close

Вход

Забыли?

вход по аккаунту

?

Persistence of scleroderma-like phenotype in normal fibroblasts after prolonged exposure to soluble mediators from mononuclear cells.

код для вставкиСкачать
54
PERSISTENCE OF SCLERODERMA-LIKE PHENOTYPE
IN NORMAL FIBROBLASTS AFTER PROLONGED
EXPOSURE TO SOLUBLE MEDIATORS FROM
MONONUCLEAR CELLS
JENNIFER G. WORRALL, THERESA L. WHITESIDE, ROBERT K. PRINCE,
ROBERT B. BUCKINGHAM, IRENE STACHURA, and GERALD P. RODNAN
Supernatants of mononuclear cells (MNC-SN)
were shown to increase synthesis of glycosaminoglycan
(GAG;) by cultured normal dermal fibroblasts. Fibroblasts from the skin of patients with progressive systemic sclerosis (PSS, scleroderma) were hyporesponsive.
We exposed fibroblasts outgrowing from explants of
normal adult skin to MNC-SN for up to 30 generations
in culture. MNC-SN were obtained by incubating nocma1 MNC with concanavalin A. Four experimental, 4
normal control, and 3 PSS control lines were passaged
by trgpsinizing and splitting the cultures 1:2 every 7
days. At the third and fifth passages, portions of the
experimental fibroblasts were removed from MNC-SN,
then passaged in medium alone. Cell counts, assays for
GAG, and electron microscopy were performed and
increases in GAG after brief reexposure to MNC-SN
were determined at the third, fifth, and eighth passages.
In normal dermal fibroblasts, baseline GAG produc_____
From the Department of Medicine, Division of Rheumatology and Clinical Immunology, and the Department of Pathology, Division of Clinical Immunopathology, University of
Pittsburgh School of Medicine, and Allegheny General Hospital,
Pittsburgh, Pennsylvania.
Supported by NIH grant AM-24019, a grant from the RGK
Foundation of Austin, TX, and the Veterans Administration.
Jennifer G. Worrall, MBBS: Division of Rheumatology and
Clinical Immunology and recipient of the Copeman Travelling
Fellowship of the Arthritis and Rheumatism Council of Great
Britain; Theresa L. Whiteside, PhD: Division of Clinical Immunopathology; Robert K. Prince, PhD: Division of Rheumatology
and Cliniical Immunology: Robert B. Buckingham, MD: Division of
Rheumalology and Clinical Immunology; Irene Stachura, MD:
Allegheny General Hospital.
IDr. Rodnan is deceased.
'4ddress reprint requests to Dr. Theresa L. Whiteside,
Room 406, Scaife Hall, University of Pittsburgh School of Medicine,
Pittsburgh, PA 15261.
Submitted for publication May 29, 1984; accepted in revised
form Junle 19, 1985.
Arthritis and Rheumatism, Vol. 29, No. 1 (January 1986)
tion, measured by 3H-glucosamineuptake, was low and
increased as much as 15 times after reexposure to
MNC-SN. In contrast, production was high in both
experimental and PSS lines, and increases after
reexposure to MNC-SN were consistently small. This
PSS-like behavior persisted in experimental fibroblasts
removed from MNC-SN at the third and fifth passages.
Growth of experimental and scleroderma fibroblasts
was slower than that of control fibroblasts. Ultrastructurally, both scleroderma and experimental dermal
fibroblasts differed from normal fibroblasts by their oval
cellular shape, indentations in nuclear membrane, numerous organelles and bundles of microfilaments, prominent Golgi, and intranuclear inclusions. These experiments indicate that normal adult dermal fibroblasts
subjected to MNC-SN in vitro acquire a sclerodermalike phenotype that persists for many generations.
Progressive systemic sclerosis (PSS, scleroderma) is a serious, life-threatening disorder of unknown
etiology. It is characterized by fibrosis that affects
many organs, in particular, the skin (1). Affected skin
is thickened and contains an abnormal excess of
glycosaminoglycans (GAG), collagen, and other connective tissue substances (2-4). Fibroblasts grown in
vitro from skin biopsy material obtained from patients
with PSS produce excess connective tissue substances, compared with normal fibroblasts, for many
generations of culture (5-8).
The skin of patients with PSS, at least in the
early, active stage of the disease, is frequently infiltrated with mononuclear cells (MNC) (1,9). These
cells, when activated, produce a range of potent
mediators of cellular interactions (lo), and supernatants of activated mononuclear cells (MNC-SN) have
MNC FACTORS INDUCE PSS-LIKE PHENOTYPE
repeatedly been shown to alter the behavior of fibroblasts in culture. MNC-SN may modulate the growth
(11,12) and synthesis of collagen (13-15) and GAG
(16,17) b y fibroblasts. Normal as well as PSS fibroblasts have been shown to respond t o these modulatory influences (16). Thus, MNC and their products
may be implicated in the pathogenesis of PSS.
PSS is an acquired disease, that is, it arises in
tissues that are initially normal. Investigations into the
etiology and pathogenesis of PSS have been hampered
b y the lack of an adequate animal model. The development of a tissue culture model which recreated
conditions existing in vivo, e.g., by causing normal
cells to adopt PSS-like characteristics, would therefore represent an important step in such an investigation.
In this study, we subjected dermal fibroblasts
grown from normal adult skin explants to prolonged
exposure to products of activated MNC. W e found
that fibroblasts exposed t o MNC-SN exhibited alterations in ultrastructure, cell growth, and GAG synthesis. In terms of GAG synthesis, both at baseline and in
response to a brief challenge with MNC products, the
experimental cultures differed from the normal control
cultures but resembled the PSS control cultures.
Moreover, this resemblance was still apparent many
weeks after termination of exposure to MNC products. Our experiments indicate that normal adult dermal fibroblasts subjected in vitro to a period of exposure to MMC-SN acquire a PSS-like phenotype that
persists for many generations.
