxidase and immunofluorescence methods were used. The fixation method - perfusive or immersive - did not change the post-lesion phenomena investigated. The postmortem delay did not influence the β-dystroglycan immunoreactivity, that is its lack delineated the area of the lesion. However, in the case of the other substances, various lengths of postmortem delay rendered the immunohistochemistry uninterpretable. The results suggest β-dystroglycan immunostaining could be applied in the neuropathology to detect cerebrovascular impairments.

Correlation Between Extravasation and Alterations of Cerebrovascular Laminin and β-Dystroglycan Immunoreactivity Following Cryogenic Lesions in Rats

The blood-brain barrier becomes "leaky" following lesions. Former studies revealed that following lesions the immunoreactivity of cerebrovascular laminin becomes detectable whereas that of β-dystroglycan disappears. These alterations may be indicators of glio-vascular decoupling that may result in the impairment of the blood-brain-barrier. This study investigates correlation between the post-lesion extravasation and the above-mentioned immunohistochemical alterations. Following cryogenic lesions, the survival periods lasted 5, 10, 30 minutes, 1 or 12 hours, or 1 day. Some brains were fixed immediately post-lesion. Immunofluorescent reactions were performed in floating sections. The extravasation was detected with immunostaining for plasma fibronectin and rat immunoglobulins. When the survival period was 30 minutes or longer, the area of extravasation corresponded to the area of altered laminin and β-dystroglycan immunoreactivities. Following immediate fixation some laminin immunoreactivity was already detected. The extravasation seemed to precede this early appearance of laminin immunoreactivity. The β-dystroglycan immunoreactivity disappeared later. When the extravasation spread into the corpus callosum, vascular laminin immunoreactivity appeared but the β-dystroglycan immunoreactivity persisted. It seems that extravasation separates the glial and vascular basal laminae, which results in the appearance of laminin immunoreactivity. The disappearance of β-dystroglycan immunoreactivity is neither a condition nor an inevitable consequence of the 2 other phenomena.

Immunohistochemical detectability of cerebrovascular utrophin depends on the condition of basal lamina

Utrophin is an autosomal homologue of dystrophin. Dystrophin is a member of the dystrophin-glycoprotein complex, which is a cell surface receptor for basal lamina components. In recent opinions utrophin occurs in the cerebrovascular endothelium but not in the perivascular glia. Cerebrovascular laminin immunoreactivity can only be detected in the subpial segments of the vessels, in circumventricular organs lacking blood-brain barrier, in immature vessels and following brain lesions. In our former experience utrophin immunoreactivity showed similar phenomena to that of laminin. The present study investigates the parallel occurrence of vascular utrophin and laminin immunoreactivity in the brain tissue, especially in the circumventricular organs, and during the parallel postnatal regression of both utrophin and laminin immunoreactivity. Their cerebrovascular immunoreactivity observed in frozen sections renders plausible the role of hidden but explorable epitopes, instead of a real absence of laminin and utrophin. The laminin epitopes are supposed to be hidden due to the fusion of the glial (i.e. brain parenchymal) and vascular basal laminae (Krum et al., Exp. Neurol. 111 (1991) 151). In all cases including its post-lesion re-appearance published formerly by us, laminin immunoreactivity may be attributed to the separation of glial and vascular basal laminae. Utrophin is localized, however, intracellularly, therefore a more complex molecular mechanism is to be assumed and it remains to be investigated how structural changes of the basal lamina may indirectly affect the immunoreactivity of utrophin. The results indicate that immunoreactivity may be influenced not only by the presence or absence of macromolecules but also by their functional state.

Immunohistochemical study of the extracellular matrix proteins laminin, fibronectin and type IV collagen in secretory meningiomas

The extracellular matrix plays a pivotal role in numerous cellular functions during normal and pathological processes. Secretory meningiomas are rare histological meningioma subtypes that have benign behavior, are highly vascularized and are frequently accompanied by massive peritumoral edema. The aim of this study was to assess in secretory meningiomas the immunohistochemical expression of laminin, fibronectin and type IV collagen, proteins found in the extracellular matrix. Extracellular matrix proteins were evaluated in samples from six secretory meningiomas using a semiquantitative scale ranging from not detected (0) to marked (3). Laminin expression was not detected in two cases, but was minimal in one, moderate in one and marked in the remaining cases. Fibronectin expression was absent in two cases, minimal in two, moderate in one and marked with generalized distribution in the remaining case. Type IV collagen expression was minimal in three cases, moderate in two and marked with generalized distribution in the remaining case. Our results are indicative of significant neoangiogenic activity. Meningiomas increase in size through increased production of extracellular matrix; furthermore, the proliferation of cells typically associated with neoplasia requires considerable interaction with the extracellular matrix.

Acute changes of nidogen immunoreactivity in the basal lamina of the spinal cord vessels following dorsal hemisection without correlative changes of nidogen gene expression

Previous studies have revealed that the immunoreactivities of basal lamina components in the central nervous system (CNS) vasculature change after lesions. The purpose of the present study was to determine the acute response of nidogen immunoreactivity (Nd-ir) in blood vessels of the spinal cord following a dorsal hemisection. Nd-ir immunoreactivity in the vascular basal lamina was dramatically increased within 24 h of injury, and returned to basal level after 1 week. This temporal profile of the Nd-ir change was almost the same as that of laminin. However, reverse transcription-polymerase chain reaction (RT-PCR) and Western blot analysis revealed that there was no significant increase in nidogen expression after the injury. These findings indicate that changes of antibody accessibility or epitope structure, but not a change in the expression of nidogen, may be responsible for the temporal change of Nd-ir in blood vessels following the spinal cord lesion.

