Sequential development of ethylnitrosourea-induced neurinomas: morphology, biochemistry, and transplantability

Proceedings of the 2019 National Toxicology Program Satellite Symposium

The 2019 annual National Toxicology Program Satellite Symposium, entitled "Pathology Potpourri," was held in Raleigh, North Carolina, at the Society of Toxicologic Pathology's 38th annual meeting. The goal of this symposium was to present and discuss challenging diagnostic pathology and/or nomenclature issues. This article presents summaries of the speakers' talks along with select images that were used by the audience for voting and discussion. Various lesions and topics covered during the symposium included aging mouse lesions from various strains, as well as the following lesions from various rat strains: rete testis sperm granuloma/fibrosis, ovarian cystadenocarcinoma, retro-orbital schwannoma, periductal cholangiofibrosis of the liver and pancreas, pars distalis hypertrophy, chronic progressive nephropathy, and renal tubule regeneration. Other cases included polyovular follicles in young beagle dogs and a fungal blood smear contaminant. One series of cases challenged the audience to consider how immunohistochemistry may improve the diagnosis of some tumors. Interesting retinal lesions from a rhesus macaque emphasized the difficulty in determining the etiology of any particular retinal lesion due to the retina's similar response to vascular injury. Finally, a series of lesions from the International Harmonization of Nomenclature and Diagnostic Criteria Non-Rodent Fish Working Group were presented.

Neural stem cells and neuro-oncology: quo vadis?

Conventionally, gliomas are assumed to arise via transformation of an intraparenchymal glial cell that forms a mass that then expands centrifugally, eventually invading surrounding tissues. We propose an alternative model in which gliomas arise via initiation and promotion of cells within the brain's subependymal layer or subventricular zone, the source of a recently characterized pool of neural cells with the properties of self-renewal and multipotentiality (i.e., stem cells) that persists into adulthood. In this model, the particular histological subtype of glioma would represent the effects of temporal and spatial environmental influences rather than the particular cell of origin and the disease's centrifugal point would be the subependymal layer. The implications of such a model are discussed.

Three-dimensional imaging of nerve tissue by x-ray phase-contrast microtomography

We show that promising information about the three-dimensional (3D) structure of a peripheral nerve can be obtained by x-ray phase-contrast microtomography (p-microCT; Beckmann, F., U. Bonse, F. Busch, and O. Günnewig, 1997. J. Comp. Assist. Tomogr. 21:539-553). P-microCT measures electronic charge density, which for most substances is proportional to mass density in fairly good approximation. The true point-by-point variation of density is thus determined in 3D at presently 1 mg/cm3 standard error (SE). The intracranial part of the rat trigeminal nerve analyzed for the presence of early schwannoma "microtumors" displayed a detailed density structure on p-microCT density maps. The average density of brain and nerve tissue was measured to range from 0.990 to 0.994 g/cm3 and from 1.020 to 1.035 g/cm3, respectively. The brain-nerve interface was well delineated. Within the nerve tissue, a pattern of nerve fibers could be seen that followed the nerve axis and contrasted against the bulk by 7 to 10 mg/cm3 density modulation. Based on the fact that regions of tumor growth have an increased number density of cell nuclei, and hence of the higher z element phosphorus, it may become possible to detect very early neural "microtumors" through increases of average density on the order of 10 to 15 mg/cm3 by using this method.

Early mutation of the neu (erbB-2) gene during ethylnitrosourea-induced oncogenesis in the rat Schwann cell lineage

