Induction of Congenital Hydrocephalus in Hamsters with Attenuated and Natural Strains of Mumps Virus

Size: px

Start display at page:

Download "Induction of Congenital Hydrocephalus in Hamsters with Attenuated and Natural Strains of Mumps Virus"

Brenda Stevenson
6 years ago
Views:

THE JOURNAL OF INFECTIOUS DISEASES VOL. 132, NO.4 OCTOBER 1975 1975 by the University of Chicago. Ail rights reserved.

1 THE JOURNAL OF INFECTIOUS DISEASES VOL. 132, NO.4 OCTOBER by the University of Chicago. Ail rights reserved. Induction of Congenital Hydrocephalus in Hamsters with Attenuated and Natural Strains of Mumps Virus Lawrence Kilham and George Margolis From the Departments ofmicrobiology and Pathology, Dartmouth Medical School, Hanover, New Hampshire The abilities of a low-passage strain and of a live, attenuated vaccine strain of mumps virus to induce congenital hydrocephalus in hamsters were tested by intraamniotic inoculation on the 10th day of pregnancy. Examination of term fetuses and neonates, with cytoplasmic inclusions, cytopathic effects, and specific immunofluorescence used as indicators, demonstrated an oronasal portal of entry for both strains. The vaccine strain appeared to be more pathogenic; it spread primarily into the respiratory tract and hence to the central nervous system. Inclusions were observed as long as 21 days after inoculation. Hydrocephalus and ependymal involvement, potentially capable of producing aqueductal stenosis, were observed in 19 of 81 animals studied days after inoculation. This report describes induction of hydrocephalus in hamsters after inoculation of mumps virus into the amniotic sac. As far as we are aware, this is the first report of induction of congenital hydrocephalus in any animal by this route. Johnson et al. [1], in work later confirmed by Ennis et al. [2], have shown that mumps virus can induce a destructive ependymitis followed by an obstructive hydrocephalus in neonatal hamsters. The strains of virus used by these investigators are.of interest. Johnson et al. used a neuro-adapted and a nonneuro-adapted (natural) strain, whereas Ennis et al., in addition to a natural one, used the Jeryl Lynn strain as it is employed in a live, attenuated vaccine (Mumpsvax; Merck, Sharp and Dohme, West Point, Pa.). Our experiments centered on the use of this commercial strain, and we used a natural strain for comparison. Materials and Methods Our methodology has been described in a recent publication [3] and thus is given below in an abbreviated form. Received for publication February 4, 1975, and in revised form June 14, This work was supported by United States Public Health Service grants no. HD07775, HD03298,and HD08184from the National Institute of Child Health and Development and by Research Career Program award no. I-K6-CA from the National Cancer Institute. Please address requests for reprints to Dr. Lawrence Kilharn, Department of Microbiology, Dartmouth Medical School, Hanover, New Hampshire Virus. A natural O'Take strain [4] and an attenuated strain, in the form of a live commercial vaccine (Mumpsvax), were used. Titrations. Titrations were performed by inoculation ofserial 1Q-fold dilutions of materials to be tested into two tubes each of a continuous line of rhesus monkey kidney cells (MKS); 0.1 ml of inoculum was used per tube. Tubes were fed again in four to five days and were tested in eight to nine days by hemadsorption with guinea pig erythrocytes. The highest dilutions that showed hemadsorption were taken as end points of titrations. In this system the O'Take strain of virus had a titer of 1O- 4TCID so/o.i ml, and the titer of the vaccine strain was 10-3 TCIDso/O.1 ml. Inoculations and harvestings, All procedures were done underanesthesia with ether. Undiluted virus was inoculated into the amniotic sacs with 30-gauge needles after exposure by laparotomies. Volumes of inocula were approximately 0.01 ml. Serology. CF tests were performed for us by Microbiological Associates, Bethesda, Md. Pathological studies. The general plan followed previously reported procedures for study of antenatal and postnatal mumps infections [3, 5]. Fetuses were inoculated on days 9-12 of gestation and were harvested in two sets. An early series of studies was made in term fetuses, neonates, and sucklings four days old or less; these animals were sacrificed five to nine days following inoculation. A later series included sucklings sacrificed five to 22 days after birth and days after inoculation. 462

Congenital Hydrocephalus and Mumps Virus 463 Results Infectivity titrations of tissues from inoculated hamsters showed that pools offetuses, placentas, and uteri were all infected, as were amniotic

