| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Dermatology, Brigham and Women’s Hospital, Boston, MA 02115; and
Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, and Harvard Skin Disease Research Center, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In humans, memory T cells observed in cutaneous inflammatory lesions express cutaneous lymphocyte-associated Ag (CLA),3 a carbohydrate epitope defined by reactivity with the unique mAb HECA-452 (2, 3). Memory T cells residing in noncutaneous tissues, in contrast, lack the CLA epitope (4, 5). HECA-452 has been shown to bind a set of sialylated fucosylated carbohydrates, similar in structure to sialyl-Lewisx, and CLA-positive populations of cells have been shown to bind to E- and P-selectin, leading to speculation that CLA may serve as a receptor for vascular selectins and as a homing receptor targeting CLA-positive T cells to skin (6, 7, 8, 9, 10). We have shown previously that the CLA carbohydrate epitope expressed by memory T cells is found predominantly on P-selectin glycoprotein ligand-1 (PSGL-1), a mucin-like glycoprotein present on all leukocytes (11).
Although the PSGL-1 peptide backbone is expressed constitutively by all leukocytes, posttranslational modification is required to produce functional selectin binding sites (12). At least three distinct functional populations can be identified among circulating T cells, a CLA-negative population that cannot bind either E- or P-selectin, a CLA-negative population that can bind P-selectin but not E-selectin, and a CLA-positive population that can bind both E- and P-selectin (8, 9, 10). Although numerous reports have established a strong correlation between CLA expression and the ability of memory T cells to interact with E-selectin (3, 6, 9, 10, 13, 14, 15, 16, 17, 18, 19), there has been no direct evidence reported showing that human T cell CLA is, itself, a functional ligand capable of supporting rolling interactions with E-selectin in physiologic shear flow. Furthermore, investigators have reported the inability of mAb HECA-452 to block T cell binding to E-selectin (20), and have identified cell populations that bind E-selectin but lack the HECA-452 epitope (21, 22). These data support an alternate hypothesis that the E-selectin binding structure on human T cells may be coordinately expressed with the HECA-452 epitope, but as a distinct cell surface component (17, 18, 19, 20, 21, 22). Although it is attractive to consider mechanisms by which one surface glycoprotein (e.g., PSGL-1) might be modified to bear distinct selectin ligand domains, characterization of such model systems depends on direct demonstration of physiologic ligand function for both E- and P-selectin on the candidate glycoproteins.
Analysis of selectin ligand function is complicated by the dynamic nature of selectin binding, mediating tethering, and rolling of leukocytes on endothelium in shear flow. Wall shear stress plays an important role in the regulation and promotion of these interactions. Static binding assays that demonstrate firm adhesion to a substrate do not assess this dynamic function or address shear-related binding issues and, as such, may not reflect physiologic ligand activity. Methods for in vitro study of leukocyte-endothelial interactions in shear flow, such as Stamper-Woodruff and parallel plate flow chamber assays, have provided valuable insights into physiologic adhesive interactions (23, 24). However, use of these methods to study the role of individual cellular components is largely limited by the character and availability of purified substrate materials that can be affixed to glass or plastic. Although purification of specific cell membrane components can be attained via the use of specific immunoaffinity reagents or with sufficient prior knowledge regarding the structure of the molecule of interest, the applicability of these assay methods to identify or study the function of ligands of unknown structure is inherently limited. As mentioned above, the production of high affinity selectin binding domains on leukocyte glycoproteins also requires one or more posttranslational modifications (e.g., tyrosine sulfation, production of core-2 O-linked glycans, and sialylation and fucosylation of O-linked and N-linked glycans) that can vary between species, among cell populations, and across stages of cellular development (8, 12, 25). Thus, the use of synthetic or recombinant materials, or material from other cell lineages, may not reflect the true nature of the ligand of interest, further limiting the availability of purified ligand materials for study.
