close

Вход

Забыли?

вход по аккаунту

?

Phosphorylation of proteoglycans. Identification of phosphorylation sites in chondroitin sulfate-rich region of core protein

код для вставкиСкачать
804
PHOSPHORYLATION OF PROTEOGLYCANS
Identification of Phosphorylation Sites in Chondroitin Sulfate-Rich
Region of Core Protein
ROGER S. ANDERSON and EDITH R. SCHWARTZ
Serine residues, which are the sites of phosphorylation in proteoglycans, were demonstrated to be located on chondroitin sulfate-containing peptides. These
peptides appeared to be derived primarily from the
chondroitin sulfate-rich region of the proteoglycan core
protein. The localization of phosphate moieties in the
chondroitin sulfate-containing peptides was observed in
all experiments. The phosphate moieties were retained
on chondroitin sulfate-containing peptides after the
protein core was treated with either papain or trypsin.
Two phosphopeptide preparations, derived from chondroitin sulfate-containing peptides of proteoglycan subunits, were extensively purified. These 2 phosphopeptide preparations were shown by carbohydrate analysis
to be free of keratan sulfate-containing peptides or
peptides from the hyaluronic acid binding region of the
core protein. One of the phosphopeptide preparations
had a phosphate:serine molar ratio of 0.40. This indicated that nearly one-half of the serine residues were
phosphorylated rather than glycosylated. Peptides derived from the core protein that contained keratan
sulfate had no phosphate moieties.
Proteoglycans, a major constituent of articular
cartilage, contribute to the unique biomechanical
properties of the tissue. Proteoglycans are found in
From the Department of Orthopedic Surgery, Tufts University School of Medicine, Boston, Massachusetts.
Supported in part by NIH grants AM-22057 and AM-22149
(Dr. Schwartz) and RCDA AM-01029 (Dr. Anderson).
Roger S. Anderson, PhD: Research Assistant Professor;
Edith R. Schwartz, PhD: Professor.
Address reprint requests to Edith R. Schwartz, PhD, 136
Harrison Avenue, Boston, MA 021 11.
Submitted for publication August 21, 1984; accepted in
revised form January 14, 1985.
Arthritis and Rheumatism, Vol. 28, No. 7 (July 1985)
cartilage as aggregates formed by their noncovalent
interactions with hyaluronic acid and link proteins. In
addition, other matrix proteins have been found closely associated with the aggregates (see refs. I and 2 for
review).
Proteoglycan subunits consist of a central protein core of M , 200,000 to which carbohydrate moieties are covalently attached. For operational purposes, the core protein has been divided into 3 regions
referred to as the hyaluronic acid binding region, the
keratan sulfate-rich region, and the chondroitin sulfate-rich region. N-linked oligosaccharides are found
in the hyaluronic acid binding region. The keratan
sulfate-rich region contains small 0-linked oligosaccharides and predominantly keratan sulfate. The chondroitin sulfate-rich region contains mostly chondroitin
sulfate.
Phosphorylation of the core protein of human
proteoglycans was first described by us in 1977 (3). In
that and subsequent studies we demonstrated that all
tested samples of proteoglycan subunits from human
articular and epiphyseal cartilages could be phosphorylated and that serine residues appeared to be the sites
of phosphorylation (3-5). Subsequent studies examined this posttranslational event in greater detail (6,7).
Examination of proteoglycan subunits purified from a
variety of normal human articular cartilages showed
that phosphorylation occurred on serine residues of
the core protein. Furthermore, chemical quantification
of phosphate showed 4 phosphate moieties/200,000dalton protein core. No phosphothreonine or phosphotyrosine residues were found (7).
These results were obtained in large part by
conducting studies with the use of 32P-proteoglycan
subunits isolated and purified from human cartilage
PROTEOGLYCAN PHOSPHORYLATION
specimens previously incubated in vitro in the presence of 32P-orthophosphate. Two other approaches
were used to confirm these findings. In the first case,
proteoglycan subunits isolated and purified from articular cartilage were analyzed for phosphate content by
chemical means (7). In the second instance, proteoglycan subunits extracted and purified from human articular and bovine nasal and articular cartilages, as well
as chondrosarcoma, were phosphorylated with the use
of cyclic AMP-dependent protein kinase (8). In all
instances, serine residues on the core protein were the
sites d phosphorylation.
Subsequent to the publication of our initial
studies, Peters et a1 (9) described the phosphorylation
of proteoglycans in bovine nasal cartilage. Recently,
Oegeirna et a1 (10) reported on the phosphorylation of
proteloglycans in chondrosarcoma, describing the
presence of phosphoxylose residues in this tumor
tissue:.
