BLOOD The Journal of - Semantic Scholar · Investigator of the American Hearr Association. Address reprint requests to Geoffrey S. Kansas, PhD, ... of VCAM-1 and ICAM-1, two other

BLOOD VOL 88, NO 9

The Journal of The Amm'can Society of Hematology

NOVEMBER 1, 1996

REVIEW ARTICLE

Selectins and Their Ligands: Current Concepts and Controversies

By Geoffrey S. Kansas

ITH THE ESSENTIALLY simultaneous cloning in 1989 of the selectins (then called ELAM-1, GMP-

140/PADGEM, and LECAM- 1ILAM- 1 ; now called E-selectin, P-selectin, and L-selectin or CD62E, CD62P, and CD62L, respectively), and the recognition of a new family of cell adhesion molecules critically involved in the regulation of leukocyte traffic, a new era dawned in the study of leukocyte-endothelia1 recognition. Rapid progress has since been made in understanding the molecular and cellular basis of selectin-mediated adhesion, the identification of cellular ligands for selectins, the biophysics of selectins, and the physiologic role played by selectins in both normal leukocyte traffic and in a range of inflammatory and other diseases. This review will highlight this progress, and define some important questions for the future.

EXPRESSION OF SELECTINS

L-selectin. L-selectin is expressed on essentially all blood neutrophils and monocytes, on the majority of blood borne T and B cells, and on a subset of natural killer (NK) cells.'-4 On T and B cells, L-selectin is expressed on essentially all virgidnaive cells in blood or tissues, but is absent from at least some memory cells2 CD4' memory helper T cells derived from the spleen in mice lack L-selectin,'" whereas a subset of human memory CD4+ cells express L- selectin at levels equivalent to those found on naive ~ e l l s . ~ * ~ ~ ~ This difference may be caused by differential retention of L-selectin after activation of T cells within distinct microen- vironments.'.' L-selectin is also expressed on immature hematopoetic cells, including the majority of myeloid col- ony-forming cells.10." In contrast, L-selectin is expressed relatively late during B-cell development, well after Ig gene rearrangement, and just before the mature, virgin, immuno- competent B cells migrate out of the bone marrow (BM)."

Immunohistologic examination of frozen sections of secondary lymphoid organs with monoclonal antibodies (MoAb) to L-selectin demonstrates a characteristic and unique appearance of an essentially complete absence of staining of germinal center cells with strong staining of man- tle zone and para~ortex.'.'.'~ This observation suggested that L-selectin might be lost with cellular activation, a hypothesis which was subsequently confirmed by several investiga- t o r ~ . ' ~ - ' ~ L-selectin is rapidly lost from the surface of normal leukocytes in response to a variety of stimuli, including cyto-

kines and phorbol esters, and a slightly smaller form of L-selectin can be detected in the supernatant of activated leukocytes concomitant with its disappearance from the sur-

Soluble L-selectin can be detected in the serum of normal healthy individuals at surprisingly high concentra- tions,I8 and may increase or decrease in certain disease states. The mechanism of shedding is not completely understood, but is clearly the result of proteolytic cleavage at a site just outside the plasma membrane. The cleavage site has been identified as between residues K283 and S284 (of the mature protein) in a reasonably well conserved sequence predicted to lie just outside the memb~ane.'~ Despite this amino acid homology in this region of L-selectin, mutational analysis failed to identify specific residues in this region required for shedding.20,2' Rather, the total number of residues and length of the sequence surrounding the cleavage site appeared to regulate shedding. However, no protease responsible for cleaving L-selectin has been identified to date, and the signal transduction pathways which lead to shedding have not been defined. The physiologic significance of shedding remains unclear, but may be a mechanism of downregulating adhesion following firm attachment to the endothelium.

E-selectin expression is limited to endothelium, and principally to endothelium in response to inflammatory stimuli such as interleukin-1 (IL-l), tumor necrosis factor-a (TNF-a), or bacterial lipopolysaccharide (LPS)." Cell-surface E-selectin expression is induced at the level of transcription, and inhibitors of either transcription (eg, Actinomycin D) or translation (eg, cycloheximide) inhibit E-selectin expression.22 Induction of E-selectin expression and transcription can also be inhibited by transforming growth factor$ (TGF-/3).2' On endothelium cultured in vitro

E-selectin.

From the Department of Microbiology-Immunology, Northwest-

Submitted May 3, 1996; accepted June 24, 1996. Supported in part by grants from the Earl M . Bane Charitable

Trust and Northwestern Medical School. G.S. K. is an Established Investigator of the American Hearr Association.

Address reprint requests to Geoffrey S. Kansas, PhD, Department of Microbiology-Immunology, Northwesiern Medical School, 303 E Chicago Ave, Chicago, IL 60611.

ern Medical School, Chicago, IL.

0 1996 by The American Society of Hematology. 0006-4971/96/8809-OO44$3.00/0

Blood, Vol 88, No 9 (November I), 1996: pp 3259-3287 3259

For personal use only.on October 27, 2017. by guest www.bloodjournal.orgFrom

http://www.bloodjournal.org/

http://www.bloodjournal.org/site/subscriptions/ToS.xhtml

3260 GEOFFREY S. KANSAS

such as human umbilical vein endothelial cells (HUVEC), E-selectin expression peaks between 3 and 6 hours after stimulation with TNF-a, and decreases thereafter, even in the continued presence of the cytokine, to basal levels within 10 to 12 hours. Endothelium cultured from other tissues can exhibit more prolonged expre~s ion .~~ This decline in expression of E-selectin in the continued presence of the stimulus is in contrast to the pattern of inducible expression of VCAM-1 and ICAM-1, two other inducible endothelial cell adhesion molecules whose expression remains elevated for up to 72 hours on TNF-a-stimulated HUVEC.25,26 How- ever, in vivo, E-selectin may be chronically expressed at sites of local inflammation, particularly in the skin during delayed hypersensitivity reaction^.^'-*^ This difference in the pattern of expression of E-selectin between endothelium cultured in vitro and dermal vessels in vivo may be caused by differences in the type and stability of the E-selectin mRNA expressed in the activated endothelium in these different setting^.^'

Cytokine-inducible E-selectin gene transcription requires the activation and nuclear translocation of NF-KB, which is involved in the induction of many genes involved in immune and inflammatory response^.^'-^^ Endothelial activation and nuclear translocation of NF-KB is preceded by degradation of the cytoplasmic inhibitor of NF-KB, I K B ~ , ~ ~ - ~ ~ a process that requires the proteosome pathway.35 Three NF-KB sites have been defined in the E-selectin promoter, each of which are required for strong induction of transcription.”,” In addition, maximal cytokine responsiveness requires an ATF-2 site, and multiple high mobility group protein HMGI(Y) site^.^^.^* The coordinate action of each of these transcription factors at the level of the proximal promoter is required for strong induction of E-selectin gene transcription.

The loss of E-selectin from the surface of activated endothelial cells is likely caused by a combination of factors. First, E-selectin gene transcription is sharply downregulated within 6 to 9 hours after induction,22 and the E-selectin mRNA has a short half-life.39 In addition, E-selectin is rapidly internalized and degraded in l y s o s ~ m e s . ~ ~ - ~ ~ The combination of these processes collectively ensures that the expression of E-selectin at the surface of cytokine-stimulated endothelium is transient. However, different species of mRNA observed in vivo have much longer half lives, leading to persistent expression in vessels in certain tissues, particularly skin.30

P-selectin expression on both endothelium and platelets is also inducible. However, P-selectin is stored preformed in the a-granules and Weibel-Palade bodies of platelets and e n d ~ t h e l i u m , ~ ~ , ~ ~ secretory storage granules which contain a variety of substances that are released or expressed at the membrane after activation of the cells. P- selectin is rapidly expressed at the cell surface as a result of fusion of these granules with the plasma This fusion event is caused by agonists such as thrombin, histamine, activators of protein kinase C, complement fragments46 or (in platelets) adenine diphosphate (ADP), and is quite rapid, with P-selectin and other constitutents of the granules (eg, von Willebrand factor47) being detectable at the cell

P-selectin.

surface within minutes. The sorting of P-selectin into these granules is controlled by sequences within the cytoplasmic tail of P-selectin which interact with the sorting machinery in cells that have the regulated sorting ~athway.~’ Transfec- tion of P-selectin into cells that do not have the regulated sorting pathway results in constitutive expression at the cell ~ u r f a c e . ~ * ~ ~ ~ P-selectin expressed at the cell surface is rapidly internalized? which accounts for its transient appearance at the surface of activated endothelium and platelets, and this activity also maps to the cytoplasmic tail.” Internalized P- selectin molecules are targeted to lysosomes for degradation, a distinct pathway regulated by distinct amino acids within the cytoplasmic tail.” Some P-selectin may be recycled back to storage granulesJ2 or shed,” consistent with an alternatively spliced transcript lacking the transmembrane do-

Thus, the cytoplasmic tail of P-selectin contains main.5?,53

multiple motifs, still incompletely defined, which cumulatively regulate the sorting to intracellular granules, internal- ization from the plasma membrane, and targeting to lysosomes. This dynamic pattern of expression may be important for regulating the expression of P-selectin at the cell surface.

The cytoplasmic tail of P-selectin is phosphorylated in a complex pattern after platelet and endothelial cell activa-

The majority of phosphorylation detectable is on tion.’?.’5

serine, and mutagenesis studies identify serine 788 as the predominant site. However, phosphorylation of threonine and tyrosine can also be detected. The tyrosine phosphorylation may be mediated by pp6O-src, a tyrosine kinase found in platelet granules.’6 Interestingly, P-selectin is also rapidly and transiently phosphorylated on histidine.” The rapid kinetics of phosphorylation suggest that phosphorylation occurs while the P-selectin is still within the storage granules, suggesting that phosphorylation may be involved in induction of cell surface expression. In addition, P-selectin is acylated at the single cysteine in the cytoplasmic tail.‘* Despite these observations, the functional significance of these modifications in the regulation of P-selectin expression or function remains unclear.

In addition to the rapid expression of P-selectin at the cell surface as a result of fusion of secretory granules with the plasma membrane, P-selectin expression on endothelium is also regulated t ranscr ip t i~na l ly .~~~~~ Analysis of P-selectin mRNA expression in vivo and in vitro shows that, like E- selectin, P-selectin gene transcription is transiently induced by LPS, TNF-a, or IL- I , leading to P-selectin expression 2 to 4 hours later. Therefore, P-selectin is likely to be expressed both early and late in the course of an inflammatory response.

The overlapping but temporally and spatially distinct pattern of expressions of selectins and their ligands allows for precise regulation of the initial phase of leukocyte recognition of endo the l i~m.~~~‘~ Distinct classes of leukocytes can use different combinations of selectins, leading to differences in their ability to interact with endothelium. Thus, essentially all blood neutrophils and monocytes express L- selectin as well as ligands for E- and P-selectin (and possibly L-selectin; see below), and can thus use all three. In contrast, phenotypically and functionally distinct subsets of T cells




SELECTINS AND THEIR LIGANDS 3261

exist which express only one or two of these molecules (see below). Furthermore, E- or P-selectin may not be expressed in vessels of all organs in all settings of inflammation or tissue damage. Differences in the ability of distinct leukocyte subpopulations to use different selectins are likely to determine in part the differences in the ability of these leukocyte populations to enter different tissues in response to inflammatory or immune stimuli.

FUNCTIONS OF SELECTINS IN LEUKOCYTE TRAFFIC

It has long been recognized that the earliest cellular response to inflammatory stimuli or tissue damage is the rolling of leukocytes along the vessel wall near the site of insult.64 This rolling behavior, which is an essential prerequisite for arrest on the lumenal surface of the endothelium and subsequent transmigration into tissues, is mediated principally by selectins. All three selectins have been shown to mediate rolling both in vivo and in vitro, and each selectin can directly and independently mediate r~lling.~’-~’ However, due to the different kinetics of expression of the selectins and/or their ligands, different selectins come into play at different times during the course of an inflammatory response. In addition, the expression of the selectins and/or their ligands is not identical for all tissues or all species. Thus, blocking studies performed with cells which can use all three selectins (ie, neutrophils and monocytes; see below) can give different results as a function of the endothelial stimulus, tissue site in vivo, species, or the method of selectin blockade. Another layer of complexity is that the utilization or activity of selectins varies with cell type (see below). In addition, apart from their role in leukocyte recruitment in inflammation, the selectins have other, more specialized functions in the regulation of leukocyte traffic.

An essential aspect of an effective immune response is the constant recirculation of normal mature T and B lymphocytes throughout the secondary lymphoid organs, including the lymph nodes and organized lymphoid tissue in the gut such as Peyer’s patches. This process ensures that the full complement of antigen receptor specificities is exposed to the full range of antigens encountered by the organism. Entry of lymphocytes from the blood into secondary lymphoid organs occurs across specialized endothelial cells in the postcapillary venules of lymph nodes. These specialized endothelial cells exhibit cuboidal morphology, unlike their flat-walled counterparts outside lymphoid organs, and are hence referred to as “high” endothelial cells, or high endothelial venules (HEV). An in vitro assay that effectively models the interaction between lymphocytes and HEV is the Stamper-Wood- ruff frozen section assay.76 This elegant assay, in which lymphocytes or other cells are overlaid onto frozen sections of lymph nodes (or other tissues) under conditions in which binding is preferentially to HEV present in the frozen section, has been instrumental in the characterization of the function of L-selectin and other HEV receptors.

Before the identification of selectins as a family of related adhesion molecules which mediate leukocyte rolling, the role of L-selectin (then known as the MEL-14 antigen in the

Role of L-selectin in lymphocyte recirculation.

mouse3) in the tissue specific migration of T and B lymphocytes to peripheral lymph nodes was well documented. Thus, MoAb to L-selectin completely blocks binding to lymph node HEV in the frozen section assay, but does not block binding to HEV from other lymphoid organ^.^.^^.'^ In vivo administration of MoAb to L-selectin or preincubation of lymphocytes with L-selectin MoAb completely blocks migration of lymphocytes to lymph nodes, but in most studies does not significantly block migration to Peyer’s patches or ~ p l e e n . ~ . ~ ~ Murine T-cell clones, unlike their human counter- part~,~’ do not generally express L-selectin, and home poorly to lymph nodes in vivo.” Transfection of nonbinding cells with cDNA encoding L-selectin confers binding to lymph node HEV.” Expression of L-selectin is necessary and (usu- ally) sufficient for binding to peripheral lymph node HEV in the frozen section assay. L-selectin is absolutely required for entry of lymphocytes into lymph nodes from the blood,83 and mice genetically deficient in L-selectin (“knockout” mice) have very small, hypocellular lymph nodes.84 This requirement for L-selectin for entry into lymph nodes is consistent with the expression of L-selectin on essentially all normal naive T and B cells (above). Taken together, these observations demonstrate convincingly that L-selectin is the principal if not sole HEV receptor for lymphocyte traffic to lymph nodes, and the principal determinant of this tissue specificity.

The central importance of L-selectin in binding to lymph node HEV is clear from the observation that MoAb to other adhesion molecules such as LFA- 1 have variable and much less significant effects on lymphocyte binding to HEV in the frozen section a s ~ a y ~ ’ ~ ~ ~ ; in addition, activation of cells with phorbol esters, which induces rapid shedding of L-selectin (above) but upregulation of LFA- 1 a ~ t i v i t y , ~ ~ - ~ ~ abrogates binding to HEV in the frozen section assay and homing in vivo. In addition, expression of LFA-1 is not necessary for binding to HEV.82,90 However, it is important to emphasize that the frozen section assay measures only binding to HEV, and that other, tissue-nonspecific adhesion molecules clearly involved in later phases of entry into tissues, including LFA- 1, are often not detected in this assay. In addition, normal neutrophils, which express L-selectin but do not normally traffic through quiescent, uninflamed lymph nodes, bind to HEV in the frozen section assay as well as lymphocytes! Therefore, binding to HEV in the frozen section assay does not necessarily predict whether a given cell type will migrate to lymph nodes in vivo, but does measure whether that cell type is capable of performing the first, critical step.

Although L-selectin MoAb do not block binding of lymphocytes to Peyer’s patch HEV in the frozen section assay, more recent studies indicate that L-selectin, while not crucial, may participate in migration to Peyer’s patch in vivo. Lymphocytes from mice genetically deficient in L-selectin (“knockout” mice) show a small and transient defect in homing to Peyer’s In some studies, L-selectin MoAb inhibit in vivo homing to Peyer’s L-selectin may play a role in lymphocyte rolling before firm adhesion to Peyer’s patch HEV?3 The basis for the discrepancies between these studies and earlier work remains unclear.





Methods for studying dynamic leukocyte-endothelial interactions. Leukocyte rolling in vivo can be directly evaluated and quantitated by intravital micro~copy. ’~-~~ Animals, usu- ally rodents or rabbits, are anesthetized, and the tissue of interest, often the mesentery or cremaster (which have readily accessible venules) is laid across the stage of a spe- cially prepared microscope, and is maintained under approximately physiologic conditions for the entire course of the experiment. “Spontaneous” rolling of the animal’s own leukocytes, which is actually induced by the surgical trauma associated with the experimental setup, can then be observed and recorded on videotape for off-line analysis. In addition, the animal can be systemically treated with proinflammatory cytokines such as TNF-a or IL-1. Alternatively, cells of interest can be introduced upstream of the point of observation, and their behavior observed and recorded. Similarly, various reagents, eg, MoAb, can be introduced upstream and their effect on leukocyte rolling determined. Intravital microscopy makes possible the cell-by-cell quantitation of the rolling behavior of cells in an essentially natural setting, and has therefore been crucial to the characterization of selectins in vivo.

