| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
The Biomedical Research Centre, University of British Columbia, Vancouver, British Columbia, Canada; and
Division of Hematology/Oncology, Tupper Research Institute, New England Medical Center, Boston, MA 02111
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
200 kDa
identified as a hyposialylated form of CD45RB preferentially expressed
on peripheral CD8, but not CD4, T cells. Our data also show that the
recognition of CD43 and CD45RB by 1B11 is differentially affected by
O-linked glycosylation and sialic acid. Whereas 1B11
recognition of CD43 on activated T cells required both core 2
O-glycan branching and sialic acid, 1B11 recognition of
CD45 only occurred in the absence of both core 2 glycosylation and
sialic acid. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-helix and maintained in an extended form by its heavy
O-linked glycosylation (1). Because of this
structure, most mAbs to CD43 are thought to recognize linear epitopes,
and binding is often influenced by changes in glycosylation or removal
of sialic acid with neuraminidase (1). CD43 is expressed
by most lymphohemopoietic cells, except resting B cells. Despite many
indications that CD43 influences cellular adhesive interactions and can
participate in signaling events (reviewed in Ref. 2),
CD43-deficient mice display relatively limited perturbations in
phenotype relating to increased leukocyte adhesion to vascular
endothelium and failure to emigrate through endothelium to sites of
inflammation (3, 4, 5).
Two glycoforms of CD43 have been distinguished by mAbs S7 and 1B11
(6), the former recognizes a 115-kDa glycoform carrying
tetrasaccharides (7), and the later recognizes a 130-kDa
glycoform carrying branched O-glycans forming
hexasaccharides and expressed predominantly on bone marrow cells and
activated T cells (6). Whether the differential
glycosylation and sulfation (8) of CD43 are relevant to
its function is currently unresolved. The presence of the 115-kDa vs
the 130-kDa CD43 glycoform corresponds with expression of the core 2
-1,6-N-acetylglucosaminyltransferase (core 2
GlcNAcT)3(9); core 2 GlcNAcT is
up-regulated during T cell activation and generates branched
O-linked oligosaccharides that are responsible for the
greater mass of the 130-kDa CD43 glycoform expressed in activated T
cells (10).
The regulation of CD43 glycoforms as determined by mAbs 1B11 and S7 was found to be complex, particularly in T cells (6, 11, 12), and when CD43-deficient mice became available (3) they were examined for expression of these epitopes. CD43-/- mice failed to express the S7 epitope on leukocytes, as expected, but surprisingly retained expression of an epitope recognized by 1B11 on CD8 T cells. To comprehensively define the specificity of 1B11 and to understand the basis for the restricted tissue distribution of the new epitope we sought to determine the identity of the CD43-independent molecule recognized by 1B11 on CD43-/- CD8 T cells.
In this report we describe the characterization of a new specificity of mAb 1B11 for a hyposialylated RB isoform of CD45. Structural similarities between the extracellular portion of CD43 and the variable N-terminal region of extracellular CD45 are consistent with the possibility of Ab cross-reaction. However, 1B11 binding to the epitopes on CD43 and CD45RB is dramatically, but, oppositely, influenced by sialic acid and glycan branching structures determined by core 2 GlcNAcT.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice were bred in-house at The Biomedical Research Centre. Spleen cells from C2GlcNAcT-/- mice (13) were supplied by Drs. L. G. Ellies and J. D. Marth (Howard Hughes Medical Institute, La Jolla, CA).
Media
Cells were maintained in RPMI (Life Technologies, Gaithersburg, MD) supplemented with 10% (v/v) FCS, 2 mM glutamine, antibiotics, and 50 µM 2-ME in a humidified atmosphere containing 5% CO2 at 37°C.
Antibodies
1B11-FITC was obtained from PharMingen (San Diego, CA) (IgG2a, 09694D); H18 polyclonal rabbit anti-peptide Ab recognizing the cytoplasmic domain of CD43 was produced as previously described (6). 1F10 is an IgG2a mAb reactive with a specific rabbit anti-peptide Ab Id and was used as an isotype control for mAbs S7 and 1B11. 1F10 was produced as ascites in our laboratory. I3/2 rat IgG (14) is a pan-CD45-specific mAb. R02 antiserum was obtained from rabbits immunized with Escherichia coli-expressed cytoplasmic domain of CD45, provided by Dr. Pauline Johnson. 4B4 is an IgG anti-CD45RB-specific Ab (15). Other reagents used were anti-CD4-PE (09005B, PharMingen), anti-CD8-biotin (01042D, PharMingen), and streptavidin cychrome (13038A, PharMingen).