The present report concerns the localization in
the core protein of the phosphorylated serine residues
and addresses the question of what features distinguish the phosphorylated serine residues from those
that are glycosylated. In the experiments described
here, phosphorylation sites were identified by subjecting purified proteoglycan subunits to enzymatic degradation and isolating and characterizing the resulting
peptides. It was found that the phosphoserine residues
were located exclusively on peptides containing chondroitin sulfate. It was also found that all tested samples
provided the same phosphoserine-containing peptides.
This !juggests a conservation of core protein structure.
MATERIALS AND METHODS
Materials. Guanidine hydrochloride (ultrapure grade)
was olbtained from Schwarz/Mann (Cambridge, MA); cesium
chloride (density gradient grade) was from Gallard-Schlesinger (Clarle Place, NY); protease inhibitors were from Sigma
(St. Louis, MO); chondroitinase ABC was from Miles (Elkhart, IN); Ham’s F-12 and Gey’s balanced salt solution were
from Gibco (Grand Island, NY); 3’P-orthophosphate and 3Hamino1 acid mixture were from New England Nuclear (Boston, MA); Sepharose C L d B and Bio-Gel P-4 were products
of Pharmacia (Piscataway, NJ) and Bio-Rad (Richmond,
CA), irespectively. All other chemicals were reagent grade.
Labeling, isolation, and purification of proteoglycan
subunits. Specimens of human articular cartilage were obtained under sterile conditions during necessary surgery or
at the time of autopsy. The samples were washed repeatedly
with Gey’s balanced salt solution, diced into l-2-mm3
pieces, and incubated for 24-48 hours in Ham’s F-12 (10 ml/
gm tissue) containing 15 mM HEPES (pH 7.6), 1 mM Lglutanline, 0.081 mM MgS04, 300 pCi/ml 32P-orthophos-
805
phate (32P),and 20 pCilml’H-amino acid mixture. The mean
donor age was 55 years, with a range of 43-79.
At the end of the incubation period, the tissue was
washed with 0. IM sodium phosphate (pH 7.0). Immediately
thereafter, proteoglycans were extracted from the labeled
tissue with 10 volumes of 4M guanidine hydrochloride in 50
mM sodium acetate (pH 5.8), containing 50 mM sodium
fluoride as a phosphatase inhibitor (11) and a mixture of
protease inhibitors including 100 mM 6-aminohexanoic acid,
10 mM disodium EDTA, 10 mM N-ethylmaleimide, 5 mM
benzamidine, and 1 mM phenylmethylsulfonyl fluoride (2).
Proteoglycan subunits were purified by subjecting
the extracts to 2 successive equilibrium density gradient
centrifugations in CsCl under dissociative conditions, followed by size-exclusion chromatography on a Sepharose
CL-6B column (2.5 x 83.0 cm) equilibrated in 500 mM
sodium acetate (pH 6.8). Fractions containing proteoglycan
subunits were pooled, dialyzed against distilled water, and
lyophilized. The subunit preparations were shown to be free
of other proteins and 32P-proteins. When subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, neither preparation exhibited proteins or 32P-proteinsin the gel,
as monitored by Coomassie blue and silver staining or
autoradiography, respectively.
Treatment of proteoglycan subunits with degradative
enzymes. Papain. Purified proteoglycan subunit preparations
were treated with papain at a concentration of 33 pg papain11
mg proteoglycan in 100 mM sodium acetate (pH 7.3) containing 5 mM disodium EDTA and 5 mM cysteine HCl. The
reaction mixtures were incubated at 65°C for 4 hours (12).
Trypsin. Proteoglycan subunits in 100 mM sodium
acetate, 100 mM Tris-HCI (pH 7.3) were treated with trypsin
at a concentration of 4 pg trypsidl mg proteoglycan for 8
hours at 37°C (13).
Chondroitinase ABC. Preparations of proteoglycan
subunits dissolved in 100 mM sodium acetate, 100 mM TrisHCl (pH 7.3) were reacted with chondroitinase ABC at a
concentration of 100 mU/400 pg proteoglycan hexuronic
acid for 2 hours at 37°C. A second aliquot of enzyme, equal
to the first, was then added to the mixture, and the reaction
was continued for an additional 6 hours (14).
Size-exclusion chromatography. Sepharose C L d B .
Proteoglycans treated with proteases alone or subsequent to
additional treatment with chondroitinase ABC were chromatographed on a 1.0 x 115.0-cm Sepharose CL-6B column
equilibrated in 500 mM sodium acetate (pH 6.8) at 1.66 ml/
hour. Fractions of 0.73 ml were collected and analyzed for
32P,3H, and hexuronic acid contents and for absorbance at
280 nm.
