close

Вход

Забыли?

вход по аккаунту

?

Synthesis secretion and deposition of fibronectin in cultured human synovium.

код для вставкиСкачать
1016
SYNTHESIS, SECRETION, AND DEPOSITION O F
FIBRONECTIN IN CULTURED HUMAN SYNOVIUM
BEVERLY B. LAVIETES, STEVEN CARSONS, HERBERT S. DIAMOND, and RICHARD S. LASKIN
We examined fibronectin synthesis, secretion, and
deposition in vitro by primary explants of rheumatoid
synovium. Primary cultures initiated from tissue with
monocytic infiltrates had higher levels of fibronectin synthesis; addition of dexamethasone at concentrations
known to stimulate other tissue fibroblasts increased
fibronectin synthesis and secretion. Newly synthesized
fibronectin recovered from primary rheumatoid culture
medium had a higher apparent molecular weight (240-245
kd), on sodium dodecyl sulfate-polyacrylamide gel electrophoresis, compared with fibronectin recovered from passaged normal and rheumatoid cultures (230 kd). Primary
rheumatoid explant cultures had a characteristicmorphology which correlated with fibronectin deposition. Dense
deposits of fibronectin extracellular matrix covered overlapping synoviocytes adjacent to esterase-positive monoPresented in part at the Annual Meeting of the American
Federation for Clinical Research, Washington, DC, May 1984, and
at the 48th Annual Meeting of the American Rheumatism Association, Minneapolis, MN, June 1984.
From the Departments of Medicine and Orthopaedic Surgery, Long Island Jewish-Hillside Medical Center, New Hyde Park,
New York, and the State University of New York, Stony Brook,
New York.
Supported by grants from the National Institutes of Health
(Multipurpose Arthritis Center grant AM-20625) and the New York
Chapter of the Arthritis Foundation.
Beverly B. Lavietes, PhD: Department of Medicine, Long
Island Jewish-Hillside Medical Center; Steven Carsons, MD: Department of Medicine, Long Island Jewish-Hillside Medical Center;
Herbert S. Diamond, MD: Department of Medicine, Long Island
Jewish-Hillside Medical Center; Richard S . Laskin, MD: Department of Orthopaedic Surgery, Long Island Jewish-Hillside Medical
Center.
Address reprint requests to B. B. Lavietes, PhD, Depart.
rnent of Medicine, Long Island Jewish-Hillside Medical Center,
New Hyde Park, NY 11042.
Submitted for publication July 31, 1984; accepted in revised
form March 14, 1985.
Arthritis and Rheumatism, Vol. 28, No. 9 (September 1985)
nuclear cells. Dexamethasone-treatedcultures showed little fibronectin deposited as extracellular matrix and did
not develop overlapping cellular networks. Characteristic
patterns of fibronectin synthesis and deposition in primary
rheumatoid cultures appear to result from interaction
between fibroblastic and monocytic cells. This culture
system may provide a model by which to study interactions between cells and extracellular matrix components
that regulate synovial cell function.
Synovial fluid from patients with rheumatoid
arthritis contains elevated levels of fibronectin (1-6),
and large deposits of fibronectin are seen in rheumatoid pannus (2,7,8). The multidomain structure of
fibronectin has been mapped by binding assays to
specific macromolecules, and fibronectin has been
associated with a variety of biologic functions, suggesting a role for this molecule in the disease process
(9).
The lack of correlation between the concentration of fibronectin and other plasma proteins in synovia1 fluid (1) is evidence against nonspecific leakage of
plasma fibronectin into the joint. It has been suggested
that there is local synthesis of fibronectin (2,3,6). We
examined factors regulating the synthesis of fibronectin in vitro by rheumatoid synovial tissue.
Type A synoviocytes resemble macrophages;
type B synoviocytes resemble fibroblasts (IC). Both
macrophages and fibroblasts isolated from other tissues have been shown to synthesize fibronectin in
vitro (for review see ref. 1 1). In a preliminary report by
Dayer et a1 (12), mononuclear cell factor was shown to
enhance the synthesis of fibronectin by rheumatoid
synovial fibroblasts. We therefore chose to study
FIBRONECTIN IN HUMAN SYNOVIUM
primary explants of rheumatoid synoviupl cultured in
medium supplemented with human serum to approximate as closely as possible the cell types and serum
proteins present in viva. Since fibronectin is heterogeneously distributed in pannus (2,7,8), we also examined the association of fibronectin with the various cell
types in the cultures.
PATIENTS AND METHODS
Tissue specimens. Synovial h u e samples were obtained from patients undergoing knee replacement surgery or
diagnostic arthroscopy for trauma. Patients classified as
having rheumatoid arthritis met the American kheumatism
Association criteria (13). Cultures were established within 30
minutes of excision.
Cultures. Tissue samples were washed and dissected
free of fat, fibrous tissue, and elastic tissue in Hanks'
balanced salt solution (HBSS; Gibco, Grand Island, NY).
