close

Вход

Забыли?

вход по аккаунту

?

Type ii collagen-induced arthritis. A Morphologic and Biochemical Study of Articular Cartilage

код для вставкиСкачать
1245
TYPE I1 COLLAGEN-INDUCED ARTHRITIS
A Morphologic and Biochemical Study of Articular Cartilage
DANIEL P. DESIMONE, DIANE B. PARSONS, KURT E. JOHNSON, and ROBERT P. JACOBS
Articular cartilage was obtained from type I1
collagen-induced arthritic rat joints. Transmission electron nticroscopy showed a gradual degeneration of
chondrocytes, disorganization of the collagenous extracellular matrix, and formation of microscars. Biochemical analyses indicated that type I1 collagen was the only
collagen present and that it was normal in regard to
hydroxylation of lysine and glycosylation of hydroxylysine. Analyses of the proteoglycan in the extracellular
matrix revealed a 50% loss of chondroitin sulfate and
keratan sulfate.
Various experimental arthritides have been
studied in an attempt to elucidate the underlying
mechanism involved in cartilage and bone destruction
during rheumatoid arthritis (RA) (1-6). Despite the
variety of methods used to produce an experimental
synovitis typical of RA, none has proven strictly
analogous to the human disease. An exception to this
appears to be type I1 collagen-induced arthritis in rats
(7-1 1). This model requires a single intradermal injection of native type I1 collagen emulsified in adjuvant
oil, and results in an inflammatory arthritis involving
Presented in part at the VIII Pan-American Congress of
Rheurnatology. Washington, DC, June 1981.
From the George Washington University School of Medicine and Allied Health Sciences, Washington, DC
Supported in part by the Washington Metropolitan Arthritis
Foundation.
Daniel P. DeSimone, PhD: Department of Anatomy; Diane
B. Parsons, PhD: Department of Orthopaedic Surgery; Kurt E.
Johnson, PhD: Department of Anatomy; Robert P. Jacobs, MD:
Department of Medicine.
Address reprint requests to Dr. Daniel P. DeSimone, 3828
25th Street N , Arlington. VA 22207.
Submitted for publication February 14, 1983; accepted in
revised form May 2, 1983.
Arthritis and Rheumatism, Vol. 26, No. 10 (October 1983)
the tarsal and interphalangeal joints within 3 weeks of
primary immunization. Many features of this model,
such as its proliferative synovitis, marginal erosion,
destruction of articular cartilage and bone, and immunologic response to collagen, closely resemble human
RA and suggest that it may be a suitable model for
human disease.
This paper provides ultrastructural and biochemical data concerning the sequential changes
which develop within the articular cartilage of inflamed rat joints during a 6-month period following
primary immunization with type I1 collagen.
MATERIALS AND METHODS
Rats. Ten-week-old female Wistar-Lewis rats, 175250 gm, were obtained from Charles River Breeding Laboratories Inc., Wilmington, MA.
Pepsin solubilization of chicken sternum. Sternal cartilage from 8-week-old chickens (Pel Freeze Laboratories,
Rogers, AR) was dissected free of perichondrium, minced,
lyophilized, and pulverized in liquid nitrogen using a freezer
mill (Spex Industries, Metuchen, NJ). The cartilage powder
was gently stirred for 3 hours in 2M MgCIIOSM Tris (pH
7.4). The resultant suspension was centrifuged for 1 hour at
10,OOOg. The residue was washed twice with cold distilled
water. All subsequent procedures were performed at 4°C.
Pepsin solubilized collagen was prepared as described by
Trentham et a1 (7) by suspending the residue in O.05M acetic
acid adjusted to pH 2.5 with formic acid. Pepsin (2x
crystallized, Worthington Biochemical Corp., Freehold, NJ)
was added at 1/50 gm wet weight and digestion continued for
72 hours with constant shaking. The solubilized collagen was
removed following centrifugation at 10,OOOg for 1 hour, and
the remaining residue was digested twice more under identical conditions,
Purification of pepsin solubilized type I1 collagen.
Pepsin solubilized collagen was purified according to a
modification of Miller’s method (12). A SO-ml aliquot of
1246
solubilized collagen was exhaustively dialyzed against 2
liters of 0.2M NaCVO.05M Tris (pH 7.4). Partial purification
of the extracted collagen was achieved by applying 50 ml at a
flow rate of 100 ml per hour to a column (2.5 x 15 cm) of
DEAE-cellulose (DE-52, Whatman Ltd. Clifton, NJ) which
was equilibrated with 0.2M NaCIIO.05M Tris (pH 7.4). Fivemilliliter fractions were collected over a total volume of 250
ml and the absorbance determined at 230 nm. The eluted
collagen was exhaustively dialyzed against O.05M acetic acid
and lyophilized. Phosphocellulose chromatography of CNBr
peptides and amino acid analysis of the lyophilized material
revealed it to be pure type I1 collagen.
Sensitization of animals. The purified collagen was
dissolved in cold O.1M acetic acid at a concentration of 5
mg/ml. Equal volumes of this collagen solution and incomplete Freund’s adjuvant were mixed at 4°C to make a stable
emulsion. A total of 0.1 ml of this cold emulsion was
immediately injected into the left hind footpad of 10-weekold rats (10).