To address these issues, we developed a method for direct real-time observation of adhesive interactions between cells in shear flow and cellular components separated by SDS-PAGE and immobilized on blotting membranes using standard Western blot techniques. This method allows for the rapid and reproducible assessment of individual components, both known and unknown, within a complex mixture without the need for prior isolation or enrichment beyond standard SDS-PAGE. In this report, we use this method to survey lysates of CLA-positive and -negative human T cells for molecules able to serve as E- and P-selectin ligands. These studies show that CLA+ PSGL-1 is, unequivocally, a ligand for both E- and P-selectin in physiologic shear flow, indicate that the P-selectin ligand activity of T cell PSGL-1 is independent of modification with CLA, and provide preliminary evidence for an additional E-selectin ligand on CLA-positive T cells that is distinct from known ligands.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Anti-CLA mAb (HECA-452, rat IgM) was purchased from BD PharMingen (San Diego, CA). Anti-PSGL-1 (CD162) reagents included mAb PL-2 (MuIgG1, nonblocking; Beckman Coulter, Miami, FL) (26), used for Western blotting; mAb PL-1 (MuIgG1, blocks P-selectin binding; Beckman Coulter) (26), used for functional blocking studies; mAb PSL-275 (MuIgG1, blocks P-selectin binding; Genetics Institute, Cambridge, MA), used for FACS staining; and 4H10 (MuIgG1, blocks P-selectin binding; Genetics Institute), used for immunoprecipitation studies. Isotype-matched control Abs were purchased from Zymed Laboratories (South San Francisco, CA). FITC-conjugated anti-rat IgM and anti-mouse IgG were purchased from Southern Biotechnology Associates (Birmingham, AL). Alkaline phosphatase conjugated anti-rat IgM and anti-mouse IgG were from Zymed Laboratories. Anti-E-selectin mAb (CD62E, clone 68-5H11) and P-selectin (CD62 P, clone AK-4) were purchased from BD PharMingen.
Cells
PBMC were prepared by density gradient separation (Ficoll-Histopaque 1.077; Sigma-Aldrich, St. Louis, MO) of peripheral blood or of cells collected during platelet pheresis of normal donors. CLA-positive and -negative T cell preparations were generated essentially as previously described (11, 16). Normal human PBMC were cultured in XVIVO15 medium (BioWhittaker, Walkersville, MD) or RPMI 1640 medium (Mediatech, Herndon, VA) with 10% FBS (Sigma-Aldrich; RPMI-10% FBS). Both media were supplemented with penicillin (100 U/ml) and streptomycin (100 µg/ml; Life Technologies, Grand Island, NY), HEPES (5 mM (pH 7.4)) and glutamine (2 mM; Mediatech), and human rIL-2 (100 U/ml; PeproTech, Rocky Hill, NJ). Cells for this study were used between 10 and 14 days after stimulation with immobilized anti-CD3 as described (11). Expression of relevant surface Ags was confirmed by indirect immunofluorescence flow cytometry performed on a FACScan IV (BD Biosciences, Mountain View, CA) using CellQuest software (version 3.1). XVIVO-cultured T cells were typically 70–80% positive for CLA while RPMI-cultured T cells were typically 3–5% positive for CLA (data not shown). Staining for PSGL-1 revealed Ag present at approximately equal levels on each population (e.g., 100% of cells positive and similar mean fluorescence intensities; data not shown). Cultured cells showed undetectable levels of CD16 or CD19, indicating lack of contamination with monocytes or B cells. Chinese hamster ovary (CHO) cells transfected with human E-selectin (CHO-E) were provided by R. Lobb (Biogen, Cambridge, MA). CHO cells transfected with human P-selectin (CHO-P) were provided by Genetics Institute. Mock-transfected CHO cells (CHO-mock) were provided by R. Finberg (University of Massachusetts Medical Center, Worcester, MA) (27). All three lines were cloned by limiting dilution. CHO-E and CHO-P cells were maintained in MEM (Mediatech) supplemented with 10% FBS, 1x nonessential amino acids, penicillin-streptomycin, and glutamine. CHO-mock were maintained in Ham’s F-12 medium (Mediatech) supplemented with 5% FBS, penicillin-streptomycin, and glutamine. FACS analysis of CHO cells using mAb to CD62E and CD62P was performed regularly to confirm stable selectin expression at approximately equal levels (100% of cells positive and similar mean fluorescence intensities; data not shown).
