вход по аккаунту


Stimulation of neovascularization by human rheumatoid synovial tissue macrophages.

код для вставкиСкачать
47 1
Synovial tissue from patients with rheumatoid
arthritis was enzymatically dissociated, and single cell
suspensions were fractionated into subpopulations by
centrifugation on continuous Percoll gradients. Five
fractions (Fl-F5) with densities of 0.991-0.998 gndml,
0.998-1.042 gm/ml, 1.042-1.062 gndml, 1.062-1.082
gmlml, and 1.082-1.180 gndml, respectively, were prepared. F3 consistently contained the highest number of
macrophages, while F2 and F4 contained substantially
fewer macrophages. Macrophages present in F2, F3,
and F4 were enriched by differential adherence to
fibronectin-coated collagen gels. These macrophageenriched cell preparations were found to be Fc and C3
p i t i v e , esterase positive, and peroxidase negative, to
stain positively with anti-HLA-DR, anti-Leu-M3,
OH;Ml, and OKM5 monoclonal antibodies, and to show
characteristic features of macrophages by electron microscopy. Macrophages from F3 consistently induced
neovascularization in rat corneas, while equal numbers
Presented in part at the 49th Annual Meeting of the Ameri c a ~Rheumatism Assocation, Anaheim, CA, June 1985.
From the Department of Internal Medicine, Section of
Arthritis and Connective Tissue Diseases, the Department of Pathology, and the Department of Oral Biology, Connective Tissue
Biology Laboratories, Northwestern University, Chicago, Illinois.
Supported by USPHS grant GM-29135 and a grant from the
3M Foundation. Dr. Koch was supported by a Schoil Foundation
Felllowship from the Illinois Chapter of the Arthritis Foundation.
Alisa E. Koch, MD: Department of Internal Medicine,
Section of Arthritis and Connective Tissue Diseases; Peter J.
Polverini, DDS, DMSc: Department of Pathology; S. Joseph
Leibovich, PhD: Department of Oral Biology, Connective Tissue
Biology Laboratories.
Address reprint requests to Dr. S. J. Leibovich, Department of Oral Biology, Northwestern University, Ward Building,
13-007, Chicago, IL 60611.
Submitted for publication October 26, 1984; accepted in
revised form October 4, 1985.
Arthritis and Rheumatism, Vol. 29, No. 4 (April 1986)
of macrophages from F2 and F4 did not. Fibroblastic
synovial cells and cells that did not adhere to fibronectin-coated collagen gels did not induce neovascularization. Within the rheumatoid synovium, there appears to be a major subpopulation of macrophages
capable of inducing neovascularization, a process vital
to the development of the rheumatoid synovial pannus.
Rheumatoid arthritis (RA) is a systemic disease; its primary lesions are characterized by proliferation and thickening of the synovial lining membrane
of diarthrodial joints. The resulting hyperplastic
synovial pannus invades and degrades cartilage and
bone (1) and is largely responsible for the irreversible
destruction of diarthrodial joints often seen in RA
(2-4). The rheumatoid synovium contains many different cell types, including synovial cells, fibroblasts,
lymphocytes, monocytes-macrophages, and endothelial cells of blood vessels (2). Monocytes and
macrophages are believed to play an important role in
mediating joint destruction (2). Macrophages have also
been shown to mediate neovascularization-the process of new blood vessel growth-in wound repair,
inflammation, and solid tumor growth (5-9). Neovascularization is a particularly prominent feature of the
developing rheumatoid synovial lesion (10). In this
study, we examined the ability of various cell populations isolated from human rheumatoid synovial tissues
to induce neovascularization in an in vivo model
Tissue disaggregation. Fresh samples of rheumatoid synovial tissue were obtained during total joint replacement (16 knees, 1 hip) from 17 patients who had classic RA
that met the diagnostic criteria of the American Rheumatism
Association (1 1). The synovial lining tissue was dissected
and minced into 2-3-mm pieces. Fifty milliliters of an
enzyme solution containing 2.4 mg/ml dispase (grade 11, 5
units/mg; Boehringer-Mannheim, Indianapolis, IN), 1 mgiml
collagenase (type 11, 200-400 d m g ; Cooper Biomedical,
Freehold, NJ), and DNase (type I, 10,000 dornase units/ml;
Calbiochem-Behring, San Diego, CA) in Ca" free and
free Hanks' balanced salt solution (Gibco, Grand
Island, NY) was used per 2 gm of tissue. The fragments were
stirred gently for 1 hour at 37°C.