Light microscopy and histochemistry. The distal hind
limbs of age-matched control and diseased rats were fixed
for a minimum of 3 days in 4% aqueous formaldehyde
containing 0.5% cetylpyridinium chloride (13) and decalcified for 12 hours in Rapid Decal (Dupage Kinetics, Naperville, IL). Specimens were dehydrated through a series of
graded ethanol, cleared with xylene, and embedded in
paraffin. Six-micrometer sections were cut with a rotary
microtome and stained with hematoxylin and eosin. In
addition, Safranin 0-fast green staining (14) was used to
demonstrate the presence of acidic proteoglycans.
Transmission electron microscopy. The articular cartilage from age- and site-matched control and diseased rat
joints was shaved with a scalpel blade from the subchondral
bone and immediately fixed according to the method of
Hirsch and Fedorka (15) as described by Holtrop (16) for
cartilage. Thin sections (40-60 nm) were stained with Reynolds lead citrate (17) and examined on a JEM 100-S or JEM
100-B electron microscope using an accelerating voltage of
60 kV.
Biochemical characterization. Articular cartilage was
studied from age- and site-matched control and diseased rat
joints. The talus and tibia were removed and soaked for 24
hours in O.5M EDTA, pH 7.4. The articular cartilage was
removed from the subchondral bone with jewelers forceps
using a dissecting microscope and pooled according to the
time of the disease and lyophilized.
Lyophilized cartilage samples (0.5-2.0 mg) were
hydrolyzed in 6N HCI in sealed pyrex tubes at 105°C for 24
hours. The hydrolysates were dried on a Buchler Evapo-Mix
(Buchler Instruments Company, Fort Lee, NJ) and washed
with distilled water 3 times. The dried hydrolysates were
diluted to 1 rnl with distilled water and a known aliquot used
for hydroxyproline determination. Hydroxyproline content
was measured by the method of Stegemann (18) and the
collagen content calculated per mg of dry tissue by the
method of Eyre et a1 (19).
Digestion with cyanogen bromide (CNBr). Lyophilized cartilage samples (4-6 mg) were suspended in 70%
formic acid in a 25-ml flask and flushed with nitrogen. An
amount of CNBr equal to the weight of the sample was
added and the flask sealed under nitrogen. The samples were
DESIMONE ET AL
digested with gentle shaking for 4 hours at 30°C. The digest
was centrifuged for 15 minutes at 40,OOOg and the supernatant containing the freed peptides was diluted 10 times with
cold distilled water and lyophilized (20).
Phosphocellulose chromatography of the CNBr peptides. The molecular identity of the collagen in the samples of
articular cartilage was determined by phosphocellulose chromatography of the CNBr peptides according to the method
of Eyre and Muir (21). The lyophilized peptides were separated on a 1.5 x 10 cm column of phosphocellulose (Whatman P11, Whatman Inc.) which was equilibrated with starting buffer (1 mM sodium acetate, pH 3.6) and maintained at
43°C. Digested samples weighing between 4 and 6 mg were
dissolved in 5 ml of the starting buffer and applied to the
column. The CNBr peptides were eluted using a linear
gradient of 0-0.3M NaCl, between 400 ml each of starting
buffer and limiting buffer (1 mM sodium acetate/0.3M NaC1,
pH 3.6) at a flow rate of 70 ml per hour. The eluent was
continuously monitored at 230 nm using a Gilson Holochrome spectrophotometer (Gilson Medical Electronics,
Middletown, WI). The specific peptides, al(I)CB2 characteristic of type I collagen and al(II)CB6 characteristic of
type I1 collagen, were identified and verified by comparison
with known elution profiles.
Hydroxylysine content. Samples weighing 2 mg were
placed in vacuum ampules containing 1 ml 6N HCI. The
ampules were flushed with nitrogen and evacuated. This
process was repeated 3 times and the ampules were sealed
under vacuum. The samples were hydrolyzed for 24 hours at
105°C and dried on a Buchler Evapo-Mix. The residue was
dissolved in 1 ml of 0.2N sodium citrate, pH 2.2 and passed
through a millipore filter (0.22 pm). Aliquots were analyzed
by chromatography on a 60 cm column of an amino acid
analyzer at 57°C. Separation of hydroxyproline and hydroxylysine was accomplished using 2 buffers. Hydroxyproline
was eluted with 0.2N sodium citrate, pH 2.92 after which the
hydroxylysine was eluted with 0.35N sodium citrate, pH
5.28. The flow rate was maintained at 90 ml per hour. The
hydroxylysine content was quantified as the number of
hydroxylysine residues per 100 residues of hydroxyproline.
Hydroxylysine glycosylation. Hydroxylysine glycosylation of all tissue samples was determined according to the
method of Eyre and Muir (22). Samples weighing approximately 2 mg were placed in 12 x 75 mm polyethylene tubes
containing 1 ml of 2N NaOH. The tubes were placed into 18
x 150 mm pyrex tubes, sealed, and hydrolyzed for 24 hours
at 110°C. Following hydrolysis, 2 ml of distilled water and 1
ml of 1M citric acid were added to the tubes and the pH
adjusted to 4.0. Aliquots were analyzed by chromatography
on a 60-cm column of the amino acid analyzer at 50°C.
Fractionation of hydroxylysine and its glycosides, glucosylgalactosylhydroxylysine (GGH) and galactosylhydroxylysine (GH) was accomplished using 0.35M sodium citrate, pH
5.28 as the buffer. The flow rate was maintained at 90 ml per
hour. The net absorbance of non-glycosylated hydroxylysine
and the net absorbance of GGH and GH were measured, and
the percentage of total hydroxylysine glycosylation determined.
