| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Veterinary Molecular Biology, Montana State University, Bozeman, MT 59717; and
Department of Pathology, University of Michigan School of Medicine, Ann Arbor, MI 48109
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
60% and P11.4 precipitated CD62L from detergent lysates of
PMA-activated Jurkat cells. In contrast, P11.4 precipitated minimal
amounts of CD62L from detergent lysates of nonactivated human PBL. As
reported previously, P-selectin glycoprotein ligand 1 and a distinct
130-kDa glycoprotein were the major species in these precipitates.
However, T cell activation on plate-immobilized anti-CD3 and growth
in low-dose IL-2 increased the percentage of CD62L molecules with
E-selectin-binding activity. After two cycles of activation and
culture, 60–70% of the CD62L was precipitated with the P11.4
chimera. These cultured T lymphoblasts rolled avidly on both E-selectin
and P-selectin at physiologic levels of linear shear stress. The DREG
56 Ab partially blocked rolling on the E-selectin substrate, whereas no
effect was seen on P-selectin. Thus, CD62L on human cultured T
lymphoblasts is one of several glycoproteins that interacts directly
with E-selectin and contributes to rolling under
flow. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
P-selectin glycoprotein ligand 1 (PSGL-1)3 is a ligand for both P-selectin and E-selectin (2, 4, 5, 6, 7). PSGL-1 purified from cultured human (8) and murine (9) T cells reacts with the HECA-452 Ab in Western blots. The HECA-452 Ab recognizes carbohydrate structures with E-selectin-binding activity and partially blocks E-selectin-mediated T cell adhesion in vitro (10, 11). In mice, PSGL-1-deficient T cells develop relatively low levels of E-selectin-binding activity during short-term culture in vitro and show reduced levels of E-selectin-mediated trafficking in vivo (12). On the other hand, PSGL-1 is not required for the development of E-selectin-binding activity on many human leukocytic cell lines (13). The PSGL-1-specific mAb KPL-1 completely blocks the interaction of human leukocytes with P-selectin, but does not affect their rolling on Eselectin (13). In addition, circulating leukocytes in PSGL-1-deficient animals are devoid of P-selectin-binding activity but still interact with E-selectin in vitro and in vivo (14). Finally, the porcine E-selectin/Ig chimera P11.4 precipitates several glycoproteins from detergent lysates of HECA-452-positive human T cells other than PSGL-1, suggesting that multiple binding sites may support rolling activity in vivo (15).
Several lines of evidence suggest that L-selectin (CD62L) serves as a ligand for E-selectin on neutrophils. CD62L on human neutrophils is recognized by the HECA-452 mAb (16) and recombinant E-selectin/Ig chimeras precipitate neutrophil CD62L from detergent lysates (15, 17). Anti-CD62L mAbs partially block neutrophil adhesion to E-selectin (15, 16, 17, 18), though blockade of leukocyte-on-leukocyte rolling may partially account for this effect (18, 19). In contrast, HECA-452 does not recognize CD62L on most circulating human lymphocytes (16) and E-selectin/Ig chimeras do not precipitate appreciable amounts of CD62L from detergent lysates of resting human, bovine, or mouse lymphocytes (9, 15). Thus, CD62L carries E-selectin binding sites on neutrophils, but not, it appears, on most circulating lymphocytes.
