Immunophenotypic characterization of infiltrating polynuclear and mononuclear cells in childhood brain tumors

During a systematic cell surface antigen expression profile analysis of 76 primary childhood brain tumors (34 medulloblastomas/primitive neuroectodermal tumors and 42 astrocytomas), we employed the following library of monoclonal antibodies (MoABs): anti-Leu-2/a; anti-Leu-3/a; anti-Leu-M5; anti-Leu-11b; anti-HLA-A, -B, -C; anti-HLA-DR; anti-HLe-1 (leukocyte common antigen); and UJ 308. The MoABs identified the expression of various leukocyte-associated, lymphocyte cell line differentiated, cell surface antigens in childhood brain tumors. The antigens were detected with an indirect, biotin-streptavidin-conjugated alkaline phosphatase (AP) immunocytochemical technique. Leu-2/a+ cells comprise the significant CD8+ cytotoxic T-lymphocyte (CTL) population of the tumor-infiltrating lymphocytes. CTLs are major histocompatibility complex (MHC) class I restricted, tumor-associated antigen-specific, cytotoxic cells and were identified in 58 of 76 (76.32%) brain tumors. CTLs usually represented 1-10% of all cells, but in some cases 30-44% of the cells were CD8+. CD4+, MHC class II restricted helper lymphocytes, detected with MoAB anti-Leu-3/a, were present in 65 of 76 (85.53%) brain tumors. Usually 1-10% of the observed cells reacted with MoAB anti-Leu 3/a. Macrophages (Leu-M5 antigen-positive cells) were expressed in 74 of 76 (97.37%) brain tumors. Their number also represented 1-10% of all observed cells in the frozen brain tumor sections. All 76 (100%) brain tumors contained cells that reacted positively with MoABs anti-HLA-A, -B, -C and anti-HLA-DR, demonstrating a strong MHC class I restriction of the tumor cell population and an overall leukocyte antigen expression.(ABSTRACT TRUNCATED AT 250 WORDS)

p53 immunoreactivity in astrocytomas and its relationship to survival

p53 is a putative tumour suppressor gene implicated in a wide range of human malignancies. Mutation of p53 gene results in a more stable product and increased quantities of p53 protein in the cell. Thus, unlike the normal situation, mutated p53 is detectable by immunohistochemistry. We stained frozen sections of 74 astrocytomas with two antibodies to p53, PAb 1801 and PAb 421. Overall 18/74 (24%) of astrocytomas showed p53 immunoreactivity. Fifteen of 47 (32%) grade IV were p53 immunopositive, as were 3/16 (19%) grade III, 0/7 (0%) grade II and 0/4 (0%) grade I astrocytomas. These findings are in agreement with previous studies in showing relatively greater numbers of high grade than low grade p53 immunopositive tumours. Although we found an expected difference in survival according to grade, there was no significant difference in survival (p > 0.1) between p53 immunopositive and immunonegative tumours. We conclude that, whilst p53 undoubtedly plays an important role in the molecular 'chain' leading to malignancy in some astrocytomas, within tumours of comparable grade it does not appear to influence survival.

Automated grading of astrocytomas based on histomorphometric analysis of Ki-67 and Feulgen stained paraffin sections. Classification results of neuronal networks and discriminant analysis

In stereotactically obtained astrocytoma biopsies, four morphometric nuclear parameters were determined with the use of an image analysis system. A special Ki-67 (MIB1)/Feulgen stain made it possible to quantify the essential characteristics of gliomas of the astrocytoma/glioblastoma group: growth pattern, cellularity, proliferation tendency and nucleus pleomorphism. A grading scale based on a cluster analysis resembling the WHO-scheme, which is suitable for automated astrocytoma grading, was developed. Large back propagation neural networks were used and their results compared with those of a classical multivariate discriminant classification analysis. It is possible to show that the neural network technology is superior to the statistical approach for automated astrocytoma grading. Based on the results of our study we believe neural network technology to be useful for tumour grading problems. The presented approach can be generalized for the automated grading of other tumour entities.