MATERIALS AND METHODS
Generation of active MNC-SN. Mononuclear cells
were obtained from the peripheral blood of normal volunteers by separation on Ficoll-Hypaque gradients (Pharmacia,
Piscataway, NJ). After washing 3 times in Medium 199
(Gibco, Grand Island, NY), MNC were incubated at a
concentration of 1.5 x 106/ml in Medium 199 supplemented
with 2% fetal calf serum and ITS (insulin, selenium, transferrin; Collaborative Research, Lexington, MA) for 24 hours in
a 5% C02 atmosphere at 37°C. Concanavalin A (Con A;
Miles Laboratories, Elkhart, IN), at a final concentration of
10 pg/ml, was added to the cultures at the start of incubation.
The MNC-SN were harvested by centrifugation,
treated with 0.1M a-methyl-D-mannoside (Sigma, St. Louis,
MO) to inactivate Con A, and assayed for biologic activity
(i.e., stimulation of GAG synthesis) on normal dermal fibroblasts as follows. The MNC-SN were diluted 1 : 10 with fresh
growth medium and added to confluent fibroblast monolayers in quadruplicate wells of multiwell plates. Fresh medium
and supernatants of MNC not activated with Con A were
used as controls. The fibroblasts were incubated with the
55
NORMAL
EXPLANT
\1
CONTROLS
n-+
~TRYPSINIZE
~TRYPSINIZE
/ASSAY
/ASSAY
@- -
-+ITRYPSINIZE
/&SAY
--------
Figure 1. Experimental protocol for exposure of fibroblasts outgrowing from explants of normal skin to mononuclear cell supernatants (MNC-SN).The numbers 1-6 = passage number; thick arrows
= cultures passaged in the presence of MNC-SN; thin arrows =
cultures passaged in nutrient medium only; GAG = glycosaminoglycan.
MNC-SN for 72 hours (16) and then assayed for GAG
accumulation as described below. Active MNC-SN were
pooled and stored at -70°C until further use. Control supernatants were generated by incubating medium with MNC but
without Con A and medium without MNC but with Con A.
Establishment of fibroblast cultures. Full-thickness
punch biopsies of dorsal forearm skin were obtained from
patients with early, rapidly advancing PSS and from normal
volunteers whq were sex-, race-, and nearly age-matched
with the patients. At the time of biopsy, none of the patients
was taking medication thought to affect connective tissue
metabolism, and all were diagnosed as having the diffuse
variant of PSS, in accordance with previously described
clinical criteria (18). Biopsy samples were obtained from 3
patients with PSS and from 4 normal volunteers.
Biopsy material was minced and placed in separate
25-cm2 plastic flasks (Falcon Labware, Oxnard, CA). After
standing at room temperature for 30 minutes, the tissues had
attached to the flasks, and nutrient medium, consisting of
CMRL 1066 with Earle’s salts (Gibco) supplemented with
15% pooled human serum, penicillip (100 unitshl), and
streptomycin (100 pglml), was added. Explant cultures were
incubated at 37°C. After 2-3 weeks, each explant was
surrounded by a dense outgrowth of fibroblasts. Subcultures
were established from these primary cultures by trypsinization. Thereafter, fibroblasts were passaged by trypsinizing
and splitting the cultures 1 :2 every 7 days.
Expssure of fibroblaststo MNC-SN. The design of the
experiment is shown in Figure 1. Each normal biopsy
specimen was halved, and 1 of the halves (normal control)
was treated as described above. The other half (experimental) was exposed to nutrient medium supplemented with 10%
MNC-SN. All control (normal and PSS) and experimental
cultures were incubated and passaged as described above,
with fresh MNC-SN added to the experimental cultures at
each nutrient medium change. At the third and fifth passages. portions of the experimental fibroblasts were removed
from MNC-SN exposure and passaged in nutrient medium
alone. while other portions continued their growth in the
56
presence of MNC-SN. At the third, fifth, and eighth passages, cell counts (Coulter counter), viability counts by
trypan blue exclusion, and assays for GAG syn?hesis were
performed on all fibroblast cultures.
Assay for GAG synthesis. Falcon 24-well tissue culture plates were inoculated with 40,000 cells in 0.5 ml of
nutrient medium per well on the day the experiment was
initialed (day 0). Multiple wells were started for each fibroblast line. Prior to assays, all cultures, including those
undergoing MNC-SN exposure during passaging, were
grown in nutrient medium alone while in tissue culture
plates,. The cells were allowed to grow for 6 days, at which
time cell numbers approached saturation density.
Nutrient medium was changed on days 1,4, and 6. At
the time of medium change on day 6, 8% (volumelvolume)
MNC-SN was added to half the wells of each fibroblast line,
control and experimental, in order to compare baseline GAG
synthlesis with alterations in GAG synthesis in response to a
further brief challenge with MNC-SN. The cells were then
allowed to grow for 72 hours without a further medium
change. At 60 hours (12 hours prior to harvest), 10 pl of
D- 1,6-3H-glucosamine hydrochloride (2 pCi/ml; New England Nuclear, Boston, MA) was added to each well to label
the GAG. Ten microliters of ascorbic acid (50 pl/ml) was
added to each well 1-2 hours before addition of the 3Hglucosamine. At harvest, 100 pl of O.15M Tris buffer, pH 8.0,
containing 30 mg/ml of pronase (Calbiochem-Behring, San
Diego, CA) Was added to each well.
The fibroblasts were rapidly detached, and cell numbers were determined in a Coulter counter, on 0.2 ml of the
resultant cell suspension. The remainidg cell suspension (0.4
ml) W a s incubated with pronase for 4 hours at 55°C and
centrifuged clear.
