|
|
|
|||||||
|
|||||||||
(Received for publication, August 17,
1994; and in revised form, November 14, 1994) From the
Yeast The
particulate yeast
Figure 1:
Schematic representation of a binding
site-specific antibody for
Figure 2:
Immunospecificity of mAb ACIA12 for the
160-kDa
The surface-specific method described for
lactoperoxidase (Sigma) and H
For nonreducing/reducing
two-dimensional SDS-PAGE, nonreduced samples of affinity-purified
proteins were resolved by electrophoresis and gel strips, 11
For carbohydrate analysis by
periodic acid-Schiff (PAS) reagent, reduced samples of purified
receptor protein were separated by SDS-PAGE, electroblotted onto
nitrocellulose, treated for 15 min with 1% periodic acid in 3% acetic
acid, and washed with several changes of distilled water. The oxidized
samples were stained for 15 min in the dark with fresh Schiff's
reagent, treated for 5 min with 0.5% sodium m-bisulfite,
washed, and stored in the dark. By this procedure, the carbohydrate in
ovalbumin (4%) and rabbit IgG (3%) was detectable in blots of 0.1
µg of electrophoresed protein. To assess for transfer efficiency,
concurrent samples of electrophoresed receptor protein were
electroblotted onto nitrocellulose and stained for protein with 0.1%
Ponceau S in 0.1% acetic acid.
Figure 3:
Subunit constituents of the 180- and
160-kDa
Figure 4:
Isoelectric points of the
Charge heterogeneity of otherwise identical proteins is frequently
attributable to the content of sialic acid residues in oligosaccharide
side chains, many of which are covalently linked to N-glycosylation sites. The contribution of glycosylation as a
source of charge heterogeneity of the 20-kDa polypeptide was approached
in two ways: enzymatic removal of N-linked glycosylated units
and chemical oxidation of monosaccharides followed by staining with
Schiff's reagent. By the first method, 12 µg of purified
Figure 5:
N-Glycosylation of the
Figure 6:
Chemical analysis of the oligosaccharide
content of the
Figure 7:
Constitutive tyrosine phosphorylation of
Figure 8:
Localization of the two species of 20-kDa
polypeptides within the 160-kDa
The present studies demonstrate the structural composition of
two The
composite biochemical properties of the 20-kDa polypeptide subunit
showed it to be an unusual polypeptide and a unique constituent of
Studies on phosphorylation were initiated to
determine whether phosphorylation was a factor contributing to the
charge heterogeneity of the 20-kDa polypeptide. Even when consideration
was given to the fact that unopsonized zymosan particles induce protein
tyrosine phosphorylation in human monocytes(16) , the finding
that The preparation of monoclonal antibodies to purified 180-
and 160-kDa
Volume 270,
Number 5,
Issue of February 3, 1995 pp. 2145-2151
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
-Glucan
Receptors on Human Mononuclear Phagocytes (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-Glucan receptors are present on mammalian leukocytes and
initiate phagocytosis of particulate yeast -glucans, such as
zymosan particles. Human monocytes and U937 cells express two membrane
proteins of 180 and 160 kDa, each of which binds particulate yeast
glucan through a 20-kDa polypeptide constituent. In this report, the
structural composition of the two -glucan receptors and the
biochemical properties of their polypeptide constituents were examined.
The 180-kDa receptor was composed of three disulfide-linked
polypeptides of 95, 60, and 20 kDa, whereas the 160-kDa receptor was a
multimer of two polypeptides of 27 and 20 kDa. Unlike other receptor
constituents, the 20-kDa polypeptide was nonglycosylated and focused at
two distinct isoelectric points. Immunoblots of the focused
polypeptides showed the two 20-kDa variants and the 95-kDa subunit to
be constitutively tyrosine-phosphorylated, a feature not previously
reported for receptors on human mononuclear phagocytes.
Dephosphorylation of the receptor proteins resulted in the loss of
antigenic phosphotyrosine without affecting the antigenicity of either
20-kDa variant for the anti-idiotypic antibody to -glucan
receptors. Separate analysis of the 160-kDa receptor showed it
contained both variants of the 20-kDa polypeptide. Thus, the 20-kDa
subunit constituent of the two -glucan receptors is a functionally
and chemically unique polypeptide with apparent microheterogeneity in
its primary structure.
-glucan is a carbohydrate polymer derived from the
cell wall of Saccharomyces cerevisiae and composed of
-D-glucopyranosyl residues with consecutive 1,3 and
branched 1,6 linkages(1) . -Glucan is a major constituent
of yeasts and fungi (2) and is the active pharmacological agent
of zymosan particles(3, 4) . The administration of
yeast -glucans to laboratory animals rapidly gives rise to a state
of increased resistance to a variety of biologic insults through
mechanisms initiated by macrophages (5, 6, 7) . Recent studies with the soluble
yeast -glucan, Betafectin, have shown it to protect against lethal
peritonitis in the rat model for intra-abdominal sepsis (8) and
have demonstrated its efficacy as an anti-infective agent against
postoperative infections in high risk surgical patients(9) .