Hexosamine content. The hexosamines, glucosamine,
and galactosamine were measured according to the method
of Ford and Baker (23). Tissue samples (2 mg) were placed in
1247
TYPE I1 COLLAGEN ARTHRITIS
pyrex tubes with 2 ml 6N HCl. The tubes were sealed and
hydrolyzed at 105°C for 7 hours. The hydrolysates were
washed once with distilled water and dried. The residues
were diluted to 5 ml with 0.2N sodium citrate, pH 2.2 and
passed through a millipore filter (0.22 pm). Aliquots were
chromatographed on a 60-cm column of an amino acid
analyzer using a buffer of 0.35M sodium citrate, pH 5.28.
The flow rate was maintained at 90 ml per hour and the
column temperature at 57°C. The concentrations of glucosamine and galactosamine were expressed as micrograms
per milligram dry weight of tissue.
RESULTS
Incidence of type I1 collagen-induced arthritis.
Typically, inflammatory disease manifested itself 1824 days after immunization and was characterized by
moderate to severe edema and erythema of the hind
paws. Weight-bearing was poorly tolerated by the
diseased animals. In the various experiments, the
percentage response ranged from 40-70%. The mean
percentage response for 8 groups of 20 animals was
56%. Bilateral swelling was observed in 82%, with no
involvement of the forepaws.
Light microscopy and histochemistry. The
pathogenesis of joint inflammation from its initial
onset through 6 months of disease is shown in Figures
1 and 2. Figure 1A shows a section of a control rat
joint. The synovium consists of a 1- or 2-cell thick
mesothelium which lies on the subsynovial fat and
connective tissue and protrudes into the joint space
from the medial and lateral malleoli. The articular
cartilage covers the articular surfaces of the bones.
Histopathologic studies of the involved proximal and
distal, tibiotalus, tarsal, and interphalangeal joints
A
show lesions which include severe periarticular inflammation. Initially the synovium becomes infiltrated
with inflammatory cells as it undergoes a diffuse
proliferative synovitis (Figure 1B). The synovial lining
cells become hypertrophic and hyperplastic, changing
the single layered synovium into a multilayered mass
of inflammatory tissue infiltrated with lymphocytes,
macrophages, and plasma cells (Figure 2A). The synovial pannus of granulation tissue extends over the
involved articular surface (Figure 2B), burrowing into
the underlying tissue and secondarily destroying cartilage at the joint margins (Figure 2C).
Age-matched control articular cartilage stained
positively with Safranin 0 dye, with particularly
strong reactions in the territorial matrices (Figure 3A).
Safranin 0 positive staining material was markedly
absent from the surface regions of the articular cartilage at the first sign of joint disease and diminished
gradually from the deeper regions by the sixth month
after the onset of type I1 collagen-induced arthritis
(Figures 3B and 3C). At times, isolated chrondrocytes
were observed surrounded by territorial matrix containing Safranin 0 positive staining material (Figure
3C).
Transmission electron microscopy. Representative chondrocytes from the superficial zone taken
from control articular cartilage are shown in Figure 4.
The cells are elongated with their long axis oriented
parallel to the articular surface. The cell surface has
uniformly short cytoplasmic processes approximately
500 nm in length and is in close apposition to the
surrounding extracellular matrix. The nuclear envelope has a smooth contour and the nucleus may have
B
Figure 1. A, Control rat joints showing the articular surfaces of the tibia and talus. The joint is free ofcellular debris and the articular surface is smooth and uninterrupted. The synovial membrane (arrowhead) is shown with the underlying subsynovial fat (hematoxylin and
eosin, original magnification x 40). B, Ankle joint at the onset of arthritis. The joint space contains an intense inflammatory exudate
which overlies areas of the articular surface (hematoxylin and eosin, orginal magnification x 40).
1248
DESIMONE ET AL
A
A
B
B
C
C
Figure 2. A , Articular surface of the talus at 4 months after the onset
of arthritis. The single layered synovium is now a multilayered mass
of inflammatory tissue infiltrated with inflammatory cells (hematoxylin and eosin, original magnification x 130). B,A synovial pannus
of granulation tissue extends over the articular surface (hematoxylin
and eosin, original magnification x 78). C, Tarsal joint 6 months
after the onset of arthritis. The inflammatory exudate and pannus of
granulation tissue have eroded the articular cartilage and are extending into the underlying subchondral bone (hematoxylin and eosin,
original magnification x 40).
Figure 3. Light microscopic histochemical examination of articular
cartilage with Safranin 0 and fast green to demonstrate acidic
proteoglycan complexes. A, Control articular cartilage showing
Safranin 0 positive material uniformly distributed throughout the
extracellular matrix (original magnification x 180). B,At the onset of
arthritis the positive staining material is diminished in the superficial
zone (original magnification x 130). C , Six months after the onset of
arthritis acidic proteoglycan complexes appear only in the territorial
matrices immediately adjacent to the remaining chondrocytes (original magnification x 130).
1249
TYPE I1 COLLAGEN ARTHRITIS
Figure 4. Transmission electron micrograph of the superficial zone of control articular cartilage showing a pair of
chondrocytes. Each has rough endoplasmic reticulum (RER).Golgi (C), and nucleus (N). Cytoplasmic processes extend
from the cell surface. Collagen fibers (arrow) are occasionally observed in the extracellular matrix (original magnification
x 10,000).
one or more nucleoli. These cells contain moderate
amounts of rough endoplasmic reticulum, one or two
Golgi zones, and several mitochondria. Frequently
these cells are observed in pairs. Crossbanding of the
collagen fibers in the extracellular matrix is not particularly apparent; however, collagen fibers with a periodicity of 64 nm are occasionally observed. The length
of the individual fibers is not determinable due to the
planes of section.
