Axon–Myelin Interactions during a Viral Infection of the
Central Nervous System
Michel Brahic1*, Jean-Pierre Roussarie2
1Department of Microbiology-Immunology, Stanford University School of Medicine, Stanford, California, United States of America, 2Molecular and Cellular Neuroscience,
The Rockefeller University, New York, New York, United States of America
Theiler’s virus offers a remarkable
example of a pathogen that navigates the
various cells of the organism to evade
immune responses and establish a persis-
tent infection. Here, we discuss the
transition from neuron to myelin and
oligodendrocyte infection, a step that is
crucial for the persistence of this virus in
the central nervous system (CNS).
CNS myelin is an extension of the
cytoplasmic membrane of oligodendro-
cytes wrapped numerous times around
axons. An oligodendrocyte sends many
such extensions and can myelinate up to
segments are separated by short unmy-
elinated regions called nodes of Ranvier.
Cytoplasm is totally extruded from myelin
except in areas where it forms channels
that are in continuity with the oligoden-
drocyte cell body. These channels form
the so-called ad-axonal inner loop and the
paranodal loops at the level of the nodes of
Ranvier. Inner and paranodal loops are in
close contact with the axon membrane
(Figure 1). (For a review of myelin and
node organization, see .)
Theiler’s virus, a mouse picornavirus, is
responsible for a peculiar neurological
disease. It infects neurons and spreads by
fast axonal transport for approximately 2
weeks following intracerebral inoculation.
Depending on their genetic background,
mice may clear this infection or remain
persistently infected. The virus is no longer
in neurons during persistent infection.
Instead, it is found in white matter, in
oligodendrocytes, in the cytoplasmic chan-
nels of myelin, and in macrophages. The
infection is focal and results in inflamma-
tion, primary demyelination, and some
axonal damage. For more detailed reviews
of Theiler’s virus and its pathogenesis, the
reader is referred to [2,3].
In the course of studying this persistent
infection, our laboratory observed that two
myelin mutants, the shiverer and rumpshaker
mice, were totally resistant to persistent
infection, whereas the parental strains were
susceptible . These strains bear muta-
tions in the Mbp and Plp1 genes, which
encode two major myelin components,
myelin basic protein (MBP) and proteolipid
protein (PLP), respectively. Shiverer mice
make very little myelin, whereas in rump-
shaker mice, the amount of myelin is
reduced and the periodicity of myelin
leaflets is slightly altered [5,6]. Theiler’s
virus infects neurons and is transported in
axons in these mutant mice just as in wild-
type mice. However, it disappears from the
CNS of the mutants after 2–3 weeks.
Various experiments showed that this
clearance was not immune mediated and
that shiverer oligodendrocytes grown in
culture were permissive to viral replication
. These results suggested that myelin
might play a crucial role in viral persis-
tence. What could that role be?
We postulated that viruses transported
in axons infect the surrounding myelin and
spread to oligodendrocyte cell bodies, then
to macrophages, where infection persists.
Accordingly, myelin would be an obliga-
tory passage that would be prevented by
the shiverer and rumpshaker mutations. We
tested this hypothesis by introducing virus
in the vitreous chamber of the eye. The
virus infected retinal ganglion cells and
was transported anterogradely in the
axons of the optic nerve. Infected oligo-
dendrocytes were observed in the nerve of
wild-type mice as early as 3 days post-
inoculation. The only possible source of
infection for these oligodendrocytes is the
axons of retinal ganglion cells. In contrast,
the infection of oligodendrocyte cell bodies
was considerably impaired in shiverer and
rumpshaker mice, although the number of
oligodendrocytes is normal in these mu-
Importantly, by performing these experi-
ments with Wldsmice, a mutant whose
axons are strikingly resistant to degenera-
tion, we showed that the virus was able to
traffic from axon to myelin in the absence
of axonal degeneration .
Enveloped viruses with their lipid bilay-
er and glycoproteins can exit the cyto-
plasm by budding from the plasma
membrane and entering a new host cell
by fusion. Classically non-enveloped virus-
es such as picornaviruses exit by causing
cell lysis. How can Theiler’s virus cross
from intact axons into the surrounding
myelin? Before presenting hypotheses it is
necessary to review briefly the salient
aspects of axon–myelin interactions.
