- Email: [email protected]
The effect of influenza C virus on the Purkinje cells of chick embryo cerebellum
~Pergamon
Int. J. DevlNeuroscience,Vol. 12, No. 5, pp. 461-470, 1994
0736-5748(94)E0003-K
ElsevierScienceLtd Copyright© 1994ISDN Printed in GreatBritain.All rightsreserved 0736-5748/94$7.00+0.00
THE EFFECT OF INFLUENZA C VIRUS ON THE PURKINJE CELLS OF CHICK EMBRYO CEREBELLUM M. SHARON PARKER,* RICHARD J. O'CALLAGHAN,* DIANE E. SMITH]" and H. ADELE SPENCE* *Department of Microbiology,Immunology and Parasitology, and tDepartment of Anatomy, Louisiana State University Medical Center, 1901 Perdido Street, New Orleans, LA 70112, U.S.A. (Received 1 October 1993; accepted 15 December 1993)
Abstraet--Intra-amniotic inoculation of influenza C virus resulted in observable and quantitatively measurable changes in the Purkinje cellsof chick embryo cerebellum. Purkinje cells were visualizedby the Golgi--Coxprocedure and prepared for statistical and computer evaluation from camera lucida drawings. Four computer-generated measurements (the area of the dendritic arbor, the perimeter of the dendritic tree, and the height and width of the cell's arborization) and two manually counted measurements (total number of branches and the number of first order branches) were made. Analysis of Purkinje ceils from influenza C virus-infected embryos showed disturbances in dendritic arborization patterns and misalignment in the arrangement of .the cells in the Purkinje cell layer compared to control cells. Statistical evaluation of Purkinje cell arborization showed significant decreases in all measured parameters for the influenza C virus-infected members when compared with the members of the uninfected control group. Key words: Purkinje cell, chick, influenza C virus.
Influenza C virus is a m e m b e r of the Orthomyxoviridae family, which also includes the influenza A and B viruses. Influenza viruses have a major epidemiological impact on health and productivity worldwide, due to their association with respiratory infections. Influenza C virus infects most humans by young adulthood and usually causes mild respiratory infection. 6,15,17,19 T h e virus is also an important cause of pneumonitis in infants. 5 With rare exceptions, none of the influenza viruses have b e e n implicated in CNS disease or in neurological sequelae in humans. 14,21,22 In previous work, we described down feather abnormalities which occurred in chick embryos infected with the prototype strain of influenza C virus (j j).27 We have also observed that inoculation of e m b r y o n a t e d eggs with a clinical isolate, designated VS, produced some hatchlings that were unable to stand or walk. Affected chicks that could walk had an ataxic gait. Although leg and gait abnormalities were confined to VS-inoculated chicks, histological examination of h a e m o t o x y l i n eosin-stained sections of both JJ and VS inoculated 19 day embryos suggested that the Purkinje cells of the chicks had been adversely affected. The present work describes visually discernible and statistically measurable changes of the Purkinje cells of chick embryos infected with either of two strains of influenza C virus.
EXPERIMENTAL PROCEDURES Propagation o f virus strains
Virus strains were p r o p a g a t e d in chicken eggs as described previously.18,19, 29 Briefly, egg-derived virus was diluted in calcium- and magnesium-free H a n k s balanced salt solution with 100 units of penicillin and 100 Ixg of streptomycin per milliliter (Grand Island Biological Company, G r a n d Island, NY). Eight day e m b r y o n a t e d chicken eggs (SPAFAS Inc., Norwich, CT) were inoculated with 0.2 ml of diluted virus and placed into a 33°C humidified still-air incubator. Egg fluids were harvested 3 days post-inoculation and tested for hemagglutinating activity with 0.5% chicken red blood cells (Whittaker Bioproducts, Walkersville, MD). Influenza C virus strain JJ was grown in the chorioallantoic cavity of chicken eggs and strain VS was grown in the chick e m b r y o amniotic cavity. Abbreviations: DDP, dichlorodiammineplatinum; DFMO, a-difluoromethyl ornithine; MAM, methylazoxymethanol.
461
462
M.S. Parker et al.
Egg inoculation for viral pathogenesis stud), Nineteen embryonated chicken eggs (9 days old) were inoculated amniotically with either JJ strain or VS strain of influenza C virus diluted in phosphate-buffered saline (pH 7.2) without Pen-Strep. Six eggs were mock infected by inoculation with phosphate-buffered saline alone, and as an additional control, two eggs were not inoculated. All eggs were incubated at 33°C for 3 days in a humidified incubator, after which time the eggs were returned to the egg incubator (39°(?) for an additional 8 days.
Golgi-Cox procedure (Ha modification) Eleven days post-inoculation, at embryonic day 20, cerebellar tissue was harvested from the embryos and processed according to the Ha modification of the Golgi-Cox procedure. 1°,26 The specimens were removed from the impregnating solution after 2 months in the dark and dehydrated using graded alcohols (70%, 80%, 95% and 100% ethanol). Dehydration was completed with ether-ethanol (1:1 ). The tissue was infiltrated with increasing concentrations of celloidin (2%, 4%, 6%, 12% ) and embedded in 12% celloidin. One hundred micrometer sagittal sections of cerebellum were cut on a sliding microtome. The sections were developed in 5% ammonium hydroxide (in 95% ethanol), counterstained with 0.25% aqueous Cresyl Fast Violet, differentiated in 95% ethanol, dehydrated in butyl alcohol and cedarwood oil, and mounted on slides with histological mounting medium (Permount; Fisher Scientific, Fairlawn, N J).
Camera lucida drawings Drawings were made of all Purkinje cells meeting selection criteria which were within folia III, IV and IX from each of three consecutive mid-sagittal cerebellar sections. Selected cells were completely impregnated, were wholly present within the selected section and folium, and were not obscured by other cells or by debris. The three folia, as well as the mid-sagittal location of the selected folia, were chosen to correspond to cerebellar areas associated with hind limbs. Camera lucida drawings were made using a Zeiss microscope equipped with a drawing tube. The drawing of the arbor of each selected cell was outlined by forming the smallest polygon around the outermost extension of the dendrites, as in the method described by Schweitzer et al..25
Quantitative analysis of Purkinje cells" Computer-aided image analysis was done on each selected Purkinje cell using a Zeiss Videoplan Computer equipped with a stylus and magnetic tablet. The stylus was used to trace the polygon encircling the arbor and the Videoplan program (Carl Zeiss, Inc., New York, NY) was used to calculate the area of the dendritic tree enclosed in the polygon (area), the distance around the perimeter of the polygon (perimeter), the width of the dendritic arbor (width) and the height of the cell's arborization (height). In addition to the computer-generated values, the number of first order branches and the total number of branches of each evaluated Purkinje cell was counted manually. The ordering of the counted branches was done according to the method of Strahler, 28 in which each outermost branch is a first order branch. Second order and subsequent ordered segments are formed when two lower ordered branches join; the ordering number of segments from two unequally numbered segments retains the higher order number. The highest ordered segment for each Purkinje cell is the stem dendrite, which exits the Purkinje cell soma.
