Keisuke Ishizawa

Immunohistochemical Demonstration of EMA/Glut1-Positive Perineurial Cells and CD34-Positive Fibroblastic Cells in Peripheral Nerve Sheath Tumors

To clarify the cellular composition of various peripheral nerve tumorous lesions (traumatic neuroma, 5 cases; schwannoma, 10 cases; neurofibroma, 14 cases; perineurioma, 3 cases; conventional malignant peripheral nerve sheath tumor (MPNST), 7 cases; perineurial MPNST, 4 cases), expression of several markers specific to nerve sheath cells, including glucose transporter protein 1 (Glut1) and CD34, were immunohistochemically investigated with highly sensitive detection methods. In normal nerves and neuromas, perineuriums were positive for Glut1 as well as for epithelial membrane antigen (EMA), and there were some CD34-positive fibroblast-like cells in the endoneurium. Schwannomas consisted principally of S-100 protein–positive Schwann cells, whereas a few CD34-positive fibroblastic cells were present in Antoni B areas. Neurofibromas and conventional MPNST exhibited a mixed proliferation of S-100 protein-, EMA/Glut1-, and CD34-positive cells, indicating a heterogeneous composition of the constituents. The catalyzed signal amplification (CSA) system demonstrated more numerous EMA-positive perineurial cells in neurofibromas than did the ENVISION+ method. Perineurial cell tumors (benign and malignant) were composed of EMA/Glut1-positive and S-100 protein–negative tumor cells. The present study confirmed the characteristic cellular composition to each nerve sheath tumor immunohistochemically and showed the usefulness of the nerve sheath cell markers. Glut1 as well as EMA are specific to perineurial cells, and CD34 seems to be immunoreactive to endoneurial fibroblasts.

Pleomorphic xanthoastrocytoma: a comparative pathological study between conventional and anaplastic types

Aims:  To facilitate the understanding and correct diagnosis of the anaplastic variant of pleomorphic xanthoastrocytoma (PXA).

Glial cytoplasmic inclusions and tissue injury in multiple system atrophy: A quantitative study in white matter (olivopontocerebellar system) and gray matter (nigrostriatal system).

Glial cytoplasmic inclusions (GCIs) and microglia were quantified in 12 cases of multiple system atrophy (MSA) with special reference to their association with histologically defined lesion severity. The targets of the analysis were white matter (cerebellum, pontine base) and gray matter (putamen, substantia nigra). First, the lesion severity was defined: for white matter, the degree of demyelination and tissue rarefaction were semi‐quantified on Klüver‐Barrera (KB) sections as grade I (mildly injured), II (moderately injured), and III (severely injured); for gray matter, neurons and astrocytes were counted on KB and glial fibrillary acidic protein‐immunostained sections, respectively. Next, the GCI burden was quantified on sections immunostained for α‐synuclein, phosphorylated α‐synuclein, and ubiquitin and the microglial burden was quantified on sections immunostained for HLA‐DR. In white matter, the GCI and microglial burdens were the greatest when the tissue injury was mild and/or moderate (grade I and/or grade II), and they became less prominent when the tissue injury became more severe (grade III). In gray matter, in contrast, the GCI and microglial burdens failed to show significant correlations with the lesion severity. Our result suggests that the amount of GCIs as well as that of microglia is reduced when the tissue injury becomes severe in vulnerable white matter areas, but not in vulnerable gray matter areas, of MSA. It also suggests that there seems to be a difference between gray matter and white matter in the way GCIs and microglia participate in the degenerative process of MSA.

Stromal cells in hemangioblastoma : Neuroectodermal differentiation and morphological similarities to ependymoma

The histogenesis of stromal cells in hemangioblastoma is inconclusive despite a long‐term controversy. An immunohistochemical and ultrastructural study was conducted for 17 cases of cerebellar hemangioblastoma. A wide range of immunohistological markers, targeting epithelial, mesenchymal, endothelial and neuroectodermal tissues, was used. In all cases, the microscopic hallmark characterizing hemangioblastomas, that is, lipid‐containing stromal cells and a fine capillary network, known as a reticular variant, was noted. Stromal cells showed a variable immunoreactivity for neuroectodermal markers, such as S‐100 protein, CD56, CD57, CD99, and neuron‐specific enolase. This result, in conjunction with the absence of immunoreactivity for epithelial, mesenchymal, and endothelial markers, likely suggests neuroectodermal differentiation of stromal cells. In three cases, another component, known as a cellular variant, where epithelioid tumor cells were arranged in nests encircled by capillaries and/or in pseudorosette‐like structures, was noted. Glial fibrillary acidic protein‐immunoreactivity, which was totally absent in cases only showing the reticular pattern, was noted in two of them, suggesting a distinctive sign of glial differentiation in a proportion of hemangioblastomas. Ultrastructurally, microvilli‐like projections in intracytoplasmic vacuoles were demonstrated in stromal cells. This result, taken together with the neuroectodermal hypothesis of stromal cells, suggests that hemangioblastomas may occasionally exhibit morphological similarities to ependymomas.