Chondrocytes from the deeper zones of articular cartilage lie in clusters of three or more (Figure 5).
These cells are usually larger than those observed in
the superficial zone and are oval in shape, with the
long axis perpendicular to the articular surface. The
cell surface has numerous cytoplasmic processes. In
addition to a large paranuclear Golgi region and numerous mitochondria, the chondrocytes of this zone
contain abundant rough endoplasmic reticulum. Numerous small vesicles are present in the area of the
rough endoplasmic reticulum. Vesicles of intermediate
and large size are also seen in the paranuclear Golgi
'One (Figure 6)*These intermediate and large
sometimes 'Ittain
fine
Some may appear empty. Usually small spherical
inclusions are present inside the large vesicles.
Aggregates of glycogen granules are also common throughout the cytoplasm of these chondrocytes.
Elongated and circular mitochondria are numerous in
the chondrocytes of this zone. Collagen fibers of this
Figure 5. Transmission electron micrograph from the deeper zone of
the control articular cartilage with a cluster of chondrocytes. The
cytoplasm is packed with rough endoplasmic reticulum (RER)and
Golgi (C). The collagen fibers (arrow) in the extracellular matrix
show a random distribution (original magnification x 8.000).
1250
DESIMONE ET AL
Figure 6. Transmission electron micrograph of a chondrocyte from
the middle zone of control articular cartilage. It contains a nucleus,
abundant rough endoplasmic reticulum (RER),glycogen (GLY), and
Golgi (G) with vesicles of varying size which contain fine granular
material (original magnification x 18,000).
region are loosely packed, resulting in a matrix whose
architecture is more random than that of the superficial zone. Collagen fibers in the pericellular zone are
identical in size to those in the interterritorial matrix.
The collagen fibers in this area, corresponding to a
perilacunar rim or territorial matrix, frequently run
parallel to the chondrocyte surface.
The ultrastructural appearance of the articular
cartilage during the pathogenesis of type I1 collageninduced arthritis reveals the development of a fibrinlike pannus, the gradual degeneration and death of
chondrocytes, the formation of microscars similar to
those described by Weiss (24) for osteoarthritic articular cartilage, and the disruption of the normal architecture of the extracellular matrix.
At the first sign of joint disease, synoviocytes
containing little rough endoplasmic reticulum and numerous mitochondria were observed on the articular
surface (Figure 7). The articular cartilage was covered
by variable thicknesses of electron dense material
which resembled fibrin. The chondrocytes from the
superficial region exhibited a loss of ultrastructure and
an increase in the length of cell surface processes
(Figure 8). By 1 month, degenerating chondrocytes
could be seen lying in wide moats of loose amorphous
material (Figures 9 and 10). Nuclear clumping and
deeply invaginated nuclear envelopes are characteristic of these cells. Chondrocytes from deeper regions
revealed myelin-like figures (Figure 1 l), perhaps representing degenerating mitochondria or invaginating
cell surface membrane.
Figure 7. Transmission electron micrograph of the articular surface of the rat talus at the onset of type I1 collagen-induced
arthritis. The articular surface is covered with a fibrous pannus (FP). A macrophage is present in the joint space (JS) along
with fibrous debris (arrow) (original magnification x 18,000).
1251
TYPE I1 COLLAGEN ARTHRITIS
Figure 8. Transmission electron micrograph of the superficial zone of articular cartilage 1 month after the onset of type I1
collagen-induced arthritis. The chondrocyte has lost all cellular detail. A peripheral moat (MT) which is devoid of collagen
fibers surrounds the cell. A fibrous pannus (FP) covers the articular surface (original magnification x 18.000).
As the disease progressed, fewer normal cells
were observed, and more cells showing signs of degeneration and death were apparent. Typically the chondrocytes exhibited nuclear pyknosis and separation of
the nucleus from the cytoplasm, followed by shrinkage
of the entire cell. Degenerate and dying cells appeared
to lie within moats of amorphous electron dense
material (Figure 12).
Figure 9. Transmission electron micrograph from the superficial zone of articular cartilage 6 months after the onset of type
11 collagen-induced arthritis, showing the joint space (JS).The remains of the degenerate chondrocytes lie in a pericellular
moat (MT). Scattered electron dense debris is present in the matrix (original magnification x 15,000).
1252
DESIMONE ET AL
Figure 10. Transmission electron micrograph of the superficial region of articular cartilage 1 month after the onset of
type I1 collagen-induced arthritis. Electron dense remnants of chondrocytes (C) remain in lacunae along with
banded collagen fibers (arrow). A third chondrocyte is present with very dense cytoplasm and a pericellular moat
(MT) (original magnification X 10,OOO).
Figure 11. Transmission electron micrograph from the deeper region of articular cartilage 1 month after the onset of
type XI collagen-induced arthritis. A pair of chondrocytes shows early degenerative changes which include
numerous myelin-like figures (arrow). Glycogen (GLY) is also present (original magnification X 15,000).
1253
TYPE I1 COLLAGEN ARTHRITIS
Figure 12. Transmission electron micrograph of a pair of chondrocytes 3 months after the onset of type I1 collagen-induced arthritis
showing varying degrees of degeneration. The nucleus (N)of the cell
on the left is present as well as mitochondria ( M ) . The peripheral
rims of the pericellular moats are present (arrows) (original magnification x lO.000).
