Immunoradiometric and Immunohistochemical Demonstration of Neuron-Specific Enolase in Experimental Rat Gliomas1

[CANCER RESEARCH 44, 2595-2599, June 1984]

Immunoradiometric and Immunohistochemical Demonstration of Neuron-specific Enolase in Experimental Rat Gliomas1

Stanley A. Vinores,2 Paul J. Marangos, Jose M. Bonnin,3 and Lucien J. Rubinstein

Division of Neuropathology, Department ol Pathology, University of Virginia School of Medicine, Charlottesville, Virginia 22908 [S. A. V., J, M. B., L J. R.], and Biological Psychiatry Branch, National Institute of Mental Health, NIH, Bethesda, Maryland 20205 [P. J. M.J

ABSTRACT demonstration of NSE in the tumors obtained following their i.e. inoculation. A number of ENU-induced neural tumors in rats were A number of neural and nonneural tumor cell lines of rat and also examined immunohistochemically for the presence of NSE. human origin were assayed for neuron-specific enolase (NSE) by radioimmunoassay. Most neural tumor cell lines had appreciably MATERIALS AND METHODS higher levels of NSE than did the nonneural tumor cell lines, the highest levels being found in two anaplastic rat glioma lines (F98 Cell Lines. A number of neural and extraneural tumor cell lines of rat and T24). These two lines contained more than twice the amount and human origin were obtained (see Tables 1 and 2) for the determi of NSE found in a rat pheochromocytoma line (PC12) and in nation of their NSE content. The tumors from which the rat lines were neuroblastoma lines derived from rats (B35 and B50) or humans derived included anaplastic gliomas, malignant schwannomas, spongio- (IMR-32 and SHSY-5Y). Several of the rat glioma and schwan- blastoma, astrocytomas, mixed gliomas, neuroblastomas, pheochromo noma lines were inoculated intracerebrally into syngeneic rats. cytoma, hepatoma, and mammary adenocarcinoma. The human lines In the resulting tumors, NSE was demonstrable by immunohis- were derived from 2 neuroblastomas and one mammary adenocarci noma. Lines C6 and IMR-32 were obtained from the American Type tochemistry only in those from the F98 and T24 cell lines. A Culture Collection (Rockville, MD); line SHSY-5Y from Dr. J. Biedler (New number of ethylnitrosourea-induced rat tumors were also ex York, NY); clone PC12 from Dr. R. Goodman (Philadelphia, PA); clones amined immunohistochemically for NSE: NSE was demonstrated F98 and D74 from Dr. W. Wechsler (Dusseldorf, Germany); lines B35 in three anaplastic gliomas; three astrocytomas; and two mixed and 850 from Dr. D. Schubert (San Diego, CA); lines T9, T13, TU, T22, gliomas. Reactive astrocytes were also positive. Fibroadenomas T24, Y1196, Y2116, Y2121, 78-3853, 78-3864, and 78-4836 from Dr. of apocrine and mammary glands in rats were weakly positive, A. Koestner (East Lansing, Ml); line Q7 from Dr. J. Chou (Bethesda, MD); but other extraneural tumors tested were negative. and lines 64/24 and MCF7 from Dr. C. Eil (Bethesda, MD). Since normal neuronal elements, axonal swellings, and amine Inoculations i.e. of Tumor Cell Lines. Nine of the glioma lines and 2 precursor uptake and decarboxylation cells are strongly positive of the schwannoma lines (Table 1) were inoculated i.e. into the brains of syngeneic male Fischer rats weighing 90 to 110 g (Charles River, for NSE, whereas glia and most other normal cells are negative, Portage, Ml) as described previously (35). PC12 cells were inoculated in we hypothesize that the elevated metabolic demands imposed the same manner into New England Deaconess rats (Deaconess Hospi on neoplastic and