вход по аккаунту


Antiserum-mediated demyelination Relationship between remyelination and functional recovery.

код для вставкиСкачать
Antiserum-Mediated Demyelination:
Relationslup Between Remyelination
and Functional Recovery
Kyoko Saida, M D , Austin J. Sumner, M D , Takahiko Saida, M D ,
Mark J. Brown, MD, and Donald H. Silberberg, M D
A focal demyelinative lesion of peripheral nerve was produced by intraneural injection of either antiserum from
rabbits with experimental allergic neuritis or experimental allergic encephalomyelitis or antiserum to galactocerebroside. We studied the relationship between clinical and electrophysiological recovery from this lesion and
the morphological pattern of remyelination. Foot muscles on the injected side weakened within an hour of injection
and remained paralyzed for 7 days; strength gradually returned to normal by 16 days after injection. Electrophysiological conduction block, apparent within a few hours of injection, persisted for about 7 days. At 8 days
we detected dispersed, very low amplitude muscle action potentials with long latency. Morphologically, demyelinated axons were surrounded by Schwann cells at 7 days after injection, but compacted myelin was not present.
After 8 days, remyelinating axons became surrounded by thickening compacted myelin. The time of onset of remyelination and the rate of remyelination u p to 14 days following the injection were independent of axon size. The
onset of clinical and electrophysiological recovery from the lesion corresponded to the appearance of 2 to 8 myelin
lamellae around each remyelinating axon. At 37 days after injection, when conduction velocities had returned to
preinjection values, myelin thickness of remyelinating fibers had increased to approximately one-third that of control nerves.
Saida K, Sumner AJ, Saida T, et al. Antiserum-mediated demyelination: relationship between remyelination
and functional recovery. Ann Neurol 8:12-24, 1980
Acquired local and generalized demyelinating neuropathies are important causes of human peripheral
nerve disorders [9]. Diagnosis is often made by
electromyographic demonstration of slowed conduction velocities or the presence of conduction block at
the site of a lesion. Functional and electrophysiological recovery after a demyelinating neuropathy is often
very good, presumably because remyelination occurs
around demyelinated but otherwise intact axons.
However, the relationship between electrophysiological recovery and remyelination is incompletely
understood. To investigate this relationship further, we examined the evolution of functional,
electrophysiological, and morphological recovery
in an experimental focal demyelinating neuropathy.
We previously described a method for producing a
readily timed acute focal peripheral nerve demyelinative lesion with negligible axonal degeneration [38-40, 431. In this model, animals are injected
intraneurally with either antisera from rabbits with
experimental allergic neuritis (WN-EAN) or experi-
mental allergic encephalomyelitis (WM-EAE) induced by whole nerve or white matter sensitization,
or antisera from rabbits sensitized with galactocerebroside (GC). T h e animals develop weakness of toe
and ankle movements on the side of the injected sciatic nerve and remain paralyzed for more than a
week. Sequential recordings of motor conduction
through the demyelinative lesion reveal that focal
conduction block begins 30 to 60 minutes after antiserum injection, progresses to completion within 3
to 4 hours, and persists for more than a week 1461.
Demyelination begins at the nodal and paranodal regions within a few hours and is completed within 5
days with the participation of phagocytic macrophages. Cross sections of the focal lesion show that
demyelinated axons occupy at least 80% of the fascicular area [38-401. Using this model of peripheral
nervous system demyelination, we studied and now
describe the relationship between clinical and electrophysiological recovery and the morphological
evolution of remyelination. T h e three measures of
recovery shared a close temporal relationship.
From the Department of Neurology, University of Pennsylvania
School of Medicine, Philadelphia, PA.
Address reprint requests to Dr Saida, Department of Neurology,
Utano National Hospital, Narutaki, Ukyo-ku, Kyoto 616, Japan.
Received Aug 10, 1079, and in revised form Oct 3. Accepted for
publication Oct 28, 1979.
0364-5134/80/070012-13$01.25 @ 1979 by Kyoko Saida
Materials and Methods
Male New Zealand albino rabbits weighing 2.3 to 2.7 kg
were used to produce antiserum. Sera were obtained from
rabbits immunized with homogenized bovine white matter
in complete Freund's adjuvant (CFA) to produce WM-EAE
[40],with homogenized bovine peripheral nerve and CFA
to produce WN-EAN [ 3 8 , 4 3 ] ,and with purified GC plus
bovine serum albumin (BSA) in CFA [39, 41, 421 to produce anti-GC antibodies. Previous work demonstrated
that the demyelinating activity in peripheral nerve induced
by these sera correlated well with anti-GC antibody titer in
each serum, and patterns of demyelination produced by the
three sera could not be distinguished from one another
[38-401. For this study we selected for injection three of
each antiserum, all with anti-GC antibody titers of 1 : 128
or greater as measured by agglutination of liposomes containing G C [ 1 I ] . Control sera were obtained from 8 rabbits
inoculated with BSA, CFA, and saline [38-40, 431.