Prognostic significance of p53 immunoreactivity in adult patients with supratentorial fibrillary astrocytic neoplasms

The prognostic significance of p53 immunoreactivity in adult patients with supratentorial fibrillary astrocytic neoplasms was examined by Kaplan-Meier survival analysis. Using a monoclonal antibody that reacts with both mutant and wild-type p53 protein (PAb 1801), reactivity was assessed immunohistochemically in specimens from the first diagnosis of astrocytic neoplasm in 95 patients: 26 astrocytomas (A), 19 anaplastic astrocytomas (AA), and 50 glioblastomas multiforme (GBM). Overall, 53% of cases exhibited any p53 nuclear immunoreactivity, with approximately the same proportion in each histologic grade. Survival was measured from diagnosis to death or last follow-up and ranged from 3 months to 9 years. Histologic grade was a powerful prognostic variable for this group of patients (p < 0.001), with median survivals of 88, 18, and 9 months for A, AA, and GBM patients, respectively. In contrast, patients with p53-immunoreactive or -nonimmunoreactive neoplasms had median survival times of 18 or 15 months, respectively (p = 0.21). These results indicate that p53 immunoreactivity was not prognostically significant in this group of adult patients with supratentorial fibrillary astrocytic neoplasms, although a small difference in survival cannot be excluded.

Prognostication of astrocytoma patient survival by Ki-67 (MIB-1), PCNA, and S-phase fraction using archival paraffin-embedded samples

The prognostic power of three proliferation estimation methods, Ki-67 (MIB-1) and PCNA immunohistochemistry, and flow cytometry (S-phase and S + G2/M fractions, respectively), were evaluated in 50 cases of astrocytoma. Each proliferation index showed a strong association with the grade of malignancy (grades I-IV). The MIB-1 labelling index (LI) provided additional information, as it could be used for the discrimination of grade II and grade III astrocytomas (P = 0.0357). All three proliferation estimation methods also had strong prognostic potential (MIB-1 LI: P < 0.0001; PCNA Li: P < 0.0001; S-phase: P = 0.0004; S + G2/M: P = 0.0124). According to the receiver operating characteristics (ROC) curve, the MIB-1 LI showed generally the best sensitivity and specificity in placing the patients correctly into groups of survivors and non-survivors, which was further confirmed in the multivariate analysis. Only 4 per cent of the patients having high MIB-1 scores (> 15.3 per cent) were alive after 2-years' follow-up. In contrast, 72 per cent of patients with tumours of low proliferation activity survived. It appears that Ki-67 (MIB-1) immunolabelling using archival paraffin-embedded samples is of value in predicting prognosis in astrocytic tumours.

Shared amino acid sequences in the ND beta N and N alpha regions of the T cell receptors of tumor-infiltrating lymphocytes within malignant glioma

The purpose of this study was to assess the V-(D)-J junctional region of the T cell receptor (TCR), the CDR3 region, which is responsible for glioma-specific antigen contact in alpha beta TCR-mediated recognition. We sequenced the TCR alpha and beta chains of V alpha 7, and V beta 13.1 cDNA derived from tumor-infiltrating lymphocytes (TIL) of 12 glioma patients and also the corresponding clones from the patients' peripheral blood lymphocytes (PBL). A shared V beta 13.1 DJ sequence of the CDR3 region, ND beta N, was demonstrated in 49 of 66 V beta 13.1+ clones (74.2%) from the glioma TIL, whereas only 4 of 33 clones (12.1%) were observed in the V beta 13.1+ clones from the PBL (p < 0.001). A common VDJ sequence, FCASS (V beta 13.1)-YRLPWGTSDS (ND beta N)-GELFF (J beta 2.2), was observed not only in the gliomas from each patient, but also among all the patients with a preference for V beta 13.1. In contrast, the amino acid sequences of the V beta 13.1+ PBL clones were diverse and random. Next, we sequenced subclones from other V beta subfamilies randomly selected to compare their VDJ region rearrangements (V beta 3 and V beta 5.1). In contrast to V beta 13.1, the amino acid sequences of these junctional regions were completely different in these subclones. The V-J junctional region of the alpha chain is dominated by a few clones in some patients, and no shared amino acid sequences were detected in the TCR V alpha junctional region. However, in the N alpha region of the V alpha 7-bearing TIL clones, arginine was used in 27 of 44 clones (61.4%) compared to only 3 of 12 clones from the PBL (p < 0.05). These results are consistent with the hypothesis that a clonal expansion/accumulation of glioma lineage-specific T cells occurred in vivo at the tumor site and that these T cells may be recognizing glioma-specific antigens.

