Tolerance of the normal canine brain to epithermal neutron irradiation in the presence of p-boronophenylalanine.

Twelve normal dogs underwent brain irradiation in a mixed-radiation, mainly epithermal neutron field at the Brookhaven Medical Research Reactor following intravenous infusion of 950 mg of 10B-enriched BPA/kg as its fructose complex. The 5 x 10 cm irradiation aperture was centered over the left hemisphere. For a subgroup of dogs reported previously, we now present more detailed analyses including dose-volume relationships, longer follow-ups, MRIs, and histopathological observations. Peak doses (delivered to 1 cm3 of brain at the depth of maximum thermal neutron flux) ranged from 7.6 Gy (photon-equivalent dose: 11.8 Gy-Eq) to 11.6 Gy (17.5 Gy-Eq). The average dose to the brain ranged from 3.0 Gy (4.5 Gy-Eq) to 8.1 Gy (11.9 Gy-Eq) and to the left hemisphere, 6.6 Gy (10.1 Gy-Eq) to 10.0 Gy (15.0 Gy-Eq). Maximum tolerated 'threshold' doses were 6.7 Gy (9.8 Gy-Eq) to the whole brain and 8.2 Gy (12.3 Gy-Eq) to one hemisphere. The threshold peak brain dose was 9.5 Gy (14.3 Gy-Eq). At doses below threshold, some dogs developed subclinical MRI changes. Above threshold, all dogs developed dose-dependent MRI changes, neurological deficits, and focal brain necrosis.

Postmortem neuropathological features secondary to boron neutron capture therapy for glioblastoma multiforme.

This postmortem study of 12 patients with glioblastoma multiforme (GBM) treated with boron neutron capture therapy (BNCT) employing an epithermal neutron beam and p-boronophenylalanine describes the neuropathological findings in patients receiving a relatively high radiation dose to the tumor, but a relatively low radiation dose to the normal brain. In addition to a standardized neuropathology panel of sections, we used individual treatment dosimetry maps to select sections along the projected maximum radiation beam pathway. We found that the normal neuroparenchyma exposed to the highest radiation dose exhibited a single instance of radiation-induced focal venular fibrinoid necrosis and a single instance of multifocal demyelination. Semiquantitative analysis of pretreatment neurosurgical and postmortem tumor samples revealed only two radiation ascribed histopathological findings to be particular to therapy, fibrinoid necrosis and vascular hyalinization. In this relatively small series of cases we found an unexpectedly high frequency of cases (3 of 12) with neurodegenerative histopathology (Lewy bodies, neurofibrillary tangles, and neuritic senile plaques), which appeared, by distribution, to be independent of the radiation beam. Two of these patients were over 70 yr of age. One was only 41. Our findings suggest an acceptable radiation-induced level of neurotoxicity at the lower doses employed, but raise the possibility of unexpected boron neurodegenerative toxicity.

Biodistribution of boronophenylalanine in patients with glioblastoma multiforme: boron concentration correlates with tumor cellularity.

Boron-10 (10B) concentrations were measured in 107 surgical samples from 15 patients with glioblastoma multiforme who were infused with 95 atom% 10B-enriched p-boronophenylalanine (BPA) intravenously for 2 h just prior to surgery at doses ranging from 98 to 290 mg BPA/kg body weight. The blood 10B concentration reached a maximum at the end of the infusion (ranging from 9.3 to 26.0 microg 10B/g) and was proportional to the amount of BPA infused. The boron concentrations in excised tumor samples ranged from 2.7 to 41.3 microg 10B/g over the range of administered BPA doses and varied considerably among multiple samples from individual patients and among patients at the same BPA dose. A morphometric index of the density of viable-appearing tumor cells in histological sections obtained from samples adjacent to, and macroscopically similar to, the tumor samples used for boron analysis correlated linearly with the boron concentrations. From that correlation it is estimated that 10B concentrations in glioblastoma tumor cells were over four times greater than concurrent blood 10B concentrations. Thus, in the dose range of 98 to 290 mg BPA/kg, the accumulation of boron in tumor cells is a linear function of BPA dose and the variations observed in boron concentrations of tumor specimens obtained surgically are largely due to differences in the proportion of nontumor tissue (i.e. necrotic tissue, normal brain) present in the samples submitted for boron analysis. The tumor:blood 10B concentration ratio derived from this analysis provides a rationale for estimating the fraction of the radiation dose to viable tumor cells resulting from the boron neutron capture reaction based on measured boron concentrations in the blood at the time of BNCT without the need for analysis of tumor samples from individual patients.