A second digest was carried out with fresh enzyme
solution (12). The cells were filtered through a 110p Nitex
mesh (Tetco, Elmhurst, NY) and dispensed into centrifuge
tubes containing an equal volume of Dulbecco's modified
Eagle's medium (DMEM), supplemented with 5 pg/ml
gentamycin and 10% heat-inactivated fetal calf serum (FCS).
The filtered digests were pooled, and the cells were washed
3 times by repeated centrifugation and resuspension in fresh
DMEM and counted. Viability was determined by suspending cells in trypan blue solution (4 mg/ml); at least 300 cells
were counted in a counting chamber, and the percentage of
cells that excluded trypan blue was determined. Cytocentrifuge preparations were made from 100-pl aliquots of
suspended cells (1 x lo6 ml), using a Shandon Cytospin 2
Isopycnic density gradient fractionation of synovial
cell preparations. Cell suspensions were fractionated by
isopycnic centrifugation through continuous Percoll gradients (Pharmacia, Piscataway, NJ). Gradients were preformed in 10-ml centrifuge tubes by centrifuging 8.5 ml of
Percoll, density 1.08 g d m l , at 17,000 revolutions per minute
in a Sorvall RC5B centrifuge fitted with an SS34 rotor.
The top milliliter of the Percoll gradient was removed, and 4-5 x lo7 cells in 2 ml of phosphate buffered
saline (PBS) was added to each gradient. The tubes were
then centrifuged at 1,800 rpm at 4°C for 20 minutes in a
Beckman TJ6R centrifuge. The density of sedimenting cells
was monitored with density marker beads (Pharmacia) centrifuged in parallel.
The Percoll gradients were aspirated from the bottom
of the tubes into 5 2-ml fractions, washed 3 times with
DMEM, and counted; cytocentrifuge preparations were then
made. The percentage of macrophages in each fraction was
determined before further purification, using the functional
markers of Fc- or complement (C3)-mediated phagocytosis,
as described below.
Determination of receptors for the Fc portion of IgG.
Cells (1 x lo6) from each fraction were placed in 35-mm petri
dishes (Falcon, Oxnard, CA) and allowed to adhere to
thermanox coverslips (Flow Laboratories, McLean, VA)
during a 1-hour incubation at 37°C. Coverslips were washed
3 times with DMEM, and 2 x lo7 sheep red blood cells
opsonized with anti-sheep cell antibody (7s IgG; Cordis
Laboratories, Miami, FL) was added to the adherent cells in
1 ml of DMEM-10% FCS. Coverslips were incubated for 1
hour at 37"C, washed 3 times with DMEM, and fixed and
stained with Diff-Quik staining solutions (American Scientific Products, McGaw Park, IL). At least 300 cells were
counted, and the percentage of adherent cells which en+
gulfed 3 or more sheep red blood cells was determined and
expressed as a percentage of total cells counted.
C3-mediated phagocytosis of yeast. Cells (1 x lo6)
from each fraction were allowed to adhere to thermanox
coverslips as above. Heat-killed yeast cells (1.5 x lo7)
opsonized by incubation with 10% FCS for 30 minutes were
added in 1 ml of DMEM, incubated for 1 hour at 37"C,
washed 3 times with DMEM, fixed, and stained with DiffQuik solution. At least 300 cells were counted, and the
percentage of cells ingesting yeast was determined as above.
Enrichment of macrophages from Percoll gradient
fractions. After the marker studies described above were
performed, macrophages were separated from other cell
types by adherence to fibronectin-coated collagen gels, as
previously described (12). The gels were prepared by dispensing neutralized sterile type I collagen (Vitrogen; Flow
Laboratories) into 100 x 20-mm tissue culture dishes (Falcon) to form a thin layer. The plates were incubated at 37°C
for 1 hour to allow gel formation. After heat-inactivated calf
serum (4 ml; Biologos, Naperville, IL) as a source of
fibronectin was added, the gels were incubated for 1 hour.
The plates were then washed with PBS to remove excess
serum. Cell fractions from the Percoll gradients were added
to the collagen gels and allowed to adhere overnight at 37°C
in an incubator gassed with 5% COz, 95% air. This procedure
resulted in attachment of macrophages and some remaining
nonmacrophage cells (fibroblasts, endothelial cells, etc.) to
the gels. Some cells did not adhere to the gels, and these
were collected and used as described below.
To remove adherent nonmacrophage cells, gels were
treated briefly (5-10 minutes) with trypsin-EDTA (Gibco).