Pepsin pretreatment allows collagen IV immunostaining of blood vessels in adult mouse brain

While the brain vasculature can be imaged with many methods, immunohistochemistry has distinct advantages due to its simplicity and applicability to archival tissue. However, immunohistochemical staining of the murine brain vasculature in aldehyde fixed tissue has proven elusive and inconsistent using current protocols. Here we investigated whether antigen retrieval methods could improve vascular staining in the adult mouse brain. We found that pepsin digestion prior to immunostaining unmasked widespread collagen IV staining of the cerebrovasculature in the adult mouse brain. Pepsin treatment also unmasked widespread vascular staining with laminin, but only marginally improved isolectin B4 staining and did not enhance vascular staining with fibronectin, perlecan or CD146. Collagen IV immunoperoxidase staining was easily combined with cresyl violet counterstaining making it suitable for stereological analyses of both vascular and neuronal parameters in the same tissue section. This method should be widely applicable for labeling the brain vasculature of the mouse in aldehyde fixed tissue from both normal and pathological states.

Disappearance of the post-lesional laminin immunopositivity of brain vessels is parallel with the formation of gliovascular junctions and common basal lamina. A double-labelling immunohistochemical study

Previous studies revealed that during development the laminin immunopositivity gradually disappeared from the brain vessels, but temporarily re-appeared in them around lesions. The question of the present study was the correlation between the post-lesional vascular immunopositivity to laminin and the glial reaction. Following stab wounds, double fluorescent immunohistochemical labelling was performed against laminin (using a polyclonal antiserum against laminin 1) and glial fibrillary acidic protein. A number of vessels exhibited intense immunopositivity to laminin within the lesioned tissue. Where these laminin immunopositive vessels entered the perilesional brain substance, the astroglia formed contacts on them, and the separate vascular and glial basal laminae fused. The disappearance of the post-lesional laminin immunopositivity seemed to coincide with these phenomena. When monoclonal antibodies were applied against the beta1 and gamma1 laminin chains, vessels proved to be immunopositive at the lesion, but none in the intact brain tissue. No immunoreactivity was detected in the cases of alpha2 and beta2 chains. The results suggest that the disappearance of laminin immunopositivity may be attributed to that the epitopes become inaccessible for antibodies owing to the formation of gliovascular junctions and common basal lamina between astroglia and vessel. The possible role of an alteration in the laminin composition and the effect of fixation are discussed.

Role of extracellular matrix in tumor invasion: migration of glioma cells along fibronectin-positive mesenchymal cell processes

OBJECTIVE

The major morbidity of glioma lies in its infiltrative growth. One of the major patterns of this invasive growth is the formation of Scherer's secondary structures associated with the blood vessels and the leptomeninges. To better understand the role of extracellular matrix (ECM) in glioma invasion, we investigated in vitro the interaction between glioma cells and the meningeal mesenchymal tissue from the brain. As an aid to this study, ECM in glioma cell line spheroids was compared with that in primary fetal brain aggregates.

METHODS

To study the expression of ECM, four glioma cell lines (U-87 MG, U-251 MG, AN1/lac-z, and HF-66) and primary cells from fetal rat brain were grown as spheroids and monolayers. To sudy the role of ECM in glioma invasion, spheroids from the glioma cell lines were grown over established cultures of fetal meningeal and mesenchymal tissue. Expression of fibronectin, laminin, tenascin, collagen VI, and chondroitin sulfate proteoglycan was studied by immunofluorescence.

RESULTS

Expression of ECM by the spheroids was variable. U-87 MG expressed most of the ECM components robustly, whereas AN1/lac-z expressed them all weakly. Fetal rat brain aggregates produced minimal ECM. In cocultures of glioma spheroids and fetal meningeal mesenchymal tissue, individual cells from the glioma spheroids that expressed least fibronectin (AN1/lac-z and U-251 MG) migrated along the fibronectin-positive mesenchymal cells in the culture dish. Cells from the other two lines (U-87 MG and HF-66) that expressed fibronectin strongly did not demonstrate such behavior. None of the other ECM components showed a similar association; mesenchymal cells did not express laminin as strongly as fibronectin, and glioma cells were not observed to align with the laminin-positive structures.

CONCLUSION

This study suggests that fibronectin may play a key role in intracerebral invasion of glioma cells.

