Rabies is a highly neurotropic virus that produces fatal encepha-
lomyelitis in humans and animals (1,2). Attenuated rabies virus
strains have been traditionally generated by serial passages through
animal brains, cell culture, or both. The attenuated SAD-B19 vaccine
strain is a derivative of the Street Alabama Dufferin strain isolated
from a rabid dog in Alabama in 1935, attenuated by multiple pas-
sages in baby hamster kidney (BHK) cells, and selected for vaccine
production on the basis of its thermostability (3). SAD-B19 has been
widely used in Europe for oral vaccination of foxes (3,4). Recently,
SAD-B19 was demonstrated to be avirulent in young foxes, dogs,
and other carnivores following different routes of inoculation,
including the intracerebral route, while pathogenicity was noted in
orally-infected wild rodents, including mice (5). Recombinant rabies
virus strains are powerful tools in rabies pathogenesis research
because they allow targeted exchange of genes and regulatory ele-
ments, while allowing the construction of modified rabies viruses
that are highly attenuated and have potential as live vaccines (6). A
recombinant clone of SAD-B19, SAD-L16 (L16), was generated from
a full-length cDNA clone (7), and L16 was observed to be highly
neurovirulent after intracerebral inoculation of adult and suckling
mice (8). This study examines the neuropathologic changes in the
encephalitis produced by intracerebral inoculation of L16 in suckling
mice and notes the differences in the infection produced by the more
virulent challenge virus standard (CVS) — 11 strain (9). L16 was
found to cause severe encephalitis associated with widespread
Materials and methods
The SAD-L16 strain of fixed rabies virus was obtained from
Teshome Mebatsion (Intervet International, Boxmeer, The
Netherlands). The generation of recombinant rabies virus L16 has
been previously described; L16 contains the authentic sequence of
the SAD-B19 vaccine strain, which was derived from a full-length
cDNA clone, and BSR-T7/5 cells expressing phage T7 RNA poly-
merase were used to recover infectious virus from cDNA (7,8).
Animals and inoculations
The experimental animal protocol followed the Canadian Council
on Animal Care Guidelines on Animal Use and was approved by
Recombinant rabies virus vaccine strain SAD-L16 inoculated intracerebrally
in young mice produces a severe encephalitis with extensive
Pamini Rasalingam, John P. Rossiter, Alan C. Jackson
Seven-day-old ICR mice were infected by intracerebral inoculation with recombinant rabies virus vaccine strain SAD-L16.
Infected mice developed severe and fatal encephalitis with rabies virus-infected neurons in widespread regions of the brain.
There was extensive neuronal death with predominant features of apoptosis, as assessed by light and electron microscopy,
terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) staining, and immunohistochemical
staining for activated caspase-3. Although SAD-L16 is a neuroattenuated rabies virus, it is fully capable of spreading efficiently
and inducing widespread neuronal apoptosis in the immature mouse brain.
Des souris ICR âgées de 7 jours ont été infectées par inoculation intracérébrale avec la souche recombinante du virus de la rage SAD-L16.
Les souris infectées développèrent une encéphalite sévère et fatale avec infection des neurones par le virus de la rage dans plusieurs régions
du cerveau. Il y avait mort neuronale extensive avec des caractéristiques évidentes d’apoptose, tel que démontré par microscopie photonique
et électronique, par coloration par marquage de l’extrémité de la coupure simple brin à la dUTP-biotine médiée par la déoxynucléotidyl
transférase, et par coloration immunohistochimique pour la caspase-3 activée. Bien que la souche SAD-L16 soit un virus rabique neuro-atténué,
elle est en mesure de se répandre efficacement et d’induire une apoptose neuronale généralisée dans le cerveau de souris immatures.
(Traduit par Docteur Serge Messier)
Department of Microbiology and Immunology (Rasalingam, Jackson); Department of Pathology and Molecular Medicine (Rossiter); Department
of Medicine (Neurology) (Jackson), Queen’s University, Kingston, Ontario.