Glial transforming growth factor (TGF)-beta isotypes in multiple sclerosis: differential glial expression of TGF-beta 1, 2 and 3 isotypes in multiple sclerosis.

We studied glial transforming growth factor (TGF)-beta isotype expression in 14 cases of multiple sclerosis. Acute active lesions exhibited selective TGF-beta 2 immunoreactivity of lesion encircling ramified microglia. In contrast, astrocytes within chronic active white matter lesions expressed all three isotypes. Chronic active lesions which extended into cortex exhibited selective cortical astrocyte TGF-beta 2 expression. This isotype was also selectively expressed by astrocytes in apparently normal white matter. A similar pattern of glial TGF-beta expression was seen in the pathological control, progressive multifocal leukoencephalopathy. The results suggest that TGF-beta cytokines are locally expressed in demyelination and that the beta 2 isotype may be uniquely regulated.

MDR1 P-glycoprotein is expressed by endothelial cells of newly formed capillaries in human gliomas but is not expressed in the neovasculature of other primary tumors.

The expression of human MDR1 P-glycoprotein (Pgp) in the capillary endothelial cells of the central nervous system has been demonstrated. The brain capillary endothelial cells maintain the structure and function of the blood-brain barrier. Recently, the human MDR1 Pgp (and its mouse homologue MDR1a Pgp) has been shown to function as an important part of this barrier, pumping out xenobiotics from endothelial cells into the lumen of capillaries resulting in the protection of the brain parenchyma. To examine whether the endothelial cells of the newly formed capillaries during neoangiogenesis within malignant human brain tumors express MDR1 Pgp, 35 adult surgical brain tumor specimens (29 gliomas and 6 tumors metastatic to the brain) were obtained from previously untreated patients and studied by a new immunohistochemical sandwich method developed in our laboratory using the JSB-1 monoclonal antibody. JSB-1 is specific for the Pgp product of the human MDR1 (and not MDR3) gene. This sensitive method allows the detection of Pgp in capillary endothelial cells of normal brain in conventional paraffin sections after formalin fixation. The endothelial cells of the newly formed capillaries in 25 of 29 gliomas (86%) and 3 of 6 metastatic tumors, immunostained positive for MDR1 Pgp. The tumor cells in 7 of 35 cases were also positive for Pgp. In the 35 brain tumor cases investigated, the endothelial cells were Pgp positive in the tumor-brain border and in the brain further from the tumor. Capillary endothelial cells of neovasculature in 137 malignant tumors (non-brain) obtained from previously untreated patients showed no MDR1 Pgp expression. These results demonstrated that MDR1 Pgp is expressed not only in the capillaries of normal brain but also in the majority of the newly formed capillaries of brain tumors. Multidrug resistance of brain tumors may result not only from the expression of resistance markers in neoplastic cells but also from the MDR1 Pgp expression in endothelial cells of tumor capillaries. Pgp in this special localization can exclude chemotherapeutic agents from tumor cells that are located around the capillaries. The therapeutic benefit and selectivity of chemotherapeutic agents in combination with a Pgp-reversing agent should be evaluated.

Differential expression of TGF-beta 1, 2 and 3 isotypes in Alzheimer's disease: a comparative immunohistochemical study with cerebral infarction, aged human and mouse control brains.

Based upon the hypothesis that growth regulatory and inflammatory mechanisms participate in the pathogenesis of Alzheimer's disease, we studied cases of Alzheimer's disease for immunoreactivity to each of the three mammalian transforming growth factor beta (TGF-beta) isotypes: TGF-beta 1, TGF-beta 2, TGF-beta 3. Results were compared with those seen in control brains and in a destructive pathological process, subacute infarction. In the cases of Alzheimer's disease, TGF-beta 1 immunoreactivity was limited to neuritic profiles within senile plaques. Neuronal neurofibrillary tangles, plaque neurites, microglia, astrocytes and macrophages expressed TGF-beta 2 immunoreactivity. TGF-beta 3 produced strikingly selective staining of Hirano bodies. In contrast, in cases with infarction, reactive astrocytes and macrophages were positive with all three antibodies. Ramified microglia labeled selectively, as in the Alzheimer brains, with the TGF-beta 2 antibody. Subtle generalized astrocyte and microglial immunoreactivity for TGF-beta 2 was seen in pathological and control brains. The localization of TGF-beta isotypes to the lesions of Alzheimer's disease supports the hypothesis that these cytokines may influence lesion expression. Their presence in reactive cells associated with cerebral infarction suggest that they may play a broader role in the pathogenesis of CNS disease.