Mice and humans with myelin muta-
tions have provided most of our under-
standing of axon–myelin interactions. PLP
is a major myelin protein. Plp1nul mice
have myelinated axons, surprisingly, and
mice are normal at birth . However, in
late life, organelles accumulate in the
axons at the level of the nodes of Ranvier
and there is axonal swelling and degener-
ation . The Plp1 gene is on the X
chromosome. Because in females one of
the X chromosomes is inactivated, the
oligodendrocytes of a female heterozygous
for a Plp1 mutation form a mosaic of wild-
type and mutant cells. In these mice, the
same axon can be myelinated alternatively
by wild-type and by mutant oligodendro-
Citation: Brahic M, Roussarie J-P (2009) Axon–Myelin Interactions during a Viral Infection of the Central
Nervous System. PLoS Pathog 5(9): e1000519. doi:10.1371/journal.ppat.1000519
Editor: Glenn F. Rall, The Fox Chase Cancer Center, United States of America
Published September 25, 2009
Copyright: ? 2009 Brahic, Roussarie. This is an open-access article distributed under the terms of the
Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any
medium, provided the original author and source are credited.
Funding: We acknowledge the financial support of NIH (grant 5R01AI65972-2) and NMSS (grant RG3924). The
funders had no role in study design, data collection and analysis, decision to publish, or preparation of the
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
PLoS Pathogens | www.plospathogens.org1September 2009 | Volume 5 | Issue 9 | e1000519
cytes. Remarkably, axonal swelling is
observed only in axon segments myelinat-
ed by mutant oligodendrocytes .
These elegant experiments show that a
myelin protein, PLP, has a direct role in
maintaining the integrity of the axon.
In the CNS, 29,39 cyclic nucleotide 39-
phosphodiesterase (CNPase) is found only in
oligodendrocyte cell bodies and in myelin
[12–14]. Cnp12/2 mice have normal
myelin. However, with age, they develop a
neurological deficit due to abnormal distri-
bution of ion channels in the nodes of
Ranvier . The CNPase mutation nicely
segregates two functions of the oligodendro-
cyte: the production of myelin, which is
unaffected, and the support of the axon,
which requires the presence of the enzyme.
In contrast to these mutant pheno-
types, and others which we did not
discuss, there is no axonal degeneration
in the shiverer mouse even though the
amount of myelin is severely restricted,
indicating that myelin supports the axon
through specific signaling and not just by
the physical presence of an electric
Figure 1. Diagrammatic view of CNS myelin. Myelin is an extension of the plasma membrane of the oligodendrocyte. Compact myelin is devoid
of cytoplasm. Cytoplasmic channels, which are continuous with the oligodendrocyte cytoplasm, form the so-called inner and outer myelin loops as
well as the longitudinal incisures.
PLoS Pathogens | www.plospathogens.org2 September 2009 | Volume 5 | Issue 9 | e1000519
Lastly, axon–myelin interactions may
include cytoplasmic exchanges between
the two compartments. The giant axons
of invertebrates, such as squids and
crayfish, are surrounded by uncompacted
multilamellar glial sheaths whose main
function appears to be axonal support
more than electric insulation. The ex-
change of macromolecules between the
axon and this periaxonal glia is well
documented . It may take place
through cytoplasmic channels that connect
both compartments as well as by an
exchange of vesicles budding from one
compartment and fusing with the other
one [17,18]. Therefore, axonal support
appears to be an ancestral function of glial
cells that predates electric insulation.
Cytosol exchanges between axon and
myelin may still take place in vertebrates.
In peripheral nerves, myelin forms com-
structures, called axon-Schwann cell net-
works, that invade the axon. They become
prominent following distal axon injury and
appear to engulf axon material such as
neurofilaments, microtubules, and mito-
chondria. The reader is referred to the
article by Spencer and Thomas  and
to the electron micrographs published by
Gatzinsky et al.  for more information
on this subject. Axon-Schwann cell net-
works could be important for clearing
retrogradely transported organelles target-
ed for degradation, thereby relieving the
neuron cell body from the burden of
recycling the large volume of cytoplasm
present in long axons and preventing toxic
products introduced into peripheral axons
from reaching the cell body [21–24].
by electron microscopy in CNS myelin
[19,22,25]. They may function for axon
clearance, as indicated by several observa-
tions. For example, Lucifer yellow and
horseradishperoxidase injected intothe eye
of mice can be found in the myelin of the
optic nerve [26,27]. In a transgenic mouse
model of Huntington disease, huntingtin
aggregates formed in neurons are found in
the myelin surrounding degenerating axons
. Multiple system atrophy is a rare
human disease characterized by the pres-
ence of alpha-synuclein inclusions in oligo-
dendrocytes . Since alpha-synuclein is
not normally expressed in oligodendro-
cytes, and since alpha-synuclein mRNA is
alpha-synuclein inclusions, the protein is
most likely imported from the axons into
the oligodendrocyte .