Unlike nearly all ligands thus far reported for mammalian cells, the
yeast -glucans are molecules which are totally foreign to animals
and, yet, are nonimmunogenic(10, 11) .-glucans have been extensively used in isolated
cell systems because of their potent and pleiotropic immunomodulatory
activating capacities. Glucan and zymosan particles stimulate rapid
phagocytic responses in human monocytes, and these are inhibited
selectively by soluble yeast -glucans(12, 13) .
Ligation of monocyte -glucan receptors by particulate yeast
glucans induces production of a broad spectrum of cellular metabolites (14) and phosphorylation of a substantial number of cellular
proteins(15, 16) , some of which appear to be linked
to early secretory and synthetic responses(15) . During serum
opsonization, the dominant complement fragment deposited onto zymosan
particles is iC3b(17) , which binds to the I domain of the
-chain subunit of complement receptor type 3 (CR3;
CD11b)(
)(18) . The deposition of C3 fragments on
particles already displaying an independent phagocytic signal augments
phagocytosis by improving contact between particle and
cell(19, 20, 21, 22, 23) .
A role for a glucan-binding site on CR3 has been
suggested(24) . However, ligand binding studies with purified
CR3 show no binding by this complement receptor to unopsonized zymosan
particles or to soluble -glucan(25) , and functional
studies with monocytes from patients genetically deficient in CR3 show
no impairment in the zymosan-induced production of platelet-activating
factor by a mechanism inhibitable by soluble
-glucans(26) .-Glucan receptors have been best
described for human monocytes in terms of ligand specificity. Monocyte
-glucan receptors do not recognize homopolysaccharides of mannose
and galactose, or homoglucosyl polysaccharides with -anomeric or
-1,4 linkages(12) . The smallest functional ligand has
been isolated and shown by mass spectrometry to be a yeast
heptaglucoside(27) . Structural conformations inherent to the
yeast glucan polymers are required for monocyte recognition, a feature
not exhibited by mammalian receptors for other carbohydrate molecules,
such as mannose (28) and galactose (29, 30) .
Steps to prepare an antibody to monocyte -glucan receptors have
focused on the ligand specificity of the receptors so as to obtain an
antibody to the ligand-binding domain(31) . As schematically
shown in Fig. 1, an anti-idiotypic antibody was prepared to the
antigen-combining site or idiotype (Id) of a monoclonal antibody (mAb
OEA10) that recognizes the biologically active epitopes on particulate
yeast glucans and the purified yeast heptaglucoside. The idiotypic
specificity of the anti-Id is apparent by its capacity to compete with
-glucan ligand for mAb OEA10 and by its selective capacity to bind
and block monocyte ingestion of particulate yeast glucans. Two
-glucan receptors of 180 and 160 kDa have been isolated from human
myelomonocytic U937 cells and normal monocytes by immunoaffinity
purification with the anti-Id(32) . On both types of cells, the
two receptor proteins contain a 20-kDa polypeptide subunit which
expresses the anti-Id epitope and glucan-binding domain, and the
160-kDa protein presents as the immunodominant -glucan receptor.
In this report, we present the structural composition of the 180- and
160-kDa -glucan receptor proteins on U937 cells and characterize
the polypeptide subunit chains of the two multimeric proteins. Each
polypeptide subunit exists as a distinct structural domain maintained
by disulfide bonds, and the 20-kDa glucan-binding domain presents as
two allelic variants differing in isoelectric points.
-glucan receptors. Soluble and
particulate yeast -glucans are recognized by plasma membrane
receptors (A) and by mouse mAb OEA10 (B). An
antibody, prepared to the antigen-combining region or Id of the mouse
monoclonal, binds to the ligand-binding region of -glucan
receptors (C).
Materials
Acrylamide, SDS, glycine, ampholytes,
nitrocellulose, and all other electrophoretic supplies were obtained
from Bio-Rad; cystatin from Aldrich; pepstatin, leupeptin,
phenylmethylsulfonyl fluoride, diisopropyl fluorophosphate, Nonidet
P-40, sodium thioglycollate, periodic acid solution, Schiff's
reagent, and general chemicals from Sigma; and the human myelomonocytic
U937 cell line was from the American Type Culture Collection
(Rockville, MD).Cell Culture
U937 cells were cultured in tissue
culture flasks (Costar Corp., Cambridge, MA) containing RPMI 1640
medium (Life Technologies, Inc.) and 10% heat-inactivated (56 °C
for 30 min) calf serum (Life Technologies, Inc.). The cell cultures
were incubated at 37 °C in a humidified atmosphere of 5% CO
and either harvested during logarithmic growth by centrifugation
or used as inoculum for scale-up production of cells in spinner flasks.