Structural analysis of E-selectin ligands on circulating human T cells is difficult because only a small fraction of the cells in most individuals express functional binding sites (20). In addition, effector T cells are rapidly cleared from the circulation thus comprise a small percentage of the circulating pool in healthy individuals. Consequently, the existing studies of circulating PBL may not detect ligands expressed transiently by recently activated T lymphoblasts. Previous studies showed that phorbol ester treatment of the Jurkat T cell line and anti-CD3/IL-2-driven culture of normal human T cells induced synthesis of E-selectin ligands (21, 22). Therefore, the current study investigated the structure and function of E-selectin binding sites induced by phorbol ester treatment of the Jurkat T cell line and anti-CD3/IL-2-driven proliferation of human PBL. In both cases, T cell activation induced synthesis of E-selectin binding sites on CD62L that supported rolling activity under flow conditions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human blood from healthy adult donors was collected into citrate anticoagulant and the mononuclear cell fraction was isolated following centrifugation through Histopaque 1077 (Sigma-Aldrich, St. Louis, MO). The cells were washed in HBSS and prepared for flow cytometric analysis, biotin-labeling procedures, or in vitro culture. For the latter, 1 x 107 cells were resuspended at 2 x 106 cells/ml in plates with immobilized anti-CD3 for 24 h, collected, and then cultured in XVIVO 15 medium containing 1.5 ng/ml IL-2, as described previously (21, 22). Cells were collected at various intervals after initiation of the cultures for analysis. After 6 days, some cultures were restimulated by another round of CD3 cross-linking and cultured with IL-2 for an additional 3 days. The human Jurkat T cell line was cultured and stimulated with phorbol ester as previously described (21). In some experiments, T cells were treated with neuraminidase (0.1 U/ml in low pH buffer (from Vibrio cholerae; Calbiochem, La Jolla, CA)) for 30 min before the flow cytometric or immunoprecipitation analyses described below. Control cells were treated with the low pH buffer alone.
Reagents and flow cytometric analysis
The following mAbs were used: DREG 56 and DREG 200 (mouse anti-human CD62L mAbs (23)), PL2 (mouse anti-PSGL-1; Serotec, Oxford, U.K.), and HECA-452 (rat IgM anti-human cutaneous lymphocyte-associated Ag (11)). The P11.4 pig E-selectin/human Ig chimera was used as described in our earlier analysis of human T cells (15). Flow cytometric analysis was done as previously described (15). Briefly, 1 x 106 cells were labeled with DREG 56 or HECA-452 followed by appropriate rat or mouse Ig-specific second-stage reagents (Jackson ImmunoResearch Laboratories, West Grove, PA). Negative controls included second-stage reagent alone or in combination with an irrelevant isotype-matched negative control. In some experiments, cells were stained with the P11.4 E-selectin chimera followed by PE-labeled anti-human Ig second stage (Jackson ImmunoResearch Laboratories). The specificity of the chimera binding was confirmed by blocking with EDTA, as previously described (15). For two-color stains, Jurkat T cells were labeled with P11.4 and HECA-452 in the first step and PE-anti-human Ig and FITC-anti-rat Ig second stages were then added. Appropriate controls were done to ensure that the reactivity of the second-stage Abs was specific. The effect of DREG 56 on the binding of P11.4 to Jurkat T cells was determined by first pretreating the cells with 50 µg/ml DREG 56 for 30 min on ice. The cells were then stained with P11.4 followed by the anti-human Ig second-stage reagent. All analyses were done on a BD FACSCalibur (BD Biosciences, Mountain View, CA).