Male Wistar rats weighing 300 to 350 gm were used as
recipients. Fifty microliters of WM-EAE serum, WN-EAN
serum, or anti-GC antiserum (subsequently referred to
collectively as antiserum) was mixed with 20% fresh guinea
pig serum (v/v) as a source of complement and was injected
into the subperineurial portion of one sciatic nerve. An
equal volume of control serum similarly mixed with guinea
pig serum was injected into the contralateral side. The
method of injection is described elsewhere [38]. Injected
rats were examined daily for signs of leg or foot weakness,
and a quantitative assessment of clinical status was recorded.
Electropbysiological Studies
Sciatic nerves of 9 Wistar rats were injected with WM-EAE
serum, which was also used for the morphological and
morphometric studies. Nine similar rats injected with antiserum to BSA served as controls. Rats were anesthetized
with sodium pentobarbital (30 mg per kilogram of body
weight intraperitoneally). Temperature in the hind limb
was carefully maintained between 37.5" and 38.5"C with
the aid of a heat lamp and intramuscular thermistor probe.
Serial electrophysiological studies were performed in the
same animals after injection of antiserum or control serum.
To follow the electrophysiological changes occurring
through the focal demyelinative lesion, the sciatic nerve
was stimulated at the sciatic notch proximal to the injection
site and the posterior tibial nerve at the ankle distal to the
lesion. Motor nerve conduction velocity and compound
muscle action potential responses were studied by methods
previously described [22]. Steel needle stimulating electrodes (Grass Instrument Company, Quincy, MA) were inserted through the skin to lie adjacent to the sciatic nerve
in the thigh and the posterior tibial nerve at the medial side
of the ankle. Stimulus pulses (0.1 to 0.2 msec duration)
were delivered through an isolated stimulator at 1-second
Muscle action potentials were recorded using an active
electrode with the tip just penetrating the skin of the plantar surface over the middle of the interosseus and lumbrical
muscle bellies; a remote electrode was placed at the base of
the lateral digit. Potentials were amplified and displayed on
one beam of a Tektronics cathode ray oscilloscope. The
distance between stimulating electrodes was measured on
the skin and used for calculation of conduction velocity.
The presence of conduction block (neuropraxia) was
determined by demonstrating partial or complete failure of
sciatic nerve conduction, measured by a fall in amplitude of
the muscle evoked potential produced by nerve stimulation
proximal to the injection site while distal stimulation produced muscle evoked responses of normal amplitude. Active demyelinating antiserum produced conduction block
that became fully established within 4 hours and persisted
for 7 days after the injection [46]. Five animals in which
injection of demyelinating antiserum into the sciatic nerve
produced complete or nearly complete conduction block
were subjected to serial electrophysiological examination
to study the evolution of recovery.
Morphological Techniques
At least 2 animals were studied at each of the following
times after injection: 7 , 8 , 10, 14, 21, 28, and 37 days. Following pentobarbital anesthesia, rat sciatic nerves were
fixed in situ for 5 minutes with 3.6% glutaraldehyde/
phosphate-buffered solution at p H 7.4. The sciatic nerves
were excised, dissected into four portions, fixed in 3.6%
glutaraldehyde solution for 2 hours, and washed overnight
in phosphate-buffered solution. Tissues were then postfixed with 2% osmium tetroxide for 2 hours, dehydrated in
graded ethanol and propylene oxide, and embedded in
epoxy. Toluidine blue-stained cross sections 1 p thick
were prepared and studied by light microscopy. Ultrathin
sections were double stained with uranyl acetate and lead
citrate and were examined with an electron microscope.
For single fiber studies, nerve segments were postfixed in
2% osmium tetroxide for 4 hours, then were placed in
66% glycerin solution and examined after 2 days. Fibers
were teased onto glass slides in a droplet of creosote, overlaid with a Permount coverslip, and examined with the light
Morpbometric Studies
To evaluate the relationship between axon size and myelin
sheath thickness in the early recovery period, ultrathin
cross sections were made from the two most extensive lesions proximal to the injection site 7 , 8 , 10, and 14 days
after injection. Electron micrographs of 30 to 6 0 nerve
fibers that did not show obvious features of being
Schmidt-Lantermann clefts, nodes, or paranodes 191 were
taken in a uniform manner to avoid selection based on
morphological appearance of the fibers, and were printed at
a final magnification of x 13,500. The number of myelin
lamellae surrounding each fiber was counted under the dissecting microscope. T o look at the late recovery period 37
days following injection, light micrographs of cross sections
were enlarged to a final magnification of ~ 3 , 1 5 0 ,and
myelin sheath thickness was measured directly with a calibrated magnifying loupe. Axon area was measured from
the coordinates of its perimeter using Simpson's rule with a
digitizer (Elographics Inc, Oak Ridge, TN). The relationship between number of lamellae or sheath thickness and
area was determined by linear regression analysis.
To examine the relationship between fiber diameter and
internode length, we teased 21 single fibers passing
Saida et al: PNS Remyelination and Functional Recovery
through the demyelinated lesion at 37 days after injection,
a time when electrophysiological measurements had returned to normal, These fibers were selected from various
portions of the demyelinated fascicle for having normalappearing internodes at both ends, indicating healed segmental demyelination and not axonal degeneration [ 141.