Total GAG (both cell-associated and that released
into supernatants) was assayed as follows, using a modification of the method of Scott (19). Camer GAG was added to
1 W p l aliquots of each sample, followed by 10 volumes of
0.5% cetj4pyridinium chloride (CPC) in 0.002M sodium
sulfate:. Reactioh tubes were mixed and allowed to stand for
20 minutes at room temperature. The resultant precipitates
were collected on Whatrnan glass-fiber filters (Fisher Scientific, ]Pittsburgh, PA). Individual tubes and filters were
washed twice with 0.1% CPC in O.05M sodium chloride, and
the filters were then washed 4 times with distilled water. The
filters were placed in liquid scintillation counting vials and
dried at 45°C. Ten milliliters of toluene with PPO and
POPOP was added, and counts were performed in a Beckman liquid scintillation counter. GAG synthesis was expressed as counts per minute of incorporated 3H-glucosamine per cell.
Electron microscopy. The control and experimental
fibroblast cultures were grown to confluence in 25-cm2
plastic flasks (Falcon). The growth mediuin was removed
with a pipette, and the monolayers were fixed with 3%
(weightholume) glutaraldehyde buffer&d with Sorensen’s
phosphate buffer to pH 7.4, for 1 hour at room temperature.
The cells were detached from plastic surfaces by gentle
scrapirig with a rubber policeman and were then transferred
to conical glass tubes. After centrifugation at low speed to
sediment the cells, the pellets were washed several times
with phosphate buffered saline and postfixed for 1 hour with
WORRALL ET AL
a 2% osmium tetroxide solution buffered with phosphate
buffer (pH 7.4).
After fixation, the cells were dehydrated in a series of
ethyl alcohols and propylene oxide and embedded in
Poly/Bed 8 12-Araldite mixture. One micron-thick sections
stained with toluidine blue were screened by light microscopy, and representative areas were selected for further
studies. Ultrathin sections mounted on 200-mesh copper
grids were stained with uranyl acetate and Reynolds lead
citrate. The fibroblasts were examined and photographed
with a Phillips EM-300 electron microscope at 60 kV.
RESULTS
GAG production in control and experimental
cultures. Contk-ol cultures of fibroblasts were grown
from 4 normal and 3 PSS skin biopsy samples. Each
normal biopsy specimen was divided in half to yield an
experimental culture, which was subjected to continuous MNC-SN exposure beginning at the explant
stage, and an untreated control culture. Table 1 shows
the spontaneous baseline accurpulations of GAG in
these cultures assayed at the eighth passage. It can be
seen that in each PSS control culture; production of
GAG, measured by incorporation of 3H-glucosamine
during a 12-hour pulse, was much greater than the
average production of GAG in all the normal control cultures. This finding is in agreement with our
previously published results (5). We have previously
demonstrated, by enzymatic digestions with fungal
hyaluronidase, that tritiated glucosamine is mostly
incorporated into hyaluronic acid in these cultures (5).
Although the normal control cultures were variable in their production of GAG, the highest normal
mean f SD)
control value (174 +- 4 cpdcell x
was still only 44% of the lowest PSS control value (399
& 31 cpndcell x
The experimental cultures each
departed from the behavior of their kppropriate normal
controls and showed a high level of baseline GAGkell,
of the order of that seen with PSS control cultures.
Also, when the GAG synthesis per culture was calculated (by multiplying cpdcell x cell number), the
experimental and PSS control cultures clearly showed
higher GAG accumulation than did the normal control
cultures.
PSS fibroblasts grew less well in culture than
did normal fibroblasts, as shown by the cell numbers
(Table 1). Normal fibroblasts cultured with MNC-SN
for a prolonged period also grew less well than did
their normal controls. Thus, in terms of cell numbers
and baseline production of GAG, it appears that
57
MNC FACTORS INDUCE PSS-LIKE PHENOTYPE
Table 1. Baseline accumulations of glycosaminoglycan (GAG) and cell numbers in fibroblasts
outgrown in the presence of MNC-SN*
GAG (cpdcell
x 10-3)
N (ChW
N + MNC-SN
N (KaL)
N + MNC-SN
N (JoM)
N + MNC-SN
N (JC)
N + MNC-SN
No. of cells
( X 10-7
0.237 ? 0.019
0.139 f 0.001
0.237 f 0.01
0.100 -+ 0.003
0.252 ? 0.028
0.129 f 0.004
0.289 ? 0.010
0.182 -+ 0.010
43.9 2 4
237 2 49
174 2 4
389 f 19
113 f 22
406 f 29
93 f 8
761 ? 35
+I24
+ 259
+718
PSS (RoH)
PSS (JeL)
PSS (PL)
0.180 -+ 0.006
0.149 2 0.006
0.170 -+ 0.01
485 2 82
399 -+ 31
670 f 45
+358t
+277t
+532t
Culture
% change
from normal
+440
-
* Each line was assayed at passage 8. See Materials and Methods for details. Values are means 2 SD
of triplicate determinations. MNC-SN = supernatants of activated mononuclear cells; N = normal
control lines; letters in parentheses are the initials of the patient from whom cells were obtained: N +
MNC-SN = experimental lines, i.e., normal cultures subjected to prolonged exposure to MNC-SN:
PSS = progressive systemic sclerosis (scleroderma) control lines.
t Represents the % GAG increase in PSS cultures over the mean GAG synthesis of the 3 normal
fibroblast lines.
experimental normal fibroblasts cultured in the presence of MNC-SN over several generations begin to
resemble PSS fibroblasts rather than their appropriate
normal control fibroblasts.