reactive glial cells and on some extraneural tal, Boston, MA). Some of these rats were inoculated s.c. with PC12 tumors necessitate the opening up of metabolic pathways that cells for comparison. The tumor-bearing brains and, when present, the are normally operative only in neurons and neuroendocrine cells, visceral mÃ©tastases, as well as s.c. PC12 pheochromocytomas, were therefore resulting in the synthesis of the more stable neuron- fixed by immersion in 10% buffered formalin after necropsy. specific form of enolase. Induction of Tumors by ENU. Tumors were induced transplacentally by a single injection of 20-mg/kg body weight of ENU in citrate buffer, pH 4.5, via the sciatic vein of pregnant Fischer rats on the 18th day of INTRODUCTION gestation or of ENU, 50 mg/kg, via the tail vein on the 16th day of NSE4 has been reported to be uniquely contained in all classes gestation. The tumors that developed in the offspring, usually between 150 and 500 days after birth, were fixed as described above after the of neurons (13, 21, 26) and in normal and neoplastic cells of the rats were sacrificed or had died from their tumors. APUD system (4, 5, 8,19, 25, 27, 30, 40, 41). Its appearance in Perfused Normal Rat Tissues. Normal rat tissues were fixed for 48 normal conditions correlates directly with structural and func hr at room temperature in 10% buffered formalin after intraventricular tional neurogenesis (15,25,37), and the enzyme levels detected gravity perfusion, first briefly with phosphate-buffered saline (0.01 M have been used as an index for neuronal differentiation (11, 19, phosphate buffer, pH 7.6, containing 0.85% NaCI) and then with 10% 25, 39). It is absent from normal glia (21, 24). The purpose of buffered formalin for about 10 min, both at 37Â°C. this study was to determine by immunoradiometric assay the RIA. Cultured cells from the various cell lines tested (Tables 1 and 2) were washed with Ca2+-, Mg2+-free Dulbecco's phosphate-buffered sa levels of NSE in a variety of rat glioma and schwannoma cell line (0.2 g KCI:0.2 g KH2POÂ«:8g NaCI:2.16 g Na2HPO4-7H2O/liter) and lines and to correlate these levels with the immunohistochemical immediately frozen on dry ice in 107 cell aliquots. These were homoge ' Supported by Research Grant CA 31271 of the National Cancer Institute and nized and assayed for NSE using a double-antibody, solid-phase RIA by Neuropathology Research Training Grant T32 NSO7236 of the National Institute procedure as described previously (14, 20, 39). Assays were done in of Neurological and Communicable Diseases and Stroke, United States Department triplicate for each cell line. Cells from rat tumor lines were assayed with of Health and Human Services. 2To whom requests for reprints should be addressed. anti-rat NSE antiserum, and those from human lines were assayed with 3 Recipient of Clinical Fellowship 5732 of the American Cancer Society. anti-human NSE antiserum. Antisera were produced as described pre 4 The abbreviations used are: NSE, neuron-specific enolase; APUD, amine viously (17). precursor uptake and decarboxylation; GFA, glial tibnllary acidic; RIA. radioimmu Immunohistochemistry. All tissues from the i.e. implants and ENU- noassay; ENU, ethylnitrosourea; TBS:NGS, 0.05 M Tris buffer, pH 7.6, containing 0.9% NaCI solution and 1% normal goat serum; i.e., intracerebral(ly). induced tumors, as well as normal rat tissues, were embedded in paraffin, Received November 22, 1983; accepted February 24, 1984. sectioned, and stained for NSE by the peroxidase:antiperoxidase tech-