Patterns of reactivity with anti-glycolipid antibodies in human primary brain tumors

Antibodies against carbohydrates of three glycolipids were used to determine patterns of immunohistochemical reactivity of histologically identifiable cell subpopulations in 101 human primary brain tumors. For all tumor types fibrillary cells, polar cells, and gemistocytes (commonly seen in astrocytomas and ependymomas) stained more frequently for galactosylcerebroside with mAbO1 than small tumor cells and macrophages. Frequency of staining for sulfatide with mAbO4 was fibrillary > polar > small cells = macrophages. Gemistocytes stained more frequently with mAbO4 than polar cells in all tumors except low grade astrocytomas. These data indicate that tumors classified on histological grounds as astrocytic are often stained with antibodies that recognize oligodendrocytes and their progenitors. Thus, anti-glycolipid antibodies used in the study of developmental lineage may offer useful tools for classification of human brain tumors. Staining of fibrillary cells, polar cells, and gemistocytes for paragloboside directly with mAb F1H11 was much less common than with mAbO1, but this increased by pretreatment of the tissues with neuraminidase (F1H11 + N). Of particular note was the finding that small tumor cells frequently stained with F1H11 + N. Evidence that these were not macrophages was obtained using double immunostaining with F1H11 + N and anti-macrophage antibodies. In astrocytomas the frequency of small tumor cells immunostained with F1H11 + N was high grade > anaplastic > low grade, demonstrating a correlation of this tumor cell population with more aggressive astrocytomas. Thus, immunostaining with F1H11 + N may be of value in identifying small, anaplastic tumor cells, especially in small biopsies or tissue taken adjacent to the main tumor mass.

Prognostic significance of proliferating cell nuclear antigen (PCNA) expression in gliomas

The relationship between proliferating cell nuclear antigen (PCNA) expression and various clinicopathological indices (age, sex, tumour location, histological type and grade and treatment) and post-operative survival were studied in patients with central nervous system gliomas using univariate and multivariate analysis. The expression of PCNA (PC10 score) was examined immunohistochemically using the monoclonal antibody PC10 on paraffin sections from 45 cases. Univariate analysis showed that a high PC10 score as well as older age, high histological grade and the histological type (astrocytoma) were associated with reduced survival. However, multivariate analysis revealed that only PC10 score and histological type had independent prognostic significance. The most important feature influencing PC10 score was the tumour grade. Regarding the patients who relapsed, the survival from the time of original diagnosis was related to the relapse-free period, while the PC10 score of the primary tumour emerged as the only independent predictor of survival following the first recurrence. These results indicate that PCNA expression is an independent prognostic indicator in CNS gliomas.

[A study on HLA-Dr expression of brain tumor cells and mononuclear cell subsets infiltrating in these tumors]

Fourty-four primary brain tumors were studied by immunohistochemistry. 29 (67.2%) expressed HLA-Dr. The incidence and number of tumor cells positive for HLA-Dr correlated with the histological type of brain tumor and increased with the degree of malignancy. The mononuclear cells infiltrating in these tumors were mostly CD45ROT cells and macrophages. The former consisted mainly of CD4 and CD8 subsets. The numbers of CD45RO+ and CD8+ T cells in HLA-Dr-group were more than any of the HLA-Dr+, +2, +3 groups. The numbers of CD4 subset and macrophages were not affected by the level of HLA-Dr expression. These results suggest that the HLA-Dr expressed by brain tumor cells selectively inhibit CD8 subset which participates in immunoreaction against brain tumors in situ.