Macrophage-enriched preparations remaining on the gels
were then harvested intact by treatment with 0.1%
collagenase (type I; Cooper Biomedical) at 37°C to dissolve
the collagen gels. The macrophage-enriched cell preparations were washed 3 times with DMEM, resuspended in
PBS, and injected into rat corneas, as described below, for
assessment of angiogenic activity. Cytocentrifuge preparations of the cells were made and were used to characterize
the cells by: 1) histochemical markers and 2) monoclonal
antibody staining for the presence of monocytemacrophage-related surface antigens, as described below.
Histochemical determination of nonspecific esterase
activity. Nonspecific esterase activity was determined with a
histozyme kit (Sigma, St. Louis, MO), using cytocentrifuge
cell preparations. The substrate used for esterase activity
was a-naphthyl acetate.
Histochemical determination of peroxidase activity.
Peroxidase activity was determined with a histozyme kit
(Sigma), using cytocentrifuge cell preparations, as described
by the manufacturers.
Staining with monoclonal antibodies. Cytocentrifuge
preparations were washed 3 times with DMEM-2% FCS,
0.1% sodium azide. Fluorescein-conjugated anti-HLA-DR
or anti-Leu-M3 antibodies (Becton-Dickinson, Mountain
View, CA) were suspended at a 1:lO dilution in DMEM-2%
FCS, 0.1% sodium azide, and applied to the slides, which
were incubated in a humidified slide chamber at 4°C for 30
minutes. The slides were then washed 3 times with
DMEM-2% FCS, 0.1% sodium azide, mounted, and examined with a Leitz Dialux 20 epifluorescence microscope.
Slides were also stained with fluorescein-conjugated
OKMS monoclonal antibody (Ortho, Raritan, NJ), as described above. In addition, slides were stained with
unconjugated OKMl monoclonal antibody, which was then
visualized using a fluorescein-labeled goat anti-mouse IgG
second antibody (Ortho).
Preparation of rheumatoid fibroblastic synovial cells.
Enzymatically dispersed synovial cell preparations were
plated at a concentration of lo6 cells/lO ml of DMEM-10%
FCS in a humidified incubator gassed with 5% COz, 95% air.
The cells were re-fed twice weekly until a confluent monolayer was formed, then they were trypsinized and subcultured at a split ratio of 1:4.
Preparation of nonadherent rheumatoid synovial cells.
After overnight incubation to allow adherence of macrophages and other adherent cells to the fibronectin-coated
collagen gels, nonadherent cells were aspirated from the
dishes, washed 3 times with DMEM, and counted. Viability
was determined by measuring the percentage of cells which
excluded trypan blue.
Corneal bioassay for neovascularization. Neovascularization was assayed in the normally avascular corneas of
100-15O-gm rats by introducing 5 x lo5 cells in 10 p1of PBS
into the corneal stroma 1-1.5 mm from the limbus, as
previously described (8,12). This procedure, first developed
by Gimbrone et a1 (13), has been used extensively as an
assaiy procedure for studying angiogenesis (9,14,15). Corneas were examined daily with a stereomicroscope to monitor capillary growth. Seven days after injection, rats were
perfused intraarterially with colloidal carbon (Pelican, Inc.,
Hanover, West Germany), which permeated and filled the
blood vessels within the corneas, thus allowing a permanent
record of individual vascular responses to be made.
The rats were then killed and corneas were excised,
fixed in 2% gluteraldehyde, flattened, and photographed.
Responses were recorded as positive when unidirectional
suslained growth of a brush-like network of capillary sprouts
and hairpin loops was observed. Controls consisted of
corneal injections of boiled, frozen, and thawed cells, and
Histologic examination. Samples of synovial tissue
fixed in formalin were dehydrated and embedded in paraffin,
sectioned, and stained with hematoxylin and eosin.
Nonperfused corneas from rat eyes injected with cells were
also fixed, dehydrated through graded ethanol solutions, and
embedded in glycol methacrylate (Polysciences, Warrington, PA). Sections ( 1 . 5 ~ )of corneas were cut through
vascularized areas, using a Sorvall JB-4 microtome, and
stained with methylene blue and basic fuchsin.
Preparation of cells for electron microscopy. Macrophage-enriched cell preparations removed from the
fibronectin-coated collagen gels were washed 3 times with
DMEM, fixed at room temperature for 1 hour with
Karnofsky’s fixative containing 1% formaldehyde-3%
glutaraldehyde in 0.1M cacodylate buffer, pH 7.4 (16), and
postfixed for 30 minutes at 4°C with 1% osmium tetroxide in
cacodylate buffer. The cell preparations were then dehydrated to 100% by passage through graded ethanol solutions
andl embedded in Epon (17).