It may be argued that the increased GAG production in the experimental and PSS cultures is a
result of the lower cell densities achieved in these
cultures rather than the result of a real change in
synthetic rate of GAG. To ihvestigate the relationship
between cell density and GAG synthesis in normal
dermal fibroblasts, we plated fibroblasts at 3 cell
densities, labeled them with 3H-glucosamine for 24
hours, and measured the radioactivity incorporated
into the glycosaminoglycans (Figure 2). These experiments were done with 2 different fibroblast lines, and
they showed that as much as a threefold difference in
incorporation of radioactivity into cellular GAG was
demonstrable between the low-density (20,000 cellplating density) and the high-density (100,000 cellplating density) cultures. Thus, production of GAG in
normal dermal fibroblasts appears to be critically
dependent on cell density in actively growing cultures
(Figure 2).
All of our control and experimental cultures
were assayed for GAG on day 6 of growth, i.e., at
confluence. Under defined growth conditions and with
a standard inoculum of 40,000 fibroblasts in the log
phase of growth, normal and PSS cultures are in
confluence by day 6 (16). Using 8 normal fibroblast
20 x 103
b
40
r03
1501
X
0
2
100
50
0
T
0
401
30
I
IGAGIWELI
X
E
a
20
10
r
0
no* CELL
500
?
P
x
400
300
E
2 200
I00
0
LINE 1 OF NORMAL DERMAL FIBROBLASTS
LINE 2
OF NORMAL DERMAL FIPROBLASTS
Figure 2. Two lines of normal dermal fibroblasts (line 1 in third
passage, line 2 in sixth passage), plated at the cell densities shown at
top, and after 24 hours of growth, labeled with 3H-glucosamine.
Cultures were harvested 24 hours later and assayed for total
glycosaminoglycan (GAG) and cell numbers. Values are means of
triplicate culture wells; bars represent standard deviations.
WORRALL ET AL
58
Table 2. Relationship between cell numbers and glycosaminoglycan (GAG) synthesis in cultures of
normal dermal fibroblasts at confluence*
Day 4
No. cells
Culture
la
lb
2a
2b
~~
(X
299 f 40
253 7
138 2 8
152 f 40
*
Day 6
GAG (cpmi
cell x
71 f 5
97 3
103 2 10
9023
*
No. cells
(X
259
GAG (cpml
cell x
10-3)
f
30
84 2 10
140 2 30
97 t 1
160 f 30
195 f 16
116 k 6
113 t 17
~~
* Two normal adult dermal fibroblast lines were each studied at 2 different times (a and b): line 1 at the
third passage and line 2 at the sixth passage. All cultures were inoculated with 40 X lo3 cells in the log
phase of growth and were grown, fed, and labeled as described in Materials and Methods. Values are
means 2 SD of triplicate determinations.
strains, we have shown that no further increases in cell
numbers occur by day 4 of culture (unpublished observations). Therefore, we also determined the relationship between cell numbers and GAG synthesis in
confuent cultures of normal dermal fibroblasts (Table
2). It appears that GAG synthesis remains inversely
related to cell density at confluence. Nevertheless, the
differences in GAG synthesis between the log-phase
cultures and the stationary-phase cultures were much
smalller than those between control and experimental
cultures, as seen in Table 1. Although a high level of
baseline GAG/cell in experimental cultures exposed to
MNG-SN may be partly related to slower growth
kinetics, it is probably also a result of increased
synthesis of GAG in these cultures.
We have previously found that incubation of
fibroblast cultures with MNC-SN for brief periods
(24-96 hours) causes an increase in fibroblast GAG
production; in normal fibroblasts with a low baseline
GAG level, this increase can be enormous (up to
I,SOO%), whereas in PSS fibroblasts with a high
baseline GAG level, the increase is consistently much
smaller (<200%) (16). We therefore investigated the
response of our experimental cultures to a further,
brief challenge with MNC-SN.
Table 3 shows the results of the representative
experiment, performed at the eighth passage. The
normal control cultures, which had. a low baseline
accuinulation of GAG, were able to increase GAG
production greatly (by 303-1,391%), whereas the PSS
controls, which already had a high baseline accumulation of GAG, increased GAG production by a much
smaller amount (5164%). Of the experimental cultures, 2 (ChH and JC) showed a dramatic reduction in
the magnitude of the response to MNC-SN, compared
with their appropriate normal controls. A third culture
(JoM) showed a moderate reduction in response, while
the response of the fourth culture (KaL) was unaffected.
Thus, while the behavior of the experimental
cultures in their response to a further challenge with
MNC-SN was variable, it was still generally similar to
that seen in the PSS cultures. A further challenge with
MNC-SN had no consistent effect on cell numbers.
Both PSS and normal controls showed small variable
changes from baseline, and the experimental fibroblasts responded in a similar manner.
Persistence of the observed changes in cultures
removed from MNC-SN. We next determined whether
the changes seen in the experimental fibroblasts continuously exposed to MNC-SN persisted in culture
after exposure to MNC-SN was terminated. At the
third and fifth passages, portions of each experimentai
culture were removed from exposure to MNC-SN and
Table 3. Response of fibroblasts outgrown in the presence of
MNC-SN to short-term reexposure to MNC-SN*
(xlo-6)
GAG ( c p d
cell x
% change from
Culture
N (ChH)
N + MNC-SN
N (KaL)
N + MNC-SN
N (JoM)
N + MNC-SN
N (JC)
N + MNC-SN
0.261 2 0.015
0.135 f 0.011
0.264 f 0.008
0.098 f 0.014
0.303 2 0.021
0.104 f 0.012
0.289 f b.012
0.156 f 0.003
570 f 50
1,213 f 92
702 f 30
1,877 -t 263
635 f 19
1,813 ? 59
1,370 2 48
1,219 f 177
+ 1,200
+412
+ 303
+ 383
+462
+347
+ 1,391
+60
PSS (RoH)
PSS (JeL)
PSS (PL)
0.196 f 0.007
0.159 f 0.004
0.166 2 0.003
No. cells
802 f 37
72
720 -t 40
1,055
*
baselinet
+65
+ 164
+7
* Each line was assayed at passage 8. See Materials and Methods
for details. Values are means f SD of triplicate determinations. See
Table I for definitions.
t See Table 1 for baseline GAG production for each normal and PSS
control culture.