Internode length and diameter of the longest remyelinating
segment in the fiber were measured directly through the
light microscope using a calibrated ocular micrometer.
When a fiber varied in thickness along its length, the widest
diameter was assumed to be the true diameter for that
fiber [8].The relationship between external diameter of individual fibers and internode length was expressed by the
method of Fullerton [121.
As in previous studies, all rats lost reflex toe spreading and developed weakness of ankle motion on the
side injected with antiserum; the disability appeared
within an hour and persisted for at least 7 days following injections. Between days 7 and 14, all animals
showed gradual daily improvement in toe and ankle
movements. Reflex toe spreading returned about 10
days after injection. By 14 days, most rats walked
normally with apparent full return of foot and ankle
strength. By 16 days, all had attained complete clinical recovery. The control serum-injected hind limbs
did not show evidence of weakness throughout the
observation period.
Elect rophyJiological Findings
Electrophysiological conduction block was apparent
within a €ew hours after injection. In general, recovery of conduction in blocked s o n s was signaled
by the abrupt appearance at 8 o r 9 days of longlatency responses to proximal stimulation (Fig 1,
Table). The time of first detection of such responses
was remarkably consistent in all 5 animals (Table).
Over the next day o r two, increasing numbers of
motor axons passing through the lesion recovered
the ability t o conduct impulses, but with a wide range
of velocities that gave rise to dispersed muscle
evoked responses (see Fig 1). The evoked responses
became progressively more synchronous as individual velocities increased, until maximum motor
conduction velocity and muscle response amplitude
were restored to control levels between days 25 and
35 after injection.
The early appearance of a small-amplitude response was identified in 2 animals between 1 and 3
days following the injection (see Fig 1). This early
recovery response occurred at a latency that fell
within the normal range, and it did not increase in
amplitude over the next several days. Muscle evoked
potentials recorded from nerve stimulation distal to
the lesion did not change appreciably throughout the
time course of the conduction block.
14 Annals of Neurology Vol 8 No 1 July 1980
F i g I . Serial sciatic nerve electrophysiological recordings from
the same rat a fmhours t o 25 days after injection of W M EAE serum. The mascle potential (M response) evoked after tibial nerve stimulation at the ankle is recorded on the left; that
after iciatic nerve stimulation at the hip, proximal to the injection site, on the right. All potentials are superimposed recordings of three or four responses. A control recording prior t o
injection of the demyelinating antiserum is shown (top line,
left). Partial conduction block of the sciatic nerve without
delay in latency i s apparent 1 hour 5 minutes after antiserum
injection (top line, center) and is complete 2 hours 4 5 minutes after the injection (top line, right). A small decrease in M
response after ankle stimulation is evident. Conduction block
persiJts untilday 7 (days 1, 3 , and 7, left), although a
small-amplitude response of normal latency is detectable with
higher gain on day 3 (days 3 and 7, center and right). A
long-latency,low-amplitude, dispersed response t o sciatic nerve
stimulation first appears on day 9 (day 9, center and right).
Stimulation proximal t o the demyelinative lesion produces
progresji?iely larger amplitude, shorter latency potentials as
time progresses, so that by day 25 the response has returned to
control values. (Calibration: longitudinal bar = I msec; vertical bar = I mv.)
D a y of Reappearance of a Major Response to Proximal Sciatic Newe Stimulation in 5 R d s Studied Serially
Experimental Rat
Days after Injection
of WM-EAE Serum
No. 1
No. 2
No. 4
No. 5
No. 3
0 = no long latency response, indicating complete or nearly complete conduction block of motor fibers; + = small response, less than 3 long
latency units; ++ = dispersed response of more than 3 units; NT = not tested.
Morphological Findings
Macroscopic observation at the time of nerve dissection 7 days after the injection revealed only a mild
increase of epineurial connective tissue. The microscopic lesion within antiserum-injected nerves
ranged between 1 and 2 cm in length, extending both
proximal and distal to the injection site. In most cases
the demyelinative lesion occupied 80 to 90% of the
cross sectional area of the fascicle, but at times
the entire fascicle was affected (Fig 2A). At the
proximal and distal ends of the lesion, demyelinated
fibers were seen most frequently at the subperineurial portion of the fascicle. Control serum-injected
nerves did not show demyelinative changes and generally appeared normal (Fig 2B).
Despite unavoidable variations in injection technique, there was great uniformity in the extent and
character of the demyelinative lesions among nerves
from different animals sacrificed on the same day. At
7 days following injection, most demyelinated axons
were reinvested by Schwann cells in which the cytoplasm stained darker than normal with toluidine blue
(Fig 3). In none of the 6 nerves examined at this time
was there compacted myelin. Debris-laden macrophages were scattered within the endoneurium, especially around venuies and in subperineurial spaces.
