Interferon-induced Mx proteins in brain tissue of multiple sclerosis patients.

BACKGROUND AND PURPOSE: Recombinant interferon-beta is proven as an effective long-term treatment in patients with multiple sclerosis (MS). Unlike in other chronic inflammatory diseases, endogenous synthesis of type I interferons (IFN-alpha and IFN-beta) has not been studied extensively in MS. Mx proteins A and B (MxA and MxB) are intracellular proteins that are induced exclusively by type I IFNs. We investigated the expression of Mx proteins in post-mortem brain tissue of IFN-beta-naïve MS patients as a marker for endogenous synthesis of type I IFNs. METHODS: By employing monoclonal antibodies specific for MxA and MxB positive staining was detectable predominantly in reactive astrocytes within the MS plaques but also in endothelial and ependymal cells as well as in lymphocytic infiltrates. RESULTS: This is of interest in view of results previously published by our group and others that Mx protein concentrations measured by ELISA increase in blood samples from MS patients after IFN-beta therapy. CONCLUSIONS: In MS, Mx proteins are detectable in plaques suggesting endogenous synthesis of type I IFNs as part of the acute inflammatory process.

The "bioartificial living nerve graft".

INTRODUCTION: Nerve tubes seeded with cultured Schwann cells have become a promising alternative to nerve autografts. However, the functional results of these bioartificial cellular grafts remain to be improved. To imitate the three-dimensional structure of peripheral nerves, we designed a Schwann cell-seeded intrinsic framework within a semipermeable biodegradable collagen nerve tube (Integra). MATERIAL AND METHODS: In 90 rats a 25 mm gap was created at the sciatic nerve of the right lower limb. In group I, the gap was treated using the "bioartificial nerve graft". In group II, the tube filled with non-seeded filaments was implanted in order to evaluate the influence of the Schwann cells on regeneration. In group III, the gap was bridged using an autologous nerve graft. For evaluation clinical testing, gait analysis, electrophysiological conduction testing, tibialis anterior muscle weight recording and axon counts from the distal nerve stump were used. RESULTS: There was a significant difference between the "bioartificial nerve graft" (group I) and the non-seeded bioartificial nerve graft (group II) indicating the importance of the living Schwann cells. Comparing the results of the "bioartificial nerve graft" (group I) with the autologous nerve grafts (group III), there was a significant difference in all the examinations indicating a still slower regeneration in the artificial graft. CONCLUSIONS: We conclude that the unique three-dimensional net allowed the settlement of Schwann cells onto the biodegradable filaments, which can be used as "artificial Bünger bands". With further refinements of the "artificial Bünger bands" and Schwann cell cultures there should be improved functional and histological results in the "bioartificial nerve graft" group.

Bridging extended nerve defects with an artifcial nerve graft containing Schwann cells pre-seeded on polyglactin filaments.

A 24 mm long bioartificial nerve graft (BNG) was created to bridge extended peripheral nerve defects of the rat sciatic nerve. In our previous studies, an identical graft had demonstrated good results over nerve gaps of up to 15 mm. The BNG device comprised a collagen-I tube filled with ten Schwann-cell-seeded polyglactin filaments and 10(6) isogenic Schwann cells suspended in Matrigel which were implanted in 27 rats (group I). Schwann-cell-free grafts (27 rats) and nerve autografts (18 rats) served as controls. Functional recovery was followed over a period of six months using walking track analysis. Terminal analyses of graft efficacy included neurophysiology, muscle weight, and histological assessment of the implants and the distal nerve stumps. In 17/27 cases, axonal regeneration into the distal nerve stump could be detected across the BNG, but all animals in group I and II failed to regain motor function of the hindlimb upon completion of the experiment. Axon diameter and axonal density in the graft and distal nerve stump were greater in group I than in group II. Although Schwann cells had a significant positive effect on axonal regeneration, either granuloma formation or the amount of the inserted foreign material may have impaired nerve regeneration by acting as a physical impediment to nerve regeneration or negatively effecting cell function.