In summary, the role of myelin is much
more complex than that of a passive
electric insulator. Through signaling path-
ways that are still largely unknown, myelin
is one of the factors that determines the
cytoarchitecture of the axon and provides
axons with support that, if abolished, leads
to axonal degeneration. The myelin of
Schwann cells and possibly that of oligo-
dendrocytes may also be important in
Figure 2. Hypothetical mechanisms for the traffic of Theiler’s virus from the axon into the surrounding myelin, in the absence of
axonal degeneration. Viral particles are shown in blue, replication complexes in red. Pathway 1: Viral particles (blue) are engulfed in double-
membrane autophagosomes. Following fusion of the autophagosome with a lysosome and digestion of its inner membrane, the particles, which are
resistant to low pH and to proteases, are in a single-membrane vesicle that fuses with the axolemma, thereby releasing the virus in the periaxonal
space. Entry in the myelin requires the presence of a viral receptor. Pathway 2: The outer membrane of the double-membrane vesicle fuses with the
axolemma. The single-membrane vesicle that is released from the axon fuses with the membrane of the myelin inner loop and delivers viral particles
into the myelin. This is an unlikely pathway since the viral RNA cannot be released from the virus particle without interaction with a receptor. Pathway
3: A pathway similar to pathway 2, but in this case replication complexes (red), instead of viral particles, are transferred from the axon into the myelin
where replication can resume. Pathway 4: Engulfment of replication complexes may take place in the axon, where autophagy is known to be
prominent. Pathway 5: Viral particles or replication complexes are transferred from the axon into the myelin by a hypothetical mechanism analogous
to the axon clearing mechanism described in peripheral nerves . A double membrane (axolemma plus myelin) engulfs axonal material, including
viral products. Two fusion events, (axolemma/axolemma) and (myelin/myelin), result in the intoduction of a double-membrane vesicle containing
viral material into the myelin inner loop.
PLoS Pathogens | www.plospathogens.org3September 2009 | Volume 5 | Issue 9 | e1000519
clearing the axon of unwanted organelles
and insoluble protein aggregates.
Theiler’s Virus Traffic within the
We will outline a few hypothetical
mechanisms by which Theiler’s virus
could traffic from intact axons into myelin
(see Figure 2 and its legend). They are
based on the fact that viruses, in particular
those with limited genetic information,
tend to hijack cellular functions for their
replication and spread.
First, it has been proposed that picor-
naviruses could exit through an intact
plasma membrane by a mechanism de-
rived from autophagy. Cellular autophagy
consists of the engulfment of cytoplasmic
material in a double-membrane vesicle
that fuses with endosomes and then
lysosomes, causing the inner membrane
and its contents to be degraded . In an
infected cell microbes may be engulfed in
the double-membrane vesicle. Therefore,
autophagy can be considered part of the
innate immune response. However, picor-
naviruses are highly resistant to proteases
and low pH and may resist degradation. If
the single-membrane vesicle were to fuse
with the plasma membrane, it would
deliver viral particles to the outside in the
absence of cell lysis  (Figure 2, path-
way 1). Interestingly, autophagy is very
active in neurons, including in axons. In a
variation on this theme, double-mem-
brane vesicles could engulf viral RNA
replication/translation complexes. They
could then fuse with the plasma mem-
brane instead of with a lysosome, releasing
a single-membrane vesicle into the extra-
cellular milieu. Such a vesicle could fuse
with the plasma membrane of a neighbor-
ing cell and deliver replication complexes
directly into the cytoplasm where they
could resume their activity (K. Kirke-
gaard, personal communication) (Figure 2,
pathways 3 and 4).