The tissue was cut into 1-2-mm pieces and plated into
Dulbecco's modified Eagle's medium (DMEM; Gibco) containing 10% pooled human serum and antibiotics. For synthesis studies, cultures were transferred to medium supplemented with fibronectin-depleted serum. Samples of each
biopsy specimen were fixed in cold 95% ethanol and processed for histologic and immunofluorescefice studies (see
below).
Cultures were fed twice each week At the first
feeding, aliquots offresh medium were added; at the second,
medium was totally replaced.
For subcultures, cells were released from flasks by
brief trypsinization for 3 minutes at 37°C (trypsin-EDTA;
Gibco), then washed, counted, and replated.
After removing medium and washing twice with
HBSS, culture dishes were fixed with cold 95% ethanol and
kept at 4°C for 4-16 hours. The dishes were then air-dried
and stored at -30°C.
Radiolabeling of newly synthesized fibronectin. Medium was removed from culture Aasks, centrifuged, brought to
1 mM with phenylmethylsulfonyl fluoride (PMSF), and
stored at -30°C for fibronectin analysis by enzyme-linked
imrnunosorbent assay (ELISA) (14). The flasks were washed
twice with HBSS before adding labeling medium, which
consisted of 100 &i/flask 35S-methionine in DMEM, with
10% fibronectin-depleted human serum (see below) and
antibiotics.
After 24 hours of incubation, the culture medium was
removed, brought to 1 mM with PMSF, and processed for
fibronectin analysis by immunoprecipitation. The cells were
washed twice with HBSS, removed from the Aasks by
trypsinization, and extracted for DNA (15). DNA was assayed colonmetrically by the diphenylamine reaction (16).
Incorporation of label into deoxycholate-insoluble
extracellhlar matrix, relative to secretion into the medium,
was also determined. In these experiments, after removal of
labeled culture medium, the cell layer was washed twice
with HBSS and then lysed with deoxycholate (0.5% in 0.05M
Tris, with I mM PMSF and 0.01% sodium azide). After
washing with 0.05M Tris buffer, any extracellular matrix
remaining in the flasks was solubilized in 0.5% sodium
1017
dodecyl sulfate (SDS), 0.1 mM P-mercaptoethanol at 100°C
for 5 minutes. Aliquots of cell extract and matrix extract
were counted. An aliquot of deoxycholate cell extract was
assayed for DNA as described above.
Fibronectin isolation by affinity chromatography. Fibronectin was isolated from human plasma, human serum,
and synovial culture medium by affinity chromatography on
gelatin-Sepharose. Fibronectin was eluted with 8M urea and
dialyzed.
Purified plasma fibronectin was used to prepare
monospecific rabbit antifibronectin antibody (17). The rabbit
serum was absorbed with fibronectin-free plasma obtained
from gelatin-Sepharose columns. The absorbed serum was
used for immunoprecipitation studies. Antifibronectin IgG
was isolated from the antiserum on fibronecth-Sepharose
(18).
Immunoprecipitation assay. Rabbit antiserum to human plasma fibronectin was added to aliquots of labeled
culture medium; appropriate dilutions were determined for
each batch of antiserum. As a control for specificity, normal
rabbit serum was added to equal volumes of ctdture medium
at the same dilution. After incubation at 37°C for 1 hour, goat
anti-rabbit IgG (Cappel, Cochranville, PA) was added to all
samples. After 1 hour at 4"C, immune complexes were
precipitated by centrifugation for 5 minutes in a microfuge
(Beckman, Palo Alto, CA).
The precipitates were washed 4 times in 1/10 phosphate buffered saline (PBS; Gibco) and resuspended in
Laemmli sample buffer (0.05M Tris, 1% SDS, I mM PMSF)
(19). Aliquots were taken for scintillation countidg, and the
remainder was stored at -80°C prior to polyacrylamide gel
electrophoresis (PAGE).
Fibronectin synthesis by the cultures was calculated
from the counts per minute after correction for isotope
decay, efficiency of counting, and nonspecific precipitation,
i.e., antifibronectin-precipitated counts minus the counts
precipitated with normal rabbit serum.
ELISA for fibronectin. Fibronectin was assayed in
culture media by an indirect competitive ELISA, which was
adapted from the method of Rennard et a1 (14). Concentrations of fibronectin in the unknowns were calculated from
standard curves after logit transformation of the data (20).
Only regression lines having slopes within 20% of the
standards and correlation coefficients >0.95 were used.
Freshly prepared, nonincubated culture medium
contained 10% pooled human serum. All fibronectin levels
reported for media from cultures have been corrected by
subtraction of the fibronectin content of nonincubated samples of corresponding media.