Brain neurotransmitter changes in three patients who had a fatal hyperthermia syndrome.

The authors examined the autopsied brains from three patients who had a fatal hyperthermia syndrome. There was marked hypothalamic noradrenaline depletion in all three patients, severe brain choline acetyltransferase deficiency with nucleus basalis cell loss in two patients, and mild to moderate brain choline acetyltransferase loss in one patient. Striatal dopamine metabolite/dopamine ratio was below normal in two patients and not elevated, as would be expected after short-term neuroleptic administration, in the third. This suggests that reduced capability (aggravated by the cholinergic deficit) of the nigrostriatal dopamine system to respond adequately to stress and/or neuroleptic-induced receptor blockade may be important in the development and course of fatal hyperthermia syndrome.

Comparison of unilateral and bilateral intrastriatal 6-hydroxydopamine-induced axon terminal lesions: evidence for interhemispheric functional coupling of the two nigrostriatal pathways.

Partial lesions of the nigrostriatal dopamine system can be induced reliably by the intrastriatal injection of 6-hydroxydopamine (6-OHDA) and are considered to be analogous to the early stages of human Parkinson's disease. Previous studies have established a clear correlation between different doses and placements of the 6-OHDA toxin and the degree of neurodegenerative changes and behavioral impairments. In the present study, the influence of the interdependence between the two nigrostriatal systems in both hemispheres on the effects on sensorimotor behavioral performances after terminal 6-OHDA lesions was investigated. The behavioral effects were correlated to the extent of nigral dopamine neuron cell and striatal tyrosine-hydroxylase (TH)-positive fiber loss. Sprague-Dawley rats receiving unilateral intrastriatal 6-OHDA injections (4 x 5 microg) exhibited a 30-70% reduction in striatal TH-positive fiber density along an anterior-posterior gradient, an 80% loss of nigral dopamine neurons and a mild degree of behavioral impairments as revealed by amphetamine-induced rotational asymmetry, and a reduced performance in the stepping and postural balance tests. When the same amount of toxin was injected twice into both hemispheres (2 x 4 x 5 microg), additional behavioral deficits were observed, consisting of a significant, but temporary, weight loss, a stable reduction in general locomotor activity and explorational behavior, and a long-term deficit in skilled forelimb use. This is interesting in light of the morphological findings, in which uni- and bilaterally lesioned animals did not differ significantly in the extent of TH-immunoreactive fiber and dopamine neuron loss within the nigrostriatal system in each lesioned hemisphere. These results indicate that the interdependent regulation of the two nigrostriatal systems may provide some compensatory support for the function and behavioral performance of the lesioned side via the normal unlesioned side, which is lost in animals with bilateral lesions of the nigrostriatal system. Therefore, this model of uni- and bilateral partial lesions of the nigrostriatal system, as characterized in the present study, may foster further exploration of compensatory functional mechanisms active in the early stages of Parkinson's disease and promote development of novel neuroprotective and restorative strategies.

CD44s-targeted treatment with monoclonal antibody blocks intracerebral invasion and growth of 9L gliosarcoma.

Glioma invasiveness is a complex process involving recognition and attachment of tumor cells to particular extracellular matrix (ECM) molecules prior to migrating into proteolytically modified matrix and inducing angiogenesis. CD44 is a group of transmembrane adhesion molecules found on a wide variety of cells including gliomas that has been suggested as the principal mediator of migration/invasion. The aim of the present study was to demonstrate whether antibody specific for the standard form of CD44 (CD44s, 85-90 kDa) might prevent invasion, thus blocking growth of the 9L gliosarcoma in vivo. High expression of CD44s on the surface of 9L cells and brain tumors was demonstrated by immunochemistry. Fluorescence-activated cell sorting (FACS) demonstrated binding saturation of anti-CD44s monoclonal antibody (mAb) to the receptor at 1 microg/5 x 10(5) cells. Blocking of CD44s in vitro resulted in a dose-dependent progressive, up to 95%+/-2.5% detachment of 9L cells from ECM-coated culture surfaces. Blocking of CD44s in vivo resulted in significantly reduced 9L brain tumors (2.5%+/-0.4%)--measured as the quotient: tumor surface (mm2)/brain surface (mm2) x 100--as compared to untreated (16.1%+/-2.2%) or sham-treated rats (16%+/-3.7% to 16.1%+/-3%). We conclude that CD44s-targeted treatment with specific mAb may be an effective means for preventing glioma progression.