Statistical analysis Mean values for the six measured parameters were calculated for each evaluated embryo of all Purkinje cells meeting selection criteria in that embryo. Of the analysed embryos, 95 Purkinje cells from the uninfected group of five embryos and 196 Purkinje cells from the infected group of nine embryos met the selection criteria and were analysed by statistical analysis using Student's t-test for unpaired data. 3
Influenza C virus affects chick embryo Purkinje cells
463
Fig. 1. Photograph of Purkinje cell from the uninfected group of chick embryos. The Purkinje cell has a balanced arbor with numerous segments, many of which demonstrate dichotomous branching. The arbor reaches toward the pial surface of the cerebellum (P) and to the external granular layer (EGL). The cell soma is located within the Purkinje cell layer (PCL). Scale bar=20 ixm.
Selection of test groups for analysis The uninfected group consisted of Purkinje cells from two uninoculated embryos and three mock infected embryos. To show that uninoculated and mock infected specimens could be treated as a single group, the P value was calculated for these two subgroups using Student's t-test for unpaired data. 3 The P value was greater than 0.15 for each of the six evaluated parameters (for total number of branches, P>0.20; for number of first order branches, P>0.25; for area, P>0.30; for width,
464
M.S. Parker eta/.
P>0.70; for height, P>0.15; and for perimeter, P>0.30). For values of P>I/.05, differences arc considered not statistically significant. 3 Therefore, the members of the two subgroups were trcalcd as a single control group. Similarly, the infected group contained all of the evaluated Purkinje cells from nine embryos which were infected with either the JJ strain (five embryos) or the VS strain (four embryos) of influenza C virus. The P value for these two groups (JJ or VS) was greater than 0.25 for all six evaluated parameters (for total number of branches, P>0.40; for number of first order branches, P>0.50; for area, P>0.25; for width, P>0.25; for height, P>0.60; and for perimeter P>0.40). The members of the two subgroups were treated statistically as a single infected group. Of 27 embryos which were originally included in the experiment (19 embryos inoculated with virus and eight embryo controls), three were excluded from evaluation when the embryos died before harvest, five were harvested for hemagglutination titers 3 days post-inoculation to confirm the efficacy of the inoculation procedure, and five were excluded when the orientation of the cerebelli in the celloidin blocks was incompatible with parasagittal sectioning. The remaining 14 embryos were analysed.
RESULTS
Qualitative alterations in Purkinje cell morphology Morphology of Purkinje cells from the cerebellum of influenza C virus-infected embryos were compared to uninfected embryos. Visual parameters included orientation of the Purkinje cell to the pial surface of the cerebellum, the position of the Purkinje cell somata within the Purkinje cell layer and the appearance of the branching patterns of the Purkinje cells. Purkinje cells of the uninfected group were generally large, with complex arborization and an orderly branching pattern comparable in maturation for chick embryos of this age and under these experimental conditions. The cell somata were located within the narrow Purkinje cell layer and most cells were perpendicular to the pial surface (Figs 1 and 2). Although controls exhibited some variation of orientation and branching patterns, severely aberrant cells were not observed in sections from the uninfected group.
|
\
-.
L
Fig. 2. Camera lucida drawing of Purkinje cell from the uninfected group. Balanced arborization with numerous examplesof dichotomousbranching of both the primary dendrites (solid arrows) and the stem dendrite (open arrow) is seen in this representative cell. The cell soma is located within the Purkinje cell layer(PCL).Scalebar= 20 ~Lm.
Influenza C virus affects chick embryo Purkinje cells
465
Fig. 3. Photograph of a Purkinje cell from the infected group of chick embryos. The Purkinje cell appears much smaller than cells from the uninfected group. The total number of branches (27) is greatly reduced and the area (2031 i~m2) of the dendritic arbor is small. The cell lacks dichotomous branching and the stem dendrite appears to terminate in a node from which secondary branches emerge (arrowhead). Scale bar=20 p,m.
In c o n t r a s t , P u r k i n j e cells r e p r e s e n t a t i v e o f t h e i n f l u e n z a C v i r u s - i n f e c t e d g r o u p w e r e s m a l l e r , with d e n d r i t e s w h i c h w e r e d e c r e a s e d in n u m b e r a n d l e n g t h (Fig. 3). In s o m e cells, d e n d r i t e s w e r e a b s e n t f r o m o n e e n t i r e side (Fig. 4). A s s e e n in Fig. 5, s o m e cells in t h e P u r k i n j e cell l a y e r w e r e a r r a n g e d so t h a t t h e s t e m d e n d r i t e w a s b e n t i n t o a n ' S ' o r a r c h e d shape. T h e o u t e r b r a n c h e s w e r e b e n t t o w a r d t h e P u r k i n j e cell layer, i n s t e a d o f u p w a r d a n d o u t w a r d t o w a r d t h e pial surface o f t h e cerebellum.
M.S. Parker et al.
466
Fig. 4. Camera lucida drawing of Purkinje cell from the infected group. The cell has a greatly reduced number of total dendrites (38) and a decided lack of balance in the dendritic arbor. The branches on the left side of the arbor (solid arrows) appear truncated and lack bifurcation. The left branch (open arrow) also appears to be out of the plane of the rest of the cell. Part of the cell soma lies outside of the Purkinje cell layer. The soma also appears smaller than somata of representative control cells. Scale bar=20 fxm.
P
J
PCL
Fig. 5. Camera lucida drawing of Purkinje cell from the infected group. This cell is also smaller than those from the uninfected group with few total branches (52) and reduced height (62 p.m). The stem dendrite and the secondary dendrites (dosed arrows) which exit from it bend away from the pial surface and into the Purkinje cell layer in a 'U' or 'S' shape. Part of the cell soma is located below the Purkinje cell layer (PCL; open arrow). Scale bar=20 p.m.