Ultramicroscopically, Schwann cell cytoplasm had
abundant polysomes, rough endoplasmic reticulum,
and Golgi complexes. Degenerating Schwann cells,
which are seen frequently in the acute phase of
antiserum-mediated demyelination, were not observed. Axons surrounded by uncompacted myelin
were infrequent despite careful search. In general,
demyelinated axons were smaller than those still
myelinated. Longitudinal sections demonstrated that
this size reduction was an accompaniment of focal
demyelination, as the diameter of individual axons
was narrower at bare herninodes than at adjacent
myelinated ones. On occasion, demyelinated axons
contained abnormal membranous profiles, suggesting
chronic axonal degeneration.
The earliest evidence of remyelination was found 8
days postinjection at the border between demyelinated and normal-appearing zones. Occasional axons
present in this region were encircled by a few turns of
compacted myelin. Only a few had uncompacted
myelin, and most axons remained unmyelinated at
this time. By 10 days increasing numbers of axons at
the periphery of the lesion were surrounded by compacted myelin (Fig 4). Myelination seemed to occur
without relationship to axon size. Most axons located
in the center of the lesion remained unmyelinated; a
few had uncompacted myelin. In distal portions of
the demyelinative lesions, evidence of acute axonal
degeneration with swollen, empty axons was observed, but usually less than 5 such fibers could be
found per fascicle. Electron microscopy showed that
most demyelinated axons in cross section were surrounded by a single Schwann cell. Supernumerary
Schwann cells and early remyelination by tunication
rarely were seen around a central rernyelinating axon,
although concentric arrays of a few Schwann cell processes and redundant folds of basal lamina were
sometimes present. Phagocytosis of myelin debris by
Schwann cells and islands of small axons suggesting
sprouts were occasionally encountered in the endoneurium.
By 14 days after the injection, both axon area and
myelin thickness had increased in remyelinating fibers. Ultramicroscopically, remyelination progressed
in a stereotyped way for the majority of axons. At
21 and 28 days, axon area and myelin thickness continued to enlarge. The density of macrophages within
the endoneurium decreased. In longitudinal sections,
short internodes with thinner than normal myelin
sheaths were conspicuous (Fig 5). By 37 days, the last
time examined, remyelinated nerve fibers appeared
normal aside from myelin sheaths relatively thinner
than would be expected for their axon size (Fig 6).
Occasional macrophages were still evident scattered
in the endoneurium.
Teased fiber studies 37 days after injection showed
Saida et al: P N S Remyelination and Functional Recovery
that most myelinated nerve fibers passing through
the lesion had many short, faintly stained internodes intercalated between normally stained long
internodes. At the center of the lesion these continuously remyelinating segments were approximately
1 cm long. At the periphery the affected segments
were shorter and were associated both proximally
and distally with paranodal remyelination over several internodes (Fig 7).
Morphometric Findings
In the early phase of remyelination, myelin thickness
was not related to axon size (Fig S), while at later
16 Annals of Neurology Vol 8 No 1 July 1980
F i g 2. (A)Whole sciatic newe fascicle in cross section I0 days
after injection of WM-EAE serum. The demyeiinative lesion,
identifed by paleness of staining and numerous dark,
punctate, lipid-flled macrophages, occupies all but the upper
right portion of the fascicle. (B) Sciatic nerve 10 duys after injection of control (unti-BSA) serum. Morphologically, this
myelinared fiber i s normal except fov rarefit5er.r .undergoing
wallerian degeneration. (Both toluidine blue; ~ 5 5 . )
F i g 3. Sciatic nerve 7 days after injection of W N - E A N sewm.
Most nervefiders, having been demyelinated, are reinvesfed by
one Schwann cell. No remyelinating axons have compacted
myelin. Debris-laden macrophages are present. (Toluidine blue;
F i g 4. Sciatic nerve 10 duys after injection of W M - E A E
serum. Thickness of the myelin sheath around remyelinating
axons appears t o be independent of axon size. Supernumeravy
Schwann cells are not seen. (Tohidine blue; X890.1
Saida et al: PNS Remyelination and Functional Recovery
F i g 5 . Longitudinal section of sciatic nerve 28 days after injection of anti-GC serum. Remyelinated axons have thin
.sheathsfor their diameters. The segmental nature ofthe proceu
is identijied by the remyelinating internode to the right of one
node of Ranuier (arrowhead), with a normal internode on the
left of the sa9ne axon. (Toluidine blue: x350.j
F i g 6. Sciatic newe 37 days after in,jection of W M - E A E
seram. MyelinatedJiber density of the rempelinating large fa.(cicle appears normal, although .sheath thicknes.r of those Jibers
is thinner than that of myelinatedjhers i n a normal, zrninjected neighboring fascicle (arrow). (Toluidine Mae; ~ 1 7 be0
fore 35 % reduction.)