Hyperphosphorylated tau deposition parallels prion protein burden in a case of Gerstmann-Sträussler-Scheinker syndrome P102L mutation complicated with dementia

Abstract. Hyperphosphorylated tau (p-tau) deposition has been documented in a limited population of patients with Gerstmann-Sträussler-Scheinker syndrome (GSS) with particular point mutations of the prion protein (PrP) gene. Although its pathogenesis is only poorly understood, p-tau in GSS is known to be identical to that in Alzheimers disease (AD). We conducted immunohistochemical and quantitative image studies on the brain from a 44-year-old man with a 7-year history of dementia, diagnosed as having GSS with a point mutation of the PrP gene at codon 102 (GSS102), the commonest mutation in GSS. Severe spongiform degeneration and numerous PrP plaques were disclosed in the cerebral cortices and hippocampus, consistent with the diagnosis. However, rarely described in GSS102, prominent p-tau deposits as pretangles, neurofibrillary tangles and degenerating neurites were demonstrated adjacent to or around PrP plaques. β-Amyloid protein (Aβ) plaques were generally sparse and appeared invariably to be of a diffuse type. Double-labeling immunohistochemistry yielded co-localization of p-tau with PrP but not with Aβ. Most PrP plaques did not contain Aβ. These results excluded a diagnosis of concomitant AD. Quantitative analysis on a fractional area density of immunoreactive pixels demonstrated that burdens of PrP and p-tau but not Aβ were significantly correlated. These results suggest that p-tau deposition in this GSS102 is secondarily induced by PrP but not by Aβ (secondary tauopathy). Our study also suggests that p-tau deposition might be a more common phenomenon in long-standing GSS.

Large motor neuron involvement in Stiff-man syndrome: a qualitative and quantitative study

Abstract Stiff-man syndrome (SMS) is characterized by fluctuating muscular rigidity and spasm. Recently, antibodies against glutamic acid decarboxylase (GAD), the enzyme catalyzing the synthesis of γ-amino butyric acid (GABA), have been detected in SMS patients. An autoimmune mechanism against GAD was thus proposed for the suppression of GABAergic inhibitory interneurons, resulting in rigidity and spasm. We conducted quantitative investigations on the ventral horn of the spinal cord and its GAD immunoreactivity, post mortem, in a SMS patient and four controls. In the spinal cord of the SMS patient, we found a 70%, 33% and 27% reduction (P < 0.05) in the density of neurons with somal areas of 1000–1500 μm2, 500–1000 μm2, and 0–500 μm2, respectively. The density of neurons with a somal area greater than 1500 μm2 was not reduced, although some neurons in this class showed central chromatolytic changes. The affected muscles exhibited neurogenic atrophy. GAD-like immunoreactivity in the spinal gray matter was not significantly decreased. The density of Purkinje cells, known to contain high amounts of GAD, was not significantly reduced. While the co-occurrence of elevation of anti-GAD antibody in the serum and reduction in the density of small spinal neurons was confirmed, that of smaller α-motor neurons and γ-motor neurons, the qualitative changes in larger α-motor neurons, and the preservation of spinal GAD-like immunoreactivity and non-spinal GAD-containing neurons suggest the involvement of factors other than autoimmune mechanisms through anti-GAD antibodies. More diverse mechanisms may be associated in the pathogenesis of SMS.

Olig2 and CD99 are useful negative markers for the diagnosis of brain tumors.

Olig2, a member of the group of basic helix-loop-helix transcription factors, is innately expressed in oligodendrocytes. In the neoplastic condition, Olig2 is widely expressed in astrocytomas and oligodendrogliomas, but its expression in ependymomas remains poorly documented. A total of 59 brain tumors including 16 ependymomas, 32 astrocytomas, and 11 oligodendrogliomas were immunohistochemically studied for the expression of Olig2 as well as other markers including epithelial membrane antigen (EMA) and CD99. In general, the Olig2-positive nuclei were only sparsely distributed in ependymomas; in contrast, they were very numerous in astrocytomas and oligodendrogliomas. Particularly in cases of glioblastoma or pilocytic astrocytoma that histologically mimicked ependymoma, the Olig2-positive nuclei were numerous as in conventional astrocytomas, which helped to differentiate them from ependymomas. The EMA-positive structures were helpful for the diagnosis of ependymoma, however, they were occasionally very modest and sparse on immunostained sections. A quantitative study showed that the Olig2-positive nuclei were much fewer in ependymomas than in astrocytomas and oligodendrogliomas. These results indicate that the Olig2-immunohistochemistry is useful and potentially more reliable than the EMA-immunohistochemistry for the diagnosis ofependymoma. CD99 is a cell surface antigen expressed in some tumors, most notably in Ewings sarcomas. In our preliminary experiment, we noted the absence of CD99-immunoreactivity in a fraction of brain tumors with clear cell morphology, including oligodendroglioma, clear cell ependymoma, and pilocytic astrocytoma (the oligodendroglioma-like component). Thus, we investigated the expression of CD99 in an additional series of brain tumors with clear cell morphology, including oligoastrocytoma (7 cases), central neurocytoma (6), and dysembryoplastic neuroepithelial tumor (9). We found that the absence of CD99-immunoreactivity was dependent on clear cell morphology rather than on tumor entities. The CD99-immunohistochemistry is unique in that it is helpful for the diagnosis of clear cell brain tumors through the visualization of CD99-negative clear cells.