Western blots
Cells of interest were collected by centrifugation, washed in protein free buffer (PBS or HBSS), and the pellet was resuspended at 4 x 108 cell equivalents/ml in lysis buffer consisting of 150 mM NaCl, 1 mM MgCl2, 50 mM Tris (pH 7.5), 1 mM PMSF, 10 µg/ml aprotinin, and 1% N-octylglucoside (Roche, Indianapolis, IN). Lysates were incubated 2 h on ice and centrifuged for 30 min at 10,000 x g. Supernatants were collected and stored at -20°C until use. Protein concentrations were determined using Bradford’s reagent (Bio-Rad, Hercules, CA). Samples representing 25–100 µg of protein were electrophoresed through 4–20% acrylamide gradient SDS-PAGE gels (Criterion; Bio-Rad) under standard reducing conditions (5% 2-ME sample buffer). Prestained molecular mass standards (Amersham Biosciences, Piscataway, NJ) were included in adjacent lanes in all experiments. SDS-PAGE separated proteins were blotted onto SequiBlot polyvinylidene difluoride (PVDF) membranes (Bio-Rad) in 100 mM Tris, 75 mM glycine, 9% methanol using standard transfer techniques. Blots were subsequently blocked with newborn calf serum (Sigma-Aldrich) for 1 h and probed with anti-CLA (HECA-452) or anti-PSGL-1 (PL-2) Abs, as indicated. Isotype control blots were performed in parallel for each condition and did not identify 240 or 140 kDa species in any samples tested. Bound Ab was visualized with species specific alkaline phosphatase coupled anti-Ig and Western Blue alkaline phosphatase substrate (Promega, Madison, WI). Following staining, the membranes were again blocked by incubation in newborn calf serum (Sigma-Aldrich) for at least 1 h at 4°C before use in flow studies.
Blot rolling assay
Before analysis, blots were equilibrated in
Ca2+-free and Mg2+-free
HBSS (Life Technologies) supplemented with HEPES (10 mM (pH 7.4); H/H)
or HEPES and CaCl2 (2 mM;
H/H/Ca2+). Where indicted, glycerol (American
BioAnalytical, Natick, MA) was added to a final concentration of 10%
(v/v) (H/H/Ca2+-10% glycerol). For rolling
studies, blots were placed in a Petri dish filled with
H/H/Ca2+-10% glycerol and a circular parallel
plate flow chamber apparatus (Cox Laboratory for Biomedical
Engineering, Rice University, Houston, TX) (internal flow chamber
dimensions 2 cm x 0.5 cm x 0.025 cm) was mounted over the
area of interest (Fig. 1
). This parallel
plate flow chamber is similar to that previously described for use with
substances coated on glass or plastic (24, 28). No special
conditions were required for attachment or use of this chamber on
blotting membranes other than the addition of glycerol to the buffers.
CHO cells were harvested by rinsing the culture flasks with H/H with 5
mM EDTA, washed, and resuspended at 1–2 x
107 cells/ml in H/H and maintained on ice.
Immediately before use, cells were diluted at least 1/10 in
binding medium (H/H/Ca2+-10% glycerol) at room
temperature, resulting in a final concentration of 1–2 x
106 cells/ml. Cells were drawn into the chamber
under defined flow conditions by a precision syringe pump (Harvard
Apparatus, Cambridge, MA) and observed in real-time for interaction
with the blot surface under stable shear force conditions. Tethering
was observed at low physiologic wall shear stresses (0.5–1.5
dyne/cm2) for short periods (one to several
minutes). Bound cells were then subjected to timed stepwise increases
in wall shear stress. Tethering was defined as reduction of forward
motion below the hydrodynamic velocity lasting a minimum of two video
frames (0.07 s), and rolling was defined as >5 cell diameters of
lateral translation below the hydrodynamic velocity. The majority of
tethered cells were observed to roll smoothly across the entire field
of view. Nonspecific interactions (i.e., cellular collisions with the
substrate that did not lead to tethering and/or rolling) were defined
as interactions lasting <0.07 s and were not included in the analysis.
Calcium-dependent binding typical of selectin-mediated interactions was
confirmed by perfusion of H/H with 5 mM EDTA which resulted in >95%
release of bound cells for all samples studied. Nonspecifically bound
cells (not rolling and/or not released by perfusion with EDTA) were
also discounted from analysis. The frequency of nonspecific events
observed was not different in experiments performed in medium with or
without 5 mM EDTA, suggesting that these were not selectin-mediated
events. There was a time-dependent increase in nonspecific (nonrolling)
attachments in the absence of flow, though very few cells formed firm
(nonrolling) attachments to the blot in continuous shear flow. All
experiments were observed in real time and videotaped for analysis.
Tethering rate was calculated as the number of cells that tethered per
field per time (usually 30–60 s) at a defined shear stress and
adjusted to per minute values. Wall shear stress (T) values
were calculated according to the formula T
(dynes/cm2) = 3µ
Q/2ba (2) where µ is the
coefficient of viscosity of the solution in the chamber (poise),
Q is the volumetric flow rate (cm3/s),
b is the channel width (0.5 cm), and a is the
half-channel height (0.0127 cm) (29). A value of 0.009
poise was used for the viscosity (µ) of water at 25°C
(30) and a value of 0.0123 poise was used for the
viscosity of 10% glycerol (v/v) in water at 25°C (31).