Another anomaly seen in the infected group was an increase in the distance between the top of the dendritic arbor and the pial surface (compare Figs 3 and 5 with Figs 1 and 2). Examples of abnormal cells with short terminal branches are seen in Figs 5 and 7. Other cells had noticeably unbalanced shapes and grossly distorted distribution of the branches (Fig. 6). In some cells the soma appeared abnormally small (Figs 4 and 7) and in others part of the soma was located below the Purkinje cell layer (Figs 4 and 5).
Quantitative analysis of Purkinje cells Quantitative differences between the uninfected control group and the influenza C virus-infected group were calculated and compared statistically (Table 1). A significant change, representing a decrease of 25%, was seen for the mean value of the total number of dendrites (from 77 for the uninfected group to 58 for the infected group). Similarly, the number of first order dendrites
Influenza C virus affects chick embryo Purkinje cells
467
f f
f
Fig. 6. Camera lucida drawing of Purkinje cell from the infected group. The upper left side of the cell is devoid of branches (solid arrows), which results in a very abnormal arborization pattern. A truncated branch is also seen (open arrow). The stem dendrite terminates in a node-like structure without dichotomous branching (solid arrowhead). Scale bar=20 ixm.
Fig. 7. Camera lucida drawing of a very small Purkinje cell from the infected group. The number of total branches (52) is reduced, the area of the arbor (2409 i~m2) is diminished and the outermost or first order dendrites appear much shorter than control ceils. The cell soma is extremely small compared to Purkinje cells of the control group. Scale bar=20 ~m.
decreased by 26% (uninfected group, 58; infected group, 39). Comparison of these two parameters showed very significant statistical differences for the influenza C virus-infected embryos (P<0.001). The comparison of the area of the dendritic arbor, 8983 I~m2 for the uninfected group and 6695 ixm2 for the infected group, represents a significant difference (25%; P<0.020). The decrease in the width, height and perimeter of the dendritic arbor of the influenza C virus-infected group when compared to the uninfected group also demonstrated statistically significant differences (Table 1). When the four computer-aided measurements were considered together (height, width, area and perimeter of the Purkinje cell arbor), reductions in these measurements were consistent with the observed decrease in the total number of dendrites.
468
M.S. Parker et al. Table I. The effect of influenza C virus on Purkinje cells of the cerebellum ot chick embryos
Name of parameter Number of total dendrites Number of first order dendrites Area of dendritic arbor Width of dendritic arbor Height of dendritic arbor Perimeter of dendritic arbor
Mean value of control group n=5 (S.D.)
Mean value of infected group n=9 (S.D.)
77 16.8 ) 53 (4.8) 8983 p~m2 (1726) 110 ~m (9.3) 120 ~m (13.8) 374 Ixm (32.5)
58 (5.2) 39 (3.6) 6695 p,m2 (875) 94 p.m (7.3) 103 ~m (7.2) 324 p~m (211.6)
Percent difference*
P value
24.7
<0.001
26.4
<0.00 l
25.5
<0.020
14.5
-::11.020
14.2
<1/.11511
13.3
<0.010
*Decrease between uninfected control group and influenza C virus-infected group.
DISCUSSION Statistically measurable and visually discernible differences were seen in cerebellar Purkinje cells of influenza C virus-infected chick embryos compared to the uninfected embryos. Purkinje cells from infected embryos were smaller, had less complex patterns of arborization, displayed bizarre shapes and patterns of branching, and had somata placed abnormally in the Purkinje cell layer. Such an effect has not been reported previously for influenza C virus. With regard to the irregular placement of the Purkinje cells in virus-infected embryos and the abnormally small somata, similar results were reported by Schweitzer et al. 25 for rats treated with cx-difluoromethyl ornithine (DFMO), which is a specific and irreversible inhibitor of ornithine decarboxylase. Woodward et al. 3° also reported decreased soma size in rats injected with the antiproliferative drug methylazoxymethanol (MAM), and Altman and Anderson 1 reported soma abnormalities in irradiated mice. Abnormal somata may affect the function of the Purkinje cell, since the output of the entire cerebellar cortex is via the axon which exits the cell on the side opposite the stem dendrite.9 Abnormalities in orientation, alignment and distance to the pial surface were also seen in Purkinje cells from influenza C virus-infected embryos. Aberrations in Purkinje cell orientation and alignment are associated with irradiation, 1,4 and with MAM 3° and DFMO 25 treatment, and with exposure to hydrogen sulfide (H2S).ll Treatment with DFMO is also associated with an abnormal increase in the distance from the top of the cell arbor to the pial surface of the cerebellum. 25 Infection with viral agents such as feline panleukopenia virus 12 and parvovirus 2° are, furthermore, associated with disorientation of Purkinje cells. Optimal functioning of Purkinje cells requires proper positioning of the dendrites for interaction with afferent axons, such as the parallel fibers of granule cells and the climbing fibers of the olivocerebellar nuclei,g'9 One of the most visually prominent anomalies seen in Purkinje cells from influenza C virus-infected embryos is abnormal arborization. Especially striking are those cells in which half or one fourth of the expected branching system is missing. Similar severe asymmetry was seen in Purkinje cells of rats treated with c/s-dichlorodiammineplatinum (c/s-DDP), which is an inhibitor of DNA synthesis.24 Also noteworthy are those cells with S-shaped, stunted dendritic arbors and cells that have branches which originate from a node-like structure, or terminate in a club-shaped ending. Club-shaped dendrite terminations have been observed in irradiated mice. 1'4 Stunted Purkinje cells have been reported in studies with the drug MAM 3° and c i s - D D P , 24 and for both experimental parvovirus infection2° and panleukopenia infection12. Statistically significant decreases in the number of dendrites (total and first order), and decreases in area, height, width and perimeter are seen in Purkinje cells from influenza C virus-infected chick embryos. Similar results are seen in a number of other experimental models. Injection with MAM is associated with an overall decrease in the area of the Purkinje cell arbor, 3° as is experimental
Influenza C virus affects chick embryo Purkinje cells
469
infection with lymphochoriomeningitis virus. 2,16 Decrease in the number of dendrites has been reported for irradiation treatment 1,4 and parvovirus infection. 2° Complex interactions exist among the neurons of the cerebellum which can be disrupted by deficiencies of even one cell type. In influenza C virus-infected chick embryos, abnormal Purkinje cells do not have maximal contact with parallel fibers and climbing fibers of the cerebellum. The Purkinje cell is the output neuron of the cerebellar cortex and injury to these neurons can have far reaching effects on the functional integrity of the whole organism. Although there are reports of neuropathology for rare strains of influenza A virus, 7,13,14,23 this is the first description of neurocytological abnormalities caused by infection with influenza C virus. The mechanism by which influenza C virus causes damage to cerebellar Purkinje cells is not yet known. Damage may be primary as a result of infection of the Purkinje cell itself, secondary due to infection of other cerebellar cells and brain cells, or both conditions may contribute to the pathology. In vitro studies utilizing cultured chick embryo brain cells are now in progress to study influenza C virus infection in neuronal cell populations. Results from these studies may help determine tropism of the virus for cells of the CNS and the mechanism by which Purkinje cells of the cerebellum are altered in vivo in influenza C virus-infected chick embryos. Acknowledgements--This work was supported in part by the Buddingh Memorial Research Fund. Virus strains were from the collection of the late G. John Buddingh, M.D.