Address all correspondence and reprint requests to Dr. Alan C. Jackson; telephone: (613) 548-1316; fax: (613) 548-1317;
Dr. Jackson’s current address is the Kingston General Hospital, Connell 725, 76 Stuart Street, Kingston, Ontario K7L 2V7.
Received September 16, 2004. Accepted January 6, 2005.
100 The Canadian Journal of Veterinary Research 2005;69:100–105
2000;64:0–00 The Canadian Journal of Veterinary Research 101
the Queen’s University Animal Care Committee. Timed pregnant
female ICR mice (pathogen-free) were purchased (Charles River
Canada, St. Constant, Quebec) and their 7-day-old offspring of either
sex were used. Mice were inoculated intracerebrally with 20 L
containing 1000 focus-forming units of L16 diluted in phosphate
buffered saline solution (PBSS) with 4% fetal bovine serum (10,11).
Uninfected control mice of the same age were inoculated with only
the diluent (mock-infected). Three to 4 infected mice and 1 to
2 mock-infected (uninfected) mice were euthanized at daily intervals
for 6 d after inoculation. The mice were anesthetized with methoxy-
flurane and perfused with buffered 4% paraformaldehyde.
Preparation of tissue sections
Brains were removed and immersion-fixed in the same fixative
for 24 h at 4°C. Coronal brain tissue sections (6 m) were prepared
after dehydration and embedding in paraffin. Tissues were stained
with cresyl violet for light microscopic histological examination.
Immunoperoxidase staining for rabies
Immunoperoxidase staining was performed on all mouse brains
for rabies virus antigen. Sections were stained for rabies virus anti-
gen by the avidin-biotin-peroxidase complex method using a mouse
monoclonal antibody against rabies virus nucleocapsid protein
immunoglobulin (Ig)G 5DF12 (obtained from Alexander I. Wandeler,
Centre of Expertise for Rabies, Canadian Food Inspection Agency,
Nepean, Ontario) as the primary antibody, as previously described
(12). Tissues from uninfected mice were used as controls. In brief,
tissue sections were deparaffinized and hydrated. Sections were
successively treated with 5% normal rabbit serum, primary antibody
diluted 1:160 in 2% normal rabbit serum, biotinylated rabbit anti-
mouse IgG (Vector Laboratories, Burlingame, California, USA)
diluted 1:100 in 2% normal rabbit serum, 1% hydrogen peroxide in
methanol, avidin-biotinylated horseradish peroxidase complex
(Vector Laboratories), 3,3-diaminobenzidine tetrachloride
(Polysciences, Warrington, Pennsylvania, USA) with 0.01% hydrogen
peroxide, and 0.5% cupric sulfate in 0.15 M sodium chloride. The
slides were counterstained with hematoxylin.
Immunoperoxidase staining for activated
Immunoperoxidase staining was performed on all mouse brains
for activated caspase-3. Tissue sections were deparaffinized and
hydrated, and then heated in a microwave for 1 min at high power
followed by 9 min at medium power in sodium citrate buffer
(pH 6.0) for antigen unmasking. Sections were successively treated
with 5% normal goat serum, rabbit polyclonal antibody directed
against cleaved (activated) caspase-3 (Asp175) (Cell Signaling
Technology, Beverley, Massachusetts, USA) diluted 1:400 in 2%
normal goat serum, biotinylated goat anti-rabbit IgG (Vector
Laboratories) diluted 1:100 in 2% normal goat serum, 1% hydrogen
peroxide in methanol, avidin-biotinylated horseradish peroxidase
complex (Vector Laboratories), 3,3-diaminobenzidine tetrachloride
(Polysciences) with 0.01% hydrogen peroxide, and 0.5% cupric
sulfate in 0.15 M sodium chloride. The slides were lightly counter-
stained with hematoxylin.