Flow rates used in this study correspond to shear stresses ranging from
0.5 to 3.5 dynes/cm2. Wall shear stress in 10%
glycerol at 25°C is 
1.37 fold greater than in water at the same
temperature.
|
250–40 kDa) were
observed for brief periods (2–10 s) and videotaped for subsequent
analysis. Initiation of viewing at low vs high molecular mass
ends of the chamber did not reveal significantly different results.
Observation of other molecular mass ranges was achieved by moving the
chamber and repeating the study. Multiple observations of unmanipulated
cell lysates did not reveal significant attachments below 40 kDa,
including the gel dye front, or above 240 kDa apparent molecular mass
(data not shown). Ab inhibition studies
For anti-PSGL-1 inhibition studies, identical blots of 50 µg CLA+ T cell lysates stained with anti-PSGL-1 (nonfunction-blocking mAb PL-2) were prepared and assembled into the flow apparatus. Each blot was assessed for normal binding function with both CHO-E and CHO-P cells using the method stated above. The chamber was then filled with H/H/Ca2+-10% glycerol medium containing mAb (anti-PSGL-1 function-blocking mAb (PL-1) or mouse IgG1 control mAb) at 50 µg/ml in H/H/Ca2+-10% glycerol and perfused with this material at 0.01 ml/min for 1 h at room temperature. The chamber was flushed with H/H/Ca2+-10% glycerol and cell binding with both CHO-E and CHO-P was repeated. All observations for blocking studies were made at the midpoint of the 140 kDa band identified by anti-PSGL-1 mAb.
Immunoprecipitations
For immunoprecipitation studies, aliquots of lysate from CLA-bearing T cells were incubated with 3 µg anti-PSGL-1 mAb 4H10, or isotype control mAb, for at least 1 h at 4°C. Reaction mixtures were transferred to fresh tubes containing 20 µl washed recombinant protein G-agarose beads that had been blocked with BSA (binding capacity 18 mg IgG/ml; Invitrogen, Carlsbad, CA) and incubated with agitation for at least 1 h at 4°C. Agarose bead immune complexes were collected by centrifugation and washed repeatedly with 1% N-octylglucoside lysis buffer. For some studies, the primary supernatant was subjected to serial immunoprecipitation by adding additional Ab and repeating the process as above. Each sample was brought to 1x SDS-sample buffer and incubated for 5 min in a boiling water bath before SDS-PAGE and Western blotting under standard conditions as outlined above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Lysates of CLA-positive and
CLA-negative T cells were subjected to SDS-PAGE under standard reducing
conditions and then blotted onto a PVDF membrane and immunostained
using standard methods. Under reducing conditions, CLA/PSGL-1 stained
with anti-CLA (HECA-452) mAb typically appears as a single band
with a relative molecular mass of 140 kDa (PSGL-1 monomer) (Fig. 2A). Differences in relative
molecular mass attributable to differential glycosylation resulting in
decoration with the CLA epitope appear to be beneath the resolution
limit of this separation technique. Immunostained blots were placed in
buffer containing 10% glycerol for use in flow analysis. Equilibration
in dilute glycerol alters the opacity of the membrane sufficiently to
allow transmission of light and the direct visualization of cells
interacting with the surface of the blot by standard light microscopy.
A parallel plate laminar flow chamber, used routinely for observing
real-time interactions under controlled shear conditions of cells with
substrates immobilized on glass or plastic, was mounted directly on the
blot via a low-pressure vacuum seal and the entire apparatus was placed
on the stage of an inverted microscope (see Fig. 1). Cells were
introduced into the chamber through an inlet port under controlled
aspiration conditions and monitored for interaction with the blot
surface by direct observation. All experiments were videotaped for
offline analysis. For this study, CHO-mock, CHO-E, or CHO-P cells were
introduced into the chamber under a variety of wall shear stress
conditions and observed for interaction with the blot surface. Under
continuous flow conditions equivalent to physiologic shear in
postcapillary venules, both CHO-E and CHO-P cells were observed to
attach primarily to a single discrete band, corresponding to the 140
kDa band of monomer CLA/PSGL-1, on blots of CLA-positive T cell lysates
(Fig. 2B). Cells could be observed, in real-time, tethering
to the upstream edge of the stained area, rolling across the band, and
releasing from the downstream edge (a digital movie showing this
activity is included as supplemental
material).4 Although
this assay is nonphysiologic by design, using vascular selectins on
nonadherent cells and immobilizing the leukocyte glycoproteins, the
rolling behavior of bound cells was qualitatively similar to that
observed in traditional binding assays using leukocyte binding to glass
or plastic immobilized selectins. Maintaining cells in continuous shear
flow substantially reduces or eliminates the background binding of CHO
cells to the blotting membrane seen in static assays (data not shown).