REFERENCES 1. Altman J. and Anderson W. J. (1972) Experimental reorganization of the cerebellar cortex. I. Morphological effects of elimination of all microneurons with prolonged X-irradiation started at birth. J cornp. Neurol. 146, 355-406. 2. Amaral D. G., Foss J., Monjan A. A. and del Cerro M. (1982) A Golgi analysis of cerebellar immunopathology in neuronatal rats after infection with lymphocytic choriomeningitis virus. Int. J. Neurosci. 17, 71-82. 3. Bancroft H. 1957) t test for small samples. In Introduction to Biostatistics, pp. 172-182. Harper and Brothers, New York. 4. Bradley P. and Berry M. (1976) The effects of reduced climbing and parallel fibre input on Purkinje cell dendritic growth. Brain Res. 109, 133-151. 5. Cherry J. D. (1987) Lower respiratory tract infections. Acute Bronchitis. In Textbook of Pediatric Infectious Disease, Chapter 10, p. 272. Saunders, New York. 6. Dykes A. C., Cherry J. D. and Nolan C. E. (1980) A clinical, epidemiologic, serologic, and virologic study of influenza C virus infection. Arch. Int. Med. 140, 1295-1298. 7. Francis T. Jr and Moore A. E. (1940) Study of the neurotropic tendency in strains of the virus of epidemic influenza. Z exp. Med. 72, 717-728. 8. Furber S. (1983) The organization of the olivocerebellar projection in the chicken. Brain Behav. Evol. 22, 198-21l. 9. Gi•man S.• B••ede• J. R. and Lechtenberg R. ( •98• ) Princip•es •f cerebe••ar •rganizati•n. •n Dis•rders •f the Cerebe••um• Chapter 4, pp. 71-76. F. A. Davis, Philadelphia. 10. Ha H. (1966) A modified Golgi-Cox method with counterstain for the study of synapses. Anat. Rec. 155, 59-64. 11. Hannah R. S. and Roth S, H. (1991) Chronic exposure to low concentrations of hydrogen sulfide produces abnormal growth in developing cerebellar Purkinje cells. Neurosci. Lett. 122, 225-228. 12. Herndon R. M., Margolis G. and Kilham L. (1971) The synaptic organization of the malformed cerebellum induced by perinatal infection with panleukopenia virus. J. Neuropath. exp. Neurol. 30, 557-569. 13. Johnson K. P., Klasnja R. and Johnson R. T. (1971) Neural tube defects of chick embryos: an indirect result of influenza A virus infection. J. Neuropath. exp. Neurol. 30, 68-74. 14. Johnson R. T. (1982) Viral infections of the developing nervous system. In Viral Infections o f the Nervous System, Chapter 9, pp. 203-236. Raven Press, New York. 15. Katagiri S., Ohizumi A. and Homma M. (1983) An outbreak of type C influenza in a children's home. J. inf. Dis. 148, 51-56. 16. Monjan A., Cole G. A., Gilden D. H. and Nathanson N. (1973) Pathogenesis of cerebellar hypoplasia produced by lymphocytic choriomeningitis virus infection of neonatal rats. I. Evolution of disease following injection at 4 days of age. J. Neuropath. exp. Neurol. 32, 110-124. 17. O'Callaghan R. J., Gohd R. S. and Labat D. D. (1980) Human antibody to influenza C virus: its age-related distribution and distinction from receptor analogs. Inf. Immun. 30, 500-505. 18. O'Callaghan R. J. and Labat D. D. (1983) Evidence of a soluble substrate for the receptor-destroying enzyme of influenza C virus. Infect. Immun. 39, 305-310. 19. O'Callaghan R. J. Sr, Loughlin M., Labat D. and Howe C. (1977) Properties of influenza C virus grown in cell culture. J. Virol. 24, 875-882. 20. Oster-Granite M. L. and Herndon R. M. (1976) The pathogenesis of parvovirus-induced cerebellar hypoplasia in the Syrian hamster. Fluorescent antibody, foliation, cytoarchitecture, Golgi and EM studies. J. comp. Neurol. 169, 481-522. 21. Protheroe S. M. and Mellor D. H. (1991) Imaging in influenza A encephalitis. Arch. Dis. Child. 66, 702-705. 22. Ravenhold R. T. and Foege W. H. (1982) 1918 influenza, encephalitis lethargica, Parkinsonism. The Lancet, October 16, 860-864. 23. Robertson G., DeBandi H. O. Jr, Williamson A. P. and Blattner R. J. (1967) Brain abnormalities in early chick embryos infected with influenza A virus. Anat. Rec. 158, 1-10. DN 12:5-D
471~
M . S . P a r k e r et al.
24. Scherini E. (1991) Permanent alterations of the dendritic tree of cerebellar Purkinje neurons in the rat following postnatal exposure to cis-dichlorodiammineplatinum. Acta neuropath. 81, 324-327. 25. Schweitzer L., Robbin A. J. and Slotkin T. A. (1989) Dendritic development of Purkinjc and granule cells in the cerebellar cortex of rats treated postnatally with alpha-difluoromethylornithine. J. Neuropath. exp. Neurol. 48, I 1-22. 26. Smith D. E. and Downs (1978) Postnatal development of the granule celt in the kitten cerebellum. Am..I. Anat. 151, 527-538. 27. Spence H. A. and O'Callaghan R. J. (1985) induction of chick embryo feather malformations by an influenza (7 virus. Teratology 32, 57-64. 28. Strahler A. N. (t%4) Quantitative geomorphology of drainage basins and channel networks. In ttandbook of Applied Hydrology led. Chow V. T.), Section 4-II. McGraw-Hill, New York. 29. Wagaman P. C., Spence H. A. and O'Caltaghan R. J. (1989) Detection of influenza C virus by using an in ,sim esterasc assay. J. clin. Microbiol. 27, 832-836. 30. Woodward D. J., Bickett D. and Chanda R. (1975) Purkinje cell dendritic alterations after transient developmental injury of the external granular layer. Brain Res. 97, 195-241.