18 Annals of Neurology
Vol 8 No 1 July 1980
F i g 7. Sequentialsegments of a teased remyelinatedfiber 3 7 days
ufter injection of W M - E A E serum. Letters indicate nodes of
Ranuier. Remyelinated internodes (a-b, c-dj e-f, g-h, and
i-j) show faint staining. (Osmium tetroxide; X80 before 5 %
times there were more myelin lamellae around larger
axons than smaller ones (Fig 9). At 7 days following
injection, only a few demyelinated axons were surrounded by uncompacted myelin as determined with
the electron microscope (Fig 8A). At 8 days fewer
than 5 % of demyelinated fibers had remyelinated,
with compacted myelin ranging from 2 to 8 lamellae
in thickness (Fig 8B). At 10 days approximately 40%
of demyelinated axons had acquired compact sheaths,
their thickness ranging from 2 to 19 lamellae (mean,
5 ? 6 SD; N = 4 3 ) (Fig SC). At 14 days most demyelinated axons had become ensheathed, myelin
thickness ranging from 6 to 26 myelin lamellae (17 ?
6; N = 32) (Fig SD). Axon sizes of these sampled
fibers were smaller than those of normal nerves (Fig
9B). A positive correlation (Y = 0.49) between axon
size and myelin thickness was found by 37 days (Fig
9A). Correlation was less than in control seruminjected nerves also examined at 37 days (Y = 0.75;
Fig 9B). The smaller regression coefficient in remyelinating fibers (slope = 0.15) than in controls
(slope = 0.48) confirqed that myelin sheaths were
thinner in remyelinating fibers. The mean axon area
of remyelinating fibers (10.5 k 7.2 p2) was not
significantly different from that of controls (11.1 t
7.6 p').
The relationship between external fiber diameter
and internode length of teased remyelinating fibers
was studied 37 days after injection (Fig 10). The
mean internode length of the fiber with the smallest
external diameter was 306 5 7 0 p, and that of the
largest, 338 k 49 p. Linear regression analysis of
mean internode length for each remyelinating fiber
plotted against its external diameter revealed a small
positive relationship (Y = 0.33, N = 21), indicating a
possible trend for larger axons to have larger internodes at this stage of recovery.
The present study demonstrated that the onset of
clinical improvement distal to a focal demyelinativc
neuropathy correlates closely with return of motor
conduction in the affected nerve and initiation of remyelination at the site of the lesion. Serial electrophysiological recordings of motor conduction
through the demyelinative lesion provided a more
sensitive and objective method than clinical observations for following the precise time course of recovery. Nerve conduction measurements also were a
more sensitive measure of persisting abnormalities
than was clinical examination. Complete clinical recovery of leg strength usually was observed by 14
days after injection, while conduction velocity did
not become normal until about 25 days. Clinical
improvement in the presence of persisting slow
conduction also has been reported in another
demyelinative disorder, experimental diphtheritic
neuropathy [22, 281.
Present experimental techniques d o not allow
study of both electrophysiological function and ultrastructural morphology in the same demyelinated
axon. Instead, information has been acquired from
population studies on similarly treated nerves [2, 10,
1 7 , 3 11. Antiserum-induced demyelination, however,
provides a better system for such correlative studies
Saida et al: PNS Remyelination and Functional Recovery
14 days
r= 0
10 days
* *
8 days
0 -
3 c o
Axonal Area ( r m 2 )
because the time course of electrophysiological and
morphological recovery in this experimental model is
predictable from animal to animal. We found that
motor nerve conduction first returns in a few axons 8
or 9 days after antiserum injection, demonstrated by
the appearance of long-latency muscle responses.
Recovery proceeds rapidly thereafter. Conduction is
restored only after axon remyelination is underway. The reappearance of conduction in previously
blocked axons on the eighth day following injection
occurs when a small number of axons are encircled
with 2 to 8 turns of myelin lamellae. Conduction
across the lesion could not be demonstrated on day 7,
when axons in the affected region were encircled
only by Schwann cell cytoplasm without compact
myelin. Therefore our observations d o not support
the contention that electrophysiological recovery
after focal demyelination occurs by a process of
continuous conduction in completely demyelinated
axons [ S ] . They do not disprove that hypothesis,
however, since demyelination in our model progresses more rapidly, and thus the mechanism of
conduction operating in these two models may be
20 Annals of Neurology
Vol 8 No 1 July 1980
F i g 8. Relationshap between axon area and myelin sheath
thickness of remyelinating axons i n cross section 7 , 8, 10, and
14 days after injection of W M - E A E serum. Open circles represent unmyelinated axons, open circles with a dot represent
axons with unconzparted myelin, and filled circles indicate
axons with rompacted myelin. At 7 days (A)most axons are
unmjielinated, but a few have uncompacted myelin. At 8 days
(B) some axons have rompacted myelin with fewer than eight
lamellae. At 10 days (Cj half of the fibers examined have compacted myelin with fewer than 20 lamellae. At 14 days (D),
nerve fibers are remyelinated by 6 to 26 turns of compacted
myelin lamellae. There is no correlation between axon area and
myelin thickness during this early phase of remyelination, and
remyelination begins at the same time around axons of all sizes.
(n = number ofjbers examined; r = correlation coefficient.?