2 Congenital Hydrocephalus and Mumps Virus 463 Results Infectivity titrations of tissues from inoculated hamsters showed that pools offetuses, placentas, and uteri were all infected, as were amniotic fluids on the two occasions when they were tested. Actual titers for fetuses ranged from to- 3 TClDso for the vaccine strain to 10-4 TClDso for the natural strain (O'Take). Brain tissues of neonatal hamsters born within hr after intraamniotic sac inoculation of their mothers were also titrated. Of eight individual brains of neonates whose mothers had received the vaccine strain, five were negative and three were positive for mumps virus (titers, lo-l lo- s TCIDsJ In contrast, seven of eight animals from mothers that had received the O'Take (natural) strain were negative, and only one was positive (titer, 1O- 2 TCID so ). These findingscorrelated with the results of histologic examinations, which revealed that only a portion of neonatal and suckling hamsters had evidence of infection and that the vaccine strain was the more virulent. Major pathological findings were (1) an oronasal route of entry, manifested by cytoplasmic inclusions in the mucosa; (2)a primary spread via respiratory mucosa to the lungs; (3) significant systemic disease only in the central nervous system; and (4) obstructive hydrocephalus as a residual effect. Lesions at the portals of entry and manifestations of early pulmonary spread regressed without sequelae, and only a small proportion of animals exhibited the full spectrum of disease, which eventuated in aqueductal stenosis and obstructive hydrocephalus (figures 1 and 2). Of the two strains used, the vaccine strain was more pathogenic. Thus, 16of80 fetuses studied in the early period after inoculation and 36 of 81 sucklings examined at the later stage were positive. Involvement of ependyma and choroid plexus was observed in four fetuses in the early stages of infection and in 13sucklings at the later stage. Obstructive hydrocephalus was demonstrated in six animals of the latter group. In the group that received O'Take virus, manifestations of infection were observed in seven of 71 fetuses or neonates in the early period after inoculation, and in three of 16 sucklings in the later period. All three positive animals in the latter group had obstructive hydrocephalus. Figure 1. Obstructive hydrocephalus in a )3-day-old hamster infected in utero 19 days earlier by intraamniotic inoculation of the attenuated (vaccine) strain of mumps virus. Both the third ventricle (lop) and aqueduct of Sylvius (bottom) are severely stenosed. A number of sucklings died or were moribund when sacrificed. The chiefcause ofdeath as identified in sections appeared to be pneumonia, due primarily to mumps virus, with secondary bacterial infection. It appeared that no animals died of encephalitis. Findings made with immunofluorescence techniques generally paralleled histopathologic observations. In the early period after inoculation, positive findings in the respiratory tractpredominated, whereas in the later period viral antigen was demonstrable only in the nervous system.

3 464 Kilham and Margolis Figure 2. Severe stenosis of the third ventricle of a IS-day-old hamster infected in utero 21 days earlier by intraamniotic inoculation ofa natural (O'Take) strain of mumps virus. Three linearly arranged ependymal rosettes, remnants of the original lumen, are linked by an inconspicuous glial scar ( X 150, modified Lendrum strain). Virus was identified by two methods. The first, given above, was the use of specific immunofluorescence, which showed a distribution identical to that of the intracytoplasmic inclusions found in histologic sections. When the second (serological) method was used, hamsters inoculated with either the natural or the vaccine strain showed a rise in titer of CF antibodies to mumps virus. Discussion Our previous work on mumps carried us in two directions, both germane to this report. Inthe first study [3], we inoculated pregnanthamstersparenterally with a hamster-adapted strain of mumps virus [6]. Virus given in this manner induced viremia, followed by infections of uteri, placentas, and fetuses. The fetal infections, however, were only sporadic. Therefore, we turned to direct inoculation of virus into amniotic sacs in an effort to learn more about the processes involved. Using immunofluorescence as well as presence of intracytoplasmic inclusions, we were then able to demonstrate that mumps virus, presumably from infected amniotic fluid, infected fetuses via the trachea and lungs, with further invasion to skeletal muscle and neural tissue. This neuroadapted virus, however, was too virulent to permit survival of fetuses after birth. It was for this reason that we resorted to the seemingly less virulent O'Take and vaccine strains, which gave better rates of survival. We also studied mumps virus infections in neonatal hamsters [5]. Our results not only confirmed the work of Johnson et al. [I] on the ability of mumps virus to induce hydrocephalus, but also clarified the nature of the accompanying encephalitis by demonstration of intracytoplasmic inclusions in neural tissue. A continuation of these studies by electron microscopy [4] showed differences between the neuro-adapted and a natural (O'Take) strain of mumps virus. A major aspect of the present report is the induction of congenital hydrocephalus in an animal model, a phenomenon that Kent et al. [8] have already reported for the rhesus monkey. We emphasize, however, that our experiments differed in the route of inoculation. Whereas Kent et al.