By 3 months, collagen fibers of extremely large
diameter were apparent, filling in the spaces surrounding the degenerating chondrocytes (Figures 13 and 14).
They ranged from 500 to 1,250 nm in diameter. These
large atypical fibrils were seen to be poorly integrated
aggregations of smaller fibrils. In certain regions these
aggregates were separate and parallel, while at other
sites they were consolidated into branched subdivisions. The periodicity was maintained across the entire aggregated fibril and ranged from 160 to 250 nm.
The anastomosing fibrils sometimes showed a helical
twisting pattern (Figure 15).
Collagen and hexosamine contents. The percent
dry weight of collagen in the articular cartilage, as
determined by the total hydroxyproline content, increased with the duration of the disease (Figure 16).
The sharpest increase in collagen content appeared
immediately at the onset of inflammation, when the
percent collagen was 69% as opposed to 54% in the
control articular cartilage. Six months after the onset
of the disease the total percent collagen of the dry
weight reached 90%. The collagen in cartilage is
closely associated with proteoglycan (25). Therefore.
in this investigation, glucosamine and galactosamine
were measured. These sugars are the respective amino
sugars of keratan sulfate and chondroitin sulfate. Figure 16 shows the total amount of hexosamine, at
various times during the pathogenesis of this disease.
By 6 months the loss of proteoglycan from the cartilage extracellular matrix amounted to over 50%.
Figure 13. Transmission electron micrograph from the superficial zone of articular cartilage 3 months after the onset of type
I1 collagen-induced arthritis. The chondrocyte shows dense cytoplasm with very little cellular detail. Elongated
cytoplasmic processes are present. The pericellular region has begun to fill in with collagen tibers (arrows) of varying
diameter size (original magnification x 15.000).
1254
DESIMONE ET AL
Figure 14. Transmission electron micrograph from the superficial
zone of articular cartilage 6 months after the onset of type I1
collagen-induced arthritis. Remnants of a degenerating cell are
seen. The pencellular moat (MT) contains fibers (arrows) of large
diameter (original magnification x 13,000).
Phosphocellulose chromatography of CNBr peptides. The elution profiles of CNBr peptides derived
from CNBr digestion of articular cartilage were identical at every time period examined. Typical elution
profiles of control and diseased tissues are represented
in Figure 17. The characteristic type I1 peptide
al(II)CB6 was demonstrated at each time interval
examined with no appearance of al(I)CB2 characteristic of type I collagen. This indicates that only type I1
collagen was detectable in the articular cartilage.
Hydroxylysine content and glycosylation of hydroxylysine. The hydroxylysine content of collagen in
the control articular cartilage was 21.1 residues per 100
hydroxyproline residues. The hydroxylysine content
did not change with increasing duration of the disease;
at 6 months after the onset of type I1 collagen-induced
arthritis, it was 20.9 residues per 100 residues of
hydroxyproline.
Type I collagen has approximately 17-36% of
its hydroxylysine residues glycosylated. In type I1
collagen, hydroxylysine is 60-70% glycosylated. It is
generally agreed that glycosylation is a better indication of collagen type than hydroxylysine content itself
(26). Total hydroxylysine glycosylyation (GGH + GH)
was quantified. At each time examined during the
pathogenesis of the disease, more than 70% of the
hydroxyIysine residues were glycosylated. The data
corroborated the findings of the phosphocellulose
chromatography and further support the presence of
Figure 15. Transmission electron micrograph of microscar present in the superficial zone of the articular cartilage 6 months
after the onset of type I1 collagen-induced arthritis. The periodicity of the fibers ranges from 160 to 250 nm. These large
atypical fibers (single arrow) appear to be poorly integrated aggregations of smaller fibers. Normal collagen fibers are
present in the interterritorial matrix (double arrow) (original magnification x 20,000).
TYPE I1 COLLAGEN ARTHRITIS
1255
.05
type I1 collagen alone in the extracellular matrix of the
diseased articular cartilage.
kL
DISCUSSION
The results of this study show that the histologic and biochemical characteristics of articular cartilage, at various times after the onset of type I1
collagen-induced arthritis, differ from those of normal
age- and site-matched control articular cartilage. Electron microscopic examination of the articular cartilage
during the disease showed the development of a fibrous pannus, the gradual degeneration of chondrocytes, the formation of microscars, and the disruption
of the normal architecture of the extracellular matrix.
Our results confirm and extend the studies of the
synovium in this disease recently reported by Caulfield et a1 (1 1).
Collagen content of the articular cartilage appears to increase from 54% in the control age-matched
articular cartilage to 90% in articular cartilage exam-
0 ’
0
,051
..
I
I
100
200
n
AGE AND SITE MATCHED
DIAPHYSEAL BONE
AGE AND SITE MATCHED
a’(n)CB6
ARTICULAR CARTILAGE
U
8
a .05
ARTICULAR CAFITLAGE
6 MONTHS FOLLOWING
THE ONSET OF TYPE II
COLLAGENINDUCED
0
lZ01
100
mmRrns
200
EFFLUENT VOLUME, ml
Figure 17. Phosphocellulose chromatograms of CNBr digested ageand site-matched control articular cartilage, and articular cartilage 6
months after the onset of type I1 collagen-induced arthritis. The
samples ( 5 mg) were applied to a column (1.5 x 10 cm) after
dissolution in 1 mM sodium acetate, pH 3.6. The CNBr peptides
were eluted with a linear gradient of 0-0.3M NaCl forced between
400 ml each of 1 m M sodium acetate and I mi4 sodium acetate/0.3M
NaCI, pH 3.6 at a flow rate of 70 ml per hour. The eluent was
continuously monitored at 230 nm.