Expression of tumour necrosis factor-alpha, -beta and interferon-gamma genes within human neuroglial tumour cells and brain specimens

Expression of cytokine genes, TNF-alpha, TNF-beta and IFN-gamma, in human astroglial cell lines and in fresh brain specimens was studied by PCR. mRNA transcripts of TNF-alpha could be detected in three out of five astrocytomas and neuroblastoma cell lines, and after stimulation with IL-1 beta/IFN-gamma or LPS/IFN-gamma all these cell lines expressed TNF-alpha genes. TNF-beta genes could not be detected in these cell lines. We were able to detect expression of IFN-gamma genes within two astrocytoma cell lines, which interestingly did not show TNF-alpha activity. In addition to the cultured cells, we also examined gene expression of these cytokines within four human malignant astrocytoma specimens, two peritumoral brain and two autopsied normal brains. The results show that tumour and surrounding reactive lesions express TNF-alpha genes (four of six) but not normal brains. The concentration of these cytokines in the supernatant of cultured cells was measured quantitatively by TNF-alpha, -beta or IFN-gamma ELISA. The combined stimulation of these neuroglial cell lines with IL-1 beta and LPS or IFN-gamma, revealed a high level of TNF-alpha activity. This was especially evident with a neuroblastoma cell line. The concentration of TNF-alpha in the supernatant of the IMR32 neuroblastoma cell line increased markedly upon stimulation with IL-1 beta in both a time- and dose-dependent fashion in the presence of LPS or IFN-gamma. Next, we examined expression of IL-1 beta and IFN-gamma genes in the brain specimens. The result shows that four in six tumour and peritumoral regions expressed IFN-gamma genes and one specimen showed IL-beta gene by PCR. From these experiments it is suspected that neuroglial cell-derived TNF-alpha induced by IL-1 beta of IFN-gamma may participate in local immune reactions of the brain in an autocrine and paracrine fashion.

Induction and regulation of IL-8 and MCAF production in human brain tumor cell lines and brain tumor tissues

In order to elucidate the role of inflammatory cytokines in the central nervous system, we examined the production of two leukocyte chemoattractants, IL-8 and monocyte chemotactic and activating factor (MCAF) in brain tumor cell lines. The glioma cell lines tested exhibited high levels of IL-8 and MCAF mRNA expression upon stimulation with IL-1 or TNF-alpha, while none of the neuroblastoma cell lines expressed these cytokine mRNA. Both IL-8 and MCAF mRNA expression depended on the dose of IL-1 alpha and TNF-alpha and appeared very rapidly, reaching maximal levels at 3-6 hr, with substantial production of these cytokines in the culture supernatants. When various immunosuppressive drugs were tested, glucocorticoids but not other immunosuppressive drugs markedly inhibited the IL-1 or TNF-alpha-induced IL-8 and MCAF mRNA accumulation, suggesting that glucocorticoid is a potent regulator of these inflammatory cytokine production in the neural tissues. In addition, reverse transcription-polymerase chain reaction (RT-PCR) revealed the expression of IL-8 and MCAF mRNA expression in resected brain tumor tissues including glioblastoma, astrocytoma grade 2, ependymoma and medulloblastoma, indicating that these inflammatory cytokines are expressed in vivo.

The generation of nitric oxide participates in gamma IFN-induced MHC class II antigen expression by cultured astrocytoma cells

The effects of interferon gamma (gamma IFN) on the MHCII antigen expression by human cultured astrocytoma cells were investigated. The co-incubation of gamma IFN with T67 astrocytoma cells produced a dose-dependent increase of MHCII antigen expression as evaluated by flow cytometric (FACS) analysis and confocal laser microscopy analysis. The number of MHCII molecules expressed by gamma IFN-pretreated astrocytoma cells was reduced by co-incubation with N omega-nitro-L-arginine methyl ester (L-NAME), a selective inhibitor of the nitric oxide (NO)-synthesizing enzyme. In addition, methylene blue, which inhibits the biological activity of NO acting at the guanylate cyclase level, strongly antagonized the MHCII antigen expression on astrocytoma cells induced by gamma IFN. Furthermore, gamma IFN increased the activity of the inducible isoform of NO-synthase as well as the concentration of nitrite, one of the breakdown products of NO and the antiplatelet activity of astrocytoma cells. In conclusion, the present data show that gamma IFN increases the synthesis and release of NO by cultured astrocytoma cells and this could co-participate in the MHCII antigen expression by this cell type. Therefore, the generation of NO by cultured astrocytoma cells may represent an important step in the development of the immunocompetent activity of astrocytes.