Int. J. DevlNeuroscience,Vol. 12, No. 5, pp. 461-470, 1994
0736-5748(94)E0003-K
ElsevierScienceLtd Copyright© 1994ISDN Printed in GreatBritain.All rightsreserved 0736-5748/94$7.00+0.00
THE EFFECT OF INFLUENZA C VIRUS ON THE PURKINJE CELLS OF CHICK EMBRYO CEREBELLUM M. SHARON PARKER,* RICHARD J. O'CALLAGHAN,* DIANE E. SMITH]" and H. ADELE SPENCE* *Department of Microbiology,Immunology and Parasitology, and tDepartment of Anatomy, Louisiana State University Medical Center, 1901 Perdido Street, New Orleans, LA 70112, U.S.A. (Received 1 October 1993; accepted 15 December 1993)
Abstraet--Intra-amniotic inoculation of influenza C virus resulted in observable and quantitatively measurable changes in the Purkinje cellsof chick embryo cerebellum. Purkinje cells were visualizedby the Golgi--Coxprocedure and prepared for statistical and computer evaluation from camera lucida drawings. Four computer-generated measurements (the area of the dendritic arbor, the perimeter of the dendritic tree, and the height and width of the cell's arborization) and two manually counted measurements (total number of branches and the number of first order branches) were made. Analysis of Purkinje ceils from influenza C virus-infected embryos showed disturbances in dendritic arborization patterns and misalignment in the arrangement of .the cells in the Purkinje cell layer compared to control cells. Statistical evaluation of Purkinje cell arborization showed significant decreases in all measured parameters for the influenza C virus-infected members when compared with the members of the uninfected control group. Key words: Purkinje cell, chick, influenza C virus.
Influenza C virus is a m e m b e r of the Orthomyxoviridae family, which also includes the influenza A and B viruses. Influenza viruses have a major epidemiological impact on health and productivity worldwide, due to their association with respiratory infections. Influenza C virus infects most humans by young adulthood and usually causes mild respiratory infection. 6,15,17,19 T h e virus is also an important cause of pneumonitis in infants. 5 With rare exceptions, none of the influenza viruses have b e e n implicated in CNS disease or in neurological sequelae in humans. 14,21,22 In previous work, we described down feather abnormalities which occurred in chick embryos infected with the prototype strain of influenza C virus (j j).27 We have also observed that inoculation of e m b r y o n a t e d eggs with a clinical isolate, designated VS, produced some hatchlings that were unable to stand or walk. Affected chicks that could walk had an ataxic gait. Although leg and gait abnormalities were confined to VS-inoculated chicks, histological examination of h a e m o t o x y l i n eosin-stained sections of both JJ and VS inoculated 19 day embryos suggested that the Purkinje cells of the chicks had been adversely affected. The present work describes visually discernible and statistically measurable changes of the Purkinje cells of chick embryos infected with either of two strains of influenza C virus.
EXPERIMENTAL PROCEDURES Propagation o f virus strains
Virus strains were p r o p a g a t e d in chicken eggs as described previously.18,19, 29 Briefly, egg-derived virus was diluted in calcium- and magnesium-free H a n k s balanced salt solution with 100 units of penicillin and 100 Ixg of streptomycin per milliliter (Grand Island Biological Company, G r a n d Island, NY). Eight day e m b r y o n a t e d chicken eggs (SPAFAS Inc., Norwich, CT) were inoculated with 0.2 ml of diluted virus and placed into a 33°C humidified still-air incubator. Egg fluids were harvested 3 days post-inoculation and tested for hemagglutinating activity with 0.5% chicken red blood cells (Whittaker Bioproducts, Walkersville, MD). Influenza C virus strain JJ was grown in the chorioallantoic cavity of chicken eggs and strain VS was grown in the chick e m b r y o amniotic cavity. Abbreviations: DDP, dichlorodiammineplatinum; DFMO, a-difluoromethyl ornithine; MAM, methylazoxymethanol.
461
462
M.S. Parker et al.
Egg inoculation for viral pathogenesis stud), Nineteen embryonated chicken eggs (9 days old) were inoculated amniotically with either JJ strain or VS strain of influenza C virus diluted in phosphate-buffered saline (pH 7.2) without Pen-Strep. Six eggs were mock infected by inoculation with phosphate-buffered saline alone, and as an additional control, two eggs were not inoculated. All eggs were incubated at 33°C for 3 days in a humidified incubator, after which time the eggs were returned to the egg incubator (39°(?) for an additional 8 days.
Golgi-Cox procedure (Ha modification) Eleven days post-inoculation, at embryonic day 20, cerebellar tissue was harvested from the embryos and processed according to the Ha modification of the Golgi-Cox procedure. 1°,26 The specimens were removed from the impregnating solution after 2 months in the dark and dehydrated using graded alcohols (70%, 80%, 95% and 100% ethanol). Dehydration was completed with ether-ethanol (1:1 ). The tissue was infiltrated with increasing concentrations of celloidin (2%, 4%, 6%, 12% ) and embedded in 12% celloidin. One hundred micrometer sagittal sections of cerebellum were cut on a sliding microtome. The sections were developed in 5% ammonium hydroxide (in 95% ethanol), counterstained with 0.25% aqueous Cresyl Fast Violet, differentiated in 95% ethanol, dehydrated in butyl alcohol and cedarwood oil, and mounted on slides with histological mounting medium (Permount; Fisher Scientific, Fairlawn, N J).
Camera lucida drawings Drawings were made of all Purkinje cells meeting selection criteria which were within folia III, IV and IX from each of three consecutive mid-sagittal cerebellar sections. Selected cells were completely impregnated, were wholly present within the selected section and folium, and were not obscured by other cells or by debris. The three folia, as well as the mid-sagittal location of the selected folia, were chosen to correspond to cerebellar areas associated with hind limbs. Camera lucida drawings were made using a Zeiss microscope equipped with a drawing tube. The drawing of the arbor of each selected cell was outlined by forming the smallest polygon around the outermost extension of the dendrites, as in the method described by Schweitzer et al..25
Quantitative analysis of Purkinje cells" Computer-aided image analysis was done on each selected Purkinje cell using a Zeiss Videoplan Computer equipped with a stylus and magnetic tablet. The stylus was used to trace the polygon encircling the arbor and the Videoplan program (Carl Zeiss, Inc., New York, NY) was used to calculate the area of the dendritic tree enclosed in the polygon (area), the distance around the perimeter of the polygon (perimeter), the width of the dendritic arbor (width) and the height of the cell's arborization (height). In addition to the computer-generated values, the number of first order branches and the total number of branches of each evaluated Purkinje cell was counted manually. The ordering of the counted branches was done according to the method of Strahler, 28 in which each outermost branch is a first order branch. Second order and subsequent ordered segments are formed when two lower ordered branches join; the ordering number of segments from two unequally numbered segments retains the higher order number. The highest ordered segment for each Purkinje cell is the stem dendrite, which exits the Purkinje cell soma.