Thin sections were cut using a Sorvall MT-2 ultramicrotome, mounted on uncoated copper grids, and stained
with uranyl acetate and lead citrate. These sections were
examined in a Hitachi HU12 electron microscope at an
operating voltage of 75 kV.
Gross histology of the samples. Histology of the
tissue samples was typical of rheumatoid arthritis. The
tissues showed a proliferative synovitis, with a heavy
mononuclear cell infiltrate, as well as the presence of
numerous blood vessels. Generally, very little fibrosis
was present.
Distribution of cells on Percoll gradients. In
general, between 2.3 X lo8 and 9.9 X lo8 cells were
isolated from each synovium and applied to the Percoll
gradients and centrifuged. From 50-80% of cells were
recovered from the gradients, which is in accord with
other authors’ data on the recovery of cells from
Percoll gradients (18). The viability of the cells recovered was found to be >99% by trypan blue exclusion.
Five fractions were obtained, with density
ranges as follows: fraction 1 (Fl), 0.991-0.998 gm/ml;
F2, 0.998-1.042 gm/ml; F3, 1.042-1.062 grdml; F4,
1.062-1.082 gm/ml; and F5, 1.082-1.180 gm/ml. F5
contained mainly red blood cells. F1 contained mainly
cellular debris and very few cells. Thus, most of the
cells were found in F2, F3, and F4.
Functional marker studies on cell populations
from Percoll. The results of the opsonized IgG sheep
red blood cell Fc-mediated phagocytosis marker studies on these 3 fractions before further purification of
macrophages by adherence to fibronectin-coated
collagen gels are summarized in Table 1. The yeast
marker studies for C3-mediated phagocytosis yielded
similar results and are not reported here. Overall, F3
contained the greatest proportion of cells as well as the
greatest percentage of macrophages. F3 thus constituted the major fraction of cells isolated from the
rheumatoid synovial tissues.
Enrichment of macrophages on fibronectincoated collagen gels. This procedure has previously
been utilized to purify macrophages from single cell
suspensions of a solid tumor (12). Several cell types,
including macrophages , fibroblastic cells, and endothelial cells, adhered to the gels, while others did not. The
nonadherent, nonmacrophage cells were removed and
used in the angiogenesis assay. Adherent, nonmacrophage cells were selectively removed from the gels by
trypsinization, which detached fibroblasts and endothelial cells from the gels while leaving macrophages
firmly attached. Dissolution of the collagen gels with
Table 1. Percentage of cells and of macrophages in fractions from Percoll gradients before
enrichment of macrophages by adherence to fibronectin-coated collagen gels*
Fraction F2
Fraction F3
% macro-
% macro-
% cells
% cells
Fraction F4
% macro-
% cells
* Determined by
assessing phagocytosis of opsonized sheep red blood cells. Fraction F2 = 0.9981.042 g d m l ; fraction F3 = I .042-1,062 g d m l ; fraction F4 = 1.062-1.082 g d m l .
clostridial collagenase released the macrophages from
the gels. These cells, prepared from F2, F3, and F4
from the Percoll gradients, were then characterized
using light and electron microscopy, and enzyme- and
immunohistochemistry. More than 90% of the cells
were found to have the characteristics of macrophages, as described below. These cells were then
used for assay of angiogenic activity.
Functional, enzyme, and immunohistochemical
marker studies on macrophage-enriched cell populations. The results of the functional marker studies for
receptors, the histochemical staining with esterase and
peroxidase, and the immunohistochemical monoclonal
antibody marker studies on the cell populations released from collagen gels by collagenase treatment are
shown in Table 2. No morphologic differences were
seen among the various macrophage-enriched cell
preparations of different density from F2, F3, and F4.
All possessed Fc and C3 receptors, were moderately
esterase positive, and did not stain for peroxidase. The
cells stained with anti-Leu-M3, OKM1, and OKM5
monoclonal antibodies and stained strongly with antiHLA-DR monoclonal antibody. These histochemical
Table 2. Functional, immunochemical, and histochemical characteristics of purified macrophage subpopulations*
AntiAntiEster- PeroxFraction Fc HLA-DR Leu-M3 OKMl OKM5 ase
++ ++
++ ++
++ ++
++ = strongly positive; + = positive; - = negative; ND = not
and immunochemical staining characteristics clearly
indicate that these cells are of monocyte-macrophage
Electron microscopy. Electron microscopic examination of macrophage-enriched cell populations
showed predominantly cells with morphologic characteristics typical of active macrophages. The cells contained numerous cytoplasmic-dense bodies, a sparse
rough endoplasmic reticulum, and an extensive network of plasma membrane invaginations and pseudopodia. A typical cell from F3 is shown in Figure 1.
