вход по аккаунту


Central nervous system remyelination Studies in chronically damaged tissue.

код для вставкиСкачать
Central Nervous System Remyelination:
Studies in Chronically Damaged Tissue
Samuel K. Ludwin, MBBCh, FRCP(C)
In this presentation, I will discuss how the oligodendrocyte responds to chronic injury, either from processes that affect the cells themselves or from those
damaging the axon. The mechanics of remyelination
will not be discussed.
Cuprizone demyelination, whch remains a very
good model of easily traceable, consistent demyelination is achieved by poisoning all ohgodendrocytes,
with no damage to other cell types. This latter point is
emphasized because of its relevance to multiple sclerosis (MS). In their elegant previous presentations, both
Lassmann et al [l] and Raine 121 have shown chronic
MS plaques with very heavy ghosis in some cases separating a background of sparse axons. In cuprizone demyelination however, the axons are all left intact and
in place, and the interstitium consists of relatively mild
gliosis. Ic is important to distinguish between the two
situations, as the degree of axonal injury or loss will
have implications for remyelination.
In the Cuprizone model, when the toxin is removed,
almost complete remyelination occurs with thinly remyelinated sheaths around axons. In a similar manner,
in typical MS plaques there are thinly remyelinated
axons at the edges; the point should be made again
that in this location, the interstitium is not all that dissimilar to that seen in the Cuprizone model, with many
axons and very little in the way of ghosis.
In looking at the requirements for remyelination, it
has become clear that the first obvious necessity is the
need for the provision of adequate numbers of oligodendrocytes. Whether these are derived from immature
cells or from surviving or mature oligodendrocytes has
been touched on by some of the other speakers and
will be discussed later. Cells may arise from endogenous cells, or from exogenously grafted cells, which
is what Dr Van Evercooren will discuss. Second, the
oligodendrocyte or its precursor has to be able to migrate and actually reach and extend processes to reach
the axon. Third, oligodendrocytes have to adhere to
axons, and finally there is a very complex oligodendrocyte-axon signaling, which will be discussed later.
It is important to test, in the intact animal, the resil-
ience of the endogenous oligodendrocyte system as
a whole, in addition to determining the regeneration
capacities of individual cells at different maturational
stages. One has to ask not only what cells survive, but
also whether the cells are surviving a direct attack on
the myelin or oligodendrocyte, or whether there is, in
addition, an indirect effect secondary to axonal damage? What cells can proliferate? What functional reserve do these cells have after they have proliferated?
Are oligodendrocytes involved in other functions such
as phagocytosis, which may compromise their capacity
to recover and remyelinate? Finally, a very important
question to be answered, as we have seen from MS, is
whether myelination always follows in situations where
axons and oligodendrocytes are present.
Dealing with the last issue first, in collaboration with
Dr Szuchet we looked at mixed cultures of dorsal root
ganglia cells from the rat to which were added, in the
first instance, rat gllal cells C31. In these cultures good
compact myelin is easily made. When the experiments
were repeated with highly purified mature sheep oligodendrocytes, we found the following things. The cells
adhere very well and preferentially to the axons rather
than to the collagen bed. This selective adherence has
been described before. Subsequently, as has been
shown before, the act of adherence immediately stimulates these cells to start production of myelin proteins,
in this case, specifically myelin basic protein (MBP),
although other myelination proteins may also be produced.
Further experiments show that proliferation of these
cells occurs, both when they are first seeded onto neurons and axons and also when they are established in
culture and futther dorsal root ganglia neurons are
added. Therefore, these cells can accomplish the necessary prerequisites for myelination; i.e., they adhere
preferentially, they turn on their proteins, they proliferate. However, although they wrapped spirally
around the axons with inner and outer tongues in the
characteristic manner, the last step was not complete
and they did not make compact myelin.
There are a small number of phagocytes, macro-
From the Department of Pathology, Queen’s University at Kingston,
Ontario, Canada
Address correspondence to Dr Ludwin, Department of Pathology,
Richardson Iaboratory, Queen’s University,Kingston, Ontario, Canada K7L 3N6.