Surface biotin labeling
For biotin labeling of surface Ags, 1 x 108 PMA-treated Jurkat cells or cultured human T cells were washed three times with sterile PBS and resuspended in 1 ml of sulfo-N-hydroxysuccinimide-LC-biotin/PBS (550 µg/ml; Pierce, Rockford, IL). After a 30-min incubation at room temperature, cells were washed three times with PBS and lysed for 2–3 h on ice with Nonidet P-40 lysis buffer (2% Nonidet P-40, 100 mM NaCl, 1 mM MgCl2, 1 mM CaCl2, 5 mM NaN3, 10 mM HEPES, and the following protease inhibitors: pepstatin A, 1,10-phenanthroline, PMSF, benzamidine, antipain, leupeptin, and chymostatin (Sigma-Aldrich)). To remove all nonspecific reactive components in the lysates, rabbit serum (5%) was added, the lysates were incubated for 1 h at room temperature, and then extensively precleared with protein G (Boehringer Mannheim, Indianapolis, IN) overnight at 4°C with constant rotation. The protein G beads were subsequently removed and the lysates were either used immediately or frozen at -80°C. Precleared biotin lysates were then incubated with 40–50 µg/ml E-selectin chimera or 100 µg/ml mAb for 1 h at room temperature. The chimera-ligand complexes were then precipitated with protein G at 4°C overnight under constant rotation. The beads were washed three times with wash buffer and mixed with nonreducing loading buffer and loaded onto 8% SDS-polyacrylamide gels. Gels were electrophoresed under nonreducing conditions and transferred to polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). After transfer, membranes were washed briefly with 1x TBST (10 mM Tris (pH 8.0), 150 mM NaCl, and 0.05% Tween 20) and then blocked with 7% BSA/TBST for 2 h at room temperature. Membranes were then washed with several changes of 1x TBST for 1.75 h. Streptavidin-HRP conjugate (Amersham Life Science, Buckinghamshire, U.K.) was diluted 1/5000 in sterile PBS and incubated with the membrane for 40 min and then washed for 1.75 h with several changes of 1x TBST. ECL detection reagents were used according to the manufacturer’s recommendations (Amersham Life Science) and incubated with the membranes for 1 min. Membranes were then covered with cellophane and exposed to X-OMAT film (Kodak, Rochester, New York) for 1–10 min, and the film was developed.
Precipitation and Western blot analysis
Detergent lysates prepared from unlabeled cells were precleared with rabbit serum and protein G, precipitated with the P11.4 chimera, and then run on an 8% nonreducing SDS-PAGE gel. For the Western blot procedures, the resolved glycoproteins were transferred to nitrocellulose and probed with DREG 56, DREG 200, or negative controls using a miniblotter apparatus (Immunetics, Cambridge, MA). The blots were developed as previously described (15).
In vitro rolling assay
In vitro rolling assays were done as previously described
(24). Briefly, for most experiments, human E-selectin
cDNA-transfected mouse L cells were grown to confluency on the internal
surface of small 1-cm long capillary tubes (1.5-mm diameter). For
others, recombinant human E- and P-selectin/IgG chimeras (prepared from
constructs provided by B. Seed, Massachusetts General Hospital, Boston,
MA), used as described in a previous report (22), were
immobilized on the internal surface of the capillary tubes. For the
latter, anti-human Ig (Sigma-Aldrich) was used to capture the E-
and P-selectin/IgG chimeras from culture supernatant fluids. The
capillary tubes were integrated into a closed loop system (120 cm)
in which 3 ml of fluid (DMEM plus 5% FBS) was recirculated via a
peristaltic pump, and mounted on the stage of an inverted microscope
(Nikon, Melville, NY) set up for video microscopy (Sony charge-coupled
device camera; Sony, Tokyo, Japan).
Lymphocytes (2 x 106/ml) were injected into
the closed system and their interaction with the E-selectin
transfectants or selectin/Ig chimeras analyzed under a flow rate that
produced shear of 2 dynes/cm2. Interactions were
monitored at the beginning of the capillary tube. Once a rolling
interaction was established, isotype-matched negative control (GD3.8)
and the anti-CD62L mAbs were injected sequentially at 50 µg/ml
concentrations. The number of cells rolling on the immobilized
E-selectin was measured at 1-min intervals for 3 min following
injection of each Ab. Since the maximal number of rolling cells varied
with each cell population (maximum ranged from 20 to 200), the data
are reported as the percentage of the maximum number of rolling cells
in a given assay. At the cell concentration used, rolling cells were in
contact with the immobilized E-selectin substrate but not with each
other. Therefore, the loss of cells from the capillary surface
reflected, primarily, inhibition of lymphocyte-E-selectin-adhesive
interactions as reported previously (24).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Jurkat T cell line was stimulated with 5 nM PMA for 3 days and
its capacity to bind P11.4 examined by flow cytometry. As shown in Fig. 1
, the P11.4 E-selectin/Ig chimera bound
40% of the activated Jurkat cells. Two-color analysis showed
linearly correlated binding of P11.4 and FITC-labeled HECA-452: P11.4
dull cells were low for HECA-452; whereas, P11.4 bright cells were
HECA-452 bright. The binding of P11.4 to human T cells was blocked by
both EDTA and anti-porcine E-selectin mAb (data not shown; .15). In
addition, neuraminidase treatment of the activated Jurkat T
cells reduced the percentage of P11.4-positive cells from 89 to 4%,
whereas the low pH buffer control had little effect (data not shown),
confirming that P11.4 attachment was sialic acid dependent.
