Journal of Neurocytology 34, 343–351 (2005)
Interaction of olfactory ensheathing cells
with astrocytes may be the key to repair of tract
injuries in the spinal cord: The ‘pathway hypothesis’
YING LI, DAQING LI and GEOFFREY RAISMAN∗
Spinal Repair Unit, Institute of Neurology, UCL, Queen Square, London, WC1N 3BG, UK
Received 9 August 2005; revised 3 October 2005; accepted 5 October 2005
Transplantation of cultured adult olfactory ensheathing cells has been shown to induce anatomical and functional repair of
lesions of the adult rat spinal cord and spinal roots. Histological analysis of olfactory ensheathing cells, both in their normal
location in the olfactory nerves and also after transplantation into spinal cord lesions, shows that they provide channels for
the growth of regenerating nerve fibres. These channels have an outer, basal lamina-lined surface apposed by fibroblasts, and
an inner, naked surface in contact with the nerve fibres. A crucial property of olfactory ensheathing cells, in which they differ
from Schwann cells, is their superior ability to interact with astrocytes. When confronted with olfactory ensheathing cells
the superficial astrocytic processes, which form the glial scar after lesions, change their configuration so that their outer pial
surfaces are reflected in continuity with the outer surfaces of the olfactory ensheathing cells. The effect is to open a door into the
central nervous system. We propose that this formation of a bridging pathway may be the crucial event by which transplanted
olfactory ensheathing cells allow the innate growth capacity of severed adult axons to be translated into regeneration across a
lesion so that functionally valuable connections can be established.
(OECs) cultured from the adult primary olfactory sys-
tem of rat, dog and man (Barnett et al., 2000; Devon
& Doucette, 1992; F´ eron et al., 1998; Kato et al., 2000;
Ram´ on-Cueto & Nieto-Sampedro, 1992; Smith et al.,
2002) and transplanted into lesions in the adult spinal
cord or spinal roots promote regeneration of severed
adult central axons across the lesion, remyelination of
axons, and resumption of functions (Barnett et al., 2000;
Boyd et al., 2005; Imaizumi et al., 1998; Li et al., 1998,
2004; Lu et al., 2001; Ram´ on-Cueto et al., 2000; Ram´ on-
Cueto & Nieto-Sampedro, 1994). Both OECs and spinal
regeneration have been the subject of recent excellent
reports and reviews (Barnett & Chang, 2004; Franklin
& Barnett, 1997; Geller & Fawcett, 2002; Silver & Miller,
2004), and it is not the purpose of this article to repeat
them. What we would like to do here is to give a per-
sonal view of what is the problem in spinal cord injury,
and how we believe the work on OECs will not only
∗To whom correspondence should be addressed.
lead to a repair of injuries in man, but can also shed
some light on the enigma of the failure of regeneration
we put forward the hypothesis that the primary prob-
lem in axonal injuries lies not with the nerve fibres, but
with the loss of the astrocytic pathway, and that OECs
produce their beneficial effects by reconstructing that
pathway. The pathway hypothesis is not more than a
hypothesis. But hypotheses are after all the stimulators
of advance in research.
Spinal cord injury
ing from the brain, and the long sensory tracts ascend-
localised injuries is a result of severing these long de-
scending and ascending tracts. Therefore, the repair of
these injuries requires that the tracts are reconstructed,
if not accurately, at least to the extent of establishing a
C ?2006 Springer Science + Business Media, LLC
344 LI, LI and RAISMAN
new circuitry capable of providing access to the unique
information which was cut off by the original injury.
Thus, the approach to the treatment of spinal injury is
one of re-connection.
The establishment of neural connections consists of
two essential elements—the ability of nerve fibres to
grow, and the provision of a pathway along which
they can grow to reach appropriate destinations. So far,
most studies have concentrated on how to promote the
growth of severed nerve fibres. However, after CNS in-
juries it is known that (a) severed nerve fibres not only
the site of the injury (Li & Raisman, 1995), and (b) den-
ervated postsynaptic sites become spontaneously rein-
nervated by axon terminals in their immediate vicinity
(Raisman, 1969). Taken together, these observations in-
dicate that neurons in the injured CNS retain both the
power of axon growth and also the ability to accept
new connections. What is lacking is the ability of the
cut axon sprouts to elongate to their distant destina-
tions and re-establish their original connections. This
focuses attention on what has happened to the path-
way at the lesion site to obstruct their passage.
up of glial cells. The effect of a lesion on the pathway
glial cells of the CNS is devastating. The normal path-
ways in white matter tracts consist of an array of at
least four different types of glial cells arranged with al-
by railroad tracks made up of a dense parallel array of
very fine elongated astrocytic processes (Fig. 1).