Expression of angiogenic factors and structural proteins in central nervous system vascular malformations

Little is known of the molecular mechanisms mediating the genesis and subsequent biological behavior of central nervous system vascular malformations. The role of angiogenic and permeability-inducing factors in the pathogenesis of these lesions has not bee previously explored. In this study, we subject specimens from 12 cases of excised vascular malformation to a battery of immunostaining for vascular endothelial growth factor, basic fibroblast growth factor, and selected structural and matrix proteins. The lesions consisted of seven arteriovenous malformations (AVMs), including one angiographically occult AVM, one arterialized vein from a dural AVM, and five cavernous malformations (CMs). Vascular endothelial growth factor was expressed by all lesions and was localized predominantly in the subendothelial layer and in perivascular spaces. Four of seven AVMs and four of five CMs demonstrated faint basic fibroblast growth factor expression that was localized to the media of AVM vessels and the subendothelial layer and intercavernous matrix of CMs. This pattern of angiogenic factor immunostaining was correlated with the expression of structural and matrix proteins in the same lesions. Laminin was not expressed in any of the CMs, confirming previous reports from our laboratory. By contrast, fibronectin expression was more prominent in CMs than in AVMs. Collagen Type IV and alpha smooth muscle actin expression occurred in every lesion. We conclude that angiogenic growth factors are expressed in all types of vascular malformations of the central nervous system. The pattern of expression suggests diffuse activation of angiogenesis without specific relation to individual vessel types or recent clinical behavior. Defining the role of angiogenesis in vascular malformations might provide insight into their pathogenesis and suggest novel strategies for modification of their behavior.

Expression of tenascin in human gliomas: its relation to histological malignancy, tumor dedifferentiation and angiogenesis

The immunohistochemical distribution of tenascin (TN), fibronectin (FN), and laminin (LN) was investigated in 56 human gliomas (8 astrocytomas, 15 anaplastic astrocytomas, and 33 glioblastomas) with regards to the histological degree of malignancy and the degree of tumor cell differentiation evaluated by the staining of glial fibrillary acidic protein (GFAP). In 8 anaplastic astrocytomas and 28 glioblastomas, TN was predominantly immunolocalized in the basement membrane zone of the proliferating tumor vessels; sections of all astrocytomas were negative for TN staining. FN was localized in the basement membrane zone of the vessels in all astrocytomas, 12 anaplastic astrocytomas, and 22 glioblastomas. In 7 anaplastic astrocytomas and 19 glioblastomas, both TN and FN were expressed to various degrees in the tumor vessels. However, most of the TN-positive vessels did not express FN, and most of the FN-positive vessels were negative for TN staining. Furthermore, in 6 anaplastic astrocytomas and 12 glioblastomas, either TN or FN, but not both, were expressed in any area on serial sections. Most of the tumor cells around TN-positive, FN-negative tumor vessels did not express GFAP. On the other hand, GFAP was present in most tumor cells around TN-negative, FN-positive vessels. LN was detected in all vascular and pial-glial basement membrane zone of the tissues examined. These findings indicate that the degree of histological malignancy and the degree of cell dedifferentiation of human gliomas correlate well with the expression of TN, but are inversely correlated with the expression of FN.(ABSTRACT TRUNCATED AT 250 WORDS)

Localization of fibronectin in the frog skin

In this review, we present structural and ultrastructural localizations of fibronectin (FN) in the larval and adult skin of the frog (Rana esculenta) either in in vivo or in in vitro conditions. The ventral skin of the tadpole contains membrane-associated FN-plaques disposed around the epidermal and dermal cells during their climactic rearrangement. Moreover, lines of fibrillar FN are detected inside the breaks opened in the derived collagen. The ventral skin of the adult frog reveals FN distributed in the three superimposed tissues forming the skin, i.e. the epidermis, the dermis and the subcutaneous tissue. In vivo, the epidermis is devoid of FN except for the mitochondria-rich cells (MRCs) which contain FN cytoplasmic granules. The dermis reveals two distinct collagenous networks showing FN localizations. A vertically-oriented network formed by thick tracts contains axis of fibrillar FN connecting the upper dermis devoid of FN to the FN-rich subcutaneous tissue. In contiguity with an horizontally-oriented network comprises thin tracts formed by clear spaces separating the superimposed collagen bundles of the dermal stratum compactum. These tracts contain aligned FN-granules. Inside the thick and thin tracts, the dermal and pigment cells present membrane-associated In vitro (in organ culture conditions) MRCs of the epidermis maintain their FN localization and, in addition, the stratum germinativum cells show cytoplasmic FN granules. Epidermal cells, in the vicinity of the cut edges of the cultivated skin fragment, modify their shape and acquire membrane-associated FN-plaques located between desmosomes. The FN localizations in these two collagenous networks of the dermis remain unchanged. In the same way, the FN-rich subcutaneous tissue is unmodified. In summary, the FN distribution in the larval skin is related to the cell rearrangement during the metamorphic climax, and, in the adult skin to the cell migration during the wound healing process and the pigment cell patterning. The cell migration is demonstrated, in organ culture conditions, by antiFN serum used as an experimental tool. FN is an important substrate used in the dermal breaks of the larval skin, and in the dermal tracts of the adult skin, both allowing the dermal and pigment cell migration.