The cells were extensively washed in 0.01 M phosphate-buffered
saline (PBS), pH 7.4, counted on a Coulter counter (Coulter
Electronics, Hialeah, FL), and assessed for viability by trypan blue
exclusion, which was >95%.Antibodies
Nonimmune rabbit IgG (Miles, Inc., West
Haven, CT), rabbit IgG anti-phosphotyrosine (Upstate Biotechnology
Inc., Lake Placid, NY), and goat IgG anti-rabbit F(ab`) (Cappel Laboratories, Westchester, PA) were purchased; rabbit IgG
anti-Id (31) and mouse IgG1 mAb BD4 anti-fibronectin (33) were prepared and purified as described
elsewhere(31, 33) . The preparation of mouse IgG1 mAb
ACIA12 was carried out as previously described(31) . BALB/c
mice were immunized with 25 µg of purified U937 cell -glucan
receptor proteins, hybridomas were prepared by polyethylene glycol
fusions with single cell suspensions of spleen cells and plasmacytoma
Sp 2/0, and positive antibody-secreting hybridomas were identified with
purified receptor in a solid phase radioimmunoassay. The antibodies
were raised in spinner cultures and purified by affinity column
chromatography with rat mAb AHF5 anti-mouse light chain antibodies that
had been coupled to Sepharose 4B (Pharmacia Biotech Inc.) by cyanogen
bromide activation (34) . mAb ACIA12 immunoprecipitated the
160-kDa but not the 180-kDa -glucan receptor from detergent
lysates of surface-radioiodinated U937 cells (Fig. 2) and
recognized conformational determinants in the receptor that were lost
by reduction of disulfide bonds.
-glucan receptor. Detergent-soluble proteins from
surface-radioiodinated U937 cells were sequentially immunoprecipitated
for 1 h at 4 °C with mouse IgG1 (mAb BD4), mouse IgG1 mAb ACIA12,
and the rabbit IgG anti-Id (left panel), or were preadsorbed
with rabbit IgG and the anti-Id before immunoprecipitation with mAb
ACIA12 (right panel). The immunoadsorbed proteins were eluted,
resolved in nonreduced samples by SDS-PAGE in 5-15% acrylamide
gradient gels, dried, and detected by exposure to x-ray film. Variable
amounts of residual 180-kDa protein were present in cell samples
heavily surface-labeled and immunoprecipitated by mAb ACIA12,
suggesting a close association between the two -glucan receptor
proteins in the cell membrane. Mobility and size of prestained protein
standards are indicated.
Protein-coupled Sepharose Beads
Bovine serum
albumin (BSA), nonimmune rabbit IgG, anti-Id, mAb BD4, and mAb ACIA12
were each coupled in 0.1 M phosphate buffer, pH 7.0, at a
concentration of 4 mg of protein/g of activated CH-Sepharose beads
(Pharmacia) with coupling efficiencies of 80-90%.Radiolabeling of Proteins and Cells
Antibodies
were radiolabeled to specific activities of 1-2 
10
cpm/µg by the chloroglycoluril method (35) in tubes
coated with 50 µg of IODO-GEN (Pierce) and containing 100-200
µg of IgG protein and 1 mCi of carrier-free NaI
(NEN-DuPont). Iodinated antibody was separated from free iodide by
filtration through a 5-ml Excellulose desalting column (Pierce) in PBS
and 0.02% NaN
.O(36) was
used to radiolabel 2 10
U937 cells in 1 ml of PBS
with 1 mCi of NaI at 4 °C for 15 min. The labeled
cells were washed, lysed at 4 °C for 1 h in PBS containing 1%
Nonidet P-40, 5 mM diisopropyl fluorophosphate, 2 mM phenylmethylsulfonyl fluoride, 1 µM pepstatin, and 1
µM leupeptin (lysis buffer), and centrifuged at 12,000
g for 30 min at 4 °C. The detergent-soluble
fractions were precleared with BSA-Sepharose beads for 18 h at 4 °C
and subjected to sequential immunoprecipitation(32) .
Scale-up
purification of the 180- and 160-kDa -Glucan Receptor Purification-glucan receptors was
performed by immunoaffinity chromatographic procedures previously
published for smaller preparations(32) . U937 cells were raised
in spinner flasks, washed, lysed at a cell density of 5
10 cells/ml in lysis buffer, and centrifuged to remove
detergent-insoluble materials. The detergent-soluble fractions were
supplemented with 0.02% NaN before sequential passage
through columns of Sepharose 4B, nonimmune rabbit IgG-Sepharose, and
rabbit IgG anti-Id-Sepharose at a flow rate of 20 ml/h at 10 °C.
The columns were washed in PBS with azide until baseline levels were
established. The anti-Id column was then washed in azide-free PBS, and
the bound materials were eluted with 0.1 M glycine, pH 2.5.
The eluted proteins were concentrated under nitrogen by positive
pressure filtration on a PM-10 membrane (Amicon Corp., Danvers, MA),
dialyzed against 0.1 M NaCl in 1 mM phosphate buffer,
pH 7.0, and stored in aliquots at -70 °C. The average yield
of receptor protein from two large preparations was about 500
µg/10 lysed U937 cells as measured by the Bio-Rad
microassay for protein with BSA as the reference standard.