When injury occurs, the various glial cells behave
differently. The astrocytes respond rapidly to injury by
closing off the injured site with a basal lamina-lined
outer layer continuous with the glia-pial surface of the
brain or spinal cord (Fig. 1D). This has the important
Fig. 1. (A) The vertical array of astrocytic processes forming the aligned parallel glial skeleton of the corticospinal tract (adult
rat). (B) Schematic representation (not to scale) of the astrocytes (grey) which form the parallel array of processes along which
the axons (black) run, and the basal lamina (dotted lines on outer surface) formed where the end feet of the astrocytes meet
either blood vessels or the pial surface. (C) A lesion which causes degeneration of the distal segments of the axons (large round
dots). (D) The astrocytic ‘scarring’ response results in the lesion site becoming completely sealed off by basal lamina-lined
astrocytic processes. (E) The requirement for axon regeneration is a re-establishment of the longitudinally aligned astrocytic
processes. Scale bar (in A) about 50μm.
effect of preserving the blood-brain barrier, protecting
the CNS from infiltration from outside, and maintain-
ing the specialised CNS ionic environment needed for
nerve function. But at the same time, this astrocytic re-
pathways along which nerve fibres can travel. Oligo-
dendrocytes die (Li et al., 1999), but may be replaced
from precursors (Levine et al., 2001), and microglia are
activated, round up and divide (Liu et al., 2003; Wu &
Ling, 1998). Thus, the normal pathway architecture is
totally lost. In view of the evident signs that the nerve
fibres are attempting to regenerate, a primary objective
in attempting to obtain repair is to reconstruct the glial
pathways in such a way as to re-establish a route along
which the sprouting axons can elongate and reach use-
ful targets (Fig. 1E).
Transplantation of Schwann cells
The idea of repair by transplantation was proposed
by Cajal (1928) on the grounds that severed periph-
eral nerve fibres are able to elongate readily and for
long distances in the Schwann cell environment of pe-
ripheral nerve. Pioneering studies by Aguayo’s group
on peripheral nerve grafts (Vidal-Sanz et al., 1987), and
the Bunge’s on Schwann cell (SC) culture and trans-
plantation (Xu et al., 1997) established that either pieces
of peripheral nerve or cultured Schwann cells trans-
planted into lesions in the optic nerve and spinal cord
do indeed provide pathways for elongation of severed
adult central axons. However, the regenerating axons
remain largely confined to the Schwann cell environ-
ment of the grafts. Re-entry of axons from the grafts to
the host CNS is minimal (Keyvan-Fouladi et al., 2005;
Li & Raisman, 1994). The interface established between
the transplanted SCs and the host astrocytes behaves
rather like the interface between the SC environment
of spinal roots and the astrocytic environment of the
The ‘pathway hypothesis.’345
spinal cord—namely, after lesions central motoneuron
axons are able to exit from the astrocytic environment
of the CNS and enter the SC territory in ventral roots,
but peripheral sensory nerve fibres are unable to cross
when sensory fibres regenerating through the SC ter-
ritory of a severed dorsal root reach the former dorsal
root entry zone they are unable to leave the SC terri-
references see Li et al., 2004. The door for re-entry into
the CNS is closed, and this raises the question of how
to open the door.
Olfactory ensheathing cell as ‘door-openers’
todermal surface of the body. To reach the brain, the ol-
the mesoderm and break through into the neurecto-
they establish remains open throughout adult life and
permits the continuous regeneration of olfactory nerve
fibres from the mucosa through the olfactory nerves
and into the olfactory bulbs. The door-opening ability
of the olfactory nerves was strikingly demonstrated in
a seminal but hardly noticed experiment of Graziadei
et al. (1979), who showed that when the neonatal olfac-
of that side continued to grow through the cranial cav-
ity until they reached the rostral surface of the cerebral
hemisphere which they then entered and formed typi-
cal glomerular structures in the frontal cortex, an area
From a morphological point of view OECs lie at the
deepest levels of the ectoderm. They are asymmetrical
cells (Fig. 3). Their superficial surfaces face the neural
tissue—i.e. the olfactory nerve fibres. Their deep sur-
ina and apposed on the mesodermal side by layers of
very thin circumferentially arranged processes belong-
ensheathment of the olfactory nerves in situ, the cells
become folded over, like the outer wrapper of a cigar,
to enclose a central channel. The basal lamina-lined
mesodermal surface lies on the outside, surrounded
by thin circumferential slivers of ONFs and ONF pro-
cesses which lie embedded in a collagen-rich extracel-
lular matrix, while the naked inner surface of the OECs
nerve fibres which they enwrap in small bundles (Field
et al., 2003).