Bio-Gel P 4 . Peptides resulting from sequential treatment of proteoglycan subunits with papain and chondroitinase ABC were chromatographed on a 1.0 x 115.0-cm column
of Bio-Gel P-4 equilibrated in 500 mM pyridine acetate ( H
6.2). Fractions of 1.O ml were collected and analyzed for P,
3H, hexuronic acid, sialic acid, and P content.
Agaros+acrylamide gel electrophoresis. Proteoglycan
subunits and peptides derived from protease-treated subunits were subjected to electrophoresis on gels containing
0.6% agarose and 1.2% acrylamide (15). Glycosaminoglycan-containing materials were detected in the gels after
electrophoresis was completed by reacting the gels with
P
806
ANDERSON AND SCHWARTZ
toluidine blue (0.2% in 0.1M acetic acid). The mobility of
proteoglycan material was calculated relative to that of
chondroitin sulfate (Rcs).
Phosphoamino acid analysis. Two-dimensional electrophoresis-chromatography was used to identify the phosphorylated amino acids released by limited acid hydrolysis
of purified proteoglycan subunits or of peptides derived from
subunits. Preparations of 32P-labeledsubunits or peptides
were hydrolyzed under a nitrogen atmosphere with 6N HCI
at 105°C for 2 hours. After the addition of standard phosphoamino acids, the mixtures were subjected to thin-layer
analysis as described previously (7,16). 32P-phosphoamino
acids were detected by autoradiography and identified by
reaction with ninhydrin.
Phosphopeptide mapping. Two-dimensional electrophoresis-chromatography on 20 x 20-cm cellulose thin-layer
plates was used to separate the 32P-peptidesproduced by
protease treatment of purified proteoglycan subunits. The
solvent systems of Elder et al(l7) were used. Electrophoresis in the first dimension was for 90 minutes at 500V before the
performance of chromatography in the second dimension.
Peptides and phosphopeptides were detected by reaction
with ninhydrin. 32P-peptideswere detected by autoradiograPhY.
Phosphate analysis. Covalently-bound phosphate was
determined by a modification (7) of the reaction method that
uses malachite green (18). Samples for analysis were prepared by heating as much as 50 a of subunit core protein or
peptide in 100 pl of 70% perchloric acid at 150°C (19).
Noncovalently-bound phosphate was determined in duplicate samples that were not subjected to the hydrolysis step.
Analytic methods. Protein was determined by using
the Lowry method, with bovine serum albumin as a standard
(20). Hexuronic acid was measured by an automated method
of the carbazole reaction (21). Total sialic acid content of
subunit and peptide material was measured according to the
method of Jourdian et al (22). Carbohydrate composition
was analyzed by the gas-liquid chromatography method of
Reinhold (23) after hydrolysis of the samples in 1N HzS04
for 4 hours at 100°C. Amino acid analyses were performed
by using a Beckman analyzer on samples that were hydrolyzed under a nitrogen atmosphere or in vacuo in 6N HC1 at
105°C for 24 hours.
the gradient. The ratio of 32P:hexuronic acid was
constant throughout the gradient. These data indicated
that 32Pwas associated with peptides from the portion
of the proteoglycan core protein containing hexuronic
acid.
The cleavage of the proteoglycan core protein
by trypsin was also monitored by subjecting the trypsin-treated proteoglycan subunits to agarose-acrylamide gel electrophoresis. All of the toluidine
blue-staining material migrated with an R,, of 0.89
compared with chondroitin sulfate. In contrast, the
intact proteoglycan subunit had R,, values of 0.64 and
0.70. Trypsin treatment had produced fragments substantially smaller than the original intact proteoglycan
subunits.
When subjected to size-exclusion chromatography on Sepharose C L d B , this same preparation eluted
in 2 fractions with K,, values of 0.03 and 0.23 (Figure
2). Essentially all of the materials containing 32P and
hexuronic acid were eluted in these 2 fractions, which
represented 25% and 70% of the total 32P,respectively.