Cells from F2 and F4 showed similar morphology.
Corneal neovascularization. Table 3 demonstrates the cumulative results obtained using the rat
corneal neovascularization assays to assess the angiogenic potentials of cell and control preparations.
Macrophage-enriched cell populations from F3
(1.042-1.062 g d m l ) stimulated angiogenesis consistently, while those from F2 and F4 individually or in
combination did not. These cell populations from F2,
F3, and F4 consisted of >90% macrophages by
morphologic, functional, and enzyme- and immunohistochemical marker studies. This suggests that the
angiogenic cell population resides exclusively in F3
and that a synergistic interaction between high- and
low-density cells was not sufficient to induce an angiogenic response.
In general, the corneal injections were performed using 5 x lo’ cells per 10 pl of PBS. In 2 of the
corneal injections, a lower concentration of macrophage-enriched cells from F3 was used. As few as 2.3
x lo5 cells induced an angiogenic response. Figure 2
shows a positive response obtained using cells in F3. A
brush-like network of capillary sprouts and hairpin
loops extending toward the cell implant is clearly seen.
Figure 2. Carbon-perfused whole mount of a rat cornea 7 days after
injection of macrophages from F3, showing a positive angiogenic
response. The site of the cell depot is indicated (arrow) (magnification x 18).
Figure 1. Electron micrograph of a typical cell isolated from F3
from the Percoll gradients. bv adherence to fibronectin-coated
collagen gels. Extensive surface membrane invaginations, dense
bodies, and membrane-bounded vacuoles typical of macrophages
are clearly seen (magnification x 7,750).
Figure 3 shows a negative response obtained
with cells from F2 (cell implant located at top of Figure
3). IVo evidence for a nonspecific inflammatory reaction was observed, either by gross visual examination
or by histologic examination of sections of corneas.
These results indicate clearly that a population of cells
from F3 containing >90% macrophages potently induced neovascularization, while similarly constituted
populations of cells from F2 and F4 did not.
Cultured rheumatoid synovial fibroblastic cells
(P&P3) did not induce an angiogenic response (Table
3). Likewise, nonadherent rheumatoid synovial cells,
Table 3. Neovascular responses induced by cell populations from rheumatoid synovial tissue
Cell population tested*
Macrophages from
F2 + F4
Cultured synovial
fibroblastic cells
Nonadherent cells from
PBS, sham implants
Boiled, frozen, and
thawed cells
Total no.
* F2 = 0.998-1.042 g d m l ; F3 = 1.042-1.062 g d m l ; F4 = 1.062-1.082 gdm l. PBS = phosphate
buffered saline.
t Eight corneas were tested; 1 showed evidence of nonspecific inflammation and was excluded.
Figure 3. Carbon-perfused whole mount of a rat cornea 7 days after
injection of macrophages from F2, showing a negative angiogenic
response. The site of the cell depot is indicated (arrow) (magnification x 18).
prepared by aspirating the cells which did not adhere
to fibronectin-coated collagen gels during an overnight
incubation, were found to be nonangiogenic. These
cells were viable, as indicated by exclusion of trypan
blue, but we have not characterized them further.
Control injections of boiled, frozen and thawed
cells, PBS, and DMEM also did not induce neovasculanzation. This suggests that the presence of viable
cells is required for the induction of an angiogenic
The primary lesion of rheumatoid arthritis is
characterized by hyperplastic thickening of the
synovial lining membrane of diarthrodial joints, with
proliferation of fibroblastic synovial cells and accumulation of large numbers of macrophages and other
mononuclear cells within the hyperplastic synovial
tissue. New capillary blood vessels are also a prominent feature of the early and developing synovial
pannus (10). The mechanisms underlying the development of this hyperplastic response and their relationship to the causative agent inducing the onset of RA
are poorly understood. Cell-mediated immune mechanisms have been implicated in the pathogenesis of this
disease, and it has also been suggested that the
synovial lesion in RA has many features in common
with invasive, nonmalignant tumors (19). Extensive
morphologic studies support this hypothesis of involvement of a tumor-like lesion in the pathogenesis of
RA (20,21), as do recent studies with the MRL/l mouse
strain, which exhibits a spontaneous, rheumatoid-like
arthritis (22).