At the sub-mucosal origin of the olfactory nerves
(the ‘lamina propria’) the complete folding over of the
Fig. 2. (A) to (C) Schematic representation of the develop-
ment of the primary olfactory projection from the olfactory
placode (p, black) in the ectoderm (e, grey) from which the
glial cells and nerve fibres grow inwards through the meso-
derm (arrow in B) to form the olfactory nerves (black in C)
which reach the cerebral vesicle of the CNS (grey). The outer
covering of basal lamina (dashed line) and fibroblasts which
initially form separate coverings of the olfactory placode and
CNS are later reflected off the nerve at its entry point into the
nerve glia. D (modified from Li et al., 2005): A summary of
the ensheathment of the olfactory axons. In the MUCOSA the
gather in bundles in the extracellular spaces between the end
feet of the horizontal basal cells (hbc) which end on the basal
lamina (bl) separating the olfactory epithelium from the lam-
ina propria and deeper tissues. Slender processes of olfactory
ensheathing cells (oec) penetrate the basal lamina and tightly
invest bundles of axons. In the NERVE adjacent OECs with
their enclosed bundles of axons are interlocked end-to-end
to form continuous longitudinal channels (see Fig. 4) that are
closely lined on their outer surface by basal lamina and out-
side that by the olfactory nerve fibroblasts (onf). On reaching
the BULB, the basal lamina and the ONFs are reflected off to
become continuous with the leptomeningeal coverings of the
brain, and the OEC processes interact with the astrocytic pro-
cesses (ast) to ‘open up’ the glomerular space (glo) in which
the olfactory axons synapse upon the primary dendrites of
the mitral cells (mc) whose axons project (arrow) into the ol-
of the CNS.
outer margins of the wrapper results in ensheathment
of nerve fibres by single OECs. Looked at in the lon-
gitudinal plane, the OECs have the shape of ragged-
ended cylinders. Placed end to end these cylinders
provide continuous channels through which the ol-
factory nerve fibres pass (Fig. 4). Progressing further
from the lamina propria into the main parts of the ol-
factory nerves, two or more OECs combine to form
larger cylinders enclosed in a common basal lamina
346 LI, LI and RAISMAN
Fig. 3. Schematic representation of the folding over of the OECs to form channels. Morphologically, the OECs have an epithelial
origin. A single OEC is represented in A as an asymmetrical cell with a basal lamina-lined inner surface (dashed line to the left
of the cell) facing the mesenchymal fibroblasts and an outer, naked surface (to the right) which is apposed to the olfactory nerve
fibres. Progressive curling over of the cell (B to E) shows the formation of an enclosed channel ensheathing the olfactory nerve
axons, which are partially divided into interconnecting territories by thin cytoplasmic sheets arising from the inner, naked
surface of the OECs. Overall cell diameter around 10μm.
Fig. 4. Schematic representation of the formation of long
channel by the end-to-end apposition of individual OECs.
Overall diameter around 10μm.
with thin fibroblastic processes and collagen outside.
Their inner, process-bearing surfaces face into large
channels in which are enclosed several thousand tiny,
unmyelinated olfactory axons (Fig. 5; Field et al., 2003).
Having passed through the cribriform plate of the
ethmoid bone, the olfactory nerves enter the cranial
cavity and make contact with the olfactory bulb. At
this level the fibroblastic encirclement and basal lam-
ina become reflected off the nerve fascicles to form the
pial surface of the olfactory bulb. The peeling off of the
fibroblastic encirclement allows the nerve fascicles to
coalesce and intermingle. This constitutes the olfactory
nerve layer of the bulb and provides for the different
receptor modalities of the olfactory receptors to sort
out and navigate to their appropriate glomeruli (Buck,
1996). The final stage of the primary olfactory path-
way (Fig. 6) occurs in the glomerular layer of the olfac-
tory bulb, where the OEC cylinders open up and their
processes interact with the astrocytic processes of the
CNS (Marin-Padilla & Amieva B, 1989; Raisman, 1985;
Valverde et al., 1992; Valverde & Lopez-Mascaraque,
1991). At this point the superficial astrocytes of the
Fig. 5. Electron micrographic montage showing a cross sec-
nerve (modified from (Field et al., 2003). The processes from
two nucleated OECs cooperate to form a continuous encir-
clement of around 2,500 olfactory axons. The whole fascicle
is tightly invested with a basal lamina and lies in an extracel-
lular space rich in collagen fibrils separating the fascicle from
adjacent fascicles (asterisks). ONF, an olfactory nerve fibrob-
last and narrow electron dense ONF processes (arrowheads)
encircle the array. Scale bar, 2μm.
olfactory glomeruli are not lined by basal lamina or
apposed by the leptomeningeal fibroblasts of the pia,
but are open to the incoming olfactory nerve axons.
