вход по аккаунту


Immunohistologic and cytochemical studies of temporal arteritis.

код для вставкиСкачать
Arteritic lesions from 14 patients with temporal
arteritis were studied by cytochemical and immunohistochemical methods to identify the nature of the inflammatory cell infiltrate and to demonstrate immunoglobulin deposition. The infiltrating cells typically seen were
histiocytes, giant cells, monocytes, and lymphocytes.
The lymphocytes were mainly T cells and the majority of
them were of the helperhnducer (Leu3a) T cell subset,
as indicated by monoclonal antibody reactivity.
Immunoglobulin deposits on the internal elastica were
identified in only 6 patients and neutrophils were either
very sparse or absent. These findings are consistent with
the hypothesis that cell-mediated immunity may be
important in the pathogenesis of temporal arteritis.
Temporal arteritis (giant cell arteritis) is a
chronic vasculitis which afflicts persons over the age
of 50 years. Its cause is unknown and its pathogenesis
From the Departments of Pathology and Internal Medicine,
Mayo Clinic/Mayo Foundation, Rochester, Minnesota.
Supported in part by gifts from Mrs. Dorothy Zinn and Mr.
and Mrs. Nathan Gumenick.
Peter M. Banks, MD: Assistant Professor of Pathology,
Mayo Medical School and Consultant, Division of Surgical Pathology, Department of Pathology; Marc D. Cohen, MD: Senior Clinical
Fellow, Division of Rheumatology and Internal Medicine. Department of Internal Medicine; William W. Ginsburg, MD: Assistant
Professor of Medicine, Mayo Medical School and Consultant,
Division of Rheumatology and Internal Medicine, Department of
Internal Medicine; Gene G. Hunder, MD: Professor of Medicine,
Mayo Medical School and Consultant, Division of Rheumatology
and Internal Medicine, Department of Internal Medicine.
Address reprint requests to Peter M. Banks, MD, c/o
Section of Publications, Mayo Clinic, Rochester, MN 55905.
Submitted for publication January 18, 1983; accepted in
revised form May 18, 1983.
Arthritis and Rheumatism, Vol. 26, No. 10 (October 1983)
is poorly understood. Pathologic examination has disclosed a patchy or continuous distribution which may
involve the proximal aorta and major elastic tissuecontaining arteries of the head, neck, and upper extremities (1-4).Microscopy of affected arteries reveals
granulomatous inflammation concentrated in the region of the internal elastic lamina, which is focally or
multifocally disrupted (1-4). The intimate intermixture
of mononuclear leukocytes, histiocytes, and giant cells
in zones of fragmented elastica has suggested a pathogenetic role for cell-mediated immunity. However,
prior studies of cellular immunity in temporal arteritis
have produced inconclusive results (5,6). Other reports of the presence of extracellular and intracellular
deposits of immunoglobulins and complement (7) in
temporal arteries and circulating serum immune complex-like material (8) have indicated the possible importance of humoral immune processes in the pathogenesis of this condition.
With the recent availability of highly specific
monoclonal antibodies directed against human cell
surface antigens, direct evaluation of intact arteritic
lesions with respect to T cell and T cell subset composition has become feasible. In the present investigation, inflamed, biopsied temporal arteries from 14
patients with giant cell arteritis were examined using a
combination of cytochemical and immunohistochemical modalities, in an attempt to characterize the cellular composition of the process. Additionally, peripheral blood from 8 of the 14 patients was studied by flow
cytophotometric analysis using monoclonal antibodies
equivalent to those utilized in the immunopathologic
methods, to immunologically measure circulating
mononuclear cell fractions.
Patients and temporal artery specimens. Temporal
artery specimens showing giant cell arteritis on frozen
section were obtained from 14 patients during a 10-month
period spannihg 1981 and 1982. Blocks of the intact artery
specimen adjacent to the portion which showed arteritis by
diagnostic intraoperative frozen section were snap-frozen in
liquid nitrogen, after being coated with hyperviscous aqueous mounting medium to prevent storage desiccation. Samples were stored at -70°C and examined later as a group.
Frozen sections from special studies were cut in a cryotome
at 4p thickness.
The inflammatory arterial lesions were also studied
by conventional methods both with thawed tissues from
intraoflerative frozen section samplings and from intact
adjacent tissue blocks. These were formalin-fixed, routinely
sequenced for paraffin embedding, and stained with hematoxylin-eosin and with a Giemsa stain (E. Merck, Darmstad,
West Germany).