Statistical analysis Mean values for the six measured parameters were calculated for each evaluated embryo of all Purkinje cells meeting selection criteria in that embryo. Of the analysed embryos, 95 Purkinje cells from the uninfected group of five embryos and 196 Purkinje cells from the infected group of nine embryos met the selection criteria and were analysed by statistical analysis using Student's t-test for unpaired data. 3
Influenza C virus affects chick embryo Purkinje cells
463
Fig. 1. Photograph of Purkinje cell from the uninfected group of chick embryos. The Purkinje cell has a balanced arbor with numerous segments, many of which demonstrate dichotomous branching. The arbor reaches toward the pial surface of the cerebellum (P) and to the external granular layer (EGL). The cell soma is located within the Purkinje cell layer (PCL). Scale bar=20 ixm.
Selection of test groups for analysis The uninfected group consisted of Purkinje cells from two uninoculated embryos and three mock infected embryos. To show that uninoculated and mock infected specimens could be treated as a single group, the P value was calculated for these two subgroups using Student's t-test for unpaired data. 3 The P value was greater than 0.15 for each of the six evaluated parameters (for total number of branches, P>0.20; for number of first order branches, P>0.25; for area, P>0.30; for width,
464
M.S. Parker eta/.
P>0.70; for height, P>0.15; and for perimeter, P>0.30). For values of P>I/.05, differences arc considered not statistically significant. 3 Therefore, the members of the two subgroups were trcalcd as a single control group. Similarly, the infected group contained all of the evaluated Purkinje cells from nine embryos which were infected with either the JJ strain (five embryos) or the VS strain (four embryos) of influenza C virus. The P value for these two groups (JJ or VS) was greater than 0.25 for all six evaluated parameters (for total number of branches, P>0.40; for number of first order branches, P>0.50; for area, P>0.25; for width, P>0.25; for height, P>0.60; and for perimeter P>0.40). The members of the two subgroups were treated statistically as a single infected group. Of 27 embryos which were originally included in the experiment (19 embryos inoculated with virus and eight embryo controls), three were excluded from evaluation when the embryos died before harvest, five were harvested for hemagglutination titers 3 days post-inoculation to confirm the efficacy of the inoculation procedure, and five were excluded when the orientation of the cerebelli in the celloidin blocks was incompatible with parasagittal sectioning. The remaining 14 embryos were analysed.
RESULTS
Qualitative alterations in Purkinje cell morphology Morphology of Purkinje cells from the cerebellum of influenza C virus-infected embryos were compared to uninfected embryos. Visual parameters included orientation of the Purkinje cell to the pial surface of the cerebellum, the position of the Purkinje cell somata within the Purkinje cell layer and the appearance of the branching patterns of the Purkinje cells. Purkinje cells of the uninfected group were generally large, with complex arborization and an orderly branching pattern comparable in maturation for chick embryos of this age and under these experimental conditions. The cell somata were located within the narrow Purkinje cell layer and most cells were perpendicular to the pial surface (Figs 1 and 2). Although controls exhibited some variation of orientation and branching patterns, severely aberrant cells were not observed in sections from the uninfected group.
|
\
-.
L
Fig. 2. Camera lucida drawing of Purkinje cell from the uninfected group. Balanced arborization with numerous examplesof dichotomousbranching of both the primary dendrites (solid arrows) and the stem dendrite (open arrow) is seen in this representative cell. The cell soma is located within the Purkinje cell layer(PCL).Scalebar= 20 ~Lm.
Influenza C virus affects chick embryo Purkinje cells
465
Fig. 3. Photograph of a Purkinje cell from the infected group of chick embryos. The Purkinje cell appears much smaller than cells from the uninfected group. The total number of branches (27) is greatly reduced and the area (2031 i~m2) of the dendritic arbor is small. The cell lacks dichotomous branching and the stem dendrite appears to terminate in a node from which secondary branches emerge (arrowhead). Scale bar=20 p,m.
In c o n t r a s t , P u r k i n j e cells r e p r e s e n t a t i v e o f t h e i n f l u e n z a C v i r u s - i n f e c t e d g r o u p w e r e s m a l l e r , with d e n d r i t e s w h i c h w e r e d e c r e a s e d in n u m b e r a n d l e n g t h (Fig. 3). In s o m e cells, d e n d r i t e s w e r e a b s e n t f r o m o n e e n t i r e side (Fig. 4). A s s e e n in Fig. 5, s o m e cells in t h e P u r k i n j e cell l a y e r w e r e a r r a n g e d so t h a t t h e s t e m d e n d r i t e w a s b e n t i n t o a n ' S ' o r a r c h e d shape. T h e o u t e r b r a n c h e s w e r e b e n t t o w a r d t h e P u r k i n j e cell layer, i n s t e a d o f u p w a r d a n d o u t w a r d t o w a r d t h e pial surface o f t h e cerebellum.
M.S. Parker et al.
466
Fig. 4. Camera lucida drawing of Purkinje cell from the infected group. The cell has a greatly reduced number of total dendrites (38) and a decided lack of balance in the dendritic arbor. The branches on the left side of the arbor (solid arrows) appear truncated and lack bifurcation. The left branch (open arrow) also appears to be out of the plane of the rest of the cell. Part of the cell soma lies outside of the Purkinje cell layer. The soma also appears smaller than somata of representative control cells. Scale bar=20 fxm.
P
J
PCL
Fig. 5. Camera lucida drawing of Purkinje cell from the infected group. This cell is also smaller than those from the uninfected group with few total branches (52) and reduced height (62 p.m). The stem dendrite and the secondary dendrites (dosed arrows) which exit from it bend away from the pial surface and into the Purkinje cell layer in a 'U' or 'S' shape. Part of the cell soma is located below the Purkinje cell layer (PCL; open arrow). Scale bar=20 p.m.
Another anomaly seen in the infected group was an increase in the distance between the top of the dendritic arbor and the pial surface (compare Figs 3 and 5 with Figs 1 and 2). Examples of abnormal cells with short terminal branches are seen in Figs 5 and 7. Other cells had noticeably unbalanced shapes and grossly distorted distribution of the branches (Fig. 6). In some cells the soma appeared abnormally small (Figs 4 and 7) and in others part of the soma was located below the Purkinje cell layer (Figs 4 and 5).