Mice were perfused with a mixture of 2.5% glutaraldehyde and
2% paraformaldehyde in 0.1 M cacodylate buffer. Brain tissues were
removed and immersion-fixed in the same fixative for several days
at 4°C and then immersed in 1% osmium tetroxide, dehydrated in
a graded series of ethanols, cleared with propylene oxide, and infil-
trated with resin (Jembed resin; J.B. EM Services, Dorval, Quebec).
Sections (1 m thick) taken from 8 mice were stained with toluidine
blue and examined using light microscopy. Ultrathin sections from
4 mice were stained with uranyl acetate and lead citrate and exam-
ined with an electron microscope (Hitachi H7000; Hitachi,
Schaumburg, Illinois, USA) at 75 kV.
Transferase-mediated dUTP-biotin nick end
labelling (TUNEL) staining
Oligonucleosomal DNA fragmentation was assessed in situ in
sections using the terminal deoxynucleotidyl TUNEL method and a
TdT-FragEL DNA fragmentation detection kit (Catalogue no. QIA33;
Oncogene Research Products, San Diego, California, USA) with the
manufacturer’s protocol for paraffin-embedded tissue sections. The
TUNEL staining was performed on brains from 11 mice.
Figure 1. Immunoperoxidase staining for rabies virus antigen in a cerebral
hemisphere 3 d postinoculation (pi) showing heavy involvement of the
cerebral cortex with relative sparing of the neostriatum (A) in cerebellar
Purkinje cells with involvement of their processes in the molecular layer
(B), in pyramidal neurons of the hippocampus, and in a few neurons in the
dentate gyrus (C), along with low background staining in the hippocampus
of a mock-infected mouse (D). Histopathology of periventricular germinal
matrix (E) 5 d pi showing a few apoptotic cells with nuclear chromatin
condensations (arrowheads), and transferase-mediated dUTP-biotin nick
end labeling (TUNEL) and staining is present in cells in periventricular
germinal matrix (F) 4 d pi, immunoperoxidase — hematoxylin (A to D),
cresyl violet (E), TUNEL staining — methyl green (F). Magnification:
A 10; B 100; C, D 25; E 385; F 230.
102 The Canadian Journal of Veterinary Research 2000;64:0–00
The L16-infected mice developed signs of limb weakness, ataxia
(present in 9 out of 19), and growth retardation on day 3 post-
inoculation (pi). Over the next 2 d there was progression to quadri-
paresis and all surviving mice became moribund by day 5 or 6 pi.
Rabies virus antigen distribution
Rabies virus antigen was first detected at 2 d pi in neurons in the
brainstem, diencephalon, cerebellum (Purkinje cells), hippocampus
(pyramidal layer), and cerebral cortex and also in a few ependymal
cells lining the lateral ventricles. At this time the involvement of
hippocampal pyramidal neurons was marked. By day 3 pi the num-
ber of infected neurons in the brainstem and cerebral cortex had
markedly increased, and viral antigen was noted in deep cerebellar
nuclei. There was also mild involvement of the neostriatum
(Figure 1A). On day 3 pi there was increased staining in Purkinje
cells (Figure 1B) and in both hippocampal pyramidal neurons and
neurons in the dentate gyrus of the hippocampus (Figure 1C), while
there was low background staining in the brains of mock-infected
mice (Figure 1D). There was very mild multifocal involvement of
the periventricular germinal matrix, but the vast majority of these
immature cells were uninfected. On day 4 pi there was marked
infection in the neostriatum and infection was observed in the inter-
nal granular layer of the cerebellum. Reduced staining was observed
in hippocampal pyramidal neurons at late time points associated
with marked neuronal loss in this region, while staining increased
in the dentate gyrus. Rabies virus antigen was not observed in neu-
rons in the external granular layer of the cerebellum.