MNC FACTORS INDUCE PSS-LIKE PHENOTYPE
Table 4. Persistence of the MNC-SN-induced alteration in fibroblast GAG after removal of MNC-SN, effects on baseline GAG*
Culture,
passage no.
N (ChH)
N + MNC-SN,
N + MNC-SN,
N (KaL)
N t MNC-SN,
N + MNC-SN,
N (JoM)
N + MNC-SN,
N + MNC-SN,
N (JC)
N + MNC-SN,
N + MNC-SN,
PSS (RoH)
PSS (JeL)
PSS (PL)
No. cells
( X 10-6)
3
5
3
5
3
5
3
5
0.237 2 0.019
0.148 f 0.005
0.136 f 0.003
0.237 f 0.01
0.127 t 0.009
0.088 f 0.003
0.252 f 0.28
0.195 f 0.005
0.177 f 0.010
0.289 f 0.010
0.343 f 0.005
0.222 t 0.009
GAG (cpm/cell
x w3)
43.9 2 4
92.1 ? 3
204 f I
174 f 4
216 2 26
499 f 48
113 f 22
222 f 32
329 2 4.5
93 2 8
282 ? 45
604 f 34
0.180 f 0.006
0.149 f 0.006
0.170 f 0.01
485
399
670
82
31
f 45
f
f
% change
from normal
+ 109
+365
-
+ 24
+ I87
+96
+ I91
-
+ 206
+556
+358t
+277t
+532t
* Experimental lines were incubated and passaged in the presence
of MNC-SN as described in Materials and Methods. Normal and
PSS control lines were passaged simultaneously, but in medium
alone. At the third and fifth passages, portions of the experimental
fibroblasts were removed from MNC-SN exposure and passaged
further in medium alone. These cells as well as control cultures were
assayed for cell counts and baseline GAG synthesis at passage 8, as
described in Materials and Methods. Values are means f SD of
triplicate determinations. See Table 1 for definitions.
t Represents the % GAG increase in PSS cultures over the mean
GAG synthesis of the 3 normal fibroblast lines.
59
passaged in nutrient medium alone. These cultures
were also assayed at the eighth passage. Table 4 shows
that for each experimental culture for which exposure
to MNC-SN was terminated at an earlier passage,
there was an increase in baseline GAG production and
a reduction in cell numbers in comparison with the
relevant normal control. Thus, the altered behavior
which was induced by MNC-SN persisted after MNCSN was removed. Moreover, those experimental cultures with the longer duration of exposure to MNC-SN
(and therefore the shorter time in culture following
termination of exposure) showed a greater increase in
baseline GAG and a greater reduction in cell numbers
over normal than those cultures with the shorter
duration of exposure.
Table 5 shows the increase in GAG production
over baseline levels in response to a further, brief
exposure to MNC-SN. In all but 1 of the experimental
cultures (KaL), the dramatic normal response to
MNC-SN was attenuated. This attenuation had been
observed in experimental cultures undergoing continuous long-term exposure to MNC-SN (Table 2), and as
shown in Table 5, it persisted in corresponding experimental cultures in which MNC-SN exposure had been
terminated at an earlier passage. The effects on cell
Table 5. Persistence of the MNC-SN-induced alteration in fibroblast GAG after removal of MNCSN, effects on GAG with reexposure to MNC-SN*
N
N
N
N
N
N
N
N
N
N
N
N
GAG (cpmi
cell x 10-7
% change from
Culture,
passage
.
- no.
No. cells
( x 10-9
(ChH)
+ MNC-SN,
+ MNC-SN,
(KaL)
+ MNC-SN,
+ MNC-SN,
(JoM)
+ MNC-SN,
+ MNC-SN,
(JC)
+ MNC-SN,
+ MNC-SN.
0.261 f 0.015
0.152 f 0.007
0.116 f 0.006
0.264 f 0.008
0.133 ? 0.008
0.091 ? 0.006
0.303 f 0.021
0.215 f 0.023
0.193 f 0.009
0.289 t 0.012
0.324 f 0.005
0.202 f 0.009
570 ? 50
864 f 58
1,593 f 132
702 f 30
1,028 ? 50
2,008 f 322
635 f 19
826 t 80
!)97 ? 86
1,370 t 48
1,166f 17
1,578 2 59
+ 1,200
+838
+681
+ 303
+ 376
+ 302
+462
+ 272
+203
+ 1,391
+313
+ 161
0.196 f 0.007
0.159 2 0.004
0.166 ? 0.003
802 t 37
1,055 f 72
'720 f 40
+ 65
+ 164
PSS (RoH)
PSS (JeL)
PSS (PL)
3
5
3
5
3
5
3
5
baselinet
+7
* Fibroblast lines were passaged as described in Materials and Methods. Cells removed from MNCSN at passages 3 and 5 were assayed for GAG synthesis after a 72-hour reexposure to MNC-SN at
passage 8, as described in Materials and Methods.Values are means f SD of triplicate determinations.
See Table 1 for definitions.
t See Table 3 for baseline GAG production for each normal and control PSS culture.
WORRALL ET AL
60
numbers were again small and variable within all
groups of control and experimental fibroblasts.
For clarity, only the results of assays performed
at the eighth passage are shown in Tables 1, 3, 4, and
5, altlhough all strains were also tested at the third and
fifth passages. The results of all assays performed on 1
representative strain (JC) are given in Tables 6 and 7
and show that the change in phenotype of the experimental fibroblasts, induced by the prolonged influence
of MNC-SN, was already present at the third and fifth
passages.
Control supernatants, generated by incubating
medium with MNC but without Con A and medium
without MNC but with Con A, were assayed for
activity on fibroblasts in several preliminary shortterm experiments (data not shown). The activity of
supernatants generated without MNC did not differ
from that of medium alone, and the activity of supernatants generated with MNC but without Con A was
persistently midway between that of medium alone
and MNC-SN.