In computer simulations of conduction in demyelinated axons, Koles and Rasminsky [ 2 4 ] predicted that conduction could proceed through an
internode with a myelin thickness only 2.7% of normal, but with a substantial delay. With two or more
internodes involved, myelin thickness greater than
4% would be required to sustain conduction. The
present results support their prediction, since a small
25 r
Y = 0.15~
t 6.20
! *
n = 108
0 ~ " " 1 " " 1 " " ! " " 1 " " I " " I "
Axonal Area ( p m 2 )
Y= 0. 40x -t 6.76
n = 77
F i g 9. Relationship between axon area and myelin sheath
thickness i n sciatic nerve fibers in cross section 3 7 days after
injection of WM-EAE serzm (A) and control serum (B). There
is now a moderate positive correlation between axon area and
myelin thickness, although the slope of the regression line i s less
than i n controlfibers. Axon areas {mean k SD) in remyelinatingfibers and controls do not ddffersignijcantly (10.5
k.7.2 cersus 11.1 k'7.6 p2).
700 -5 600-
- 5oo 0
0 300-
;2 o o -
} 1 fi p 1
5 100-
Saida et al: PNS Remyelination and Functional Recovery
(see Fig 1). This rapid recovery of conduction could
represent a reversible effect on axonal conduction
independent of demyelination, perhaps due to
nonspecific local inflammatory reactions such as
edema or as a direct effect of the demyelinating
serum on nodal axolemma. Alternatively, it could be
the result of recovery from a minor degree of
paranode demyelination. This phenomenon may be
the equivalent in the peripheral nervous system of a
similar mechanism thought to occur in the early
phase after central nervous system demyelination, in
which recovery appears to be unrelated to remyelination C23, 291.
The remyelinative patterns after antiserummediated demyelination resemble those reported in
the recovery phase of EAN [4,44],chronic recurrent
EAN [33, 481, the peripheral nervous system in
chronic EAE [35],other experimental demyelinating
neuropathies [ 3 , 18, 25, 261, and chronic relapsing
polyneuropathy in humans [341. However, supernumerary Schwann cells, multiple mesaxons, and irregular paranode patterns were infrequent in our
model. Well-developed onion bulb formations were
not found, probably because demyelination induced
by intraneural injection of antiserum is a single
monophasic event. Remyelination that occurs after
compression by a tourniquet [lo, 321 or nylon cord
[37] is limited to paranodal regions of larger fibers
and takes place over several weeks o r months. Remyelination produced by intraneural injection of
diphtheria toxin [20, 2 11 usually starts about 20 days
after injection. In both of these experimental systems
the demyelinative and remyelinative processes commence slowly [ l o ,20,21,32, 371, so neither model is
suitable for precise studies of the relationship between functional recovery and remyelination in the
peripheral nervous system. Demyelination produced
by the perineurial window [7,45],although morphologically similar in some ways to our model, is associated with striking herniation of nerve fibers through
the defective perineurium, and axon swelling or distortion often precedes demyelination. Further, the
demyelination that occurs following a perineurial
window is slower than that of antiserum-mediated
demyelination, which progresses in hours [38-401.
Intraneural injection of lysophosphatidyl choline
(lysolecithin) produces rapidly evolving demyelination and remyelination with a time course only
slightly slower than that of antiserum injection [ 1 5 ,
191. However, on most occasions lysolecithin demyelinates only a small portion of the injected fascicle, and most fibers remain unaffected. Unlike injection of serum with high anti-GC antibody titers,
lysolecithin reportedly does not affect Schwann cells,
and remyelination by tunication is prominent [151.
The injection procedure used in this study pro-
Annals of Neurology Vol 8 No 1 July 1980
duces wallerian degeneration, but with current techniques this affects fewer than 5% of myelinated
fibers in the fascicle. If a substantial amount of wallerian degeneration were induced by the injection procedure, it would have been apparent from the electrophysiological studies, in which motor responses
from degenerating axons would have disappeared
within 24 to 36 hours and the amplitude of the distally evoked response would have decreased [ I S , 16,
22, 301. Preservation of the ankle evoked responses
to distal stimulation over the period of this study correlates closely with the morphological observation
that an antiserum-induced lesion is a relatively pure
demyelinative lesion without significantly associated
axonal degeneration 138-401.
Remyelination after antiserum-induced demyelination starts around 7 days postinjection, regardless
of axon size. Each axon is surrounded by one
Schwann cell in a one-to-one relationship similar to
that almost invariably found in normal peripheral
nerve development [6, 471. Our morphometric
studies and observations of longitudinal sections
demonstrated that axons demyelinated focally with
antiserum have smaller diameters than normal at the
lesion site. Axons of the peripheral nervous system
being remyelinated after antiserum-mediated demyelination also have smaller diameters than normal;
the reduction in caliber occurs during the initial demyelinative event [ l ,27, 361.
The number of myelin lamellae around remyelinating fibers does not correlate with axon size
for up to 14 days after injection. The internode
length of remyelinating rat sciatic nerve fibers, approximately 300 p, also does not relate to axon size
in the early weeks of remyelination. The mean remyelinated internodal length of rat sciatic nerve
fibers studied 165 days after injection of diphtheria
toxin similarly was 300 p , also apparently unrelated
to axon diameter [21].