The development of malignant tumors of the peripheral nervous system (schwannomas) within a defined intracranial section of the rat trigeminal nerve ("trigeminal box") was used as a model to identify genetic alterations typically associated with the process of cell-lineage-specific oncogenesis induced by exposure to N-ethyl-N-nitrosourea on postnatal day 1. All 47 trigeminal schwannomas (and 12 extracranial neurinomas) investigated carried a T.A----A.T transversion mutation at nucleotide 2012 of the neu (erbB-2) gene sequence encoding the transmembrane domain of pg185neu. This mutation was absent in all 18 tumors in the brain and spinal cord (central nervous system) isolated from the same animals. Identical observations were made in cell lines derived from N-ethyl-N-nitrosourea-induced rat schwannomas vs. brain tumors. By asymmetric PCR and mutant-specific Mnl I restriction fragment length analyses, cells carrying the mutant neu allele became detectable and could be localized within the trigeminal box as early as 7 days after the carcinogen pulse. The proliferation rate of the mutant cells strongly exceeded that of the wild-type cells up to the time of maturation of the trigeminal nerve around postnatal day 30 and thereafter to a lesser extent until the appearance of schwannomas. A specific mutation of the neu gene thus represents a very early, probably the first, step in the malignant conversion of immature rat Schwann cells exposed to N-ethyl-N-nitrosourea in vivo and is diagnostic for a subset of proliferative cells at high risk of progressing toward the expression of fully malignant phenotypes. Loss of heterozygosity for the mutant neu allele is a candidate event for a critical second step in the process.

Cellular and molecular aspects of neurocarcinogenesis

Although the morphology of neural tumors induced in rats by N-ethyl-N-nitrosourea (NEU) and related alkylating agents has been extensively investigated, their histogenesis and the molecular basis of malignant transformation are still largely unknown. This review gives an account of the interaction of neurocarcinogenic agents with cellular DNA, the possible role of promutagenic O6-alkyldeoxyguanine and their deficient repair by the cerebral O6-alkylguanine-DNA alkyltransferase. A new experimental model is described in which neural tumors are induced in fetal brain transplants. Pregnant rats received a single iv dose of NEU (50 mg/kg) on the 14th day of gestation. One day later, suspensions were prepared from the fetal forebrain and stereotactically injected into the caudoputamen of adult rats. After additional exposure to NEU of the host animals 8 days and 9 weeks post transplantation, all rats developed brain tumors within the neural graft. Histopathologically, all neoplasms were classified as olidogdendrogliomas. Other neoplasms typically induced by NEU transplacentally (astrocytomas, mixed gliomas, ependymomas) were absent. The selective induction of oligodendrogliomas indicates that neoplastic transformation in the nervous system can occur in a differentiated glial cell or a precursor cell committed to oligodendrocytic differentiation, and that transformation of a pluripotential stem cell is not necessary. Transplacental exposure of the donor fetuses to NEU alone, i.e., without additional postgrafting exposure, did not produce brain tumors in any of the experimental animals indicating that in the microenvironment of fetal brain transplants the multistep development of gliomas requires additional mutational events. Malignant schwanomas perinatally induced by NEU carry a point mutation in the transmembrane domain of the neu gene. The mode of oncogene activation in NEU-induced CNS gliomas has not yet been elucidated. We have used cerebral grafting techniques to study the effects of known oncogenes on the developing nervous system, taking advantage of efficient gene transfer by replication-defective retroviral vectors and of the extraordinary capacity of fetal CNS to differentiate in and fully integrate with the host brain. Rats carrying transplants exposed in vitro to the polyoma medium T-antigen developed endothelial hemangiomas in the graft which often led to fatal cerebral hemorrhage within 13-50 days after transplantation. Introduction of the viral src gene caused astrocytic and mesenchymal tumors after latency periods of 2-6 months. Following infection of fetal donor cells with a vector encoding the v-myc oncogene, only a single embryonal CNS tumor was observed whereas exposure to v-H-ras produced a low incidence of gliomas.(ABSTRACT TRUNCATED AT 400 WORDS)