SDS-Polyacrylamide Gel Electrophoresis
(PAGE)
Protein samples for SDS-PAGE were electrophoresed in
1.5-mm discontinuous slab gels with a 3% gel stacked on a 5-15%
acrylamide gradient resolving gel(37) . The protein samples
were denatured by heating at 100 °C for 5 min in Laemmli sample
buffer containing 1% SDS with and without 100 mM dithiothreitol. The mobilities of mixtures of prestained protein
standards (Life Technologies, Inc.) containing myosin (200 kDa),
phosphorylase b (97 kDa), BSA (68 kDa), ovalbumin (43 kDa),
carbonic anhydrase (29 kDa), -lactoglobulin (18 kDa), and lysozyme
(14 kDa) were used as markers. 2
cm, containing the resolved proteins were excised from the gel. The gel
strips were incubated at 21 °C for 20 min in sample buffer
containing 2% SDS and 10 mM dithiothreitol and inserted into a
13-cm sample well of the second gel. Prestained standards were loaded
into a separate 7-mm well, and the running buffer for the second gel
contained 0.1 mM sodium thioglycollate. The separated proteins
were detected by silver staining as recommended by the manufacturer
(Bio-Rad).Immunoblotting
Proteins resolved by
two-dimensional gel analysis were electrophoretically transferred onto
nitrocellulose and analyzed by the Western blot method described (33) with 10 µg/ml anti-Id or 4 µg/ml
anti-phosphotyrosine and 10
cpm/ml goat anti-rabbit
F(ab`). The primary and secondary antibodies were diluted
in 0.01 M Tris, 0.15 M NaCl, 0.02% NaN,
2% BSA, pH 7.4, and blots of receptor proteins probed with nonimmune
rabbit IgG and the labeled secondary antibody were negative by
radioautography.Isoelectric Focusing
Two-dimensional
electrophoresis with isoelectric focusing in the first dimension and
SDS-PAGE in the second was carried out in slab gels as previously
described(38) , except that the final concentration of ammonium
persulfate was 2.4% (w/v) and the gel overlay was distilled water. The
focusing gels contained 5% acrylamide, 8 M urea, 1% 3/10
ampholytes (w/v), and 0.5% Nonidet P-40. The protein samples were
prepared in 8 M urea, 1% Nonidet P-40, 1% ampholytes, 0.5 M 2-mercaptoethanol, and 0.18% phenol red. The cathode (upper)
and anode (lower) compartments contained 0.02 N NaOH and
0.085% phosphoric acid, respectively; electrophoresis was carried out
at a constant voltage of 400 for 18-20 h. Strips (11 2
cm) containing the focused proteins were excised from the gel, rinsed,
and treated for 20 min in Laemmli sample buffer containing 2% SDS and
10 mM dithiothreitol before insertion into the troughlike well
of a 5-15% acrylamide gradient slab gel. A sample-free strip was
cut into 1-cm segments, and the ampholytes in each were eluted with
distilled water to determine the pH gradient. The gel-separated focused
proteins were either transferred onto nitrocellulose for immunoblot
analysis or fixed and stained with Coomassie Blue.Carbohydrate Analyses
For deglycosylation of
Asn-linked oligosaccharides, 12 µg of purified receptor protein
were denatured by boiling for 5 min in 0.1% SDS, 20 mM 2-mercaptoethanol, 0.08 M Tris buffer, pH 8.0, and
digested at 37 °C for 18 h in the presence of 1.3% Nonidet P-40 and
0.8 unit of recombinant N-glycanase enzyme (Genzyme Corp.,
Cambridge, MA) per 50 µl of reaction mixture. Reaction mixtures of
receptor protein without enzyme and parallel samples of 10 µg of
ovalbumin with and without enzyme were treated in a similar fashion.
The sample proteins were denatured in 8 M urea, 1% SDS, and
0.2 M 2-mercaptoethanol, incubated for 5 min in a boiling
water bath, and subjected to SDS-PAGE.Amino Acid Composition
Receptor proteins in
reduced samples were resolved by SDS-PAGE, transferred to Immobilon
polyvinylidene difluoride membranes (Millipore Corp., Bedford, MA) in
0.01 M 3-cyclohexylamino-1-propanesulfonic acid, pH 11.25,
containing 10% methanol, stained briefly with Coomassie Blue, and
destained. The stained bands were excised and submitted to William S.
Lane, Harvard Microchemistry Facility, Cambridge, MA, for acid
hydrolysis and analysis of phenylthiocarbamyl-amino acids. Analysis was
carried out for 1.5, 3.8, 16.6, and 23.2 pmol of the 95-, 60-, 27-, and
20-kDa polypeptides, respectively.