Scanning of the entire blot area visible in the laminar flow chamber
showed smaller, but reproducible, numbers of CHO-E cells tethering and
rolling on an area of the blot at 95 kDa and a few cells between 140
and 90 kDa (Fig. 2B). This lower molecular mass band did not
stain with anti-CLA or anti-PSGL-1 mAb (Figs. 2B and
4A). Observations made after relocating the chamber over
higher and lower molecular masses did not reveal other bands supporting
selectin-mediated attachment and rolling. Blots of CLA-negative T cell
lysates, in contrast, supported binding of CHO-P cells at levels
equivalent to those seen on the CLA-positive blot, but did not support
significant binding of CHO-E cells (Fig. 2C). CHO-mock cells
did not form rolling adhesions on either blot (data not shown).
|
, A and C), but markedly different for CHO-E cells
(Fig. 3, A and B), indicating that the P-selectin
ligand activity of the two preparations was similar while effective
E-selectin ligand was present only in the CLA-positive T cell lysate.
Blots prepared from gels loaded with decreasing amounts of CLA-positive
cell lysate showed proportionally reduced binding of both CHO-E and
CHO-P cells and did not identify a minimum threshold for supporting
selectin-mediated attachment on blots (Fig. 3D). Bound CHO-E
(Fig. 3E) and CHO-P (Fig. 3F) cells showed
similar shear resistance profiles, with substantial rolling activity
noted on blots loaded with as little as 25 µg of total cell protein
from CLA-positive T cells. Duplicate blots of CLA-positive T cell
lysate stained for CLA (mAb HECA-452) or PSGL-1 (nonblocking mAb PL-2)
showed staining of a single band each at 140 kDa (Fig. 4A) and supported equivalent
levels of CHO-E and CHO-P cell tethering and rolling activity on the
respective labeled bands (Fig. 4B). Control CHO cells
transfected with vector only (CHO-mock; Figs. 3, A and
D, and 4B) infused under identical conditions, as
well as CHO-E and CHO-P cells infused in medium with 5 mM EDTA (data
not shown), did not form rolling interactions on any areas of the blots
examined.
|
|
C).
Observations were performed on identical blots before and after
blocking with control mAb (IgG1) or mAb PL-1 (P-selectin blocking
anti-PSGL-1 mAb). Treatment of CLA-positive T cell blots with
excess anti-PSGL-1 Ab (mAb PL-1) resulted in complete inhibition of
CHO-P cell binding and had no effect on CHO-E cell binding relative to
control IgG-treated blots. Control IgG1 had no effect on CHO-E or CHO-P
cell binding relative to untreated blots (data not shown). A similar
experiment using HECA-452 showed only partial inhibition of CHO-E
binding relative to control IgM Ab, with no effect on CHO-P binding
(data not shown). The relative inability of HECA-452 to block
E-selectin mediated adhesion to CLA-positive cells has been reported by
several other investigators. Knibbs et al. (20) showed
that HECA-452 had no effect on the ability of T lymphoblasts to roll on
E-selectin in a conventional flow chamber assay, and Wagers et al.
(21, 22) have described cell lines that lack detectable
staining with HECA-452 yet bind avidly to E-selectin. Preincubation of
CHO-E and CHO-P cells with function-blocking anti-E- or P-selectin
mAb, respectively, resulted in complete abrogation of tethering and
rolling on both CLA-positive and -negative substrates, confirming that
the activity observed was mediated by selectins (data not shown).
Although both P- and E-selectin binding activity were coincident on
these Western blots, it is possible that two distinct proteins that
comigrate in SDS-PAGE under these conditions could serve as separate
ligands for E- and P-selectin. To address this possibility, PSGL-1 was
immunoprecipitated from lysates of CLA-positive T cells, subjected to
SDS-PAGE, and transferred to PVDF membranes. Staining with Abs to
either CLA (HECA-452) or PSGL-1 (PL-2), revealed a single band at
140 kDa molecular mass (Fig. 5A; data not shown).