Ultrastructural appearance of cultured fibroblasts, Cultured fibroblasts from normal skin biopsy
samples displayed minimal variability. The cells were
spindle-shaped, elongated, with regular oval nuclei,
model-ate amount of rough endoplasmic reticulum
(RER), scattered mitochondria, and rare intracytoplasmic lipid droplets. Golgi apparatus and intracellular
and extracellular procollagen bundles were not conspicuous (Figure 3). A few cells had slightly irregular
nuclei and contained rare small cisternae of RER.
Apprciximately 20% of cultured fibroblasts from normal skin biopsies exhibited prominent nucleoli, and
fewer than 10% of these cells contained simple
Table 6. Baseline accumulations of GAG and cell numbers in
fibroblasts outgrown from a normal skin biopsy sample in the
Presence of MNC-SN. for different numbers of oassages*
Culture,
passage no.
N (JC)
3
5
8
N t MNC-SN
3
5
8
GAG (cpml
cell x lo-))
7% change from
normal
0.373 ? 0.012
0.455 t 0.025
0.289 t 0.010
74 t 10
101 t 12
93 t 8
0
0
0
0.250
0.281
0.182
369 5 7
159 t 6
761 t 35
+ 398
No. cells
(X
10-9
?
?
4
0.015
0.022
0.011
+ 57
+ 728
* Control and experimental cultures were assayed for GAG and cell
numbers at passages 3, 5 , and 8, as described in Materials and
Methods. Values are means t_ SD of triplicate determinations. See
Table 1 for definitions.
Table 7. Responses of fibroblasts outgrown from a normal skin
biopsy sample in the presence of MNC-SN to short-term reexposure
to MNC-SN, for different numbers of passages*
Culture,
passage no.
N (JC)
3
5
8
N
+ MNC-SN
3
5
8
No. cells
(X I O P )
GAG (cpml
% change from
cell x
baseline
0.422 5 0.015
0.460 t 0.018
0.289 ? 0.012
448 ? 10
356 t 40
1,370 t 48
+ 502
0.252 t_ 0.003
0.287 5 0.008
0.156 t 0.003
769 5 43
389 t 51
1,219 t 177
+ 108
+ 252
+ 1,391
+ 145
+60
* Control and experimental cultures were assayed for GAG and cell
numbers, following a 72-hour reexposure to MNC-SN, at passages
3, 5 , and 8, as described in Table 2. Values are means t SD of
triplicate determinations. See Table 1 for definitions.
intranuclear 240A inclusions composed of conglomerates of spherical granules surrounded by low electron
density ‘‘halo.
In contrast, cultured fibroblasts from PSS patients as well as experimental (MNC-SN-treated) cells
consisted of mixed populations of cells. In addition to
elongated cells with low or moderate numbers of
organelles and smooth cell outlines, there were plump
cells with irregular shaggy cell margins, prominent
Golgi apparatus, dilated cisternae of RER, and bundles of microfilaments extending beyond the cell
boundaries (Figures 4 and 5). The nuclei were elliptical
or round with indented nuclear membranes. Approximately 25% of these fibroblasts exhibited prominent
nucleoli, and 45% contained simple intranuclear inclusions. In addition, a few cultured fibroblasts from
normal skin biopsies treated with MNC-SN contained
membrane-bound 7,500A intranuclear inclusions (Figure 4a, inset). The ultrastructural characteristics of
many fibroblasts in the experimental cultures resembled those of cultured PSS fibroblasts. However, in
every culture, there were a few cells which looked like
control fibroblasts.
”
DISCUSSION
Experiments performed in this laboratory have
shown that PSS and normal fibroblasts behave differently in culture (5,6) and respond differently to the
modulatory factors in supernatants of activated MNC
(15,16). It is not known what accounts for the observed differences in the growth and the biochemical
properties of these fibroblasts, or more generally, how
previously normal fibroblasts come to express dif-
61
MNC FACTORS INDUCE PSS-LIKE PHENOTYPE
for many weeks in medium containing products of
activated MNC. The resultant population showed a
high level of spontaneous GAG synthesis and a further
small increase in the level of GAG synthesis in response to a challenge with MNC-SN. These are features of PSS fibroblasts in culture ( 5 ) and are in
marked contrast with the behavior of normal control
fibroblasts, which show a low level of spontaneous
Figure 3. Control, untreated cultured fibroblasts outgrowing from
normal skin biopsy material (sixth passage). Note the elongated
shapes of cells and nuclei, scanty intracytoplasmic organelles, and
prominent nucleoli (Nc). RER = cisternae of rough endoplasmic
reticulum (magnification x 9,240).
B
ferent phenotypes in disease. Our results indicate that
manipulation of the environment of a population of
normal adult dermal fibroblasts can result in the permanent adoption of PSS-like characteristics by that
population.
Normal adult dermal fibroblasts were cultured
Figure 4. A and B, Cultured fibroblasts outgrowing from normal
skin biopsy material treated with supernatants of activated mononuclear cells (third passage). Note the oval or rounded cells with
irregular cell outlines, indented nuclei (N), simple intranuclear
inclusions (Inc), membrane-bound intranuclear inclusions (M inc)
(inset), prominent rough endoplasmic reticulum (RER),abundant
pinocytotic vesicles (Pv), and intracellular and extracellular bundles
of microfilaments (Bf). (Original magnification x 9,240; inset
X 12,540.)
62
WORRALL ET AL
B
Figure :5. A and B, Cultured progressive systemic sclerosis (PSS,
scleroderma) fibroblasts (PSS control, third passage). Note the oval,
rounded cells with shaggy cell outlines, irregularly indented nuclear
membrane (N), dilated cisternae of rough endoplasmic reticulum
(RER),and bundles of filaments (Bf) extending beyond the cell
boundaries. (Original magnification of A X 10,900 and of B X
11,800.)