Quantitative analysis of Purkinje cells Quantitative differences between the uninfected control group and the influenza C virus-infected group were calculated and compared statistically (Table 1). A significant change, representing a decrease of 25%, was seen for the mean value of the total number of dendrites (from 77 for the uninfected group to 58 for the infected group). Similarly, the number of first order dendrites
Influenza C virus affects chick embryo Purkinje cells
467
f f
f
Fig. 6. Camera lucida drawing of Purkinje cell from the infected group. The upper left side of the cell is devoid of branches (solid arrows), which results in a very abnormal arborization pattern. A truncated branch is also seen (open arrow). The stem dendrite terminates in a node-like structure without dichotomous branching (solid arrowhead). Scale bar=20 ixm.
Fig. 7. Camera lucida drawing of a very small Purkinje cell from the infected group. The number of total branches (52) is reduced, the area of the arbor (2409 i~m2) is diminished and the outermost or first order dendrites appear much shorter than control ceils. The cell soma is extremely small compared to Purkinje cells of the control group. Scale bar=20 ~m.
decreased by 26% (uninfected group, 58; infected group, 39). Comparison of these two parameters showed very significant statistical differences for the influenza C virus-infected embryos (P<0.001). The comparison of the area of the dendritic arbor, 8983 I~m2 for the uninfected group and 6695 ixm2 for the infected group, represents a significant difference (25%; P<0.020). The decrease in the width, height and perimeter of the dendritic arbor of the influenza C virus-infected group when compared to the uninfected group also demonstrated statistically significant differences (Table 1). When the four computer-aided measurements were considered together (height, width, area and perimeter of the Purkinje cell arbor), reductions in these measurements were consistent with the observed decrease in the total number of dendrites.
468
M.S. Parker et al. Table I. The effect of influenza C virus on Purkinje cells of the cerebellum ot chick embryos
Name of parameter Number of total dendrites Number of first order dendrites Area of dendritic arbor Width of dendritic arbor Height of dendritic arbor Perimeter of dendritic arbor
Mean value of control group n=5 (S.D.)
Mean value of infected group n=9 (S.D.)
77 16.8 ) 53 (4.8) 8983 p~m2 (1726) 110 ~m (9.3) 120 ~m (13.8) 374 Ixm (32.5)
58 (5.2) 39 (3.6) 6695 p,m2 (875) 94 p.m (7.3) 103 ~m (7.2) 324 p~m (211.6)
Percent difference*
P value
24.7
<0.001
26.4
<0.00 l
25.5
<0.020
14.5
-::11.020
14.2
<1/.11511
13.3
<0.010
*Decrease between uninfected control group and influenza C virus-infected group.
DISCUSSION Statistically measurable and visually discernible differences were seen in cerebellar Purkinje cells of influenza C virus-infected chick embryos compared to the uninfected embryos. Purkinje cells from infected embryos were smaller, had less complex patterns of arborization, displayed bizarre shapes and patterns of branching, and had somata placed abnormally in the Purkinje cell layer. Such an effect has not been reported previously for influenza C virus. With regard to the irregular placement of the Purkinje cells in virus-infected embryos and the abnormally small somata, similar results were reported by Schweitzer et al. 25 for rats treated with cx-difluoromethyl ornithine (DFMO), which is a specific and irreversible inhibitor of ornithine decarboxylase. Woodward et al. 3° also reported decreased soma size in rats injected with the antiproliferative drug methylazoxymethanol (MAM), and Altman and Anderson 1 reported soma abnormalities in irradiated mice. Abnormal somata may affect the function of the Purkinje cell, since the output of the entire cerebellar cortex is via the axon which exits the cell on the side opposite the stem dendrite.9 Abnormalities in orientation, alignment and distance to the pial surface were also seen in Purkinje cells from influenza C virus-infected embryos. Aberrations in Purkinje cell orientation and alignment are associated with irradiation, 1,4 and with MAM 3° and DFMO 25 treatment, and with exposure to hydrogen sulfide (H2S).ll Treatment with DFMO is also associated with an abnormal increase in the distance from the top of the cell arbor to the pial surface of the cerebellum. 25 Infection with viral agents such as feline panleukopenia virus 12 and parvovirus 2° are, furthermore, associated with disorientation of Purkinje cells. Optimal functioning of Purkinje cells requires proper positioning of the dendrites for interaction with afferent axons, such as the parallel fibers of granule cells and the climbing fibers of the olivocerebellar nuclei,g'9 One of the most visually prominent anomalies seen in Purkinje cells from influenza C virus-infected embryos is abnormal arborization. Especially striking are those cells in which half or one fourth of the expected branching system is missing. Similar severe asymmetry was seen in Purkinje cells of rats treated with c/s-dichlorodiammineplatinum (c/s-DDP), which is an inhibitor of DNA synthesis.24 Also noteworthy are those cells with S-shaped, stunted dendritic arbors and cells that have branches which originate from a node-like structure, or terminate in a club-shaped ending. Club-shaped dendrite terminations have been observed in irradiated mice. 1'4 Stunted Purkinje cells have been reported in studies with the drug MAM 3° and c i s - D D P , 24 and for both experimental parvovirus infection2° and panleukopenia infection12. Statistically significant decreases in the number of dendrites (total and first order), and decreases in area, height, width and perimeter are seen in Purkinje cells from influenza C virus-infected chick embryos. Similar results are seen in a number of other experimental models. Injection with MAM is associated with an overall decrease in the area of the Purkinje cell arbor, 3° as is experimental
Influenza C virus affects chick embryo Purkinje cells
469
infection with lymphochoriomeningitis virus. 2,16 Decrease in the number of dendrites has been reported for irradiation treatment 1,4 and parvovirus infection. 2° Complex interactions exist among the neurons of the cerebellum which can be disrupted by deficiencies of even one cell type. In influenza C virus-infected chick embryos, abnormal Purkinje cells do not have maximal contact with parallel fibers and climbing fibers of the cerebellum. The Purkinje cell is the output neuron of the cerebellar cortex and injury to these neurons can have far reaching effects on the functional integrity of the whole organism. Although there are reports of neuropathology for rare strains of influenza A virus, 7,13,14,23 this is the first description of neurocytological abnormalities caused by infection with influenza C virus. The mechanism by which influenza C virus causes damage to cerebellar Purkinje cells is not yet known. Damage may be primary as a result of infection of the Purkinje cell itself, secondary due to infection of other cerebellar cells and brain cells, or both conditions may contribute to the pathology. In vitro studies utilizing cultured chick embryo brain cells are now in progress to study influenza C virus infection in neuronal cell populations. Results from these studies may help determine tropism of the virus for cells of the CNS and the mechanism by which Purkinje cells of the cerebellum are altered in vivo in influenza C virus-infected chick embryos. Acknowledgements--This work was supported in part by the Buddingh Memorial Research Fund. Virus strains were from the collection of the late G. John Buddingh, M.D.