JUNE 1984 2595

Tabtel with fresh buffered saline (except after the last centrifugation). Diluted Detection of NSE in rat neural tumors by RIA and peroxidase:antiperoxidase anti-NSE serum was added to the pellet in a ratio of 1 ml of 1:500 Cultured cells were assayed by RIA for NSE In aliquots of 107 cells. The standard antiserum/100 mg of washed liver powder (dry weight). This was incu curve for NSE detection ranged from 100 pg to 5 ng (20). The minimal detectable bated for 30 min at 37Â°and centrifuged as above. The supernatant was level was 100 pg/ml of solution. The resulting tumors were assayed by peroxidase:antiperoxidase following I.e. inoculation of these cells. Paraffin-embedded added to 50 mg of washed liver powder (dry weight)/ml, incubated, and sections were incubated with a 1:500 dilution of rabbit anti-rat NSE serum. centrifuged as before. The supernatant was then applied to deparaffin- ized sections as described above. of NSE by from which the NSE levels in out- peroxidase: cell tured cells (ng/mg antiperoxidase in CelllineF98T24T9Y211678-3864T13Y1196T22Y2121C6T1478-385378-4836D74PC12TumorderivedAnaplastialine was implants1407.1protein) i.e. RESULTS gliomaAnaplastic Â±1330.0 glioma0Anaplastia Radioimmunoassay Â±751 gliomaAnaplastic .3Â±61 gliomaMalignant 2.2Â±271.1 The quantitative values for NSE obtained from several cultured schwannomaMalignant Â±754.9 lines of neural tumors (Tables 1 and 2) ranged from a low of schwannomaMalignant Â±373.4 schwannomaSpongioblastomaAstrocytomaAstrocytomaMixedÂ±741 267.7 Â±75.6 (S.D.) ng/mg protein in the C6 astrocytoma to a .7Â±470.1 high of 1407.1 Â± 102.8 ng/mg protein in the F98 anaplastic Â±267.7 glioma. Most neural tumor lines had NSE levels that were appre Â±1041.7 gliomaMixed Â±421 ciably higher than those of the 3 nonneural tumors examined. gliomaMixed .6Â±289.8 However, the cells from one rat malignant schwannoma, 2 rat gliomaDifferentiated Â±820.4 gliomas (including the C6 astrocytoma), and one human neuro gliomaPheochromocytomaDetectionÂ±649.1 Â±102.8a58.817.96.14.214.222.626.221.375.65.428.12.335.244.9+"+â€”â€”â€”â€”NTâ€”â€”â€”NTâ€”â€”NT-blastoma had low levels of NSE that were within the range of ' Mean Â±S.D. that of the rat hepatoma. " +, positive; -, negative; NT, not tested. c The T24 line was derived from a tumor which was originally diagnosed as an Immunohistochemistry astrocytoma, but the absence of GFA protein in the implants grown after numerous in vitro passages has prompted us to call the tumor an anapÃ¬asticglioma. Controls. No staining was seen in any tissue when normal Table 2 rabbit serum was substituted for NSE antiserum. When sections NSE levels by RIA in peripheral neuroblastomas and in nonneural tumors of neural or extraneural tumors were incubated with NSE anti- Cultured cells were assayed by RIA for NSE in aliquots of 10' cells. serum that had been preadsorbed to hydrated liver powder, the from which the cell line (ng/mg staining intensity was reduced proportionally to the resulting CelllineB35 derivedNeuroblastomawas protein)653.3 increased dilution, but not eliminated. Â±46.1* Normal Cells. With the exception of Purkinje cells, both central B50 Rat Neuroblastoma 641 .5 Â±22.3 and peripheral neurons (Fig. 1) and their processes were con Q7 Rat Hepatoma 255.7 Â±26.6 64/24IMR-32 Rat Mammary adenocarcinoma 3.6 Â± 0.5 sistently positive for NSE in normal or tumor-infiltrated tissues. Human Neuroblastoma 390.4 Â±49.6 Axonal swellings were conspicuously stained. NSE