Characteristics of peptides from papain treatment. Proteoglycan subunits were isolated and puri-
-
-
1 4
c
.-0
c
0
c
L
0
1
-
0
e
3
. 6
.c
N
2
0
0
*
f
X
2
- 4
8
U
0
Y
RESULTS
n
N
n
U
Y
- * za
I
Characteristics of peptides from trypsin treatment. A preparation of 32P-proteoglycan subunits was
purified from normal human articular cartilage. It was
treated with trypsin, and the resulting reaction mixture
was subjected to equilibrium density gradient centrifugation in CsCl under associative conditions. At equilibrium, the gradient was fractionated into 5 1-ml
aliquots, each of which was analyzed for radioactive
and chemical contents. As shown in Figure 1, the
distributions of 32Pand hexuronic acid paralleled each
other throughout the gradient, and 85% of the 32Pand
hexuronic acid was located in the bottom two-fifths of
E
3
0
-0
AS
A4
A3
A2
TOP
Al
BOTTOM
GRADIENT FRACTION
Figure 1. Distribution of tryptic products of 32P-proteoglycans after
centrifugation in a CsCl density gradient under associative conditions. 32P-proteoglycan subunits, treated with trypsin as described
in Materials and Methods, were centrifuged for 48 hours at 35,000
rpm. At equilibrium, the gradients were divided into 5 1-ml fractions, A1-A5 (bottom to top), and each fraction was analyzed for
hexuronic acid (0)
and 32P(W) content.
807
PROTEOGLYCAN PHOSPHORYLATION
fied from normal human articular cartilage that had
been incubated in vitro in a medium containing a
3H-aimino acid mixture and 32P. This preparation of
32P-, 3H-proteoglycan subunits was treated with papain, antd the resulting reaction mixture was subjected to
Chromatography on a Sepharose CL-6B column (Figure 3). The material was eluted in 2 fractions, designated 1 ,and 2. Fraction 1 contained 83% of the 32P-and
hexuronic acid-containing peptides. In contrast, the
protein content of fraction 1, as represented by 3Hradioactivity, was small. Over 75% of the proteinconta.ining material eluted in fraction 2.
The material contained in fraction 1, as well as
the original 32P-,3H-proteoglycan subunit preparation,
was characterized by electrophoresis on agaroseacryliamide gels. Whereas the original, intact proteoglycan subunit preparation migrated as 2 bands with
R,, v,aluesof 0.64 and 0.70, the toluidine blue-staining
material in fraction 1 migrated with an R,, of 0.99,
compared with chondroitin sulfate. This suggested
that papain treatment of proteoglycan subunits yielded
smaller peptides than trypsin treatment.
Phosphoamino acid analyses by 2-dimensional
TUBE NUMBER
Figure 2. Size-exclusion chromatography on Sepharose CL-6B, of
peptides from trypsin treatment of 32P-proteoglycan subunits (see
Materials and Methods for details). Fractions of 0.86 ml were
collected. Aliquots of each fraction were analyzed for 32P(0-0)
and hexuronic acid (0- -0)
content.
6
5
c
0
5
4
e
,
o!
0
3
%
-s
2
;
u
g
1
40
50
60
70
80
90
I00 110
I20
3
0
130 I40
TUBE NUMBER
Figure 3. Size-exclusion chromatography on Sepharose CL-6B,
of peptides derived from papain treatment of "P-, 'H-proteoglycan subunits (see Materials and Methods for details). Fractions of
0.73 ml were collected and aliquots of each were analyzed for "P
( L O ) , 'H (0-0), and hexuronic acid (UA)(O- - -0)
content. Contents of tubes numbered 68-100 were combined to give
fraction 1 and contents of tubes 101-130 to give fraction 2.
electrophoresis-chromatography showed the presence
of only 32P-phosphoserine residues in the peptides
contained in fraction 1 (Figure 4).
The peptides contained in fraction 1 were analyzed for chemical content. This was compared with
the chemical composition of the proteoglycan subunit
preparation, which served as starting material. As
shown in Table 1, the peptides in fraction I contained
essentially all of the hexuronic acid and phosphate
content present in the original proteoglycan subunit
preparation. The ratio of phosphate to hexuronic acid
remained unchanged for the isolated fraction 1 peptides compared with the original subunit preparation
(Table 1). This indicated that there was copurification
of phosphate and hexuronic acid in peptide fraction 1.
In contrast, the phosphate :protein ratio increased
fourfold in fraction 1 compared with that in the original
proteoglycan subunit preparation (Table 1). These
data show a significant purification of phosphorylated
peptides in fraction 1. The conservation of hexuronic
acid in isolated peptide fraction 1 indicated that essentially all of the peptides from the chondroitin sulfaterich region of the proteoglycan core protein were
contained therein.
Additional chemical data are shown in Table 2.