Cell lines
Core 2 GlcNAcT-transfected EL4 cells were generated as previously described (9) with the murine core 2 GlcNAcT cDNA obtained from the plasmid pmC2-251 (C. E. Warren, D. S. Smookles, and J. W. Dennis, unpublished observations; GenBank accession no. U19265). EL4 cells transfected with pMSCVneoEB (16) were used as controls. The myelomonocytic cell line D26 (17) and the T cell line CTLL (18) were obtained from Dr. John Schrader.
Flow cytometry
Cells were suspended in FACS buffer containing 2% (v/v) FCS in PBS and were incubated for 40 min at 4°C with Abs in 96-well round-bottom plates (Nunclon, InterMed, Roskilde, Denmark). Cells were washed twice and were analyzed on a FACScan IV flow cytometer (Becton Dickinson, Mountain View, CA).
Lysates
Briefly, cells were washed in PBS and lysed at 
5 x
107/ml in lysis on ice in buffer containing 20 mM
Tris (pH 7.5), 0.15 M NaCl, 0.5% Nonidet P-40, 10 µg/ml leupeptin,
174 µg/ml PMSF, and 5 µg/ml pepstatin. Lysates were mixed by
rotation for 30 min at 4°C and were spun for 2 min at top speed in an
Eppendorf centrifuge. Supernatants were then recovered for SDS-PAGE or
used for immunoprecipitations.
Immunoprecipitations
Abs were bound to protein G-coupled Sepharose beads (Pharmacia,
Piscataway, NJ) at a ratio of 1–2 µg of purified Ab/µl of packed
beads. After washing beads free of unbound Ab with PBS, cell lysates
prepared in 0.5% Nonidet P-40 were mixed with beads and incubated at
4°C for 1–2 h. Immunoprecipitates were then washed in TBS containing
0.1% Tween-20 (Fisher, Nepean, Canada). Immunoprecipitated protein was
extracted from beads with Laemmli loading buffer. 2-ME was added to
reduce the Ab where necessary, and samples were then loaded for
SDS-PAGE. To demonstrate 1B11-CD45 expression in freshly isolated CD8 T
cells, shown in Fig. 6
C, CD4 and CD8 lymph node cells from
12 mice were purified by depletion of surface
IgG+ and CD8+ cells or
surface IgG+ and CD4+ cells
with magnetic beads to 92 and 78% purity, respectively. Lysates from
15 x 106 cells were prepared and
preabsorbed with sheep anti-mouse Ig-coupled Sepharose, subjected
to immunoprecipitation with mAb I3/2 (pan-CD45) for 60 min, and then
captured with protein G-Sepharose. Immunoprecipitates were washed and
reduced before immunoblotting with 20 µg/ml 1B11.
|
Blots were generated and probed using standard methods. Lysates or immunoprecipitates were combined with Laemmli sample buffer and loaded into minigels (Bio-Rad, Richmond, CA) with a 4% stacking gel and a 6% resolving gel. Resolved protein was transferred to PROTRAN-BA85 nitrocellulose membrane (Schleicher & Schuell, Keene, NH) that was subsequently blocked with 5% BSA for rat mAb, or 5% skim milk powder for rabbit antiserum in TBS. Blots were probed with Abs diluted in TBS containing 0.5% of the corresponding blocking protein solution (BSA or milk) and 0.5% Tween for 60 min. Blots were washed in TBS with 0.1% Tween and were detected with goat anti-mouse Ig-HRP or goat anti-rabbit HRP where appropriate (Life Technologies), washed, and developed with enhanced chemiluminescence reagent (Amersham, Oakville, Canada) for autoradiography with Biomax film (Eastman Kodak, Rochester, NY) according to the manufacturer’s instructions. Kaleidascope or high range prestained m.w. standards were used (Bio-Rad).