Cytochemistry. Frozen sections were air-dried and
Subsequently fixed for 30 seconds in cold (4°C) formolacetone solution (25% formalin, 45% acetone in phosphate
buffer pH 6.6). They were then exposed to respective
substrates for detection of cellular enzyme activity according to previously published methods (9,lO). These were used
to detect inflammatory cell types as follows: acid phosphatase-intense diffuse staining in macrophages (histiocytes)
and monocytes (focal staining in some T lymphocytes);
naphthol AS-D chloroacetate esterase-so-called “specific”
esterase in neutrophilic granulocytes and mast cells; alphanaphthyl butyrate esterase-so-called “nonspecific” esterase in macrophages (histiocytes) and monocytes, exclusively.
Immunohistochemistry. Frozen sections were airdried and subsequently fixed for 5 minutes in acetone (20°C).
The 2-step peroxidase-conjugation method of McMillan et al
(1 1) was used for color labeling of commercially obtained
mouse hybridoma monoclonal antibodies (Becton-Dickinson, Sunnyvale, CA). Antibodies directed against the following human antigenic determinants were used: Leu-I (pan-T
cell), Leu-2a (suppressorkytotoxic T cell), Leu-3a (helper/inducer T cell), HLA-DR (Ia-like major histocompatibility
antigen), kappa light chain (more prevalent of the two human
immunoglobulin light chain types).
For purposes of morphologic comparison, frozen
sections from each patient were also formalin-fixed and
stained with hematoxylin-eosin and with Giemsa stain.
Blood studies. Heparinized whole blood samples
were obtained from 8 of the patients prior to corticosteroid
therapy. Mononuclear cell fractions were derived from Ficoll-Hypaque density gradient centrifugation. Cytofluorographic analysis of cell suspensions was performed with
direct immunofluorescence labeling using fluorescein-conjugated monoclonal antibodies that were identical to those
applied in immunohistochemical procedures, with the following exceptions: Leu-4 (Becton-Dickinson)was substituted for Leu-I as a pan-T cell determinant, and anti-kappa
light chain was not used. Suspension specimens were analyzed with an automated fluorescence-activated cell sorter
(FACS) (Becton-Dickinson).
Clinical. Important clinical features and the
results of selected laboratory tests are listed in Table
1. Among the 14 patients studied, initial symptoms
were those of polymyalgia rheumatica in 6, malaise in
5 , headaches in 2, and temporal artery tenderness in 1.
Twelve of the 14 had either symptoms or signs related
to temporal arteritis prior to biopsy, while in 2 patients
clinical manifestations were limited to polymyalgia
rheumatica only. Only a single patient suffered permanent visual loss. Only 2 patients had sedimentation
rates less than 80 mmlhour. A final diagnosis of
polymyalgia rheumatica was made in 9 cases.
Six of the 14 patients had received corticosteroid therapy prior to biopsy. Four of these had received high-dose (60 mg) prednisone daily for an
average of 4 days before biopsy; the 2 others had
received 4 mg and 20 mg prednisone daily, respectively, for more than 4 weeks before biopsy. All 6 of these
patients showed clinical manifestations of active disease at the time of biopsy.
Histopathology. In all cases paraffin and frozen
section preparations showed a remarkably consistent
and distinctive arteritic process. An inflammatory
infiltrate was concentrated most in the inner half of the
media centering upon the internal elastica, but a more
dispersed infiltration continued through the outer half,
so that a transmural distribution with adventitial involvement was observed in 10 of the 14 patients.
Luminal constriction or obstruction was due to massive edema of the intima, which itself was devoid of
inflammatory cells (Figure 1). So-called fibrinoid necrosis was identified in 4 patients. It consisted of zones
of eosinophilic acellular material deposited within inflammatory infiltrate of the media.
Table 1. Clinical features of 14 patients with temporal arteritis
Age, mean (range)
Symptoms related to arteries, no.
Signs related to arteries, no.
Estimated duration of disease at biopsy, mean
months (range)
Presence of temporal artery symptoms or findings
at time of biopsy, no.
Polymyalgia rheumatica, no.
Receiving corticosteroids at time of biopsy, no.