>I
-COLLAGEN
0
G
=
m
40-
---
CONTENT
HEXOSAMINE CONTENT
-
I
I
I
I
I
I
Figure 16. Changes in collagen content expressed as micrograms
hydroxyproline per milligram dry articular cartilage, and the total
hexosamine content (glucosamine + galactosamine) expressed as
micrograms per milligram dry articular cartilage at sequential time
intervals following the onset of type I1 collagen-induced arthritis.
All values represent pooled tissue at each time examined.
ined 6 months after the onset of joint inflammation.
Phosphocellulose chromatography of CNBr liberated
peptides .from the diseased articular cartilage at various times during the progression of the disease indicates that the collagen present in the extracellular
matrix is type 11. Analysis of the hydroxylation of
lysine and the glycosylation of hydroxylysine provides
further evidence that only type I1 collagen is present
and that it is normal with regard to these post-transitional modications.
The observed increase in the percent collagen
may indicate one of two activities. Either there is an
increase in the amount of collagen present or there is a
substantial loss of another component of the extracellular matrix. Additional biochemical data reported
1256
here reveal that this steady increase in collagen content is accompanied by a steady decline in the hexosamine content, presumably representing a loss of both
keratan sulfate and chondroitin sulfate.
Histochemical demonstration of acidic proteoglycans in the extracellular matrix of articular cartilage
was studied at various times following the onset of the
disease. Although an absolute numeric correlation is
not possible, microscopic examination of tissue from
normal and diseased animals showed a crude relationship between the levels of proteoglycans and the
intensity of the positive staining material in the extracellular matrix. It therefore seems most likely that the
apparent increase in collagen is not a true increase, but
rather a result of the loss of proteoglycan from the
matrix. We cannot definitively rule out the possibility
that the collagen content did increase. However, our
alternative interpretation is in accord with the findings
of Mankin et a1 (27) and Nimni and Deshmukh (28) for
osteoarthritic human articular cartilage, in which collagen content was determined after extraction of proteoglycan and was shown to be unchanged.
Although biochemical studies have been performed using animal models for inflammatory arthritis,
ultrastructural observations of articular cartilage during the progression of inflammatory arthritis are fragmentary. Electron microscopic studies undertaken
here reveal that chondrocytes from diseased articular
cartilage show nuclear clumping and a marked loss of
the biosynthetic machinery necessary to maintain normal protein synthesis. These results indicate that
perhaps the reason for the loss of biosynthetic activity
demonstrated by previous investigators (6,29,30) is the
degeneration of the chondrocytes in all zones of the
articular cartilage during inflammatory arthritis.
Degeneration of chondrocytes is also a well
known feature of human rheumatoid arthritis (31,32).
Numerous factors may lead to the observed chondrocyte degeneration. The fibrous pannus may play a role
in the degeneration of the chondrocytes, merely by
preventing the diffusion of nutrient material from the
synovial fluid into the extracellular matrix of the
articular cartilage. This does not seem to be a major
factor however, since degenerate chondrocytes were
observed in all areas of the articular cartilage examined and no qualitative differences in the degree of
degeneration were apparent in relation to the amount
of fibrous pannus present.
The degree of degeneration was, however, related to the proteoglycan loss-the greater the proteoglycan loss, the more advanced the chondrocyte de-
DESIMONE ET AL
generation. This suggests, but does not establish, that
proteoglycan serves a protective role for the chondrocytes. Abnormal compressive forces due to the significant loss of proteoglycan (33) may be responsible for
physically injuring surface chondrocytes. Normal cartilage is impermeable to large molecules (34,35) and
therefore occupies a somewhat privileged site in the
body. Milroy and Poole (36) have shown that enzymatic depletion of the cartilage matrix allows large molecules to enter the tissue. Diffusion of high molecular
weight cytotoxic factors or immunoglobulins into the
extracellular space may be facilitated by the large
proteoglycan loss. Thymic derived T cells in the
synovial membrane, which are known to accelerate
both proteoglycan breakdown and inhibition of proteoglycan synthesis (37,38), may be responsible for the
degenerative changes seen in chondrocytes. Chondrocytes may be more directly acted upon by the elevated
levels of proteases derived from the polymorphonuclear leukocytes present in the inflamed joints (39,40).
Increased activity of lysosomal enzymes leading to the degradation of the articular cartilage matrix
has been observed in the synovial fluid and cartilage of
joints involved in inflammatory arthritis (5). .If lysosoma1 enzymes produced either by the synovial membrane or the chondrocytes themselves alter the microenvironment of the cells causing stimulation of type I
collagen synthesis, as suggested by Deshmukh and
Nimni (41) and Deshmukh and Hemrick (3,it is
plausible that during the initial, highly inflammatory
stages of disease in this experimental model, type I
collagen would have been detected in addition to type
I1 collagen. It is reasonable that at further advanced
stages of this disease, when the entire articular surface
is eroded to the subchondral bone, fibrocartilage is
formed by cellular proliferation, and hence a mixture
of type I and type I1 collagens would be present in the
extracellular matrix. Certainly the findings of Deshmukh and Nimni (41) and Deshmukh and Hemrick ( 5 )
do not represent a complete resynthesis of the articular surface with the inappropriate type of collagen.