SDS-PAGE analysis. Electrophoresis of fibronectin
samples in SDS-polyacrylamide was performed according to
the method of Laemmli (19), in the presence of dithiothreitol. Molecular weight markers used were myosin (200 kd), pgalactosidase (1 16 kd), phosphorylase B (94 kd), bovine
serum albumin (66 kd), and ovalbumin (43 kd) (Bio-Rad,
Richmond, CA). Purified plasma fibronectin was added to
the molecular weight markers.
Gels were stained with Coomassie blue. After destaining, radiolabeled gels were treated with water-soluble
scintillation phosphor (Autofluor; National Diagnostics,
Somerville, NJ). Gels were dried and exposed on Kodak
XAR-2 X-omat film for 2 weeks at -80°C.
LAVIETES ET AL
1018
Table 1. Characteristics of primary rheumatoid cultures with higher and lower rates of fibronectin synthesis*
Relative no. of
Specimen
Day of
culture
pol ykaryocytes/
precipitable Fn
TCA-precipitable
protein
13
22
15
28
28
17
21
2.38
2.73
2.62
5.95
4.64
2.85
2.77
3.42
1.37
+++
+++
+
+++
+
++
Group B
Group B
Group B
Group B
*
252
537
356
ND
ND
83
72
260 195
21
7
15
15
15
0.43
0.27
1.27
0.11
0.70
0.57 ? 0.45ll
341
55
79
14
206
139 134
-
Group I
Group A
Group 1
Group A
Group 1
lmmuno-
Group 1
83-9
83-17
83-18
83-21
83-23
84-8
84-9
Mean ? SD
Group 2
83-7
83-10
83-14
83-15
83-19
Mean ? SD
*
*
culture t
-
-
-
Histopathologic
classification$
NDI
Group B
Group I
* Values for immunoprecipitable fibronectin (Fn) and trichloroacetic acid (TCA)-precipitable protein are given in pmoles of 35Smethionine/pg of DNA/24 hours. ND = not done.
t + + + = epithelioid colonies; + + = intermediate no. of polykaryocytes;+ = scattered polykaryocytes; - = no polykaryocytes.
$ Classified according to the method of Malone et al (24): group A = lymphocyte infiltrate, numerous plasma cells, >7
proliferative synovial lining cells; group B = nonlymphoid infiltrate, sparse lymphocytes, heavy fibrin deposits; group I =
intermediate between group A and group B.
§ Although this specimen was not examined, tissue from patient’s contralateral knee was group B type.
11 P < 0.001 by Student’s t-test.
Gels of unlabeled fibronectin isolated from culture
media were silver stained, using a commercial kit (Bio-Rad).
Immunofluorescence studies. Biopsy samples, fixed
in cold 95% ethanol overnight, were processed for immunofluorescence analysis according to the method of SainteMarie (21). Blocks and slides were stored at 4°C prior to
staining. Slides were deparaffinized, quickly hydrated, and
washed twice (10 minutes each) with PBS, pH 7.2.
Fibronectin was localized using fluorescein isothiocyanate (FITCtconjugated rabbit anti-human fibronectin
F(ab’)*, prepared from our own antibody or obtained commercially (lot 15857; Cappel). Antibody diluted in PBS (19
pg of protein/25 pl) was pipetted directly over sections and
allowed to incubate for 30 minutes at room temperature.
Sections were washed 3 times with PBS, for 5 minutes each,
and mounted in glycerol-Verona1 NaCl ( l : l ) , pH 8.6 (22).
Adjacent sections were incubated similarly with FITC-conjugated nonimmune rabbit IgG F(ab’)* (lot 14685; Cappel).
Fixed tissue culture plates were examined by phase
microscopy, and areas of differing morphology were circled.
These areas were isolated by sealing a ring of Tygon tubing
(1-cm outside diameter, 0.7-cm inside diameter) to the dish
surface with petroleum jelly. The area within the ring was
washed twice with PBS and stained as described above for
slides.
Preparations were examined and photographed with
a Nikon Labophot microscope equipped with EFA epifluorescence (495-nm excitation filter) and UFX photography systems. Micrographs were taken on Kodak Tri-X film,
at ASA 400 or 800, and developed accordingly.
In addition to the nonimmune rabbit IgG F(ab‘)z
control, specificity for fibronectin staining was shown by
preabsorbing FITC-antifibronectin F(ab’)* with fibronectin. This was accomplished with fibronectin isolated from
human plasma by gelatin-Sepharose chromatography and
coupled to CNBr-activated Sepharose beads (Pharmacia,
Piscataway, NJ) (see above). Aliquots (100 pl) of fibronectin-Sepharose beads were added to a series of 1.5-ml
microfuge tubes and washed 3 times with PBS, by centrifugation.
After the final wash, 1 ml of FITC-antifibronectin
F(ab’):! was added to the first tube and incubated at room
temperature for 15 minutes, with periodic gentle inversion of
tubes. After centrifugation, the supernatant was transferred
to the second tube, and the beads were examined for
fluorescence. This procedure was repeated until fibronectinSepharose beads showed no fluorescence. Parallel aliquots
were treated identically with ethanolamine-blocked Sepharose to control for dilution artifacts.