Assessment of selectin ligand function confirmed that CLA-positive
PSGL-1 immunopurified from human T cells has both functional E- and
P-selectin ligand activity (Fig. 5B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). This is in agreement
with a report by Knibbs et al. (20) showing that
incubation with both mAb HECA-452 and mAb CSLEX1
(anti-sLex) did not significantly inhibit the
binding of CLA-positive T lymphoblasts to E-selectin in a conventional
flow chamber assay. Although the current study does not identify the
epitope recognized by HECA-452 mAb as the E-selectin binding site, it
establishes a direct correlation between expression of CLA on PSGL-1
and the acquisition of functional E-selectin ligand activity by PSGL-1
that was not possible with prior methods. Moreover, these data suggest
that the E-selectin binding domain on CLA+ PSGL-1
is distinct from both the P-selectin binding domain and the HECA-452
epitope.
The findings in this report represent an attempt at global assessment
of individual T cell glycoproteins that function as E- and P-selectin
ligands in physiologic shear flow. Although this survey is limited to
those ligands able to function after reducing SDS-PAGE, both the E- and
P-selectin binding functions of CLA+ PSGL-1 were
preserved and a potential novel E-selectin ligand was identified. We
observed E-selectin ligand activity on a structure migrating at 95
kDa apparent molecular mass in lysates of human CLA-positive T cells
separated by SDS-PAGE under reducing conditions. This area did not
stain with HECA-452, anti-PSGL-1 (PL-1 or PL-2), or anti-CD62L
(L-selectin) mAb and supported E-, but not P-, selectin-mediated
binding. Although this may be a processing or degradation product of
PSGL-1, the lack of reactivity with two mAbs directed to separate
domains, membrane-proximal in the case of PL-2 and membrane-distal in
the case of PL-1, suggests that this is a distinct structure. The
presence of an E-selectin ligand activity on human T cells separate
from PSGL-1 and CLA has not been evident in prior studies using
conventional leukocyte rolling assays. The structure of this ligand and
its role in T cell homing is currently under investigation.
This report also describes and demonstrates the utility of a relatively simple method for the rapid and reproducible characterization of both known and unknown cell adhesion molecules within a complex mixture without the need for purification or enrichment before immobilization. This method allows for direct real-time observation of interaction parameters (e.g. rolling vs firm attachment, specificity, and reversibility with inhibitors) in both physiologic and nonphysiologic shear conditions, thus permitting a unique user interface for the observation of adhesive events on membrane-immobilized materials. Ligands under investigation can be immobilized directly or segregated by gel electrophoresis (e.g., SDS-PAGE, isoelectric focusing) or other methods before transfer to the membrane, providing structural information that is not evident from previously available methods. This technique also allows for real-time manipulation of interaction conditions including wall shear stress, ion requirements, temperature, metabolic inhibitors, and the presence of activating agents or inhibitors of cell function. As shown in this study, blot-immobilized substrates can be used repeatedly, allowing direct comparison of different cell populations in shear flow or in situ manipulation of the substrate under continuous direct visualization. The capacity to observe sequential experimental and control conditions on a single substrate, and to observe physiologic behaviors and responses to manipulations in real-time, provides distinct advantages over conventional static and flow-based binding assays. As additional proof of principle, we have used this method to characterize, purify, and confirm the activity of CD44 as an E- and L-selectin ligand expressed on hemopoietic progenitor cells (32, 33).
As protein components may lose function as a result of denaturation in SDS-PAGE, the method, as described in this report, is more suited to the study of carbohydrate components mediating binding to selectins or other cell surface lectins. In this study, however, CHO-E and -P cells showed comparable levels of binding to denatured and blot immobilized CLA/PSGL-1. E-selectin has been shown previously to bind carbohydrate alone, but P-selectin has been shown to require both a carbohydrate component and a protein component incorporating one or more sulfotyrosine residues near the N terminus. This suggests that either P-selectin binding does not require native conformation of the PSGL-1 protein backbone or that sufficient renaturation occurs after transfer to the blotting membrane to support selectin binding. Modification of this method to maintain function of more sensitive protein components (e.g., nondenaturing PAGE, direct adsorption to membrane) may be desirable to study other adhesion events (e.g., integrin binding to cell adhesion molecules or extracellular matrix proteins).