By 37 days after injection the myelin sheath
thickens as the axon enlarges, although the correlation is not as close as in normal nerves. Myelin thickness reaches approximately one-third that in control serum-injected nerves. At that time, normal
maximum conduction velocity is reestablished and
the proximal evoked muscle action potential amplitude has returned to control values. This indicates
that conventional motor conduction velocity studies
are relatively insensitive for detecting a uniform reduction in myelin sheath thickness in a focal remyelinating lesion extending over as many as 30 consecutive internodes.
Our correlative studies using intraneural injection
of antiserum to produce a timed focal demyelinative
lesion demonstrated that electrophysiological abnormalities d o not improve until remyelination takes
place, about 7 or 8 days after injection. This suggests
that at least one mechanism underlying amelioration
of conduction block is remyelination, and that electrophysiological improvement in demyelinating neuropathies heralds a measure of morphological and
possibly functional recovery.
Supported by Grants NS08075 and NS11307 from the National
Institutes of Health, by Grant 894-B from the National Multiple
Sclerosis Society, and by the Muscular Dystrophy Association and
the Kroc Foundation.
The authors thank Miss Lois Stefanowitz and Miss Leslie Hansen
for technical assistance and Mrs Barbara J. Whiting for secretarial
1. Aguayo AJ, Bray GM, Perkins SC: Axon-Schwann cell relationships in neuropathies of mutant mice, in Harris JB (ed):
Muscular dystrophy and other inherited diseases of skeletal
muscle in animals. Ann N Y Acad Sci 317:512-531, I979
2. Aguayo A, Nair CPV, Midgley R: Experimental progressive
compression neuropathy in the rabbit. Histologic and electrophysiologic studies. Arch Neurol24:358-364, 1971
3. Allt G: An ultrastructual analysis of remyelination following
segmental demyelination. Acta Neuropathol (Berl) 22:333344, 1972
4. Ballin RHM, Thomas PK: Electron microscope observations
on demyelination and remyelination in experimental allergic
neuritis. Part 2. Remyelination. J Neurol Sci 8:225-237,
5. Bostock H, Sears TA: Continuous conduction in demyelinated mammalian nerve fibers. Nature 263:786-787, I976
6. Brown MJ, Radich SJ: Polyaxonal myelination in developing
dystrophic and normal mouse nerves. Muscle Nerve 2:217222, 1979
7. DeBaecque C, Raine CS, Spencer PS: Copper binding at PNS
nodes of Ranvier during demyelination and remyelination in
the perineurial window. Neuropathol Appl Neurobiol 2:
459-470, 1976
8. Dyck PJ, Gutrecht JA, Bastron JA, et al: Histologic and
teased-fiber measurements of sural nerve in disorders of lower
motor and primary sensory neurons. Mayo Clin Proc 43231135, 1968
9. Dyck PJ, Thomas PK, Lambert EH: Peripheral Neuropathy,
vols 1 and 2. Philadelphia, Saunders, 1975
10. Fowler TJ, Danta G, Gilliatt RW: Recovery of nerve conduction after a pneumatic tourniquet: observations on the hindlimb of the baboon. J Neurol Neurosurg Psychiatry 35:638647, 1972
11. Fry JLM, Lisak RP, Manning MC, et aI: Serological techniques
for detection of antibody to galactocerebroside. J Neurol Sci
8:397-411, 1976
12. Fullerton PM, Gilliatt RW, Lascelles RG, et al: The relation
between fiber diameter and internodal length in chronic
neuropathy. J Physiol (Lond) 178:268, 1965
13. Gilliatt RW, Hjorth RJ: Nerve conduction during wallerian
degeneration in the baboon. J Neurol Neurosurg Psychiatry
35:335-341, 1972
14. Gledhill RF, McDonald WI: Morphological characteristics of
central demyelination and remyelination: a single fiber study.
Ann Neurol 1:552-560, 1977
15. Gregson NA, Hall SM: A quantitative analysis of the effects
of the intraneural injection of lysophosphatidyl choline. J Cell
Sci 13:257-277, 1973
16. Gutmann E, Holubar J: The degenerarion of peripheral nerve
fibers. J Neurol Neurosurg Psychiatry 13:89-105, 1950
17. Hall JI: Studies on demyelinated peripheral nerves in guinea
pigs with experimental allergic neuritis. A histological and
electrophysiological study. Part 11. Electrophysiological observations. Brain 90:313-332, 1967
18. Hall SM: Some aspects of remyelination after demyelination
produced by the intraneural injection of lysophosphatidyl
choline. J Cell Sci 13:461-477, 1973
19. Hall SM, Gregson NA: The in vivo and ultrastructural effects
of injection of lysophosphatidyl choline into myelinated peripheral nerve fibers of the adult mouse. J Cell Sci 9:769-789,