No staining was detected with nonimmune serum or
antiserum absorbed with fibronectin-Sepharose.
Nonspecific esterase activity. Cytochemical identification of mononuclear phagocytes in cultures was performed
according to the method of Kaplow (23). Areas of cultures
were postfixed, as described by Kaplow (23), for smears and
imprints. Cultures were then incubated 4 hours with naphthol-aniline sauer acetate, yielding a blue reaction product,
and counterstained.
Smears of adherent peripheral blood monocytes, run
in parallel, were always 100% positive.
1019
FIBRONECTIN IN HUMAN SYNOVIUM
2.34 2 1.08
a
2.52 2 0.50
2.01 20.58
a
a
a
0.45t0.11
a
a
PRIMARY SECOND
PASSAGE
RHEUMATOID ARTHRITIS
PRIMARY SECONO
PASSAGE
NORMAL
Figure 1. Fibronectin secretion into medium of rheumatoid and
normal, primary and second-passage human synovial cultures.
(Rheumatoid synovium samples shown here are group 2 specimens
in Table 1.) Fibronectin was quantitated by an indirect competitive
enzyme-linked immunosorbent assay, against a plasma standard. An
aliquot of nonincubated culture medium corresponding to each
sample was run with every assay, and the fibronectin content of the
media due to serum was subtracted from each experimental value.
vial tissues from which the cultures were established
showed that the cultures that were more actively
synthesizing fibronectin came from synovium that had
a histologic appearance resembling group B tissues
described by Malone et a1 (24). These synovial specimens had nonlymphoid infiltrates and heavy fibrin
deposits.
The cultures with lower rates of fibronectin
synthesis were derived from specimens which histologically resembled Malone and coworkers’ group A
tissues (24), having more lymphocytes, plasma cells,
and a more hyperplastic synovial lining. No significant
differences were seen in general protein synthesis,
which was assayed as the incorporation of labeled
methionine into trichloroacetic acid-precipitable material (Table 1).
Figure 1 shows a comparison of rates of fibronectin synthesis in primary and second-passage rheumatoid and normal synovial cultures. The media from
rheumatoid cultures showed a significant increase
( P = 0.008) in fibronectin content for the second
passage compared with the primary cultures. That the
increase in fibronectin in the medium represented
increased fibronectin synthesis was confirmed by measuring 35S-methionine incorporation into immunoprecipitable fibronectin (Table 2). Cultures derived from
normal synovium did not change in their rate of
fibronectin synthesis with passage.
SDS-PAGE of 35S-labeled immunoprecipitated
fibronectin showed a difference in apparent molecular
weight between fibronectin from primary rheumatoid
cultures and passaged rheumatoid or normal synovium
(Figure 2). Passaged rheumatoid or normal cultures
secreted a molecule similar to fibroblast cellular fibronectin (227 I+_ 6 kd, mean 5 SD of 5 cultures).
Fibronectin secreted by primary rheumatoid cultures
RESULTS
Fibronectin synthesis and secretion. Fibronectin
synthesis and secretion into culture medium was assayed both by immunoassay of the culture medium
and by incorporation of 35S-methionine into immunoprecipitable fibronectin. Rates of fibronectin synthesis
by primary rheumatoid cultures were heterogeneous
(Table 1).
Comparison of fibronectin synthesis rates with
culture morphology showed that cultures with rates of
fibronectin synthesis that were >2 pmoles of 35Smethionine incorporatedlpg of DNA/24 hours were
characterized by the presence of more polykaryocytes
(Table 1). Further, histologic examination of the syno-
Table 2. Fibronectin synthesis and secretion by cultured human
svnovium*
3SS-methionineincorporated into
immunoprecipitable fibronectin
Primary culture Second-passage culture
Rheumatoid synoviumt
Group I
Group B
Normal synovium
0.42
2.38
2.05
3.62
6.28
2.26 L 0.28i
* Values are pmoles of 35S-methionine/pgof DNA/24 hours.
t Classified according to the method of Malone et al(24). See Table
1 for definitions.
.%
Mean +. SD of 3 cultures.
LAVIETES ET AL
1020
Figure 2. Fluorograms of sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (6% gels), showing fibronectins labeled with 35Smethionine and immunoprecipitated from synovial culture medium.
Lanes 1 and 3, primary rheumatoid synovium culture (specimen
84-8, Table 1); lanes 2 and 5 , passaged normal synovium culture;
lane 4, second-passage rheumatoid synovium culture (specimen
84-8, Table 1). The primary rheumatoid and passaged normal
cultures were labeled and processed at the same time. Small aliquots
of these immunoprecipitates were stored at -7o"C, while the
rheumatoid culture was carried through 2 passages, and the fibronectin was labeled and immunoprecipitated. Despite the difference
in counts loaded, it can be seen that the fibronectin from the
passaged rheumatoid culture (lane 4) had the same molecular weight
as the passaged normal culture fibronectin (lanes 2 and 5 ) . The lines
indicate the tops of the bands.
had a significantly (P < 0.002) higher apparent molecular weight (241 k 4 kd, mean k SD of 7 cultures).