REFERENCES 1. Altman J. and Anderson W. J. (1972) Experimental reorganization of the cerebellar cortex. I. Morphological effects of elimination of all microneurons with prolonged X-irradiation started at birth. J cornp. Neurol. 146, 355-406. 2. Amaral D. G., Foss J., Monjan A. A. and del Cerro M. (1982) A Golgi analysis of cerebellar immunopathology in neuronatal rats after infection with lymphocytic choriomeningitis virus. Int. J. Neurosci. 17, 71-82. 3. Bancroft H. 1957) t test for small samples. In Introduction to Biostatistics, pp. 172-182. Harper and Brothers, New York. 4. Bradley P. and Berry M. (1976) The effects of reduced climbing and parallel fibre input on Purkinje cell dendritic growth. Brain Res. 109, 133-151. 5. Cherry J. D. (1987) Lower respiratory tract infections. Acute Bronchitis. In Textbook of Pediatric Infectious Disease, Chapter 10, p. 272. Saunders, New York. 6. Dykes A. C., Cherry J. D. and Nolan C. E. (1980) A clinical, epidemiologic, serologic, and virologic study of influenza C virus infection. Arch. Int. Med. 140, 1295-1298. 7. Francis T. Jr and Moore A. E. (1940) Study of the neurotropic tendency in strains of the virus of epidemic influenza. Z exp. Med. 72, 717-728. 8. Furber S. (1983) The organization of the olivocerebellar projection in the chicken. Brain Behav. Evol. 22, 198-21l. 9. Gi•man S.• B••ede• J. R. and Lechtenberg R. ( •98• ) Princip•es •f cerebe••ar •rganizati•n. •n Dis•rders •f the Cerebe••um• Chapter 4, pp. 71-76. F. A. Davis, Philadelphia. 10. Ha H. (1966) A modified Golgi-Cox method with counterstain for the study of synapses. Anat. Rec. 155, 59-64. 11. Hannah R. S. and Roth S, H. (1991) Chronic exposure to low concentrations of hydrogen sulfide produces abnormal growth in developing cerebellar Purkinje cells. Neurosci. Lett. 122, 225-228. 12. Herndon R. M., Margolis G. and Kilham L. (1971) The synaptic organization of the malformed cerebellum induced by perinatal infection with panleukopenia virus. J. Neuropath. exp. Neurol. 30, 557-569. 13. Johnson K. P., Klasnja R. and Johnson R. T. (1971) Neural tube defects of chick embryos: an indirect result of influenza A virus infection. J. Neuropath. exp. Neurol. 30, 68-74. 14. Johnson R. T. (1982) Viral infections of the developing nervous system. In Viral Infections o f the Nervous System, Chapter 9, pp. 203-236. Raven Press, New York. 15. Katagiri S., Ohizumi A. and Homma M. (1983) An outbreak of type C influenza in a children's home. J. inf. Dis. 148, 51-56. 16. Monjan A., Cole G. A., Gilden D. H. and Nathanson N. (1973) Pathogenesis of cerebellar hypoplasia produced by lymphocytic choriomeningitis virus infection of neonatal rats. I. Evolution of disease following injection at 4 days of age. J. Neuropath. exp. Neurol. 32, 110-124. 17. O'Callaghan R. J., Gohd R. S. and Labat D. D. (1980) Human antibody to influenza C virus: its age-related distribution and distinction from receptor analogs. Inf. Immun. 30, 500-505. 18. O'Callaghan R. J. and Labat D. D. (1983) Evidence of a soluble substrate for the receptor-destroying enzyme of influenza C virus. Infect. Immun. 39, 305-310. 19. O'Callaghan R. J. Sr, Loughlin M., Labat D. and Howe C. (1977) Properties of influenza C virus grown in cell culture. J. Virol. 24, 875-882. 20. Oster-Granite M. L. and Herndon R. M. (1976) The pathogenesis of parvovirus-induced cerebellar hypoplasia in the Syrian hamster. Fluorescent antibody, foliation, cytoarchitecture, Golgi and EM studies. J. comp. Neurol. 169, 481-522. 21. Protheroe S. M. and Mellor D. H. (1991) Imaging in influenza A encephalitis. Arch. Dis. Child. 66, 702-705. 22. Ravenhold R. T. and Foege W. H. (1982) 1918 influenza, encephalitis lethargica, Parkinsonism. The Lancet, October 16, 860-864. 23. Robertson G., DeBandi H. O. Jr, Williamson A. P. and Blattner R. J. (1967) Brain abnormalities in early chick embryos infected with influenza A virus. Anat. Rec. 158, 1-10. DN 12:5-D
471~
M . S . P a r k e r et al.
24. Scherini E. (1991) Permanent alterations of the dendritic tree of cerebellar Purkinje neurons in the rat following postnatal exposure to cis-dichlorodiammineplatinum. Acta neuropath. 81, 324-327. 25. Schweitzer L., Robbin A. J. and Slotkin T. A. (1989) Dendritic development of Purkinjc and granule cells in the cerebellar cortex of rats treated postnatally with alpha-difluoromethylornithine. J. Neuropath. exp. Neurol. 48, I 1-22. 26. Smith D. E. and Downs (1978) Postnatal development of the granule celt in the kitten cerebellum. Am..I. Anat. 151, 527-538. 27. Spence H. A. and O'Callaghan R. J. (1985) induction of chick embryo feather malformations by an influenza (7 virus. Teratology 32, 57-64. 28. Strahler A. N. (t%4) Quantitative geomorphology of drainage basins and channel networks. In ttandbook of Applied Hydrology led. Chow V. T.), Section 4-II. McGraw-Hill, New York. 29. Wagaman P. C., Spence H. A. and O'Caltaghan R. J. (1989) Detection of influenza C virus by using an in ,sim esterasc assay. J. clin. Microbiol. 27, 832-836. 30. Woodward D. J., Bickett D. and Chanda R. (1975) Purkinje cell dendritic alterations after transient developmental injury of the external granular layer. Brain Res. 97, 195-241.