This is the crucial region of the ‘open-door,’ which we
CNS lesions (see below).
During the cycle of degeneration and regeneration
which occurs after lesions of the adult olfactory nerves,
and are probably strengthened physically by the sur-
rounding basal lamina and collagen fibrils produced
by interaction with the olfactory nerve fibroblasts (Li
et al., 2005; Williams et al., 2004). As in normal devel-
opment, once the advancing tips of the regenerating
The ‘pathway hypothesis.’347
Fig. 6. (A to D) Schematic representation (not to scale) of the ‘door-opening’ interaction between the OECs and the astrocytic
coverings (ast, opening at∗in B) of the olfactory glomeruli (glom) which allows the ingrowth (arrow in C) of the olfactory
axons (ax) to make contact with the mitral cell dendrites (mcd) of the CNS. Overall diameter of a glomerulus around 100μm.
adult olfactory nerves reach the olfactory bulbs, the
OECs open the astrocytic door and provide a chan-
nel for the regenerating axons to enter the glomeruli
and make synaptic contacts with the mitral, tufted and
1978; Li et al., 2005; Schwob et al., 1999).
OECs in transplants
ment of the CNS that is lacking in the SCs of cut dorsal
roots or in SCs transplanted into lesions of the CNS.
And it is this door-opening ability that makes OECs an
attractive subject for transplantation into spinal cord
lesions. Tissue culture protocols have been established
to obtain OECs from adult olfactory tissue (either from
samples from the mucosa or from the outer layers of
the olfactory bulbs) in rat (Devon & Doucette, 1992;
Ramon-Cueto & Nieto-Sampedro, 1992), dog (Smith
et al., 2002) and man (Barnett et al., 2000; Kato et al.,
2000; Feron et al., 1998). Although it has still not for-
mally been established that the OECs of the primary
olfactory pathway in situ are capable of being gener-
ated in the adult either from mucosal or other stem
cells, it is certainly the case that they are generated by
cell division in adult tissue samples in culture (Jani &
There are a number of reports from our laboratory
and others that cultured OECs transplanted into either
spinal cord lesions (Barnett et al., 2000; Imaizumi et al.,
1998; Li et al., 1998, 2004; Lu et al., 2001; Ramon-Cueto
et al., 2000; Ramon-Cueto & Nieto-Sampedro, 1994)
or dorsal root lesions (see below) provide a bridge
which allows severed nerve fibres to regenerate back
into the CNS. We have transplanted OECs into corti-
cospinal tract lesions (Keyvan-Fouladi et al., 2003) at a
time when the astrocytes at the lesion site have already
formed a closed off scar (Keyvan-Fouladi et al., 2003;
Li & Raisman, 1995). We believe that the transplanted
OECs act, as they do in the primary olfactory system,
by interacting with the CNS astrocytes in such a way
as to open up the astrocytic interface at the lesion site
(see below for the similar effect at the dorsal root en-
try zone). This permits the entry of regenerating nerve
fibres into the graft and their elongation through the
graft—as is the case after transplantation of SCs. But,
unlike SCs, after the regenerating fibres have crossed
the graft, the OECs allow their re-entry back into the
CNS to reach target areas through which they are able
to make connections mediating important functional
with transplanted SCs (see also Franklin & Barnett,
1997; Franklin & Blakemore, 1993) has a parallel in
tissue culture. When co-cultured with astrocytes, SCs
occupy separate territories, whereas OECs intermingle
with the astrocytes (Lakatos et al., 2000).
As to how transplanted OECs achieve this repair,
we must look to the tissue rearrangements which they
engender, and in particular the question of how they
can open up the damaged glial pathways so as to re-
most direct demonstration of the interaction of trans-
planted OECs with the host astrocytes is seen in ex-
periments with rat dorsal roots (Fig. 7A; Li et al., 2004;
Ram ’ on-Cueto&Nieto-Sampedro,1994;butseeGomez
et al., 2003; Riddell et al., 2004). If a severed dorsal root
is simply replaced in contact with the dorsal root en-
try zone, there is very little interaction between the
SCs of the root and the adjacent spinal cord astrocytes
(Fig. 7B). The astrocytes smooth over the injury by
forming a ‘closed,’ glia-pial membrane which is con-
tinuous with the adjacent glia-pial membrane of the
spinal cord and which has a basal lamina-lined sur-
face facing outwards towards the replaced root. Only a
small number of very fine astrocytic processes extend
for short distances into the SC territory of the dorsal
root. The SCs make no attempt to enter the astrocytic
able to cross into the spinal cord (see Li et al., 2004, for
In contrast, after transplantation of OECs into the
junction between a severed dorsal root and the spinal
cord there is a massive outgrowth of thick astrocyte
processes which interweave intimately with the OEC
and SC processes forming a series of bridges like the
rungs of a ladder (Fig. 7C). These bridges convey the
regenerating dorsal root axons from the SC territory of
the peripheral nerve into the astrocytic environment of
the spinal cord where they are able to arborise in the
grey matter of the dorsal horn and also run rostrally
and caudally in the white matter of the sensory dorsal
columns (Fig. 7D).