Leukocyte rolling can also be studied in vitro, by means of a parallel plate flow ~ h a m b e r . ~ ~ , ~ ’ , ’ ~ The cells of choice are introduced into the chamber and allowed to pass over the substrate present on the lower “floor” of the chamber under conditions of defined shear force. Although not identical to the in vivo situation with respect to the geometry of the interactions, the maximum shear force which permits binding, the presence of erythrocytes, and several other hemodynamic parameters, flow chambers have the advantage that the substrate on which the cells roll can be easily manipulated before the assay, and can consist of cells, eg, endothelium or stably transfected cell lines, or plastic or glass to which purified adhesion molecules of interest can be attached. In addition, as with intravital microscopy, the rolling cells can be manipulated externally before the experiment. This system was instrumental in showing the dominant role of selectins in mediating rolling, and the inability of leukocyte integrins (j32 integrins) to do SO.“

Role of L-selectin in leukocyte rolling and recruitment in inflammation. In addition to its role in normal lymphocyte recirculation, multiple lines of evidence clearly show that L- selectin directly mediates rolling of leukocytes on endothelium at sites of inflammation in vivo. MoAb to L-selectin or recombinant soluble L-selectin significantly (-80%) blocks the spontaneous rolling of introduced human or endogenous rodent neutrophils in exteriorized mesenteric venules in vi^^.^^^"^'^^'' MoAb that block binding of lymphocytes to lymph node HEV in the frozen ~ e c t i o n ~ ~ . ~ ‘ assay also block rolling of neutrophils in Transfection of lymphoid cells, which cannot bind E- or P-selectin, with cDNA encoding L-selectin confers the ability to roll in mesenteric venules, and this rolling is completely blocked by the same MoAb which block lymphocyte binding to lymph node HEV.82.98.I(JlJ L-selectin knockout mice exhibit sharply impaired “spontaneous” rolling in postcapillary venules in vivo.84 Monocytes and neutrophils both roll on HUVEC

stimulated with proinflammatory cytokines, and this rolling is significantly inhibited by preincubation of the leukocytes with blocking MoAb to L - ~ e l e c t i n . ~ ~ . ~ ‘ . ’ ~ Collectively, these studies firmly establish that L-selectin directly mediates rolling on endothelium in sites of inflammation or tissue injury in vivo.

The availability of both a given selectin plus its ligand does not always mean that it will be used. As indicated above, normal human peripheral blood T cells bind well to lymph node HEV in the frozen-section assay, and therefore express at their surface functional L-selectin. However, rolling of T cells on cytokine stimulated HUVEC in vitro is not affected by MoAb to L-sele~tin,~’.’~ in contrast to rolling of both normal neutrophils and monocyte^.^".^'." Similarly, rolling of monocytes is not significantly affected by MoAb to E-selectin, despite the ability of these cells to roll on E- selectin when expressed on stably transfected cells“” and the presence of E-selectin on TNF-a-stimulated HUVEC.“” The absence of L-selectin-dependent T-cell rolling does not appear to be due to the lack of a functional L-selectin ligand on cytokine-stimulated HUVEC,”’ as monocyte and neutrophil rolling is inhibited by L-selectin MOA^.^".".'^ Similarly, the lack of monocyte rolling on E-selectin expressed on TNF-a-stimulated HUVEC is not due to a lack of a functional E-selectin ligand on monocytes.“” The lack of effect of L-selectin MoAb on T-cell rolling on TNF-a stimulated HUVEC may be due in part to the lack of a functional L- selectin ligand on most T lymphocytes (see below). Alterna- tively, because the ability to roll on TNF-a-stimulated HU- VEC is confined almost exclusively to the memory subset, the ability of these (mostly L-selectin-negative) T cells to use E- and P-selectin may make a contribution of L-selectin difficult to detect. Interestingly, T-cell rolling on cytokine stimulated HUVEC is not completely blocked by MoAb to any of the known selectins or to the integrin VLA-4 (which can mediate lymphocyte rolling under certain circum- s t a n c e ~ ~ ’ , ’ ~ ~ ) , even when used in ~ o m b i n a t i o n , ~ ’ . ~ ~ suggesting the possibility of additional “rolling receptors” on T cells. These observations are consistent with the idea that cell-type specific differences in the use of selectins (and other adhesion molecules involved in leukocyte traffic) may influence the pattern of migration of different leukocyte types.“

L-selectin acts both early and late in an inflammatory response, consistent with its constitutive expression on essentially all blood leukocytes. Early, “spontaneous” rolling in exteriorized mesenteric venules, which is induced by surgical trauma (ie, without the addition of exogenous cytokine or other stimulus) and is likely related to stimulation of tissue mast cells,“’5 is evident quite rapidly, and continues for at least 2 hoUrS,hh.Yh.Y”.”..I)h Experiments with blocking MoAb, cell lines that use only L-selectin or only E- and P-selectin, or mice which are genetically deficient in L-, P-, or E-selectin, clearly show that the very earliest phase (<20 minutes) of this rolling response is mediated principally by P-selectin. with only a minor component contributed by L-selectin,6h.9X. I0h The dependence on P-selectin of this very rapid rolling response is consistent with the rapid expression of P-selectin from intracellular storage granules (above). Subse-





quently (>20 minutes), the contribution of P-selectin to leukocyte rolling declines, and rolling is mediated predomi- nantly by L-selectin. No contribution of E-selectin is detected until at least 2 hours postsurgery, consistent with the kinetics of induction of E-selectin (above). Thus, L- selectin and P-selectin function in a separable but temporally overlapping fashion to mediate the earliest response to an inflammatory stimulus. Similarly, in animals treated systemically with IL-I or TNF-a, MoAb to L-selectin significantly blocks neutrophil rollingkw; this effect is particularly pro- nounced in P-selectin knockout mice.66 Taken together, these observations show an important role for L-selectin during most phases of an inflammatory response.

The importance of L-selectin in the recruitment of leukocytes to inflammatory sites in vivo has been investigated in several specific models of inflammation and leukocyte- mediated disease. Neutrophil recruitment into the inflamed peritoneal cavity of mice is impaired by intravenous administration of MoAb to L - s e l e ~ t i n , ~ ~ , ' ~ ~ by soluble recombinant L-selectin,'08 or in L-selectin knockout mice.84 Later mononuclear cell recruitment into the peritoneum is also impaired by in vivo administration of MoAb to L-selectinIw or in L-selectin knockout mice. Neutrophil recruitment into the lung"' or skin4 is blocked by MoAb to L-selectin. L-selectin knockout mice exhibit impaired inflammatory responses in several models of chronic inflammation, including delayed type hypersensitivity (DTH) i.eactions and LPS-induced toxic shock."' MoAb against L-selectin inhibit ischemia/ reperfusion injury in several animal models.'''-' l5 Therefore, L-selectin is important for migration of multiple classes of leukocytes into a variety of tissues during acute and chronic inflammation.

Finally, recent evidence has suggested that L-selectin may play a role in dynamic interactions between leukocytes, particularly neutrophils, and that these leukocyte-leukocyte interactions may be important in the amplification of an inflammatory response. Neutrophils can roll in vitro on neutrophils bound either to endothelium or a plastic substrate, and this rolling is blocked by MoAb to L-selectin."' Ligands for L-selectin can be detected on some hematopoietic cell lines."7 Furthermore, nrmtrophil aggregation in sus- pension can be inhibited by MoAb to L-selectin or by carbohydrates that inhibit L-selectin activity."8-'20 Because aggregation in these assays could also be inhibited by MoAb to p2 integrins, these investigators concluded that neutrophil aggregation was mediated by a direct interaction between L-selectin and ,B2 integrins. However, the potential role of the known ligands for the p2 integrins, ICAM-1, -2, and - 3, was not evaluated in these s1udies."8-'20 In addition, the requirement for both L-selectin and ,B2 integrins in neutrophil aggregation strikingly parallels the requirements for both of these molecules in neutrophil binding to endothelium under and conforms well to multistep models of leukocyte/endothelial recognition.62 Therefore, it seems likely that neutrophil aggregation is initiated by L-selectin recognition of its still unidentified leukocyte ligand (also see below), followed by cellular activation and engagement of 02 integrins. The role of L-selectin in leukocyte interactions

with other leukocytes may contribute significantly to the broad importance of L-selectin in acute and chronic inflammation.

As indicated above, P-selectin mediates the very earliest leukocyte rolling during an inflammatory response. Spontaneous rolling in several animal models is blocked by MoAb to P - s e l e ~ t i n , 6 ~ , ~ ~ ~ and spontaneous rolling is initially absent in P-selectin knockout mice.Iz2 A significant proportion of this "spontaneous" rolling is probably induced by mediators released from tissue mast cells, including histamine, and histamine-induced rolling of leukocytes is mediated by P - ~ e l e c t i n . ~ ~ ~ " ~ ~ " ~ Consistent with an important role in early inflammation, recruitment of neutrophils into the inflamed peritoneum of mice is delayed -2 to 4 hours in P-selectin knockout mice, but ultimately reaches near normal levels.122 This pattern of delayed recruitment of leukocytes into the peritoneal cavity is nearly identical to that seen in L-selectin knockout mice,84 consistent with a role for both these selectins in the early response (above), and recruitment into the peritoneal cavity can be completely blocked by a combination of L-selectin and P-selectin MoAb.Io7 These observations reinforce the dual importance of L-selectin and P-selectin in leukocyte recruitment during acute inflammation.

In addition to a crucial role in the earliest cellular response to inflammation, P-selectin is also involved in several types of chronic inflammation, consistent with its transcriptional upregulation by LPS and proinflammatory cytokines. Mono- cyte entry into the inflamed peritoneum at later time points is impaired in P-selectin knockout mice,Iz6 as are T-cell- dependent contact hypersensitivity response~."~ P-selectin is expressed by rheumatoid synovial endothelium, and supports the binding of monocytes to these vessels.IZ8 P-selectin is involved in some models of inflammatory neutrophil recruitment into the l~ng , ' '~ . '~ ' but not others.13' Like L-selectin, P-selectin can therefore also contribute to chronic inflammation, but may be selectively involved in accumulation of leukocytes in certain tissues in response to different inflammatory stimuli.

The unique (among selectins) expression of P-selectin on activated platelets suggests that P-selectin plays an important role in leukocyte-platelet interactions during wound healing and hemostasis. P-selectin mediates adhesion of activated platelets to monocytes, neutrophils, NK cells, and memory/ activated T ~ e l l s , ' ~ ' - ' ~ ~ and this interaction serves to amplify the recruitment of both leukocytes, especially monocytes and neutrophils, and platelets to sites of vascular injury. Platelet binding to monocytes induces the expression of tissue factor, thereby initiating the blood coagulation cascade, and this interaction is mediated directly by P-se1e~tin.l~~ Fibrin deposition within a developing thrombus requires leukocytes, and this recruitment of leukocytes and subsequent fibrin deposition is blocked by MoAb to P-selectin on activated platelets within the P-selectin may therefore constitute an important molecular interface between the inflammatory, thrombotic, and wound-healing systems. Despite these observations, no obvious defects in normal hemostasis were initially observed in P-selectin-deficient mice,"' although

P-selectin.




3264

JJ

30 -

25-

20 -

15 -

10 -

5 -

0

GEOFFREY S. KANSAS

wildtype L-selectin KO P-selectin KO

- -A*- mild defects were detected in subsequent studies.'3x The exact role played by P-selectin in thrombosis and hemostasis remains to be elucidated.

Similar to the other selectins, E-selectin also supports rolling of leukocytes at sites of inflammation and tissue injury."" Purified E-selectin and E-selectin transfectants support rolling in However, because of the requirement for de novo gene transcription for expression, E-selectin plays essentially no role in the earliest phases (<2 hours) of leukocyte recruitment during acute inflammation. In addition, because of concordant transcription of P-selectin in many tissues, at least in the mouse, E- and P-selectin are coexpressed in many inflamed tissues. Consequently, inhibition of activity of either selectin often results in little or no effect on leukocyte recruitment. Thus, in animals which have been systemically pretreated with inflammatory cytokines such as TNF-a or IL-I, contributions from E- selectin can be easily detected in some settings,'"' but not in others." In E-selectin knockout mice, no defect of leukocyte recruitment to the inflamed peritoneum is evident unless P- selectin function is simultaneously blocked with MoAb.I4" Similarly, no effect on DTH reactions induced by oxazo- lone are detectable unless both endothelia1 selectins are bl~cked. '~" Despite this apparent redundancy (in mice). leukocyte recruitment clearly involves E-selectin in several models of inflammatory disease, including airway inflammation and obstruction in monkeysl'l and lung injury in rats,"" and E-selectin is expressed in numerous sites of chronic inflammation. including the rheumatoid syno- vium,l?x.l'2 sites of allograft rejection,'J'"'" and DTH reac- ti ons .27.29

E-selectin appears to serve an important role as a tissue- specific homing receptor for leukocyte recruitment specifically to the skin, particularly for memory T cells. Leukocyte recruitment into human skin grafted onto severe combined immunodeficient (SCID) mice is dependent on E-selec-

T-cell infiltration into sites of cutaneous DTH is inhibited by MoAb to E-selectin.'") E-selectin expression is associated with T-cell infiltration in a variety of dermatologic disorders, including malignancy, and multiple sites of chronic inflammation in the ~kin. '~.~' ' . l~~) T cells found in skin lesions strongly express a unique marker, the cutaneous lymphocyte antigen (CLA), a carbohydrate (or family of related carbohydrates) defined by the MoAb HECA-452'") (see below), which is expressed by less than 5% of T cells in blood or other tissues. The CLA' subset of memory T cells is highly enriched in T cells which bind E-selectin." Chronic expression of E-selectin is frequently seen in the skin.'" Taken together, these observations strongly support the hypothesis that E-selectin serves as a tissue-specific "homing receptor" for leukocyte recruitement to the skin.

E-selectin.

t in . 147.148

COOPERATIVITY BETWEEN SELECTINS IN LEUKOCYTE RECRUITMENT

An important aspect of selectin function is the cooperativity or synergy between selectins, especially on populations of leukocytes, eg, neutrophils. which can potentially use more than one selectin simultaneously.'" A clear example

Fig 1. Spontaneous leukocyte rolling flux in normal, L-selectin knockout (KO), and P-selectin KO mice. Experiment performed and analyzed as described in the report by Ley et aI.= Note that the flux values of wild-type mice are at nearly all timepoints higher than the sum of the flux values from the individual KO mice, indicative of cooperativity between L- and P-selectin. E-selectin is not involved in rolling under these conditions. See text for details. (Reproduced from The Journal of Experimental Medicine, 1995, vol 181, pp 669-675, by copyright permission of The Rockefeller University Press.66)

of this is the analysis of spontaneous rolling flux in normal mice or mice deficient in either L- or P-selectin" (Fig I ) . The rolling flux values observed in wild-type mice, where L- and P-selectin are both operating, are at virtually all timepoints higher than the sum of the flux values obtained in each knockout mouse, where only a single selectin is operative. Interaction of neutrophils with endothelium, via simultaneous use of two selectins, is therefore significantly more efficient than interaction via a single selectin. In part, this is related to differences in the characteristic rolling velocities of the different selectins (in vivo): L-selectin mediates the fastest rolling (SO to 150 pm per second), whereas E-selectin

selectin velocities occupy much of the intermediate ranges (20 to SO pm per second). This difference in velocity classes mediated by each selectin would allow for capture and fast rolling by L-selectin, followed by slower rolling, by P-selectin and/or by E-selectin. Rolling (and capture) via L-selectin is therefore likely to facilitate and enhance rolling via the endothelial selectins. Consistent with this hypothesis, spontaneous rolling in normal mice occurs at the lower velocity of P-selectin, even though both L-selectin and P-selectin are involved." Velocity differences between selectins would also allow for another layer of regulation of leukocyte-endo- thelia1 recognition: the higher the interacting fraction a n d or the lower the velocity of the interaction, the higher the likelihood that leukocytes will be activated by endothelial signals (eg, chemokines), leading to firm arrest and entry into tissues. Therefore, in addition to the combinatorial use of different selectins, activation signals and integrins by dif-

mediates the slowest (3 to I O pm per ~ e c o n d ) . ~ ~ ~ ' ~ ~ - l ~ ~ P-





Fig 2. Structure and homolo- gies of selectins. Individual domains are depicted as different shadings, with vertical lines in- dicating positions of cysteines. EGF, epidermal growth factor; SCR, short concensus repeat; TM, transmembrane; cyto, cytoplasmic domain. The double slash between the SCR domains is to indicate that selectins vary in the number of SCR: L-selectin has 2, but E- and P-selectin can vary from 4-9. Approximate amino acid identities are given for each domain, based on published sequences for human, mouse, and cow.

C-type lectin EGF SCRs TM Cyto

. . . . . . . . . . . . . . . . . . . . . NH 2 . . . . . . . . . . . . . . . . . . . . .

-36 -63 -23 17-35 Residues: -120

same selectin, different species: -72%

different selectin, same species: -52%

same domain, other proteins: -2040%

ferent classes of leukocytes,"' intrinsic differences in the rolling properties of the three selectins will influence the efficiency and nature of leukocyte accumulation in tissues.

This hypothesis is dramatically illustrated by analysis of mice deficient in both endothelial selectins.'55."" Although L-selectin function is intact on circulating lymphocytes, as evidenced by normal lymph node architecture and cellu- larity, recruitment of neutrophils to experimentally induced sites of inflammation is completely absent at early timepoints. Unlike mice deficient in a single selectin, a proportion of these mice also exhibit a severe immunodeficiency, mani- fested by severe and ultimately fatal mucocutaneous bacterial infections. Analysis of leukocyte rolling in these mice shows a corresponding near total absence of rolling in un- treated or TNF-a-treated venules up to at least 4 hours, with the few residual rolling cells exhibiting the characteristically high velocity of L-selectin.'s'.'5h The diminished expression of L-selectin on blood neutrophils in these mice makes it difficult to draw firm conclusions about the actual biologic capacities of L-selectin in the absence of the other selectins. Nonetheless, the phenotype of these mice suggests that the ability of L-selectin to mediate capture and fast rolling cannot be translated into effective leukocyte arrest and transmigration in the absence of at least one endothelial selectin. Consequently, neutrophils and other leukocytes are extremely impaired in their ability to escape from the circulation, and blood leukocyte counts become progressively extremely high, especially in the setting of concurrent infection. Other hematologic abnormalities and perturbations of leukocyte homeostasis, which are probably secondary to the inability of leukocytes to escape the circulation, also en- sue."" The phenotype of these endothelial selectin deficient mice offers strong support for the physiologic importance of synergy between selectins during leukocyte recruitment in inflammation.