was not SHSY-5Y Human Neuroblastoma 282.8 Â±31.3 MCF7SpeciesRatHumanTumor Mammary adenocarcinomaNSE 141.0 + 21.9 demonstrable in normal glial cells or in the melanocytes of the a Mean Â±S.D. normal skin. The adenohypophysis, the pineal gland, and the adrenal medulla were strongly positive. nique (29). Sections were incubated with 10% normal goat serum in 0.05 In normal perfused rat kidney, the nuclei and walls of the M Tris buffer, pH 7.6, containing 0.9% NaCI solution (saline) for 30 min collecting tubules stained for NSE, as did the macula densa and at room temperature. This was followed by overnight incubation at 4Â° the juxtaglomerular cells. Foci of endothelial cells in the thin loop with anti-rat NSE rabbit serum at a dilution of 1:500 in TBSrNGS. Controls of Henle also stained positively. In the intestine, some of the were incubated with the same dilutions of normal rabbit serum. Sections cells in the lamina propria were positive. Positive staining was were warmed to room temperature for 1 hr, following which they were also found in some of the blood vessel walls, in smooth muscle rinsed briefly using TBSiNGS and washed in Coplin jars containing Tris- fibers, and in macrophages present in various tissues. buffered saline. They were then incubated with a 1:40 dilution of goat Reactive Astrocytes. Reactive astrocytes stained positively anti-rabbit IgG serum in TBS:NGS for 30 min, washed as before, and in the areas adjacent to both primary ENU-induced central nerv incubated for 30 min with a 1:100 dilution of peroxidase:antiperoxidase conjugated to rabbit IgG. They were washed in 0.05 M Tris buffer, pH ous system tumors and tumors resulting from the i.e. inoculation 7.6, and incubated in the dark with freshly made 0.07% 3,3'-diamino- of tumor cell lines. benzidine-4 HCI in 0.05 M Tris buffer, pH 7.6, containing 0.0185% H2O2 Implants i.e. Of the neural lines inoculated i.e. (Table 1), only for 10 to 30 min (until a brown color appeared). Some sections from rat the 2 containing the highest in vitro levels of NSE (F98 and T24 tumors and normal rat tissues were subsequently treated with 1% Os04 anaplastic gliomas) showed immunohistochemically demonstra in H2O for 1 min and rinsed with H2O. All sections were lightly counter- ble NSE in the i.e. tumor implants. In the T24 tumor implants, stained with hematoxylin. NSE-containing cells were seen only at the periphery of areas of A number of sections were incubated with a 1:500 dilution of anti- necrosis. In the F98 tumor implants, positive NSE staining was NSE serum that had been preadsorbed to pig liver acetone powder (ICN, seen in polyhedral cells that contained moderately abundant Plainview, NY; Catalogue NO. 101002) according to the following pro cytoplasm but no cell processes (Fig. 2). cedure in order to eliminate nonspecific binding. Liver powder was ENU-induced Tumors. In experimental rat tumors, positive hydrated with 0.01 M Tris:0.01 M sodium phosphate buffer, pH 7.6, containing 0.85% NaCI solution (buffered saline) for 2 hr on ice. This was staining was found in 3 of 6 anaplastic gliomas and in 3 of 6 followed by three 30-min incubations with the buffered saline at 37Â°, astrocytomas (Table 3). Two of these astrocytomas, described followed each time by a 10-min centrifugation (12,000 x g) in a Beckman in detail elsewhere (28), were positive for GFA protein by both Microfuge 12, after which the supernatant was removed and replaced immunofluorescence and immunoperoxidase (6). Of 40 mixed