Mock-infected mice showed only occasional apoptotic neurons in
the regional areas of the brain, likely due to naturally occurring
neuronal death during postnatal development. The earliest patho-
logic changes in infected mice were observed at 2 d pi in scattered
pyramidal neurons of the hippocampus and cerebellum, including
neurons in deep cerebellar nuclei and Purkinje cells (Table I), along
with cytological features of apoptosis, including karyorrhectic con-
densation of nuclear chromatin and cytoplasmic shrinkage. By day
3 pi, there was extensive apoptotic cell death of hippocampal pyra-
midal neurons, most prominently in the CA2 field (Figures 2A
and B). Apoptotic changes were also present at this time in neurons
in the brainstem, diencephalon, dentate gyrus of the hippocampus
(particularly involving the inner layers), and cerebral cortex. By
days 3 to 4 pi mononuclear inflammatory changes were present in
the leptomeninges, perivascular regions, and brain parenchyma, and
there was neuronal apoptosis in all of the regional areas of the brain.
Ongoing loss of hippocampal pyramidal neurons on days 4 and
5 pi resulted in a near-complete loss of neurons in all CA fields
(Figures 2C and 3B). Neuronal apoptosis also became prominent in
the brainstem, diencephalon, cerebellum (deep cerebellar nuclei,
internal granular layer, and Purkinje cells) (Figures 3C and D),
dentate gyrus of hippocampus (Figure 2D), and cerebral cortex
(Figures 2E to H) at late time points. A small number of cells also
showed apoptotic changes in the periventricular germinal matrix
(Figure 1E). Apoptosis of neurons in the external granular cell layer
of the cerebellum was not noted to be more frequent than in mock-
Ultrastructural examination of both the cerebellum and hip-
pocampus showed a predominant pattern of apoptotic cell death,
characterized by dense nuclear chromatin condensations (Figure 4A),
cell shrinkage with intact cytoplasmic membranes and relative
preservation of organelle integrity. Some dying cells also contained
numerous autophagic-type vacuoles (Figure 4A). In addition, some
shrunken cerebellar Purkinje cells (Figure 4A) and hippocampal
pyramidal neurons (Figure 4B) lacked prominent nuclear chro-
matin condensations. These cells exhibited dilation of the nuclear
envelope; endoplasmic reticulum; Golgi apparatus; and, to a lesser
extent, of mitochondria, in combination with preservation of cyto-
plasmic membrane integrity, a pattern characteristic of type 3B
Table I. Morphological changes of apoptosis were evaluated in neurons in the midbrain, cerebellum (deep
cerebellar nuclei and Purkinje cells), thalamus, hippocampus (CA1 and CA3 regions), and cerebral cortex
(40 objective). A semiquantitative evaluation of the severity of apoptotic changes was performed on
days 2 to 6 postinoculation (pi) with the following grading scheme: 0, no significant changes; 1, mild
changes; 2, moderate changes; 3, severe changes; and 4, very severe changes or neuronal loss. The
identity of all slides was masked during scoring in order to prevent bias in the evaluation. Rating scale
scores are expressed as the mean score, standard error: 0, no significant changes; 1, mild changes;
2, moderate changes; 3, severe changes; and 4, very severe changes or neuronal loss
2000;64:0–00 The Canadian Journal of Veterinary Research 103
(“cytoplasmic”) cell death described by Clarke and colleagues
(13,14). Swollen astrocyte processes were frequently seen adjacent to
dying neurons (Figure 4A) and accounted for much of the neuropil
vacuolation seen by light microscopy (Figures 3B and D).