Copyright 0 1994 by the American Neurological Association
phages, and astrocytes in the cultures, although it is
unclear whether the proportions of these cells were
optimal for myelination. The inability of these particular cells to myelinate contrasted with the rat ohgodendrocytes, which did so very well. For a long time, species restriction was thought to be the problem, and the
study was not completed until recently. The issue of
species specificity was solved by transplanting the
sheep oligodendrocytes into neonatal shiverer mice,
which lack oligodendrocyte MBP. In this in vivo situation, myelination occurs, demonstrating that crossspecies restriction is not the block to myelination. This
study pointed out that the system of myelination or
remyelination is a very delicately controlled one and
that the numerous steps are probably independently
regulated. Therefore, the mere presence of bare axons
and an adequate number of cells does not necessarily
guarantee that myelination will be a success. This is of
tremendous importance for MS where it is unclear
what subtle changes in axonal membrane, in oligodendrocyte membrane, or in environmental factors, may
be present that may be hindering remyelination, even
though all the cellular constituents are in place.
In dealing with regenerative capacity in chronic systems, two situations will be demonstrated. The first is
a model of olgodendrocyte damage, and the other is
one in which the effect of the oligodendrocyte is
through indirect axonal damage.
In chronic Cuprizone toxicity, which severely depletes oligodendrocytes from the brain, two interesting
points emerge in animals kept on the toxin for about
a year. First of all, there is severe depletion of myelin
with only a very subtle increase in the amount of astroghosis between the axons as measured ultrastructurally
or with gllal fibrillary acidc protein (GFAP) staining,
compared with the more acutely demyelinated animal.
Second, there is still no significant axonal loss or disruption of the architecture even after a year of demyelination. On removing the animals from the toxin, in
contrast to short-term demyelination, there is only
sparse remyelination in up to 20% of axons. There is
no difference between those axons that are myelinated,
compared with those that remain unmyelinated.
Proliferation studies have been done that also show
diminished numbers of dividing cells in the chronic
situation, and the amount of remyelination correlates
with the number of oligos that are available for remyelination. In the Cuprizone model where one depletes
mature oligodendrocytes, we have previously shown
that the newly generated oligodendrocytes are derived
from immature ohgodendrocytes. When animals are
repetitively demyelimated and remyelinated, the result
is a markedly decreased amount of remyelination, due
presumably to an eventual depletion of precursors. It
is important, however, that whatever oligodendrocytes
are still around under these very extreme conditions
still do myelinate axons. A similar finding has been
shown in chronic experimental allergic encephalomyelitis (EAE) by Raine et alC4) and probably is of very
great relevance in MS. The lesson is that the longer an
offending agent acts, the more regenerating cells will
be depleted, and the less successful remyelination will
The next experimental situation was designed to
produce a situation where axons were lost or damaged
and the effect on the oligodendrocytes explored. In
this set of experiments the functional capacity of the
adult myelination system to recover from injury was
tested. Long-term wallerian degeneration was produced in the optic nerve of rats by enucleation, and the
fate of the oligodendrocytes followed {5J The results
should be compared with the peripheral nervous system where we know there is a very large functional
capacity to regenerate after similar episodes. Within 5
days after enucleation, a very small proliferative response is found in the oligodendrocyte. The animals
are then kept alive for periods of up to 2 years, with
their optic nerves deprived of axons. An interesting
observation is that all the axons degenerate and the
optic nerve becomes one large glial scar. In addition,
unusual-looking cells appear that have some of the
characteristics of oligodendrocytes but with sparse
shrunken cytoplasm and notched nuclei.
These are all the characteristics of quiescent cells.
Cell identification was difficult, but we found that the
markers that distinguished them as oligodendrocytes
were the myelin oligodendrocyte glycoprotein (MOG)
described by Linington and carbonic anhydrase. Of interest also is that staining of the optic nerves with MBP
showed no positive cells. The important question to
be answered was whether these quiescent cells could
be functionally resuscitated? Because a gliotic optic
nerve cannot be hooked up to a normal nerve to test
this, as in the peripheral nervous system, fragments of
the gliotic optic nerves were transplanted into the
brains of neonatal shiverer mice. After about 4 weeks
there was some ingrowth of axons into the grafts, and
the cells in the grafts that had previously been entirely
MBP negative, now stained positively. Not only were
there MBP-positive cells in the graft, but MBP-positive
myelin was formed within and at the edge of the graft,
in host tissue. This demonstrated that these cells, which
have lain quiescent for 2 years, when faced with the
stimulus of exposure to naked axons, could in fact be
functionally reactivated. It is unclear whether these
cells derived from less mature precursors or from fully
mature previously myelinating cells, although the finding that they were MOG positive tended to suggest
that they were cells that had achieved a certain level
of maturity.