Pretreatment of the Jurkat cells with an anti-CD62L mAb (DREG 56)
reduced the percentage of P11.4-positive cells by 60% (Fig. 2). In contrast, isotype-matched negative
control mAb had no effect in this assay.
|
|
The P11.4 chimera precipitated two major biotinylated molecules
from the detergent lysate of phorbol-treated, surface-biotinylated
Jurkat cells. These species migrated as 70- to 80- and 120- to 130-kDa
glycoproteins under nonreducing conditions on SDS-PAGE (Fig. 3). A minor band migrating at 180–220
kDa was detected in some precipitates as well (Fig. 3A,
lane 1). DREG 56 also precipitated a biotinylated species
migrating at 70–80 kDa from the detergent lysate (Fig. 3A,
lane 2). Preclearing experiments using DREG 56 confirmed the
identity of the 70- to 80-kDa glycoprotein precipitated by P11.4 as
CD62L (Fig. 3B). A human P-selectin/Ig chimera showed no
reactivity for CD62L (data not shown), indicating that this interaction
was specific to E-selectin. Thus, CD62L is one of two major
glycoproteins that react with the P11.4 chimera in extracts of
surface-biotinylated, phorbol ester-treated Jurkat cells.
|
Our previous study found that P11.4 precipitated three major
structures from the surface of biotinylated human PBL that migrated at
120–130, 220–230, and 260 kDa (15). In the current
study, PBL were activated on plate-immobilized anti-CD3 and
cultured in IL-2 (two cycles) to up-regulate selectin ligand synthesis
before biochemical analysis (21, 22). P11.4 precipitated
four major structures from cultured human PBL that migrated at 70–80,
120–130, 220–230, and 260 kDa under nonreducing conditions (Fig. 4A, lane 2). The
DREG 56 mAb precipitated material that comigrated with the 70- to
80-kDa band in the P11.4 precipitate (Fig. 4A, lane
1). In addition, DREG 56 reacted with the 70- to 80-kDa band in
Western blots prepared from the P11.4 precipitate (Fig. 4B,
lanes 2 and 3). Neuraminidase treatment reduced
L-selectin size by 5–10 kDa and eliminated all reactivity with the
P11.4 chimera in the precipitation assay (data not shown). Finally, the
anti-PSGL-1-specific mAb PL-2 (Fig. 4C) and a
P-selectin/Ig chimera (data not shown) both precipitated a single band
that comigrated with the 220- to 230-kDa band in the P11.4 precipitate.
Thus, P11.4 reacts with CD62L, PSGL-1, and two as yet unidentified
structures on cultured human PBL.
|
10–20% HECA-452-positive T cells were
cultured for either one (3 days) or two cycles (9 days) using
immobilized anti-CD3 and IL-2. DREG 56 and the P11.4 precipitations
were performed on detergent extracts prepared from equal numbers of
biotinylated fresh and cultured cells. Extracts of both the fresh (0
days) and one-cycle (3 day) blasts contained relatively small amounts
of the 70- to 80-kDa P11.4-reactive material, despite large overall
recoveries of CD62L with the DREG 56 mAb (Fig. 5A). In contrast, P11.4 and
DREG 56 precipitated nearly similar quantities of CD62L from the
two-cycle (9 days) T cell blasts. If DREG 56 precipitated nearly all
soluble CD62L in the extracts, then 60–70% of the available CD62L
from two-cycle blasts was also precipitated by P11.4 (Fig. 5B). This compared with 20% in either the fresh or
one-cycle blasts.
|
The relationship between time in culture and the development of
CD62L-mediated rolling on human E-selectin cDNA transfected L cells was
investigated. The greatest number of rolling cells was seen in the
9-day culture, whereas time zero had the fewest (data not shown).