Evidence of recurrent atypical meningioma with rhabdoid transformation and expression of pyrogenic cytokines in a child presenting with a marked acute-phase response: case report and review of the literature.

Children presenting with acute systemic illnesses that lack specific clinical or serological defining features may be diagnosed as having a chronic infection, an atypical systemic vasculitis or a connective tissue disease, but often turn out to have occult neoplasias. Cytokines have been implicated in causing many of the systemic effects in such cases. In this study, we describe the case of a 9-year-old boy presenting at an interval of 18 months with a marked acute-phase response due to a recurrent atypical meningioma with rhabdoid transformation of the tentorium cerebelli. Resection of the recurrent tumor was curative. We evaluated in detail the local and systemic production of cytokines released by the primary and the recurrent tumor. Blood and CSF samples were taken pre-, intra-, and postoperatively, and the production of IL-6, IL-1beta, and TNF-alpha was measured by enzyme-linked immunosorbent assays (ELISA). The level of IL-6 in CSF was about 150-fold increased before tumor resection, normalizing postoperatively. On the contrary, the levels of IL-1beta and TNF-alpha in CSF and of IL-6, IL-1beta, and TNF-alpha in serum were pre-, intra-, and postoperatively within normal limits. Cytokine production was also evaluated immunohistochemically, and confirmed strong IL-6 and TNF-alpha expression in the primary and the recurrent tumor, while expression of IL-1beta was lacking. The scattered MHC class II- and leukocyte common antigen (LCA)-expressing inflammatory cells, which were infiltrating exclusively the tumoral stroma, had no detectable cytokine immunoreactivity. We conclude that chronic IL-6 and TNF-alpha production by the tumor cells in this patient was responsible for the severe systemic illness with which he presented.

Expression of granulocyte colony-stimulating factor in recurrent glial tumors is inversely correlated with tumor progression.

We have previously reported that in vivo-expression of granulocyte colony-stimulating factor (G-CSF) is a characteristic feature of reactive and neoplastic astrocytes. The aim of the present study was to analyze the expression of G-CSF protein in primary and recurrent astroglial, oligodendroglial and mixed glial-differentiated tumors. G-CSF expression was present in all GFAP-positive tumors excepting glioblastomas. G-CSF expression was significantly reduced in recurrent tumors more dedifferentiated than their primary counterparts. G-CSF expression was absent in all GFAP-negative tumors such as oligodendrogliomas. Our results demonstrate that G-CSF is a highly sensitive differentiation marker of neoplastic astrocytes, which is lost during tumor progression.

Cellular mechanisms of rejection and regeneration in peripheral nerve allografts.

A model of rejection and regeneration of peripheral nerve allografts in rats is presented. A 2.5-cm segment of 28 right sciatic nerves was transplanted orthotopically from LEW.1W to DA and from DA to LEW.1W. With a microsurgical technique, proximal and distal coaptations were performed. In an autologous control group the same surgical procedure was applied. Evaluation included clinical estimation of motor recovery and macroscopic appearance of the graft, electrophysiological examination, conventional histology, and immunohistology. The latter concentrated on demonstration of monomorphic and polymorphic determinants of MHC class I and II antigens and of macrophages. By functional, electrophysiological, and histological parameters it was demonstrated that after rejection a certain degree of regeneration took place in the allografts. Both rejection and subsequent regeneration were studied in detail by immunohistology. During the course of Wallerian degeneration MHC class I expression on myelin sheaths could be demonstrated. When the rejection response occurred, additional MHC class II expression on myelin sheaths and on vascular endothelial was observed. Recipient specific class I-positive macrophages were infiltrating the graft from the epineurium and the coaptation sites, and were later present at the sites of myelin degradation. At 6 weeks postoperatively donor-specific MHC products were no longer detectable, but recipient-specific Schwann cells were present in the allograft tissue. We conclude that a rejection response renders a peripheral nerve allograft acellular but does not destroy the nerve architecture, still enabling it to function as an axon conduit. The regeneration in the rejected allograft however lacks the positive neurotropic and -trophic influence physiologically provided by viable Schwann cells.