GAG synthesis, but an enormous proportional increase in response to challenges with MNC-SN. In our
studies, prolonged exposure to MNC-SN changed the
behavior of normal fibroblasts, causing them to resemble PSS fibroblasts, and more importantly, this
changed behavior was evident for many weeks after
the removal of MNC-SN and return to normal culture
conditions.
One of the changes brought about by MNC-SN
was a (consistent decrease in growth of normal dermal
fibroblasts, as evidenced by the lower cell counts in
the experimental cultures. Postlethwaite and Kang
(12) have reported that leukocyte-derived factors in-
creased proliferation of foreskin fibroblasts, as measured by 3H-thymidine uptake into cells and cell
enumeration. With adult dermal fibroblasts in log or
confluent cultures, we do not observe increased proliferation following exposure to MNC-SN (16). The
differences in the proliferative response may be due to
the nature of responding fibroblasts or to the different
content of the stimulatory versus inhibitory factors of
crude or partially purified MNC-SN used in these
experiments (12,16).
A number of mechanisms may be postulated to
explain the changed behavior of normal fibroblasts
treated with MNC-SN. For example, if the initial
population of normal fibroblasts is heterogeneous with
respect to rates of GAG synthesis, then exposure to
MNC-SN may constitute an environmental change
which selectively favors those cells with high rates of
GAG synthesis or disfavors those with low rates of
synthesis, or both. The favored fibroblasts will overgrow others, and the rate of GAG synthesis of the
population as a whole will eventually be increased.
Clones of human dermal fibroblasts are known
to be heterogeneous with respect to collagen synthesis
(20) and response to MNC-SN in terms of growth and
prostaglandin Ez (PGE2) synthesis (21,22). Moreover,
Botstein et a1 (20) showed that serum from PSS
patients augmented the growth of high collagenproducing fibroblast clones while inhibiting low
collagen-producers; that is, the high collagen-producers were selectively favored. It is possible that
MNC-SN, like serum from PSS patients, can function
as a mechanism of selection, and the evidence we have
provided is consistent with this possibility. However,
we cannot rule out the possibility that during exposure
to MNC-SN, more rapidly proliferating fibroblasts are
lost from the culture, and a concomitant loss of
feedback inhibition in cultures with lower cell densities leads to increased rates of GAG synthesis. Recently, Korn et al have shown that MNC-SN are
capable of increasing PGE2 synthesis and suppressing
growth of fibroblast clones derived from human neonatal foreskin and then exposed transiently to MNCSN (22). In our experiments featuring long-term exposure to MNC-SN, the observed increases in GAG
synthesis per cell and per culture in experimental lines
appear to be more consistent with the selection hypothesis rather than with there being a lack of feedback inhibition in cultures with lower cell density.
A second explanation of the permanent change
in behavior of fibroblasts exposed to MNC-SN could
be the passage of “toxin” from the cells to their
MNC FACTORS INDUCE PSS-LIKE PHENOTYPE
progeny (23). In the work reported here, each experimental culture originated from the same skin biopsy as
its corresponding normal control, and all cultures
except 1 (JC) were passaged and assayed in synchrony. It is unlikely, therefore, that exogenous
“toxin” would contaminate only the experimental
cultures or that endogenous toxin (presumably present
in the original skin biopsy) would be selectively transmitted to experimental cells and not to their normal
controls. Passage of material present in the MNC-SN
applied to the experimental cells is, however, a possibility which cannot be discounted, but it is one which,
at present, cannot be tested.
A third possible explanation is that MNC-SN
may contain substances which either alter the genetic
code or alter its expression in experimental fibroblasts.
Although at this time we have no evidence for genetic
changes, they could be involved. Certain known
mutagens, such as cytotoxic drugs, radiation, and free
radicals, are known to cause gross changes in behavior
and morphology. Cells of abnormal shape and size are
generated, and there may be death of some cells with
malignant change and formation of an “immortal”
strain by the survivors. Such dramatic changes were
not observed in our cultures. In contrast, we found, as
Korn had reported (2 l), that experimental fibroblasts
showed no gross changes in morphology (by light
microscopy) or growth, and normal senescence after
approximately 30 cell divisions in culture was shown
by experimental and control cultures alike. On the
ultrastructural level, definite changes were observed in
a majority of experimental fibroblasts, but these
changes appeared to be more consistent with cellular
hyperactivity rather than with genetic alterations (24).
If, indeed, exposure to MNC-SN acts as a
selective mechanism favoring high producers of connective tissue substances, then the following model for
the pathogenesis of fibrosis in PSS may be proposed.
Fibroblasts in normal tissues exhibit individual diversity in rates of synthesis of connective tissue substances, but the average rate of synthesis is low.
Activated mononuclear cells, which are effectors of
the stimulated immune system, infiltrate the tissues
and secrete in the microenvironment potent factors
known to modulate behavior of fibroblasts. The immediate result is acute inflammation, but prolonged exposure to these factors over many months may result
in overgrowth by a population of fibroblasts committed to high rates of synthesis of connective tissue
substances. This would lead to the fibrosis, with
63
thickening and loss of elasticity, which is seen clinically in PSS.
Results of this work show that mediators of
inflammation can cause persistent changes in the behavior of populations of fibroblasts, and lend support
to the growing body of circumstantial evidence in
favor of an immunologic selection mechanism in the
pathogenesis of fibrosis in PSS. It is clear that disorders of immunity may be important in this disease.
However, although there is some evidence for aberrant responsiveness to Cundidu (25), a skin antigen
(26), and collagen (27) in PSS patients, the original
stimulus to the immune system, which is presumed to
initiate the disease process, remains unknown.
REFERENCES
1 . Rodnan GP: Progressive systemic sclerosis (scleroderma), Arthritis and Allied Conditions. Ninth edition.