348 LI, LI and RAISMAN
green, and the glial fibrillary protein (GFAP)-positive astrocytes of the spinal cord tissue in red. (B) ‘Closed’glia-pial membrane
2 weeks after direct re-apposition of a severed rootlet with fibrin glue, but without OECs. (C) The reconstituted interface
between central and peripheral tissue 2 weeks after transplantation of OECs into the zone of re-apposition of the dorsal root
to the spinal cord consists of a periodic, ladder-like structure formed by the intimate interleaving of outgrowing astrocytic
processes (GFAP, red) with processes arising from the peripheral nerve tissue (laminin, green; overlay shows as yellow). (D)
Reconstructed dorsal root entry zone after transplantation of OECs (labelled with adenovirally transfected green fluorescent
protein). The regenerating dorsal root axons (labelled with biotin dextran, red) pass through the region of the transplanted
OECs and into the dorsal columns. Blue, counterstain, diamidinophenylindole (DAPI). Horizontal sections. Scale bars, 100μm.
Modified from Li et al., 2004.
We propose that the reparative ability of transplanted
OECs is due to their interaction with astrocytes. Both
OECs and astrocytes have a similar asymmetrical ar-
and an outer basal lamina-lined surface facing either
the fibroblasts of the pial surface and/or blood vessels.
The door-opening action that transplanted OECs exert
on the astrocytes at a lesion site involves a rearrange-
ment of their processes so as to match up their inner
surfaces to form an open channel for the axons to pass
through, while matching their outer surfaces so that
the basal lamina is reflected over the outer aspect of the
channels. The formation of an OEC bridge between the
host astrocytic pathways on either side of the lesion re-
stores the continuity of the longitudinal railroad tracks
The ‘pathway hypothesis.’349
across the lesion and reach tissue targets that can me-
diate useful functional return. We refer to this as the
‘pathway hypothesis’of neural repair.
The molecular signalling mechanisms underlying
the opening up of the astrocytic scar, and the subse-
therapeutic benefits. In fact, most molecular interven-
tions have so far been focussed on the axons which,
according to the pathway hypothesis, would be a sec-
ondary target. The regeneration of axons through OEC
transplants indicates that there is a sufficient endoge-
nous supply of axonal growth factors present in the
host/graft system to support a baseline of repair lead-
ing to a significant level of functional recovery. Addi-
tional exogenous factors may enhance growth (Blesch
et al., 2002; Cao et al., 2005; Ruitenberg et al., 2003).
The view that the failure of axon regeneration is due
to inhibitory molecules produced either by oligoden-
drocytes and/or by astrocytes largely originates from
observations on growth cones in tissue culture. But in
the highly organised environment of the CNS it is nec-
essary to know where these molecules are expressed
(Raisman, 2004). It seems unlikely that inhibitory
molecules associated with degenerating oligodendro-
cytes and myelin (Filbin, 2003; Schnell & Schwab, 1990)
will be present within the enclosed environment of
With regard to the inhibitory molecules demonstrated
Snow et al., 1990), it seems most likely they are associ-
ated with the outer, basal lamina-lined surfaces. The
union of the OEC and astrocytic channels would mean
outwards, with the inner surface of the channels being
devoid of inhibitory molecules and therefore permis-
sive to axonal elongation.
A number of neurosurgical teams are now undertak-
ing treatments for repair of human spinal cord injuries
by transplantation of OECs. Given the understandable
public interest and expectations aroused by this, it is
important to entertain some caution (Anderson et al.,
2005). It would add confidence if the selection of the
sources of tissue (e.g., embryonic vs. adult) and the de-
the tissue (as in the published animal experiments) or
simply to transfer biopsied material directly into the
spinal cord were supported by published scientific lab-
oratory findings. Given the natural history of recovery
in spinal cord injury (Frankel et al., 1969; Katoh & el
dependently verified neurological assessment for suf-
ficiently long periods both before and after treatment
would be necessary to ensure that any observed im-
provements are due to the transplanted cells, and to
control for the effects of extraneous medical, physical
and psychological interventions.