However, neutrophil recruitment into the peritoneum of EIP-selectin-deficient animals in response to bacterial chal- lenge is quantitatively normal at 24 hours, and numerous leukocytes can be found in the dermis of infected animals,"' suggesting that other, possibly L-selectin-dependent, mechanisms are still effective. In addition, the phenotypes of the

-60% -40% -80% >75% (but P, >90%) (but L, >95%)

-47% -35% None None

-3040% -2535% - NA-

single selectin knockout animals and the E P double-knockout animals are distinct: L-selectin-deficient mice have detectable defects in leukocyte recruitment at timepoints up to 48 hours,"' P-selectin-deficient mice have partial defects at early timepoints but not significant defects later,'" and E- selectin-deficient mice have no significant defects in leukocyte recruitment dete~table.'~" EP-selectin doubly deficient mice have near total defects in leukocyte recruitment early, but not by 24 hours. Hence, although the synergy between selectins is important for effective leukocyte recruitment during acute inflammation, the functions of the individual selectins are distinguishable, and not completely essential, in all settings of chronic inflammation.

STRUCTURE OF SELECTINS

The selectins are typical type I proteins composed of a tandem array of discrete protein d~mains.".".'~'~'"' These include an amino terminal C-type lectin domain, a single epidermal growth factor (EGF)-like domain, from two to nine short concensus repeat (SCR) domains, a single membrane spanning region, and a cytoplasmic tail (Fig 2). The amino acid identity between the lectin domains of the three selectins within the same species is --52%, but increases to -72% when comparing the same selectin between species. Similarly, the amino acid identity between the EGF-like domains of the three selectins within the same species is -47%, increasing to -60% or higher when comparing the same selectin between species. The identity between SCR domains is generally lower, approximately 35% to 40%. In addition, the number of SCR domains varies: although L-selectin in all species described has two SCR domains, the number of SCR domains in E- and P-selectin can vary from four to nine. However, selectin SCR domains all have six conserved cysteine residues, in contrast to SCR domains found in other proteins, eg, complement receptors, which have four.'"'.'"' No homology exists between the transmembrane or cytoplasmic domains of different selectins, but these regions are well conserved for a given selectin between species. A re- markable example of this homology is the predicted membrane spanning residues of L-selectin, which are identical





in 22/23 residues between humans, mice, and cows. The cytoplasmic tails of the selectins are also well conserved beween species, suggesting that selectin-specific functions reside in these regions (see below).

Consistent with the modular structure of selectins, each of the individual domains is encoded by separate exons within the genes which encode the selectins. Thus, the translational start site, the lectin domain, the EGF-like domain, each SCR domain, and the transmembrane domain (and flanking residues) are each encoded by a separate exon, with the cytoplasmic tail of each selectin encoded by two distinct exons~53.161.16J The selectins have been mapped to syntenic regions of chromosome 1 in both humans and mice, and are clustered together over a region of -300 kb in the order P/ L E , reflecting their common evolutionary origin.'"

C-type lectin domains, EGF-like domains, and SCR domains are each found in numerous other proteins. However, no other known proteins incorporate all three of these protein modules in a single protein. This makes it likely that each of the domains present in selectins has a specific contribution towards the function of the molecule as a whole, and that the arrangement of these domains is also important.

MOLECULAR BASIS OF CELL ADHESION BY SELECTINS

The first indication that selectins functions as mammalian lectins came from studies in which various simple carbohydrates inhibited the binding of lymphocytes to lymph node HEV in a stereospecific f a ~ h i o n . ' " ~ ' ~ ~ Binding could also be inhibited by complex carbohydrates, including polyphospho- mannan ester (PPME) and f ~ ~ ~ i d i n . ~ ~ " ~ ~ ~ ~ ~ ~ ~ ~ These soluble complex carbohydrates were useful also because they could be labeled with fluorescein or radioisotopes and used to as- sess the activity or integrity of the lectin domain independent of actual cell adhesion a ~ s a y s . ~ ~ . ' ~ ~ . ' ~ ~ . ' ~ ~ Binding of lymphocytes to lymph node HEV in the frozen-section assay could also be inhibited by treatment of the frozen section, but not the lymphocytes, with neuraminidase, which cleaves sialic acid moieties present at the termini of oligosaccharide

Homing of lymphocytes to lymph nodes in vivo could also be inhibited by intravenous administration of n e ~ r a m i n i d a s e . ' ~ ~ . ' ~ ~ Subsequent to the molecular cloning of the selectins, numerous investigators showed that adhesion mediated by E- and P-selectin could also be substantially blocked by a variety of simple and complex carbohydrates, most of which are sialylated, fucosylated, or b ~ t h . ~ ~ " " ~ A hallmark of cell adhesion mediated by all three selectins is the inhibition of adhesion by treatment of ligand-bearing cells with neuraminidase.lX6 In addition, most MoAb that block adhesion by selectins define epitopes located within the lectin domain.174,178.'87.'"8 Finally, adhesion by selectins is characteristically blocked by chelation of Ca2+ by EGTA.'" Collectively, these observations offer strong evidence that the selectins function in a calcium-dependent, lectin-like fashion, consistent with the presence of a prominent amino- terminus C-type lectin domain.

The crystal structure of the E-selectin lectin plus EGF- like domain has been solved at 2.0 A resolution.'s' The lectin domain of E-selectin aligns closely with that of the mannose

&ains. 175-177

binding protein,"" fortifying earlier studies based on this presumed alignment.1y".'y2 However, the nature of Ca2' ion binding is quite different from the other known C-type lectins, in that selectin lectin domains have only one bound calcium ion, compared with two in the other C-type lectins; the second Ca2+ binding site is replaced in E-selectin with a water molecule.'"~'93 In addition, the calcium-binding loop of E-selectin is distinct from that of the mannose-binding protein. Mutagenesis studies of the lectin domains of the selectins have identified a number of residues involved in the binding of model carbohydrate ligands (eg, sLex), myeloid cells, and MoAb which block binding of neutro-

and these residues map at or near this Ca2+- binding face of the molecule.'*' Importantly, a number of positively charged residues, most of which are identical in all three selectins, have been identified by these studies. Some of these invariant residues identified by mutagenesis studies are likely to recognize sialic acid or fucose, both of which appear to be essential components of selectin ligands (see below). The identified residues are derived from non- contiguous sequences at both the amino and carboxy terminus of the lectin domain. In a properly folded lectin domain, these residues appear to collectively form the ligand binding surface of the lectin d ~ m a i n . ' " , ' ~ ' . ~ ~ ~ Finally, using chimeric selectins created by domain swapping, exchange of lectin domains between selectins results in an exchange of adhesive specificity. '94 Collectively, these studies firmly establish a prominent and essential role for the lectin domain in adhesion by selectins.

An important question concerns the role of the EGF-like and SCR domains in the function of selectins. Certain MoAb against L-selectin which inhibit adhesion, including the Ly- 22 MoAb against murine L-selectin,Iy5 and the LAMI-I MoAb against human L-selectin, define epitopes composed in part of residues in the EGF-like domain^.^'.'^^ A common property of these MoAbs is their lack of effect on binding of soluble carbohydrate ligands such as PPME or fucoidin to the lectin domain of L-selectin. Deletion of the EGF- like domains of selectins either abolishes or sharply inhibits

These deletions appear to have global effects on protein structure, because epitopes localized to the lectin domain are absent in these deletion mutant^.'^^^''^^'^^ However, replacement of the EGF-like and SCR domains of L- or P-selectin with the corresponding domains of the other selectin had no significant effect on L-selectin function, and only mild effects on P-selectin f~nct ion ."~ Furthermore, replacement of the E-selectin EGF-like domain with the EGF domain of factor IX similarly had no significant effect on cell adhesion. 18' Therefore, the EGF-like domains may contribute essential structural information, perhaps required for the con- formation of the lectin domain, without contributing directly to ligand specificity. However, chimeric selectins containing the P-selectin EGF domain plus the L-selectin lectin domain could, when expressed in transfected cells, mediate binding both to myeloid cells bearing P-selectin ligands (ie, HL-60 cells) and to endothelium bearing L-selectin ligands (ie, lymph node HEV and postcapillary venule^).''^ In addition, cell adhesion to chimeric selectins containing only the P-

I78.1'1.1Y2

adhesion. 178. (87.1%





selectin lectin domain (on a background of L-selectin) was suboptimal; full binding required both the lectin and EGF domain^.'^.'^^ Furthermore, peptides derived from the EGF domain of P-selectin inhibit leukocyte adhesion to thrombin stimulated platelets, and support binding of U937 cells.’98 Given these functional observations, the extraordinarily high conservation of the P-selectin EGF domain compared with those of the other two selectins (89% v -60%), and the observation from the E-selectin crystal structure that there is very little interaction between the lectin and EGF domains,’” it is plausible that the EGF domain of P-selectin is unique in directly contributing to ligand specificity.‘” An attractive hypothesis is that this region of P-selectin interacts with the tyrosine sulfation motifs of PSGL- 1, the principal ligand for P-selectin (see below).

Similar observations have been made regarding a requirement for the SCR domains of selectins. One MoAb, designated EL-246, binds to both L-selectin and E-selectin and inhibits the adhesive function of both molecules.’99 EL-246 defines an epitope in the SCR of both select in^.'^' In other studies, deletion of the SCR sharply impairs recognition of ligands, as measured by binding of soluble recombinant forms of the selectins.’yh.2W~2n’ H owever, exchange of the two L-selectin SCR for the nine P-selectin SCR had no discern- ible effect on the function of either m~lecule.”~ Therefore, the SCR domains also make an important contribution to function, but the nature of this contribution remains unclear. That some unique function resides in the SCR domains in suggested not only by the observations cited above, but by the fact the selectin SCR domains have six conserved cysteine residues, whereas SCR domains present in numerous other proteins (eg, complement receptors’623202) typically have four. Extension of the lectin and EGF domains away from the cell surface might be important for effective cellular interactions, and the multiple tandem SCR domains, which are predicted to form linear rodlike arrays, are well suited for this function. It is also possible that the SCR domains serve to oligomerize otherwise unassociated selectin molecules, thereby potentially increasing the avidity of interaction with ligand. Studies that replace selectin SCR domains with SCR domains of other proteins have not been reported.

CELL BIOLOGIC ASPECTS OF ADHESION BY SELECTINS

An interesting and important aspect of selectin biology is the cell-surface topology of their expression. At the electron microscope level, leukocytes have a complex surface appearance, with numerous ruffles, microvilli, and related structures projecting away from the cell surface. Interestingly, L- selectin is selectively localized to the microvilli, a location hypothesized to facilitate initiation of endothelial binding.203-2” In comparison, most other cell-surface molecules are uniformly distributed. Other adhesion molecules, including p 2 integrins and CD44, are specifically localized away from microvilli onto the cell body.205.207 Recently, it has been shown that PSGL-1, the principal ligand for P-selectin (see below), is also localized to microvilli in a manner essentially identical to the pattern seen with L-selectin.208 Similarly, a 4 integrins on lymphocytes, which can under certain circum-

stances mediate attachment to endothelium under flow conditions,7xzn9 are also localized to microvilli and related surface structures.2w In each case, these “rolling receptors” are present largely in clusters, rather than isolated individual molecules. The biologic significance of this clustering is unclear. However, the preferential appearance of these three leukocyte ‘‘rolling receptors” on microvilli suggests that localization to microvilli is important for effective cell adhesion. Consistent with this, one study reported that an L- selectin mutant redirected away from the microvilli by replacement of the cytoplasmic plus transmembrane regions with those of CD44 was less efficient in initiation of rolling (‘ ‘tethering”).207 These studies collectively suggest that one facet of the control of endothelial recognition may involve regulation of the surface topography of receptors which are responsible for leukocyte tethering and rolling. Whether E- and P-selectin expressed on endothelial cells also exhibits preferential localization to microvilli, or whether this property is limited to leukocyte rolling receptors, is presently unknown.

The molecular basis for sorting of specific receptors to microvilli (or other discrete domains) of leukocytes is poorly understood. For L-selectin, this microvillar localization pattern appears to depend on the cytoplasmic domain, transmembrane domain, or both, as exchange of a segment that includes both of these regions between L-selectin and CD44 resulted in an exchange of localization properties.’” For PSGL-1 and VLA-4, no comparable studies have been reported, and it is worth noting that L-selectin, PSGL- 1, and VLA-4 share no homology or obvious motifs in their cytoplasmic domains. The cellular machinery responsible for sorting of L-selectin, PSGL- 1, and VLA-4 to microvilli has not been identified, but is likely to be universally expressed in hematopoietic cells, because this pattem of localized expression is seen in normal neutrophils, lymphocytes, and numerous stably transfected leukemic cell lines representing multiple Iineages20”2“~2’n (Kansas GS, et al, manuscript in preparation). As the presence of microvilli on leukocytes (and other cells) is dependent on an intact actin microfilament system, it is tempting to hypothesize that interaction between the cytoplasmic tails of these rolling receptors and the actin cytoskeleton is involved in sorting to microvilli. However, a mutation of the L-selectin cytoplasmic tail which blocks association of L-selectin with a-actinin and vinculin, cytoskeletal proteins which link transmembrane receptors with the cytoskeleton,2”,2’2 fails to block sorting of the mutant L-selectin to microvilli of transfected In combination with the data described above, this suggests that the highly conserved membrane-spanning residues of L-selectin may somehow be responsible for sorting of L-selectin to microvilli. Elucidation of the molecular mechanisms for sorting of leukocyte adhesion receptors to specific domains of the cell surface is important to a full understanding of the function of these molecules.

A second important aspect of the cell biology of selectins involves their interaction with the cytoskeleton. Although well documented for integrins and cadherins, an interaction between the cytoskeleton and selectins has only recently





been e ~ t a b l i s h e d . ~ ~ ~ . ~ ~ ~ Truncation of the 17 residue L-selectin cytoplasmic tail by 11 residues blocks adhesion to HEV and rolling in mesenteric venules in vivo without altering cell-surface expression or carbohydrate recognition.82 L-selectin constitutively (ie, without overt activation or cell adhesion) interacts with the actin microfilament system via a- actinin and vinculin?” and truncation of the L-selectin cytoplasmic tail blocks this interaction without preventing sorting to microvilli.210 Taken together, these observations suggest that the constitutive association between L-selectin and the actin cytoskeleton is necessary for adhesion via L- selectin, and that this functional importance is unrelated to ligand recognition or sorting to microvilli. Cytoskeletal association of L-selectin may be important for retention of L- selectin in the membrane, maintenance of proper receptor positioning, or signaling functions during interaction with endothelial ligands.

In contrast, E- and P-selectin do not appear to constitutively interact with the actin cytoskeleton or a-actinin, and deletion of the cytoplasmic tails of either E- or P-selectin has no effect on cell a d h e ~ i o n . ~ ~ ” ~ ~ ~ Rather, linkage of E- selectin to the actin cytoskeleton is inducible following leukocyte adhesion to endothelium through E-~electin.~‘~ Leukocyte adhesion, or crosslinking E-selectin with MoAb, induces association with several actin-associated cytoskeletal proteins, including a-actinin, vinculin, filamin, and paxil- lin, as well as focal adhesion kinase (FAK),2’1 a tyrosine kinase localized to focal adhesions of adherent ~ e l l s . ~ ~ ~ . * ’ ~ MoAb binding to E- and P-selectin on activated endothelial cells has also been observed to result in changes in cell shape,2” presumably also a cytoskeletally driven process. These observations raise the possibility that linkage between E-selectin and the actin microfilament system following leukocyte adhesion is involved in leukocyte traction and signaling events during leukocyte transmigration into tissues, suggesting a broader role for E-selectin in leukocyte traffic than previously appreciated.

LIGANDS FOR SELECTINS

Preliminary considerations. As indicated above, the ligands for selectins are at least in part carbohydrate, consistent with the presence of the lectin domain and the sensitivity of cell adhesion mediated by selectins to neuraminidase. As summarized in an excellent recent selectins appear to recognize quite a structurally diverse array of carbohydrates, which are sialylated, fucosylated, andor sulfated (or all three). However, the data must be interpreted with certain caveats. The term “ligand” may refer to the molecule being recognized, or to the precise structures that directly and phys- ically interact with the selectin. In addition, as pointed out by Varki?I8 distinctions must be drawn between structures that can interact with selectins under certain conditions in vitro and structures that actually do interact with selectins under physiologic conditions in vivo. Thus, the assay conditions used can have a strong influence on the nature of the “ligands” which are “identified.” Assays performed with whole cells under nonstatic/flow conditions, or binding assays using stringent washing conditions, eg, flow cytome-

try, affinity chromatography, or blotting, seem more likely to identify ligands of biologic significance than static cell adhesion assays or assays employing artificially high concen- trations of immobilized glycoconjugates and soluble selectins in enzyme-linked immunosorbent assay (EL1SA)-type formats.’” Putative ligands may act differently when immobilized on plastic, expressed at the surface of a cell, or in solution. Ultimately, any candidate ligands must be shown to be present and functional in a physiologic setting.

In direct cell-cell adhesion assays, adhesion by all selectins is uniformly dependent on sialic acid, as evidenced by a strong sensitivity to neuraminidase treatment of ligand- bearing cells. Similarly, several lines of evidence strongly indicate that all three selectins require fucose for biologically relevant recognition (see below). Direct functional evidence has also emerged that, for L- and P-selectin, but apparently not for E-selectin, the biologic ligand requires s~ l fa te”~~”’ (see below). However, it is not necessarily the case that each of these moieties must be present on the terminus of the same carbohydrate chain. In the case of PSGL- 1. the sulfate is directly linked to 1 to 3 tyrosines of the protein back-

(see below). Taken together, these findings are consistent with the concept that what selectins actually recognize is a “discontiguous carbohydrate epitope””‘.’’’ composed of carbohydrates attached to distinct carbohydrate chains or amino acids of a glycoprotein, analogous to MoAb- defined epitopes composed of amino acids which are not contiguous in the linear sequence of the target protein. Thus, certain simple tetrasaccharides, eg, sLex (NeuAca2-3Gal/?I- 4(Fucal-3)GlcNAc), which were originally proposed as ligands for selectins based on blocking studies with IgM MoAb or ELISA-type assays using glycoconjugates,’7”’7‘~~’8’~~~~ and which may contribute to or form part of the biologically relevant ligand, may nonetheless be more accurately thought of as markers for the enzymatic machinery required for construction of biologically relevant selectin ligands (see below). This model is consistent with the rather low affinities of sLex-based tetrasaccharide inhibitors of selectin-mediated adhesion, even when expressed as neoglycoproteins, and with the variable results obtained from blocking studies with anti-carbohydrate IgM MoAb.