Neuraminidase treatment
Cells were pelleted and resuspended in HBSS supplemented with 5% FCS and 250 mU/mL of neuraminidase (Clostridium perfringens, 1585886, Boehringer Mannheim, Indianapolis, IN) at room temperature and mixed gently for 50 min. Untreated control cells were treated identically, except that neuraminidase was omitted. Cells were pelleted, washed once in PBS, and dissolved in lysis buffer or processed for flow cytometry.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous investigations had determined that mAb 1B11 was specific
for the 130-kDa sialomucin CD43 (6). In the process of
further 1B11 characterization, CD43-/-
lympho-myeloid tissues were assayed for 1B11 binding using flow
cytometry. Binding of 1B11 was compared with binding of mAb S7 reactive
with 115-kDa CD43 and anti-Dd (negative
control). When lymphoid and myeloid cells from
CD43+ and CD43-/- mice
were compared for 1B11/S7 binding, the S7 epitope was lost on all
tissues examined, as expected (data not shown) (3),
whereas a subset of LNC and splenocytes retained a significant level of
1B11 staining as shown in Fig. 1 (and
data not shown). The persistent 1B11 binding was most obvious in the
lymph node, where 1B11 staining did not significantly differ between
CD43+ and CD43-/- cells.
Among thymocytes, the absence of CD43 effectively reduced the intensity
of 1B11 staining, whereas in bone marrow cells from
CD43-/- mice 1B11 staining was eliminated.
CD43-/- T cells activated by Con A initially
lost all 1B11 binding (data not shown), but regained low levels of 1B11
binding after several days of incubation and subculture. These results
reinforce the conclusions that CD43 was indeed absent in
CD43-/- mice and that 1B11 bound to a
CD43-indpendent epitope that was present predominantly on a subset of
LNC and was not expressed by myeloid cells from bone marrow.
|
demonstrated that virtually all the
1B11 staining occurred on CD8+ T cells, whereas
CD4+ T cells and
CD4-8- cells (B cells)
failed to stain significantly. The unusual restriction of 1B11 staining
to CD8+ T cells indicated that 1B11 Ab might
distinguish a new specificity of some importance to the function of
this cell subpopulation and prompted us to pursue the identification of
the new epitope.
|
Previous analysis of the carbohydrate dependence of 1B11 binding
had shown that neuraminidase pretreatment of lymphocytes enhanced Ab
binding (6). CD43-/- lymphocytes
were treated with C. perfringens neuraminidase, which
cleaves 2-3, 2-6, and 2-8 sialic acids from N- or
O-glycans. Flow cytometric analysis shown in Fig. 3 demonstrated that neuraminidase
effectively enhanced 1B11 binding in LNC from
CD43-/- mice, whereas neuraminidase-treated
bone marrow cells from the same mice remained 1B11 negative. 1B11
staining of neuraminidase-treated LNC yielded two fluorescence peaks, a
minor 1B11low peak and a major
1B11high peak. Three-color analysis demonstrated
that the major 1B11high peak was composed of both
CD4 and CD8 T cells; CD8 T cells were the most positive for 1B11,
whereas the minor 1B11low peak consisted of
CD4+8- and
CD4-8- LNC, which are
known to be predominantly B cells. Thus, neuraminidase-enhanced 1B11
binding on CD4 and CD8 T cells was observed preferentially on lymph
node T cells, while the neuraminidase effect was not observed on
myeloid cells from bone marrow. 1B11 staining of untreated or
neuraminidase-treated spleen cells was similar in intensity to that
observed in LNC, except that the numbers of 1B11+
cells paralleled the reduced proportions of T cells in the spleen (data
not shown). We also observed that neuraminidase treatment had little
effect on 1B11 staining of Con A-activated T cells from
CD43-/- mice (data not shown).
|
Initial efforts to precipitate the CD43-independent 1B11 target
failed, and we concluded that either the epitope was too low in
abundance or it was destroyed during processing of samples for Western
blotting (see below). The finding that neuraminidase enhanced 1B11
staining prompted renewed immunoprecipitation efforts. Western blot
analyses exemplified in Fig. 4
demonstrated that 1B11 could indeed detect and precipitate an
200-kDa protein from lysates of neuraminidase-treated lymph node
cells from both CD43+ and
CD43-/- tissue. The detection of p200 required
that samples not be denatured before electrophoresis; while either
boiling under nonreducing conditions or reduction alone was tolerated
with only minor loss of reactivity, boiling under reducing conditions
destroyed the p200 epitope recognized by 1B11.
|
EL4 thymoma cells express p200 after neuraminidase treatment
To facilitate the characterization of p200, different cell lines
were examined for neuraminidase-inducible expression of 1B11 binding.