DIETZ, V., DOBKIN, B., FAWCETT, J., FEHLINGS,
M., FISCHER,I., GROSSMAN,
HAGG, T., HALL, E. D., HOULE, J., KLEITMAN,
TETZLAFF, W., TUSZYNSKI, M. H., WAXMAN,
S. G., WHITTEMORE, S., WOLPAW, J., YOUNG,
W. & ZHENG, B. (2005) Recommended guidelines for
studies of human subjects with spinal cord injury. Spinal
Cord 43, 453–458.
BARBER, P. C. & RAISMAN, G. (1978) Replacement of
receptor neurones after section of the vomeronasal
nerves in the adult mouse. Brain Research 147,
BARNETT, S. C., ALEXANDER, C. L., IWASHITA, Y.,
GILSON, J. M., CROWTHER, J., CLARK, L., DUNN,
L. T., PAPANASTASSIOU, V., KENNEDY, P. G. &
FRANKLIN, R. J. (2000) Identification of a human
mediated remyelination of demyelinated CNS ax-
ons. Brain 123, 1581–1588.
BARNETT, S. C. & CHANG, L. (2004) Olfactory en-
sheathing cells and CNS repair: going solo or in
need of a friend? Trends in Neurosciences 27, 54–60.
BLESCH, A., LU, P. & TUSZYNSKI, M. H. (2002) Neu-
rotrophic factors, gene therapy, and neural stem
cells for spinal cord repair. Brain Research Bulletin 57,
BOYD, J. G., DOUCETTE, R. & KAWAJA, M. D. (2005)
Defining the role of olfactory ensheathing cells in
facilitating axon remyelination following damage
to the spinal cord. FASEB Journal. 19, 694–703.
BUCK, L. B. (1996) Information coding in the verte-
brate olfactory system. Annual Review of Neuroscience
CAJAL, S. R. Y. (1928) Degeneration and Regeneration of the
Nervous System, pp. 1–769. Oxford: Oxford Univer-
CAO, Q., XU, X. M., DEVRIES, W. H., ENZMANN, G.
U., PING, P., TSOULFAS, P., WOOD, P. M., BUNGE,
M. B. & WHITTEMORE, S. R. (2005) Functional re-
covery in traumatic spinal cord injury after trans-
plantation of multineurotrophin-expressing glial-
restricted precursor cells. Journal of Neuroscience 25,
DE CARLOS,J. A.,L´OPEZ-MASCARAQUE,
VALVERDE, F. (1995)Thetelencephalicvesiclesare
innervated by olfactory placode-derived cells: A
possible mechanism to induce neocortical devel-
opment. Neuroscience 68, 1167–1178.
DEVON, R. &DOUCETTE, R. (1992)Olfactoryensheath-
ing cells myelinate dorsal root ganglion neurites.
Brain Research 589, 175–179.
350 LI, LI and RAISMAN
FERON, F., PERRY, C., MCGRATH, J. J. & MACKAY-
SIM, A. (1998) New techniques for biopsy and
culture of human olfactory epithelial neurons.
Archives of Otolaryngology-Head & Neck Surgery. 124,
FIELD, P. M., LI, Y. & RAISMAN, G. (2003) Ensheath-
ment of the olfactory nerves in the adult rat. Journal
of Neurocytology 32, 317–324.
FILBIN, M. T. (2003)Myelin-associatedinhibitorsofax-
onal regeneration in the adult mammalian CNS.
Nature, Reviews Neuroscience 4, 703–713.
FRANKEL, H. L., HANCOCK, D. O., HYSLOP, G.,
MELZAK, J., MICHAELIS, L. S., UNGAR, G. H.,
VERNON, J. D. & WALSH, J. J. (1969) The value
of postural reduction in the initial management of
closed injuries of the spine with paraplegia and
tetraplegia. I. Paraplegia 7, 179–192.
FRANKLIN, R. J. & BARNETT, S. C. (1997) Do olfactory
glia have advantages over Schwann cells for CNS
repair? Journal of Neuroscience Research 50, 665–672.
FRANKLIN, R. J. M. & BLAKEMORE, W. F. (1993) Re-
environments: A viewpoint. International Journal of
Developmental Neuroscience 11, 641–649.
GELLER, H. M. & FAWCETT, J. W. (2002) Building a
bridge: Engineering spinal cord repair. Experimental
Neurology 174, 125–136.
GOMEZ, V. M., AVERILL, S., KING, V., YANG, Q.,
PEREZ, E. D., CHACON, S. C., WARD, R., NIETO-
SAMPEDRO, M., PRIESTLEY, J. & TAYLOR, J.
(2003) Transplantation of olfactory ensheathing
cells fails to promote significant axonal regener-
ation from dorsal roots into the rat cervical cord.