Fucosyltransferases as crucial regulatory enzymes in selectin ligand biosynthesis. The crucial role of carbohydrates in adhesion by selectins has served to focus attention on glycosyltransferases responsible for construction of selectin ligands. Transfection of nonhematopoietic cells with an epithelial a 1,3 fucosyltransferase’” (now designated FucT- 111) conferred cell surface expression of sLex and the ability to bind E-selectin.*26 The lack of expression of FucT-I11 in leukocytes prompted the search for similar enzymes that are expressed in leukocytes and might therefore be better candidates for physiologically relevant enzymes. A myeloid cell a 1,3 fucosyltransferase (originally designated ELFT,227 now termed FucT-IV2‘*) was identified by expression cloning using an IgM MoAb (against a still undefined, nonsialy- lated carbohydrate) that had blocking activity for HL60 cell binding to E-~electin.~~’ This enzyme could confer binding to E-selectin when expressed in some but not all lines of

bone220-222





Chinese hamster ovary (CHO) In addition, FucT- IV can create effective E-selectin ligands when stably expressed in some, but not all, human leukemic cell lines (see below). Although these data suggested that FucT-IV might be involved in E-selectin ligand biosynthesis, further investigation revealed the existence of three other a 1,3 fucosyltransferases. The genes for two of these enzymes, designated FucT-V and FucT-VI, are closely related to FucT- 111, and are located together on human chromosome 19, but are not expressed to a significant degree in normal leukocytes or most leukocyte cell lines.231s232

In contrast, the most recently cloned a 1,3 fucosyltransferase, designated FucT-VII, is located on chromosome 9, and is expressed in normal leukocytes and in myeloid cell lines which bind E- and P - ~ e l e c t i n . ~ ~ ~ ~ ~ ~ ~ In addition, the substrate specificity of FucT-VI1 is clearly distinct from the other a1,3 fucosyltransferases. In COS cells, most CHO cell lines, and all human leukemic cell lines analyzed to date, transfection with FucT-IV cDNA confers high levels of Lex but not sLex, whereas FucT-VI1 gives sLex but not Lex, and FucT-I11 gives not only both Lex and sLex, but also their stereisomers,

These substrate preferences are largely recapitulated using synthetic oligosaccharide ac- ceptors in vitro. Because the addition of fucose is thought to be the last step in the biosynthesis of sialylated, fucosylated lactosamines such as sLex, and because a2,3 sialyltransfer- ases capable of constructing the sialylated type I1 lactosamine precursor are widely expressed, expression of fucosyltransferases is an attractive candidate for an important regulatory step in the expression of biologically relevant selectin ligands. Taken together, these enzymatic and expression data suggested FucT-VI1 and possibly FucT-IV as good candidates for enzymes critically involved in selectin ligand biosynthesis in leukocytes.

Data strongly supporting this hypothesis have recently emerged. Transfection of human lymphoid cell lines with cDNA encoding FucT-VI1 confers binding to E-selectin in all cell lines t e ~ t e d ~ ” , ~ ~ ~ (Wagers AJ, et al, submitted). In addition, activation of normal human resting blood T cells induces the coordinate expression of sLex-like carbohydrates, FucT-VI1 expression, and binding to E-~electin.”~ In contrast, expression of FucT-IV confers binding to E-selectin in only some cell lines, and is not induced in activated T cells that bind E-se1ecti1-1’~~ (Wagers AJ, et al, submitted). Similarly, in transfected hematopoietic cell lines, binding to P-selectin requires expression of FucT-VI1 (plus PSGL- 1 ; see below) (Snapp KR, et al, submitted). FucT-IV does not confer binding to P-selectin in any (PSGL-l+) human hematopoietic cell line thus far examined (Snapp KR, et al, submitted). Results obtained with FucT-IV in transfected CHO

may be complicated by activation of endogenous fucosyltransferase genesz3’ and variation in the glycosylation phenotype of different strains of CHO cells.227-230 These data suggest a critical role for FucT-VI1 in construction of T-cell and myeloid ligands for both E- and P-selectin, and also imply a role for FucT-IV in myeloid cell E-selectin ligand biosynthesis. The recent demonstration that expression of FucT-VI1 in mouse lymphoid organ HEV correlates perfectly

Lea and sLea.225.227,229,230,233

with expression of L-selectin ligands239 supports a role for FucT-VI1 in L-selectin ligand construction. By implication, this may potentially include leukocyte ligands for L-selectin (see below). FucT-VI1 may therefore be involved in construction of ligands for all three selectins. These observations cumulatively make clear the critical role of FucT-VI1 in selectin-mediated leukocyte traffic.

The importance of FucT-VI1 in construction of ligands for all three selectins on normal cells in vivo has recently been confirmed through the analysis of mice with targeted deletions of FUCT-VII.’~~ These mice have nearly absent functional E- and P-selectin ligands on their leukocytes, high white blood cell (WBC) counts, and drastically reduced spontaneous leukocyte rolling in postcapillary venules. In addition, leukocyte entry into the inflamed peritoneum is sharply impaired, to a significantly greater degree than for any single selectin knockout mouse, consistent with a role for FucT-VI1 in construction of ligands for all three selectins. These mice also exhibit no detectable L-selectin ligands on their lymph node HEV (although staining with MECA-79 is retained; see below), and a severe reduction in short-term lymphocyte homing in vivo to lymphoid organs. These studies firmly establish a crucial role for FucT-VI1 in construction of ligands for all three selectins, and implicate FucT- VI1 as a crucial enzyme in the regulation of leukocyte traffic.

Finally, the availability of cloned a l , 3 fucosyltransferases has permitted a reevaluation of the putative role of MoAb- defined carbohydrates in adhesion mediated by selectins, and an analysis of the relationship between various related MoAb-defined carbohydrates, a 1,3 fucosyltransferase expression, and leukocyte adhesion to E- and P-selectin. Analy- sis of a panel of stably transfected human hematopoietic cell lines expressing FucT-VII, FucT-IV, or neither, shows that epitopes defined by the “sLex-like” MoAb HECA-452,ls0 CSLEXl?’ 2F3,242 and AM-3243 are closely associated with FucT-VI1 expression and adhesion to E-selectin (Wagers AJ, et al, submitted). In contrast, epitopes defined by SNH3,244 FH6,245 or 2H5,246 previously also described as “sLex-like” or “variant sLex,” were present on all cell lines examined, including multiple cell lines which did not bind to E-selectin, certain murine hematopoietic cells, and at least two cell lines with no detectable fucosyltransferase activity (Wagers AJ,et al, submitted). Similarly, the VIM-2 MOA^,"^ which defines the CDw65 and is reportedly directed against the internally monofucosylated sialylated lactosamine Neu- Aca2-3Galfl1-4GlcNAcfl1-3Galfl1-4GlcNAc(Fuca 1-3)pl- 3Ga1, also stains numerous cell lines which do not bind to E-selectin (Wagers AJ, et al, submitted), but is closely associated with FucT-IV expression. However, the VIM-8 and VIM-1 1 MoAb, initially also classified as CD~65 ,2~’ stained essentially all cell lines examined, including two cell lines with no detectable fucosyltransferase activity, similar to the SNH3, FH6, and 2H5 MoAb (Wagers AJ, et al, submitted). VIM-8 and VIM-1 1 clearly recognize distinct, non- sialylated carbohydrates.”’ Importantly, the inability of any given cell line to bind to E-selectin was not due to underlying defects in the expression of any other proteins essential for ligand activity, because transfection of nonbinding cells with





FucT-VI1 cDNA alone always conferred high levels of adhesion to E-selectin (Wagers AJ, et al, submitted). Finally, a stable variant of the HL60 cell line was identified which was missing the “FucTVII-associated” epitope(s) defined by HECA-452, CSLEX1, 2F3, and AM-3, but which retained full activity for adhesion to E- and P-selectin. Taken together, these results indicate that expression of the FucT- VII-associated epitopes corresponds closely with, but is not required for, binding to E-selectin. This suggests that these carbohydrates are good markers for the glycosyltransferases required for binding to E- and P-selectin, especially FucT- VII, but that these MoAb-defined. carbohydrates are not themselves the actual physiologic ligands.

Glycoproteins as biologically relevant selectin ligands. Although the selectins can each recognize sLex and related simple tetrasaccharides, their specificities, and therefore by implication their ligands, are clearly distinct at the level of cell-cell adhesion. To identify selectin ligands, a number of groups have used purified selectins or soluble recombinant selectins in the form of Ig fusion proteins (receptor globulins or RIg) in affinity isolation studies to identify candidate glycoprotein ligands for all three select in^.'^^^^^' An important aspect of these studies is that, under the conditions of the experiments, the selectins themselves, when used as affinity reagents, exhibit a higher level of specificity than MoAb against their putative carbohydrate “ligands,” similar to true cell-cell adhesion assays. Thus, the MECA-79 MoAbZs4 immunoprecipitates 7- 10 bands from mouse lymph node HEV, whereas L-RIg brings down only three (see below). In Westem blotting type studies, P-selectin recognizes only a single band in lysates from neutrophils and HL60 cells,25’ and E-selectin recognizes 2-4, clearly less than is recognized by either CSLEXl or HECA-452 (see below), which are directed against sLex and/or related sialylated and fucosylated lactosamines. The binding requirements and specificities of the selectins can also be distinguished from those of these anti-carbohydrate MoAb as well, even for MoAb that block the relevant adhesion event, such as MECA-79. These observations reinforce the idea, described above and e l ~ e w h e r e , ~ ’ ’ ~ ~ ~ ~ that what selectins actually see is not a simple oligosaccharide such as sLex, but a three- dimensional surface composed of carbohydrate moieties contributed by several molecular species, attached to a specific glycoprotein. Candidate glycoprotein ligands for each of the selectins have been identified (see below).

Because selectins clearly recognize carbohydrates, the precise role of the protein “core” of the glycoprotein ligand is of considerable interest. Several, not mutually exclusive, functions of the protein portion of a selectin ligand glycoprotein can be imagined. First, the protein may be required for the appropriate “display” of carbohydrates to the lectin domains of selectins. This may occur via generation of a ‘‘clustered oligosaccharide patch,’ ’2’8.223 whose appropriate orientation would necessarily depend on the underlying protein, but whose individual carbohydrate elements might not be unique to that glycoprotein.’’8 Alternatively, or in addition, the information required for proper glycosylation and/ or other posttranslational modifications might be contained

in the protein itself, thus guaranteeing that only certain proteins are modified to function as biological ligands. This last possibility seems likely at least for the tyrosine sulfation motif of the P-selectin ligand, PSCL-1 (see below). Second, the protein may be essential for sorting of the ligand to the appropriate domains of the cell surface, ie, the microvilli. It is difficult to envisage how non-protein carbohydrate-bearing molecules, eg, glycolipids, might be localized to microvilli. Third, proteins are obviously well-suited for any “outside- in” signaling function that may occur through selectin l i - gands, as appears to be the case for PSGL-I and L-selectin (see below), presumably via interactions between their cytoplasmic tails and cytoplasmic effector molecules. Finally, the protein portion of a selectin glycoprotein ligand may be directly involved in protein-protein interactions. A single protein could potentially fulfill any or all of these functions.

Probably the most extensively characterized selectin ligand is P-selectin glycoprotein ligand- 1 (PSGL- I). PSGL-I is a disulfide-linked homodimer of two identical - 120-kD chains first identified by Moore et by Western blotting of extracts of membrane glycoproteins from neutrophils and HL-60 cells with purified soluble P-selectin. Subse- quent molecular cloning of PSGL-lZs5 showed a type 1, mucin-like protein with a unique structure. Following a typical signal sequence, PSGL-I has a stretch of -100 residues containing a consensus PACE cleavage site followed closely by a potential tyrosine sulfation motif. Following the first -100 residues are 15*” or 16’”* decameric repeats of the sequence A-T/M-E-A-Q-T-T-X-PLAIT, where X can be P, A, Q, E, or R. This is followed by a short stretch of residues before the transmembrane region, and a 69-residue cytoplasmic tail.*” The mouse homologue of PSGL-I retains each of these features, except for having 10 decameric repeats instead of 15 to 16,”‘ and exhibits significantly greater homology in the transmembrane and cytoplasmic domains than in the external portion of the molecule.256 Interestingly, the mouse decameric repeat sequence appears distinct: E-T- S-QIK-P-A-P-TN-E-A, but actually aligns well with the human if the human sequence is considered to begin with the first glutamine.’“ Although mouse PSGL- I has five less decameric repeats than the human, the predicted mouse protein is only five residues shorter, by virtue of having longer sequences than the human protein both in front of and behind the decameric repeats, in the extracellular region. Both mouse and human PSGL- I have serinekhreonine-rich stretches of residues outside the decamer repeats; these potential sites of 0-linked carbohydrates may be important for binding to P-selectin.

Detailed biochemical characterization of PSGL- 1 shows that, consistent with the demonstrated requirements for sialic acid and fucose, PSGL-1 has extensive branched-chain, fucose- and sialic acid-containing polylactosamine side chains, most of which are presumably attached to serines and/or threonines of the decamer repeats, and many of which termi- nate in sLex~?2’.2’7.257 This arrangement of clustered, sialylated 0-linked sugars makes PSGL-1 sensitive to digestion with the enzyme 0-sialoglycoprotease (OSGP),*57~2s’ a protease isolated from Pasteurella hemolytica that also cleaves

PSGL-1.





other mucin-like proteins such as CD45, CD43, and (334.259-261 The sensitivity of PSGL-1 to OSGP, and the specificity of this protease for glycoproteins bearing clustered sialylated 0-linked carbohydrates, has proven to be useful for studies of P-selectin binding before development of blocking MoAb. PSGL-1 also has 2 to 3 N-linked sugars, but these do not appear to be necessary for binding to P- electi in.*^^ The structure of the O-linked sugars that are involved in binding P-selectin are unknown, and the serine andor threonine residues on which these carbohydrate chains are displayed have not been definitively identified.

PSGL-1 appears to be the principal ligand for P-selectin on all major classes of leukocytes. The PL1 MoAb, devel- oped by Moore et al?08 essentially completely blocks binding of soluble PSGL- 1 to immobilized P-selectin, soluble P- selectin to neutrophils, adhesion of neutrophils to immobilized P-selectin under static conditions, and rolling of neutrophils on P-selectin under shear stress.2o8 Rolling of neutrophils in vivo is also blocked by PL- 1 .262 Variable results have been obtained with a polyclonal antibody to PSGL-1 using T-cell clones or activated T How- ever, binding of T cells, mononuclear cells, neutrophils, and HL60 cells to COS cells expressing P-selectin under conditions of low shear force is essentially totally blocked by a functionally similar MoAb, KPL-1 (Snapp KR, Ding H, Luscinskas FW, Kansas GS, submitted). Similarly, rolling of all classes of leukocytes on P-selectin is blocked by KPL- 1 MoAb (Snapp KR, Ding H, Luscinskas FW, Kansas GS, submitted). Transfection of hematopoietic cell lines of appropriate glycosylation phenotype with cDNA encoding PSGL- 1 confers binding to P-selectin, and this binding is completely blocked by the KPL-1 MoAb (Snapp KR, Wagers AJ, Craig R, Stoolman LM, Kansas GS, submitted). These studies collectively establish the essential role of PSGL-1 in leukocyte recognition of P-selectin.

In addition to fucose and sialic acid, binding of P-selectin to recombinant PSGL-1 requires at least one tyrosine sulfate located within a concensus sequence near the amino terminus of the mature, PACE-cleaved glycoprotein.220~222 Inhibition of sulfation by incubation of cells in 10 mmovL sodium chlorate inhibits binding of treated HL60 cells to P-selec-

Cleavage of the first 10 amino acids of the mature tin.220-222

PSGL-1 with a cobra venom-derived protease named mocarhagin, which removes the tyrosine sulfation motif, abolishes binding to P-selectin.’@ Simultaneous mutation of all three tyrosines to phenylalanine also abolishes binding to P-selectin. Whether all three tyrosines are sulfated in normal leukocytes, and whether sulfation of a particular tyrosine is essential, is not yet known. The epitope defined by the function-blocking PL1 MoAbZo8 binds close to this tyrosine sulfate motif.”’ Interestingly, none of these three tyrosines in the human sequence is conserved at the exact corresponding position in the mouse, which instead has two tyrosine resides just downstream from the positions in the human sequence, also within a concensus tyrosine sulfation Fol- lowing the conserved PACE cleavage site, the human PSGL- 1 sequence is Q-ATEYEYLDYDFLPET, whereas the mouse PSGL- 1 sequence is QVVGDDDFEDPDYTYNT (potential

tyrosine sulfation sites bolded). The lack of conservation of the precise location of the tyrosine sulfation motif suggests some flexibility in the configuration of the actual recognition structure.

PSGL-1 is expressed on essentially all blood leukocytes, including essentially all T cells, most of which cannot bind P - s e l e ~ t i n . ~ ~ ~ * * ~ ~ Among T cells, the capacity for P-selectin binding is confined to a subset of activatedmemory T ~ e l l s . ~ ~ ~ * ~ ~ Activation of T cells induces, over the course of several days, the ability to bind P- (and E-) s e l e ~ t i n ? ~ ~ . ’ ~ ~ and this change in selectin binding function is associated with the de novo induction of glycosyltransferase activity, including F u c T V I I ~ ~ ~ (see above) and possibly C2GnT.263 In mice, Northern analysis of murine tissues indicates that expression of PSGL-1 is widespread,256 suggesting that PSGL- 1 may have a physiologic role unrelated to the interaction of cellular elements within the vasculature. One report7’ indicated that PSGL- 1 is not expressed on human fibroblasts or keratinocytes. Precisely which cells, apart from leukocytes, express PSGL-1, and what the role of PSGL-1 is on these cells, is presently unknown, but is an important area for future investigation. The metastatic potential of a variety of malignant cells that bind P-~electin’~~.’~~ may be related to the functional expression of PSGL-1. These considerations reinforce the importance of tightly regulated control of glycosyltransferases and other proteins which induce functionally relevant posttranslational modifications of widely expressed selectin ligand proteins. The unregulated expression of glycosyltransferases may underlie the metastatic potential of a diverse array of tumor cell types.