Anderson et a1 (42) have demonstrated many
large collagen fibers in the extracellular matrix during
chondrodystrophy, suggesting that chondroblasts secrete a molecule capable of regulating the width of the
collagen fibers and that this material might well be
proteoglycan. Seegmiller et a1 (43), Stephens and
Seegmiller (44), and Byers et a1 (45) have suggested
that abnormalities in fibrillogenesis might be due to
aberrations of the noncollagenous extracellular matrix
components that influence collagen fiber formation.
TYPE I1 COLLAGEN ARTHRITIS
More recently Vogel et a1 (46) have demonstrated a
similar morphologic appearance of the collagen fibers
in the gravis form of Ehlers-Danlos syndrome. Large
extracellular banded structures have likewise been
described by Chew et a1 (47) in paralytic scoliotic
intervertebral disc.
Cartilage stiffness is partly dependent upon
proteoglycan. Depletion of this component of the
extracellular matrix as demonstrated in this study may
result in the collapse of the collagenous architecture of
the articular cartilage. Degenerating chondrocytes
leave behind a microscar similar t o those described by
Weiss (24) in osteoarthritic articular cartilage. Rather
than leaving behind a flaw in the articular cartilage,
these microscars appear t o assist in maintaining the
integrity of the collagenous network.
T h e use of type I1 collagen-induced arthritis in
rats as a model for human rheumatoid arthritis has
certain advantages over other animal models. Type I1
collagen is a logical choice for the inciting antigen
because it is the major constituent of articular cartilage. It is also more applicable because of the increased appearance of antibodies t o type I1 collagen in
the serum and synovial fluid of patients with RA.
Retrospective clinical studies by Trentham et a1 (48)
report that humans exhibit the same cellular and
humoral responses a s the diseased rats. T h e single
intradermal administration of the antigen in an oil
vehicle avoids the necessity of presensitization and
repeated intraarticular injections required of other
antigen induced arthritides. Finally and most importantly, the chronic nature of the disease may be
explained by the slow elimination of type I1 collagen,
which is unmasked by the loss of proteoglycan from
the articular cartilage into the joint space, where
complement binding antigen-antibody complexes may
trigger the observed inflammatory response.
REFERENCES
1. Ungar G, Damgaard E, Weinstein HG: Endocrine control of inflammation: effect of hormones on anaphylactic
arthritis. Am J Physiol 166:340-348, 1951
2. Dumonde DC, Glynn LE: The production of arthritis in
rabbits by an immunological reaction to fibrin. Br J Exp
Pathol 43:373-383, 1962
3. Consden R, Doble A, Glynn LE, Nind AP: Production
of a chronic arthritis with ovalbumin. Ann Rheum Dis
30:307-315, 1971
4. Jasin H, Ziff M: Immunoglobulin and specific antibody
synthesis of chronic inflammatory focus: antigen induced synovitis. J Immunol 102:355-369, 1969
5 . Deshmukh K, Hemrick S: Metabolic changes in rabbit
1257
articular cartilage due to inflammation. Arthritis Rheum
19: 199-208, 1976
6. Gillard GC, Lowther DA: Carageenin-induced arthritis.
11. Effect of intraarticular injection of carageenin on the
synthesis of proteoglycan in articular cartilage. Arthritis
Rheum 19:918-922, 1976
7. Trentham DE, Townes AS, Kang AH: Autoimmunity to
type I1 collagen: an experimental model of arthritis.
J Exp Med 146:857-868, 1977
8. Trentham DE, Townes AS, Kang AH, David JR: Humoral and cellular sensitivity to collagen type I1 collagen-induced arthritis in rats. J Clin Invest 61239-96, 1978
9. Trentham DE, Dynesius RA, David JR: Passive transfer
by cells of type I1 collagen-induced arthritis in rats. J
Clin Invest 62:359-366, 1978
10. Stuart JM, Cremer MA, Kang AH, Townes AS: Collagen-induced arthritis in rats: evaluation of early immunologic events. Arthritis Rheum 22: 1344-1351, 1979
11. Caulfield JP, Hein A, Dynesius-Trentham RA,
Trentham DE: Morphologic demonstration of two stages
in the development of type I1 collagen-induced arthritis.
Lab Invest 46:321-343, 1982
12. Miller EJ: Isolation and characterization of a collagen
from chick cartilage containing three identical a-chains.
Biochemistry 10: 1652-1658, 1971
13. Engfeldt B, Hjertquist SO: The effect of various fixatives on the preservation of acid glycosarninoglycans in
tissues. Acta Pathol Microbiol Scand 71:219-232, 1967
14. Rosenberg L: Chemical basis for the histologic use of
safranin-0 in the study of articular cartilage. J Bone
Joint Surg 53A:69-82, 1971
15. Hirsch JG, Fedorka ME: Ultrastructure of human leukocytes after simultaneous fixation with glutaraldehyde
and osmium tetroxide and post-fixation in uranyl acetate. J Cell Biol 38:615-627, 1968