Journal of Neurocytology 32, 53–70.
IMAIZUMI, T., LANGFORD, K. L., WAXMAN, S. G.,
GREER, C. A. & KOCSIS, J. D. (1998) Transplanted
olfactory ensheathing cells remyelinate and en-
hance axonal conduction in the demyelinated dor-
sal columns of the rat spinal cord. Journal of Neuro-
science 18, 6176–6185.
JANI, H. R. & RAISMAN, G. (2004)Ensheathingcellcul-
tures from the olfactory bulb and mucosa. Glia 47,
KATO, T., HONMOU, O., UEDE, T., HASHI, K. &
KOCSIS, J. D. (2000) Transplantation of human ol-
factory ensheathing cells elicits remyelination of
demyelinated rat spinal cord. Glia 30, 209–218.
KATOH, S. &E L MASRY, W. S. (1995)Motorrecoveryof
patients presenting with motor paralysis and sen-
Paraplegia 33, 506–509.
KEYVAN-FOULADI, N., RAISMAN, G. & LI, Y. (2005)
Transplanted Schwann cells are less effective than
olfactory ensheathing cells for delayed repair of
corticospinal tract lesions. Glia 51, 306–311.
KEYVAN-FOULADI, N., RAISMAN, G. & LI, Y. (2003)
Delayed repair of corticospinal tract lesions by
transplantation of olfactory ensheathing cells
in adult rats. Journal of Neuroscience 23, 9428–
LAKATOS, A., FRANKLIN, R. J. M. & BARNETT, S. C.
(2000) Olfactory ensheathing cells and Schwann
cells differ in their in vitro interactions with astro-
cytes. Glia 32, 214–225.
LEVINE, J. M., REYNOLDS, R. & FAWCETT, J. W. (2001)
The oligodendrocyte precursor cell in health and
disease. Trends in Neurosciences 24, 39–47.
LI, Y., CARLSTEDT, T., BERTHOLD, C.-H. & RAISMAN,
G. (2004) Interaction of transplanted olfactory-
vides a bridge for axons to regenerate across the
dorsal root entry zone. Experimental Neurology 188,
LI, Y., FIELD, P. M. & RAISMAN, G. (2005) Olfactory
ensheathing cells and olfactory nerve fibroblasts
maintain continuous open channels for regrowth
of olfactory nerve fibres. Glia 52, 245–251.
LI, Y., FIELD, P. M. & RAISMAN, G. (1999) Death of
oligodendrocytes and microglial phagocytosis of
myelin precede immigration of Schwann cells into
the spinal cord. Journal of Neurocytology 28, 417–427.
LI, Y., FIELD, P. M. &RAISMAN, G.(1998)Regeneration
of adult rat corticospinal axons induced by trans-
planted olfactory ensheathing cells. Journal of Neuro-
science 18, 10514–10524.
LI, Y. & RAISMAN, G. (1994) Schwann cells induce
sprouting in motor and sensory axons in the adult
rat spinal cord. Journal of Neuroscience 14, 4050–4063.
LI, Y. & RAISMAN, G. (1995) Sprouts from cut corti-
cospinal axons persist in the presence of astrocytic
scarring in long-term lesions of the adult rat spinal
cord. Experimental Neurology 134, 102–111.
LIU, P. H., WANG, Y. J. & TSENG, G. F. (2003) Close
axonal injury of rubrospinal neurons induced tran-
sient perineuronal astrocytic and microglial reac-
tion that coincided with their massive degenera-
tion. Experimental Neurology 179, 111–126.
LU, J., F´ERON, F., HO, S. H., MACKAY-SIM, A. &
WAITE, P. M. E. (2001) Transplantation of nasal ol-
factory tissue promotes partial recovery in para-
plegic adult rats. Brain Research 889, 344–357.
MARIN-PADILLA, M. & AMIEVA B, M. R. (1989) Early
neurogenesis of the mouse olfactory nerve: Golgi
and electron microscopic studies. Journal of Compara-
tive Neurology 288, 339–352.
MENDOZA, A. S., BREIPOHL, W. & MIRAGALL, F.
(1982) Cell migration from the chick olfactory pla-
of Embryology and Experimental Morphology. 69, 47–59.
RAISMAN, G. (1985) Specialized neuroglial arrange-
ment may explain the capacity of vomeronasal ax-
ons to reinnervate central neurons. Neuroscience 14,
RAISMAN, G. (1969) Neuronal plasticity in the septal
nuclei of the adult rat. Brain Research 14, 25–48.
RAISMAN, G. (2004)DoesNOmeanGO?Nature,Reviews
Neuroscience 5, 157–161.