Peptide fraction 1 contained sugars found in both
808
ANDERSON AND SCHWARTZ
Table 2. Chemical composition of fraction 1, obtained by
Sepharose CL-6B chromatography of papain-treated proteoglycan
subunits*
Constituent
nmolesigm cartilage
Hexuronic acid
Xylose
N-acetylgalactosamine
Galactose
N-acetylglucosamine
Sialic acid
Mannose
Fucose
Glucose
Phosphate
Serine
2,430
129
979t
1,052
443t
~_____
100
18
55
80
13
187
* Material in fraction 1 was obtained as described in Figure 3.
t The conditions used to hydrolyze the samples before carbohydrate
analyses were chosen to give quantitative recovery of neutral
sugars. For this reason, values for hexosamines are lower than
expected.
Figure 4. Characterization of phosphoamino acids in fraction 1
peptides isolated from proteoglycan subunits. A sample containing
fraction 1 peptides (Figure 3) was hydrolyzed for 2 hours in 6N HC1
at 105°C and then subjected to 2-dimensional thin-layer electrophoresis-chromatography. 32P-labeled amino acids were detected
by autoradiography and identified by comparing their migration
with standards stained with ninhydrin. 32P-phosphoserine (P-SER)
was the only phosphoamino acid detected in fraction 1 peptides.
P-THR = 32P-threonine; P-TYR = 32P-tyrosine.
chondroitin sulfate and in keratan sulfate. In addition
to hexuronic acid, xylose, N-acetylgalactosamine, and
galactose from chondroitin sulfate, analyses showed
the presence of N-acetylglucosamine, galactose, and
sialic acid from keratan sulfate. Trace amounts of
mannose, glucose, and fucose were also detected in
peptide fraction 1 . These data indicate that the peptides in fraction 1 were a mixture originating from
regions of the proteoglycan core protein rich in chondroitin sulfate and keratan sulfate.
Amino acid analysis of the peptides in fraction 1
showed the predominant amino acids (making up 55%
of the total) to be serine, glutamic acid, glycine, and
alanine. Two residues, aspartic acid and threonine,
contributed an additional 13%. The remaining 32%
were evenly distributed among other amino acid residues. Analyses of these data show that the molar ratio
of phosphate :serine :xylose was 1 : 14 : 10.
Characteristics of peptides from papain and
chondroitinase treatments. To delineate more precisely
the sites of phosphorylation, the phosphopeptides in
fraction 1 were further purified. Fraction 1 was treated
with chondroitinase ABC, and the resulting reaction
mixture was chromatographed on a column of Bio-Gel
P-4. The elution profiles for 3H-, 32P-, and hexuronic
acid-containing materials are shown in Figure 5. Materials recovered from the column were divided into 4
pools, as indicated in the figure. Pools 2 and 3 contained essentially all of the phosphorylated peptides
(Figure 5). No 32Pwas detected in pool 4 and only a
trace amount was in pool 1 .
Chemical analyses of materials contained in
Table 1. Chemical composition of proteoglycan subunit preparation and fraction 1
Analysis
Sample
Proteogl ycan
subunits
Fraction 1
*
P
(nmoles)
Hexuronic acid
(nmoles)
Protein
(kd
P:hexuronic acid
(nmoles/nmole)
P: protein
(nmoles/kg)
15.0
2,687
324
5.6 x 10-3
46.3 x 10-3
13.2
2,430
71
5.5 x 10-3
185.9
X
Fraction 1 represents pooled samples from chromatography on Sepharose CL-6B (Figure 3) of
papain-treated proteoglycan subunits.
PROTEOGLYCAN PHOSPHORYLATION
poolr; 1 through 4 (Figure 5) are summarized in Table
3 . Pool 4, representing small molecules eluting near
Vt, contained 90% of the hexuronic acid. This material
represents disaccharides formed by the action of chondroitinase ABC on chondroitin sulfate. In contrast,
pool 1 contained no hexuronic acid but did contain
94% of the sialic acid. This indicated that the peptides
in pool 1 were keratan sulfate-containing peptides of
the core protein. On the other hand, pools 2 and 3 were
essentially free of sialic acid but contained residual
amounts of hexuronic acid (Table 3 and Figure 5). This
indicated that chondroitin sulfate-containing peptides
were in pools 2 and 3. Chemical analysis showed that
56% and 41% of the total phosphate were present in
poolrj 2 and 3, respectively (Table 3). These data,
coupled with the finding that essentially all 32P-labeled
molecules were in these pools, indicated that phosphate moieties were located almost exclusively in
peptides containing chondroitin sulfate.
Amino acid analyses of the peptides contained
in pools 2 and 3 are presented in Table 4. Serine was
the predominant amino acid residue in both pools 2
and 3 . The other major amino acids included aspartic
acid, glutamic acid, and glycine. The molar ratios of
serine to all of the other amino acids were 1 :3.17 for
809
pool 2 peptides and 1 : 1.84 for pool 3 peptides. These
data indicate that if, on the average, each peptide
contained one serine residue, then pools 2 and 3
contained tetrapeptides and tripeptides, respectively.