Ligands for L-selectin were first identified on mouse lymph node HEV using the MECA-79 MOA^?'^ MECA-79 specifically blocks binding of lymphocytes to lymph node HEV in the frozen section assay and homing of lymphocytes to lymph node in vivo, but does not block binding of lymphocytes to Peyer’s patch HEV or homing to Peyer’s Further work has shown that MECA-79 detects an epitope on multiple glycoprotein species, only some of which bind to L - s e l e ~ t i n ~ ~ ~ . ~ ~ ~ (see below). Like L-selectin binding?I9 the MECA-79 epitope is sulfation-de~endent,’~’ but unlike L-selectin binding, the MECA-79 epitope is independent of both sialic acid and f ~ c o s e , ~ ~ ~ ~ ~ ~ ’ both of which are part of the biological ligand(s) for L-selectin on mouse lymph node HEV. Taken together, these data suggest that MECA-79 detects an epitope that includes the sulfate moiety of the actual L-selectin ligand, but that is distinct from, and overlapping with, the structure(s) recognized by L-selectin. MECA-79 also recognizes the HEV-like vessels that arise in the NOD mouse, and blocks lymphocyte recognition of these vessels.272 Interest- ingly, MECA-79 also recognizes human HEV, both in lymph node and tonsil, and HEV-like vessels in multiple sites of chronic inflammation, and blocks binding of human lymphocytes to HEV from these tissues.273

HEV glycoproteins that carry functional sialylated, fucosylated, and sulfated ligands for L-selectin have been identified with the use of a soluble recombinant form of L-selectin, LEC-IgG.250.274 Three such ligands have been identified, ini-

Ligands for L-selectin.





tially termed sgp50, sgp90, and sgp200. Each of these glycoproteins carries the MECA-79 epitope. Molecular cloning of sgp50Z7' shows it to be a novel mucin-like molecule, containing two regions of clustered serinelthreonine residues, but no apparent transmembrane region. Further work has shown this molecule, designated GlyCAM-1, to be primarily a secreted m ~ l e c u l e , ~ ~ ~ ~ " ~ making it unlikely to be the adhesive ligand for L-selectin on mouse HEV. However. Gly- CAM-I may have important signaling functions (see below). Sequencing of peptides derived from sgp90 showed it to be a glycoform of CD34,27' a mucin-like protein previously described on hematopoietic stem cells.2x" In the mouse, CD34 is universally expressed on endothelium both during development and in the mature anima1,28'~'8' and is therefore perfectly placed to serve as an endothelial ligand for L- selectin, when appropriately modified. CD34 in humans is also widely distributed on vascular endothelium.2x3 However, HEV from CD34 knockout mice support normal levels of lymphocyte binding and homing to lymph nodes, and leukocyte rolling and recruitment during inflammation are normal.2x4 CD34 accounts for -50% of the L-selectin ligand activity purified from human tonsil with MECA-79."' No GlyCAM- 1 homologue in humans has been reported. Taken together, these studies suggest that sgp200, not yet identified at the molecular level, may constitute the principal adhesive ligand(s) for L-selectin on lymph node HEV.

Analysis of GlyCAM- 1 knockout mice and GlyCAM- 1 / CD34 double-knockout mice further implicate sgp200 as an important L-selectin ligand on lymph node HEV. These mice have normal lymphoid organ architecture, normal lymphocyte homing to lymph nodes, and normal leukocyte rolling in postcapillary venules and recruitment to sites of inflammation (Larry Lasky, personal communication, April 1996). Interestingly, genetic deletion of either CD34 or GlyCAM- 1 induces a 5- to IO-fold upregulation of sgp200 expression, as assessed by LEC-IgG affinity isolation, and knockout of both CD34 and GlyCAM-1 increases sgp200 expression even further (Larry Lasky, personal communication, April 1996). This apparent compensatory mechanism leading to elevated spg200 induced by the absence of the other L- selectin ligands suggests that strong, dynamic homeostatic mechanisms have evolved to maintain adequate lymphocyte traffic and recirculation.

The 0-linked carbohydrate structures on GlyCAM-I have been extensively characterized. Two of the more prominent structures of @eliminated chains of GlyCAM-I are novel sulfated derivatives of sLex: 6' sulfo-sLex, ie, NeuAca2- 3(S04-6)Galpl-4(Fuca 1-3)GlcNAc, in which the sulfate is attached at the 6 position of the galactose, and 6-sulfo-sLex, ie, NeuAca2-3Gal~l-4(Fuca 1-3)(S04-6)GlcNAc, in which the sulfate is attached to the 6 position of G ~ C N A C . ' " - ~ ~ ~ Both of these structures are contained within larger, core 2- containing branched 0-linked glycans, the complete structures of which have not yet been completely deduced.'" These 0-linked carbohydrate structures potentially account for all of the L-selectin binding requirements (ie, sulfation, sialylation, and fucosylation) in a single structure; importantly, these structures are compatible with the known distri-

bution and properties of the relevant glycosyltransferases.'X'' Whether these precise structures constitute the actual recognition units for L-selectin binding. and whether these or other structures are present on CD34, sgp200 or other potential ligands, is presently unknown.

Independently, Sawada et alZ4' identified an MoAb, termed 2H5, which was raised against a complex sLex-reactive glycolipid and which selectively stains primarily lymph node HEV, but not postcapillary venules in other organs. The 2HS MoAb inhibits binding of lymphocytes to lymph node HEV in the frozen section assay, and detects glycoproteins of 90, 110, and 250 kD by Western blotting. The possible relationship between these glycoproteins and those detected by MECA-79 or L-selectin are presently unknown. but the apparent molecular weights of these glycoproteins are similar to some of those recognized by MECA-79. Unlike MECA- 79, which does not stain normal leukocytes or leukocytic cell lines, 2H5 stains all or nearly all human hematopoietic cell lines (Wagers et al, submitted), and the 2HS epitope is sensitive to neuraminidase. The identity of the glycoproteins recognized by 2H5 at the molecular level has not been reported.

As discussed above, the existence of L-selectin ligands on cytokine-stimulated cultured endothelium in vitro, and postcapillary venules in vivo, has been inferred from the ability of L-selectin transfectants to roll in vitro and in vivo, and from blocking studies using MoAb against L-selec-

These ligands have not been identified or tin,"'.Y7.'x. 101.l03.lrK,

well characterized. MECA-79 does not recognize cultured endothelium, eg, HUVEC, whether or not the cells have been stimulated with cytokines,"" suggesting that MECA-79 does not recognize all endothelial L-selectin ligands, especially those expressed in acute inflammation. Rather, the pattern of expression of MECA-79 staining in multiple sites of chronic inflammation, which typically arise over days and weeks, and in HEV. which are an excellent example of an actively maintained state of endothelial differentiation,'"' suggests that MECA-79 staining is indicative of endothelial differentiation towards a morphologically recognizable, cell adhesive phenotype.

Finally, it is important to distinguish between several distinct roles played by L-selectin (Fig 3). Interaction between leukocyte L-selectin and endothelial ligands for L-selectin will directly mediate attachment ("tethering" or "capture") and rolling. As described above, this action of L-selectin will facilitate molecularly distinct leukocyte-endothelial interactions, through the action of the endothelial selectins with their leukocyte ligands, eg, P-selectinlPSGL- 1. Blockade of L-selectin could therefore inhibit cellular interactions which at the molecular level do not directly involve L-selectin. The sequential yet synergistic nature of these two distinct types of interactions has not always been appreciated, or incorpo- rated into interpretations of results from a variety of systems. Furthermore, both of these distinct pathways of leukocyte- endothelial interaction will be amplified by leukocyte-leukocyte interactions, also mediated by L-selectin""."' (Fig 3). These leukocyte-leukocyte interactions may be at least partially responsible for the observation that, in most rolling





systems, newly arriving cells tend to bind downstream from bound cells, rather than randomly. This mechanism would tend to concentrate leukocytes together at sites of insult, and would avoid the problem of bound cells blocking access to endothelial ligands of later arriving cells. Again, blockade of L-selectin, leading to inhibition of leukocyte-leukocyte interactions, will indirectly inhibit leukocyte adhesion to endothelium (or other substrate, eg, transfectants). Distinguish- ing between these distinct modes of L-selectin action is difficult or impossible in the absence of firm knowledge of the identity of the L-selectin ligands operating at each level, or reagents which selectively and specifically inhibit one of these pathways. This problem is particularly acute for those assays, even those performed under shear stress, in which the readout is solely a measurement of bound leukocytes, with no direct knowledge of the interactions which led to this endpoint. However, this consideration is also relevant to "pure" flow assays, whether in vitro or in vivo, because simultaneous visualization of leukocyte-leukocyte interactions (above the plane of view) and leukocyte-endothelial/ substrate interactions (in the plane of view) is difficult if not impossible. Direct evaluation of leukocyte tethering rates, ie, initiation of interaction with the substrate, can potentially distinguish to a degree effects on these different categories of cellular interactions. The use of L-selectin-negative cells or cell lines, and the construction of cells which can utilize only a single selectin, is also likely to simplify interpretation of these kinds of data. The difficulties inherent in the inter-

pretation of these complex data are likely to continue to confound the analysis of leukocyte-endothelial recognition, until the relevant L-selectin ligands are identified and characterized.

As discussed above, the initial "identification" of sLex as the "ligand" for E-selectin has given way to a more sophisticated appreciation of the complex nature of selectin ligands, and a recognition of sLex and/ or related carbohydrates as markers of enzymatic machinery required for selectin ligand biosynthesis. In this context, the characteristics of the HECA-452 MoAb's".'"' are of interest. This MoAb detects a carbohydrate or family of carbohydrates expressed on HEV in lymphoid organs, and on neutrophils, monocytes, and a subset of T cells'"~'5"~'"' (see below), a distribution which is perfectly consistent with the close association between FucT-VI1 expression and HECA-452 staining (see above). The precise structures of the sialylated and fucosylated carbohydrates detected by HECA-452 are still unclear, but appear to include both sLex and its stereo- isomer, sLea, as well as other, less well defined carbohydrates found on sLex-negative/low T cells.'"'.'"' Whether this difference in staining between CSLEX I and HECA-452 is due to different affinities for the same structure or different fine specificities in their target antigen is unclear. Similarly, the identity of the differing glycoproteins detected by HECA-452 on different classes of leukocytes"" or HEV is unknown.

HECA-452' T cells constitute a subset of memory T

Lignnh ,fiw E-selectin.

Fig 3. Leukocyte-endothelial interactions mediated by selectins. Depicted are known and predicted adhesion pathways mediated by selectins and their ligands, some of which have not been identified. Endothelial cells will express E-selectin, P-selectin, andlor a still unidentified L-selectin ligand 0). Leukocytes, depicted as neutrophils, will express L-selectin, PSGL-1 (with the tyrosine sulfate motif indicated as Y-SO,), and still unidentified ligands for E-selectin 0) or L-selectin. Leukocyte L-selectin ligands are presented as mucin-like, based on known L-selectin ligands GlyCAM-1 and CD34; no data exist as t o whether the endothelial and leukocyte L-selectin ligands are the same or different. Interactions which are well documented are indicated by bold, filled arrowheads; binding between PSGL-1 and E-selectin is given an open arrowhead for reasons summarized in the text. As indicated, interactions can occur between flowing leukocytes and endothelium, flowing leukocytes and other flowing leukocytes, or between flowing leukocytes and bound leukocytes. Each of these can independently augment leukocyte recruitment. Note the diversity of roles played by L-selectin, due to the presence of L-selectin ligands on both endothelium and leukocytes. See text for further details.





E-selectin ligand precursor E-selectin ligand

FucT-VI1 Fig 4. Correlation of FucT-VI1 expression with ex-

pression of sLex and functional E-wlectin ligands does not mean that sLex is the actual ligand. Expres- sion of sufficient levels of FucT-VI1 will generate detectable sLex expression in essentially any human cell type, due to the ubiquitous presence of tha sialylated type II precursor. In this figure, “R” could be any glycoprotein or glycolipid. The actual E-selectin

\r3@p1’ h glycoprotein ligand, which also requires FucT-VI1 activity for function, may also display $Lex or related carbohydrate. See text for details.

@a1

(GlcNAf) I,

4 lsAl a2

lsAl a2

R sialylated type I1 sialyl-Lewis-X

precursor

cells:29’ and the HECA-452 antigen on these T cells has been termed CLA, for Cutaneous Lymphocyte Antigen.’” Normal and malignant T cells infiltrating the skin are nearly all HECA-452+, whereas this phenotype is rare among T cells in the blood or in sites of inflammation outside

The observation that E-selectin is often expressed in these same skin lesions has led to the hypothesis that CLA is the ligand for E-selectin on these memory T cells. Consistent with this, E-selectin binding is largely confined to the HECA-452+ subset of T cell^.'^ T-cell lines and clones that bind E-selectin are HECA-452+, whereas those that bind P-selectin but not E-selectin are HECA-452-nega- ti~e.’~.’~* Immunoaffinity isolated HECA-452 antigen(s) from tonsils supports binding of E-selectin tran~fectants.’~~ Taken together, these results indicate that HECA-452 staining is an excellent and selective marker of the ability of leukocytes, especially T cells, to bind E-selectin. Given the essential role of FucT-VI1 in binding of T cells to E-selec- tin235 (see above), the expression of HECA-452 epitopes on E-selectin binding T cells is not surprising. However, these results do not establish whether the carbohydrate(s) identified by HECA-452 form part of the actual physiologic ligand (Fig 4). The inability of the HECA-452 MoAb to significantly block leukocyte binding to endothelium or E-selectin transfectants, and the lack of identification of the -200-kD glycoprotein which carries the HECA-452 epitope on T cells, makes it difficult to definitively assign direct ligand activity to HECA-452-defined carbohydrates in T-cell binding to E- selectin.

Picker et alZo3 have proposed that L-selectin on neutrophils functions as a “ligand” for E-selectin (and P-selectin) by “presenting” sLex to the vascular selectins. In this model, the localization of L-selectin to microvilli is hypothesized to confer on L-selectin the preferential ability to “present” sLex over other, identically glycosylated neutrophil glycoproteins that are not localized to microvilli. In support of

.29,150,2%,297

this hypothesis, MoAb to L-selectin and E-selectin do not show additive blocking in some assays,’w although they do in other^.'^"^^ MoAb to L-selectin partially inhibit binding of neutrophils to E-selectin transfe~tants,2~~.’~~ and to P-selectin expressed on activated platelets3” or histamine-treated HUVEC.’23 MoAb to L-selectin inhibit tethering (ie, initiation of rolling) but not rolling on E-selectin in controlled flow assays.302 Characteristically, MoAb to L-selectin block interaction of leukocytes with E- or P-selectin only partially (-SO%), which has previously been interpreted as evidence for the existence of other ligands. E-selectin transfectants bind to L-selectin isolated from neutrophils but not from lymphocytes.203 However, E-selectin transfectants will bind to virtually any immobilized glycoprotein containing high enough levels of sLex. Furthermore, expression of L-selectin is neither necessary nor sufficient for binding to E-selectin. In addition, the anti -L-selectin MoAb that exhibit partial blocking of leukocytes to E-selectin exhibit complete blocking of lymphocyte binding to HEV in the frozen section assay and rolling of L-selectin lymphoid transfectants in vivo, suggesting that the lectin activity of L-selectin is required for each of these activities. Also, L-selectin MoAb do not block rolling on E-selectin once cells are interacting directly with E-~electin.~” Finally, it is becoming clear that E-selectin is the least efficient selectin in initiation of rolling (“tethering”), whereas L-selectin is the most efficient (see below). As discussed above, L-selectin will facilitate rolling on E- or P-selectin by virtue of its ability to initiate contact with its own endothelial ligand, and will also amplify these leukocyte-endothelial interactions via its role in leukocyte- leukocyte interactions. Therefore, the likeliest interpretation of these differing studies is that the inhibition of binding to E-selectin by MoAb against L-selectin is due principally, if not exclusively, to inhibition of leukocyte-leukocyte interactions, and not to direct inhibition of leukocyte recognition of E-selectin per se. It is important to note that the existence





of a functional leukocyte L-selectin ligand on neutrophils was not known when the studies of Picker et alZo3 were first reported.

PSGL-1 has also been proposed as an E-selectin ligand. CHO cells transfected to express E-selectin (CHO-E) bind to purified immobilized recombinant glycosylated PSGL- 1 .25s Whether CHO-E can bind to other similarly glycosylated molecules under similar conditions was not tested in that PSGL-1 MoAbs partially block binding of neutrophils to E-selectin under certain conditions303 but not oth- e r ~ . ~ " The PL5 MoAb described by Ma et al'@' reportedly blocks HL60 binding to both P- and E-~electin.~"~ However, the specificity of this IgM MoAb for PSGL-1 has not been adequately documented, and the epitope is at least partially carbohydrate, given its sensitivity to neuraminidase.'@' In any case, the requirements for binding of PSGL-1 to E- selectin are distinct from those of P-selectin, because PSGL-1 does not require tyrosine sulfation for binding to ~ - ~ ~ l ~ ~ t i ~ . 2 2 u . 2 2 1.237 The binding of E-selectin to PSGL-1 is also apparently of much lower affinity.257 E-selectin may bind to carbohydrates attached to residues in the mucin-like decamer repeats of PSGL-1. However, binding of neutrophils, HL60 cells or T cells to E-selectin is unaffected by OSGP treatment, which destroys P-selectin binding.72,208.223.258.306 CHOE-selectin cells, but not CHOP-selectin cells, bind to COS cells transfected with FucTVII c D N A . ~ ~ ~ ~ ~ ~ ~ Similarly, several PSGL- 1 -negative cell lines, which bind P-selectin only after transfection with PSGL-1 cDNA, bind E-selectin without expressing PSGL- 1 (Snapp KR, Wagers AJ, Craig R, Stoolman LM, Kansas GS, submitted). These findings make it clear that PSGL-1 is not essential for binding to E-selectin. Whether PSGL-1 functions as a physiologic ligand for E-selectin on normal leukocytes remains to be determined.

Walchek et a1252 have described an -250-kD E-selectin glycoprotein ligand on bovine y/ST cells, which was isolated by affinity chromatography with purified E-selectin. The binding is sensitive to neuraminidase and EGTA, and binding is blocked by chymotrypsin treatment of the y/6T cells. Importantly, this glycoprotein, which has not yet been characterized at the molecular level, does not appear to be the bovine homologue of PSGL-1, which can be isolated from the same cells in parallel using recombinant P-selectin (Mark Jutila, personal communication, April 1996). Whether this 250-kD glycoprotein can support adhesion has not been determined, and blocking MoAb have not yet been described.