Over the past decade, Folkman and coworkers
have established a model for solid tumor growth,
which postulates that solid tumors require the presence of blood vessels to establish their growth
(9,14,15,23). Thus, tumors appear to possess the capacity to produce angiogenesis factors, which induce
new capillary blood vessel growth toward the developing tumor. Once invaded by new blood vessels, the
nutrients, growth factors, and other constituents required for cellular proliferation are available, and
extensive proliferation of tumor cells takes place.
Recently, Polverini and Leibovich (12) demonstrated, using a transplantable, cloned, chemically
induced fibrosarcoma of rats, that much of the angiogenic activity of this solid tumor is attributable to the
presence of activated macrophages within the tumor.
It seems likely, therefore, that the neovascularization
that occurs in the development of solid tumors is
closely analogous to that which occurs as an integral
part of fibroproliferative processes, including wound
healing, inflammation, and as demonstrated in this
study, rheumatoid arthritis.
Despite the apparent importance of neovascularization in the development of the rheumatoid
synovial lesion, to our knowledge there have been few
studies on the angiogenic potential of synovial tissue
or its constituent cells. Brown et a1 (24) studied 8
rheumatoid synovial fluids from which the inflammatory cells had been removed and found that 2 of these
fluids contained a low molecular weight angiogenesis
factor, apparently identical to that derived from tumors. Subsequently, they extended these observations
to 23 patients with RA and found that 4 of these
samples contained angiogenic activity (25). The source
of the angiogenic activity was not determined in those
In our study, we have shown that synovial
tissue from patients with RA contains cells which are
potently angiogenic, as demonstrated by their ability
to induce neovascularization in the rat cornea model
system. Using isopycnic Percoll density gradients,
these cells are found in the fraction of density of
1.042-1.062 g d m l (F3), are adherent, and have the
characteristics of mature macrophages, as discussed
below. As can be seen from Table 1, the majority of
the cells, as well as most of the macrophages, were
contained in F3 (1.042-1.062 gmlml), and the angiogertic macrophage-enriched cell population was purified from this fraction by differential adherence to
fibronectin-coated collagen gels. In contrast, the same
doses of macrophage-enriched cells prepared from F2
(0.9!)8-1.042 g d m l ) and F4 (1.062-1.082 gm/ml) were
not angiogenic.
Despite this marked functional difference, we
have as yet been unable to demonstrate any phenotypic differences between the angiogenic and nonangiogenic macrophage-enriched cell populations. These
populations contained >90% cells which possessed Fc
and C3 receptors, stained for nonspecific esterase, but
did ;not stain for peroxidase, all of which are characteristic features of mature macrophages. These cells
stained with OKMl and OKM5, both antibodies to
monocytemacrophage cell surface constituents. They
also stained with anti-Leu-M3 monoclonal antibodies,
which were raised against peripheral blood monocytes
from a patient with RA, and which react with cell
surface constituents of monocytes and macrophages
(26,:!7). HLA-DR expression of these cells, determined using anti-HLA-DR monoclonal antibodies,
was intense. Electron microscopic examination
showed cells with the morphologic characteristics of
active macrophages (Figure 1).
Together, these functional, histochemical, immunochemical, and morphologic characteristics
clearly indicate that these cells are mononuclear
phagocytes of bone marrow origin, similar to those
described by Burmester et al(28). The intense staining
withi anti-HLA-DR distinguishes these cells from
blood monocytes, as does their morphology by electron microscopy, suggesting that modulation or differentiation has caused the surface membranes of these
cells to resemble tissue macrophages.
Although the basis for the intriguing functional
hete:rogeneity between macrophage-enriched subpopulations is not yet clear, there are several possibilities that could explain it. Monocytes and unstimulated
macrophages require activation to produce angiogenic
activity (8,29), and it is possible that in vivo in the
rheumatoid synovium, some of the macrophages have
been activated while others have not. Conversely, it is
possible that the macrophages within F2 and F4,
which are nonangiogenic, have been inhibited in terms
of their expression of angiogenic activity, while those
in F3 have not. Finally, it is possible that the ability to
produce angiogenic activity develops during differentiation and that precursor cells may give rise to clones
of differentiated macrophages with different functional
properties, some being capable of expressing angiogenic activity and some not. Suggestive evidence for
this theory comes from studies we have carried out
with macrophage cell lines (30). It may be that the ratio
of the angiogenic to nonangiogenic subpopulations of
macrophages within a particular diseased synovium
dictates the extent of neovascularization within that
particular tissue.