The superficial basal lamina of ciliary processes: binding studies with IgG-immunogold and ferritins.

In this study, we employed two ultrastructurally visible probing systems, IgG-immunogold and ferritin molecules which differ by surface charge, to study binding activity of the aqueous face of the posterior ciliary processes in short term tissue baths. With these probes we demonstrated that the superficial basal lamina of the non-pigmented epithelium binds monomeric and heat aggregated IgG but not IgG F(ab')2. Binding was inhibited by preincubation with monosaccharides and NaCl suggesting that IgG binding was determined by lectin-like and electrostatic interactions. Anionic binding domains within the basal lamina, capable of exerting an electrostatic influence, were directly demonstrated by selective binding of cationic ferritin species. At high concentrations of cationic ferritin, anionic binding sites were saturated and tracer penetrated the basal lamina to reach intercellular spaces between non-pigmented epithelial cells. We concluded that the superficial basal lamina of the ciliary processes, which is bathed by the aqueous humor, may bind and immobilize IgG, IgG-opsonized antigens and accessible carbohydrate or cations on other molecules in this fluid. This binding may be important in the maintenance of normal aqueous humor composition and in the pathogenesis of infectious and immune-mediated ocular disease.

Localization of tissue inhibitor of matrix metalloproteinases in Alzheimer's disease and normal brain.

Based upon the hypothesis that metalloproteinases and their inhibitors might be involved in the pathogenesis of Alzheimer's disease, we studied brain samples of eight cases of Alzheimer's disease, six other pathological entities and three elderly controls for tissue inhibitor of metalloproteinase (TIMP) immunoreactivity. Specificity was supported by a loss of immunoreactivity following antigen preabsorption of antisera. Areas studied included ependyma, choroid plexus, frontoparietal, hippocampal and cerebellar cortex, n. basalis of Meynert, basal ganglia, midbrain, pons, and medulla. TIMP positivity was localized to neuritic senile plaques, neurofibrillary tangles and Purkinje cells. The pattern of TIMP plaque staining was similar to that observed with anti tau and SP18 antibodies. It differed from that observed with anti SP40, HAM 56 and GFAP antibodies. The selective localization of TIMP to the neuritic lesions of Alzheimer's disease in a codistribution with the amyloid precursor protein and abnormally phosphorylated and truncated tau supports a possible role for metalloproteinases and their inhibitors in the evolution of these lesions.

In situ detection of polymerase chain reaction-amplified HIV-1 nucleic acids in skeletal muscle in patients with myopathy.

The purpose of this study was to determine if the myopathy that commonly occurs in patients with AIDS is associated with active HIV-1 infection in the muscle tissues. Seven muscle biopsies from patients infected by HIV-1 and six controls were tested for HIV-1 DNA and RNA using polymerase chain reaction in situ hybridization and reverse transcriptase in situ polymerase chain reaction. HIV-1 DNA was detected in rare cells in only one case by standard in situ hybridization. However, after polymerase chain reaction amplification HIV-1 DNA was detected in many cells in four of seven muscle tissues from patients with the viral infection and in none of the controls. The number of cells with detectable provirus in the tissue positive by standard in situ hybridization increased up to 100-fold after amplification. Most of the HIV-1 infected cells were macrophages, as determined by colabeling experiments that were localized mainly in the areas of myocyte necrosis. Myocyte nuclei that contained amplified HIV-1 nucleic acids were also noted. Most virally infected cells contained HIV-1 transcripts, which is consistent with activated infection. The demonstration of many HIV-1 infected macrophages and myocytes in muscle biopsies from HIV-1 infected patients with myopathy suggests that active viral infection may play a role in the clinical disease state.

TGF-beta 2 and TGF-beta 3 immunoreactivity within the ciliary epithelium [corrected].

PURPOSE: To determine whether the ciliary epithelium exhibits immunoreactivity for antibodies to transforming growth factor beta (TGF-beta) 2 and TGF-beta 3. The hypothesis was that because the aqueous humor contains mainly biologically active TGF-beta 2, with little TGF-beta 1, the epithelium largely responsible for its composition would also contain this isoform of TGF-beta. The authors anticipated TGF-beta 3 immunoreactivity because TGF-beta 3 often co-localizes with TGF-beta 2. METHODS: The authors followed a standard immunohistochemical protocol using the avidin-biotin complex and newly available rabbit antibodies to synthetic peptide sequences of TGF-beta 2 and TGF-beta 3. Formalin-fixed, paraffin-embedded samples of freshly obtained rabbit and human autopsy eyes were studied. Specificity was supported by specific peptide absorption of antisera before tissue incubation. RESULTS: The pigmented and nonpigmented ciliary epithelia of rabbit and human eyes were stained by antibodies to both TGF-beta 2 and TGF beta-3, and the staining was inhibited by preabsorption of antibodies by peptides of TGF-beta 2 and TGF-beta 3. CONCLUSIONS: The authors conclude that the ciliary epithelium exhibits TGF-beta 2- and TGF-beta 3-like immunoreactivity that, based upon complementary work from other laboratories, is probably synthesized by this epithelium and is not simply absorbed by it from the aqueous humor.