Using a similar biochemical approach with an E-selectin- Ig chimera, Leninovitz et a1251 have identified a 150-kD glycoprotein from murine neutrophils and myeloid cell lines, which is also apparently expressed by HL60 cells.307 Molecu- lar cloning of the murine protein,"' designated ESL-1, shows it to be nearly identical (94%) to a chicken fibroblast growth factor (FGF) receptor,309 and over 90% identical to a chicken Golgi membrane protein.310 ESL-1 differs from the FGF receptor in only the first 70 amino acids, and may be a splice variant of the FGF receptor. E-selectin apparently recognizes only myeloid cell glycofoms of ESL-1 (or the FGF receptor), and this recognition appears to be controlled by fucosy-

lation:"' consistent with the crucial role for these tightly regulated enzymes in selectin ligand construction (see above). The anti-ESL-1 antiserum partially blocks binding of leukocytes to a recombinant form of E-selectin, suggesting that ESL-1 is capable of binding E-selectin when expressed on cell surfaces. Because the rabbit antiserum used does not distinguish between ESL-1 and the closely related FGF re~eptor,~" the distribution of ESL-1, if different from the FGF receptor, is unclear. However, the anti-ESL-1 antiserum recognizes numerous cell types, including cell lines representing endothelium, fibroblasts, epithelial cells, and hematopoietic cells.3o8 These studies are consistent with the idea, described above, that widely expressed selectin ligand proteins can serve as selectin ligands when appropriately modified by enzymes whose expression is tightly controlled.

OUTSIDE-IN SIGNALING THROUGH SELECTINS AND THEIR LIGANDS

Just as integrins are now recognized as dual-adhesion/ signaling molecules,311 a series of recent reports have suggested that outside-in signaling in leukocytes can be directly mediated by L-selectin and PSGL-1. On neutrophils, MoAb to L-selectin induce increases in intracellular Ca", augmen- tation of the respiratory burst, production of oxygen radicals, and an increase in Mac- 1 surface e x p r e s s i ~ n . ~ ~ ~ - ~ ' ~ Sulfatides, but not other artificial L-selectin ligands (eg, PPME or fu- coidan), can mimic most of these effect^,"^'^^^."^ and can also induce increases in mRNA for TNF-a and IL-8."' Sig- naling through L-selectin may also activate MAP kinase, as well as other, still undefined, tyrosine-phosphorylated sub- s t r a t e ~ . ~ ' ~ The physiologic relevance of results gathered with sulfatides, which bind to L- (and P-) selectin in a Ca2+- independent way and are only effective at high concentra- tions, is unclear. Nonetheless, the ability to initiate an array of intracellular events with MoAb to L-selectin, but not with MoAb to most other cell surface molecules, suggests some specificity in the initiation of these signals. The observation that many of these effects are augmented by cytochalasin treatment of cells, which disrupts the actin cytoskeleton, is consistent with the constitutive interaction between L-selectin and several cytoskeletal proteins which link transmembrane receptors to the actin and suggests that signaling through L-selectin may be regulated in part by cytoskeletal associations, as has been observed with integrins. However, the timescale of these signaling events (minutes) suggests that they may not be directly relevant to initial adhesive interactions occurring between leukocytes and endothelium.

Binding of neutrophils to E-selectin can upregulate the adhesive activity of neutrophil integrins, particularly Mac- 1."' More recently, it has been shown that crosslinking of L-selectin can also directly upregulate the adhesive activity of Mac-1 integrins on neutrophils.'" Thus, L-selectin may be important in leukocyte-endothelia1 interactions apart from its role in mediating leukocyte rolling. Similar results have been obtained with the L-selectin ligand GlyCAM-1 (above). Because GlyCAM-1 appears to be a secreted molecule, attention has focused on its potential role in other events related





to adhesion to HEV. Recent studies by Rosen et a1 have shown that soluble GlyCAM-1 can potently and rapidly activate both PI and 82 integrins on T cells (Steven D. Rosen, University of California at San Francisco, personal communication, April 1996). Interestingly, this induction of integrin activity is far stronger on T cells than on neutrophils, raising the possibility that GlyCAM-1 serves as a lineage- and tissue-specific adhesion regulator important in normal lymphocyte recirculation. The importance of these observations lies in the observation of a physiologic ligand of L-selectin activating integrins within a timescale relevant to leukocyte recognition in the vasculature.

An array of related observations implicates PSGL-1 in signal transduction, especially via P-selectin on activated platelets binding to PSGL- 1 on monocytes. This interaction induces superoxide anion release, enhances PAF synthesis and phagocytosis, and induces the expression of tissue fac-

In addition, monocyte binding of activated platelets through P-selectin/PSGL- 1 interactions costimulates secretion of MCP-1 and TNF-a, and is associated with the nuclear translocation of N F - K B , ~ ” . ~ ~ ~ a widely expressed transcription factor involved in the expression of numerous inducible genes involved in inflammati~n.’’~’~~ These events are inhibited by MoAb to P-selectin. Although the precise signal transduction pathways responsible for these effects have yet to be elucidated, the data are persuasive that P- selectin binding to PSGL-1 induces or amplifies a number of monocyte responses in settings of inflammation or vascular injury.

tor. 136.319.320

BIOPHYSICS OF SELECTIN-BASED ADHESION

How is it that selectins can mediate leukocyte rolling? Work by several groups has shown some fundamental properties that are likely to underlie the ability of selectins to mediate this unique form of intercellular interaction. It is clear that one property exhibited by selectins which allows for this dynamic type of adhesion is fast on rates and off rates.324 In this context, it must be emphasized that the on rates and off rates referred to here are cellular rates mediated by these molecules, and not molecular kinetic values more appropriate to estimations of true thermodynamic affinity. Selectins are often referred to as mediating “low-affinity” interactions, as compared to, eg, integrins, but this is mis- leading; as pointed out,”’ affinity per se is meaningful only at equilibrium, which is obviously not the situation that occurs with molecules on apposing cell surfaces. The intrinsic affinity of selectins for their ligands, when measured in solution, can in principle be relatively constant within a wide range of on and off rates, and is therefore unlikely to accurately predict their kinetic properties, or their biological behavior when expressed on cell surfaces.

The ability to initiate rolling, often referred to as “capture” or “tethering,” appears to be most closely related to on rate, and is relatively insensitive to off rate. L-selectin is more efficient than the vascular selectins at initiating rolling, because its intrinsic cellular on rate appears to be higher than those of the vascular selectins. In vivo, initiation of recognition of endothelium, rather than rolling per se, may be the

most important function of L-selectin, with the next phases of interaction mediated first by E- and/or P-selectin (for neutrophils) or VLA-4 (for lymphocytes and monocytes), and subsequently by p2 i n t e g r i n ~ . “ ~ ~ ~ ~ ~ ” ~ ~ ~ ~ ~ The inability of neutrophils to use VLA-4 or other molecules to “bridge” the velocity “gap” between L-selectin mediated rolling and p2 integrin mediated firm adhesion and arrest probably explains the defects observed in E-selectin/P-selectin double-knockout mice, where neutrophil rolling and recruitment is drastically affected, with mononuclear cell accumulation at later timepoints affected somewhat less severely (Fig 5).

In contrast, actual rolling velocity is determined principally by the off rate: the faster the off rate, the higher the rolling velocity. Again, L-selectin exhibits the fastest off rate, and has the correspondingly fastest rolling velocity, whereas E- selectin appears to have the slowest velocity. The higher off rate of L-selectin may be due in part to shedding of L-selectin during interaction with ligand, as inhibition of shedding by pretreatment of neutrophils with a new class of hydroxamic acid-based metalloprotease inhibitors lowers the velocity of neutrophil rolling in vitro,’2h although this hypothesis has not yet been tested with nonshedding mutants of L-selectin. These differences in off rate between selectins are reflected in bond lifetimes: the shorter the lifetime of the selectin-ligand bond, the higher the off rate, and the faster the velocity. Each of these parameters appears to be essentially a different manifes- tation of the same fundamental property of selectins. It is of interest that the hierarchy of on rates and tetherindcapture efficiency appears to parallel that of off rates and velocities, with L-selectin exhibiting the fastest velocity and the most efficient capture rate, and E-selectin exhibiting the slowest velocity and weakest capture rate.~.6y.98.1“.154.327 Although there is no necessary biophysical reason for these parameters to go in parallel, one result is to ensure overlap over a wide range of velocities, such as are found in vivo, which span the range from freely flowing leukocytes traveling at > 1,000 pm per second to leukocytes rolling on E-selectin at velocities as low as 2 to 3 pm per second, a velocity which typically precedes arrest.

Apart from on rates and off rates, one property thought to be important to the ability of selectins to mediate rolling is resistance to an increase in the off rate with increasing force applied to the b ~ n d . ~ * * ~ ’ ~ ” This property has been termed “tensile strength” by one group.’28 Measurements of the properties of bonds formed between P-selectin and its ligand are consistent with this concept, and may require constraints on bond A resistance to increased off rate with increasing bond stress may be important for rolling, because without this property, selectin bonds would have vanishly short lifetimes, lower than the minimum required to maintain rolling (as bonds on the trailing edge of the cell break, there would be no time for new ones on the leading edge to form). As discussed,328 this property may distinguish selectins from other molecules, eg, certain cell- bound antibodies and hapten^,'^' which have on and off rates in the range found for selectins, but cannot mediate rolling. The ability to maintain resistance to increased off rate with increasing force (both stress and strain324) on the bond may





Fig 5. Distinct steps in leukocyte-endothelial recognition involve distinct velocity ranges and distinct adhesion molecules. L-selectin predominates in capture, with P-selectin contributing significantly, and E-selectin relatively poor at capture, especially at shear forces and velocities seen in vivo. Fast rolling will be mediated by L-selectin, with slower rolling mediated by P-selectin. The dowest rolling is mediated by E-selectin, or for mononuclear cells, by the a4 integrins VLA-4 or a4f37. Neutrophils must use E- or P-selectin to roll slowly enough for the 82 integrins to mediate stable arrest; hence, neutrophils cannot escape the circulation in E/P-selectin double- knockout mice. In contrast, monocytes (and eosino- phils, not pictured) and subsets of memory T calls can use E-selectin, P-selectin, and VLA-4; in addition, these cells can use both p1 and p2 integrins for arrest. Naive T-cell homing to lymph nodes via L-selectin is probably made more efficient by the lower shear forces present in HEV and the higher density of L-selectin ligand. See text for more details.

Neutrophils

L>P>E

L>P

P < E

82 integrins

be a common property of other molecules which can mediate leukocyte rolling as well, eg, VLA-42w and perhaps CD44.332 In addition, the requirement for interaction of L-selectin with the cytoskeleton*'" may be related both to the generation of this tensile strength, as well as the ability of the molecule to resist extraction from the membrane.

An interesting aspect of the response of selectin-ligand bonds to shear stress is the paradoxical loss of adhesion at very low shear. This requirement for a threshhold hydrodynamic shear is most easily demonstrated for L - ~ e l e c t i n , ~ ~ ~ but can be demonstrated for the other selectins as well, at lower levels of shear force and ligand density than for L- selectin (Lawrence M, et al, submitted). This property is likely related to the shorter bond lifetime of L-selectin, because very low shear may be insufficient to promote new bond formation on the leading edge of the cell before existing bonds break. This difference between L-selectin and E- or P-selectin may partially explain why L-selectin works poorly in static adhesion assays, compared with E- and P-selectin. A threshhold shear requirement may serve to inhibit leukocyte accumulation in situations of abnormally low shear in the vasculature.

OTHER FUNCTIONS FOR SELECTINS

Several recent reports suggest that selectins may be involved in other biologic phenomena apart from their well defined role in inflammation. Nguyen et a1334 showed that tube formation by bovine capillary endothelial cells, a well- established in vitro model of angiogenesis, could be inhibited by MoAbs against sLex or sLea, or by polyclonal antibody directed against bovine E - ~ e l e c t i n . ~ ~ ~ Inhibition of tube formation was also observed with an MoAb against E-selectin which essentially completely blocks leukocyte adhesion (Joyce Bischoff, personal communication, April 1996). In addition, E-selectin mRNA was sharply upregulated in capillary endothelial cells undergoing tube formation.334 These results suggest that E-selectin may be involved in some types

Monocyte8

b R E

L>P

P < E and

a4 inkgrins

82 inkgrins +

81 integrins

Naive T cells

L-selectin

t I

L-selectin

gZ > pl integrins

M m o y T cells

L > b E (subsets) Clohlrr/tcthennP

L>P (subsets)

82 integrins

81 intgrins + brrept

of angiogenesis. Consistent with this hypothesis, E-selectin is expressed at very high levels on hemangiomas, an endothelial tumor of childhood, during the proliferative phase of the disease,335 and expression of E-selectin in this disease closely correlates with endothelial proliferation. E-selectin was also detected on dividing endothelial cells in placenta and foreskin. Although the viability and fertility of E-selectin knockout micela show that E-selectin is not essential for angiogenesis during development, this does not rule out a role for E-selectin in other settings of angiogenesis, such as that occurring in the adult animal during wound healing, metastasis, endometrial proliferation or chronic inflammatory disorder^.^^^,^^' The expression of E-selectin on most proliferating endothelium in vitro61,156.335.338 is also strongly suggestive of a biological function unrelated to leukocyte adhesion.

Selectins may play a role in hematopoiesis. L-selectin is expressed on the majority of hematopoietic stem cells," although its function on these cells is unclear. No hematologic defects have been identified in L-selectin knockout mice.- Hematopoietic stem cells bind to P - ~ e l e c t i n , ~ ~ ~ although the expression of P-selectin on BM endothelium has not been examined. In contrast, E-selectin is constitutively expressed on endothelial cells of hematopoietic tissues, and supports binding of hematopoietic progenitor cells.338 Thus, hematopoietic stem cells express L-selectin and functional ligands for both E- and P-selectin, and can therefore potentially use all three selectins. Although single-selectin knockout mice do not exhibit any obvious hematologic defects, this issue has not yet been closely examined. It is possible that some hematologic defects may be uncovered in circum- stances of severe hematopoietic stress, such as chronic infection, where high levels of granulocyte production are typically observed, sometimes in extramedullary sites such as spleen or liver. Additionally, defects in stem cell traffic might be found, especially in settings in which hematopoietic stem cells have been mobilized into the circulation by systemic





cytokine treatment,340 or during BM transplantation. Whether sublethally irradiated E- or P-selectin knockout animals can be rescued by BM transplantation, or whether L-selectin- deficient BM cells can rescue normal, sublethally irradiated animals, has not yet been determined.

CONCLUDING REMARKS

It is evident that the study of selectins and their ligands is far from over. Although rapid progress has been and continues to be made, a number of important issues remain. A partial list of questions for the future would include: the molecular basis for sorting of L-selectin, PSGL-I, and VLA- 4 to microvilli; identification of the signal transduction pathways and cytoplasmic effectors engaged by L-selectin and PSGL- 1 ; the identification and characterization of glycoprotein ligands for E-selectin and L-selectin; the biochemical basis, mechanisms of regulation, and functional importance of interactions between selectins and the cytoskeleton; the role of broadly expressed selectin ligand proteins on cells outside the vascular system; the role, if any, of secreted forms of the selectins; a fuller understanding of the biophysics which underlie selectin-based cell adhesion; a more complete description of the role of specific glycosyltransferases in construction of selectin ligands, and the mechanisms by which their expression is regulated; and the potential role of selectins in biological events unrelated to leukocyte recruitment. It can be safely anticipated that rapid progress on these and related questions will be achieved in the future.

ACKNOWLEDGMENT

I would like to acknowledge numerous colleagues for insightful discussions and provision of manuscripts and data before publica- tion. Apologies to those whose work was not cited due to space considerations.