Species diversity of neuronectin and cytotactin expression patterns in the vertebrate central nervous system

Two extracellular matrix proteins of brain tissue, neuronectin (NEC1) and cytotactin (CT), are disulfide-bonded multimers of M(r) 180,000-250,000 subunits. The previously known distribution of these molecules is, however, very different. Human NEC1 is found throughout the white matter of rostral segments of the adult central nervous system (CNS) but not in rostral gray matter or in caudal CNS segments, including the cerebellum. In contrast, CT is absent or expressed at a low level in the adult chicken cerebrum but highly expressed in the cerebellum. Despite these differences in distribution, results obtained with antibodies that recognize NEC1 and CT in several vertebrate species indicate that these molecules are identical or at least closely related: (1) alpha NEC1 antibodies recognize proteins affinity-purified with CT-binding proteoglycan; (2) proteins recognized by alpha NEC1 and alpha CT antibodies in cells constitutively expressing the molecules, cells in which expression is induced by growth factors and phorbol ester and cells treated with tunicamycin (to block glycosylation) are identical in subunit composition and mobility on SDS gels; (3) the removal of NEC1 from culture supernatants by immunoprecipitation removes all molecules reactive with alpha CT antibodies and vice versa; (4) immunoblots of brain extracts with alpha NEC1 and alpha CT antibodies yield identical results. Having demonstrated the structural similarity between NEC1 and CT, we reexamined their distribution in the CNS. Surprisingly, the temporal and spatial distribution pattern of NEC1/CT varied greatly among species. Immunohistochemical and immunoblot experiments with adult human CNS tissues revealed significant levels of NEC1/CT in rostral but not caudal segments. In contrast, in cows and pigs the molecule is found throughout the CNS. Adult rat and mouse brains show regionally restricted expression of NEC1/CT in several areas of the cerebrum--distinct from those showing NEC1/CT in the human--and in the molecular layer of the cerebellum. Tests with fetal and newborn tissues revealed that CNS development in humans, cows and pigs is not accompanied by the marked decline in NEC1/CT levels or the changes in subunit composition found in the chicken CNS. The marked species diversity in temporospatial expression patterns suggests that intrinsic and/or extrinsic elements controlling the expression of NEC1/CT have diverged during vertebrate evolution.

Laminin immunohistochemistry in brain is dependent on method of tissue fixation

Normal adult and lesioned rat and mouse brains were fixed by formaldehyde perfusion by two methods that differ primarily in the length of the post-fixation period. Sections were subsequently immunostained using monoclonal and polyclonal antibodies to laminin. With relatively short post-fixation periods (up to 4 h), vascular basement membrane (BM)-laminin was immunostained, but intraneuronal laminin-like immunoreactivity was faint. With longer post-fixation periods (18-24 h), intraneuronal laminin-like immunoreactivity was distinct, while vascular BM-laminin immunoreactivity was reduced drastically. These findings are particularly relevant to studies examining laminin immunoreactive blood vessels in response to lesions, especially ischemic stroke. In fact, the present results suggest that the apparent neovascularization or up-regulation of vascular BM-laminin following CNS injury likely relates to differences in regional tissue fixation.

Increased reactivity of laminin in the basement membranes of capillary walls in AIDS brain cortex

To verify how the components of the capillary wall are modified in the course of AIDS we studied the brain cortex from nine cases with AIDS. Cellular and extracellular components were delineated using antibodies for laminin and collagen IV for basement membranes and glial fibrillary acidic protein for astrocyte foot processes. We found a marked increase in reactivity for laminin in the basement membranes of capillary walls and hypertrophy and hyperplasia of astrocyte foot processes around vessels, when compared to control cortical tissue. We suggest that modifications of brain capillary wall may have a role in the pathogenesis of neurological disfunction in AIDS.

Immunocytochemical study of macrophages and microglial cells and extracellular matrix components in human CNS disease. 1. Gliomas

We have used an immunocytochemical approach to investigate the inter-relationships between astrocytes, macrophages and microglia and the extracellular matrix components fibronectin and laminin, in 27 gliomas. Using recently available markers to macrophages and microglia (monoclonals Mac387, KP1 and the lectin RCA-1) and antisera to GFAP, fibronectin and laminin, we have described the reactions of these cells and the extracellular matrix. We found RCA-1 to be the superior marker for detecting most macrophages and microglia. There were more macrophages and microglia in high-grade gliomas than in low-grade. RCA-1 also reacted with endothelial cells in normal and reactive brain but showed markedly less affinity for endothelium in an close to gliomas. A possible role for the extracellular matrix in the astrocyte, macrophage and microglial reactions is discussed in the broader context of their role in the immune response.

Immunohistochemical localization of the matrix glycoproteins--tenascin and the ED-sequence-containing form of cellular fibronectin--in human permanent teeth and periodontal ligament

The expression of two matrix glycoproteins, tenascin and cellular fibronectin (cFN), has been studied in fully developed human permanent teeth, periodontal ligament, and alveolar bone, in both frozen and paraffin-processed material. Polyclonal antibodies to tenascin and a monoclonal antibody recognizing the ED sequence specific to at least some forms of cFN were used. Staining for both tenascin and cFN was positive in the dental pulp, odontoblastic layer, cementoblast-pre-cementum zone, and on the periosteal as well as endosteal surfaces of the alveolar bone. In the periodontal ligament, cFN was evenly distributed, whereas tenascin was accumulated in the attachment zones. Pre-dentin stained for tenascin but not for cFN. Mineralized dentin and cementum were tenascin- and cFN-negative. The relative staining intensity for tenascin was greater than that for cFN in the cementoblast-pre-cementum layer and in the attachment zones of the periodontal ligament, whereas cFN stained more intensely in the pulp. In frozen material, antigenicities were well-preserved. Paraffin processing facilitated precise recognition of tissue morphology, but the antigenicity of cFN was lost. The co-expression of tenascin and cFN in the dental pulp, cementogenic zone, and on the surfaces of the alveolar bone may reflect the ability of the cells to deposit mineralized tissue matrices. The pronounced expression of tenascin in the interfaces between mineralized and non-mineralized tissues suggests that it is functionally associated with mechanical stress and may thus have at least two distinct functions. The relative amounts of the two matrix glycoproteins may contribute to regulation of tissue structure.