Subunit Composition of the
By SDS-PAGE under reducing conditions, the 180- and
160-kDa -Glucan Receptor
Proteins-glucan receptors present as a complex mixture of smaller
reduction products (32) . To identify the subunit constituents
of each receptor molecule, the receptor proteins were purified from
detergent-lysed U937 cells by affinity column chromatography with the
anti-Id, resolved by two-dimensional SDS-PAGE under nonreducing
conditions in the first dimension and reducing conditions in the
second, and detected by staining with silver. Electrophoretic
separation of 5 µg of purified protein showed both receptors
consisted of several disulfide-linked polypeptides (Fig. 3). The
180-kDa receptor contained polypeptides of 95, 88, 60, and 20 kDa,
whereas the 160-kDa receptor consisted of two polypeptides of 27 and 20
kDa at molar concentrations of 2 and 5, respectively. Analysis of four
different preparations showed variable levels of the 88-kDa
polypeptide, which was absent in smaller batches, suggesting that this
molecule represented a degraded form of the 95-kDa polypeptide. The
concentrations of two apparent aggregates of high molecular weight also
varied from preparation to preparation, and these were most prominent
when large batches of cells were processed for receptor purification.
Doubling the concentrations of protease inhibitors during column
loading, including them in the wash buffers prior to receptor elution,
and adding another thiol protease inhibitor, cystatin, to the mixture
of protease inhibitors during cell lysis failed to alter the molecular
nature or concentrations of the 95-(88-), 60-, 27-, and 20-kDa subunit
constituents of the 180- or 160-kDa receptor. Similar two-dimensional
analysis of receptors immunoprecipitated from detergent lysates of
surface-radioiodinated U937 cells by the anti-Id revealed the same
polypeptide constituents in radioautographs (not shown), indicating
that each subunit of the 180- and the 160-kDa -glucan receptors
was surface-expressed. The amino acid composition of the four subunit
constituents of the two -glucan receptors is shown in Table 1. The 95- and 27-kDa polypeptides had amino acid
compositions nearly identical to each other and these were more similar
to the 60-kDa than the 20-kDa polypeptide subunit.
-glucan receptor. Two-dimensional SDS-PAGE in 5-15%
acrylamide gradient gels was performed with 5 µg of
affinity-purified U937 cell receptor protein under nonreducing (NR) conditions in the first dimension and reducing (R) conditions in the second. The reference parent molecules
and their protein subunits were detected by staining with silver.
Analyses of four preparations of purified -glucan receptors showed
variations in the proportions of the apparent aggregates of high
molecular size and the 88-kDa subunit. The mobility of a BSA
contaminant falls midpoint on the diagonal and was detectable in lanes
with and without loaded sample. Mobility and size of prestained protein
standards are indicated.
Charge Characteristics and Glycosylation of
Two-dimensional electrophoresis
with isoelectric focusing in the first dimension and SDS-PAGE in the
second was carried out with 12 µg of the immunoaffinity-purified
U937 cell -Glucan
Receptor Subunit Constituents-glucan receptor proteins and the separated polypeptides
detected by staining with Coomassie Blue. The subunit polypeptides each
focused as discrete molecules with different isoelectric points (Fig. 4). The 95-kDa polypeptide focused at pI 5.0-5.2,
the 60-kDa at pI 5.6-5.7, and the 27-kDa at pI 4.8-5.1. The
20-kDa polypeptide focused as two species: a major one with a pI of
6.0-6.2 and a slightly more negative molecule with a pI of
5.7-5.9. Extending the electrophoretic separation on the focusing
gel from 8,000 to 12,000 V-h did not alter the migration of the subunit
constituents or result in comigration of the two 20-kDa moieties.
-glucan
receptor subunit constituents. Two-dimensional electrophoresis with
isoelectric focusing in the first dimension and SDS-PAGE in the second
was carried out under reducing conditions with 12 µg of
immunoaffinity purified U937 cell -glucan receptors. The pH
gradient in the focusing gel was determined by measuring the pH of
ampholytes eluted into distilled water from 1-cm segments
of the 5% slab gel, and the focused receptor polypeptides were
identified by staining the 5-15% acrylamide gradient sizing gel
with Coomassie Blue. Each of the subunit polypeptides exhibited the
same pI when focusing was increased from 8,000 to 12,000 V-h,
indicating that equilibrium had been
reached.
-glucan receptor protein or 10 µg of ovalbumin were incubated
for 24 h at 37 °C in the absence and presence of N-glycanase, and the resulting products were resolved by
SDS-PAGE. Under these enzymatic and electrophoretic conditions, the
ovalbumin control exhibited a downward shift in electrophoretic
mobility of about 1,000 daltons after treatment with N-glycanase (Fig. 5). The 95-kDa polypeptide contained
substantial amounts of N-linked carbohydrate, with the
predominant N-deglycosylated form of this molecule having a
molecular size of 76 kDa. Digestion with N-glycanase resulted
in less dramatic decreases in the apparent molecular weights of the 60-
and 27-kDa polypeptides and produced no detectable change in the 20-kDa
polypeptide. A likely dimer of incompletely reduced 20-kDa polypeptide (32) presented as a band of 40 kDa and, as with the fully
reduced 20-kDa subunit, showed no shift in electrophoretic mobility
after enzymatic digestion. Carbohydrate analysis by PAS staining can
potentially detect any oligosaccharide side chain of mammalian proteins
irrespective of linkage or sugar modification. For this analysis, 20
µg of purified -glucan receptor protein were resolved by
SDS-PAGE under reducing conditions, transferred onto nitrocellulose,
and assessed for carbohydrate by the PAS assay. A concurrent sample of
12 µg of receptor protein was separated on the same gel,
transferred onto nitrocellulose, and stained for protein. The rank
order of glycosylation for the four receptor constituents was the same
by PAS staining as that detected by N-deglycosylation (Fig. 6). The 95-kDa polypeptide had the highest content of
oligosaccharide, whereas the 20-kDa polypeptide had no detectable
carbohydrate despite its being the most prominent transferred
polypeptide subunit.