Immunohistochemical characterization of axonal sprouting and reactive tissue changes after long-term implantation of a polyimide sieve electrode to the transected adult rat sciatic nerve.

The development of artificial microstructures suited for interfacing of peripheral nerves is not only relevant for basic neurophysiological research but also for future prosthetic approaches. Aim of the present study was to provide a detailed analysis of axonal sprouting and reactive tissue changes after implantation of a flexible sieve electrode to the proximal stump of the adult rat sciatic nerve. We report here that massive neurite growth after implantation, steadily increasing over a period of 11 months, was observed. Parallel to this increase was the expression of myelin markers like Po, whereas non-myelin-forming Schwann cells did not change. Compared to five weeks post-implantation. where both Schwann-cell phenotypes were intermingled with each other, non-myelin-forming Schwann cells occupied a peripheral position in each microfascicle after 11 months. After an initial increase, hematogenous macrophages were down-regulated in number but maintained close contact with the implant. However, at no time were signs of its degradation observed. It is concluded that the introduced flexible polyimide electrode is suitable for contacting peripheral nerves since it permits substantial neurite growth and offers excellent long-term stability.

Temporal evolution of electroencephalographic abnormalities in Creutzfeldt-Jakob disease.

Frequent serial EEG investigations of three patients with neuropathologically confirmed Creutzfeldt-Jakob disease lasting 13, 24 and 68 weeks revealed typical periodic activity of short duration with stereotyped bilateral sharp waves at the 7th, 8th, and 12th week, respectively, after the onset of symptoms. During the later stages, there were several deviations from this typical pattern. However, periodic activity was preceded between the 3rd and 9th week by intermittent localized or lateralized delta rhythms, which gradually changed into periodic activity. This early temporal evolution of EEG abnormalities may be helpful in the early diagnosis of Creutzfeldt-Jakob disease when accompanied by other investigations to exclude other causes of intermittent delta rhythms.

Unusual EEG findings in a case of Creutzfeldt-Jakob disease.

A case is reported of histopathologically verified Creutzfeldt-Jakob disease of long duration (more than 3 years) with some clinical peculiarities. The prominent peculiarity was a nearly normal EEG during repeated examinations, even in the terminal stage.

Immunohistochemical detection of female sex hormone receptors in meningiomas: correlation with clinical and histological features.

Sixty-one meningiomas from 60 patients were screened for estrogen receptors and progesterone receptors (PgR) with monoclonal antibodies in an immunohistochemical assay. In addition, 43 of the cases were evaluated for tumor size and peritumoral edema, as seen on computed tomographic scans and magnetic resonance images. Sixty-one percent of the tumors contained significant amounts of PgR, whereas no estrogen receptor-positive tumor was observed. Thirteen percent of all tumors were classified as nonbenign variants (atypical and anaplastic meningiomas) and were more frequently found in male patients (P < 0.05). Nonbenign tumors more frequently showed an absence of PgR (P < 0.05), and there was a tendency for PgR-negative tumors to be larger than PgR-positive ones. No correlation was found between PgR status and edema. It is concluded that PgR status in meningiomas is related to tumor differentiation and may be of prognostic value with regard to biological behavior and clinical outcome.

Arterial hypertension and neurovascular compression at the ventrolateral medulla. A comparative microanatomical and pathological study.