Anti-CD62L had no effect on the rolling of fresh (day 0) and one cycle
(day 3) cells. In contrast, anti-CD62L dissociate 40% of the T
lymphoblasts cultured for 9 days (Fig. 6). The rolling speeds of lymphocytes
before and after the injection of anti-CD62L mAbs were highly
variable and showed no statistically significant differences (data not
shown). Subsequent injection of an anti-E-selectin mAb blocked the
interaction completely (data not shown). The 9-day cultured T
lymphoblasts rolled on human P- and E-selectin chimeras as well (Fig. 7). The DREG 56 mAb reversed 40% of
the rolling on E-selectin but did not inhibit interactions with
P-selectin. Therefore, CD62L on 9-day cultured T lymphoblasts
contributes to rolling on E-selectin, but not P-selectin under flow
conditions.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The P11.4 chimera did not appreciably interact with CD62L on freshly isolated human PBL; however, the prevalence of the E-selectin binding variant of CD62L increased with time in culture following anti-CD3/IL-2 stimulation. Approximately 40% of the cultured T lymphoblasts that rolled on immobilized human E-selectin dissociated from the substrate and the formation of new attachments was blocked by anti-CD62L mAb. Importantly, the anti-CD62L mAb did not inhibit rolling on P-selectin. This finding rules out anti-CD62L-mediated inhibition of leukocyte-leukocyte-adhesive interactions as a contributing factor in our flow system. In such interactions, the carbohydrate-binding domain of CD62L interacts with PSGL-1 on adherent leukocytes (25). If the impact of anti-CD62L on E-selectin rolling reflected inhibition of leukocyte-leukocyte interactions then one would expect the mAb to inhibit T lymphoblast rolling on immobilized P-selectin equally well. It does not; therefore, we conclude that the E-selectin-binding variant of CD62L identified in extracts of cultured T lymphoblasts supports rolling activity on immobilized E-selectin. This ligand is not, it appears, a major E-selectin binding site on circulating HECA-452-positive memory cells, but develops during T lymphoblast differentiation in vitro.
The activation conditions used in the current study share several important features. Phorbol ester stimulation of Jurkat T cells and anti-CD3/IL-2-driven T cell proliferation in the serum-free medium XVIVO-15 induce synthesis of the fucosyltransferase VII enzyme (FucTVII) (21, 22). This enzyme is required for construction of all selectin ligands (26) and its activity determines, in part, the density of E- and P-selectin ligands on T cells (22). Consequently, the current findings imply that under appropriate in vitro conditions human CD62L is one of several substrates for the FucTVII enzyme in human T cells.
Previous studies showed that phorbol ester treatment of both Jurkat T cells and human T cell cultures induces CD62L shedding initially, followed by increased CD62L synthesis and restoration of surface levels (27). Therefore, the E-selectin-binding variant of CD62L accumulates under activation conditions that promote the synthesis of both CD62L and FucTVII. Such conditions may occur during T cell differentiation in human lymph nodes and persist on the effector T cells that accumulate in E-selectin-mediated cutaneous immunologic lesions. Specifically, high levels of CD62L and the FucTVII-dependent epitope HECA-452 are expressed on a subset of actively differentiating T cells recovered from peripheral lymph nodes and on most T cells recruited into skin blisters (28, 29). Thus, effector cells that are rapidly cleared from the circulation into tissues may transiently synthesize the E-selectin-binding variant of CD62L. This hypothesis as well as how the in vitro activation parameters used in this study resemble those in vivo are currently under investigation. Nonetheless, the current findings indicate that CD62L is one of very few glycoproteins on human T cells that develop E-selectin-binding activity during T cell proliferation/differentiation in vitro and support rolling on E-selectin at physiologically meaningful levels of shear stress.