Flow cytometry revealed variable effects of neuraminidase treatment on
1B11 staining of various cell lines; three examples are illustrated in
Fig. 5A. EL4 thymoma cells
exhibited a significant increase in 1B11 staining after neuraminidase
treatment, as was observed in LNC, whereas 1B11 staining was unaffected
by neuraminidase treatment of the mature T cell line CTLL and was
reduced in myelo-monocytic D26 cells. Western analysis of lysates
derived from these cell lines (Fig. 5B) confirmed that p200
was expressed in EL4 cells, but was not detected in CTLL or D26 cells.
These results also illustrate that the effect of neuraminidase on
expression of the 130-kDa CD43 1B11 epitope was influenced by the cell
line used; 130-kDa CD43 1B11 expression was resistant to neuraminidase
in CTLL cells but sensitive in D26 and LNC (Fig. 4
). Finally, the
effect of neuraminidase on 1B11 epitope expression was complex and
varied considerably from cell line to cell line. Subsequent analysis
focused on identification of the CD43-independent p200 observed on EL4
and LNC using EL4 as a relatively accessible and abundant source of
neuraminidase-inducible p200 Ag.
|
The tissue distribution and apparent molecular mass of 200 kDa
suggested that p200 might be CD45. To address this possibility, 1B11 or
anti-CD45 immunoprecipitates were prepared from
neuraminidase-treated lymph node cells and EL4 cells. These
immunoprecipitates were then immunoblotted and probed with
anti-CD45 or 1B11, respectively. Fig. 6A shows that CD45 (200kDa)
was precipitated by 1B11 from lysates of neuraminidase-treated LNC. It
was also noted that despite the presence of multiple CD45 isoforms in
LNC lysates (lane WCL in Fig. 6A), 1B11
appeared to selectively precipitate a single CD45 isoform. As shown in
Fig. 6B Abs to CD45 precipitated the same p200 recognized by
1B11 from neuraminidase-treated EL4 lysates. These results confirm that
a form of CD45 is indeed recognized by 1B11 and also suggested that a
low level of 1B11-detectable p200 was present in EL4 before
neuraminidase digestion. Two major species and one minor species of
CD45 were detected with anti-CD45 (R02) in EL4 lysates, and
neuraminidase treatment only marginally altered their mobility on
electrophoresis (Fig. 6B). The signal detected by 1B11
corresponded to the intermediate form of the three species detected by
anti-CD45 (R02). Multiple high molecular mass isoforms of CD45
exist that span 180–240 kDa in different cell lineages. Insofar as
expression of the lower (180–200 kDa) molecular mass isoforms of CD45
are associated with T cells and EL4 cells, we examined whether p200
might be a CD8 lineage-restricted single exon isoform of CD45. In both
LNC and EL4 cells 1B11 did not detect the CD45 isoform with the lowest
apparent mass, presumably CD45R0 (180 kDa). 1B11 had primary reactivity
with the next largest (single exon form) of CD45. It was therefore of
interest to determine which single-exon CD45 isoform was recognized
(see below).
Given the importance of CD45 in lymphocyte signaling we sought to
confirm that the p200-CD45 observed after neuraminidase treatment was
indeed present in CD8+ LNC before neuraminidase
treatment. Flow cytometry data presented in Fig. 2 indicated that the
CD43-independent epitope of 1B11 was expressed on CD8-lineage LNC, but
not on CD4 LNC. Therefore, lysates were prepared from purified CD4 or
CD8 cells and probed for differential expression of the 1B11-CD45
epitope. In our hands, p200-CD45 was not detectable by immunoblotting
of whole cell lysates from CD8 LNC, presumably due to its low
abundance. However, when CD45 was immunoprecipitated from lysates of
purified CD4 and CD8 cells with pan-CD45-specific mAb, as shown in Fig. 6C, we were able to confirm that 1B11-CD45 was indeed
present in ex vivo CD8 cells, while no 1B11 signal was detected in CD4
cells.