Transferase-mediated dUTP-biotin nick end
At day 2 pi, positive TUNEL staining was observed in scattered
hippocampal pyramidal neurons; in neurons in the internal granular
layer of the cerebellum; and, rarely, in neurons in the cerebral cortex
and diencephalon. By day 3, and increasing on days 4 and 5 pi,
TUNEL staining involved many neurons in the brainstem, dien-
cephalon, cerebellum (deep cerebellar nuclei, internal granular layer,
and Purkinje cells) (Figure 5A), cerebral cortex (Figure 5B), and hip-
pocampus (including dentate gyrus) (Figure 5C), while there was
low background staining in the brains of mock-infected mice
(Figure 5D). The TUNEL staining was also noted in a few neurons
in the periventricular germinal matrix (Figure 1F) and in the external
granular layer of the cerebellum (Figure 5A). Morphological features
of apoptotic cell death above the background level in mock-infected
animals were observed in all of these regions except the external
granular layer of the cerebellum.
Activated caspase-3 staining
Immunohistochemical staining for activated caspase-3 was first
observed on day 3 pi in neurons in the diencephalon, hippocampus
(pyramidal neurons and dentate gyrus), and cerebral cortex.
Increased staining was present on days 4 to 6 pi and also involved
the cerebellum, including Purkinje cells and the internal granular
layer (Figures 5E to G) with low background staining in mock-
infected mouse brains (Figure 5H). Staining was not noted in neurons
in deep cerebellar nuclei and staining was only present in the brain-
stem neurons of 2 mice.
L16 produced severe and fatal encephalitis after intracerebral
inoculation of young mice. There was rapid viral spread within the
brain and infection of numerous neurons in multiple brain regions
developed by 3 d pi. The neuronal infection was associated with
neuronal death with features of apoptosis and marked neuronal loss.
Typical morphologic features of apoptotic cell death were observed
using both light and electron microscopy. In addition, ultrastructural
elements of autophagic cell death and also of type 3B (“cytoplasmic”)
cell death were seen in some cells, consistent with the morphologi-
cal diversity of cell death that can be seen in vivo (13,14). The
encephalitis resulted in severe neuronal loss in regional areas, includ-
ing the hippocampus and cerebellum, which was much more severe
than previously observed in CVS-infected suckling mice (9). Positive
Figure 2. Histopathologic examination showing hippocampi of a (A) mock-
infected and (B) L16-infected mouse 3 d postinoculation (pi) showing
selective severe neuronal loss in the CA2 region and (C) L16-infected mouse
5 d pi showing almost complete loss of pyramidal neurons. (D) Dentate gyrus
of the hippocampus 5 d pi showing condensation of nuclear chromatin of
many neurons in the inner layer. Cerebral cortex of a mock-infected mouse
(E) and L16-infected mice 4 d pi showing condensation of nuclear chroma-
tin of a few neurons (F) and 5 d pi showing condensation of nuclear chro-
matin of many neurons (G) and severe neuronal loss and perivascular
inflammatory inflitrates (H). Cresyl violet. Magnification: A, B 20; C 30;
D 100; E, F 320; G 390; H 40.
Figure 3. Histopathology in pyramidal neurons of the hippocampus
(CA3 field) (A and B) and in the cerebellum (C and D) after mock-infection
(A and C) and 4 to 5 d postinnoculation (pi) with L16 (B and D). The
L16-infected pyramidal neurons show condensation of nuclear chromatin
and cytoplasmic shrinkage and associated neuropil vacuolation (B). In the
cerebellum there are nuclear condensations in many neurons in the internal
granular layer and in scattered Purkinje cells (inset). There is also promi-
nent vacuolation of the neuropil, especially in the molecular layer (D).
Toluidine blue stained resin embedded sections. Magnification: A, B 370;
C, D 230, inset 1200.
104 The Canadian Journal of Veterinary Research 2000;64:0–00
TUNEL staining, indicating oligonucleosomal fragmentation of
DNA, and immunohistochemical staining for activated caspase-3, a
downstream executioner of the apoptotic program (15), provide
strong supporting biochemical evidence of a predominant apoptotic
mechanism of neuronal death in this model.