2596 CANCER RESEARCH VOL. 44

Table3 obtained after i.e. inoculation of the lines. Assuming that the NSE Immunohistochemical demonstration of NSE in ENU-induced rat neural tumors levels detected by RIA on various neural tumor cell lines in vitro Immunoperoxidase staining for NSE was done in paraffin-embedded sections are comparable to the levels present in i.e. implants, the threshold of experimental rat neural tumors. Sections were incubated with a 1:500 dilution of rabbit anti-rat NSE serum. for the detection of NSE by immunohistochemistry using our

Tumors containing procedure appears to be between 755 and 1330 ng/mg protein. NSE-positive cells/ This holds true even for the PC12 pheochromocytoma, an APUD total tumors tumor. When implanted s.c., the PC12 cells in the implants and Tumor type examined in the resulting lung mÃ©tastases were immunohistochemically Anaplastia gliomas 3/6 strongly positive; however, the cells in the i.e. implants were Astrocytomas 3/6 Diffuse meningea! gliomatosis 0/1 negative. The reason for this difference is not clear. It may Spongioblastoma 0/1 perhaps be correlated with the findings in another report (30), Oligodendrogliomas 0/9 Ependymomas 0/6 according to which NSE was detected immunohistochemically in Mixed gliomas 2/40 primary human APUD tumors, even though the concentration of Malignant schwannomas 0/13 NSE in these tumors calculated by RIA was lower than the Moderately differentiated schwannoma 0/1 threshold of immunohistochemical positivity that we estimated in the i.e. implants of rat tumor cell lines. Table 4 It is difficult to compare the values for the NSE levels we found Immunohistochemical demonstration of NSE in extraneural tumors Immunoperoxidase staining for NSE was done in paraffin-embedded sections in the tumor cells with those reported in normal brain tissue of experimental rat and mouse extraneural tumors. Sections were incubated with because of the considerable developmental and regional differ a 1:500 dilution of rabbit anti-rat NSE serum. Single tumors of each type were ences that have been demonstrated in the brain. NSE is not examined except for the mouse teratoma, of which 3 were examined. Rat tumors were induced by END. Mouse teratomas were derived from the OTT-6050 (3, 7, detectable in the brains of 11-day-old rat embryos. At birth, the 31, 32). The mouse lymphoma was a spontaneous tumor arising in a NIH nude level of NSE in whole brain is 600 ng/mg soluble protein, but this mouse. increases to 11,000 ng/mg soluble protein by 40 days of age. Demonstration of NSE Species Regional differences are also apparent: newborn rats have NSE NSE positive levels of 600 ng/mg soluble protein in the forebrain and of 1800 Hyalinizing fibroadenoma of the apocrine skin gland Rat ng/mg soluble protein in the brain stem (15). Differences are also Fibroadenoma of the mammary gland Rat apparent under culture conditions. Dissociated 17-day-old rat NSE negative embryonic cerebra, which initially contain 260 ng of NSE/mg of Bronchial adenoma Rat Arrhenoblastoma (Sertoli cell tumor of the ovary) Rat soluble protein, reach a maximum level of 650 ng of NSE/mg of Lipoma Rat soluble protein after 25 days in vitro (23). Myelogenous leukemia Rat The immunohistochemical detection of NSE in rat gliomas and Cavernous hemangioma Rat Teratoma Mouse reactive astrocytes indicates that this isoenzyme may be present Lymphoma Mouse in neoplastic or reactive cells, the normal counterparts of which do not contain it. We have also demonstrated NSE in a number gliomas composed of various astrocytic, oligodendroglial, and of human gliomas (glioblastomas, astrocytomas, oligodendro- ependymal elements with varying degrees of anaplasia, only 2 glioma, ependymomas) and in human reactive astrocytes as well had detectable NSE staining. In one of these, the staining oc as in several non-APUD-derived extraneural tumors, such as curred only in a small periventricular area; in the other, it occurred mammary fibroadenoma and carcinoma, renal carcinoma, chor- in one small area adjacent to the ependyma. Other rat central doma, and giant cell tumor of the tendon sheath (34). Using nervous systf ,.i tumors and various malignant schwannomas antiserum to bovine 14-3-2 protein, Haglid ef al. (9) found NSE were negative. to be present in a series of human gliomas and other intracranial Of the extraneural rat tumors tested, only 2 showed weak tumors. 14-3-2 protein was also reported to be found in 2 rat positive staining in the tumor cells: a hyalinizing fibroadenoma of gliomas, composed of neoplastic astrocytes and oligodendro- the apocrine skin glands and a fibroadenoma of the mammary cytes of varying degrees of anaplasia, and in a malignant schwan gland (Table 4). noma (2). 14-3-2 protein and NSE were later shown to be Mouse Teratomas. NSE was not demonstrable in 3 tumor identical, but the cross-reactivity of their antisera is limited to fractions of a transplantable experimental mouse teratoma de only 15% (18). The purity and specificity of the antiserum to 14- rived from the OTT-6050 line (IB-8, IB-9, and IB-2) (3,7, 31, 32), 3-2 protein have not been as well established as in the case of which included extensive areas of neuroepithelial differentiation. the NSE antiserum used in this study. The latter has been well characterized both biochemically and with regard to its cellular DISCUSSION localization (13, 42). Using a different antiserum to NSE, Royds ef al. (22) failed to demonstrate the enzyme in a series of gliomas NSE is a glycolytic enzyme (13, 16) that under normal condi and other intracranial tumors similar to those examined by us tions has been reported to be detectable only in neurons and and previously by Haglid et al. (9). The titer of their antiserum cells of the APUD system (18, 21, 25, 26). The amounts of NSE was not mentioned, but it could have been considerably weaker determined by RIA in the 2 cell lines with the highest levels, both than the antiserum used in the present study. Both our present anaplastic rat gliomas (F98 and T24), were more than twice as findings and those obtained by us on human material (34) have high as those determined in 2 rat and 2 human neuroblastoma led us to the conclusion that NSE is demonstrable in neoplastic lines and in the rat pheochromocytoma line. Of all the neural cell and reactive glial cells in the rat and the human species as well lines tested, only those 2 glioma cell lines had immunohistochem- as in cells of a number of extraneural non-APUD tumors in both ically demonstrable NSE in the cells of the tumor implants species.