Identification of Fc gamma RI, II and III on normal human brain ramified microglia and on microglia in senile plaques in Alzheimer's disease.

Using monoclonal antibodies to the three known human leukocyte IgG receptors, Fc gamma R, we examined the expression of Fc gamma R in normal brains and in Alzheimer's disease. We found Fc gamma RI, II and III immunoreactivity in senile plaques and on ramified microglia throughout the cortex and white matter of normal and Alzheimer's disease brains. Fc gamma RI expression was independently confirmed by a murine isotype binding study. These findings suggest that intrinsic Fc gamma R may play an important role in normal and disordered immune-related processes in the brain. They support the idea that microglia are brain macrophages.

A common epitope is recognized by monoclonal antibodies prepared against purified human neutrophil Fc gamma RIII (CD16).

Fc gamma RIII is one of two Fc gamma R constitutively expressed by human neutrophils. We have prepared a panel of anti-Fc gamma RIII mAb following immunization of mice with Fc gamma RIII purified from human neutrophils. Ten mAb which reacted with neutrophils, NK cells, and monocyte-derived macrophages were produced. Immunohistochemical staining demonstrated that these mAb also identified macrophages in the red pulp of spleen. Competitive cross-inhibition binding assays demonstrated that nine of the ten mAb reacted with a common epitope that is spatially associated with the ligand binding site. These nine mAb blocked the binding of immune complexes to neutrophils by 65 to 90%. In addition, two other anti-CD16 mAb, which also blocked immune complex binding to neutrophils, inhibited the binding of each of these nine mAb to neutrophils. One of the mAb produced here, 214.1, failed to block immune complex binding. In addition to immunoprecipitating the native Fc gamma RIII glycoprotein, mAb 214.1 was capable of immunoprecipitating a 28-kDa polypeptide following deglycosylation of Fc gamma RIII isolated from neutrophils. The results of cross-competition experiments suggest that mAb 214.1 may recognize the epitope identified by mAb BW209/2. Thus mAb 214.1 identifies a polypeptide epitope distinct from the ligand binding site of Fc gamma RIII on neutrophils.

Muscle biopsy and the clinical course of infantile spinal muscular atrophy.

Eight infants with severe early infantile spinal muscular atrophy diagnosed by clinical presentation and muscle biopsy were studied. The extent of alterations in muscle histology, histochemistry, and ultrastructure did not reflect the relative severity of the clinical presentation or the course of the illness. In seven biopsies, ultrastructural studies demonstrated empty sleeves of basal lamina projecting from the surface of small myofibers. We conclude that severe infantile spinal muscular atrophy often results in myofiber atrophy similar to that found in other motor neuron diseases, and it is not solely a hypotrophic process. Muscle biopsy findings are important because they help to establish the diagnosis, but they do not help predict the severity of disease among infants with this condition.

Immunoglobulin G receptor-mediated phagocytosis by the pigmented epithelium of the ciliary processes.

The current ultrastructural study of cultured explants of bovine and rabbit posterior ciliary processes using immunoglobulin (Ig) G-opsonized red blood cells (E-IgG) showed specific IgG Fc receptor-mediated attachment, ingestion, and digestion of IgG-coated erythrocytes by the pigmented epithelium (PE). Nonpigmented epithelial cells (NPE) in the cultured explants and a transformed cell line of bovine NPE, with and without lymphokine stimulation, did not have IgG-receptor activity. The interaction between PE and E-IgG involved the extension of micropseudopods toward adherent E-IgG, the formation of a linear uniform cap of roughly 200 A between opposing cell membranes, the ingestion of E-IgG by PE into a membrane-lined compartment, and the disintegration of the ingested ligand into membranous debris. Disintegration of some surface-associated E-IgG was also observed and was consistent with the release of a lytic substance by the receptor-activated PE.

The neuromuscular pathology of the Eosinophilia-Myalgia syndrome.