The attenuated L16 strain produced neuronal apoptosis in the
mouse brain in the present model, and the more neurovirulent
CVS-11 strain also produced widespread neuronal apoptosis in the
brains of 6-day-old mice when inoculated by the intracerebral route
(9). The CVS-infected mice survived for 1 d less (4 d), but, unlike
L16-infected mice, they did not demonstrate marked loss in hip-
pocampal pyramidal neurons or death of Purkinje cells. In addition,
the apoptotic neuronal changes were less severe in the internal
granular layer of the cerebellum in the CVS-infected mice. However,
CVS-infected mice demonstrated apoptosis in neurons in the exter-
nal granular layer of the cerebellum. This likely occurred by indirect
mechanisms because these neurons were uninfected. In summary,
L16 and CVS demonstrated some differences in their neuronal targets
and in the severity of neuronal injury that they induced in specific
regional areas of the brain.
The L16 vaccine strain is considered to be highly attenuated
in vitro and in vivo under certain conditions. It is highly attenu-
ated in carnivores (5), which makes it an effective vaccine strain.
L16 remains highly neurovirulent in both young and adult mice
after intracerebral inoculation, as well as in young mice after
peripheral inoculation (8,16). The present study indicates that
neuronal apoptosis is extensive in the fatal infection of young
mice. The destruction of virus infected cells by apoptosis has been
proposed as an innate host cellular response that acts to limit viral
propagation during infection (17–19), and it is likely that this is
an important mechanism by which neuroinvasion is limited in
the adult mouse. However, once L16 gains access to the mouse
brain, by either a peripheral route or by intracerebral inoculation,
it spreads efficiently and induces widespread neuronal apoptosis.
This study suggests that neuroattenuation of L16 predominantly
occurs by a restriction in susceptibility of carnivore hosts and also
by restriction of neuroinvasiveness, particularly in adult hosts,
because the virus remains highly neurovirulent and capable of
inducing severe cytopathology by an apoptotic mechanism in vivo.
Hopefully, a better understanding of how rabies virus injures and
kills neurons will put us one step closer to conquering this ancient
The authors are grateful for recombinant rabies virus vaccine
strain SAD-L16 from Teshome Mebatsion (Intervet International,
B.V., Boxmeer, The Netherlands) and for monoclonal antibody 5DF12
Figure 4. Ultrastructure of cerebellar neurons (A) and hippocampal neurons
(CA3 field) (B) 4 to 5 d postinoculation (pi) with L16. A shrunken Purkinje
cell (P) and a hippocampal pyramidal neuron (H) show characteristic fea-
tures of type 3B (“cytoplasmic”) cell death, with expansion of the perinu-
clear space and prominent dilation of the endoplasmic reticulum and Golgi
apparatus (B, inset). Dense nuclear chromatin condensations are seen in
2 internal granule cell neurons (*), one of which contains numerous
autophagic-type vacuoles (arrowheads). Swollen astrocyte processes are
present (). Magnification: A 3000; B 2300, inset 14 000.
Figure 5. Transferase-mediated dUTP-biotin nick end labelling (TUNEL) and
staining (A to D) and immunoperoxidase staining for activated caspase-3
(E to H) in L16-infected mouse brains. The TUNEL staining is present in
neurons in the cerebellar external and internal granular layers (A) 6 d
postinoculation (pi), in many neurons in the cerebral cortex (B) 4 d pi,
and in hippocampal pyramidal neurons and neurons in the dentate gyrus of
the hippocampus (C) 5 d pi, while there is low background staining in the
hippocampus of a mock-infected mouse (D). Activated caspase-3 staining
is present in many neurons in the internal granular layer of the cerebellum
(E) 5 d pi, in some Purkinje cells (F) 4 d pi, in both hippocampal pyramidal
neurons and in neurons in the dentate gyrus of the hippocampus (G) 4 d pi,
and there is low background staining in the hippocampus of a mock-infected
mouse (H). A to D, TUNEL staining — methyl green; E to H, Immunoperoxidase
— hematoxylin. Magnification: A 150; B 95; C, D 25; E 130;
F 390; G, H 20.