Normal and rheumatoid synovial cells, passaged and analyzed in pairs, showed differences in
growth (Table 3). The rheumatoid cultures, in which
cells overlap (see below), achieved a higher saturation
density, which was indicated by larger amounts of
DNA (Table 3). Rheumatoid cultures synthesized
twice as much extracellular matrix per cell than did
normal synovial cultures, but the proportion of fibronectin in the deoxycholate-insoluble matrix did not
differ.
Addition of dexamethasone increased the rate
of fibronectin synthesis by rheumatoid primary cultures (Figure 3). Significant increases (P= 0.027) were
seen with 1OP6Mand 10-7M dexamethasone, but not
with 10-5M dexamethasone.
Morphology and fibronectin distribution of cultured rheumatoid synovium. Primary explant cultures
of rheumatoid synovium were heterogeneous in morphology, and fibronectin distribution differed among
the various cell types and configurations. In primary
cultures, fibroblastic cells appeared both in monolayers (Figure 4a) and in overlapping networks (Figure
4b).
Fibronectin deposition patterns, demonstrated
by binding of FITC-labeled antifibronectin F(ab')*,
were different on these differing cellular configurations. Parallel-oriented fibroblast monolayers were
covered with a fine network of extracellular fibronectin (Figure 4b). In areas of the monolayers where
round cells were located, the fibronectin deposits were
more dense (Figure 4b), and were similar to fibronectin deposits in the overlapping cell straps.
The pattern of nonspecific esterase reaction
(Figure 5a) was inverse to that of the fibronectin
distribution. The greatest esterase reaction, similar to
peripheral blood monocytes (Figure 5a, inset), was
seen in the round cell population. Spindle-shaped cells
had numerous small grains which were often asym-
Table 3. Fibronectin (Fn) secretion and deposition in paired, second-passage rheumatoid and normal synovial cultures
35S incorporated
Day of
culture
Normal synovium
Rheumatoid synovium
9
17
31
9
17
31
pg of DNA/
flask
Amount of Fn
in culture
medium*
7.0
11.6
14.9
13.3
25.8
32.9
0.6
3.5
2.7
6.9
-
into
extracellular
matrixt
% Fn in
deoxycholateinsoluble
matrix$
12.0
12.2
* Determined by enzyme-linked immunosorbent assay, and expressed as pg of Fn/pg of DNA/24 hours.
t Values expressed as pmoles of "S-methionine/pg of DNA/24 hours.
$ Fluorograms were scanned with a Beckman DU-B spectrophotometer, and the area under the fibronectin peak was calculated as
a percentage of the total area of the scan.
FIBRONECTIN IN HUMAN SYNOVIUM
4
10-7~
io-
M
10-5~
Dexamethasone Concentration
Figure 3. Effect of dexamethasone on fibronectin synthesis and
secretion by primary rheumatoid synovial cultures. Fibronectin
synthesis and secretion were quantitated by immunoprecipitationof
"S-methionine-iabeled material from culture media. Aliquots of
media were processed similarly with nonimmune rabbit sera, and
the counts in the precipitates were subtracted from the specific
samples. Lines connect parallel cultures treated with different
concentrations of dexamethasone.
metrically distributed (Figure 5a); flattened cells
showed pale perinuclear deposits.
The differences in fibronectin deposition were
apparent from the beginning of the cultures. Monolayers arose as outgrowths from tissue fragments. Cell
processes extending from the tissue fragments were
covered by a fine network of extracellular fibronectin
(Figure 6b). Overlapping networks of spindle-shaped
cells appeared to derive from small round cells that
1021
spilled out of the tissue (Figure 6c). Over the first week
of culture, they adhered, elongated, made contact with
one another, and formed straps which increased in size
(Figure 4c). Large, extensively spread polykaryocytes
also appeared to arise from this cell population (Figure
7a).
On day 2, round or polygonal cells showed
blebs of fibronectin at points of cell process extension
(Figure 6d). These processes were much broader than
those emerging from the tissue fragments (Figure 6a).
No extracellular fibronectin fibers were seen over the
round cells (Figures 6b and d), nor was extracellular
fibronectin matrix associated with polykaryocytes
(Figure 7b).
After subculture, only fibroblastic cells remained (Figure 8); only an occasional esterase-positive cell could be found (Figure 5b). Both overlapping
and parallel orientations were seen before confluence,
but postconfluent cultures were multilayered. Fibronectin distribution was typical of cultured fibroblasts
(Figure 8b). Extracellular fibronectin deposits were
denser where cells overlapped (Figure 8d).
Explants of normal synovium resulted in monolayered fibroblastic outgrowth and retained that morphology over passage.