In summary, we report the direct visualization of both E- and P-selectin-mediated binding and support of rolling in shear flow to the single human T cell surface glycoprotein, CLA/PSGL-1, using a unique assay of cell binding to ligands immobilized on Western blots. These data definitively identify PSGL-1 as an E-selectin ligand on CLA-positive human T cells and provide preliminary evidence for a second HECA-452-nonreactive E-selectin ligand. The blot rolling method described will be particularly useful in the pursuit of novel adhesion molecules, in that it allows direct real-time observation of interaction parameters (e.g., rolling vs firm attachments, specificity, and reversibility with inhibitors) on materials separated by standard electrophoretic methods (e.g., SDS-PAGE), but without requirements for further purification.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert Sackstein, Harvard Institutes of Medicine, 77 Avenue Louis Pasteur, Boston, MA 02115. E-mail address: rsackstein{at}rics.bwh.harvard.edu
3 Abbreviations used in this paper: CLA, cutaneous lymphocyte-associated Ag; PSGL-1, P-selectin glycoprotein ligand-1; CHO, Chinese hamster ovary; CHO-E, CHO cell transfected with human E-selectin; CHO-P, CHO cell transfected with human P-selectin; CHO-mock, mock-transfected CHO cell; H/H, HBSS supplemented with HEPES; H/H/Ca2+, H/H and CaCl2; PVDF, polyvinylidene difluoride.
4 The on-line version of this article contains supplemental material.
Received for publication January 9, 2002. Accepted for publication March 25, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/
and 
/
T lymphocyte subsets in shear flow with E- and P-selectin. J. Exp. Med. 183:1193.-stimulated endothelial cells. Immunology 81:359.[Medline]
(1,3)-fucosyltransferase VII-dependent synthesis of P- and E-selectin ligands on cultured T lymphoblasts. J. Immunol. 161:6305.2 subunit of human VLA-2. J. Virol. 67:6847.This article has been cited by other articles:
|
|
 |
|
  T. Bianchi, L. B. Pincus, M.-A. Wurbel, B. E. Rich, T. S. Kupper, R. C. Fuhlbrigge, and M. Boes Maintenance of Peripheral Tolerance through Controlled Tissue Homing of Antigen-Specific T Cells in K14-mOVA Mice J. Immunol., April 15, 2009; 182(8): 4665 - 4674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Thomas, R. L. Schnaar, and K. Konstantopoulos Podocalyxin-like protein is an E-/L-selectin ligand on colon carcinoma cells: comparative biochemical properties of selectin ligands in host and tumor cells Am J Physiol Cell Physiol, March 1, 2009; 296(3): C505 - C513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. A. Resto, M. M. Burdick, N. M. Dagia, S. D. McCammon, S. M. Fennewald, and R. Sackstein L-selectin-mediated Lymphocyte-Cancer Cell Interactions under Low Fluid Shear Conditions J. Biol. Chem., June 6, 2008; 283(23): 15816 - 15824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Nonomura, J. Kikuchi, N. Kiyokawa, H. Ozaki, K. Mitsunaga, H. Ando, A. Kanamori, R. Kannagi, J. Fujimoto, K. Muroi, et al. CD43, but not P-Selectin Glycoprotein Ligand-1, Functions as an E-Selectin Counter-Receptor in Human Pre-B-Cell Leukemia NALL-1 Cancer Res., February 1, 2008; 68(3): 790 - 799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Napier, Z. R. Healy, R. L. Schnaar, and K. Konstantopoulos Selectin Ligand Expression Regulates the Initial Vascular Interactions of Colon Carcinoma Cells: THE ROLES OF CD44V AND ALTERNATIVE SIALOFUCOSYLATED SELECTIN LIGANDS J. Biol. Chem., February 9, 2007; 282(6): 3433 - 3441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.L. Seneviratne, A.P. Black, L. Jones, A.S. Bailey, and G.S. Ogg The role of skin-homing T cells in extrinsic atopic dermatitis QJM, January 1, 2007; 100(1): 19 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Abbal, M. Lambelet, D. Bertaggia, C. Gerbex, M. Martinez, A. Arcaro, M. Schapira, and O. Spertini Lipid raft adhesion receptors and Syk regulate selectin-dependent rolling under flow conditions Blood, November 15, 2006; 108(10): 3352 - 3359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Carlow and H. J. Ziltener CD43 Deficiency Has No Impact in Competitive In Vivo Assays of Neutrophil or Activated T Cell Recruitment Efficiency J. Immunol., November 1, 2006; 177(9): 6450 - 6459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hirahara, L. Liu, R. A. Clark, K.