Intraoperative observations and animal experiments suggest that neurovascular compression at the left ventrolateral medulla is a possible etiological factor in essential hypertension. In pursuing this hypothesis, the authors examined the neurovascular relations in the posterior cranial fossa of 24 patients with essential hypertension, of 10 with renal hypertension, and of 21 normotensive control patients. Artificial perfusion of the vessels and microsurgical investigations during autopsy identified the vascular relations at the brain stem and at the root entry zone of the caudal cranial nerves. There was no evidence of neurovascular compression at the ventrolateral medulla on the left side in any patient from the control group or among those with renal hypertension. Two normotensive patients had neurovascular compression at the right ventrolateral medulla by the posterior inferior cerebellar artery. In contrast, all patients with essential hypertension had definite neurovascular compression at the left ventrolateral medulla. Additional compression of the right side was seen in three of these patients. Based on the anatomical appearance, it was possible to define three distinct types of neurovascular compression at the ventrolateral medulla. Common to all three types is the compression of the medulla oblongata at its rostral part just caudal to the pontomedullary junction and lateral to the olive in the retro-olivary sulcus. Comparative histopathological study of the microsurgically examined brain-stem specimens revealed no differences between patients with essential hypertension, those with renal hypertension, and normal controls. There was a structural integrity at the site of neurovascular compression at the ventrolateral medulla. The microanatomical findings of this study show that neurovascular relations at the ventrolateral medulla in essential hypertension give rise to pulsatile compression on the left. This supports Jannetta's hypothesis of neurovascular compression at the left ventrolateral medulla as an etiology of essential hypertension.

The influence of nimodipine on cerebral blood flow autoregulation and blood-brain barrier.

Twenty anesthetized rats were randomly assigned to a nimodipine-treated group or a control group of 10 rats each. Local cerebral blood flow (lCBF) was measured by means of a surface electrode using the hydrogen clearance technique. Systemic arterial pressure (SAP) was varied with administration of norfenefrine or by hemorrhage in order to obtain SAP/cerebral blood flow (CBF) curves under different conditions. In the control group, a typical autoregulation curve was obtained with an lCBF plateau between 70 and 120 mm Hg SAP. The nimodipine-treated animals, however, showed only a slight diminution in the slope of the curve but no real plateau, indicating impairment of CBF autoregulation. In another series, 20 anesthetized rats were randomly assigned to a treatment group or a control group of 10 animals each. Intravenous Evans blue dye was used as a tracer for blood-brain barrier (BBB) function. In both groups, SAP was raised to a level of 180 mm Hg with administration of norfenefrine for 6 minutes. Extravasation of significantly more Evans blue dye was observed in the nimodipine group than in the control group, indicating impairment of the BBB. It is concluded that nimodipine may impair CBF autoregulation, allowing damage to the BBB under hypertensive conditions.

In vivo inhibition of angiogenesis and growth of the human U-87 malignant glial tumor by treatment with an antibody against basic fibroblast growth factor.