Hemoglobin, gm/dl, mean (range)
Erythrocyte sedimentation rate, mm/hour, mean
73 (59-85)
6 ( I /2-30)
11.2 (9.3-13.0)
95 (30-136)
Figure 1. Temporal arteritis. Total occlusion of artery due to intima1 proliferation and edema (black arrow); typical distribution of
inflammation in media and adventitia (white arrows) (hernatoxylin
and eosin, original magnification x 50).
Elastic lamellae appeared light azure blue with
Giemsa stain and were readily discerned. Disruption
of the internal elastica was multifocal in 8 patients and
widespread (circumferential) in 6. Fragments of elastics could be identified within giant cells of 5 patients
(Figure 2). In the medial regions of dense cellular
infiltration there was a predominance of histiocytes
and giant cells over small mononuclear cells, whereas
the adventitial infiltrates consisted almost exclusively
of small mononuclear cells. Plasma cells, when present, were found in only the adventitial region (Table
2 ) . Eosinophilic granulocytes were present in the medial lesions of 3 patients. By direct
staining methods, no neutrophilic (polymorphonuclear) granulocytes could be detected in any of the
lesions, although in areas of inflammatory cell necrosis, nuclear fragments suggested the appearance of
leukocytoclasis, which is often associated with neutrophilic accumulation. Giant cells were demonstrated in
immediate proximity to the internal elastica.
Cytochemistry. Chloracetate esterase (CLE)
confirmed the absence of neutrophilic granulocytes
from the inflammatory lesion (Table 2 ) . In 3 cases,
small numbers of neutrophils were identified within
small vascular spaces. In regions of cell necrosis with
nuclear fragmentation, not even faint CLE positivity
was seen, although such positivity is usually demonstrable in suppurative processes with leukocytoclasis.
Nuclear fragments may instead have derived from the
degeneration of lymphocytes. CLE demonstrated
small to moderate numbers of mast cells in the outer
adventitia, as are normally present in arteries. Scant
infiltration of mast cells was present in the inflammatory lesion of only 1 patient (Figure 3).
Acid phosphatase (AP) was found to be a more
useful marker of mononuclear-phagocyte cell types
than butyrate esterase, due to strong activity of the
latter enzyme among smooth muscle cells of the
media. Strong, diffuse AP positivity was present
among large numbers of histiocytes and giant cells
(Figure 4). A small proportion of intensely staining
small mononuclear cells, presumably monocytes, was
present in both media and adventitia.
Immunohistochemistry. Immunoperoxidase hybridoma antibody preparations consistently showed a
predominance of T lymphocytes among the inflammatory infiltrates of both medial and adventitial regions,
and a majority of the T lymphocytes were helper/
inducer T cells (Leu-3a positive) (Table 3). Leu-1
antigen was demonstrated on the surface of a majority
Table 2. Cytologic and cytochemical analysis of 14 temporal
arteries (no. in which cell tvDe mesent)
Figure 2. Elastic fragment present within giant cell (arrow). This
appears light azure blue in this Giemsa-stained section (original
magnification x 640).
Cell type
Giant cells
Pol ymorphs
Mast cells in adventitia
Mast cells in inflammatory lesion
Plasma cells
Table 3. Immunochemical analysis of 14 temporal arteries (no. in
which cell type present)
Proportion of total
mononuclear cells
T cell type
Pan-T cells (Leu-1)
Helperhnducer (Leu-3a)
Suppressorkytotoxic (Leu-2a)
B cells
Figure 3. Chloroacetate esterase stain showing mast cells (but no
neutrophils) predominantly in adventitia (patient 7) (original magnification x 100).
of mononuclear cells in all cases, and on greater than
75% of the cells in 8 of 14 cases (Table 3). In all cases
there was a predominance of helper over suppressor T
cells, as indicated by Leu-3a and Leu-2a staining
respectively. Precise quantitation could not be calculated due to varying intensity of staining among cells.
However, careful examination of the entire cellular
population of arteritic lesions, with comparison of
identical microscopic fields within the same tissue
region, permitted recognition of consistent predominance (approximately 3:2 to 3: 1 ratio) of Leu-3a to
Figure 4. Acid phosphatase stain showing histiocytes and monocytes. Note giant cells along elastica (arrow) (original magnification
X 250).
75-100% 50-75% 25-5W0 0-25%
Leu-2a positive cells (Figures 5,6,7). Since Leu-3a
reacts also with cytoplasmic antigens among some
histiocytes, it was not an unexpected observation that
faint, diffuse staining of macrophages and of giant cells
was also noted with this reagent. No differential distribution or separate intramural grouping of helper or
suppressor T cells could be detected. In general, T
lymphocytes were segregated within the medial lesions to more peripheral zones surrounding central
clusters of histiocytes and giant cells (Figure 5 ) .