Rheumatoid explants plated into and maintained in 10-6M dexamethasone did not show either
the overlapping networks or polykaryocytes. Fibroblast outgrowth was more extensive and started earlier
in treated cultures; fewer adherent round cells were
seen (Figure 9a). On day 14, fibroblastic cells showed
much more intense intracellular stain (Figure 9b) compared with similar cells in untreated cultures (Figure
4b); relatively little fibronectin extracellular matrix
was seen on day 14.
DISCUSSION
Our results demonstrate synthesis of fibronectin by synovial tissue explants. Synovial explants in
fibronectin-depleted medium incorporated 35S-methionine into fibronectin, which was both secreted into
medium (Table 1) and incorporated into extracellular
matrix (Table 3). Although the amount of fibronectin
synthesized per cell was not found to be greater in
rheumatoid synovium compared with normal synovium, the increased number of synthesizing cells in
proliferative synovium may indeed contribute to the
elevated levels of fibronectin observed in rheumatoid
synovial fluid. Primary cultures initiated from tissue
Figure 4. Heterogeneity of cell types and fibronectin deposition in 14-day primary rheumatoid synovial cultures. a, Fibroblastic
monolayer with superimposed round cells (arrow). b, Fluorescein isothiocyanate (FITC)-conjugated antifibronectin F(ab’)2 labeling
shows a fine network of fibronectin-containing extracellular matrix over the monolayered cells (thick arrow). Matrix deposits are denser
over cells in contact with round cells (thin arrow). c, Overlapping spindle-shaped cells and associated round cells. d, FITC-conjugated
antifibronectin F(ab’)2staining shows dense deposits of matrix. The fibers criss-cross, as do the cell processes. Round cells that were not
directly above spindle-shaped cells did not stain (arrow, and arrow in c). a and c, examined by phase microscopy; b and d. examined by
epifluorescence with a 495-nm excitation filter. Bar = 50 pm.
Figure 5. Nonspecific esterase activity in rheumatoid primary and
secondary cultures (specimen 83-9, Table 1). a, Primary culture on
day 13, showing heterogeneity of cells and intensity of staining. Inset
shows a peripheral blood monocyte. The round cells in the lower left
comer look like monocytes (thick arrow). Spindle-shaped cells and
more flattened cells stain, but the number, size, and distribution of
granules vary (thin arrows). b, First passage on day 11, showing a
positive spindle-shaped cell. Granules are distributed at only 1 pole
of the cell (arrow). Bar = 25 pm.
1022
Figure 6. Day 2 of primary rheumatoid synovial cultures, showing that the c-lls both grow out of (a) and spill out of (c) the rheumatoid
tissue fragments. a, Fine-cell processes emerge from tissue fragments. Round cells spill out and lie nearby. b, Fluorescein isothiocyanate
(FITC)-conjugated antifibronectin F(ab’), labeling shows extracellular fibronectin deposited over the processes extending from the
tissue fragment (arrow). c, Round cells extending processes and contacting each other (arrow). The processes are much broader than
those extending from fragments shown in a. d, FITC-conjugated antifibronectin F(ab’), stains blebs on the distal ends of cell processes
(arrows). No extracellular matrix fibers are seen over these cells. Bar = 50 pm.
Figure 7. a, Extensively spread cells, many of which are multinucleated, form an epithelioid sheet. b, Fluorescein isothiocyanateconjugated antifibronectin F(ab’), label is intracellular and mainly
perinuclear. Day 14 cultures; bar = 50 prn.
1023
LAVIETES ET AL
1024
Figure 8. Rheumatoid synovial cells in second passage, showing the cells to be more spread than those in primary cultures.
Both overlapping (a and b) and parallel (c and d) cell orientations are seen at this time (day 2 after passage). Fibronectin
deposits are heavier at sites of cell overlap (a). Perinuclear intracellular staining is seen in flattened monolayer cells (b). Bar =
50 pm.
with monocytic infiltrates had higher levels of fibronectin production (Table I), consistent with Dayer and
coworkers' preliminary observation that mononuclear
cell factor enhanced fibronectin synthesis by rheumatoid synovial fibroblasts (12).
Several different mechanisms may be responsible for the higher apparent molecular weight of newly
synthesized fibronectin recovered from primary rheu-
matoid culture medium. Differences between plasma
and cellular fibronectins within and between species
have been shown to result from differences in both
primary structure and posttranslational modification,
e.g., glycosylation, sulfation, phosphorylation (25).
Differences in molecular weight of 10-20 kd between
fibronectins from different human tissues have been
attributed to increased glycosylation (26,27).
Figure 9. Primary rheumatoid synovial explants maintained in medium containing lO-'M dexamethasone, on day 14. a, Fibroblastic
cells (arrows). b, Fluorescein isothiocyanate-conjugated antifibronectin F(ab'), labeling shows intense intracellular stain (arrows).
The extracellular matrix over fibroblastic cells is no heavier than that of untreated cultures (Figure 4b). Bar = 50 pm.