The effectiveness of in vivo suppression of neovascularization and growth of malignant glial tumors by in situ administration of an antibody directed against basic fibroblast growth factor (bFGF), a strong mitogen for cells of mesodermal origin, was tested. One hundred fifty congenitally athymic nude rats (han rnu/rnu) were implanted intracerebrally with U-87MG tumor cells, known constitutive producers of bFGF. The animals were randomly assigned to six groups of 25 animals each. Animals were treated by in situ application of saline (Group F), control antibody (Group D), or polyclonal anti-bFGF antibody (Group B). In additional groups a putative effect on tumor growth caused by the treatment application device itself (between growth control Groups A and E), and the effect of heat-inactivated tumor cells (negative control Group C) were tested. After 3 weeks of treatment, tumor progression and degree of neovascularization were morphometrically recorded. In the untreated Groups A and E massive tumor growth was recorded, consisting of 19.9% +/- 0.4% and 27.1% +/- 0.5%, respectively, of the total brain cross-sectional area. In Group C, no tumor growth occurred. In control Groups D and F tumor progression consisted of 18.6% +/- 0.4% and 18.5% +/- 0.4%, respectively, of the total brain cross-sectional area; whereas in the anti-bFGF treated Group B, significantly smaller tumor masses measuring 7.2% +/- 0.1% were recorded. New blood vessels were located both peritumorally and intratumorally and defined as numerical density and area fraction (number/area and area/area). Significantly more new blood vessels were found in Groups A, D, E, and F, ranging from 41,380/mm2 +/- 464/mm2 to 53,442/mm2 +/- 150/mm2 peritumorally and 51,846/mm2 +/- 495/mm2 to 64,660/mm2 +/- 183/mm2 intratumorally than in the anti-bFGF treated Group B, which numbered 8220/mm2 +/- 225/mm2 peritumorally and 16,554/mm2 +/- 236/mm2 intratumorally. The authors conclude that treatment with anti-bFGF antibody is effective in inhibiting tumor-induced angiogenesis and correlated tumor progression. However, owing to the character of the experimental system used, one cannot exclude the possibility that application of the specific anti-bFGF antibody also counteracts device-induced neovascularization. The authors suggest that combined surgical excision and adjuvant immunotherapy of tumors such as glioblastoma and other malignant brain tumors that express bFGF might prevent tumor recurrence.

Induction of heat shock protein 70 in the rat brain following intracisternal infusion of autologous blood: evaluation of acute neuronal damage.

OBJECT: Investigation into a potential treatment for the acute period following onset of spontaneous subarachnoid hemorrhage (SAH) is hampered by the lack of a standardized experimental model. For that purpose the authors elaborated on a small-animal model in which computer-controlled intracisternal blood infusion is used and investigated whether this model can reliably reproduce acute neuronal injury after SAH. METHODS: Whole autologous blood (blood-infused group) or isotonic saline (control group) was infused into the cisterna magna or olfactory cistern of rats. The infusions decreased exponentially during a 5-minute period. Throughout the infusion period, intracranial pressure (ICP) was monitored. Neuronal injury was quantified by observing tissue immunoreactivity to a 70-kD heat shock protein (HSP70) and comparing this with the tissue's reaction to hematoxylin and eosin staining. On Days 1, 3, and 5, the CA1, CA3, and dentate gyrus regions of the hippocampus were analyzed, respectively. During saline infusion ICP increased within seconds beyond 80 mm Hg and afterward decreased in accordance with the infusion rate. During the infusion of blood, the same initial pressure peak was found, but the ICP remained increased beyond this pressure level throughout the 5-minute infusion period. The HSP70 immunoreactivity in the saline-infused group was found only on Day 1 in the CA1 region and the dentate gyrus, but not in the CA3. After injection of whole blood, there was HSP70-positive staining in the CA1, CA3, and dentate gyrus regions throughout the observation period. CONCLUSIONS: The controlled cisternal infusion of blood caused neuronal injury that resembled that of previous experimental models that produce SAH by rupture of intracranial vessels with endovascular techniques. Unlike those experiments, the intracisternal infusion technique presented by the authors provides more standardized bleeding with regard to ICP, the volume of subarachnoid blood, and the extent of acute cellular injury.

Disruption of intracerebral progression of C6 rat glioblastoma by in vivo treatment with anti-CD44 monoclonal antibody.