Although these cell populations from F2, F3,
and F4 consisted largely of macrophages, we cannot
totally rule out the possibility that a minor subpopulation of potently angiogenic nonmacrophage adherent
cells might be concentrated within F3 and absent from
the other fractions. To eliminate the possibility that
residual contaminating fibroblastic synovial cells
might account for the angiogenic responses we have
seen, we tested the ability of early-passage cultured
synovial fibroblastic cells to induce neovascularization. These cells were found not to be angiogenic. We
cannot totally rule out the possibility, however, that
the subculturing procedures used in passaging the
fibroblastic cells might result in the loss of angiogenic
activity present in an initial population of cells. Likewise, nonadherent synovial cells did not induce
neovascularization. Since these nonadherent cells
were found to be nonangiogenic, we have not characterized them further in this study.
These studies suggest that the growth and development of the synovial pannus is mediated at least
in part by infiltrating macrophages that induce the
neovascularization required for its extensive growth.
Healing wounds, rheumatoid arthritis, and solid tumors thus appear to share a common modality, namely
that stimulation of neovascularization in these situations is mediated, at least in part, by the presence of
angiogenic macrophage populations. In a healing
wound, the control mechanisms to switch off the
neovascular process are clearly present. In RA and
solid tumor growth, however, the controls limiting
neovascularization appear to have gone awry, and this
could be due, at least in part, to the continued recruitment and presence of angiogenic macrophages.
Several investigators have described inhibitors
of angiogenesis in model systems (31). Cartilages have
been shown to be a rich source of angiogenesis inhibitors (32,33). Substances such as protamine (34) and a
combination of heparin and hydrocortisone (35) have
been shown to have anti-angiogenic properties and, in
certain cases, to impede tumor growth in vivo (36). As
has been suggested for solid tumor growth, development of methods to interfere selectively with the
production of angiogenic factors may be a potential
approach to therapy in rheumatoid arthritis.
We thank our surgical colleagues, Drs. S. D. Stulberg, C. Schwartz, M. Kreger, and G. Landon, for supplying
the synovial tissues and Todd Albinger for technical assistance in preparing samples for electron microscopy. We also
thank Drs. Arthur Veis and Frank Schmid for helpful comments. Finally, we would like to especially thank Dr.
Herbert Rubinstein for his tireless efforts in obtaining
synovial tissue and for helpful discussions.
1 . H a m s ED Jr: Recent insights into the pathogenesis of
the proliferative lesion in rheumatoid arthritis. Arthritis
Rheum 19:68-72, 1976
2. Krane SM, Golding SR, Dayer JM: Interactions among
lymphocytes, monocytes, and other synovial cells in the
rheumatoid synovium. Lymphokines 7:75-136, 1982
3. Zvaifler NJ: The immunopathology of joint inflammation
in rheumatoid arthritis. Adv Immunol 16:265-336, 1973
4. Gardner DL: The Pathology of Rheumatoid Arthritis.
London, Edward Arnold Publishers, Ltd., 1972
5. Leibovich SJ: Mesenchymal cell proliferation in wound
repair: the role of macrophages, Soft and Hard Tissue
Repair. Edited by TK Hunt, RB Heppleston, E Pines,
D Rovee. New York, Praeger Press, 1984, pp 32!&351
6. Clark RA, Stone RD, Leung DYK, Silver DC, Hahn
DC: Role of macrophages in wound healing. Surg Forum
27:16-18, 1976
7. Thakral KK, Goodson WH, Hunt TK: Stimulation of
wound blood vessel growth by wound macrophages. J
Surg Res 26:430436, 1979
8. Polverini PJ, Cotran RS, Gimbrone MA Jr, Unanue ER:
Activated macrophages induce vascular proliferation.