16. Holtrop M: The ultrastructure of the epiphyseal plate. I.
the flattened chondrocyte. Calcif Tissue Res 9: 131-139,
1971
17. Reynolds ES: The use of lead citrate at high pH as an
electron opaque stain in electron microscopy. J Cell Biol
171208-2 13, 1963
18. Stegemann T: Mikrobestimmung von Hydroxyprolin mit
chloramin-T-Und-P-dimethylaminobezaldehyd.Z Physiol Chem 311:41, 1958
19. Eyre DR, Brickley-Parsons DM, Glimcher MJ: Predominance of type I collagen at the surface of avian articular
cartilage. FEBS Lett 85:259-263, 1978
20. Bornstein P, Peiz K: Collagen: structural studies based
on the cleavage of the methionyl bonds. Science 148:
1353-1355, 1965
21. Eyre DR, Muir H: Collagen polymorphism: two molecular species in pig intervertebral disc. FEBS Lett 42: 192196, 1974
22. Eyre DR, Muir H: The distribution of different molecular species of collagen in fibrous, elastic and hyaline
cartilages of the pig. Biochern J 151:595-602, 1975
DESIMONE ET AL
1258
23. Ford JD, Baker JR: Procedures for the automated
analysis of proteoglycans and glycosaminoglycans. Anal
Biochem 84539-550, 1978
24. Weiss C: Ultrastructural characteristics of osteoarthritis. Fed Proc 32:1459-1466, 1973
25. Miller EJ, Matukas VJ: Chick cartilage collagen: a new
type of a1 chain not present in bone or skin of the
species. Proc Natl Acad Sci USA 64: 1264-1268, 1969
26. Eyre DR, Muir H: Quantitative analysis of type I and I1
collagens in human intervertebral discs at various ages.
Biochim Biophys Acta 492:29-42, 1977
27. Mankin HJ, Thrasher AZ, Hall D: Biochemical and
metabolic characteristics of articular cartilage from osteoarthritic human femoral heads. J Bone Joint Surg
59A:724-728, 1977
28. Nimni ME, Deshmukh K: Differences in collagen metabolism between normal and osteoarthritic articular
cartilage. Science 181:751-752, 1973
29. Mohr W, Wild A, Wolf HP: Incorporation of 'H-proline
into hyaline articular cartilage. Virchows Arch [Cell
Path011 28:l-12, 1979
30. Carmichael DJ, Gillard GC, Lowther DA, Handley CJ,
Santer VB: Carageenin-induced arthritis. IV. Rate
changes in cartilage matrix proteoglycan synthesis. Arthritis Rheum 20:834-840, 1977
31. Mitchell N, Shepard N: The ultrastructure of articular
cartilage in rheumatoid arthritis. J Bone Joint Surg
52A:1405-1423, 1970
32. Muirden KD, Deutschmann P, Philips M: Articular
cartilage in rheumatoid arthritis: ultrastructure and
enzymology. J Rheumatol 1:24-33, 1974
33. Freeman MA, Kempson GE: Load carriage, Adult Articular Cartilage. Edited by MA Freeman. London,
Pitman Medical, 1973, pp 228-246
34. Maroudas A: Distribution and diffusion of solutes in
articular cartilage. Biophys J 10:365-379, 1970
35. Poole AR, Barratt ME, Fell HB: The role of soft
connective tissue in the breakdown of pig articular
cartilage cultivated in the presence of complementsufficient antiserum to pig erythrocytes. 11. Distribution
of immunoglobulin G (IgG). Int Arch Allergy Appl
Immunol 44:469-488, 1973
36. Milroy SJ, Poole AR: Pig articular cartilage in organ
culture: effect of enzymatic depletion of the matrix on
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
response of chondrocytes to complement sufficient antiserum against pig erythrocytes. Ann Rheum Dis 33500508, 1974
Herman JH, Musgrave DS, Dennis MV: Phytomytogeninduced, lymphokine-mediated cartilage proteoglycan
degradation. Arthritis Rheum 20:922-932, 1977
Taplits MS, Crissman JD, Herman JH: Histologic assessment of lymphokine-mediated suppression of chondrocyte glycosaminoglycan synthesis. Arthritis Rheum
22:66-70, 1979
Weissmann G: Lysosomes. New Engl J Med 273: 11431149, 1965
Oronsky A, Ignarro L, Perper R: Release of cartilage
mucopolysaccharide-degrading neutral protease from
human leukocytes. J Exp Med 138:461-472, 1973
Deshmukh K, Nimni ME: Effects of lysosomal enzymes
on the type of collagen synthesized by bovine articular
cartilage. Biochem Biophys Res Comm 53:424-431,
1973
Anderson JC, Labedz RI, Kewley MA: The effect of
bovine tendon glycoproteins on the formation of fibrils
from collagen solutions. Biochem J 167:345-351, 1977
Seegmiller R, Fraser FC, Sheldon H: A new chondrodystrophic mutation in mice: electron microscopy of
normal and abnormal chondrogenesis. J Cell Biol 48:
580-593, 1971
Stephens TB, Seegmiller RE: Normal production of
cartilage glycosaminoglycans in mice homozygous for
the chondrodystrophic gene. Teratology 13:317-326,
1976
Byers PH, Holbrook KA, Hall JG, Bornstein P, Chandler JW: A new variety of spondyloepiphyseal dysplasia
and abnormal dermal collagen fibrils. Hum Genet 40:
157-169, 1978
Vogel A, Holbrook KA, Steinmann B, Gitzelmann R,
Byers PH: Abnormal collagen fibril structure in the
gravis form of Ehlers-Danlos syndrome. Lab Invest
40:20 1-206, 1979
Chew FC, Low WD, Yau NK: Banded structures in the
intervertebral disks with paralytic scoliosis. Proc Electron Microscopy SOC Am 292, 1979
Trentham DE, Dynesius RA, Recklin RE, David JR:
Cellular sensitivity to collagen in rheumatoid arthritis. N
Engl J Med 299:327-332, 1978
Документ
Категория
Без категории
Просмотров
6
Размер файла
2 103 Кб
Теги
morphology, induced, stud, arthritis, typed, cartilage, biochemical, articular, collagen
1/--страниц
Пожаловаться на содержимое документа