Edited by DJ McCarty. Philadelphia, Lea & Febiger,
1979, pp 762-809
2. LJitto J, Helin G, Helin P, Lorenzen I: Connective tissue
in scleroderma: a biochemical study on the correlation
of fractionated glycosaminoglycans and collagen in human skin. Acta Derm Venereol (Stockh) 51:401406,
1971
3. Fleischmajer R, Perlish JS: Glycosaminoglycans in scleroderma and scleredema. J Invest Dermatol 58:
129-132, 1979
4. Rodnan GP, Lipinski E, Luksick J: Skin collagen content in progressive systemic sclerosis and localized
scleroderma. Arthritis Rheum 22: 130-140, 1979
5. Buckingham RB, Prince RK, Rodnan GP: Progressive
systemic sclerosis (PSS, scleroderma) dermal fibroblasts
synthesize increased amounts of glycosaminoglycan. J
L,ab Clin Med 101:659-669, 1983
6. Buckingham RB, Prince RK, Rodnan GP, Taylor F:
Increased collagen accumulation in dermal fibroblast
cultures from patients with progressive systemic sclerosis (scleroderma). J Lab Clin Med 925-21, 1978
7. L,eRoy EC: Increased collagen synthesis by scleroderma
skin fibroblasts in vitro: a possible defect in the regulation or activation of the scleroderma fibroblasts. J Clin
Invest 54:880-889, 1974
8. Jiminez SA, Bashey RI: Collagen synthesis by scleroderma fibroblasts in culture (letter). Arthritis Rheum
201902-903, 1977
9. Fleischmajer R, Perlish JS, Reeves JRT: Cellular infiltrates in scleroderma skin. Arthritis Rheum 20:975-984,
1977
10. Dumonde DC, Wolstencroft RA, Panayi GS, Mathew
M, Morley J, Dawson WT: Lymphokines: non-antibody
mediators of cellular immunity generated by lymphocyte
activation. Nature 224:3842, 1969
WORRALL ET AL
1 1 . Anastassiades TP, Wood A: Effect of soluble products
from lectin-stimulated lymphocytes on the growth, adhesiveness and glycosaminoglycan synthesis of cultured
synovial fibroblastic cells. J Clin Invest 68:792-802, 1981
12. Postlethwaite AE, Kang AH: Induction of fibroblast
proliferation by human mononuclear leukocyte-derived
proteins. Arthritis Rheum 26:22-27, 1983
13. Johnson R L , Ziff M: Lymphokine stimulation of
colllagen accumulation. J Clin Invest 58:240-252, 1976
14. Jirnenez SA, McArthur W, Rosenbloom J: Inhibition of
collagen synthesis by mononuclear cell supernates. J
Exp Med 150:1421-1431, 1979
15. Whiteside TL, Buckingham RB, Prince RK, Rodnan
GP: Products of activated mononuclear cells modulate
accumulation of collagen by normal dermal and scleroderma fibroblasts in culture. J Lab Clin Med 104:
355-370, 1984
16. Whiteside TL, Worrall JG, Prince RK, Buckingham RB,
Rodnan GP: Soluble mediators from mononuclear cells
increase the synthesis of glycosaminoglycan by dermal
fibroblast cultures derived from normal subjects and
progressive systemic sclerosis patients. Arthritis Rheum
28:18&197, 1985
17. Herman JH, Nutman TB, Nozoe M, Mowery CS,
Dennis MV: Lymphokine-mediated suppression of
chondrocyte glycosaminoglycan and protein synthesis.
Arthritis Rheum 242324834, 1981
18. Rodnan GP, Jablonska S, Medsger TA: Classification
and nomenclature of progressive systemic sclerosis
(sclieroderma). Clin Rheum Dis 5:5-13, 1979
19. Scott JE: Aliphatic ammonium salts in the assay of
acidic polysaccharides from tissues, Methods of Bio-
chemical Analysis. Vol. VIII. Edited by D Glick. New
York, Interscience Publishers, 1960, pp 145-197
20. Botstein GR, Sherer GK, LeRoy EC: Fibroblast selection in scleroderma: an alternative model of fibrosis.
Arthritis Rheum 25: 189-195, 1982
21. Korn JH: Fibroblast PGE2 synthesis: persistence of an
abnormal phenotype after short term exposure to
mononuclear cell products. J Clin Invest 71: 1240-1246,
1983
22. Korn JH, Torres D, Downie E: Clonal heterogeneity in
the fibroblast response to mononuclear cell derived
mediators. Arthritis Rheum 27: 174-179, 1984
23. Buckingham RB, Castor CW: Rheumatoid behavior in
normal human synovial fibroblasts induced by extracts
of gram-negative bacteria. J Lab Clin Med 85:422435,
1975
24. Williams G: The late phases of wound healing: histological and ultrastructural studies of collagen and elastic
tissue formation. J Pathol 102:6148, 1970
25. Whiteside TL, Kumagai Y , Medsger TA Jr, Rodnan GP:
Discrepancies between in vivo and in vitro responses to
Candida antigen in patients with progressive systemic
sclerosis (PSS, scleroderma). J Clin Immunol 1 :250-256,
1981
26. Kondo H , Rabin BS, Rodnan GP: Cutaneous antigen
stimulating lymphokine production by lymphocytes of
patients with progressive systemic sclerosis (scleroderma). J Clin Invest 58:138&1394, 1976
27. Stuart J, Postlethwaite AE, Kang AH: Evidence of
cell-mediated immunity to collagen in patients with
progressive systemic sclerosis. J Lab Clin Med 88:
601-607, 1976
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 160 Кб
Теги
like, persistence, exposure, phenotypic, soluble, prolonged, mediators, norman, scleroderma, mononuclear, cells, fibroblasts
1/--страниц
Пожаловаться на содержимое документа