Staining for immunoglobulin (kappa light chain)
was ubiquitously present as scattered foci and strands
within the inflammatory lesions. A regular line of
immunoglobulin staining of moderate intensity along
the surface of the internal elastica was identified,
conspicuous in 2 cases and faint in 4 others (Figure 8).
Cellular staining for immunoglobulin was generally
faint and dispersed, corresponding to incorporation of
Figure 5. Temporal arteritis specimen treated with anti-Leu-1
monoclonal antibody in peroxidase technique, showing T cells
surrounding zone of histiocytes (original magnification X 250).
Figure 6. Anti-Leu-2A monoclonal antibody showing, with peroxidase technique, that a minority of lymphocytes are presumptive
suppressor/cytotoxic T cells (same field as Figure 5) (original
magnification x 250).
Figure 8. Temporal arteritis section treated with anti-kappa light
chain monoclonal antibody showing, with peroxidase technique,
deposition of immunoglobulin (arrows)on internal elastica (original
magnification x 250).
plasma globulins by the cells of the mononuclearphagocyte system. A few cells showed strong cytoplasmic positivity, corresponding to plasma cells. Surface staining was present among only a small
percentage of mononuclear cells, and the significance
of such staining was unknown, since monocytes are
capable of immunoglobulin coating by means of avid
Fc fragment receptors (12). The proportion of B lymphocytes could only be inferred indirectly by exclusion, i.e.. the remainder of small mononuclear cells
not staining for T cell (Leu-1) antigen or for monocyte
(AP) enzyme activity. Only a small proportion (less
than 25%) of lymphoid cells were interpreted as being
of possible B cell nature by this method (Table 3).
HLA-DR staining was not specific and was
present among histiocytes, giant cells, and a large
proportion of mononuclear cells (presumed monocytes, B cells, activated T cells). Furthermore, it
showed staining along the internal elastica, identical to
that seen for immunoglobulin, among the positive
No difference in either the severity or cellular
composition of the inflammatory lesions could be
perceived between patients who had received prebiopsy corticosteroid therapy and those who had been
untreated, nor between those whose symptoms were
of shorter or longer duration.
Blood studies. All blood samples from the 8
untreated patients showed absolute T cell percentages
(64-88%) and proportions of helperhppressor T cells
(1.1-1.8) within normal range. The percentage of
mononuclear cells positive for HLA-DR ranged from
24-4 1.
Figure 7. Anti-Leu-3a monoclonal antibody showing, with peroxidase technique, that the majority of lymphocytes are presumptive
helperlinducer T cells (same field as Figures 5 and 6) (original
magnification x 250).
The present study was performed to further
characterize the nature of the inflammatory reaction in
temporal arteritis. We found in our specimens, as
others have previously reported (1-4,13,14), a relatively uniform histologic picture of a granulomatous
inflammatory process throughout the vessel wall, centered in the region of the internal elastic lamina. Of
particular interest was the cytochemical demonstration of the absence of neutrophilic granulocytes.
With the use of labeled monoclonal antibodies
which were specific for human T cells, it was possible
to demonstrate that the majority of infiltrating mononuclear cells were T cells and that these were predominantly Leu-3a positive, i.e., presumptive helperhnducer T. These T cell findings were not, however,
manifest in peripheral blood from 8 untreated patients,
in which normal ratios of T to B lymphocytes were
Although morphologic evidence has suggested
cell-mediated mechanisms in giant cell arteritis, lymphocyte studies to date have failed to document a
specific role for cellular immunity in giant cell arteritis.
Peripheral blood lymphocyte functions were normal,
as measured by uptake of tritiated thymidine after
exposure to common infectious antigens, and did not
show a consistently increased response to arterial or
muscular tissues when compared with normal specimens (5,6).
Other reports have suggested that humoral immune mechanisms may be important in the development of temporal arteritis. Using immunofluorescence
techniques, Liang and coworkers (7) and other groups
(15,16) identified deposits of immunoglobulins and
complement in temporal artery specimens from some
patients with temporal arteritis and polymyalgia rheumatica. However, a recent study by Gallagher and
Jones (17) suggests that such deposits are an uncommon finding. These authors used the immunoperoxidase method and found varying amounts of IgA, IgM,
and IgG in scattered plasma cells and macrophages in
temporal arteritis specimens, but extracellular IgG
deposits were present in only 1 of 15 patients with
active arteritis. No complement staining was noted.