BENITO, F. F. & AVILA, J. (2000) Functional recov-
ery of paraplegic rats and motor axon regeneration
in their spinal cords by olfactory ensheathing glia.
Neuron 25, 425–435.
RAM´ON-CUETO, A. & NIETO-SAMPEDRO, M. (1994)
Regeneration into the spinal cord of transected
The ‘pathway hypothesis.’351 Download full-text
dorsal root axons is promoted by ensheathing glia
transplants. Experimental Neurology 127, 232–244.
RAM´ON-CUETO, A. & NIETO-SAMPEDRO, M. (1992)
tochemical properties of pure cultures of ensheath-
ing cells. Neuroscience 47, 213–220.
RIDDELL, J. S., ENRIQUEZ-DENTON, M., TOFT, A.,
FAIRLESS, R. & BARNETT, S. C. (2004) Olfactory
ensheathing cell grafts have minimal influence on
rhizotomy. Glia 47, 150–167.
RUITENBERG, M. J., PLANT, G. W., HAMERS, F.
P., WORTEL, J., BLITS, B., DIJKHUIZEN, P. A.,
GISPEN, W. H., BOER, G. J. & VERHAAGEN, J.
(2003) Ex vivo adenoviral vector-mediated neu-
rotrophin gene transfer to olfactory ensheathing
glia: effects on rubrospinal tract regeneration, le-
sion size, and functional recovery after implanta-
tion in the injured rat spinal cord. Journal of Neuro-
science 23, 7045–7058.
SCHNELL, L. & SCHWAB, M. E. (1990) Axonal regen-
eration in the rat spinal cord produced by an anti-
body against myelin-associated neurite growth in-
hibitors. Nature 343, 269–272.
IWEMA, C. L. & MEZZA, R. C. (1999) Reinnerva-
tion of the rat olfactory bulb after methyl bromide-
induced lesion: timing and extent of reinnervation.
Journal of Comparative Neurology 412, 439–457.
ASHER, R. A., HOLTMAAT, A. J., LEVINE, J. M.,
VERHAAGEN, J. & FAWCETT, J. W. (2003) The
astrocyte/meningeal cell interface is a barrier to
neurite outgrowth which can be overcome by ma-
nipulation of inhibitory molecules or axonal sig-
nalling pathways. Molecular and Cellular Neuroscience
SILVER, J. & MILLER, J. H. (2004)Regenerationbeyond
the glial scar. Nature, Reviews Neuroscience 5, 146–156.
SMITH, P. M., LAKATOS, A., BARNETT, S. C., JEFFERY,
N. D. & FRANKLIN, R. J. M. (2002) Cryopreserved
cells isolated from the adult canine olfactory bulb
S.L., RING, G.,
are capable of extensive remyelination following
transplantation into the adult rat CNS. Experimental
Neurology 176, 402–406.
CAPLAN, A. I. & SILVER, J. (1990) Sulfated
proteoglycans in astroglial barriers inhibit neu-
rite outgrowth in vitro. Experimental Neurology 109,
SUZUKI, M. &RAISMAN, G. (1992)Theglialframework
of central white matter tracts: Segmented rows
solitary astrocytes give rise to a continuous mesh-
work of transverse and longitudinal processes in
the adult rat fimbria. Glia 6, 222–235.
VALVERDE, F. & LOPEZ-MASCARAQUE, L.(1991)
of the hedgehog. Journal of Comparative Neurology 307,
VALVERDE, F., SANTACANA, M. & HEREDIA, M.
(1992) Formation of an olfactory glomerulus: mor-
phological aspects of development and organiza-
tion. Neuroscience 49, 255–276.
VIDAL-SANZ, M., BRAY, G. M., VILLEGAS-PEREZ, M.
P., THANOS, S. & AGUAYO, A. J. (1987) Axonal
colliculus by retinal ganglion cells in the adult rat.
Journal of Neuroscience 7, 2894–2909.
WILLIAMS, S. K., FRANKLIN, R. J. & BARNETT, S.
C. (2004) Response of olfactory ensheathing cells
to the degeneration and regeneration of the pe-
ripheral olfactory system and the involvement of
the neuregulins. Journal of Comparative Neurology 470,
WU, Y. P. & LING, E. A. (1998) Induction of microglial
and astrocytic response in the adult rat lumbar
spinal cord following middle cerebral artery occlu-
sion. Experimental Brain Research 118, 235–242.
XU, X. M., CHEN, A., GU´ENARD, V., KLEITMAN, N. &
BUNGE, M. B. (1997) Bridging Schwann cell trans-
plants promote axonal regeneration from both the
rostral and caudal stumps of transected adult rat
spinal cord. Journal of Neurocytology 26, 1–16.