-glucan receptor constituent polypeptides. Reaction mixtures (50
µl) of reduced samples containing 12 µg of purified receptor
protein were incubated without (Control) and with 0.8 unit of N-glycanase (N-g'ase) for 24 h at 37 °C.
The resulting products were resolved by SDS-PAGE and stained with
Coomassie Blue. Parallel samples containing 10 µg of ovalbumin (OA) were treated in a similar fashion. Under these conditions
of hydrolysis, the ovalbumin control exhibited a downward shift in
electrophoretic mobility of about 1,000 daltons after treatment with N-glycanase. Mobility and size of prestained protein standards
are indicated.
-glucan receptor polypeptides. Samples of 20 µg
of purified receptor protein were resolved under reducing conditions in
5-15% acrylamide gradient gels by SDS-PAGE, transferred onto
nitrocellulose, and stained for carbohydrate (CHO) by
treatment with PAS reagent. Concurrent samples of 12 µg of purified
receptor proteins were stained for protein with Ponceau S. The lower
level of sensitivity of the PAS method was 5-10 ng of CHO for
glyco-protein standards electrophoresed and immobilized on
nitrocellulose.
Constitutive Phosphorylation of the
Constitutive phosphorylation has been
described for a number of proteins, among which are several
nonglycosylated proteins of the CD3 complex(39) . To determine
whether the charge heterogeneity exhibited by the 20-kDa polypeptide
was associated with differences in phosphorylation, 12 µg of
purified receptor protein were subjected to two-dimensional separation
by isoelectric focusing and SDS-PAGE, and the resolved polypeptides
were probed in immunoblots with anti-phosphotyrosine. The 95-kDa
subunit and both species of the 20-kDa polypeptide presented as
tyrosine phosphoproteins, whereas the 60- and 27-kDa polypeptides did
not (Fig. 7). Similar analysis of purified receptor proteins
pretreated with alkaline phosphatase (40) before
electrophoretic separation and immunodetection with
anti-phosphotyrosine showed no immunoreactive molecules (not shown).
The negativity of this control blot provided a convenient sample for
determining whether the two species of 20-kDa polypeptide exhibited the
same differential electrophoretic mobilities as their tyrosine
phosphorylated counterparts. For this analysis, the blot bearing the
dephosphorylated sample was washed and probed with the anti-Id. Both
species of dephosphorylated 20-kDa polypeptide were detectable with the
anti-Id and both were slightly shifted to more basic isoelectric
points, but the charge difference between the two variants was
unchanged (not shown).-Glucan Receptor
Subunit Polypeptides
-glucan receptor polypeptides. Two-dimensional gel analysis by
isoelectric focusing followed by SDS-PAGE was carried out under
reducing conditions with 12 µg of purified -glucan receptors.
The polypeptide subunits were transferred onto nitrocellulose, probed
with anti-phosphotyrosine, and detected by radioautography with I-labeled goat anti-rabbit F(ab`)
. Control
blots on Immobilon treated with 1 M KOH before incubation with
the primary antibody revealed reactivity by each of the three
polypeptide subunits with anti-phosphotyrosine. The pH gradient of the
focusing gel was determined as described in Fig. 4.
Receptor Protein Distribution of the 20-kDa
Constituents
Purification of the 180- and 160-kDa -glucan
receptors is achieved with the anti-Id by its binding the 20-kDa
polypeptides in both receptor molecules. To determine whether the
receptors were homogeneous in their content of 20-kDa polypeptide
variants, the 160-kDa receptor was purified by immunoadsorption with
mAb ACIA12 (Fig. 2) from a preparation of receptors isolated
with the anti-Id. The eluted proteins were subjected to two-dimensional
gel analysis by isoelectric focusing followed by SDS-PAGE and the
focused constituents of the 160-kDa receptor detected by Coomassie Blue
staining. The purified preparation was rendered free of the 180-kDa
-glucan receptor by adsorption with mAb ACIA12 and the resulting
160-kDa receptor was found to be heterogeneous in its content of 20-kDa
molecules (Fig. 8). Approximately 80% of the 20-kDa polypeptides
focused at a pI of 6.1-6.3 and 20% at a pI of 5.9-6.0. Both
of these polypeptides presented as more discrete and intense bands than
the 27-kDa subunit, which focused as a more diffuse band with a pI of
4.8-5.2. By densitometry, the relative proportion of the 20- and
27-kDa polypeptides of the 160-kDa parent protein was 2:1.