Some of the differences may be explained by the
sensitivity of the reagents used and interpretation of
the findings. Our results are consistent with those of
Gallagher and Jones.
Although dispersed immunoglobulin deposition
was present in zones of inflammation among all our
patients, we interpreted this as the result of plasma
diffusion through porous endothelium rather than as a
primary directed deposition related to the pathogenesis of the arteritis. However, selective immunoglobd i n staining along the surface of the internal elastica,
which was dense in 2 patients and faint in 4 others, was
also found. The meaning of these deposits remains
In this study we used both standard methods
and newer monoclonal antibody staining techniques to
characterize the nature of the inflammatory reaction in
temporal arteritis. The predominance of T cells and
lack of polymorphonuclear leukocytes suggests a cellmediated reaction. Further morphologic and functional studies are needed for insight into the exact pathogenetic processes. Similar methods applied to lesions
from very early stages of disease may be particularly
1. Wilkinson IMS, Russell RWR: Arteries of the head and
neck in giant cell arteritis: a pathological study to show
the pattern of arterial involvement. Arch Neurol27:378391, 1972
2. Klein RG, Hunder GG, Stanson AW, Sheps SG: Large
artery involvement in giant cell (temporal) arteritis. Ann
Intern Med 83:806-812, 1975
3. Parker F, Healey LA, Wilske KR, Odland GF: Light
and electron microscopic studies on human temporal
arteries with special reference to alterations related to
senescence, atherosclerosis and giant cell arteritis. Am J
Pathol 7957-80, 1975
4. Klein RG, Campbell RJ, Hunder GG, Carney JA: Skip
lesions in temporal arteritis. Mayo Clin Proc 51504-510,
5. Papaioannou CC, Hunder GG, McDuffie FC: Cellular
immunity in polymyalgia rheumatica and giant cell arteritis: lack of response to muscle or artery homogenates.
Arthritis Rheum 22:740-745, 1979
6. Zilko P, Currey HLF, Vernon-Roberts B: Polymyalgia
rheumatica (PMR) and giant cell arteritis: lack of response of peripheral blood lymphocytes to arterial wall
homogenate. Ann Rheum Dis 36:286-287, 1977
7. Liang GC, Simkin PA, Mannik M: Immunoglobulins in
temporal arteries: an immunofluorescent study. Ann
Intern Med 81:19-24, 1974
8. Papaioannou CC, Gupta RC, Hunder GG, McDuffie FC:
Circulating immune complexes in giant cell arteritis and
polymyalgia rheumatica. Arthritis Rheum 23: 1021-1025,
9. Li CY, Lam KW, Yam LT: Esterases in human leukocytes. J Histochem Cytochem 2l:l-12, 1973
10. Yam LT, Li CY, Crosby WH: Cytochemical identification of monocytes and granulocytes. Am J Clin Pathol
55~283-290, 1971
11. McMillan EM, Wasik R , Everett MA: Identification of
T-lymphocytes and T-subsets in human tonsil using
monoclonal antibodies and the immunoperoxidase technic. Am J Clin Pathol 76:737-744, 1981
12. Rowlands DT Jr, Daniele RP: Surface receptors in the
immune response. N Engl J Med 293:26-32, 1975
13. Hamilton CR Jr, Shelley WM, Tumulty PA: Giant cell
arteritis: including temporal arteritis and polymyalgia
rheumatica. Medicine (Baltimore) 50: 1-27, 1971
14. Ostberg G: On arteritis with special reference to polymyalgia artentica. Acta Pathol Microbiol Scand [A]
(suppl) 237: 1-59, 1973
15. Park JR, Hazleman BL: Immunological and histological
study of temporal arteries. Ann Rheum Dis 37:238-243,
16. Plouvier B, Francois M, Wattre P, Francois P, Devulder
B: Examen en immunofluorescence directe de coupes
d’artere temporale: interet et limites (rapport preliminaire). Nouv Presse Med 7:1719-1721, 1978
17. Gallagher P, Jones K: Immunohistochemical findings in
cranial arteritis. Arthritis Rheum 25:75-79, 1982
Без категории
Размер файла
767 Кб
arteritis, temporal, studies, cytochemical, immunohistological
Пожаловаться на содержимое документа