p53 protein expression in central nervous system tumors: an immunohistochemical study with CM1 polyvalent and DO-7 monoclonal antibodies

Formalin-fixed, paraffin-embedded surgical specimens from 137 primary central nervous system tumors, including 26 astrocytomas (21 fibrillary, 1 protoplasmic, 1 gemistocytic and 3 pilocytic), 26 anaplastic astrocytomas, 9 glioblastomas, 1 gliosarcoma, 8 oligodendrogliomas, 4 ependymomas, 1 anaplastic ependymoma, 2 subependymomas, 3 paragangliomas, and 57 meningiomas, were immunostained with the CM1 polyclonal (pAb) and the DO-7 monoclonal (mAb) antibodies against the p53 protein, using the streptavidin/peroxidase method. In addition, two series of 17 and 9 medulloblastomas were also immunostained with the above pAb and mAb, respectively. p53 protein expression was observed in 7 fibrillary astrocytomas, 17 anaplastic astrocytomas, 5 glioblastomas, 1 gliosarcoma, 1 oligodendroglioma, 1 anaplastic ependymoma, and 4 meningiomas with the CM1 pAb. An additional 10 cases (i.e., 3 anaplastic astrocytomas and 7 meningiomas) were found to be p53 protein positive with the DO-7 mAb. Of the medulloblastomas, 8 (of the 17) and 4 (of the 9) were found to express p53 protein with CM1 pAb and DO-7 mAb, respectively. Our results indicate that p53 protein is expressed in a number of central nervous system neoplasms, and suggest that in astrocytic tumors a possible association may exist between p53 protein expression and tumor progression through increasing histological grades of malignancy.

Tumorigenic, invasive, karyotypic, and immunocytochemical characteristics of clonal cell lines derived from a spontaneous canine anaplastic astrocytoma

Tumor cells from a spontaneously arising canine astrocytoma were isolated and cloned. Three clonally derived cell lines (DL3580 clone 1, DL3580 clone 2, and DL3580 clone 3) were developed and found to express glial fibrillary acidic protein (GFAP) as well as epidermal growth factor receptor (EGFR/c-erbB1). The cell lines were tumorigenic as subcutaneous xenografts or as intracranial implants in athymic mice, or both. Both the monolayer astrocytoma cells and the xenograft tumor cells from clone 2 were aneuploid, with a modal number of 84 chromosomes per metaphase; clones 1 and 3 were also aneuploid with modal numbers of 82 and 75/79, respectively. The histology of both the initial spontaneously occurring tumor in the dog and the intracranial astrocytoma in athymic mice demonstrated features of diffuse infiltration into normal brain. These newly developed canine glioma cell lines are karyotypically stable for 1 yr in culture and carry the same marker chromosomes as the parental lines. These glioma cell lines may serve as models for investigating mechanisms of glioma invasion into brain. Additionally, clonal cell lines with divergent properties isolated from the same tumor may assist in studies of the molecular basis of astrocytoma progression and heterogeneity.

Interleukin-6 biosynthesis is increased by histamine in human B-cell and glioblastoma cell lines

An enhancing effect of histamine on the biosynthesis and gene expression of interleukin-6 (IL-6) by human cell lines, Epstein-Barr virus (EBV)-infected human B lymphoma line BMNH and the glioblastoma line SK-MG4 has been found. No similar effect of histamine has been detected on the IL-6 production by any other B-cell line, CESS or human peripheral monocytes. Histamine stimulates the IgM production of BMNH cells by autocrine action of IL-6 induced by histamine, since either neutralization of IL-6 by polyclonal antibody or blocking the IL-6 receptor by specific monoclonal antibody abolished the effect of histamine.