1B11 recognizes desialylated CD45RB
Monoclonal anti-CD45RB Ab 4B4 (19) was used to
further evaluate 1B11-CD45 specificity. However, we found that the
CD45RB epitope recognized by 4B4 was destroyed by neuraminidase (Fig. 7A). Therefore, use of 4B4 as
a probe of 1B11 specificity would have to accommodate the differential
impact of sialic acid on epitope recognition. To accommodate these
differences, neuraminidase treatment was applied after 4B4
immunoprecipitation of CD45RB, and samples were then processed for
immunoblotting with 1B11. As shown in Fig. 7B, 4B4 was
effective at precipitating a p200 that was readily detected with 1B11
only after digestion of the immunoprecipitate with neuraminidase. This
result confirmed that 1B11 recognized a hyposialylated form of
CD45RB.
|
Previous analysis of 1B11 reactivity with 130-kDa CD43
demonstrated that core 2 GlcNAcT expression was required
(9). To further delineate the contribution of carbohydrate
to 1B11 epitope expression, we evaluated 1B11 recognition of 130-kDa
CD43 and CD45RB in Con A-activated T cells from mice deficient in core
2 GlcNAcT (13). The results shown in Fig. 8A confirmed that 1B11 readily
detected 130-kDa CD43 from normal Con A blasts and that this signal was
lost in lysates from core 2 GlcNAcT-/- Con A
blasts, consistent with the view that the 1B11 epitope on CD43 required
branching of O-linked oligosaccharides. As noted above (Fig. 4), 1B11 recognition of 130-kDa CD43 on LNC was also dependent on the
presence of sialic acid; exposure to neuraminidase destroyed the
130-kDa CD43 1B11 signal from core 2 GlcNAcT+
cells. In contrast, the 200-kDa CD45RB signal detected by 1B11 was only
apparent in core 2 GlcNAcT-/- Con A blasts that
had been exposed to neuraminidase. These results demonstrate a
differential requirement for sialic acid and core 2 branched
oligosaccharides in the effective presentation of 1B11 epitopes on
130-kDa CD43 and CD45. Although the 1B11 epitope on CD45RB is obscured
by either core 2 branched oligosaccharides or sialic acid, both
carbohydrate modifications are required for 1B11 recognition of the
130-kDa CD43 epitope.
|
). Furthermore, in earlier work we
showed that EL4 cells transfected with the core 2 GlcNAcT
gene express the 130-kDa form of CD43 and gain 1B11 staining
(9). As shown in Fig. 8B, neuraminidase
treatment of these core 2 GlcNAcT-EL4 cells resulted in the loss of the
130-kDa CD43 epitope, while the 200-kDa 1B11-CD45RB epitope was exposed
at considerably lower levels than those seen with the untransfected EL4
cells. These effects were observed despite the presence of equal
amounts of CD43 and CD45 as detected by carbohydrate-independent
anti-CD43 and anti-CD45 antisera H18 and R02, respectively. In
summary, results obtained with Con A blasts from core 2
GlcNAcT-/-mice and from EL4 cells transfected
with core 2 GlcNAcT were both consistent with the view that the 1B11
epitopes on CD43 and CD45 were differentially affected by sialic acid
and core 2 O-glycan branching. 1B11 reactivity with 130-kDa
CD43 required both sialylation and core 2 GlcNAcT branched
oligosaccharides, whereas the 1B11 epitope on CD45RB was eliminated
when O-glycans where either sialylated or core 2 GlcNAcT
branched.