Distribution of collagen type IV in brain tumors: an immunohistochemical study

One hundred-twenty seven human brain tumors were examined by an immunoperoxidase technique for the expression of collagen Type IV, a major constituent of basement membrane. The parenchymal components were negative for the marker protein in all tumors except for neurilemmomas which were positively stained. In every case, the antibody to collagen Type IV showed distinct staining of the vascular pattern. In gliomas, capillaries increased in number and the vascular staining increased in intensity. Fine branching capillaries and endothelial glomeruloid proliferations characteristic each of oligodendrogliomas and glioblastomas could be distinctly illustrated. In two ependymomas, marked capillary proliferation was noted in periventricular areas. Fibrillar staining was observed between the tumor cells in seven of 34 meningiomas. Pericapillary lamellar deposition of collagen Type IV suggests a vascular origin of psammoma bodies. In some malignant tumors, pial-glial membranes were disrupted and the Virchow-Robin spaces were filled with malignant cells. Collagen Type IV was absent around the stromal cells of hemangioblastomas, suggesting that these stromal cells were unrelated histogenetically with endothelial cells. Collage Type IV may be useful in the differential diagnosis between meningiomas and neurilemmomas.

The vascular response to tumor infiltration in malignant gliomas. Morphometric and reconstruction study

Neo-vascularization and endothelial hyperplasia have been shown to be very active in malignant gliomas. In this contribution the vascularization of the cortex infiltrated by malignant gliomas is morphometrically studied and the endothelial proliferations are immunohistochemically investigated and reconstructed by a three-dimensional computer-assisted procedure. Vessel density increases after tumor infiltration in some cases only. The diameter of vessels increases and so does the number of nuclei/vessel after the complete invasion of the cortex when vascular glomeruli develop. In completely infiltrated cortex with development of glomeruli and circumscribed necroses, vessel density is very low. No neoformation of vessels takes place before the complete infiltration of the cortex by the tumor. The hyperplastic formations, usually arranged parallel to the deep or outer cortical layers, take origin from the radially penetrating vessels from the meninges and their lateral branching. The hyperplasia deforms the vascular network, making it often inadequate to supply tumor cells. Immunohistochemically, the cells composing the hyperplastic structures are variably positive for factor VIII/RAg and, at a lesser extent, for alpha-smooth muscle actin. The poorness of the vascular network in many instances of completely infiltrated cortex is responsible for the development of circumscribed necroses.

The extracellular matrix of the central and peripheral nervous systems: structure and function

The extracellular matrix (ECM) is the naturally occurring substrate upon which cells migrate, proliferate, and differentiate. The ECM functions as a biological adhesive that maintains the normal cytoarchitecture of different tissues and defines the key spatial relationships among dissimilar cell types. A loss of coordination and an alteration in the interactions between mesenchymal cells and epithelial cells separated by an ECM are thought to be fundamental steps in the development and progression of cancer. Although a substantial body of knowledge has been accumulated concerning the role of the ECM in most other tissues, much less is known of the structure and function of the ECM in the nervous system. Recent experiments in mammalian systems have shown that an increased knowledge of the ECM in the nervous system can lead to a better understanding of complex neurobiological processes under developmental, normal, and pathological conditions. This review focuses on the structure and function of the ECM in the peripheral and central nervous systems and on the importance of ECM macromolecules in axonal regeneration, cerebral edema, and cerebral neoplasia.

Neovascularization of pericellular fibrosis in alcoholic liver disease

The relationship between sinusoidal capillarization and pericellular fibrosis was studied in 29 specimens of human liver from patients with alcoholic liver disease. Immunohistochemically, factor VIII-related antigen was not observed in the normal sinusoidal lining cells, but was localized in the capillaries which proliferated in the pericellular fibrotic region. Fibronectin was localized in the proliferated endoplasmic reticulum of newly formed vascular endothelial cells and hepatocytes with the development of pericellular fibrosis, laminin became apparent in the plasmalemmal vesicles of capillary endothelial cells and pericytes and in the endoplasmic reticulum of mesenchymal cells in Disse's space. From these results, it was revealed that pericellular fibrosis was closely related to sinusoidal capillarization. Furthermore, it is suggested that this sinusoidal capillarization may be caused by neovascularization, and that the extracellular matrix produced by endothelial cells, hepatocytes and mesenchymal cells may promote this process.

Intraneuronal laminin-like molecule in the central nervous system: demonstration of its unique differential distribution

Laminin, an extracellular matrix glycoprotein rich in basement membrane, is a multifunctional molecule of approximately 1000 kDa and is known to possess a potent neurotrophic activity. Laminin-like immunoreactivity (LLI) was for the first time demonstrated in mouse and rat CNS neurons by a sensitive immunohistochemical technique. Transblotting of SDS-PAGE of the supernatant of the mouse and rat brain homogenate identified distinct 180 kDa and weak 380 kDa bands immunoreactive to anti-laminin and these molecules differed from authentic laminin subunits. The intraneuronal distribution of LLI disclosed two distinct patterns; LLI-1 (diffuse perikaryal stain) and LLI-2 (coarse granular stain). By immunoelectron microscopy, LLI was localized to the ERs in LLI-1 neurons, whereas it appeared to be confined to lysosomes in LLI-2 neurons. LLI-1 neurons were found predominantly in hippocampal pyramidal, granule and neocortical layers 1-3, 6 neurons, in most of the striatal and thalamic neurons, and Purkinje cells. The majority of neurons in neocortical layers 4-5, medial septal and Meynert neurons, somatic motor neurons, and neurons of the deep cerebellar nuclei were classified as LLI-2 cells. No LLI was found in hypothalamic mammillary, habenular and vagal dorsal motor neurons (LLI-3). These observations may indicate intraneuronal production of laminin-related molecules in central neurons. We speculate that the laminin-related molecules (neurolaminin) play important roles in trophic or servo mechanisms in the CNS.