-glucan receptor. The 160-kDa
receptor was isolated by affinity purification with mAb ACIA12 and
reduced samples with 20 µg of purified protein were subjected to
two-dimensional gel electrophoresis by isoelectric focusing followed by
SDS-PAGE. The protein constituents in the sizing gel were detected by
staining with Coomassie Blue, and the pH gradient in the focusing gel
was determined as described in Fig. 4.
-glucan receptors on human myelomonocytic U937 cells and
investigate the biochemical properties of their disulfide-linked
subunit constituents. Immunologic studies on human monocyte and U937
cell -glucan receptors have shown both receptors to be multimeric
and have in common a 20-kDa polypeptide subunit which contains the
glucan binding domain(32) . The 20-kDa polypeptide is a key
subunit constituent for receptor function and contains epitopes for the
anti-idiotypic antibody previously shown to bind human monocytes and
selectively block their phagocytosis of particulate yeast
glucans(31) . As with anti-idiotypic antibodies to other
receptors(41) , ligand rather than membrane protein is used to
generate antibody, thereby limiting the antigenic specificity of the
final product to an antibody directed against ligand-binding sites in
the receptor (Fig. 1). The capacity of mAb OEA10
anti--glucan to recognize a haptenic ligand (31) , the
yeast heptaglucoside(27) , greatly increased the likelihood
that the anti-idiotypic antibody would be reactive and specific for
human -glucan receptors. The availability of U937 cells made it
possible to prepare the large amounts of purified -glucan receptor
protein needed for biochemical analyses. By two-dimensional analysis,
each of the two nonreduced receptors was reduced to its constituent
subunits, and each polypeptide constituent was detected by a sensitive
silver staining technique. The 180- and 160-kDa -glucan receptors
both presented as complex dulfide-linked structures with the
composition of one, indicating it to be multivalent (Fig. 3).
The 180-kDa receptor was composed of three disulfide-linked
polypeptides of 95, 60, and 20 kDa and was estimated to contain these
subunits at equimolar ratios. The composition of the 160-kDa receptor
was limited to two polypeptides of 27 and 20 kDa, which were linked to
each other by disulfide bonds at molar ratios of about 2 and 5,
respectively. The presence of as many as five 20-kDa subunits in each
160-kDa receptor molecule may relate to the homopolymeric nature of the
-glucan ligands and effect high avidity binding between
-glucan receptors and ligands. No significant differences in the
number or sizes of the receptor subunits were apparent by the inclusion
of additional protease inhibitors at various points during
purification, indicating that both -glucan receptors were cleaved
and bridged by disulfide bonds in their native state. The anti-Id
recognizes primary rather than conformational determinants in the
20-kDa polypeptide, making it possible to enrich for denatured or
native forms of the 20-kDa subunit during receptor purification.
Scale-up preparations of isolated receptor proteins, however, contained
no monomeric or partially assembled 20-kDa polypeptide subunits (Fig. 3), suggesting that each of the two -glucan receptors
is derived from a proreceptor by proteolytic processing at regulated
cleavage sites rather than by post-translational assembly.-glucan receptors. Compositional analysis of the four polypeptide
subunits from the two receptors revealed high similarities between the
amino acids of the 95-kDa subunit of the 180-kDa receptor and the
27-kDa subunit of the 160-kDa receptor protein, whereas relatively
little similarity in amino acid composition was apparent between the
20-kDa subunit and the three other subunit constituents (Table 1). Except for the 20-kDa polypeptide, each subunit
constituent of the 180- and 160-kDa -glucan receptors exhibited a
single isoelectric point, with the 95- and 27-kDa polypeptides focusing
at nearly the same pI (Fig. 4). The 20-kDa polypeptide focused
as 2 molecules with pI more basic than those of the other subunits,
which may reflect its higher content of lysine residues. An average
charge differential of 0.3 pH unit was expressed by the two variants of
the 20-kDa subunit, the more minor of which had the lower pI and
accounted for about 15-20% of the two 20-kDa molecules. Since
charge heterogeneity often resides with the quantities of sialic acid
residues in oligosaccharide side chains or modifications by sulfation
or phosphorylation of sugar residues, glycosylation of the -glucan
receptor constituents was the first parameter examined. The molecular
size of the 20-kDa polypeptide was unaffected by hydrolysis with N-glycanase, suggesting the absence of N-linked
oligosaccharides in this molecule (Fig. 5). A sensitive and more
general assay for carbohydrates was devised based on PAS staining. By
this method, monosaccharides containing vincinal hydroxyls are required
for positivity, a criterion met by nearly all oligosaccharide chains of
mammalian glycoproteins. Blots containing the 20-kDa polypeptide in
amounts 80-100-fold greater than the lower level of detection of
the typically glycosylated reference proteins showed no carbohydrate by
PAS staining (Fig. 6), indicating that the primary structure
and/or folding of the 20-kDa polypeptide precluded glycosylation.