Antigenic and differentiative heterogeneity among human glioblastomas

Increasing evidence suggests that in mammals, astrocytes are a heterogenous family of cells all of which share certain properties, but differ in lineage, biochemical and functional aspects. It seems likely that glioblastomas, arising from glial precursors, may also represent a family of related but distinct cell types. We have examined the antigenic characteristics and differentiative potential of 7 different human glioblastoma cell lines in vitro. All the cell lines were labeled with a monoclonal antibody 7B11 which labels all classes of astrocytes and their precursors in the rat CNS. U138MG and Tm3 cells expressed antigens on their surfaces recognized by the monoclonal antibodies A2B5 and HNK-1. When grown in serum-free medium in the presence of cAMP and theophylline, U138MG cells assumed a process-bearing morphology and some cells expressed the Gal-C antigen specific for oligodendrocytes. Under identical conditions, Tm3 cells converted to process-bearing cells, some of which expressed glial fibrillary acidic protein (GFAP) specific for astrocytes. Other cell lines with similar antigenic characteristics did not respond similarly to cAMP and theophylline. Finally, A2781 cells were GFAP immunoreactive and unlabeled by either A2B5 or HNK-1 antibodies. These observations suggest that individual glioblastoma cell lines may be derived from distinct glial precursor cells in the vertebrate CNS.

Antitumour activity of an immunoconjugate composed of anti-human astrocytoma monoclonal antibody and neocarzinostatin

Neocarzinostatin (NCS) linked to the thiol group on the hinge region of the Fab' fragment of GA-17, a murine monoclonal antibody reacting with tyrosine-specific phosphorylated antigens, which are exclusively expressed on the cell surface of human astrocytomas, was evaluated for in vivo activity. GA-17-NCS immunoconjugates significantly suppressed the growth of human malignant glioma cell line U87-MG subcutaneous xenografts in nude mice until day 50 when administered intravenously into the tail vein. Disulphide- and thioether-linked GA-17-NCS were nearly equipotent immunoconjugates, but thioether-linked GA-17-NCS was more effective than disulphide-linked conjugates with 250 U/kg NCS content on day 50 (P < 0.05). Thioether-linked GA-17-NCS was significantly more effective on day 50 than free NCS with 500 U/kg or 250 U/kg NCS content (P < 0.05, P < 0.01, respectively). These results suggest that GA-17-NCS may prove useful in the treatment of human malignant gliomas.

Distribution of transforming growth factor-beta 1 in human astrocytomas

We used immunohistochemical techniques to study the distribution of transforming growth factor-beta 1 (TGF-beta 1) and infiltrating lymphocytes and macrophages in human astrocytomas. Thirteen of 15 grade 4 astrocytomas (glioblastomas) showed staining with anti-TGF-beta 1 antibody, predominantly in proliferating endothelial complexes and surrounding small and medium-sized blood vessels. Brain tissue microscopically free of tumor cells (n = 8) and more differentiated astrocytomas of varying grade (1 to 3; n = 6) devoid of endothelial proliferation did not stain with anti-TGF-beta 1. Normal brain contained only rare lymphoreticular cells. The majority of astrocytomas studied, however, contained T lymphocytes and macrophages with smaller numbers of B lymphocytes. The lymphoreticular infiltrates were concentrated primarily in close proximity to blood vessels. Within an individual tumor perivascular regions staining for TGF-beta 1 never contained more than occasional T lymphocytes. Perivascular regions not staining for TGF-beta 1 frequently contained low to high numbers of T lymphocytes. The inverse relationship in the distribution of TGF-beta 1 and lymphocyte infiltrates is compatible with a functional relationship between this cytokine and an immune effector cell response to glioblastomas.