The phagocytic potential of the oligodendrocyte was
also investigated. For years there was a major contro-
S144 Annals of Neurology Supplement to Volume 36, 1994
versy both in the peripheral and the central nervous
system as to whether myelinating cells were phagocytic. In the earlier stages there was a brisk macrophage/micro@ phagocytosis (demonstrated both by
GSA lectins and EDI), which constitute the major
phagocytic activity as has been previously observed in
the peripheral nervous system. Oligodendrocyte
phagocytosis played a much smaller role. There appeared to be a retraction of debris and the myelin
membrane into ohgodendrocyte processes, rather than
classical phagocytosis. This debris can persist in the
processes for a long time,
This was tested further in the Ola mouse. This mutant has a defect in the axonal membrane, which makes
it unable to recruit macrophages, and therefore has a
very much delayed wallerian degeneration. Even in this
situation where the macrophages are not recruited, oligodendrocytes behave as they do in normal animals
and do not become facultative macrophages [b].
Therefore, it is hghly likely that the survival characteristics of oligodendrocytes or their ability to regenerate
are not affected by phagocytosis.
In many respects, therefore, one can draw many parallels between the behavior of the oligodendrocytes
and the Schwann cells in wallerian degeneration. Earlier on, one finds there is an immediate reaction to
the axonal section in that both Schwann cells and now
oligodendrocytes have a very quick burst of proliferative activity. During the active phase, phagocytosis is
mainly a macrophage activity, although there is a very
minor component from Schwann cells and from ohgodendrocytes. Both the cell types then become quiescent and turn off their myelination proteins in the dormant phase. Exposure to axons can stimulate the
Schwann cell or the oligodendrocyte to reexpress these
proteins and to remyelinate. Although there is a differ-
ence obviously in the quantity, there is a behavioral
analogy between the two cell types.
We therefore see that there is a considerably greater
capacity for regeneration and functional survival by the
endogenous oligodendrocyte system than was previously thought. The challenge will be to devise methods
to utilize this capacity to promote more substantial and
meaningful remyelination in human and experimental
demyelinating diseases. In addition, each individual
model of disease, experimental or clinical, must be
looked at on its own merits. In one model where the
mature cells have been knocked out, proliferation and
remyelination may have to depend on precursor cells,
whereas in another model there may be proliferation
of mature cells. Finally, the events taking place in the
astrocytic environment and any possible damage to the
axon must be considered before one can predict remyelination.
1. Iassman H, Suchanek G, Ozawa K. Histopathology and the
blood-cerebrospinal fluid barrier in multiple sclerosis. Ann
Neurol 1994;36S42-S46
2. Raine CS. The Dale E. McFarlin memorial lecture: the immunology of the multiple sclerosis lesion. Ann Neurol 1334;36:S61s72
3. Ludwin SK, Szuchet S. Myelination by mature ovine oligodendrocytes in vivo and in vitro: evidence that different steps in the
myelination process are independently controlled. Glia 1333;8:
219-2 3 1
4. Raine CS,Hintzen R, Trangott U, Moore GR. Oligodendrocyte
proliferation and enhanced CNS remyelination after therapeutic
manipulation of chronic relapsing EAE. Ann NY Acad Sci 1988;
540:712-7 14
5. Ludwin SK. Ohgodendrocytes from optic nerves subjected to
long term wallerian degeneration retain the capacity to myelinate.
Acta Neuropathol 1992;84:530-5 37
6. Ludwin SK, Bisby MA. Delayed wallerian degeneration in the
central nervous system of Ola mice: an ulva~tructuralstudy. J
Neurol Sci 1992;109:140-147
Ludwin: CNS Remyelination S145
Без категории
Размер файла
315 Кб
central, chronically, remyelination, nervous, damage, system, tissue, studies
Пожаловаться на содержимое документа