The molar ratio of phosphate :serine residues was 0.40
for the material in pool 2 and 0.09 for that in pool 3.
Peptides found in pools 2 and 3 were analyzed
by 2-dimensional thin-layer electrophoresis-chromatography . Autoradiography of the resultant thin-layer
peptide maps revealed the presence of several 32Pphosphopeptides in both pools (Figure 6). Although
the 32P-phosphopeptides in pool 2 were heterogeneous, the predominant species could be separated
into 2 major groups of peptides based on a similar
charge:mass ratio. The location of these 2 groups is
indicated in Figure 6A. Two predominant groups of
peptides were also found in pool 3 (Figure 6B). The
charge: mass ratios were similar to those for the predominant peptides in pool 2. When compared, the 2
predominant groups of peptides detected by autoradiography in pool 3 corresponded in mobility with the 2
predominant groups of peptides that were located by
reaction with ninhydrin. Coversely , no ninhydrinreactive peptides could be detected on peptide maps of
material in pool 2.
400
POOL
X
30
35
40
45
50
55
60
65
70
75
80
85
90
FRACTION NO.
Figure 5. Elution profile of peptides and phosphopeptides from Bio-Gel P-4. The applied sample was prepared by treating fraction I (Figure 3)
with chondroitinase ABC. The column was eluted with O.5M pyridine acetate (pH 6.2), and aliquots of 1 ml were collected. Each fraction was
analyzed for 32P(0-O),
’H (0-0), and hexuronic acid (UA) (0- - -0) content as described in Materials and Methods. The contents of
fractions 30-34 were combined to give pool 1, those of fractions 35-47 to give pool 2, those of fractions 48-60 to give pool 3, and those of fractions 70-85 to give pool 4.
810
ANDERSON AND SCHWARTZ
Table 3. Composition of pools 1-4 prepared by Bio-Gel P-4
chromatography of chondroitinase-treated fraction 1 peptides*
Chemical analvsis
Bio-Gel
fraction
Pool I
Pool 2
Pool 3
Pool 4
P
Sialic acid
0.26
4.70
3.55
0
55.0
Hexuronic
acid
0
25.6
215.2
2,079.4
2.8
0.9
0
* Fraction 1 (Figure 3) was treated with chondroitinase and chromatographed on Bio-Gel P-4. Materials eluted from the column were
divided into pools 1-4 (Figure 4) and aliquots of each were analyzed
for chemical content as described in Materials and Methods. Values
shown are nmoles.
Table 4. Amino
preparations
acid
analysis
of
phosphopeptide-enriched
Amino acid
Pool 2*
Pool 3*
Pool 2t
Pool 3 t
Aspartic acid
Threonine
Serine
Glutamic acid
ProIin e
Glycine
Alanine
All others
5.6
3.1
12.0
7.3
3.1
9.9
5.3
4.6
15.7
3.7
40.8
11.6
9.7
23.8
3.6
5.7
0.47
0.26
0.38
0.09
0.61
0.26
0.83
0.44
0.38
0.28
0.24
0.58
0.09
0.14
4.7
3.6
0.40
0.09
Phosphate
1 .oo
1 .oo
* Pools
2 and 3 were recovered from a column of Bio-Gel P-4
(Figure 4). Values shown are nmoles/gm cartilage.
't Values shown are moles. Molar concentrations were normalized
to a value of 1.OO for serine.
DISCUSSION
The data presented demonstrate that the phosphate moieties in proteoglycan subunits are located
primarily in the chondroitin sulfate-rich region of the
core protein. Chemical analyses of phosphorylated
peptides showed that these peptides have high serine
contents. These results were confirmed after peptides
containing chondroitin sulfate and keratan sulfate
were separated from each other. The separation of
these peptides by size-exclusion chromatography was
possible because the peptides containing keratan sulfate were so much larger than those from the chron-
droitin sulfate linkage region after chondroitinase ABC
treatment.
Limited information is available regarding amino acid sequences about serine residues that are
glycosylated or phosphorylated. A common feature
that has been described for glycosylated serine residues is the presence of an adjacent, C-terminal glycine
residue (24,25). Amino acid sequences adjacent to
serine residues that are phosphorylated by cyclic
AMP-dependent protein kinases have been elucidated. These phosphorylation sites have the general ami-
A
2
(-1
(+)
-0-
I
(-1
-0-
I
(3-1
Figure 6. Autoradiograms of 2-dimensional maps of 32P-peptides from the chondroitin sulfate-rich region of proteoglycan core protein.