Experimental model of tumors associated with neurofibromatosis

Four groups of animals, each composed of 22 pregnant Wistar rats, were used in this study. Single intraperitoneal (IP) injections of ethylnitrosourea (ENU) at a dose of 15 mg/kg body weight (BW) were given to the animals of two groups on days 15 and 21 of pregnancy, respectively. The progeny of a third group received by subcutaneous injection (SC) the same dose of ENU, 15 mg/kg BW, on day 1 postnatally. The descendants of the fourth group served as untreated controls. The most striking findings were observed in the progeny of the mothers treated on day 15 of pregnancy, in which group 64 of 180 descendants developed peripheral nervous system (PNS) tumors, 30% of which had plexiform pattern. One hundred fifteen of the 180 descendants developed central nervous system (CNS) gliomas, mainly oligodendrogliomas, and five animals presented with Wilms' tumors. No tumors of these types were observed in the untreated controls. Although descendants of mothers treated on day 21 of pregnancy had the highest number of PNS tumors (130 of 172 animals), only 21% of these tumors were plexiform; CNS gliomas were observed in 78 animals and Wilm's tumors in one animal. The lowest percentage of PNS tumors with plexiform pattern (16%) was found in the group of 157 descendants treated postnatally on day 1, in which 88 animals developed PNS tumors, 76 developed CNS gliomas, and no animals developed Wilms' tumors. The higher percentage of plexiform PNS tumors found in the descendants treated prenatally on day 15 of pregnancy was statistically significant (P less than 0.05) when compared with the percentage found in the group treated postnatally. This significance was also valid for the plexiform tumors that developed selectively from branches of the trigeminal nerves (of the PNS tumors from this location, 48% showed a plexiform pattern), but only in the progeny exposed to ENU on day 15 of pregnancy. This same progeny also had the highest numbers of CNS and Wilms' tumors. Because in humans, plexiform neurofibromas are considered to be the neoplastic markers of neurofibromatosis, and CNS gliomas as well as Wilms' tumors are associated with this disease, it is suggested that exposure to ENU on day 15 of pregnancy, under the experimental conditions described here, may offer a model for investigating tumors associated with neurofibromatosis, as well as aspects of the spontaneous, noninherited forms of this disease.

Nerve growth factor modification of the ethylnitrosourea model for multiple schwannomas

The administration of ethylnitrosourea (ENU) to pregnant rats late in gestation or to neonatal rats results in the induction of Schwann cell tumors in a high percentage of perinatally exposed animals. Exogenous administration of nerve growth factor (NGF) significantly reduces the number of Schwann cell tumors and other neurogenic tumors developing in ENU-treated rats. Administration of antibodies directed against NGF prior to neonatal ENU exposure results in a substantial increase in the incidence of Schwann cell tumors, particularly in the trigeminal nerves of both rats and mice. Transplacental ENU treatment causes early neoplastic proliferation (ENP) at 90 days of age in the Schwann cell population of trigeminal nerves in nearly all exposed rats. A variety of NGF treatment protocols (single or multiple inoculations or microinfusion prior to or following ENU exposure) resulted in a significant reduction in ENU-induced ENP in trigeminal nerves. These results indicate that NGF may convey protection either directly or indirectly, by an unknown mechanism, to Schwann cells and other supportive neural cells by reducing their sensitivity to ENU-induced neoplastic transformation.

The brain-tumour issue in long-term toxicity studies in rats

Problems arise in the measurement of the neurocarcinogenic potential of chemical substances in chronic toxicity studies because of the spontaneous occurrence of neoplasms in the brain and other organs of rats from the age of about 2 yr. Statistical analysis may be equivocal and must be accompanied by a thorough biological analysis, to determine the presence or absence of the following characteristic effects of neurocarcinogenic agents: a reliable and consistent increase in brain-tumour incidence beyond the expected control level, a decrease in the age at which tumours appear and/or in survival, a dose-effect relationship, a greater effect on embryonal and foetal brain cells than on those of adults, a shift to more anaplastic types of tumour and the finding of preneoplastic lesions. These criteria have been met in chronic tests on the neurocarcinogens ethyl- and methylnitrosourea and some have been met in results obtained with the weak carcinogen methyl methanesulphonate. However, none of these criteria were met in the case of a test compound subjected to two 2-yr studies (one involving transplacental exposure), although brain tumours occurred in the controls and in all the experimental groups. The test substance is not considered to be a neurocarcinogenic agent.