197 1
20. Jacobs JM: Experimental diphtheritic neuropathy in the rat.
Br J Exp Path01 48:204-216, 1967
21. Jacobs JM, Cavanagh JB: Species differences in internode
formation following two types of peripheral nerve injury. J
Anat 105:295-306, 1969
22. Kaeser HE, Lambert EH: Nerve function studies in experimental polyneuritis. Electroencephalogr Clin Neurophysiol
SUPPI 22129-35, 1962
23. Kayan A, Earl CJ: Compressive lesions of the optic nerves and
chiasm. Pattern of recovery of vision following surgical treatment. Brain 98:13-28, 1975
24. Koles ZJ, Rasminsky M: A computer simulation of conduction in demyelinated nerve fibers. J Physiol (Lond) 227:351364, 1972
25. Lampert P, Garro F, Pentschew A: Tellurium neuropathy.
Acta Neuropathol (Berl) 15:308-317, 1970
26. Lampert PW, Schochet SS: Demyelination and remyelination
in lead neuropathy. J Neuropathol Exp Neurol 27:527-545,
27. Lubinska L Region of transition between preserved and regenerating parts of myelinated fibers. J Cornp Neurol
113:315-335, 1959
28. McDonald WI: The effects of experimental demyelination on
conduction in peripheral nerve: a histological and electrophysiological study. 11. Electrophysiological observations.
Brain 86:501-524, 1963
29. McDonald WI: Remyelination in relation to clinical lesions
of the central nervous system. Br Med Bull 30:186-189,
30. Miledi R, Slater CR: On the degeneration of rat neuromuscular junctions after nerve section. J Physiol (Lond) 207:507528, 1970
3 1. Morgan-Hughes JA: Experimental diphtheritic neuropathy: a
pathological and electrophysiological study. J Neurol Sci
7~157-175, 1968
32. Ochoa J, Fowler TJ, Gilliatt RW: Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet. J Anat
113:433-455, 1972
33. Pollard JD, Kmg RHM, Thomas PK: Recurrent experimental
allergic neuritis: an electron microscope study. J Neurol Sci
24:365-383, 1975
34. Prineas JW,
McLeod JG: Chronic relapsing polyneuritis. J
Neurol Sci 27:427-458, 1976
35. Raine CS, Wisniewski H, Dowling PC, et al: An ultrastructurd study of experimental demyelination and remyelination.
IV. Recurrent episodes and peripheral nervous system plaque
formation in experimental allergic encephalomyelitis. Lab Invest 25:28-34, 1971
36. Raine CS, Wisniewski H, Prineas J: An ultrastructural study
of experimental demyelination and remyelination. 11. Chronic
experimental allergic encephalomyelitis in the peripheral nervous system. Lab Invest 21:316-327, 1969
37. Rudge P, Ochoa J, Gilliatt RW: Acute peripheral nerve compression in the baboon. J Neurol Sci 23:403-420, 1974
Saida et al: PNS Remyelination and Functional Recovery
38. Saida K, Saida T, Brown MJ, et al: Antiserum-mediated demyelination in vivo: a sequential study using intraneural injection of experimental allergic neuritis. Lab Invest 39:449462, 1978
39. Saida K, Saida T, Brown MJ, et al: In vivo demyelination induced by intraneural injection of anti-galactocerebroside
serum: a morphologic study. Am J Pathol95:99-110, 1979
40. Saida T, Saida K, Brown MJ, et al: Peripheral nerve demyelination induced by intraneural injection of experimental
allergic encephalomyelitis serum. J Neuropathol Exp Neurol
41. Saida T, Saida K, Dorfman SH, et al: Experimental allergic
neuritis induced by sensitization with galactocerebroside. Science 2041103-1106, 1979
42. SaidaT, Saida K, Silberberg DH: Demyelination produced by
experimental allergic neuritis serum and antigalactocerebroside antiserum in CNS cultures: an ultrastructural study.
Acta Neuropathol (Berl) 48:19-25, 1979
43. Saida T, Saida K, Silberberg DH, et al: Transfer of demye-
24 Annals of Neurology
Vol 8
No 1 July 1980
lination with experimental allergic neuritis serum by intraneural injection. Nature 272:639-641, 1978
44. Schroder JM, Krucke W: Zur Feinstruktur der experimentell-allergischen Neuritis beim Kaninchen. Acta Neuropathol (Berl) 14261-283, 1970
45. Spencer PS, Weinberg HJ, Raine CS, et al: The perineurial
window-a new model of focal demyelination and remyelination. Brain Res 96:323-329, 1975
46. Sumner AJ, Saida K, Saida T, et al: Acute conduction block
produced by intraneural injection of EAE serum: a correlative
electrophysiological and morphological study. Neurology
29:581, 1979
47. Webster HdeF: The geometry of peripheral myelin sheaths
during their formation and growth in rat sciatic nerves. J Cell
Biol 48:348-367, 1971
48. Wisniewski HM, Brostoff SW, Carter H , et al: Recurrent experimental allergic polyganglioradiculoneuritis.Multiple demyelinating episodes in rhesus monkey sensitized with rabbit
sciatic nerve myelin. Arch Neurol 30:347-358, 1974
Без категории
Размер файла
1 335 Кб
antiserum, remyelination, recovery, demyelination, function, relationships, mediated
Пожаловаться на содержимое документа