Second, we discussed above how CNS
myelin may play a role in clearing the
axon of unwanted materials, in particular
organelles targeted for degradation. We
summarized the evidence suggesting that
such clearance involves the transport of
axonal cytosol into the internal and
paranodal loops of myelin. Theiler’s virus
could have found ways to introduce itself
into such a pathway, thereby gaining
access to the cytoplasmic channels of
myelin. In such a scenario, the viral
material transferred would be replica-
tion/translation complexes, not virions.
Indeed, virions introduced directly into
the cytoplasm are not infectious because
decapsidation requires interaction with the
viral receptor. Viral replication/transla-
tion complexes, on the other hand, could
resume their activity upon entry into
myelin (Figure 2, pathway 5).
Microorganisms, and viruses in partic-
ular, are masters at using the peculiarities
of the organ which they infect. For this
reason, they can often be used as probes to
uncover new, unsuspected, physiological
mechanisms. Therefore, pathogenesis is as
much a study of the normal functioning of
the organ as it is a study of the pathogen.
Intercellular communication, which is
essential for all multicellular organisms, is
paramount to the functioning of the
nervous system. Communication between
neurons at the level of synapses has been
known as the basis of CNS functioning for
a long time. Communication between glial
cells and between glial cells and neurons
has been appreciated only more recently.
Dissecting, at the molecular level, the
mechanism by which Theiler’s virus cross-
es from the axon into the myelin, a step
required for its persistence in CNS, will
help fill a gap in this important chapter of
We thank Ben Barres, Julia Edgar, Eric
Freundt, and Karla Kirkegaard for critical
reading of the manuscript and many helpful
suggestions. M.B. wishes to thank Karla Kirke-
gaard for her generous support at Stanford
1. Salzer JL (2003) Polarized domains of myelinated
axons. Neuron 40: 297–318.
2. Brahic M (2002) Theiler’s virus infection of the
mouse, or: of the importance of studying animal
models. Virology 301: 1–5.
3. Brahic M, Bureau JF, Michiels T (2005) The
genetics of the persistent infection and demyelin-
ating disease caused by Theiler’s virus. Annu Rev
Microbiol 59: 279–298.
4. Bihl F, Pena-Rossi C, Guenet JL, Brahic M,
Bureau JF (1997) The shiverer mutation affects
the persistence of Theiler’s virus in the central
nervous system. J Virol 71: 5025–5030.
5. Rosenbluth J (1980) Central myelin in the mouse
mutant shiverer. J Comp Neurol 194: 639–648.
6. Fanarraga ML, Griffiths IR, McCulloch MC,
Barrie JA, Kennedy PG, et al. (1992) Rumpsha-
ker: an X-linked mutation causing hypomyelina-
tion: developmental differences in myelination
and glial cells between the optic nerve and spinal
cord. Glia 5: 161–170.
7. Roussarie JP, Ruffie C, Brahic M (2007) The role
of myelin in Theiler’s virus persistence in the
central nervous system. PLoS Pathog 3: e23.
8. Roussarie JP, Ruffie C, Edgar JM, Griffiths I,
Brahic M (2007) Axon myelin transfer of a non-
enveloped virus. PLoS ONE 2: e1331.
9. Klugmann M, Schwab MH, Puhlhofer A,
Schneider A, Zimmermann F, et al. (1997)
Assembly of CNS myelin in the absence of
proteolipid protein. Neuron 18: 59–70.
10. Griffiths I, Klugmann M, Anderson T, Yool D,
Thomson C, et al. (1998) Axonal swellings and
degeneration in mice lacking the major proteo-
lipid of myelin. Science 280: 1610–1613.
11. Edgar JM, McLaughlin M, Yool D, Zhang SC,
Fowler JH, et al. (2004) Oligodendroglial modu-
lation of fast axonal transport in a mouse model of
hereditary spastic paraplegia. J Cell Biol 166:
12. Braun PE, Sandillon F, Edwards A, Matthieu JM,
Privat A (1988) Immunocytochemical localization
by electron microscopy of 2939-cyclic nucleotide
39-phosphodiesterase in developing oligodendro-
cytes of normal and mutant brain. J Neurosci 8:
13. Nishizawa Y, Kurihara T, Masuda T,
Takahashi Y (1985) Immunohistochemical local-
ization of 29,39-cyclic nucleotide 39-phosphodies-
terase in adult bovine cerebrum and cerebellum.
Neurochem Res 10: 1107–1118.