REFERENCES

1. Kansas GS, Wood GS, Engleman EG: Maturational and functional diversity of human B lymphocytes delineated with anti-Leu- 8. J Immunol 134:3003, 1985

2. Kansas GS, Wood GS, Fishwild DM, Engleman EG: Func- tional characterization of human T lymphocyte subsets distinguished by monoclonal anti-Leu-8. J Immunol 134:2995, 1985

3. Gallatin WM, Weissman IL, Butcher EC: A cell-surface molecule involved in organ-specific homing of lymphocytes. Nature 304:30, 1983

4. Lewinsohn DM, Bargatze RF, Butcher EC: Leukocyte-endo- thelia1 cell recognition: Evidence of a common molecular mechanism shared by neutrophils, lymphocytes, and other leukocytes. J Immunol 138:4313, 1987

5. Bradley LM, Atkins GG, Swain SL: Long-term CD4+ memory T cells from the spleen lack MEL-14, the lymph node homing receptor. J Immunol 148:324, 1992

6. Bradley LM, Duncan DD, Tonkonogy S, Swain SL: Character- ization of antigen-specific CD4+ effector T cells in vivo: Immuniza- tion results in a transient population of MEL-14-, CD45RB- helper cells that secretes interleukin 2 (IL-2), IL-3. IL-4, and interferon gamma. J Exp Med 174:547, 1991

7. Picker LJ, Treer JR, Ferguson-Damell B, Collins PA, Buck D, Terstappen LW: Control of lymphocyte recirculation in man. I. Differential regulation of the peripheral lymph node homing receptor

L-selection on T cells during the virgin to memory cell transition. J Immunol 150:1105, 1993

8. Tsuji T, Nibu R, Iwai K, Kanegane H, Yachie A, Seki H, Miyawaki T, Taniguchi N: Efficient induction of immunoglobulin production in neonatal naive B cells by memory CD4+ T cell subset expressing homing receptor L-selectin. J Immunol 152:4417, 1994

9. Picker LJ, Treer JR, Ferguson-Damell B, Collins PA, Berg- stresser PR, Terstappen LW: Control of lymphocyte recirculation in man. 11. Differential regulation of the cutaneous lymphocyte- associated antigen, a tissue-selective homing receptor for skin-homing T cells. J Immunol 150:1122, 1993

10. Griffin JD, Spertini 0, Emst TJ, Belvin MP, Levine HB, Kanakura Y, Tedder TF: GM-CSF and other cytokines regulate surface expression of the leukocyte adhesion molecule-I on human neutrophils, monocytes, and their precursors. J Immunol 145:576. 1990

1 1 . Kansas GS, Muirhead MJ, Dailey MO: Expression of the CDI 1/CD18, LAM-I, and CD44 adhesion molecules during myeloid and erythroid differentiation in man. Blood 76:2483, 1990

12. Kansas GS, Dailey MO: Expression of adhesion structures during B cell development in man. J Immunol 142:3058, 1989

13. Reichert RA, Gallatin WM, Weissman IL, Butcher EC: Ger- minal center B cells lack homing receptors necessary for normal lymphocyte recirculation. J Exp Med 157:813, 1983

14. Jung TM, Gallatin WM, Weissman IL, Dailey MO: Down- regulation of homing receptors after T cell activation. J lmmunol 141:41 IO, 1988

15. Jung TM, Dailey MO: Rapid modulation of homing receptors induced by activators of protein kinase C. Receptor shed-

ding due to accelerated proteolytic cleavage at the cell surface. J Immunol 144:3130, 1990

16. Kishimoto TK, Julita MA, Berg EL, Butcher EC: Neutrophil Mac- I and MEL-14 adhesion proteins inversely regulated by chemotactic factors. Science 245: 1238, 1989

17. Berg M, James SP: Human neutrophils release the Leu-8 lymph node homing receptor during cell activation. Blood 76:2381, 1990

18. Schleiffenbaum BE, Spertini 0, Tedder TF: Soluble L-selectin is present in human plasma at high levels and retains functional activity. J Cell Biol 119:229, 1992

19. Kahn J, Ingraham RH, Shirley F, Migaki GI, Kishimoto TK: Membrane proximal cleavage of L-selectin: Identification of the cleavage site and a 6-M) transmembrane peptide fragment of L- selectin. J Cell Biol 125:461, 1994

20. Migaki GI, Kahn J, Kishimoto TK: Mutational analysis of the membrane proximal cleavage site of L-selectin: Relaxed sequence specificity surrounding the cleavage site. J Exp Med 182:549, 1995

2 I . Chen A, Engel P, Tedder TF: Structural requirements regulate endoproteolytic release of the L-selectin (CD62L) adhesion receptor from the cell surface of leukocytes. J Exp Med 182:5 19, 1995

22. Bevilacqua MP, Stengelin S, Gimbrone MA Jr, Seed B: Endo- thelial leukocyte adhesion molecule I : An inducible receptor for neutrophils related to complement regulatory proteins and lectins. Science 243: 1 160, 1989

23. Gamble JR, Khew-Goodall Y, Vadas MA: Transforming growth factor-beta inhibits E-selectin expression on human endothelial cells. J Immunol 150:4494, 1993

24. Haraldsen G, Kvale D, Lien B, Farstad IN, Brandtzaeg P: Cytokine-regulated expression of E-selectin, intercellular adhesion molecule-l (ICAM- I) , and vascular cell adhesion molecule- 1 (VCAM- 1 ) in human intestinal microvascular endothelial cells. J Immunol 156:2558, 1996

25. Osborn L, Hession C, Tizard R, Vassallo C, Luhowskyj S, Chi-Rosso G, Lobb R: Direct expression cloning of vascular cell

(gp90Mel-14) .





adhesion molecule-1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell 59:1203, 1989

26. Wertheimer SJ, Myers CL, Wallace RW, Parks T P Intercellu- lar adhesion molecule-1 gene expression in human endothelial cells. J Biol Chem 267:12030, 1992

27. Cotran RS, Gimbrone MA Jr, Bevilacqua MP, Mendrick DL, Pober JS: Induction and detection of a human endothelial activation antigen in vivo. J Exp Med 164:661, 1986

28. Munro JM, Pober JS, Cotran RS: Tumor necrosis factor and interferon-? induce distinct patterns of endothelial activation and associated leukocyte accumulation in skin of papio anubis. Am J Pathol 135:121, 1989

29. Picker LJ, Kishimoto TK, Smith CW, Warnock RA, Butcher EC: ELAM-1 is an adhesion molecule for skin-homing T cells. Nature 349:796, 1991

30. Chu W, Presky DH, Swerlick RA, Bums DK: Alternatively processed human E-selectin transcripts linked to chronic expression of E-selectin in vivo. J Immunol 153:4179, 1994

31. Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T: Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J 93399, 1995

32. Read MA, Whitley MZ, Williams AJ, Collins T NF-KB and IKBW An inducible regulatory system in endothelial activation. J Exp Med 179:503, 1994

33. Thanos D, Maniatis T: NF-KB: A lesson in family values. Cell 80:529, 1995

34. Beg AA, Baldwin AS: The IKB proteins: Multifunctional regulators of Rel/NF-KB transcription factors. Genes Dev 7:2064, 1993

35. Read MA, Neish AS, Luscinskas FW, Palombella VJ, Ma- niatis T, Collins T: The proteosome pathway is required for cytokine- induced endothelial-leukocyte adhesion molecule expression. Immu- nity 2:493, 1995

36. Schindler U, Baichwal VR: Three NF-kappa B binding sites in the human E-selectin gene required for maximal tumor necrosis factor alpha-induced expression. Mol Cell Biol 14:5820, 1994

37. Lewis H, Kaszubska W, DeLamarter JF, Whelan J: Coopera- tivity between two NF-kappa B complexes, mediated by high-mobility-group protein I(Y), is essential for cytokine-induced expression of the E-selectin promoter. Mol Cell Biol 14:5701, 1994

38. Whitley MZ, Thanos D, Read MA, Maniatis T, Collins T: A striking similarity in the organization of the E-selectin and beta interferon gene promoters. Mol Cell Biol 14:6464, 1994

39. Ghersa P, van Huijsduijnen RH, Whelan J, DeLamarter F: Labile proteins play a dual role in the control of endothelial leukocyte adhesion molecule-1 (ELAM-I) gene regulation. J Biol Chem 267:19226, 1992

40. Smeets EF, de Vries T, Leeuwenberg JF, van den Eijnden DH, Buurman WA, Neefjes JJ: Phosphorylation of surface E-selectin and the effect of soluble ligand (sialyl Lewisx) on the half-life of E-selectin. Eur J Immunol 23:147, 1993

41. von Asmuth EJU, Smeets EF, Ginsel LA, Onderwater JJM, Leeuwenberg JFM, Buurman WA: Evidence for endocytosis of E- selectin in human endothelial cells. Eur J Immunol 22:2519, 1992

42. Subramaniam M, Koedam JA, Wagner DD: Divergent fates of P- and E-selectins after their expression on the plasma membrane. Mol Biol Cell 4:791, 1993

43. McEver RP, Beckstead JH, Moore KL, Marshal-Carlson L, Bainton DF: GMP-140, a platelet alpha granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest 84:92, 1989

44. Berman CL, Yeo EL, Wencel-Drake JD, Furie BC, Ginsberg MH, Furie B: A platelet alpha granule membrane protein that is

associated with the plasma membrane after activation. J Clin Invest 78:130, 1986

45. Hsu-Lin S-C, Berman C, Furie B, August D, Furie B: A platelet membrane protein expressed during platelet activation and secretion. J Biol Chem 259:9121, 1984

46. Foreman KE, Vaporciyan AA, Bonish BK, Jones ML, John- son KJ, Glovsky MM, Eddy SM, Ward PA: C5a-induced expression of P-selectin in endothelial cells. J Clin Invest 94: 1147, 1994

47. Wagner DD, Olmsted JB, Marder VJ: Immunolocalization of von Willebrand protein in Weibel-Palade bodies of human endothelial cells. J Cell Biol 95:355, 1982

48. Disdier M, Momssey JH, Fugate RD, Bainton DF, McEver RP: Cytoplasmic domain of P-selectin (CD62) contains the signal for sorting into the regulated secretory pathway. Mol Biol Cell 3:309, 1992

49. Koedam JA, Cramer EM, Briend E, Furie B, Furie BC, Wagner DA: P-selectin, a granule membrane protein of platelets and endothelial cells, follows the regulated secretory pathway in AtT- 20 cells. J Cell Biol 116:617, 1992

50. Green SA, Setiadi H, McEver RP, Kelly RB: The cytoplasmic domain of P-selectin contains a sorting determinant that mediates rapid degradation in lysosomes. J Cell Biol 124:435, 1994

51. Dunlop LC, Skinner MP, Bendall LJ, Favaloro ET, Castaldi PA, Gorman JJ, Gamble JR, Vadas MA, Berndt MC: Characteriza- tion of GMP-140 (P-selectin) as a circulating plasma protein. J Exp Med 175:1147, 1992

52. Johnston GI, Cook RG, McEver R P Cloning of GMP-140, a granule membrane protein of platelets and endothelium. Cell 56:1033, 1989

53. Johnston GI, Bliss GA, Newman PJ, McEver RP: Genomic structure of GMP-140, a member of the selectin family of adhesion receptors for leukocytes. J Biol Chem 34:21381, 1990

54. Crovello CS, Furie BC, Furie B: Rapid phosphorylation and selective dephosphorylation of P-selectin accompanies platelet activation. J Biol Chem 268:14590, 1993

55. Fujimoto T, McEver RP: The cytoplasmic domain of P-selectin is phosphorylated on serine and threonine residues. Blood 82:1758, 1993

56. Modderman PW, von dem Bome AE, Sonnenberg A: Tyro- sine phosphorylation of P-selectin in intact platelets and in a disul- phide-linked complex with immunoprecipitated pp6Oc-src. Biochem J 299:613, 1994

57. Crovello CS, Furie BC, Furie B: Histidine phosphorylation of P-selectin upon stimulation of human platelets: A novel pathway for activation-dependent signal transduction. Cell 82:279, 1995

58. Fujimoto T, Stroud E, Whatley R, Prescott SM, Muszbek L, Laposata M, McEver RP: P-selectin is acylated with palmitic acid and stearic acid at cysteine 766 through a thioester linkage. J Biol Chem 268:11394, 1993

59. Gotsch U, Jager U, Dominis M, Vestweber D: Expression of P-selectin on endothelial cells is upregulated by LPS and TNF-alpha in vivo. Cell Adhesion Commun 2:7, 1994

60. Sanders WE, Wilson RW, Ballantyne CM, Beaudet AL: Mo- lecular cloning and analysis of in vivo expression of murine P- selectin. Blood 80:795, 1992

61. Bischoff J, Brasel C: Regulation of P-selectin by tumor necrosis factor-alpha. Biochem Biophys Res Commun 210:174, 1995

62. Butcher EC: Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 67:1033, 1991

63. Springer TA: Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76:301, 1994 64. Atherton A, Bom GVR: Quantitative investigation of the ad-

hesiveness of circulating polymorphonuclear leukocytes to blood vessel walls. J Physiol 222:447, 1972





65. Ley K, Gaehtgens P, Fennie C, Singer MS, Lasky LA, Rosen SD: Lectin-like cell adhesion molecule 1 mediates leukocyte rolling in mesenteric venules in vivo. Blood 77:2553, 1991

66. Ley K, Bullard D, Arbones ML, Bosse R, Vestweber D, Tedder TF, Beaudet AL: Sequential contribution of L- and P-selectin to leukocyte rolling in vivo. J Exp Med 181:669, 1995

67. Dore M, Korthuis RJ, Granger DN, Entman ML, Smith CW: P-selectin mediates spontaneous leukocyte rolling in vivo. Blood 82:1308, 1993

68. Lawrence MB, Springer TA: Leukocytes roll on a selectin at physiologic flow rates: Distinction from and prerequisite for adhesion through integrins. Cell 65359, 1991

69. Lawrence MB, Springer TA: Neutrophils roll on E-selectin. J Immunol 151:6338, 1993

70. Abbassi 0, Lane CL, Krater S, Kishimoto TK, Anderson DC, McIntire LV, Smith CW: Canine neutrophil margination mediated by lectin adhesion molecule-1 in vitro. J Immunol 147:2107, 1991

71. Abbassi 0, Kishimoto TK, McIntire LV, Anderson DC, Smith CW: E-selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 92:2719, 1993

72. Alon R, Rossiter H, Wang X, Springer TA, Kupper TS: Dis- tinct cell surface ligands mediate T lymphocyte attachment and rolling on P and E-selectin under physiological flow. J Cell Biol 127:1485, 1994

73. Jones DA, McIntire LV, Smith CW, Picker LJ: A two-step adhesion cascade for T celVendothelial cell interactions under flow conditions. J Clin Invest 94:2443, 1994

74. Luscinskas FW, Ding H, Lichtman AH: P-selectin and vascular cell adhesion molecule 1 mediate rolling and arrest, respectively, of CD4+ T lymphocytes on tumor necrosis factor-a activated vascular endothelium under flow. J Exp Med 181:1179, 1995

75. Kubes P, Kanwar S: Histamine induces leukocyte rolling in postcapillary venules. A P-selectin-mediated event. J Immunol 152:3570, 1994

76. Stamper HB Jr, Woodruff JJ: Lymphocyte homing into lymph nodes: In vitro demonstration of the selective affinity of recirculating lymphocytes for high-endothelial venules. J Exp Med 144:828, 1976

77. Kishimoto TK, Jutila MA, Butcher EC: Identification of a human peripheral lymph node homing receptor: A rapidly downregulated adhesion molecule. Proc Natl Acad Sci USA 87:2244, 1990

78. Spertini 0, Kansas GS, Reimann KA, Mackay CR, Tedder TF: Functional and evolutionary conservation of distinct epitopes on the leukocyte adhesion molecule-I (LAM- 1) that regulate leukocyte migration. J Immunol 147:942, 1991

79. Hou S, Hyland L, Bradley LM, Watson SR, Doherty PC: Subverting lymph node trafficking by treatment with the Mel-14 monoclonal antibody to L-selectin does not prevent an effective host response to Sendai virus. J Immunol 155:252, 1995

80. Navarro RF, Jalkanen ST, Hsu M, Soenderstrup-Hansen G, Goronzy J, Weyand C, Fathman CG, Clayberger C, Krensky AM, Butcher EC: Human T cell clones express functional homing receptors required for normal lymphocyte trafficking. J Exp Med 162:1075, 1985

8 I . Dailey MO, Fathman CG, Butcher EC, Pillemer E, Weissman I: Abnormal migration of T lymphocyte clones. J Immunol 128:2 134, 1982

82. Kansas GS, Ley K, Munro JM, Tedder TF: Regulation of leukocyte rolling and adhesion to HEV through the cytoplasmic domain of L-selectin. J Exp Med 177:833, 1993

83. Bradley LM, Watson SR, Swain SL: Entry of naive CD4 T cells into peripheral lymph nodes requires L-selectin. J Exp Med 180:2401, 1994

84. Arbones ML, Ord DC, Ley K, Ratech H, Maynard-Curry C,

Otten G, Capon DJ, Tedder TF: Lymphocyte homing and leukocyte rolling and migration are impaired in L-selectin (CD62L) deficient mice. Immunity 1:247, 1994

85. Hamann A, Jablonski-Westrich D, Duijvestijn A, Butcher EC, Baisch H, Harder R, Thiele H-G: Evidence for an accessory role of LFA-I in lymphocyte-high endothelium interaction during homing. J Immunol 140:693, 1988

86. Jutila MA, Rott L, Berg EL, Butcher EC: Function and regulation of the neutrophil MEL-14 antigen in vivo: comparison with LFA-I and MAC-I. J Immunol 143:3318, 1989

87. Rothlein R, Dustin ML, Marlin SD, Springer TA: A human intercellular adhesion molecule (ICAM-I ) distinct from LFA- I , J Immunol 137: 1270, 1986

88. Dustin ML, Springer TA: T-cell receptor cross-linking transiently stimulates adhesiveness through LFA-I. Nature 341:619, I989

89. van Kooyk Y, van de Wiel-van Kemenade P, Weder P, Kuijp- ers TW, Figdor CG: Enhancement of LFA-1-mediated cell adhesion by triggering through CD2 or CD3 on T lymphocytes. Nature 342:811, 1989

90. Stoolman LM, Ebling H: Adhesion molecules of cultured hematopoietic malignancies. A calcium-dependent lectin is the principle mediator of binding to the high endothelial venule of lymph nodes. J Clin Invest 84:1196, 1989

9 I . Hamann A, Jablonski-Westrich D, Jonas P, Thiele H: Homing receptors reexamined: Mouse LECAM-I (MEL-14 antigen) is involved in lymphocyte migration into gut-associated lymphoid tissue. Eur J Immunol 21:2925, 1991

92. Hamann A, Andrew DP, Jablonski-Westrich D, Holzmann B, Butcher EC: Role of a 4 integrins in lymphocyte homing to mucosal tissues in vivo. J Immunol 152:3282, 1994

93. Bargatze RF, Jutila MA, Butcher EC: Distinct roles of L- selectin and integrins a4p7 and LFA-1 in lymphocyte homing to Peyer’s patch HEV in situ: The multistep model confirmed and refined. Immunity 3:99, 1995

94. Ley K, Pries AR, Gaehtgens P: A versatile intravital microscope design. Int J Microcirc Clin Exp 6:161, 1987

95. Ley K, Gaehtgens P: Endothelial, not hemodynamic differences are responsible for preferential leukocyte rolling in rat mesenteric venules. Circ Res 69:1034, 1991

96. von Andrian UH, Chambers JD, McEvoy LM, Bargatze RF, Arfors K-E, Butcher EC: Two-step model of leukocyte-endothelial cell interaction in inflammation: Distince roles for LECAM-I and the leukocyte p 2 integrins in vivo. Proc Natl Acad Sci USA 88:7538, 1991

97. Luscinskas FW, Kansas GS, Ding H, Pizcueta P, Schlieffen- baum B, Tedder TF, Gimbrone MA: Monocyte rolling, arrest and spreading on IL-4 activated vascular endothelium under flow is mediated via sequential action of L-selectin, 01 and p 2 integrins. J Cell Biol 125:1417, 1994

98. Ley K, Tedder TF, Kansas GS: L-selectin can mediate rolling independent of E- and P-selectin in mesenteric venules in vivo. Blood 82:1632, 1993

99. von Andrian UH, Hansel1 P, Chambers JD, Berger E, Torres- Filho I, Butcher EC, Arfors K-E: L-selectin function is required for 0 2 integrin-mediated neutrophil adhesion at physiologic shear rates in vivo. Am J Physiol 262:H2320, 1992