-i. Yamanaka, R. C. Fuhlbrigge, and T. S. Kupper The Majority of Human Peripheral Blood CD4+CD25highFoxp3+ Regulatory T Cells Bear Functional Skin-Homing Receptors J. Immunol., October 1, 2006; 177(7): 4488 - 4494. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Ni, J. J. Campbell, G. Niehans, and B. Walcheck The Monoclonal Antibody CHO-131 Identifies a Subset of Cutaneous Lymphocyte-Associated Antigen T Cells Enriched in P-Selectin-Binding Cells J. Immunol., October 1, 2006; 177(7): 4742 - 4748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Burdick, J. T. Chu, S. Godar, and R. Sackstein HCELL Is the Major E- and L-selectin Ligand Expressed on LS174T Colon Carcinoma Cells J. Biol. Chem., May 19, 2006; 281(20): 13899 - 13905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Fuhlbrigge, S. L. King, R. Sackstein, and T. S. Kupper CD43 is a ligand for E-selectin on CLA+ human T cells Blood, February 15, 2006; 107(4): 1421 - 1426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zou, V. R. Shinde Patil, N. M. Dagia, L. A. Smith, M. J. Wargo, K. A. Interliggi, C. M. Lloyd, D. F. J. Tees, B. Walcheck, M. B. Lawrence, et al. PSGL-1 derived from human neutrophils is a high-efficiency ligand for endothelium-expressed E-selectin under flow Am J Physiol Cell Physiol, August 1, 2005; 289(2): C415 - C424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Dimitroff, L. Descheny, N. Trujillo, R. Kim, V. Nguyen, W. Huang, K. J. Pienta, J. L. Kutok, and M. A. Rubin Identification of Leukocyte E-Selectin Ligands, P-Selectin Glycoprotein Ligand-1 and E-Selectin Ligand-1, on Human Metastatic Prostate Tumor Cells Cancer Res., July 1, 2005; 65(13): 5750 - 5760. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. D. Hanley, M. M. Burdick, K. Konstantopoulos, and R. Sackstein CD44 on LS174T Colon Carcinoma Cells Possesses E-Selectin Ligand Activity Cancer Res., July 1, 2005; 65(13): 5812 - 5817. [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. J. Dimitroff, M. Lechpammer, D. Long-Woodward, and J. L. Kutok Rolling of Human Bone-Metastatic Prostate Tumor Cells on Human Bone Marrow Endothelium under Shear Flow Is Mediated by E-Selectin Cancer Res., August 1, 2004; 64(15): 5261 - 5269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. F. Chong, J.-E. Murphy, T. S. Kupper, and R. C. Fuhlbrigge E-Selectin, Thymus- and Activation-Regulated Chemokine/CCL17, and Intercellular Adhesion Molecule-1 Are Constitutively Coexpressed in Dermal Microvessels: A Foundation for a Cutaneous Immunosurveillance System J. Immunol., February 1, 2004; 172(3): 1575 - 1581. [Abstract] [Full Text]
|
|
|
|
|
|
R. M. Rao, D. O. Haskard, and R. C. Landis Enhanced Recruitment of Th2 and CLA-Negative Lymphocytes by the S128R Polymorphism of E-Selectin J. Immunol., November 15, 2002; 169(10): 5860 - 5865. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). CHO-P, but not CHO-E, cells
bound in a similar fashion to the corresponding area on a blot from
CLA-negative cells (
). CHO-mock cells did not form rolling
attachments on either blot. B and C,
Cells remaining bound to CLA-positive or -negative substrate blots
during timed stepwise increases in wall shear stress. CHO-E cells bound
to CLA-positive, but not to CLA-negative, substrates and were able to
remain attached and roll throughout the shear range tested
(B). CHO-P cells, in contrast, showed nearly identical
resistance to detachment by increasing shear on CLA-positive as on
CLA-negative substrate (C). Western blots of SDS-PAGE
loaded with decreasing amounts of CLA-positive T cell total cell lysate
protein were also assessed for E- and P-selectin ligand activity.
D, The number of cells observed rolling on the 140 kDa
band per visual field (10x objective) at 1.06 dyne/cm2
wall shear stress. The number of CHO-E (
).
CHO-mock cells were not observed to form rolling attachments on any
areas of the blot observed. The field of view for each assessment was
located at the darkest area of the band on each blot. Results represent
the mean and range of three independent experiments. C,
CHO-E, -P, and -mock cells were observed for formation of rolling
attachments on the 140 kDa band on blots of CLA-positive T cell lysate
stained with mAb PL-2 (nonfunction-blocking) and treated with control
IgG1 (