2000;64:0–00 The Canadian Journal of Veterinary Research 105 Download full-text
from Alexander I. Wandeler (Centre for Rabies Expertise, Canadian
Food Inspection Agency, Nepean, Ontario). This work was supported
by a research contract with Intervet International, B.V. (Boxmeer,
The Netherlands), Canadian Institutes of Health Research grant
MOP — 64376, and the Queen’s University Violet E. Powell Research
Fund (all to A.C. Jackson).
1. Jackson AC. Human disease. In: Jackson AC, Wunner WH, eds.
Rabies. San Diego: Academic Press, 2002:219–244.
2. Niezgoda M, Hanlon CA, Rupprecht CE. Animal rabies. In:
Jackson AC, Wunner WH, eds. Rabies. San Diego: Academic Press,
3. Blancou J, Meslin F-X. Modified live-virus rabies vaccines for
oral immunization of carnivores. In: Meslin F-X, Kaplan MM,
Koprowski H, eds. Laboratory Techniques in Rabies. 4th ed.
Geneva: World Health Organization, 1996:324–337.
4. Schneider LG. Rabies virus vaccines. Dev Biol Stand 1995;84:
5. Vos A, Neubert A, Aylan O, et al. An update on safety studies of
SAD B19 rabies virus vaccine in target and non-target species.
Epidemiol Infect 1999;123:165–175.
6. Schnell MJ, Tan GS, Dietzschold B. The application of reverse
genetics technology in the study of rabies virus (RV) pathogen-
esis and for the development of novel RV vaccines. J Neurovirol
7. Schnell MJ, Mebatsion T, Conzelmann KK. Infectious rabies
viruses from cloned cDNA. EMBO J 1994;13:4195–4203.
8. Mebatsion T. Extensive attenuation of rabies virus by simultane-
ously modifying the dynein light chain binding site in the
P protein and replacing Arg333 in the G protein. J Virol 2001;75:
9. Jackson AC, Park H. Apoptotic cell death in experimental rabies
in suckling mice. Acta Neuropathol 1998;95:159–164.
10. King AA. Cell culture of rabies virus. In: Meslin F-X, Kaplan
MM, Koprowski H, eds. Laboratory Techniques in Rabies. 4th ed.
Geneva: World Health Organization, 1996:114–130.
11. Mims CA. Intracerebral injections and the growth of viruses in
the mouse brain. Br J Exp Pathol 1960;41:52–59.
12. Jackson AC, Ye H, Phelan CC, et al. Extraneural organ involve-
ment in human rabies. Lab Invest 1999;79:945–951.
13. Borsello T, Mottier V, Castagne V, Clarke PG. Ultrastructure of
retinal ganglion cell death after axotomy in chick embryos.
J Comp Neurol 2002;453:361–371.
14. Clarke PG. Developmental cell death: morphological diversity
and multiple mechanisms. Anat Embryol (Berl) 1990;181:
15. Friedlander RM. Apoptosis and caspases in neurodegenerative
diseases. N Engl J Med 2003;348:1365–1375.
16. Rasalingam P, Rossiter JP, Mebatsion T, Jackson AC. Comparative
pathogenesis of the SAD-L16 strain of rabies virus and a mutant
modifying the dynein light chain binding site of the rabies virus
phosphoprotein in young mice. Virus Res 2005 (In press).
17. Alcami A, Koszinowski UH. Viral mechanisms of immune eva-
sion. Trends Microbiol 2000;8:410–418.
18. Allsopp TE, Fazakerley JK. Altruistic cell suicide and the special-
ized case of the virus-infected nervous system. Trends Neurosci
19. Fazakerley JK, Allsopp TE. Programmed cell death in virus
infections of the nervous system. Curr Top Microbiol Immunol