Definition of an extracellular matrix protein in rostral portions of the human central nervous system

We have identified and characterized an extracellular matrix (ECM) glycoprotein of cultured astrocytomas, NEC1, that is expressed in normal human brain parenchyma. Detailed immunohistochemical analysis reveals a region-specific NEC1 pattern along the rostrocaudal axis of the central nervous system (CNS), with strong expression throughout the white matter of telencephalon and diencephalon, scant expression in some areas of mesencephalon, and no expression in pons, cerebellum, medulla, spinal cord, and peripheral nervous system. NEC1 is not co-distributed with any known neural cell type, suggesting that expression of specific ECM proteins in the CNS is segmentally controlled.

Distribution of extracellular matrix proteins in primary human brain tumours: an immunohistochemical analysis

Using immunohistochemical techniques, we localized several glycoproteins of the extracellular matrix in paraffin-embedded sections of 4 normal brain and 38 primary intracranial tumour specimens. All specimens were positively immunostained to various degrees by monoclonal antibodies to type IV collagen and procollagen III and by antisera to laminin and fibronectin. Staining was consistently most intense at sites of contact between neuroepithelial and mesenchymal or leptomeningeal elements; there was no demonstrable staining within or between neuroepithelial elements in the neuropil. Tumour cells from meningiomas and from the sarcomatous portion of a gliosarcoma were positively immunostained for fibronectin and laminin. The integrity of the glial limitans externa was demonstrated by the positive linear reaction product produced by immunostains for type IV collagen and laminin, even in the most malignant gliomas. The deposition of extracellular matrix glycoproteins at the glial-mesenchymal interface observed in this study of primary human brain tumours is a manifestation of one of the interactions between tumour and stromal cells in the central nervous system. A loss of coordination and an alteration in the interactions between epithelial cells and stromal cells across extracellular matrices such as basement membranes are thought to be fundamental steps in the development and progression of cancer. Further characterization studies focusing on other markers of the extracellular matrix are needed to elucidate completely the function of this structure in the central nervous system.

Immunocytochemical localization of factor VIII-related antigen in tumors of the human central nervous system

Using both immunohisto- and immunocytochemical techniques with periodate-lysine-paraformaldehyde (PLP) fixation, we have studied the distribution of Factor VIII-related antigen (FVIIIR:Ag) in 12 cases of tumors of the human central nervous system (CNS) and one sample of non-tumor brain tissue. FVIIIR:Ag was found both extracellularly and intracellularly. It was localized in the vascular lumen, between endothelial cells, and in the endothelial cell basement membrane. In the endothelial cell cytoplasm, FVIIIR:Ag was found in the endoplasmic reticulum, perinuclear space, and in intracytoplasmic vacuoles and vesicles. Characteristic of malignant tumors (six out of seven) was a strongly-positive dilated endoplasmic reticulum. This may reflect increased FVIIIR:Ag synthesis in the endothelial cells of malignant tumors. Only one of five benign tumors showed such staining. Six of 12 tumors and the non-tumor brain showed perinuclear FVIIIR:Ag. Both ad- and abluminal vesicles in the tumor endothelial cells contained FVIIIR:Ag suggesting that endocytosis, transcellular transport, and/or endocytosis, as well as FVIIIR:Ag synthesis occurs. The non-tumor brain showed normal capillary structure and very little FVIIIR:Ag immunoreactivity. The relationship of these FVIIIR:Ag abnormalities to the hypercoagulable state seen in some malignant brain tumor patients remains to be clarified.

An immunocytochemical method for the localization of fibronectin in Araldite-embedded specimens

The detection of fibronectin (FN) in osmium-fixed and Araldite-embedded frog skin fragments was studied using a modification of Baskin's procedure (Baskin et al. 1979). Following the removal of Araldite from the semi-thin sections (0.5-1.0 micron) with ethanol-NaOH solution, the sections were bleached with hydrogen peroxide. FN was detected by indirect immunoperoxidase method. For precise localization of FN, careful attention was paid to the temperature, antibody concentrations and the quality of the ethanol-NaOH solution. Our results were in agreement with those that we had obtained previously for polyethylene glycol (PEG) sections, suggesting that the present procedure is useful for the detection of FN in Araldite-embedded biological specimens.