FIBRONECTIN IN HUMAN SYNOVIUM
1025
Rheumatoid synovial fluid and primary rheumatoid culture medium fibronectins display similar patterns of reactivity to peroxidase-conjugated lectins
(28). These patterns differ from those of plasma fibronectin suggesting differences in glycosylation of the
molecules. Since pure human monocyte cultures produce and secrete fibronectin with an apparent molecular weight that is 10-15 kd higher than fibroblast
celiular fibronectin (29), the 240-245 kd fibronectin
recovered from monocyte-rich cultures might be a
product of that cellular component.
The morphologic arrangement of synovial cells
in primary rheumatoid cultures appears to correlate
with fibronectin deposition. Overlapping synovial cells
adjacent to esterase-positive round cells were found to
be covered with dense extracellular matrix (Figure
4d). The multilayered arrangement of cells that is
typical of primary rheumatoid cultures has been induced in cultures of normal human synovial fibroblasts
by mononuclear cell supernatants (30). Dexamethasone inhibits this induction (30).
We have shown that primary rheumatoid explants plated into and maintained in 10-6M dexamethasone do not deposit much fibronectin extracellular
matrix and do not develop overlapping cellular networks (Figure 9). At concentrations known to stimulate fibronectin synthesis by other cultured fibroblasts
(3 l ) , dexamethasone did increase fibronectin synthesis
and secretion in these cultures. Together, these observations suggest that synthesis and deposition of fibronectin are independently regulated.
Krane et a1 (32) reviewed data from in vitro
systems which suggested that mononuclear cells and
their products regulate fibroblast function in rheumatoid pannus. Our data for primary rheumatoid explants, with respect to fibronectin synthesis and deposition, are consistent with this hypothesis (32).
Fibronectin has been associated with connective tissue remodeling in both normal development and disease (1 1,33). Alteration of extracellular matrix components such as fibronectin may further regulate
expression of cell receptors and phagocytic activity
(34). Fibronectin modulates the morphologic arrangement of cells in culture; for example, fibronectins of
slightly different molecular weights have been shown
to alter the morphologic arrangement of swine smoothmuscle cells in culture (35).
Whether the molecular weight differences seen
in fibronectin produced by primary rheumatoid cultures affect the characteristic morphology of these
cultures remains to be demonstrated. The primary
rheumatoid explant cultures may provide a model
system that more closely approximates the complex
interactions between cells and extracellular matrix
components that regulate synovial function in vivo.
ACKNOWLEDGMENTS
We are grateful to Dr. J. Roofeh, Department of
Orthopaedic Surgery, Long Island Jewish-Hillside Medical
Center, for providing arthroscopy specimens, and we acknowledge with pleasure the dedicated and skillful assistance provided by Barbara Diamond in the histologic studies
and photography, Ellen Berkowitz and Sandra Kinney in the
gels and assays, and Ann Scozzari in the preparation of the
manuscript.
REFERENCES
1 . Carsons SE, Mosesson MW, Diamond HS: Detection
and quantitation of fibronectin in synovial fluid from
patients with rheumatic disease. Arthritis Rheum 24:
1261-1267, 1981
2. Scott DL, Delamere JP, Walton KW: The distribution of
fibronectin in the pannus in rheumatoid arthritis. Br J
Exp Pathol 62:362-368, 1981
3. Scott DL, Wainwright AC, Walton KW, Williamson N:
Significance of fibronectin in rheumatoid arthritis. Ann
Rheum Dis 40:142-153, 1981
4. Clemmensen I, Andersen RB: Different molecular forms
of fibronectin in rheumatoid synovial fluid. Arthritis
Rheum 25:25-31, 1982
5. Lu-Steffes M, Iammartino AJ, Schmid FR, Castor CW,
Davis L , Entwhistle R, Anderson B: Fibronectin in
rheumatoid and non-rheumatoid arthritic synovial fluids
and in synovial fluid cryoproteins. Ann Clin Lab Sci
12:178-185, 1982
6. Vartio T, Vaheri A, von Essen R, Isomaki H, Stenman
S: Fibronectin in synovial fluid and tissue in rheumatoid
arthritis. Eur J Clin Invest 11:207-212, 1981
7. Holund B, Clemmensen I: The value of hyaluronidase
treatment of different tissues before demonstrations of
fibronectin by the indirect immunoperoxidase technique. Histochemistry 86:517-525, 1982
8. Shiozawa S, Ziff M: Immunoelectron microscopic demonstration offibronectin in rheumatoid pannus and at the