OBJECT: Glioblastoma multiforme (GBM) invasiveness is a complex process that involves recognition and attachment of GBM cells to particular extracellular matrix (ECM) molecules before migrating into proteolytically modified matrix and inducing angiogenesis. The CD44 molecule, which is a transmembrane adhesion molecule found on a wide variety of cells including GBM, has been suggested as the principal mediator of migration and invasion. The aim of the present study was to demonstrate whether an antibody specific to the standard form of CD44 (CD44s, 85-90 kD) might prevent invasion and thus disrupt progression of C6 GBM in vivo. METHODS: Immunostaining demonstrated homogeneous expression of CD44s on the surface of C6 GBM cells and tumors. Flow cytometric analysis demonstrated binding saturation of anti-CD44s monoclonal antibody (mAb) to the receptor at 1 microg/5 x 10(5) cells. Blocking of CD44s in vitro resulted in a dose-dependent progressive (up to 94+/-2.7%; mean +/- standard deviation [SD]) detachment of C6 cells from ECM-coated culture. Blocking of CD44s in vivo resulted in significantly reduced C6 brain tumors (3.6+/-0.4% [SD])--measured as the quotient: tumor surface (mm2)/brain surface (mm2) x 100--compared with untreated (19.9+/-0.9%) or sham-treated (19.2+/-1.1 to 19.3+/-2.5% [SD]) rats. Disruption of C6 GBM progression correlated with an improved food intake; treated rats were significantly less cachectic (166.6+/-16.4 g [SD]) than those that were untreated (83+/-2.7 g [SD]) or sham-treated (83.4+/-1.1 to 83+/-2.2 g [SD]) rats. CONCLUSIONS: The authors conclude that CD44s-targeted treatment with specific mAb may represent an effective means for preventing progression of highly invasive GBMs.

Expression of MMP-2, MMP-7, MMP-9, MMP-10 and MMP-11 in human astrocytic and oligodendroglial gliomas.

The members of the matrix metalloproteinase family (MMP) have the ability to degrade macromolecules of the extracellular matrix and are responsible for tumor invasion and infiltration, limiting the effectiveness of the neurosurgical resection of brain tumors. Among the glial brain tumors, astrocytomas and oligodendrogliomas are the most important tumor entities and require a different therapeutic approach. To determine the pattern of MMP expression in astrocytic and oligodendroglial tumors, sections of astrocytic and oligodendroglial differentiated glioblastomas (WHO grade IV), as well as of anaplastic oligodendrogliomas (WHO grade III) and anaplastic gemistocytic astrocytomas (WHO grade III) were immunostained for MMP-2, MMP-7, MMP-9, MMP-10 and MMP-11. MMP-7, MMP-10 and MMP-11 were strongly expressed by neoplastic gemistocytic astrocytes while oligodendrocytic tumor regions showed only a low immunoreaction. In contrast, MMP-2 and MMP-9 mainly immunolabeled vascular structures. These data indicated that MMP-7, MMP-10 and MMP-11 contribute to the worse prognosis of astrocytic tumors when compared to oligodendrogliomas, while MMP-2 and MMP-9 might play an important role in neo-angiogenesis and tumor vascularization.

Doxorubicin-induced cell death in highly invasive human gliomas.

Glioblastoma is the most invasive form of primary brain tumors, and is often refractory to chemotherapy. Herein, we provide evidence that two highly invasive human glioma cell lines U-87 MG and U-373 MG, entered apoptosis after 48 hours following 24 h growth arrest induced by Doxorubicin (10 micrograms/2 x 10(5) cells/ml). Apoptosis depended solely on the level of intracellular drug accumulation, and it was not related to a functional p53 tumor suppressor factor. The multidrug resistance gene 1 (mdr-1) encoded P-glycoprotein (P-gp) was weakly expressed in these cells upon exposure to Doxorubicin, and exerted no influence on the extent of cellular drug efflux. Drug efflux occurred only in U-373 MG glioma cells subsequent to physical damage of the membrane upon exposure to Doxorubicin. Pretreatment of tumor cells with 10 micrograms/ml Doxorubicin precluded tumor formation on the chorioallantoic membrane (CAM) of embryonated hen eggs. Single-dose application of 0.4 microgram Doxorubicin on CAM/U-87 MG and CAM/U-373 MG tumor transplants inhibited tumor invasion in CAM tissue by 40 to 50%. These data suggest that highly invasive glioblastomas can be driven to apoptosis following growth arrest induced by Doxorubicin, providing that intracellular drug accumulation suffices cytotoxic levels.