14. Trapp BD, Bernier L, Andrews SB, Colman DR
(1988) Cellular and subcellular distribution of
29,39-cyclic nucleotide 39-phosphodiesterase and
its mRNA in the rat central nervous system.
J Neurochem 51: 859–868.
15. Lappe-Siefke C, Goebbels S, Gravel M,
Nicksch E, Lee J, et al. (2003) Disruption of
Cnp1 uncouples oligodendroglial functions in
axonal support and myelination. Nat Genet 33:
16. Grossfeld RM (1991) Axon-glia exchange of
macromolecules. Ann N Y Acad Sci 633:
17. Peracchia C (1981) Direct communication be-
tween axons and sheath glial cells in crayfish.
Nature 290: 597–598.
18. Buchheit TE, Tytell M (1992) Transfer of
molecules from glia to axon in the squid may be
mediated by glial vesicles. J Neurobiol 23:
19. Spencer PS, Thomas PK (1974) Ultrastructural
studies of the dying-back process. II. The
sequestration and removal by Schwann cells and
oligodendrocytes of organelles from normal and
diseases axons. J Neurocytol 3: 763–783.
20. Gatzinsky KP, Holtmann B, Daraie B,
Berthold CH, Sendtner M (2003) Early onset of
degenerative changes at nodes of Ranvier in
alpha-motor axons of Cntf null (2/2) mutant
mice. Glia 42: 340–349.
21. Gatzinsky KP, Persson GH, Berthold CH (1997)
rat lumbosacral alpha-motor axons by paranodal
axon-Schwann cell networks. Glia 20: 115–126.
22. Berthold CH, Fabricius C, Rydmark M,
Andersen B (1993) Axoplasmic organelles at
nodes of Ranvier. I. Occurrence and distribution
in large myelinated spinal root axons of the adult
cat. J Neurocytol 22: 925–940.
23. Fabricius C, Berthold CH, Gatzinsky KP (1996)
Horseradish peroxidase in axons of the dorsal
funiculi of the cat: distribution after an injection
of the enzyme into the dorsal column nuclei. Cell
Tissue Res 283: 67–73.
24. Gatzinsky KP (1996) Node-paranode regions as
local degradative centres in alpha-motor axons.
Microsc Res Tech 34: 492–506.
25. Novotny GE (1984) Formation of cytoplasm-
containing vesicles from double-walled coated
invaginations containing oligodendrocytic cyto-
plasm at the axon-myelin sheath interface in adult
PLoS Pathogens | www.plospathogens.org4 September 2009 | Volume 5 | Issue 9 | e1000519
mammalian central nervous system. Acta Anat
(Basel) 119: 106–112.
26. Gaton DD, Solomon A, Skutelsky E, Wolman M
(1992) Does a cytoplasmic bridge connect the
CNS axon with the inner loop of myelin? Cell
Mol Biol (Noisy-le-grand) 38: 597–599.
27. Duncan A, Ibrahim M, Berry M, Butt AM (1996)
Transfer of horseradish peroxidase from oligo-
dendrocyte to axon in the myelinating neonatal
rat optic nerve: artefact or transcellular exchange?
Glia 17: 349–355.
28. Yu ZX, Li SH, Evans J, Pillarisetti A, Li H, et al.
(2003) Mutant huntingtin causes context-depen-
dent neurodegeneration in mice with Hunting-
ton’s disease. J Neurosci 23: 2193–2202.
29. Dickson DW, Lin W, Liu WK, Yen SH (1999)
Multiple system atrophy: a sporadic synucleino-
pathy. Brain Pathol 9: 721–732.
30. Miller DW, Johnson JM, Solano SM,
Hollingsworth ZR, Standaert DG, et al. (2005)
Absence of alpha-synuclein mRNA expression in
normal and multiple system atrophy oligoden-
droglia. J Neural Transm 112: 1613–1624.
31. Klionsky DJ (2007) Autophagy: from phenome-
nology to molecular understanding in less than a
decade. Nat Rev Mol Cell Biol 8: 931–937.
32. Jackson WT, Giddings TH, Jr, Taylor MP,
Mulinyawe S, Rabinovitch M, et al. (2005)
Subversion of cellular autophagosomal machinery
by RNA viruses. PLoS Biol 3: e156. doi:10.1371/
PLoS Pathogens | www.plospathogens.org5 September 2009 | Volume 5 | Issue 9 | e1000519