Establishment and characterization of five cell lines derived from human malignant gliomas

We established and characterized five cell lines derived from human malignant gliomas (four glioblastomas multiforme and one highly anaplastic astrocytoma). All cell lines exhibited tumor cell morphology and growth kinetics, and anchorage-independent growth in soft agar. Cytogenetic analysis revealed significant aneuploidy in all five cases as well as clonal chromosomal alterations unique to each cell line. No cell line was tumorigenic in athymic mice. Two of the cell lines were sensitive to carmustine (BCNU) in monolayer and soft-agar cultures. Electron microscopy showed marked variability between cell lines in the number and structure of intracytoplasmic organelles; SF-126 formed collagen fibers in vitro. Immunohistochemical analysis of the surgical specimens showed variable expression of glial fibrillary acidic protein (GFAP) in malignant astrocytes; positive immunostaining for glycoproteins of the extracellular matrix was found predominantly in perivascular regions. In early-passage cultures, only cell line SF-295 expressed GFAP; at establishment, none of the cell lines expressed GFAF or glutamine synthetase. Fibronectin and laminin were expressed by all cell lines in early-passage culture, but expression of these glycoproteins at establishment was variable. Only SF-126 was positively identified by immunostains for procollagen III; this was also the only cell line in which DEAE-cellulose chromatography and SDS-PAGE demonstrated interstitial collagen synthesis. These well-characterized glioma-derived cell lines may now serve as useful tools with which to study the cell biology of gliomas. The synthesis of interstitial collagen by a glioma-derived cell line may suggest a derivation from vascular mesenchymal elements, either reactive or transformed, in the original heterogeneous malignant glioma, rather than from a glial precursor cell.

The K.G. McKenzie award lecture--1986. Effects of extracellular matrix proteins on the growth and differentiation of an anaplastic glioma cell line

Efforts to determine the factors responsible for reversing malignancy in the central nervous system may not only increase our understanding of the growth of primary human brain tumors, but may eventually prove to be of therapeutic benefit as well. We therefore devised a model system to study the effects of extracellular matrix (ECM) proteins on the malignant phenotype of an anaplastic glioma line, U-343 MG-A. Well-characterized cultures derived from normal human leptomeninges were grown to confluence and maintained for 2 weeks. The pia-arachnoid cells were then removed with detergent and base, leaving behind an ECM enriched in laminin, fibronectin, types I and IV collagen, and procollagen III. U-343 MG-A tumor cells planted on top of this normal ECM were profoundly growth inhibited, developed multiple slender cytoplasmic processes similar to those of normal astrocytes, and expressed more GFAP per cell than did tumor cells growing on plastic alone. The growth of U-343 MG-A tumor cells in flasks coated with purified fibronectin or laminin was not significantly inhibited. However, U-343 MG-A cultures grown in flasks coated with type I or IV collagen showed decreased cellular proliferation and altered cell morphology. Conditioned medium from U-343 MG-A tumor cells growing on plastic alone contained a 64 kD activated metalloprotease. U-343 MG-A tumor cells growing on the pia-arachnoid ECM do not demonstrate such proteolytic activity. We conclude that the tumor cell microenvironment is extremely important in modulating the growth and differentiation of an anaplastic glioma cell line.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunohistochemical study of chondroitin sulfate in human gliomas

A polyclonal rabbit antiserum was utilized to localize chondroitin sulfate in human gliomas. Tissue sections were digested with chondroitinase ABC to create the antigenic determinant on the chondroitin sulfate proteoglycan molecule. Normal CNS tissue showed a positive immunohistochemical staining both in white and gray matter, sparing the cytoplasm of glial and neuronal cells. Differentiated astrocytomas presented the same pattern as the normal CNS. Anaplastic astrocytomas and glioblastomas showed progressive reduction of parenchymal positivity as anaplasia increased. These data suggest that chondroitin sulfate is a character expressed by differentiated CNS cells and that it is lost with dedifferentiation. Vascular structures presented positive material in the adventitia in all the oncotypes. A discontinuous positivity was observed in the basal membrane zone of the vessels.

Simultaneous and successive localization of two antigens in the same tissue section using 3,3'-diaminobenzidine and 1-naphthol basic dye

We describe two procedures for the simultaneous and successive localization of two antigens in the same tissue section. In the simultaneous staining procedure, the first antigen was localized using 3,3'-diaminobenzidine (DAB), while the second antigen was stained using the 1-naphthol basic dye (1-NBD) method. The colour of the second antigen depended on the basic dye used, and no mixing of colours was observed when the two antigens were localized in different cells or structures. However, sequential double staining proved to be more convenient for the demonstration of two antigens in the same cell. In this procedure, the first antigen was stained using 1-NBD, and the interesting microscopic fields were photographed. The basic dye was then completely removed, and the second antigen was stained using DAB.

Immunolocalization of monoclonal antibody-defined extracellular matrix antigens in human brain tumors