JUNE 1984 2597

specific enolase in the Merkel cells of mammalian skin. Am. J. Pathol., 704: In thÃ©more anaplastic gliomas of both humans and rats, NSE 63-68,1981. was demonstrated in a very patchy manner; staining was ob 9. Haglid, K., Carisson, C. A., and Stavrou, D. An immunologicalstudy of human served in clusters of cells in apparently similar stages of differ brain tumors concerning the brain specific proteins S-100 and 14-3-2. Acta Neuropathol., 24: 187-196, 1973. entiation. One of these rat tumor lines, F98, has been morpho 10. Ko, L., and Koestner, A. Morphologic and morphometric analysesof butyrate- logically characterized as an anaplastic glioma (10). It has, how induced alterations of rat glioma cells in vitro. J. Nati. Cancer Inst., 65: 1017- ever, also been shown to respond to nerve growth factor (34- 1027, 1980. 11. Marangos, P. J., Goodwin, F. K., Parma,A., Lauter, C., and Trams, E. Neuron 36), and NSE can be induced in F98 cells in culture by butyrate specific protein (NSP)in neuroblastoma cells: relation to differentiation. Brain (38). In conjunction with those findings, the patchy distribution Res., 745: 49-58, 1978. 12. Marangos, P. J., Parma, A. M., and Goodwin, F. K. Functional properties of of NSE staining in the cerebral implants reported in the present neuronal and glial isoenzymesof brain enolase.J. Neurochem.,37: 727-732, study and the high level of NSE found by RIA in the cells in vitro 1978. suggest that further investigation on the cytogenesis of this line 13. Marangos, P. J., and Schmechel,D. E. The neurobiologyof the brain enolases. Essays Neurochem. Neuropharmacol.,4: 211-247, 1980. and its potential for divergent cellular differentiation may be 14. Marangos, P. J., Schmechel, D. E., Parma, A. M., Clark, R. L., and Goodwin, warranted. F. K. Measurementof neuronal and non-neuronalenolaseof rat, monkey, and Astrocytic tumor cells and reactive astrocytes were found to human tissues. J. Neurochem.,33: 319-329,1979. 15. Marangos, P. J., Schmechel, D. E., Parma, A. M., and Goodwin, F. K. stain positively for both NSE and GFA protein. This has also Developmentalprofile of neuron-specific (NSE) and non-neuronal (NNE) eno been shown to be the case in human material (1, 34). The lase. Brain Res., 790:185-193, 1980. coexistence of both proteins in the same neoplastic cell could 16. Marangos, P. J., and Zomzely-Neurath,C. Determinationand characterization of neuronspecificprotein (NSP)associatedenolaseactivity. Biochem.Biophys. not be ascertained, but the NSE positivity of the reactive astro Res. Commun., 68: 1309-1316, 1976. cytes, which are invariably GFA protein positive, indicates that 17. Marangos, P. J., Zomzely-Neurath, C., and Goodwin, F. K. Structural and immunological properties of neuron specific protein (NSP) from rat, cat, and the presence of the 2 proteins is not mutually exclusive. human brain: comparison to bovine 14-3-2. J. Neurochem., 28: 1097-1107, It seems reasonable to assume that, for cells such as glial 1977. cells, which normally contain only the nonneuronal form of eno- 18. Marangos, P. J., Zomzely-Neurath, C., and York, C. Immunologicalstudies of a nerve specific protein. Arch. Biochem. Biophys., 770: 289-293, 1975. lase (15, 22, 24), the change to a reactive or neoplastic state 19. Odelstad, L., Pohlman,S., Nilsson, K., Larsson, E., Lackgren, G., Johansson, necessitates an increase in metabolic activity. We speculate that K., Hjerten, S., and Grotte, G. Neuron specific enolase in relation to differen tiation in human neuroblastoma. Brain Res., 224: 69-82, 1981. the elevated metabolic rate may result in the synthesis of NSE 20. Parma, A. M., Marangos, P. J., and Goodwin, F. K. A more sensitive radioim- to accommodate the greater metabolic demands imposed on munoassay for neuron-specific enolase suitable for cerebrospinal fluid deter these cells. The greater stability of the neuron-specific form of minations.J. Neurochem.,36: 1093-1096,1981. 21. Pickel, V. M., Reis, D. J., Marangos, P. J., and Zomzely-Neurath, C. Immuno- enolase compared to its other isoenzymes (12) might normally cytochemical localization of nervous system specific protein (NSP-R) in rat be beneficial to neurons and APUD cells in permitting them to brain. Brain Res., 705: 184-187,1976. meet their specific metabolic demands. These demands may be 22. Royds, J. A., Parsons, M. A., Taylor, C. B., and Timpertey, W. R. Enolase isoenzyme distribution in the human brain and its tumours. J. Pathol., 737: similar to those of gliomas and reactive astrocytes, thus account 37-49, 1982. ing for the synthesis of NSE in neoplastic and reactive glia despite 23. Schengrund, C.