The biotinylation procedure used in the current study ensures that the receptors precipitated by the P11.4 chimera are expressed on the cell surface. Therefore, CD62L and PSGL-1 derived from the cell surface of cultured PBL are readily precipitated by the E-selectin chimera yet mAbs directed at these receptors do not completely inhibit tethering/rolling on immobilized E-selectin under flow conditions. As previously reported, the PSGL-1 mAb KPL-1 did not block T lymphoblast rolling on E-selectin (data not shown) and the anti-CD62L mAb DREG 56 showed only partial inhibition (40%). The adhesive domains of the selectins and their ligands are generally localized to specialized regions of the molecules. Consequently, not all mAbs that react with the molecules inhibit adhesive activity. Furthermore, the E-selectin-binding domains of both molecules may consist primarily of carbohydrate thus constitute poor targets for mAb production. In addition, multiple glycoconjugates may contribute to rolling activity under flow conditions so that blockade of individual ligands does not dissociate all cells. For example, the P11.4 precipitates from resting (15) and cultured PBL contain a major band migrating at 130 kDa in SDS-PAGE gels that is distinct from both CD62L and PSGL-1. The structure and physiologic relevance of this as yet unidentified E-selectin ligand is currently under study.
In summary, we show for the first time that lymphocyte CD62L is one of four surface molecules on activated T lymphoblasts that interact directly with E-selectin. Blockade of CD62L alone is sufficient to partially block the rolling of cultured PBL on immobilized E-selectin under conditions that minimize the impact of CD62L-mediated leukocyte-leukocyte rolling interactions. Thus, as previously shown for neutrophils (15, 16, 17), CD62L is both a carbohydrate-binding receptor for ligands on target cells and a scaffold for oligosaccharides that interact directly with E-selectin.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Mark A. Jutila, Veterinary Molecular Biology, Montana State University, Bozeman, MT 59717. E-mail address: uvsmj{at}montana.edu
3 Abbreviations used in this paper: PSGL-1, P-selectin glycoprotein ligand 1; FucTVII, fucosyltransferase VII.
Received for publication February 28, 2002. Accepted for publication June 12, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells. J. Immunol. 159:3574.[Abstract]
(1, 3)-fucosyltransferase VII-dependent synthesis of P- and E-selectin ligands on cultured T lymphoblasts. J. Immunol. 161:6305.(1,3)fucosyltransferase Fuc-TVII controls leukocyte trafficking through an essential role in L-, E-, and P-selectin ligand biosynthesis. Cell 86:643.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. E. Eriksson No detectable endothelial- or leukocyte-derived L-selectin ligand activity on the endothelium in inflamed cremaster muscle venules J. Leukoc. Biol., July 1, 2008; 84(1): 93 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Matsumoto, A. Shigeta, Y. Furukawa, T. Tanaka, M. Miyasaka, and T. Hirata CD43 Collaborates with P-Selectin Glycoprotein Ligand-1 to Mediate E-Selectin-Dependent T Cell Migration into Inflamed Skin J. Immunol., February 15, 2007; 178(4): 2499 - 2506. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Matsumoto, K. Atarashi, E. Umemoto, Y. Furukawa, A. Shigeta, M. Miyasaka, and T. Hirata CD43 Functions as a Ligand for E-Selectin on Activated T Cells J. Immunol., December 15, 2005; 175(12): 8042 - 8050. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Martinez, M. Joffraud, S. Giraud, B. Baisse, M. P. Bernimoulin, M. Schapira, and O. Spertini Regulation of PSGL-1 Interactions with L-selectin, P-selectin, and E-selectin: ROLE OF HUMAN FUCOSYLTRANSFERASE-IV AND -VII J. Biol. Chem., February 18, 2005; 280(7): 5378 - 5390. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. E. Green, D. N. Pearson, N. B. Christensen, and S. I. Simon Topographic requirements and dynamics of signaling via L-selectin on neutrophils Am J Physiol Cell Physiol, March 1, 2003; 284(3): C705 - C717. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