Nature 269904-806, 1977
9. Folkman J: Tumor angiogenesis. Adv Cancer Res
43: 175-203, 1985
10. Kulka JP, Blocking D, Ropes MW, Bauer W: Early joint
lesions of rheumatoid arthritis. Arch Pathol59: 12%150,
11. Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9: 175-176, 1958
12. Polverini PJ, Leibovich SJ: Induction of neovascularization in vivo and endothelial cell proliferation in vitro by
tumor associated macrophages. Lab Invest 51:635-642,
13. Gimbrone MA, Cotran RS, Leapman SB, Folkman J:
Tumor growth and neovascularization: an experimental
model using the rabbit cornea. J Natl Cancer Inst
52:413427, 1974
14. Folkman J, Ausprunk D, Langer R: Connective tissue:
small blood vessels and capillaries, Textbook of
Rheumatology . Second edition. Edited by WN Kelley,
ED Harris Jr, S Ruddy, CB Sledge. Philadelphia, WB
Saunders, 1985, pp 197-21 1
15. Folkman J: Towards a new understanding of vascular
proliferative disease in children. Pediatrics 74:850-856,
16. Karnofsky MJ: A formaldehyde-glutaraldehyde fixative
of high osmolarity for use in electron microscopy (abstract). J Cell Biol 27:137A, 1965
17. Luft JH: Improvement in epoxy resin embedding methods. J Biophys Biochem Cytol 9:40W14, 1961
18. Hester RB, Walker WS: Separation of human murine
phagocytes by density gradients of Percoll, Methods for
Studying Mononuclear Phagocytes. Edited by D Adams.
New York, Academic Press, 1981, pp 195-198
19. Hamilton J: Hypothesis: in vitro evidence for the
invasive and tumor-like properties of the rheumatoid
pannus. J Rheumatol 10:845-851, 1983
20. Fassbender HG, Simmling-Annefeld M, Stofft E: Transformation der Synovialzellen bei rheumatoider Arthritis.
Verh Dtsch Ges Pathol 64:193-212, 1980
21. Fassbender HG, Simmling-Annefeld M: The potential
aggressiveness of synovial tissue in rheumatoid arthritis.
J Pathol 139:399406, 1983
22. O’Sullivan FX, Fassbender H-G, Gay S, Koopman WJ:
Etiopathogenesis of the rheumatoid arthritis-like disease in MRLll mice. I. The histomorphologic basis of
joint destruction. Arthritis Rheum 283529-536, 1985
23. Ausprunk D, Folkman J: Migration and proliferation of
endothelial cells in preformed and newly formed blood
vessels during tumor angiogenesis. Microvasc Res
1453-65, 1977
24. Brown RA, Weiss JB, Tomlinson IA: Angiogenesis
factor from synovial fluid resembling that from tumors.
Lancet i:682-685, 1980
25. Brown RA, Tomlinson IW, Hill CR, Weiss JB, Phillips
P, Kumar S: Relationship of angiogenesis factor in
synovial fluid to various joint diseases. Ann Rheum Dis
42:301-307, 1983
26. Dimitriu-Bona A, Burmester GR, Waters SJ, Winchester RJ: Human mononuclear phagocytic differentiation
antigens. J Jmmunol 130:145-152, 1983
27. Hermann F, Komischke B, Odenwald E, Ludwig WG:
Use of monoclonal antibodies as a diagnostic tool in
human leukemia and acute phase of chronic myeloid
leukemia. Blut 47: 157-163, 1983
28. Burmester GR, Dimitriu-Bona A, Waters SJ, Winchester RJ: Identification of three major synovial lining cell
populations by monoclonal antibodies directed to Ia
antigens and antigens associated with monocytesl
imacrophages and fibroblasts. Scand J Immunol 17:
(59-82, 1983
Koch AE, Leibovich SJ, Polverini PJ: Induction of
ineovascularization by activated human rnonocytes. J
Leukocyte Biol 39:233-238, 1986
‘PolveriniPJ, Leibovich SJ: Induction of neovascularization and non-lymphoid rnesenchymal cell proliferation
Iby macrophage cell lines. J Leukocyte Biol37:279-288,
Maugh TH: Angiogenesis inhibitors link many diseases.
Science 212: 1374-1375, 1981
Lee A, Langer R: Shark cartilage contains inhibitors of
tumor angiogenesis. Science 221:1185-1 187, 1983
33. Kuettner KE, Pauli BU: Inhibition of neovascularization by a cartilage factor, Development of the Vascular
System. Edited by J Nugent, M O’Connor. London,
Pitman, 1983, pp 163-173
34. Taylor S, Folkman J: Protamine is an inhibitor of
angiogenesis. Nature 297:307-3 12, 1982
35. Folkman J, Langer R, Linhardt R, Haudenschild C,
Taylor S: Angiogenesis inhibition and tumor regression
caused by heparin or a heparin fragment in the presence
of cortisone. Science 221:719-725, 1983
36. Heuser LS, Taylor S, Folkman J: Prevention of
carcinomatosis and bloody malignant ascites in the rat
by an inhibitor of angiogenesis. J Surg Res 36:244-250,
Без категории
Размер файла
902 Кб
macrophage, neovascularization, tissue, synovial, human, stimulating, rheumatoid
Пожаловаться на содержимое документа