-L., and Marangos, P. J. Neuron-specific enolase levels in its absence from the normal cells from which these pathologically primary cultures of neurons. J. Neurosci. Res., 5: 305-311,1980. 24. Schmechel, D. E., Brightman, M. W., and Marangos, P. J. Neurons switch altered cells are derived. from non-neuronal enolase to neuron specific enolase during differentiation. Brain Res., 790: 195-214,1980. 25. Schmechel, D. E., Marangos, P. J., and Brightman, M. W. Neuron-specific ACKNOWLEDGMENTS enolase is a marker for peripheral and central neuroendocrine cells. Nature (Lond.),276: 834-836, 1979. We thank Dr. L. Ko for providing pellets from several cell lines to be processed 26. Schmechel, D. E., Marangos, P. J., Brightman, M. W., and Goodwin, F. K. for RIA and A. Parma for technical assistance with the RIAs. Brain enolasesas specific markers of neuronal and glial cells. Science(Wash. DC), 799:313-315,1978. 27. Sheppard, M. N., Johnson, N. F., Cole, G. A., Bloom, S. R., Marangos, P. J., REFERENCES and Polak, J. M. Neuron specific enolase (NSE)immunostaining.A useful tool for the light microscopicaldetection of endocrine cell hyperplasia in adult rats 1. Bonnin, J. M., Rubinstein, L. J., Papasozomenos,S. C., and Marangos, P. J. exposed to asbestos. Histochemistry, 74: 505-513, 1982. Subependymal giant cell astrocytoma. Significance and possible cytogenetic 28. Sipe,J. C., Rubinstein,L. J., Herman,M. M., and Bignami,A. Ethylnitrosourea- implications of an immunohistochemical study. Acta Neuropathol., 62: induced astrocytomas. Morphologic observations on rat tumors maintained in 185-193,1984 tissue and organ culture systems. Lab. Invest., 37: 571-579,1974. 2. Braun, M., Grasso, A., and Wechsler, W. 14-3-2 protein in rat primary and 29. Sternberger, L. A. Immunocytochemistry. Englewood Cliffs, NJ: Prentice-Hall, transplanted gliomas and neurinomas and in clonal cell lines. Exp. Brain Res., 1974. 25: 93-97, 1976. 30. Tapia, F. J., Polak, J. M., Barbosa, A. J. A., Bloom, S. R., Marangos, P. J., 3. DeArmond, S. J., VandenBerg,S. R., and Herman,M. M. Neuraldifferentiation Dermody, C., and Pearse, A. G. E. Neuron-specific enolase is produced by in the OTT-6050 mouse teratoma: effects of intracerebralenvironment on the neuroendocrinetumours. Lancet, 7. 808-811,1981. neural differentiation of embryoid bodies. Virchows Arch. A Pathol. Anat. 31. VandenBerg, S. R., Chatel, M., Griffith, 0. M., DeArmond, S. J., Pappas, C., Histol., 393. 39-52, 1981. and Herman, M. M. Neural differentiation in the OTT-6050 mouse teratoma: 4. Dhillon, A. P., and Rode, J. Patterns of staining for neurone specific enolase production of a tumor fraction restricted to stem cells and neural cells after in benign and malignant melanocytic lesions of the skin. Diagn. Histopathol., centrifugal elutriation. Virchows Arch. A Pathol. Anat. Histol., 392: 281-294, 5:169-174,1982. 1981. 5. Dhillon, A. P., Rode, J., and Leathern, A. Neurone specific enolase: an aid to 32. VandenBerg,S. R., Hess,J. R., Herman,M. M., DeArmond,S. J., Halks-Miller, the diagnosis of melanoma and neuroblastoma. Histopathology, 6: 81-92, M., and Rubinstein, L. J. Neural differentiation in the OTT-6050 mouse tera 1982. toma: production of a tumor fraction showing melanogenesisin neuroepithelial 6. Eng, L. F., and Rubinstein, L. J. Contribution of immunohistochemistry to cells after centrifugal elutriation. Virchows Arch. A Pathol. Anat. Histol., 392: diagnostic problems of human cerebral tumors. J. Histochem.Cytochem., 26: 295-308,1981. 513-522, 1978. 33. Vinores, S. A. Increasedadhesionresponse of anaplasticglioma cells to nerve 7. Erdelyi, E., VandenBerg, S. R., RÃ¡ese,J., Barchas, J. D., Rubinstein, L. J., growth factor and the presence of specific receptors. J. Neurosci. Res., 70: and Herman, M. M. Neural differentiation in the OTT-6050 mouse teratoma: 381-395,1983. enzymatic and immunofluorescencecharacterizationof a tumor fraction show 34. Vinores, S. A., Bonnin, J. M., Rubinstein, L. J., and Marangos, P. J. Immuno ing melanogenesis in neuroepithelial cells. Virchows Arch. A Pathol. Anat. histochemical demonstration of neuron-specific enolase in neoplasms of the Histol., 393: 27-37, 1981. central nervous system and other tissues. Arch. Pathol. Lab. Med., in press, 8. Gu, J., Polak, J. M., Tapia, F. J., Marangos, P. J., and Pearse,A. G. E. Neuron- 1984.