Finally, of note in the data presented in Fig. 8A is the
failure of core 2 GlcNAcT deficiency to substantially reduce the
electrophoretic mobility of 130-kDa CD43 in LNC as detected by H18
antisera. This was unexpected because the 130-kDa CD43
hexasaccharide form is believed to be distinguished from the
115-kDa CD43 unbranched tetrasaccharide form primarily by antennae that
extend from the core 2 branch. In EL4 cells, however, the presence of
core 2 GlcNAcT did have a substantial effect on the apparent mass of
CD43 as detected by H18 (9).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD45 phosphatase has a major role in regulating signaling events through the T and B cell receptors (20, 21). The extracellular portion of CD45 consists of a membrane-proximal cysteine-rich domain (22) and a distal (N-terminal) mucin-type domain (23). As the most outward projecting part of CD45, the mucin domain would have the greatest proclivity for interaction with extracellular ligands. In addition to its mucin-like structure, differential splicing of exons forming the N-terminal domain of CD45 generate substantial variability in protein sequence (24). The regulated expression of these isoforms during differentiation/activation (15, 24, 25) together with the phylogenetic conservation of both mucin structure and variable exon structure (26) reinforce the view that CD45 isoforms are of functional importance (27, 28). Moreover, T cells from mice genetically deficient in CD45 and reconstituted with either CD45R0 or CD45RABC isoforms failed to support positive selection in the HY TCR transgenic model and failed to home normally to spleen, skin, and liver. CD45RABC T cells also failed to mature normally into cytolytic T cells in response to alloantigen or TCR cross-linking. Although the ligand(s) and precise function of the variable N-terminal domain of CD45 remain obscure, these results demonstrate that these isoforms play different roles and are important for normal development, function, and homing properties of T cells (29). Identification by 1B11 of a hyposialylated form of CD45RB on peripheral CD8 T cells raises the possibility that this CD45RB glycoform might be of functional significance in these cells.
The distinctive feature of CD45RB associated with 1B11 staining is that 1B11 recognizes only a hyposialylated form of CD45RB. Relevant 1B11-CD45RB epitopes exist on CD8 and most CD4 cells, but those expressed by CD4 cells are invisible to 1B11 due to the effects of sialic acid. Although CD45RB hyposialylation has been described in HIV infection (30), the significance of 1B11-CD45RB being selectively present on CD8 cells and absent from CD4 cells in normal mice is unknown. It would seem unlikely that glycan microheterogeneity would account for this difference insofar as CD4 cells expressing CD45RB would be expected to exhibit similar microheterogeneity. Recent experiments using in vivo models of immunological memory have provided evidence that through the action of endogenous neuraminidase (31) CD4 and CD8 T cells can desialylate glycoproteins thought to include CD45 and CD43 upon activation and that the desialylated state can persist in memory CD8 T cells (32). We are currently investigating whether 1B11 can be used to distinguish CD8 memory cells in appropriate in vivo models, or whether this minor subpopulation of CD45RB differentially associates with other molecules that might participate in TCR signaling, such as CD8.
Previous research on the impact of O-glycans on protein structure and the nature of epitopes associated with mucins such as CD43 and CD45 offers some insight into the likely 1B11-CD43 and 1B11-CD45 epitopes. The extracellular portion of CD43 and the N-terminal variable domain of CD45 including the RB exon share structures typical of mucins (33). This structure can be characterized as heavily O-glycosylated and enriched in serine, threonine, and proline, but deficient in cysteine. These features promote a loose coil topology with an extended conformation (1). The effect of this structure on antigenicity is to skew the Ab response toward linear epitopes over conformational epitopes. This skewing is evident in the proportion of Abs raised against CD43 and the N-terminal mucin domain of CD45 that recognize linear epitopes (1, 34). However, our observation that the 1B11 epitope on CD45RB was destroyed by denaturation is not consistent with the view that the 1B11-CD45RB epitope is present in a simple linear form. The absence of cysteines in the N-terminal segments of murine CD45 implies that the membrane-proximal domain of CD45, which contains multiple cysteine residues (23) and fibronectin repeats (22), contributes to formation of the 1B11-CD45 epitope. The nature of this contribution is unresolved and reconciling the structural considerations described above together with the apparent specificity of 1B11 for CD45RB and the involvement of the membrane-proximal domain implied by denaturation sensitivity is a challenge! Our data argue that there is an interaction between the mucin domain and the cysteine-rich domain contrary to most current models of CD45 structure (1, 35).