Materials in pools 2 (A) and 3 (B), which had been eluted from a Bio-Gel P-4 column (Figure 5 ) , were subjected to 2-dimensional
electrophoresis-chromatography on cellulose thin-layer plates. 32P-peptides were detected by autoradiography . Vertical arrows identify the 2
predominant groups of phosphopeptides in each pool. Position of origin is indicated by 0 on figures.
PROTEOGLYCAN PHOSPHORYLATION
no acid sequence H-Lys-Arg-X-X-Ser-OH or H-ArgArg-X-Ser-OH (26). The common feature is the
presence of at least 1 arginine residue within 2-5
residues on the N-terminal side of phosphorylated
serine residue (27). Sequence studies of phosphopeptides purified from pools 2 and 3 are planned so that
the sequences of proteoglycan-derived phosphopeptides may be compared with the above general sequences.
Treatment of intact proteoglycans with phosphatases did not result in the release of all the covalently bound phosphate moieties (7). The localization
of the phosphoserine residues to the chondroitin sulfate-rich region provides an explanation for this observation. The presence of the highly charged chondroitin
sulfate chains probably sterically hindered the phosphatase from reacting with some of the serine-phosphate bonds.
Phosphorylation-dephosphorylation of proteins
has been shown to be an important mechanism for
regulating the activity of enzymes (26), as well as the
functioning of membrane receptors (28). The ability of
muscle (29) and cytoskeletal proteins (30) to interact
with leach other is also modulated by phosphorylatondephosphorylation. Although the role of phosphate
moieties on the core protein of proteoglycan subunits
has not yet been defined, limitation of the phosphate
moieties to the chondroitin sulfate-rich region suggests that their function is connected with this part of
the molecule. Since the high charge:density ratio of a
complete complement of chondroitin sulfate chains
may inegate a charge difference arising from phosphorylation-dephosphorylation, the physiologic role of the
latter may be expressed before glycosylation. Through
intracellular action, phosphorylation might direct the
addition of chondroitin sulfate and keratan sulfate to
the newly synthesized proteoglycan core protein.
On the other hand, the physiologic function for
serinle-linked phosphate groups may be expressed extracellularly. Possible extracellular roles include: 1)
modulation of proteoglycan breakdown in the extracellullar matrix, 2) regulation of the endocytotic uptake
of partially degraded proteoglycans, or 3) participation
in the interaction of proteoglycans with other components,, such as collagen or noncollagen proteins, in the
extracellular matrix.
The techniques developed to isolate phosphopeptides from proteoglycan subunits and the data
presented here are the basis for further studies to
determine the amino acid sequence about phosphorylated serine residues. The findings from these project-
81 1
ed studies should aid in elucidating the physiologic
role of phosphate moieties in proteoglycans.
ACKNOWLEDGMENTS
We thank Keyes Linsley and Dr. Roger Jeanloz of
Harvard Medical School for the gas-liquid chromatography
analyses for sugars and Dr. RennC Lu of the Boston Biomedical Research Institute for the amino acid analyses.
REFERENCES
1. Rosenberg L: Structure of cartilage proteoglycans,
Dynamics of Connective Tissue Macromolecules. Edited by PMC Burleigh, AR Poole. New York, Elsevier,
1975, pp 105-128
2. Hascall VC, Kimura JH: Proteoglycans: isolation and
characterization, Methods in Enzymology. Vol. 82. Edited by LW Cunningham, DW Frederickson. New York,
Academic Press, 1982, pp 769-800
3. Schwartz ER, Miller KS: Phosphorylation of sulfated
proteoglycans, Fourth International Congress of Glycoconjugates. Woods Hole, MA, 1977, p 26
4. Schwartz ER, Miller KS: Phosphorylation of the protein
core of proteoglycans from human cartilage. Orthop
Trans 2:118-119, 1978
5. Schwartz ER, Miller KS: Phosphorylation of proteoglycans in human articular cartilage. Connect Tissue Res
5:225-235, 1978
6. Anderson RS, Schwartz ER: Properties of phosphorylated proteoglycans from human articular cartilage.
Orthop Trans 6:243-244, 1982
7. Anderson RS, Schwartz ER: Phosphorylation of proteoglycans from human articular cartilage. Arthritis Rheum
27:58-71, 1984
8. Anderson RS, Schwartz ER: Phosphorylation of proteoglycans from human articular cartilage by a CAMPdependent protein kinase. Arthritis Rheum 27: 10231027, 1984