Congenital Hydrocephalus and Mumps Virus 465 inoculated influenza virus directly into the brains of fetal monkeys through the exposed uterine wall, we did not inoculate fetuses, but only their

4 Congenital Hydrocephalus and Mumps Virus 465 inoculated influenza virus directly into the brains of fetal monkeys through the exposed uterine wall, we did not inoculate fetuses, but only their associated amniotic sacs. Thus mumps virus, after enteringthe fetus via the oronasal route, had to reach and penetrate the blood-brain barrier in a natural manner. This action is more meaningful than bypassing of the barrier with a needle. The experiments of Kent et al. were actually little different, in essence, from those of Johnson et al. [1], who induced hydrocephalus by direct inoculation of influenza and other viruses into the brains of neonatal hamsters. The only other work known to us that involved the oronasal portal of entry is that of Phillips et al. [9]; they succeeded in inducing hydrocephalus in neonatal mice with reovirus type 1. Is there any evidence that mumps virus can cause congenital hydrocephalus in man? This problem has been discussed [6, 10]. The five references cited in these papers on hydrocephalus as a sequela of mumps suggest a provocative but unproved linkage. Moreover, the etiologic lead is reinforced by the recent observations of Herndon et al. [11], who found ependymal cells in the centrifuged sediments of five consecutive cases of mumps meningoencephalitis. These cells, when examined by electron microscopy, were found to contain cytoplasmic aggregates resembling nucleocapsids of mumps virus. A postulate that some viral infections of the nervous system may be uncommon manifestations of common viruses [11] fits in with the low incidence of congenital hydrocephalus in man. It would seem that for a common virus to induce hydrocephalus the following criteria would have to be satisfied: (1) a low frequency of transplacental passage; (2) fetal systemic disease of benign character, allowing survival without major extraneural residence of infection; (3) viral invasion of the developing nervous system; (4) neural disease of mild character, not productive of parenchymal destructive effects; and (5) productive infection of ependyma, resulting in the sequence of cytolysis, gliovascular repair, aqueductal stenosis, and obstructive hydrocephalus. Virtually all of these criteria have been met in our experimental studies of transplacental mumps infections. First, when mumps virus was inocu- lated into pregnant hamsters by a parenteral route, the virus infected only a small portion of fetuses [3]. Second, as documentedby the present experiments in which a viral strain of low pathogenicity was used, benign, self-limited systemic fetal disease was induced. Third, passage of virus though the blood-brain barrier occurred in relatively few infected fetuses. Fourth, infection progressed in the central nervous system in the face of regression of extraneural infection without systemic sequelae. Fifth, the neural disease was essentially limited to the ventricular lining tissues, thereby producing an obstructive hydrocephalus unassociated with parenchymal lesions. A question one would like to ask in conclusion is why an attenuated vaccine strain should have appeared more virulent than the natural strain of mumps virus that we used. A possibility is that the vaccine strain was attenuated in respect to adult hosts only. In the case of fetal infections, its virulence may have remained unchanged or even have been enhanced, since the strain was attenuated in embryonated hens' eggs as well as in chick embryo cultures [12]. References I. Johnson, R. T., Johnson, K. P., Edmonds, C. J. Virusinduced hydrocephalus: development of aqueductal stenosis in hamsters after mumps infection. Science 157: , Ennis, F. A., Hopps, H. E., Douglas, R. D., Meyer, H. M., Jr. Hydrocephalus in hamster: induction by natural and attenuated mumps viruses. J. Infect. Dis. 119:75 79, Kilham, L., Margolis, G. Intrauterine infections induced by mumps virus in hamsters. Lab. Invest. 31:34-41, Wolinsky,1. S., Baringer, J. R., Margolis, G., Kilham, L. Ultrastructure of mumps virus replication in newborn hamster central nervous system. Lab. Invest. 31: , Margolis. G., Kilham, L., Baringer, J. R. A new look at mumps encephalitis: inclusion bodies and cytopathic effects. J. NeuropathoI. Exp. NeuroI. 33: 13-28, Kilham, L., Overman, J. R. Natural pathogenicity of mumps virus for suckling hamsters on intracerebral inoculation. J. ImmunoI. 70: , Kent, S. G., London, W. T., Sever, J. L. Influenza virus-induced congenital hydrocephalus in rhesus monkeys: neuropathologic findings. J. NeuropathoI. Exp. NeuroI. 33:564, Phillips, P. A., Alpers, M. P., Stanley, N. F. Hy~

5 466 Kilham and Margolis drocephalus in mice inoculated neonatally by the oronasal route with reovirus type I. Science 168: , Margolis, G., Kilham, L. Problems of human concern arising from animal models of intrauterine and neonatal infections due to viruses: a review. II. Pathologic studies. Prog. Med. Viro!., 1975 (in press). 10. Herndon, R. M., Johnson, R. T., Davis, L. E., Descalzi, L. R. Ependymitis in mumps virus meningitis. Arch. NeuroI. 30: , Johnson, R. T., Mims, C. A. Pathogenesis of viral infections of the central nervous system. N. EngI. J. Med. 278:23-30, 84-92, Buynak, E. G., Hilleman, M. R. Live attenuated mumps virus vaccine. I. Vaccine development. Proc. Soc. Exp. BioI. Med. 123: , 1966.

Measles, Mumps and Rubella. Ch 10, 11 & 12

Measles, Mumps and Rubella. Ch 10, 11 & 12 Measles, Mumps and Rubella Ch 10, 11 & 12 Measles Highly contagious viral illness First described in 7th century Near universal infection of childhood in prevaccination era Remains the leading cause of