The Eosinophilia-Myalgia Syndrome (EMS) is a recently reorganized disorder in patients ingesting pharmacologic doses of L-tryptophan. We studied the lesions of skeletal muscle, peripheral nerve and skin in 12 cases of EMS. Perimyositis was severe in four, moderate in two, mild in three and absent in three cases. The lesions contained many eosinophils, T-helper cells, mast cells and activated macrophages. Type 2 myofiber atrophy was present in five cases and in one, this was the only pathologic finding. Severe epineurial inflammation was seen in the three sural nerve biopsies. Indirect evidence for peripheral neurologic involvement in three other cases consisted of inflammation surrounding intramuscular nerve twigs (two cases) and neurogenic atrophy (one case). Phlebitis accompanied the connective tissue inflammation in five cases and endarteritis in one. Fasciitis was present in three of four skin biopsies and dermal fibrosis in one.

The effects of chronic serum sickness on albumin distribution and glucose utilization in rat brain.

The level of cerebrospinal fluid (CSF) protein is elevated in diseases and disease models that are associated with circulating immune complexes such as serum sickness. Circulatory immune complexes are known to deposit in the basal lamina of fenestrated capillaries and may, as a result, affect both capillary bed and parenchymal function. Since the brain has both fenestrated and unfenestrated capillaries and immune complexes deposit to a varying extent in the fenestrated capillaries in chronic serum sickness, cerebral capillary permeability to protein may be altered in some brain areas and lead to the elevation of CSF proteins. In addition various other cerebrovascular and metabolic functions may also be affected by this condition. In this study either radio-iodinated serum albumin (RISA) or 2-[14C]deoxyglucose (14C-2DG) was intravenously injected into control Wistar rats and Wistar rats with chronic serum sickness; subsequently the tissue levels of radioactivity were measured by quantitative autoradiography in 4 brain areas with fenestrated capillaries and 11 brain areas with unfenestrated capillaries. The 2-min distribution of RISA, which demarcates the volume of circulating plasma in perfused microvessels and is generally proportional to local plasma flow, was the same in control and experimental rats. The passage of RISA from blood into brain over 30 min was negligible in both groups; thus cerebral capillary permeability to albumin was not detectably increased in any of these 15 brain areas by chronic serum sickness. The rate of local cerebral glucose utilization, an indicator of local metabolic and neural activity, was calculated from the 14C-2DG data and was virtually identical in control and experimental rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Eosinophilic neuritis, perimyositis, and vasculitis associated with ingestion of L-tryptophan.

Four cases are described of a clinical syndrome which developed in the setting of L-tryptophan ingestion. The major manifestations consisted of myalgias, neuropathy, weakness, and profound eosinophilia. Pathologically a vasculitis involving predominantly small veins was observed along with a mixed cellular infiltrate in the perimysium and epineurium. Clusters of eosinophils were characteristically noted in the tissue specimens. The clinical course appears to be chronic although further longterm followup will be required. One patient pursued a relentless downhill course with progressive neurologic impairment and death. Although the mechanism of tissue injury in these individuals is speculative, the possible association of this widely used nonprescription medication with this syndrome should be recognized.

Human immunodeficiency virus type 1 expression in the central nervous system correlates directly with extent of disease.

To investigate human immunodeficiency virus type 1 (HIV-1) pathogenesis in infected individuals and examine the correlation of HIV-1 expression with extent of clinical and pathologic disease, we studied spinal cords from acquired immunodeficiency syndrome patients with a wide range of spinal cord pathology. By performing in situ hybridization with HIV-1-specific riboprobes, we detected HIV-1 RNA in all 10 cords from acquired immunodeficiency syndrome patients with a common, characteristic pathologic entity called vacuolar myelopathy but not in 10 control cords from HIV-1-infected and uninfected patients. In the cords from individuals with vacuolar myelopathy, the level of HIV-1 RNA expression correlated directly with extent of spinal cord pathology and clinical findings. These data support a role for HIV-1 in the pathogenesis of tissue damage and related clinical disease in infected individuals.

Lewy bodies in tyrosine hydroxylase-synthesizing neurons of the human cerebral cortex.

A population of neurons situated in the human cerebral neocortex contains mRNA coding for tyrosine hydroxylase, the key enzyme for catecholamine biosynthesis. Phosphorylated neurofilament-containing cytoplasmic inclusions occur in these neurons in diffuse Lewy body disease, indicating a tendency for selective involvement that is shared with subcortical catecholamine-containing neurons. These findings are relevant to the pathophysiology of several neurologic and psychiatric illnesses in which the monoamine-containing neurons of the neocortex may participate.