9. Peters C, Schwermann J, Schmidt A, Buddecke E:
Phosphate ester groups in proteoglycans from bovine
articular cartilage. Biochim Biophys Acta 673:270-278,
1981
10. Oegema TR Jr, Kraft EL, Jourdian GW, VanValen TR:
Phosphorylation of chondroitin sulfate in proteoglycans
from the Swarm rat chondrosarcoma. J Biol Chem
259: 1720-1726, 1984
11. Hollander VP: Acid phosphatases, The Enzymes. Vol.
IV. Edited by PD Boyer. New York, Academic Press,
1971, pp 449-498
12. De Luca S, Heinegard D, Hascall VC, Kimura JH,
Caplan AI: Chemical and physical changes in proteoglycans during development of chick limb bud chondrocytes grown in vitro. J Biol Chem 252:6600-6608, 1977
13. Heinegard D, Axelsson I: Distribution of keratan sulfate
812
in cartilage proteoglycans. J Biol Chem 252:1971-1979,
1977
14. Roughley PJ, White RJ: Age-related changes in the
structure of proteoglycan subunits from human articular
cartilage. J Biol Chem 255:217-224, 1980
15. McDevitt CA, Muir H: Gel electrophoresis of proteoglycans and glycosaminoglycans on large-pore composite
polyacrylamide-agarose gels. Anal Biochem 44:612-622,
1971
16. Reynolds FH Jr, Todaro GJ, Fryling C, Stephenson JR:
Human transforming growth factors induce tyrosine
phosphorylation of EGF receptors. Nature 292:259-262,
1981
17. Elder JH, Pickett RA 11, Hampton J, Lerner RA:
Radioiodination of proteins in single polyacrylamide gel
slices: tryptic peptide analysis of all the major members
of complex multicomponent systems using microgram
quantities of total protein. J Biol Chem 252:6510-6515,
1977
18. Itaya K, Ui M: A new micromethod for the colorimetric
determination of inorganic phosphate. Clin Chim Acta
14:361-366, 1966
19. Hams WH Jr, Lux SE: Structural characterization of
the phosphorylation sites of human erythrocyte spectrin. J Biol Chem 255:11512-1 1520, 1980
20. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ:
Protein measurement with the folin phenol reagent. J
Biol Chem 193:262-275, 1951
21. Bitter T, Muir H: A modified uronic acid carbazole
reaction. Anal Biochem 4:330-334, 1962
22. Jourdian GW, Dean L , Roseman S: The sialic acids. XI.
A periodate-resorcinol method for the quantitative esti-
ANDERSON AND SCHWARTZ
mation of free sialic acids and their glycosides. J Biol
Chem 246:430-435, 1971
23. Reinhold VM: Gas-liquid chromatographic analysis of
constituent carbohydrates in glycoproteins, Methods in
Enzymology. Vol. 25. Edited by CHW Hirs, SN Timasheff. New York, Academic Press, 1972, pp 244-249
24. Johnson AH, Baker JR: Amino acid sequences of peptides from the linkage region of chondroitin sulfate.
Biochem SOCTrans 1:277-279, 1973
25. Isemura M, Hanyu T , Kosaka H , Ono T, Ikenaka T:
Comparative study of carbohydrate-protein complexes.
111. Peptide structures of the linkage region in proteoglycans of human, porcine and shark cartilages. J Biochem
(Tokyo) 89: 11 13-1 119, 1981
26. Krebs EG, Beavo JA: Phosphorylation-dephosphorylation of enzymes. Annu Rev Biochem 48:923-959, 1979
27. Nimmo HG, Cohen P: Hormonal control of protein
phosphorylation, Advances in Cyclic Nucleotide Research. Vol. 8. Edited by P Greengard, GA Robison.
New York, Raven Press, 1977, pp 145-266
28. Rosen OM, Herreta R, Olowe Y, Petruzzelli LM, Cobb
MH: Phosphorylation activates the insulin receptor tyrosine protein kinase. Proc Natl Acad Sci USA 80:32373240, 1983
29. Stull JT: Phosphorylation of contractile proteins in relation to muscle function, Advances in Cyclic Nucleotide
Research. Vol. 13. Edited by P Greengard, GA Robison.
New York, Raven Press, 1980, pp 39-83
30. Jameson L, Caplow M: Modification of microtubule
steady-state dynamics by phosphorylation of microtubule-associated proteins. Proc Natl Acad Sci USA
78:3413-3417, 1981
Документ
Категория
Без категории
Просмотров
0
Размер файла
843 Кб
Теги
corel, site, proteoglycans, phosphorylation, sulfate, chondroitin, protein, identification, regions, rich
1/--страниц
Пожаловаться на содержимое документа