Unlike the 20-kDa polypeptide, each of the other receptor constituents
was glycosylated as assessed by the enzymatic and chemical assays. In
both assays, the 95-kDa polypeptide presented as the most glycosylated
subunit, the 60-kDa as the least, and the 27-kDa as the most complex.
These data suggested that much of the carbohydrate detected by PAS
staining in these receptor constituents consisted of N-linked
oligosaccharides.-glucan receptors from unstimulated cells were phosphorylated
on tyrosine residues of the 95- and 20-kDa constituent subunits was
unexpected (Fig. 7). As indicated in the text, dephosphorylation
with alkaline phosphatase rendered these molecules free of phosphates
but had little if any effect on their overall electrophoretic
differences. That both variants of the 20-kDa polypeptide retained
reactivity with the anti-Id suggested that phosphoamino acids were not
located within the glucan-binding domain. Although constitutive
phosphorylation has been reported for a number of receptors, the vast
majority of these protein phosphorylations occur through serine or
threonine residues. For receptors on human mononuclear phagocytes,
cross-linking of FcRI (CD64) and Fc
RII (CD32) on U937 (42) and THP-1 (43) cells by receptor antibodies
stimulates tyrosine phosphorylation, but neither phagocytic receptor
appears to be constitutively phosphorylated. Constitutive rather than
stimulus-inducible phosphorylation is a feature of the
-chains
(CD11a, b, and c) of the 2 integrin proteins (CD11/CD18), for
which the phosphorylated substrate is serine (40) . Conversely,
the common -chain (CD18) is not phosphorylated in its basal state,
but treatment of human monocytes with soluble agents, such as phorbol
esters, rapidly and dramatically stimulates serine phosphorylation of
CD18. To our knowledge, no receptor on human mononuclear phagocytes has
been shown to be constitutively tyrosine phosphorylated prior to this
report.-glucan receptors resulted in a mAb specific for the
160-kDa receptor and provided a means for separating the two receptors (Fig. 2). Analysis by isoelectric focusing clearly showed the
purified 160-kDa receptor to contain two variants of the 20-kDa
polypeptide (Fig. 8). Comparison of the 20-kDa variants before
and after purification of the 160-kDa protein from receptor
preparations revealed no significant differences in their isoelectric
points or relative proportions, thereby diminishing the possibility
that each of the two receptors contained a different type of 20-kDa
polypeptide with a distinguishing isoelectric point. Since isoelectric
focusing is sensitive to a difference of a single changed amino acid,
the charge heterogeneity observed for the 20-kDa polypeptide was a
likely reflection of microheterogeneity in its primary structure. That
the compositions and relative concentrations of each 20-kDa variant
were comparable regardless of whether samples were stained with protein
dyes or antibodies indicated them to be highly homologous constituents
and occurring at fixed frequencies in the -glucan receptors. These
data were supported by those obtained with the purified 160-kDa
receptor protein, all of which are consistent with the 20-kDa
polypeptides being allelic variants.
)
We thank Drs. David S. Adams, Gary Ostroff, and Spiros
Jamas for their helpful discussions and support.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. G. Ma, M. Y. Cho, M. Zhao, J. W. Park, M. Matsushita, T. Fujita, and B. L. Lee Human Mannose-binding Lectin and L-Ficolin Function as Specific Pattern Recognition Proteins in the Lectin Activation Pathway of Complement J. Biol. Chem., June 11, 2004; 279(24): 25307 - 25312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. T. Borchers, C. L. Keen, and M. E. Gershwin Mushrooms, Tumors, and Immunity: An Update Experimental Biology and Medicine, May 1, 2004; 229(5): 393 - 406. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Masuoka Surface Glycans of Candida albicans and Other Pathogenic Fungi: Physiological Roles, Clinical Uses, and Experimental Challenges Clin. Microbiol. Rev., April 1, 2004; 17(2): 281 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. P. Fraser, L. Stuart, and R. A. B. Ezekowitz TLR-independent pattern recognition receptors and anti-inflammatory mechanisms Innate Immunity, April 1, 2004; 10(2): 120 - 124. [Abstract] [PDF]
|
|
|
|
 |
|
 C. Fradin, D. Poulain, and T. Jouault beta -1,2-Linked Oligomannosides from Candida albicans Bind to a 32-Kilodalton Macrophage Membrane Protein Homologous to the Mammalian Lectin Galectin-3 Infect. Immun., August 1, 2000; 68(8): 4391 - 4398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Zimmerman, J. Lindermuth, P. A. Fish, G. P. Palace, T. T. Stevenson, and D. E. DeMong A Novel Carbohydrate-Glycosphingolipid Interaction between a beta -(1-3)-Glucan Immunomodulator, PGG-glucan, and Lactosylceramide of Human Leukocytes J. Biol. Chem., August 21, 1998; 273(34): 22014 - 22020. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |