вход по аккаунту


Class II antigen expression and inflammatory cells in the guillain-barr syndrome.

код для вставкиСкачать
Class I1 Antigen Expression and
Inflammatory Cells in the
Gdain-Barr6 Syndrome
J. D. Pollard, PhD, FRACP, Julia Baverstock, and J. G. McLeod, DPhil(Oxon), FRACP
Monoclonal antibodies to T-lymphocyte subsets and Class 11 molecules (Ia) have been used to characterize the inflammatory infiltrate in nerve tissue biopsied from 2 patients in the acute phase of Guillain-Barre syndrome; the findings
were compared with those in control nerve specimens. Normal control nerve was treated in the same way. In normal
nerves, Class I1 molecules are expressed on endothelial cells and on occasional mononuclear and perineurial cells. In
Guillain-Barre nerves the inflammatory infiltrate consisted mainly of Class 11-positive cells of the monocytemacrophage lineage and of lesser numbers of T4 and T8 lymphoid cells. T4 cells predominated in perivascular
collections. In the more severely affected patient, there was a marked increase of Class I1 molecules expressed on
endothelial cells, perineurial cells, and most Schwann cells. Schwann cells associated with unmyelinated fibers and the
Schwann cell processes of denervated Bungner bands all expressed Class I1 molecules. These histological changes were
less marked in the more mildly affected patient. It is suggested that the expression of Class I1 molecules on the myelin
forming cell, the Schwann cell, has important implications for the pathogenesis of the demyelination that occurs in
Guillain-Barre syndrome.
Pollard JD, Baverstock J, McLeod JG: Class I1 antigen expression and inflammatory cells in the
Guillain-Barre syndrome. Ann Neurol 21:337-341, 1987
Guillain-Barrk syndrome (GBS) is considered an autoimmune disorder; the pathological features are typical
of a delayed hypersensitivity reaction [I]. The target
antigen of the immune response is unknown, and current evidence does not implicate those antigens, P2
protein or galactocerebroside, that are involved in
animal model experimental allergic neuritis [I, 71. By
analogy with experimental allergic neuritis, the role of
myelin has been emphasized as the probable inciting
antigen in the human disease, but the cell that produces and maintains myelin, the Schwann cell, has received little attention.
The induction of most immune reactions involves
the presentation of antigen to T cells by cells bearing
Class I1 (Ia, DR) molecules. Helper T cells can recognize foreign antigen only in association with Class I1
molecules of the same haplotype as they themselves
express. There is now considerable evidence that a
quantitative variation in Class I1 expression may play a
major role in immunoregulation and immunologically
mediated disease 181. In normal tissue, expression of
Class I1 molecules is confined to macrophages,
dendritic cells, B cells, and vascular endothelium.
Neurons and Schwann cells, the component cells of
the peripheral nervous system, do not normally ex-
press Class I1 (or Class I) molecules C2, 91. The induction of Class II molecules on cells normally negative
for HLA DR has been shown in inflammatory disease
of joints 141 and bowel ClZ} and in autoimmune disease of the thyroid [31. We have therefore undertaken
a study of the expression of Class I1 molecules in the
acute inflammatory demyelinating polyneuropathy
From the Department of Medicine, University of Sydney, Sydney
Address correspondence to Dr Pollard, Department of Medicine,
University of Sydney, NSW 2006, Australia.
Sural nerve was removed at biopsy from 2 patients with
GBS. Both patients presented with a predominantly motor
neuropathy with a rapidly progressive course requiring respiratory support within a week of onset. Sural nerve biopsies
from 4 control patients without neuropathy and 5 patients
with toxic neuropathy (organophosphate,nitrous oxide, hydralazine) were also studied.
Mouse monoclonal antibody to the nonpolymorphic region
of the HLA D/DR (Ia) molecule (Clone 12) [13], all T cells
(Clone T1 l), T-helper cells (Clone T4), T-cytotoxic/
suppressor cells (Clone T8), and all monocytes (Clone M 0 2 )
were purchased from Coulter Immunology, Hialeah FL. A
monoclonal antibody to all monocytes, FMC 32, was pur-
Received Apr 16, 1986, and in revised form July 15. Accepted for
publication Juiy 17, 1986.
chased from Australian Monoclonal Developments, Sydney,
Australia. Peroxidase-conjugatedF (ab*)2fragment of sheep
antimouse immunoglobulin G was obtained from Sigma Co,
St. Louis, MO.
Immunological Techniques
In our peroxidase-antiperoxidase
technique, 6-p acetone-fixed, air-dried cryostat sections of
surd nerve were pretreated with swine serum diluted 1:5 for
5 minutes and then reacted with antibodies in the following
1. The primary antibody (e.g., mouse antihuman Ia) was diluted 1:80 for 30 minutes
2. Rabbit antimouse immunoglobulin was absorbed with human spleen cells and diluted 1:750
3. Swine antirabbit immunoglobulin was diluted 1 :50
4. Rabbit peroxidase-antiperoxidase was diluted 1: 100.
In our indirect immunoperoxidase technique, similarly
fixed and pretreated cryostat sections were incubated in the
following sequence: the primary antibody was diluted 1:80,
then peroxidase-conjugated sheep antimouse IgG F (ab')z
fragment was diluted 1: 50. The reaction product was visualized using diaminobenzidine and hydrogen peroxide. Sections were counterstained with Mayer's hemalum.
Two negative controls were used: omission of primary
antibody, and substitution of primary antibody with normal
rabbit serum diluted 1:20. The specificity of all antibodies
used was tested on frozen sections of human lymphoid tissue, some of which were reacted with antibodies of another
The numbers of T cells and monocyte-macrophages were
counted in the largest fascicle in the adjoining 6-p. sections.
ELECTRON MICROSCOPY. Tissue for electron microscopy
was fixed in periodate-lysine-paraformaldehyde [101 and in
2% paraformaldehyde alone, and with 0.05%, 0.25%, and
0.5% glutaraldehyde, for 3 hours at 4"C, then taken through
7% and 15% sucrose in 0.05 M phosphate buffer and finally
through 20% sucrose containing 10% glycerol. Treated
nerves were then snap-frozen in isopentane cooled in liquid
nitrogen. Cryostat sections (6 to 10 p thick) were cut and
collected on plastic coverslips. Sections were then immunostained by the four-layer peroxidase-antiperoxidase method
and the two-step indirect immunoperoxidase method (described above) and examined by light microscopy; they were
finally stained with 2% osmic acid for 30 minutes, then dehydrated and embedded in Spurr resin by inverting the coverslip over a Beem capsule containing resin. Coverslips were
detached by brief immersion in liquid nitrogen. Sections
for electron microscopy were stained with uranyl acetate
and lead citrate and examined in a Philips 300 electron
Tissue fixed with even the most dilute glutaraldehyde
could not be immunostained, but staining was plentiful
with the other reagents. Fixation with paraformaldehyde produced better tissue preservation for electron microscopy than did the periodate-lysine-para-
338 Annals of Neurology Vol 21 No 4 April 1987
formaldehyde method, but the results were inferior to
those obtained when glutaraldehyde was used.
Control Nerves: Light Microscopy
In normal control nerves, T cells were seen only rarely
within the fascicles, and only occasional macrophage-monocytes were seen. Class I1 (1a)-positive staining
occurred only on endothelial cells of endoneurial capillaries, to a minor degree within the perineurium,
and in occasional mononuclear cells (Fig 1A). In control nerves with axonal degeneration, macrophage-monocytes were present in proportion to the degree
of active breakdown in nerve fibres. The expression of
Class I1 molecules within these nerves paralleled the
presence of monocyte-macrophages. Their expression
on Schwann cells was not seen in these nerves.
Patient Nerves
The sural nerve of the more severely afflicted patient
(Patient 1) was diffusely infiltrated with inflammatory
cells, and active demyelination was widespread. In the
other sural nerve (Patient 2), these changes were most
pronounced in a perivascular area within one fascicle.
Small numbers of T cells (25 and 5 per fascicle in
Patients 1 and 2 , respectively) were present in both
nerves. No significant differences were found between
the numbers of T4 cells and T8 cells within the fascicles. The dominant component of the cellular infiltrate
consisted of cells of monocyte-macrophage series, and
a large number (32 1) of these were present throughout
the nerve fascicle in Patient 1 (Fig 1B). In Patient 2 ,
108 monocytes were counted. In both nerves, dense
staining for Class I1 molecules was found within the
endoneurium and perineurium and on endothelial cells
(Fig lC, D). Staining was far more dense within the
nerve fascicle than could be accounted for by mononuclear cells alone (Fig l B , D). Class I1 staining was
sometimes seen immediately adjacent to strongly staining cell surfaces, i.e., the collagen around macrophagrts
and collagen pockets surrounded by Schwann cell
lamellae (Fig 2).
Electron Microscopy
The ultrastructural localization of Class I1 positivity
was on mononuclear cells, endothelial cells (Fig 2A,
B), some perineurial cells, and Schwann cells (Fig 3).
In Patient 2 only a small proportion of Schwann cells
in the region of the infiltrate were Class I1 positive,
but in Patient 1 most Schwann cells expressed these
molecules, including Schwann cells associated with unmyelinated fibers and even denervated Biingner bands
(Fig 3). It may be seen from Figure 2 that the antibody
binds in a particulate fashion to the surface of the
Schwann cell membrane; the appearance was similar to
that on endothelial cells and monocytes. Reaction
product was not found on fibroblasts, and some mononuclear cells were Class I1 negative (Fig 4).
Fig I . (A)Transverse section of normal human sural nerve immunostained with mouse monoclonal antibody to human common
spec;fcity la, and peroxidase-conjugated goat antimouse lgG.
Note reaction product (black) is present sparseh within the nerve
fascicle and perineurium (p). Capillary endothelium ( e ) accounts
for most of the positive staining. (B) Transverse section from a
patient with Guillain-Bawk syndrome (Patient I ) immuno-
stained with FMC 32 (antimonocyte)as primary antibody. Note
a moderate infiltrate of mononuclear cells. (C and 0)Transverse
sectionsfrom two different fascicles immunostained as in A. D is
an adjacent section to B. Note very dense expression of Class I1
molecules within nerve fascicle and endothelium (e). (Bar =
100 IJ..)
Pollard et al: Antigen Expression and Inflammatory Cells in GBS
Fig 2. Electron micrograph of section from sural nerve qfpatient
with Guillain-Bawi syndrome. Nerve has been immunostained
with a four-layer peroxidase-antiperoxidase technique using
monoclonal antibody t o human common spec;fcity la. (A)Endothelium (en). (B) Monocyte-macrophage (m). Note some adjacent collagen. (c = reaction product; a = axon. Bar = I p,.)
This study has confirmed Prineas’s finding [111 that
the macrophage is the predominant cell in the inflammatory infiltrate in GBS nerve. The new finding of this
study is that Class I1 antigen, which is not normally
expressed on neural cells [Z, 71, is abundantly present
in nerve in GBS and in particular is expressed on the
Schwann cell. Although reaction product was sometimes seen immediately adjacent to strongly stained
cell surfaces, i.e., on collagen surrounding macrophages (Fig 2 ) , this would not account for the staining
of Schwann cells, many of which were adjacent to
340 Annals of Neurology Vol 21 No 4 April 1987
Fig 3. Electron micrographs of sections from sura2 nerve of patient with Guillain-Bawi syndrome immunostained as in Figure 2 for human common specif;city la. (A)Without primary
antibody. (B) With primary and secondary antibodies. Note particulate staining of Scbwann cell surface membrane in B. (Sc =
Schwann cell; a = axon; bm = basement membrane. Bar =
1 p,.)
Class 11-negative cells with no positively stained
mononuclear cell intervening (Fig 4). Similarly, collagen pockets enclosed by Schwann cell cytoplasm were
often stained. In vitro studies are planned to determine
whether the Class I1 molecules on the Schwann cell
surface have been absorbed from surrounding mononuclear cells or synthesized by the Schwann cells themselves. Nonspecific binding of anti-Class I1 antibody
or peroxidase to Schwann cells is unlikely as staining
was not seen on control nerve and omission or substitution of the primary antibody produced a negative
result. It is an important point that the fixative most
useful in nerve histology, glutaraldehyde, abolishes the
detectability of Class I1 molecules on cell surfaces in
cells are restricted to target cells that express the same
Class I1 antigens as the antigen-presenting cells
stimulating the response 157. It is likely that the induction of Class I1 antigen on Schwann cells results from
the action of gamma interferon. Bartlett and associates
121 have induced Class I antigens on these cells with
gamma interferon. Interferons are produced by T lymphocytes in response to viral infections 1141, and an
association of GBS with preceding viral infection is
well established El].
Supported by the National Health and Medical Research Council of
The authors are most grateful to the Department of Electron Microscopy, University of Sydney, for use of their facilities.
Fig 4. Electron micrograph of tissue from Guillain-Bad nerve
(Patient 1) immunostained as in Figure 3 against human common specificity la. Note reaction product on Schwann cell (Sc)
surface but not on adjacent mononuclear cell (f), possibly a
fibroblast. (a = axon. Bar = 1 p.)
nerve tissue. A similar effect has been observed when
1% paraformaldehyde is used 1151, so ideal immunostaining of nerve tissue may necessitate binding of the
primary antibody before fixation.
Gamma interferon has been shown to induce Class I
and I1 antigens on neural and nonneural cells and on
thyroid epithelial cells [161, rendering the latter able to
present thyroid microsomal antigens. With regard to
autoimmune disease of the thyroid, Bottazzo and coworkers {37 have postulated that, since thytoid microsomal antigens are normally recognized infrequently by T cells because of their tissue location and
low concentration in the circulation, T-cell tolerance is
unlikely. Hence, induction of autoreactive T cells
would occur with presentation of antigen. Similarly,
one important implication of the presence of Class I1
antigen on Schwann cells in GBS is the possibility that
this cell may present neural antigen and activate the
autoimmune destruction of myelin. Another implication is that the Schwann cell itself may be a target
of immune destruction. Delayed hypersensitivity
responses (such as those involved in GBS or experimental allergic neuritis that are initiated by effector T
1. Arnason BW: Acute inflammatory demyelinating polyradiculoneuropathies. In Dyck PJ, Thomas PK, Lambert EH;
Bunge R (eds): Peripheral Neuropathy. Philadelphia, Saunders,
1984, pp 2050-2100
2. Bartlett PF, Wycherley K, Wong GH: Induction of histocompatability antigens on neural cells by interferon-gamma. Lack of
expression on sensory neurons. Neurosci Lett (suppl 15):S46,
3. Bottazzo GF, Puja-Borrell, Hanafusa T, Feldmann M: Role of
aberrant HLA-DR expression and antigen presentation in induction of endocrine autoimmunity. Lancet 2: 1115-1 119, 1983
4. Burmester GR, Menche 0 , Merryman P, et al: Application of
monoclonal antibodies to the characterisation of cells selected
from articular cartilage. Arthritis Rheum 26:1187-1195, 1983
5. Fitch Fw.T lymphocyte clones having defined immunological
functions. Transplantation 32:171-176, 1981
6. Hanafusa T, Chiovato L, Donach D, et al: Aberrant expression
of HLA-DR antigen on thymocytes in Graves’ disease. Relevance for autoimmunity. Lancet 2:1111-1115, 1983
7. Hughes RAC, Winer JB: Guillain-Barre syndrome. In Matthews WB, Glaser GH (eds): Recent Advances in Clinical Neurology, Vol4. Edinburgh, ChurchillLivingstone, 1984, pp 19-49
8. Janeway CA, Bottomly K, Babich J, et al: Quantitative variation
in la antigen expression plays a central role in immune regulation. Immunol Today 5:99-105, 1984
9. Lisak RP, Hirayama M, Kuchmy D, et al: Cultured human and
rat oligodendrocytes and rat Schwann cells do not have immune
response gene-associated antigen (Ia) on their surface. Brain Res
298:285-292, 1983
10. McLean JW, Nakane P K Periodate-lysine-paraformaldehyde
fixative. A new fixative for immunoelectron microscopy. J Histochem Cytochem 22:1077-1083, 1974
11. Prineas JW:
Acute idiopathic polyneuritis-an electron microscope study. Lab Invest 26:133, 1972
12 Selby WS, Janossy G, Mason DY, Jewell D P Expression of
HLA-DR antigens by colonic epithelium in inflammatory bowel
disease. Clin Exp Immunol 53:614-618, 1983
13 Stashenko P, Nadler LM,Hardy R, Schlossman SF: Characterization of a human B lymphocyte-specific antigen. J Immunol
125:1506-1511, 1980
14 Stewart WE: The Interferon System. New York, SpringerVerlag, 1979, p 421
15. Walker WS, Beelen RA, Buckley PJ, et al: Some fixation reagents reduce or abolish the detectability of Ia antigen and
HLA-DR on cells. J Immunol Meth 67:89-99, 1984
16. Wong, GHW, Bartlett PF, Clark-Lewis I, et al: Interferongamma induces the expression of H2 and la antigens on brain
cells. J Neuroimmuhol 7:255-278, 1985
Pollard et al: Antigen Expression and Inflammatory Cells in GBS
Без категории
Размер файла
2 166 Кб
class, expressions, barry, syndrome, antigen, guillain, inflammatory, cells
Пожаловаться на содержимое документа