Expression of E-cadherin and catenins in meningioma: Ubiquitous expression and its irrelevance to malignancy

The expression of cell adhesion molecules in 107 meningiomas was analyzed with immunohistochemical methods using antibodies to epithelial (E)‐cadherin and catenins (α, β and γ). According to the provided World Health Organization (WHO) grading, 84, 18 and five cases were classified as grade I, II and III, respectively. In addition, hemangioblastoma (15 cases) and hemangiopericytoma (four cases) were also evaluated. In most meningiomas, E‐cadherin, α‐ and β‐catenins were expressed along the cell membrane or inside the cytoplasm. The tumor cells constituting whorls and glandular structures of secretory type showed a strong immunoreactivity. γ‐Catenin expression tended to be weak and infrequent in fibrous meningiomas, while other types exhibited diffuse stainings. Even in meningiomas of more than grade II, the expressions of cell adhesion molecules were detected in all cases. Hemangiopericytoma was positive for α‐ and β‐catenins, and hemangioblastomas were positive for β‐catenin alone, which was distinct from the expression pattern in meningiomas. Quantitatively, there were no correlations between the histological variants, Ki‐67 indexes, or grades of meningiomas and the immunoreactive scores except for γ‐catenin scores of fibrous meningiomas. The present study demonstrates that cell adhesion molecules are ubiquitously expressed in all variants of meningioma and may be involved in the tumor morphogenesis. This result suggests that the expression of cell adhesion molecules is not a reliable indicator of malignancy in meningiomas. The present study also suggests that these markers may be useful for the differential diagnosis of meningioma.

Immunoexpression of 14-3-3 proteins in glial cytoplasmic inclusions of multiple system atrophy

Glial cytoplasmic inclusions (GCIs) are the histological hallmark of multiple system atrophy (MSA). In six postmortem brains of patients with MSA, 14-3-3-protein immunoreactivity was identified in GCIs predominately in the white matter tissue of the basal forebrain and cerebellum. Using double immunohistochemistry, co-localization of 14-3-3-protein and α-synuclein immunoreactivities in the GCIs was confirmed. The immunolabeling rate of GCIs with 14-3-3 proteins varied regionally from approximately 40% to 90%. Semiquantitative analysis yielded a significant negative correlation between degree of tissue degeneration and density of 14-3-3-protein-immunoreactive GCIs. The 14-3-3 proteins are active cofactors involved in cellular regulation through binding to phosphorylated motifs in target proteins and α-synuclein is a known target of 14-3-3. Our study suggests that 14-3-3 proteins are closely associated with α-synuclein in GCIs and 14-3-3 proteins may be candidate cofactors of α-synuclein in GCI formation.

Dedifferentiated liposarcoma with rhabdomyoblastic differentiation

Dedifferentiated areas of dedifferentiated liposarcoma (DDL) usually show malignant fibrous histiocytoma (MFH)- or fibrosarcoma-like features and lack any histologic signs of specific differentiation. However, some reports have demonstrated specific differentiation in these areas, with histologic features resembling those of rhabdomyosarcoma, leiomyosarcoma, and osteosarcoma. We report here a pathologic and genetic analysis of three cases of DDLs with rhabdomyosarcomatous areas. MFH- or fibrosarcoma-like areas of one primary DDL and two recurrent DDLs contained various amounts of rhabdomyoblasts, which were immunoreactive for desmin, myoglobin, muscle actin (HHF-35), and myogenin. An ultrastructural examination demonstrated rhabdomyoblasts with abundant cytoplasm containing thin and thick filaments and Z-bands. By real-time PCR, amplification of mdm2 and cdk4 was confirmed in both well-differentiated and dedifferentiated areas with rhabdomyoblasts of all cases. To our knowledge, only seven cases of DDLs with rhabdomyosarcomatous components have been reported, and furthermore, the genetic profiles of the rhabdomyosarcomatous components in DDLs have not been investigated. This study demonstrates that DDLs with rhabdomyosarcomatous areas have genetic alterations that are common to well-differentiated/dedifferentiated liposarcomas.