2598 CANCER RESEARCH VOL. 44

35. Vinores, S. A., and Koestner, A. The effect of nerve growth factor on undiffer- nerve growth factor treatment. J. Neurochem., 37: 597-600, 1981 . entiated glioma cells. Cancer Lett., 10: 309-318,1980. 40 Wharton, J., Polak, J. M., Cole, G. A., Marangos, P. J., and Pearse, A. G. E. 36. Vinores, S. A., and Koestner, A. Effect of nerve growth factor producing cells Neuron-specific enolase as an immunocytochemical marker for the diffuse on anaplastic glioma and pheochromocytoma clones, involvement of other neuroendocrine system in human fetal lung. J. Histochem. Cytochem., 29: factors. J. Neurosa. Res., 6: 389-401,1981. 1359-1364,1981. 37. Vinores, S. A., and Marangos, P. J. A developmental study of neuron-specific 41 Wick, M. R., Scheithauer, B. W., and Kovacs, K. Neuron-specific enolase in enolase in rat adrenal medulla. J. Neurochem., 39.- 1748-1750,1982. neuroendocrine tumors of the thymus, bronchus, and skin. Am. J. Clin. Pathol., 38. Vinores, S. A.. Marangos, P. J., and Ko, L. Butyrate-induced increase in 79: 703-707, 1983. neuron-specific enolase and omithine decarboxylase in anaplastic glioma cells. 42 Zomzely-Neurath, C. E., and Walker, W. A. Nervous system-specific proteins: Dev. Brain Res., 5: 23-28,1982. 14-3-2 protein; neuron-specific enolase; and S-1 00 protein. In: R. A. Bradshaw 39. Vinores, S. A., Marangos, P. J., Parma. A. M., and Guroff, G. Increased levels and D. M. Schneider (eds.), Proteins of the Nervous System, Ed. 2, pp. 1-57. of neuron-specific enolase in PC12 pheochromocytoma cells as a result of New York: Raven Press, 1980.

-* * * V '** . ,'*>Â».C- Â» Â¿r Â» . â€¢Â»V *" f *

Fig. 1. Immunoperoxidase staining for NSE demon strated in neurons of dorsal root ganglion invaded by an ENU-induced malignant schwannoma. Counterstained with hematoxylin, x 300. Fig. 2. Immunoperoxidase staining for NSE in a tumor implant resulting from the i.e. inoculation (105 cells) of the F98 rat anaplastic glioma cell line. Counterstained with hematoxylin, x 640.

JUNE 1984 2599

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1984 American Association for Cancer Research. Immunoradiometric and Immunohistochemical Demonstration of Neuron-specific Enolase in Experimental Rat Gliomas

Stanley A. Vinores, Paul J. Marangos, Jose M. Bonnin, et al.

Cancer Res 1984;44:2595-2599.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/44/6/2595

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/44/6/2595. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.