The majority of Abs raised against CD43 and the variable N-terminal domain of CD45 recognize the protein backbone, not carbohydrate per se (1, 34). Thus, despite the abundance of O-glycans and the frequency of Abs whose recognition is affected by carbohydrate, the influence of O-glycans on Ab recognition of mucins appears to reflect perturbation of backbone structure or direct obstruction of backbone accessibility. Indeed, peptide structure is considered to be most heavily influenced by the first and second O-glycan residues, that would presumably include branched residues if present, but is minimally perturbed by carbohydrate residues three positions from the backbone (1).
Abs whose accessibility is subject to the influence of carbohydrate are
of interest as probes for visibility of the backbone. It is possible
that such visibility could also be relevant for the physiological
ligands of CD43 and CD45. We have examined 1B11 reactivity primarily in
resting and activated T cells. The O-glycan structures
present on CD43 in these cells have been described (36)
and are shown in Fig. 9. Presuming that
glycans on CD45RB and CD43 are similar, our data are consistent with a
model in which 1B11 can recognize the protein backbone of CD43 and
CD45RB with O-glycan structures 1 and 2, respectively, as
shown.
|
and 5B indicated that neuraminidase
treatment of LNC and myelomonocytic D26 cells prevented recognition of
CD43 by 1B11. These results led us to conclude that sialic acid
residues were required to generate the 1B11 epitope on the 130-kDa
isoform of CD43. Sialic acid residues could contribute either bulk or
charge effects required for expression of the 1B11 epitope. In contrast
to D26 and LNC, neuraminidase had no effect on the expression of
1B11-CD43 on CTLL cells. One possible explanation for the differential
neuraminidase sensitivity of 1B11-CD43 may relate to differences in
glycosyl extensions from the core 2 branch point, e.g.,
polylactosamines (configuration 3 in Fig. 9) present on CTLL cells and
absent on D26 cells, thereby providing bulk to the branch that would be
resistant to the effects of neuraminidase. Consistent with this idea is
our observation that 1B11 detects CD43 in CTLL cells with an apparent
molecular mass of 140 kDa compared with that of 130 kDa for D26-derived
CD43 (6).
Finally, in the analysis of activated T cells from core 2
GlcNAcT-deficient mice (Fig. 8), it was apparent that CD43 did not
exhibit a broad reduction in molecular mass. This result was
unexpected, because the 115-kDa CD43 detected by the mAb S7 is thought
to represent CD43 carrying unbranched O-glycans
(6). The fact that in activated T cells derived from core
2 GlcNAcT-deficient mice a substantial proportion of CD43 has an
apparent molecular mass above 115 kDa reinforces the contention that
there may be redundant glycosyltransferase(s) that can in some cases
compensate for a loss of core 2 enzyme GlcNAcT activity (37, 38). Such redundancy may account for the limited phenotype of
core 2 GlcNAcT-deficient mice (13).
It is also interesting to note that an Ab against human CD45 exhibits major cross-reaction with a 130-kDa protein that is up-regulated in activated T cells (39), probably CD43, suggesting that 1B11 may identify a phylogenetically conserved structure shared by CD43 and CD45.
In summary, this report describes our observations on the specificity of mAb 1B11 for both CD43 and CD45RB. The functional activity of hyposialylated CD45RB is currently unknown, but the restriction of its expression to resting, peripheral CD8 T cells provides a basis for the use of 1B11 as a probe for lineage-dependent differences in CD45 glycosylation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Hermann Ziltener, The Biomedical Research Centre, University of British Columbia, 2222 Health Sciences Mall, Vancouver, British Columbia, Canada V6T-1Z3. E-mail address:
3 Abbreviations used in this paper: core 2 GlcNAcT, core 2 -1,6-N-acetylglucosaminyltransferase; LNC, lymph node cells.
Received for publication April 2, 1999. Accepted for publication May 21, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1–3GalNAc-R: specificities of core 2, UDP-GlcNAc: Gal1–3 GalNAc-R(GlcNAc to GalNAc)6-N-acetylglucosaminyltransferase and CMP-sialic acid: Gal1–3GalNAc-R3-sialyltransferase. Glycoconj. J. 10:381.[Medline]
6N-acetylglucosaminyltransferase. J. Biol. Chem. 266:23863.This article has been cited by other articles:
|
|
 |
|
  E. L. Stone, M. N. Ismail, S. H. Lee, Y. Luu, K. Ramirez, S. M. Haslam, S. B. Ho, A. Dell, M. Fukuda, and J. D. Marth Glycosyltransferase Function in Core 2-Type Protein O Glycosylation Mol. Cell. Biol., July 1, 2009; 29(13): 3770 - 3782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-W. Lee, Y. Park, A. Song, H. Cheroutre, B. S. Kwon, and M. Croft Functional Dichotomy between OX40 and 4-1BB in Modulating Effector CD8 T Cell Responses J. Immunol., October 1, 2006; 177(7): 4464 - 4472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Zelinskyy, S. J. Robertson, S. Schimmer, R. J. Messer, K. J. Hasenkrug, and U. Dittmer CD8+ T-Cell Dysfunction due to Cytolytic Granule Deficiency in Persistent Friend Retrovirus Infection J. Virol., August 15, 2005; 79(16): 10619 - 10626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. Berger, A. A. S. Akha, and R. A. Miller A glycoprotein endopeptidase enhances calcium influx and cytokine production by CD4+ T cells of old and young mice Int. Immunol., August 1, 2005; 17(8): 983 - 991. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Merzaban, J. Zuccolo, S. Y. Corbel, M. J. Williams, and H. J. Ziltener An Alternate Core 2 {beta}1,6-N-Acetylglucosaminyltransferase Selectively Contributes to P-Selectin Ligand Formation in Activated CD8 T Cells J. Immunol., April 1, 2005; 174(7): 4051 - 4059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Carlow, M. J. Williams, and H. J. Ziltener Modulation of O-Glycans and N-Glycans on Murine CD8 T Cells Fails to Alter Annexin V Ligand Induction by Galectin 1 J. Immunol., November 15, 2003; 171(10): 5100 - 5106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Onami, L. E. Harrington, M. A. Williams, M. Galvan, C. P. Larsen, T. C. Pearson, N. Manjunath, L. G. Baum, B. D. Pearce, and R. Ahmed Dynamic Regulation of T Cell Immunity by CD43 J. Immunol., June 15, 2002; 168(12): 6022 - 6031. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Carlow, S. Y. Corbel, M. J. Williams, and H. J. Ziltener IL-2, -4, and -15 Differentially Regulate O-Glycan Branching and P-Selectin Ligand Formation in Activated CD8 T Cells J. Immunol., December 15, 2001; 167(12): 6841 - 6848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-C. Lim, H. Xie, C. E. Come, S. I. Alexander, M. J. Grusby, A. H. Lichtman, and F. W. Luscinskas IL-12, STAT4-Dependent Up-Regulation of CD4+ T Cell Core 2 {beta}-1,6-n-Acetylglucosaminyltransferase, an Enzyme Essential for Biosynthesis of P-Selectin Ligands J. Immunol., October 15, 2001; 167(8): 4476 - 4484. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Snapp, C. E. Heitzig, L. G. Ellies, J. D. Marth, and G. S. Kansas Differential requirements for the O-linked branching enzyme core 2 {beta}1-6-N-glucosaminyltransferase in biosynthesis of ligands for E-selectin and P-selectin Blood, June 15, 2001; 97(12): 3806 - 3811. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Biller, K. Mardberg, H. Hassan, H. Clausen, A. Bolmstedt, T. Bergstrom, and S. Olofsson Early steps in O-linked glycosylation and clustered O-linked glycans of herpes simplex virus type 1 glycoprotein C: effects on glycoprotein properties Glycobiology, December 1, 2000; 10(12): 1259 - 1269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Harrington, M. Galvan, L. G. Baum, J. D. Altman, and R. Ahmed Differentiating between Memory and Effector Cd8 T Cells by Altered Expression of Cell Surface O-Glycans J. Exp. Med., April 3, 2000; 191(7): 1241 - 1246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Galvan, S. Tsuboi, M. Fukuda, and L. G. Baum Expression of a Specific Glycosyltransferase Enzyme Regulates T Cell Death Mediated by Galectin-1 J. Biol. Chem., May 26, 2000; 275(22): 16730 - 16737. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