Chronic toxicity/carcinogenicity study of FD & C Blue No. 2 in rats

FD & C Blue No. 2 was fed to rats in the diet in a long-term toxicity/carcinogenicity study. The study included an in utero phase in which the compound was administered to groups of 60 male and 60 female Charles River CD albino rats at levels of 0.5, 1.0 and 2.0%. Two concurrent control groups, each containing 60 rats of each sex, received the basal diet. After random selection of the F1 animals, the long-term phase was initiated at the same dietary levels, with 70 rats of each sex in each dose group and in each of two control groups. Maximum exposure was 30 months. No consistent compound-related biologically adverse effects were noted. There were random statistically significant differences from the controls with respect to body weight, food consumption and clinical chemistry tests. Food consumption by the test groups showed a dose-related increase. This was probably due to the non-nutritive character of the colouring. A statistically significant increase in gliomas in the high-dose male rats was not found to be biologically significant, since none of the criteria for determining the neurocarcinogenic potential of chemical substances was met. The overall brain-tumour incidence in this study was within the range typical for 2-yr-old CD rats. Under the conditions of this study, FD & C Blue No. 2 did not produce evidence of any toxicity, including carcinogenicity.

Prognostic role of cell morphology of animal tumors

The purpose of this presentation is to determine the prognostic role of cellular morphology in animal neoplasia. With some exceptions, cellular morphology is the single most accurate predictor of the prospective behavior of neoplasms. There is generally a positive correlation between the degree of malignancy and prognosis. The exceptions are a) morphologically malignant-appearing tumors following a benign course (e.g., canine histiocytoma, canine seminoma, equine sarcoid) and b) morphologically differentiated tumors exhibiting an unpredictable prognosis (e.g., canine pericytoma, acanthomatous epulis, myxoma, follicular thyroid cell carcinoma, etc.). Anaplasia, an important characteristic of most malignant neoplasms, may be less stable than generally assumed. Sodium butyrate may reverse it intermittently and anaplastic gliomas may loose all morphologic and cytokinetic characteristics of anaplasia following sodium butyrate exposure. Host factors, such as nerve growth factor, have similar and more lasting effects upon anaplastic cells derived from the neural crest. Such factors may act as reverse transformation agents and may represent prospective therapeutic agents for anaplastic tumors.

The effect of nerve growth factor and antibodies to nerve growth factor on ethylnitrosourea-induced neoplastic proliferation in rat trigeminal nerves

Sprague-Dawley (CD) rats were injected intravenously with ethylnitrosourea at a dose of 20 mg/kg on day 20 of gestation. This exposure resulted in early neoplastic proliferation or development of a neurinoma of the trigeminal nerve in 58% of the offspring at 90 days of age. Implantation of osmotic microinfusion pumps containing 2.5S nerve growth factor prior to ethylnitrosourea administration significantly reduced the incidence of early neoplastic proliferation. Postnatal implantation of microinfusion pumps containing 2.5S nerve growth factor also resulted in a significant but less pronounced reduction of early neoplastic proliferation. Immunoglobulin G directed against nerve growth factor (anti-nerve growth factor) did not influence the incidence of early neoplastic proliferation when administered via microinfusion pumps implanted on day 15 postnatally. These findings suggest that nerve growth factor has a protective effect on the developing nervous system against ethylnitrosourea-induced carcinogenesis.

Plexiform structures in malignant schwannomas after prenatal exposure to ethylnitrosourea

75 Schwann cell tumors were induced in the offspring of Wistar rats which had been treated intraperitoneally with 15 mg/kg b.w. ethylnitrosourea on the 15th day of pregnancy. Plexiform structures characterized by poorly-developed, distorted and bizarre bundles of nerves were seen in 2 of 33 benign tumors, and in 17 of 42 malignant schwannomas. Since when seen in man similar findings are distinctive of neurofibromatosis, it is suggested that this experiment may offer a model for studying non-inherited forms of neurofibromatosis and, more particularly, the malignant changes often associated with this disease.

Nerve growth factor and neural oncology

The precise role of the nerve growth factor protein (NGF) during the growth and development of the human nervous system is not determined. Although it appears to influence a number of neural functions, its mechanism of action is poorly understood. A number of researchers have proposed that NGF may be involved in several pathological conditions including cancer. It has been shown that NGF is secreted by certain sarcoma (23), neuroblastoma (113), and glioma (7,102,136) cell lines and can bind to neuroblastoma and metastatic melanoma cell lines (42). Neuroblastoma (136,181) and pheochromocytoma (165) cells in vitro can be induced by NGF to differentiate toward a morphologically "more benign" state and appropriate NGF treatment of rats can reduce the number of chemically induced gliomas and neurinomas (174,178). NGF can also reduce the growth of intracerebrally inoculated anaplastic glioma cells (172). Anti-NGF treatment of rats (178) and mice (179) can alter the tumor distribution observed following ethylnitrosourea or benzo(a)pyrene treatment (10). In humans, it has been reported that serum levels of NGF are usually elevated in persons "at risk" for neurofibromatosis (156). The precise nature of the NGF role is not known in these instances. Further understanding of the action of NGF could be of clinical importance.

LDH isozyme analyses of ethylnitrosourea-induced central nervous system tumors in rats

To provide the significance of LDH isozymes in rat CNS tumors, the changes in lactic dehydrogenase isozyme and calculated ratios of H- to M- subunit were studied by means of polyacrylamide gel enzymoelectrophoresis in tumor extracts from CNS tumors (7 astrocytomas, 4 oligodendrogliomas, 7 mixed gliomas, 6 anaplastic gliomas, 3 glioependymomas, 1 astroblastoma, 11 neurinomas, 8 anaplastic neurinomas and 1 meningioma in Wistar rats which were induced by ethylnitrosourea). The isozyme patterns were compared to those obtained from normal rat CNS tissues. Among the glioma group, oligodendroglioma showed the highest H/M ratio followed by mixed glioma, glioependymoma, astrocytoma, astroblastoma and anaplastic glioma in order of decreasing of H/M ratios. On the other hand, the H/M ratio of neurinoma was significantly higher than that of anaplastic neurinoma. These observation suggested that determination of LDH isozyme patterns could supplement the histological evaluation of brain tumors.

Early effects of ethylnitrosourea on the LDH isozyme of rat fetal central nervous system

The sequential changes in isozyme distribution of lactate dehydrogenase and calculated ratios of H- to M-subunit were studied by means of acrylamide gel enzymoelectrophoresis in Wister rat fetal brains from pregnant rats which were administered ethylnitrosourea with a dose of 75 mg/kg on day 14 of gestation, to search for early effects on rat fetal brains of ENU which was transplacental carcinogen and teratogen, from a view point of metabolic aspects. The isozyme patterns were compared to those obtained from normal rat fetal brain. The H/M ratios of ENU-treated group were significantly lower than those of control group. These data suggested that retarded differentiation of rat fetal CNS was induced by administration of ENU as far as LDH was concerned. The present data were discussed in relation to teratogenesis and carcinogenesis.

The histochemical behaviour of zinc-activated tartrate-resistant phosphatase (ZnTP) in early stages of experimental tumors in the rat trigeminal nerve

The activity of the ZnTP (Felicetti und Rath 1975; Rath and Felicetti 1975) was investigated histo- and biochemically in neurinomas of trigeminal nerves of rats. The tumors were induced by a single transplacental pulse of 30 mg ENU/kg on the 17th day of gestation. The rats were killed at the age of 40 to 160 days. The activity of the ZnTP is demonstrated histochemically by a coupling azo dye technique with 1-naphthylphosphate after Rath and Felicetti (1975) as well as a lead nitrate method with 4-nitrophenyl phosphate at the age of 40 to 140 days. Biochemical investigations of the trigeminal nerves are performed at the age of 140 and 160 days only. The separation of ZnTP is carried out by isoelectrofocusing in polyacrylamide gels after Felicetti and Rath (1975). Together 14 neurinomas are found in the trigeminal nerves of 53 rats. All neurinomas show a high activity of the ZnTP. The ZnTP activity is absent in normal Schwann cells. Three trigeminal nerves of rats subjects to the transplacental pulse of ENU harbour small foci of ZnTP activity without detectable neoplastic cell proliferation. Similar spots of enhanced ZnTP activity have never been found in control rats. This rises suspicion that the neoplastic proliferation of cells is preceded by the activation of ZnTP.

Acid hydrolase and cytochrome oxidase activities in nitrosourea induced tumors of the nervous system

Nitrosourea induced tumors of the nervous system in rats have proven useful for biochemical studies combined with morphological approaches. The pattern of enzyme activities for acid hydrolases and cytochrome oxidase resemble those previously observed in spontaneous nervous system tumors of man. The activities of 4 acid hydrolases were generally high in the gliomas. This could not be attributed solely to zones of regression or necrosis but was a general characteristic of the neoplasms. The activities were predominantly particulate and most likely lysosomal in localization. In schwannomas a similar increase in hydrolases was found in comparison with normal neural tissues but aryl-sulfatase was not increased. Cytochrome oxidase activities were markedly reduced in all tumors studied. The proportionate reduction with respect to normal brain was comparable to that noted in man. No differences were found with respect to fairly well differentiated gliomas.

The origin of experimental brain tumours: a sequential study

A sequential study of rat brains treated transplacetally with the neurotropic carcinogen ethylnitrosourea reveals small foci of cell proliferations from the age of 8 weeks. These lesions consist mainly of undifferentiated cells of the subependymal plate type. They occur in those areas in which gliomas develop and represent the earliest, histologically detectable, changes in the development of brain tumours.

Light and electron microscopic investigations on a transplantable tumour of the rat, induced by transplacental administration of ENU (ethylnitrosourea)

With a single dose of 80 mg/kg ENU, a tumour originating from the trigeminal nerve was induced transplacentally in the offspring of a Sprague-Dawley rat. The neoplasm diagnosed as a neurinoma, was transplanted over 20 passages to 5 six-month-old Sprague-Dawley rats in each case. During this time, the biological behaviour of the tumour as well as its histological appearance were followed. Ultrastructural observations were made of the 20th passage. The tumour caused multiple metastases in the lung and in the lymph nodes of the neck in 30 animals. Recurrences were formed within a week, metastases of the lung appeared from then on in all cases. During the time of transplantation, the neurinoma dedifferentiated histologically so markedly that it was no longer possible to classify the neoplasm as a neurogenous tomour after 20 passages. Ultrastructurally, poorly differentiated Schwann cells could nevertheless be recognized.

Studies on experimental malignant nerve sheath tumors maintained in tissue and organ culture systems. I. Light microscopy observations

Explants from seven nerve sheath tumors (four cranial and three spinal) induced in rats by transplacental ethylnitrosourea were grown on collagen-coated coverslips and in organ culture systems, using sponge foam matrices and Millipore filter platforms. Their sequential morphological features in vitro were compared to those of a human acoustic Schwannoma maintained in similar culture systems. The original experimental tumors were either undifferentiated or poorly malignant Schwannomas. In cultures on collagen-coated coverslips the explants demonstrated cellular features that were considerably more anaplastic than those of the human acoustic Schwannoma. On the other hand, in organ culture systems, in which viable cultures were maintained up to 82 days, many of the experimental tumor explants exhibited progressive differentiaiton, with nuclear palisading, increasing whorl formation and abundant reticulin fibers, and their pattern of histological organization came therefore closely to resemble that of the cultured acoustic Schwannoma. Unlike the latter, however, the cultured experimental tumors infiltrated the sponge foam matrices, a feature that mimicked the invasive character of the original tumors in vivo. Included normal ganglion cells remained demonstrable in the cultures up to 69 in vitro.