Effect of the expression of transforming growth factor-beta 2 in primary human glioblastomas on immunosuppression and loss of immune surveillance

Glioblastomas are malignant brain tumors that are attended by an immunosuppressed state. The authors have studied the expression of transforming growth factor-beta 2, which is known to have potent immunosuppressive and angiogenic properties. Transforming growth factor-beta 2 messenger ribonucleic acid and its protein product are both found to be greatly overexpressed in these tumors and are absent from normal brain tissue. The overexpression of this growth factor may contribute to the escape of neoplastic astrocytes from immune surveillance and, furthermore, to the immunosuppressed state that is characteristic of many of these patients.

Generation of cytotoxic immune responses against a rat glioma by in vivo priming and secondary in vitro stimulation with tumor cells

Cytotoxic T lymphocyte (CTL) responses to most antigens are generated by in vivo priming and secondary stimulation with antigen in vitro. The present studies were designed to determine whether that strategy could be used to stimulate development of CTL against brain tumors. Rats were primed with one of two tumors, RT2, an astrocytoma, or 9L, a gliosarcoma, and Corynebacterium parvum. Spleen cells from primed rats were stimulated with tumor cells and interleukin-2 in vitro to generate CTL. CTL generated against RT2 killed RT2 and 9L, but not allogeneic or histopathologically unrelated tumor cells, suggesting that the killing was brain tumor-specific and major histocompatibility complex gene product-restricted. Similar results were obtained with rats primed and secondarily stimulated with 9L. Specific cytotoxic cells only developed when syngeneic brain tumor cells were used for both priming and secondary stimulation. The cytotoxic cell populations were composed of OX-19+ T cells with a mixed CD4/CD8 phenotype. Controls consisting of spleen cells from unprimed or primed rats tested before culture exhibited low levels of cytotoxicity against brain tumor targets. Culturing unprimed or primed cells with interleukin-2 alone stimulated cell proliferation, but the cells that grew out exhibited only low levels of cytotoxicity for brain tumor cells. Cell populations exhibited consistent cytotoxicity against natural killer cell targets. None of the cell populations killed lymphokine-activated killer cell targets. The results demonstrated that brain tumor-specific CTL could be produced by priming in vivo followed by secondary stimulation with brain tumor cells in vitro. The results further demonstrated that RT2 and 9L share antigens that both induce and serve as target structures for specific cytotoxic cells.

Human glioma-specific antigens detected by monoclonal antibodies

Three murine monoclonal antibodies, designated GA-17, GB-4, and GC-3, were prepared by the hybridization of murine myeloma cells (NS-1) and spleen cells of BALB/c mice immunized with the crude membrane fraction of cultured human gliosarcoma cells (GI-1). Two of them (GA-17 and GB-4) reacted exclusively with the membrane of glioma cells, and the other (GC-3) reacted with the membrane of glioma cells and a T cell line (MOLT-4). Although these antibodies reacted with almost all of the gliomas, the reactions differed. GA-17 reacted equally well with all glioblastoma (17 cases) and low-grade astrocytoma (10 cases), whereas GB-4 reacted poorly with 7 cases of glioblastoma and GC-3 did not react with 7 cases of low-grade astrocytoma. The antigens, exclusively expressed on the cell surface, were analyzed by surface labeling with 125I followed by a cell lysis and immunoprecipitation with these antibodies. The findings obtained by sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that GA-17, GB-4, and GC-3 reacted with Mr 140,000-145,000, Mr 160,000, and Mr 145,000-150,000 proteins, respectively. Some evidence has been obtained indicating that these antigens are composed of the same polypeptide chain (Mr 120,000) with the carbohydrate chains being different.