The extracellular matrix is involved in many aspects of tumor cell biology, including tumor invasion and metastasis. 2A6 and 81C6 are murine monoclonal antibodies that identify glioma-mesenchymal extracellular matrix antigens. The 81C6 antigen is a high molecular weight glycoprotein composed of Mr 230,000 subunits. The expression of 2A6 antigen, 81C6 glycoprotein, fibronectin (FN), and laminin (LN) was examined immunohistochemically in ten malignant gliomas (MG) and four medulloblastomas (MBT). 2A6 and 81C6 were expressed in similar patterns by the neoplastic neuroepithelial cells in 9/10 MG and 1/4 MBT. The staining was typically diffuse and amorphous, without visualization of distinct cell bodies or processes. Less frequently, antigen was detected within tumor cell cytoplasm. In most tumors the staining was greatest in the perivascular regions. In two MG, 2A6 and 81C6 were expressed only by a subpopulation of neoplastic cells. Although intense staining was also associated with hyperplastic vascular and mesenchymal cells, many small and medium size blood vessels stained weakly or not at all. In contrast, FN and LN were expressed uniformly and intensely in the tumor vasculature, but were not expressed by neoplastic neuroepithelial cells. The 2A6 antigen and 81C6 glycoprotein are immunohistochemically distinct from FN and LN. These monoclonal antibody-defined antigens are heterogeneously expressed by neoplastic neuroepithelial cells and hyperplastic vascular-mesenchymal elements in MG and MBT. The 2A6 and 81C6 monoclonal antibodies will be useful reagents in the investigation of the extracellular matrix of malignant neuroepithelial neoplasms.

1-Naphthol basic dye (1-NBD). An alternative to diaminobenzidine (DAB) in immunoperoxidase techniques

The usefulness of 1-naphthol as substrate for horseradish peroxidase (HRP) in immunohistochemistry was studied using the peroxidase-antiperoxidase (PAP) and avidin-biotin-complex (ABC) methods in the demonstration of glial fibrillary acidic protein (GFAP), vimentin, carbonic anhydrase C (CA.C), and factor VIII-related antigen (FVIII/RAg) in central nervous tissue and cerebral tumors. In the presence of ammonium carbonate, 1-naphthol is oxidized by HRP and hydrogen peroxide, producing a fine gray-violet precipitate. The oxidation product of 1-naphthol proved capable of binding a great number of basic dyes. For each stain the final reaction product had a characteristic color that was different from the spontaneous color of the dye and from the color displayed by nuclei. The final color obtained with this procedure was alcohol resistant and could be mounted in solvent-based mounting media. The results obtained with the 1-naphthol basic dye (1-NBD) method were compared with those obtained using the diaminobenzidine (DAB) reaction in the demonstration of GFAP-positive astrocytes. The DAB reaction produced a more intense staining but also a coarser precipitate than the 1-NBD reaction. The 1-NBD procedure showed more morphological detail of fine structures and did not obscure nuclei and mitosis. The very low toxicity of 1-naphthol compared with DAB (a suspected carcinogen) is an important advantage of the 1-NBD method, as is its high specificity and sensitivity.

Affinity cytochemistry of vascular endothelia in brain tumors by biotinylated Ulex europaeus type I lectin (UEA I)

The vascularization of 50 tumors of the central nervous system (CNS) including 17 meningiomas, 25 neuroectodermal tumors, i.e., astrocytomas, oligodendrogliomas, mixed gliomas, glioblastomas, medulloblastomas, seven metastatic carcinomas, and one malignant hemangioendothelioma were investigated using biotinylated Ulex europaeus type I lectin (UEA I) in an indirect avidinbiotin-peroxidase procedure. The cytochemical staining pattern of UEA I on paraffin sections was compared with that of biotinylated Dolichos biflorus lectin (DBA), and with the immunocytochemical staining of factor VIII related antigen (F VIII/RAG) by polyclonal antisera using the PAP technique. UEA I visualized the endothelia of blood vessels with equal intensity, sensitivity, and reliability in normal brain and in tumor tissue with neovascularization. While large, medium, and small vessels were equally well demonstrated by UEA I and antibodies against FVIII/RAG, capillaries and endothelial sprouts were stained more consistently and intensely by UEA I. No reliable cytochemical staining could be obtained by DBA regardless of tissue or cell type investigated. It is concluded that UEA I is a highly useful cytochemical marker for the identification of vascular endothelia in paraffin sections of human brain tumors.

The distribution of laminin in human brain tumors: an immunohistochemical study

The immunohistochemical distribution of laminin in 70 cases of primary and secondary brain tumors has been studied. The sections were fixed in Carnoy, embedded in paraffin, and treated with collagenase before applying the specific immune serum. In normal neural tissue and in gliomas, the distribution of laminin overlaps with that of basement membrane, as is known from observations on ultrastructure in the literature. The overlapping was observed also in neurinomas, papillomas of the choroid plexus, ependymomas, and meningiomas. Problems arise in the interpretation of the results in the other oncotypes. In hemangiopericytomas and hemangioblastomas, laminin demonstrates a basement membrane around the vessels only; it is absent around stromal cells. In desmoplastic medulloblastomas, laminin forms a network among the tumor cells. In primary lymphomas, concentric rings of laminin are evidenced around some vessels. In discussing the results, the ultrastructural distribution of basement membranes is taken into account.

GFAP, F VIII/RAg, laminin, and fibronectin in gliosarcomas: an immunohistochemical study

GFAP, Factor VIII/RAg, laminin, and fibronectin were immunohistochemically investigated in 15 glioblastomas and 15 gliosarcomas . GFAP was found variably positive in the glial areas. F VIII/RAg characterizes the endothelial cells and in gliosarcomas suggests the origin of the sarcomatous component from the endothelial proliferations. Laminin separates the two components and characterizes the inner and the outer basement membranes in the vessels. It is multiplied and thickened in endothelial proliferations, while it is often fragmented in the larger vessel wall proliferations. Our observations confirm that gliosarcoma represents the last stage of a process which starts with the endothelial hyperplasia of glioblastoma.