cartilage-pannus junction. Ann Rheum Dis 42:254-263,
1983
9. Carsons SE, Lavietes BB, Diamond HS: Role of fibronectin in rheumatic disease, Fibronectin. Edited by DF
Mosher. New York, Academic Press (in press)
10. Barland P, Novikoff AB, Hamerman D: Electron microscopy of the human synovial membrane. J Cell Biol
14:207-220, 1962
1 1 . Akiyama SK, Yamada KM: Fibronectin in disease,
Connective Tissue Diseases. Edited by B Wagner, R
Fleischmajer, N Kaufman. Baltimore, Williams & Wilkins, 1983, pp 55-96
1026
12. Dayer J-M, Byrne MH, Krane SM: Effects of a mononuclear cell factor, indomethacin and prostaglandin E2 on
collagen and fibronectin synthesis by cultured rheumatoid synovial cells (abstract). Clin Res 28:343A, 1980
13. 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
14. Rennard SI, Berg R, Martin GR, Foidart JM, Robey PG:
Enzyme-linked immunoassay (ELISA) for connective
tissue components. Anal Biochem 104:205-214, 1980
15. Rutter WJ: Protein determination in embryos, Methods
in Developmental Biology. Edited by FH Wilt, NK
Wessells. New York, Thomas Y . Crowell Co., 1967, pp
671-684
16. Burton K: A study of the conditions and mechanisms of
the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem J 62:315-323,
1956
17. Carsons SE, Lavietes BB, Diamond HS: Factors influencing the incorporation of fibronectin into synovial
fluid cryoprotein. J Lab Clin Med 102:722-731, 1983
18. Cuatrecasas P: Agarose derivative for purification of
protein by affinity chromatography. Nature 228: 13271328, 1970
19. Laemmli UK: Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature
227:680-682, 1970
20. Ritchie DG, Nickerson JM, Fuller GM: Two simple
programs for the analysis of data from enzyme-linked
immunosorbent (ELISA) assays on a programmable
desk-top calculator. Anal Biochem 110:281-290, 1981
21. Sainte-Marie G : A paraffin-embedding technique for
studies employing immunofluorescence. J Hhstochem
Cytochem 10:250-256, 1962
22. Stenman S, Vaheri A: Distribution of a major connective
tissue protein, fibronectin, in normal human tissue. J
Exp Med 147:1054-1064, 1978
23. Kaplow LS: Cytochemical identification of mononuclear macrophages, Manual of Macrophage Methodology.
Edited by HB Herscowitz, HT Holden, JA Bellanti, A
Ghaffar. New York, Marcel Dekker, Inc., 1981, pp 199205
LAVIETES ET AL
24. Malone DG, Wahl SM, Tsokos M, Cattell H, Decker JL,
Wilder RL: Immune function in severe, active rheumatoid arthritis: a relationship between peripheral b l o d
mononuclear cell proliferation to soluble antigens and
synovial tissue immunohistologic characteristics. J Clin
Invest 74:1173-1185, 1984
25. Paul JI, Hynes RO: Multiple fibronectin subunit$ and
their post-translational modifications. J Biol Chem 259:
13477-13487, 1984
26. Zhu BC-R, Fisher SF, Pande H, Calaycay J, Shively JE,
Laine RA: Human placental (fetal) fibronectin: increased glycosylation and higher protease resistance
than plasma fibronectin. J B i d Chem 259:3962-3970,
1984
27. Bumol TF, Reisfeld RA: Biosynthesis and secretion of
fibronectin in human melanoma cells. J Cell Biochem
21:129-140, 1983
28. Carsons SE, Lavietes BB, Diamond HS, Berkowitz E:
Carbohydrate heterogeneity of fibronectins: identification of plasma and synovial forms by lectin binding
(abstract). Clin Res 33:504A, 1985
29. Alitalo K, Hovi T, Vaheri A: Fibronectin is produced by
human macrophages. J Exp Med 151:602-613, 1980
30. Hamilton JA, Slywka J: Stimulation of human synovial
fibroblast plasminogen activator production by mononuclear cell supernatants. J Immunol 126:851-855, 1981
31. Oliver N , Newby RF, Furcht LT, Bourgeois S: Regulation of fibronectin biosynthesis by glycocorticoid in
human fibrosarcoma cells and normal fibroblasts. Cell
33:287-296, 1983
32. Krane SM, Goldring SR, Dayer J-M: Interactions among
lymphocytes, monocytes, and other synovial cells in the
rheumatoid synovium. Lymphokines 7:75-136, 1982
33. Yamada KM: Cell surface interactions with extracellular
materials. Ann Rev Biochem 52:761-799, 1983
34. Bevilacqua MF, Amrani D, Mosesson MW, Bianco C:
Receptors for cold-insoluble globulin (plasma fibronedtin) on human monocytes. J Exp Med 153:42-60, 1981
35. Brennan MJ, Millis AJT, Mann D, Fritz KE: Structural
alterations in fibronectin correlated with morphological
changes in smooth muscle cells. Dev Biol 97:391-397,
1983
Документ
Категория
Без категории
Просмотров
3
Размер файла
1 286 Кб
Теги
synthesis, synovium, fibronectin, deposition, secretion, culture, human
1/--страниц
Пожаловаться на содержимое документа