Epithelial and melanoma antigens in gliosarcoma. An immunohistochemical study

Gliosarcomas are mixed tumors with malignant glial and mesenchymal elements. The number of GFAP-positive tumor cells decreases with the increase of sarcomatous components, until whole areas may be GFAP negative. These distinct differentiations may, however, lead to false interpretations in small tissue samples. In this connection, it is of interest that, according to other reports, glial tumors may be positive for different anti-keratin antibodies and this prompted us to undertake a systematic investigation of the immunoreactivity of gliosarcomas using a panel of well-characterized monoclonal antibodies against cytokeratins (KL1, AE 1/3, Lu-5, CK-19, CK MNF 116 and Ma-903). These cases were further studied with the anti-epithelial non-cytokeratin antibodies EMA, HEA 125, Ber-EP4, CEA as well as the melanoma-antibody HMB-45, Leu-M1, GFAP and vimentin. As screening study we examined 20 cerebral metastatic carcinomas, 21 malignant gliomas (including 6 gliosarcomas) and 3 metastatic melanomas with the monoclonal antibodies KL1 and HMB-45. All cerebral metastatic carcinomas and 4/6 gliosarcomas were positive for KL1, whereas all melanomas, 2 metastatic carcinomas and 3 gliosarcomas showed an immunostaining with HMB-45. All gliosarcomas were positive with at least one of the tested anti-cytokeratin antibodies. The gliosarcomas did not show an immunoreaction in any of the cases when CEA, HEA 125, Ber-EP4, EMA or Leu M1 were applied. In our opinion, the monoclonal antibodies HEA 125 and Ber-EP4 could obviously be helpful in differentiating gliosarcomas from metastatic carcinomas.

Modulation of the immune response by transforming growth factor beta

For the past several years immunologists have been fascinated by a series of experiments showing that transforming growth factor beta (TGF beta) suppresses T- and B-lymphocyte growth as well as IgM and IgG production by B cells. Moreover, while exerting chemotactic activity on monocytes and inducing expression of interleukin-1 and interleukin-6 by these cells, TGF beta interferes with bacterially induced tumor necrosis factor alpha production, oxygen radical formation and the adhesiveness of granulocytes to endothelial cells. These mechanisms may provide the basis for the effect of TGF beta to prevent the microvascular changes associated with brain edema formation in bacterial meningitis. Given the potential of lymphocytes as well as macrophages to produce TGF beta 1, this cytokine may exert negative feedback signals on the immune response, provided the cytokine is processed from its latent form to the bioactive homodimer. Potent effects of TGF beta have been observed in experimental animals including the inhibition of the generation of virus-specific cytotoxic T cells and antiviral antibodies as well as the diminution of cellular infiltrates with decreased major histocompatibility complex class-II expression and CD8+ T cells in the tissue of virally infected animals. TGF beta may also be of importance in tumor immunology. By the production of bioactive TGF beta as detected in glioblastoma and acute T-cell leukemia, tumor cells may induce an immunodeficiency state and escape immune surveillance. In inflammation, monitoring of TGF beta in the tissue will bring light on the immune regulation in acute and chronic inflammatory diseases.

Platelet-derived growth factor receptor (beta-subunit) immunoreactivity in soft tissue tumors

Using a well characterized monoclonal antibody (PR7212) to the beta-subunit of the platelet-derived growth factor receptor (PDGF-R(beta) and the avidin-biotin peroxidase method on frozen sections, we analyzed PDGF-R(beta) expression in 71 nonepithelial lesions as well as normal mesenchymal tissues. PDGF-R(beta) reactivity was observed in normal salivary gland, normal cutaneous and visceral fibroblasts, muscularis mucosa of bowel, and endothelial cells; squamous carcinoma was negative. Interestingly, hepatocytes and lymph node histiocytes were also positive. Positive tumors included malignant fibrous histiocytoma (6/6), benign and malignant smooth muscle tumors (5/6 leiomyoma, 8/9 leiomyosarcoma), liposarcoma (4/4), synovial sarcoma (6/7), angiosarcoma (2/2), and sarcoma NOS (2/2). Fibromatosis cases were also positive (2/2). In many tumors, the reactive fibroblasts and vascular components were also reactive. The characteristic pattern of reactivity in fibroblastic lesions highlighted thin cytoplasmic extensions or strands not visible in normal hematoxylin and eosin-stained sections. Expression of PDGF-R(beta) was not necessarily correlated with the presence of PDGF. We conclude that PDGF-R(beta) expression can be identified in a wide variety of mesenchymal lesions and postulate that its presence may be important in the mechanism of growth of these tumors.