100. von Andrian UH, Chambers JD, Berg EL, Michie SA, Brown DA, Karolak D, Ramezani L, Berger EM, Arfors K-E, Butcher EC: L-selectin mediates neutrophil rolling in inflamed venules through sialyl Lewis-X dependent and -independent recognition pathways. Blood 82:182, 1993

101. Luscinskas FW, Ding H, Tan P, Cumming D, Tedder TF, Gerritsen ME: L- and P-selectins but not VLA-4 ( a 4 p l ) integrins





mediate monocyte initial attachment to TNF-a activated endothelium under flow in vitro. J Immunol 157:326, 1996

102. Bevilacqua MP, Pober JS, Mendrick DL, Cotran RS, Gim- brone MA Jr: Identification of an inducible endothelial-leukocyte adhesion molecule. Proc Natl Acad Sci USA 84:9238, 1987

103. Spertini 0, Luscinskas FW, Munro JM, Kansas GS, Griffin JD, Gimbrone MA Jr, Tedder TF: Leukocyte adhesion molecule-1 (LAM-1, L-selectin) interacts with an inducible endothelial cell ligand to support leukocyte adhesion and transmigration. J Immunol 147:2565, 1991

104. Alon R, Kassner PD, Carr MW, Finger EB, Hemler ME, Springer TA: The integrin VLA-4 supports tethering and rolling in flow on VCAM-1. J Cell Biol 128:1243, 1995

105. Thorlacius H, Raud J, Rosengren-Beezley S, Forrest MJ, Hedqvist P, Lindbom L: Mast cell activation induces P-selectin- dependent leukocyte rolling and adhesion in postcapillary venules in vivo. Biochem Biophys Res Commun 203:1043, 1994

106. Ley K, Zakrzewicz A, Hanski C, Stoolman LM, Kansas GS: Sialylated 0-glycans and L-selectin on myeloid cells sequentially mediate rolling in vivo. Blood 85:3727, 1995

107. Bosse R, Vestweber D: Only simultaneous blocking of the L- and P-selectin completely inhibits neutrophil migration into mouse peritoneum. Eur J Immunol 24:3019, 1994

108. Watson SR, Fennie C, Lasky LA: Neutrophil influx into an inflammatory site inhibited by a soluble homing receptor-IgG chimera. Nature 349:164, 1991

109. Pizcueta P, Luscinskas FW: Monoclonal antibody blockade of L-selectin inhibits mononuclear leukocyte recruitment to inflammatory sites in vivo. Am J. Pathol 145:461, 1994

110. Mulligan MS, Miyasaka M, Tamatani T, Jones MJ, Ward PA: Requirement for L-selectin in neutrophil-mediated lung injury in rats. J Immunol 152:832, 1994

11 1. Tedder TF, Steeber DA, Pizcueta P: L-selectin-deficient mice have impaired leukocyte recruitment into inflammatory sites. J Exp Med 181:2259, 1995

112. Kubes P, Jutila M, Payne D: Therapeutic potential of inhib- iting leukocyte rolling in ischemidreperfusion. J Clin Invest 95:2510, 1995

11 3. Mihelcic D, Schleiffenbaum B, Tedder TF, Sharar SR, Har- lan JM, Winn RK. Inhibition of leukocyte L-selectin function with a monoclonal antibody attenuates reperfusion injury to the rabbit ear. Blood 84:2322, 1994

114. Kurose I, Anderson DC, Miyasaka M, Tamatani T, Paulson JC, Todd RF, Rusche JR, Granger DN: Molecular determinants of reperfusion-induced leukocyte adhesion and vascular protein leak- age. Circ Res 74:336, 1994

115. Ma XL, Weyrich AS, Lefer DJ, Buerke M, Albertine KH, Kishimoto TK, Lefer AM: Monoclonal antibody to L-selectin attenuates neutrophil accumulation and protects ischemic reperfused cat myocardium. Circulation 88:649, 1993

116. Bargatze RF, Kurk S, Butcher EC, Jutila MA: Neutrophils roll on adherent neutrophils bound to cytokine-induced endothelial cells via L-selectin on the rolling cells. J Exp Med 180:1785, 1994

117. Oxley SM, Sackstein R: Detection of an L-selectin ligand on a hematopoietic progenitor cell line. Blood 84:3299, 1994

118. Simon SI, Chambers JD, Butcher EC, Sklar LA: Neutrophil aggregation is p2 integrin- and L-selectin-dependent in blood and isolated cells. J Immunol 149:2765, 1992

119. Simon SI, Rochon YP, Lynam EB, Smith CW, Anderson DC, Sklar LA: p2 integrin and L-selectin are obligatory receptors in neutrophil aggregation. Blood 82: 1097, 1993

120. Rochon YP, Simon SI, Lynam EB, Sklar LA: A role for lectin interactions during human neutrophil aggregation. J Immunol 152:1385, 1994

121. Nolte D, Schmid P, Jager U, Botzlar A, Roesken F, Hecht R, Uhl E, Messmer K, Vestweber D: Leukocyte rolling in venules of striated muscle and skin is mediated by P-selectin, not by L- selectin. Am J Physiol 267:H1637, 1994

122. Mayadas TN, Johnson RC, Raybum H, Hynes RO, Wagner DD: Leukocyte rolling and extravasation are severely compromised in P-selectin-deficient mice. Cell 74:541, 1993

123. Jones DA, Abbassi 0, McIntire LV, McEver RP, Smith CW: P-selectin mediates neutrophil rolling on histamine-stimulated endothelial cells. Biophys J 65:1560, 1993

124. Asako H, Kurose I, Wolf R, DeFrees S, Zheng ZL, Phillips ML, Paulson JC, Granger DN: Role of H1 receptors and P-selectin in histamine-induced leukocyte rolling and adhesion in postcapillary venules. J Clin Invest 93:1508, 1994

125. Ley K: Histamine can induce leukocyte rolling in rat mesenteric venules. Am J Physiol 267:H1017, 1994

126. Johnson RC, Mayadas TN, Frenette PS, Mebius RE, Subra- maniam M, Lacasce A, Hynes RO, Wagner DD: Blood cell dynamics in P-selectin deficient mice. Blood 86:1106, 1995

127. Subramaniam S, Saffaripour S, Watson SR, Mayadas TN, Hynes RO, Wagner Dd: Reduced recruitment of inflammatory cells in a contact hypersenstivity response in P-selectin-deficient mice. J Exp Med 181:2277, 1995

128. Grober JS, Bowen BL, Ebling H, Athey B, Thompson CB, Fox DA, Stoolman LM: Monocyte-endothelial adhesion in chronic rheumatoid arthritis. In situ detection of selectin and integrin-dependent interactions. J Clin Invest 91:2609, 1993

129. Mulligan MS, Polley MJ, Bayer RJ, Nunn MF, Paulson JC, Ward PA: Neutrophil dependent acute lung injury. Requirement for P-selectin. J Clin Invest 90:1600, 1992

130. Mulligan MS, Paulson JC, De Frees S, Zheng Z-L, Lowe JB, Ward PA: Protective effects of oligosaccharides in P-selectin- dependent lung injury. Nature 364:149, 1993

131. Bullard DC, Qin L, Lorenzo I, Quinlin WM, Doyle NA, Bosse R, Vestweber D, Doerschuk CM, Beaudet AL: P-selectid ICAM-I double mutant mice: Acute emigration of neutrophils into the peritoneum is completely absent but is normal into pulmonary alveoli. J Clin Invest 95:1782, 1995

132. Larsen E, Celi A, Gilbert GE, Furie BC, Erban JK, Bonfanti R, Wagner DD, Furie B: PADGEM protein: A receptor that mediates the interaction of activated platelets with neutrophils and monocytes. Cell 59:305, 1989

133. Hamburger SA, McEver RP: GMP-140 mediates adhesion of stimulated platelets to neutrophils. Blood 75550, 1990

134. Moore KL, Thompson LF: P-selectin (CD62) binds to subpopulations of human memory T lymphocytes and natural killer cells. Biochem Biophys Res Commun 186: 173, 1992

135. de Bruijne-Admiraal LG, Modderman PW, Von dem Bome AEGK, Sonnenbeg A: P-selectin mediates Ca”-dependent adhesion of activated to many different types of leukocytes: detection by flow cytometry. Blood 80: 134, 1992

136. Celi A, Pellegrini G, Lorenzet R, De Blasi A, Ready N, Furie BC, Furie B: P-selectin induces the expression of tissue factor on monocytes. Proc Natl Acad Sci USA 91:8767, 1994

137. Palabrica T, Lobb R, Furie BC, Aronovitz M, Benjamin C, Hsu Y-M, Sajer SA, Furie B: Leukocyte accumulation promoting fibrin deposition is mediated in vivo by P-selectin on adherent platelets. Nature 359:848, 1992

138. Subramanian M, Frenette PS, Saffaripour S, Johnson RC, Hynes RO, Wagner DD. Defects in hemostasis in P-selectin-deficient mice. Blood 87:1238, 1996

139. Olafsson AM, Arfors K-E, Ramezani L, Wolitzky BA, Butcher EC, von Andrian UH: E-selectin mediates leukocyte rolling in interleukin- 1 treated mesentery venules. Blood 84:2749, 1994





140. Labow MA, Norton CR, Rumberger JM, Lombard-Gilloly KM, Shuster DJ, Huhhard J, Bertko R, Knaack PA, Terry RW, Harbison ML, Kontgen F, Stewart CL, McIntyre KW, Will PC, Burns DK, Wolitzky BA: Characterization of E-selectin-deficient mice: Demonstration of overlapping function of the endothelial selectins. Immunity 1:709, 1994

141. Gundel RH, Wegner CD, Torcellini C, Clarke CC, Haynes N, Rothlein R, Smith CW, Letts LG: Endothelial leukocyte adhesion molecule- 1 mediates antigen-induced acute airway inflammation and late-phase airway obstruction in monkeys. J Clin Invest 88:1407, 1991

142. Mulligan MS, Varani J, Lane CL, Dame MK, Smith CW, Anderson DC, Ward PA: Role of endothelial-leukocyte adhesion molecule (ELAM-I) in neutrophil-mediated lung injury in rats. J Clin Invest 88:1396, 1991

143. Koch AE, Burrows JC, Haines GK, Carlos TM, Harlan JM, Leibovich SJ: Immunolocalization of endothelial and leukocyte adhesion molecules in human rheumatoid and osteoarthritic synovial tissues. Lab Invest 64:313, 1991

144. Allen MD, McDonald TO, Himes VE, Fishhein DP, Aziz S, Reichenback DD: E-selectin expression in human cardiac grafts with cellular rejection. Circulation 88:243, 1993

145. Brockmeyer C, Ulbrecht M, Schendel DJ, Weiss EH, Hille- brand G, Burkhardt K, Land W, Gokel MJ, Riethmuller G, Feucht HE: Distribution of cell adhesion molecules (ICAM-1, VCAM-I, ELAM- 1) in renal tissues during allograft rejection. Transplantation 55:610, 1993

146. Taylor PM, Rose ML, Yacoub MH, Pigott R: Induction of vascular adhesion molecules during rejection of human cardiac allografts. Transplantation 54:45 1, 1992

147. Yan HC, Delisser HM, Pilewski JM, Barone KM, Szklut PJ, Chang XJ, Abern TJ, Langer-Safer P, Albelda SM: Leukocyte recruitment into human skin transplanted onto severe combined immunodeficient mice induced by TNF-alpha is dependent on E-selectin. J Immunol 152:3053, 1994

148. Yan HC, Juhasz I, Pilewski J, Murphy GF, Herlyn M, Albelda SM: Humadsevere combined immunodeficient mouse chi- meras. An experimental in vivo model system to study the regulation of human endothelial cell-leukocyte adhesion molecules. J Clin In- vest 91:986, 1993

149. Silber A, Newman W, Reimann KA, Hendricks E, Walsh D, Ringler DJ: Kinetic expression of endothelial adhesion molecules and relationship to leukocyte recruitment in two cutaneous models of inflammation. Lab Invest 70:163, 1994

150. Picker LJ, Michie SA, Rott LS, Butcher EC: A unique phenotype of skin-associated lymphocytes in humans. Am J Pathol 136:1053, 1990

151. Ley K, Tedder TF: Leukocyte interactions with vascular endothelium. New insights into selectin-mediated attachment and rolling. J Immunol 155525, 1995

152. Kunkel EJ, Jung U, Bullard DC, Norman KE, Wolitzky BA, Vestweher D, Beaudet AL, Ley K: Absence of trauma-induced leukocyte rolling in mice deficient in both P-selectin and intercellular adhesion molecule-1 (ICAM-I). J Exp Med 18357, 1995

153. Jung U, Bullard DC, Tedder TF, Ley K: Velocity differences between L-selectin and P-selectin dependent neutrophil rolling in venules of the mouse cremaster muscle in vivo. Am J Physiol 1996 (in press)

154. Kunkel EJ, Jung U, Ley K: Slow leukocyte rolling in TNF- a-treated mouse cremaster muscle venules is mediated by E-selectin. FASEB J 10:A1280, 1996 (abstr)

155. Bullard DC, Kunkel EJ, Kubo H, Hicks MJ, Lorenzo I, Doyle CA, Doerschuk CM, Ley K, Beaudet AL: Infectious susceptibility and severe deficiency of leukocyte rolling and recruitment in

E-selectin and P-selectin double mutant mice. J Exp Med 183:2329, 1996

156. Frenette PS, Mayads TN, Rayburn H, Hynes RO, Wagner DD: Susceptibility to infection and altered hematopoiesis in mice deficient in both P- and E-selectins. Cell 84563, 1996

157. Lasky LA, Singer MS, Yednock TA, Dowhenko D, Fennie C, Rodriguez H, Nguyen T, Stachel S, Rosen SD: Cloning of a lymphocyte homing receptor reveals a lectin domain. Cell 56:1045. 1989

158. Bowen B, Nguyen T, Lasky LA: Characterization of a human homologue of the murine peripheral lymph node homing receptor. J Cell Biol 109:421, 1989

159. Siegelman MH, van de Rijn M, Weissman IL: Mouse lymph node homing receptor cDNA clone encodes a glycoprotein revealing tandem interaction domains. Science 243: 1 165, 1989

160. Siegelman MH, Weissman IL: Human homologue of mouse lymph node homing receptor: Evolutionary conservation at tandem cell interaction domains. Proc Natl Acad Sci USA 865562, 1989

161. Tedder TF, Emst TJ, Demetrj GD, Isaacs CM, Adler DA, Disteche CM: Isolation and chromosomal localization of cDNAs encoding a novel human lymphocyte cell-surface molecule, LAM 1 : Homology with the mouse lymphocyte homing receptor and other human adhesion proteins. J Exp Med 170:123, 1989

162. Aheam JM, Fearon DT: Structure and function of the complement receptors, CRI (CD35) and CR2 (CD21). Adv Immunol 46:183, 1989

163. Weis JJ, Toothaker LE, Smith JA, Weis JH, Fearon DT: Structure of the human B lymphocyte receptor for C3d and the Epstein-Barr virus and relatedness to other members of the family of C3/C4 binding proteins. J Exp Med 167:1047, 1988

164. Collins T, Williams A, Johnston GI, Kim J, Eddy R, Shows T, Gimbrone MA, Bevilacqua MP: Structure and chromosomal location of the gene for endothelial-leukocyte adhesion molecule-l . J Biol Chem 266:2466, 1991

165. Ord DC, Ernst TJ, Zhou LJ, Ramhaldi A, Spertini 0, Griffin JD, Tedder TF: Structure of the gene encoding the human leukocyte adhesion molecule-I (TQ1, Leu-8) of lymphocytes and neutrophils. J Biol Chem 265:7760, 1990

166. Watson ML, Kingsmore SF, Johnston GI, Siegelman MH, Le Beau MM, Lemons RS, Bora NS, Howard TA, Weissman IL, McEver RP, Seldin MF: Genomic organization of the selectin family of leukocyte adhesion molecules on human and mouse chromosome 1. J Exp Med 172:263, 1990

167. Geoffroy JS, Rosen S: Demonstration that a lectin-like receptor (gp90mel) directly mediates adhesion of lymphocytes to high endothelial venules of lymph nodes. J Cell Biol 109:2463, 1989

168. Imai Y, True DD, Singer MS, Rosen SD: Direct demonstration of the lectin activity of gp90"'. a lymphocyte homing receptor. J Cell Biol 1 11:1225, 1990

169. Stoolman LM, Rosen SD: Possible role of cell-surface carbohydrate-binding molecules in lymphocyte recirculation. J Cell Biol 96:722, 1983

170. Stoolman LM, Yednock TA, Rosen SD: Homing receptors on human and rodent lymphocytes- evidence for a conserved carbohydrate-binding specificity. Blood 70:1842, 1987

171. Stoolman LM, Tenforde TS, Rosen SD: Phosphomannosyl receptors may participate in the adhesive interactions between lymphocytes and high endothelial venules. J Cell Biol 99:1535, 1984

172. Yednock TA, Stoolman LM, Rosen SD: Phosphomannosyl derivatized beads detect a receptor involved in lymphocyte homing. J Cell Biol 104:713, 1987

173. Yednock TA, Butcher EC, Stoolman LM, Rosen SD: Recep- tors involved in lymphocyte homing: Relationship between a carho-




SELECTINS AND THEIR LIGANDS

hydrate-binding receptor and the MEL-14 antigen. J Cell Biol 104:725, 1987

174. Kansas GS, Spertini 0, Stoolman LM, Tedder TF: Molecular mapping of functional domains of the leukocyte receptor for endothelium, LAM-1. J Cell Biol 114:351, 1991

175. Rosen SD, Singer MS, Yednock YA, Stoolman LM: Involvement of sialic acid on endothelial cells in organ-specific lymphocyte recirculation. Science 228: 1005, 1985

176. Rosen SD, Chi SI, True DD, Singer MS, Yednock TA: Intravenously injected sialidase inactivates attachment sites for lymphocytes on high endothelial venules. J Immunol 142:1895, 1989

177. True DD, Singer MS, Lasky LA, Rosen SD: Requirement for sialic acid on the endothelial ligand of a lymphocyte homing receptor. J Cell Biol 11 1:2757, 1990

178. Walz G, Aruffo A, Kolanus W, Bevilacqua M, Seed B: Recognition by ELAM-1 of the Sialyl-Le" determinant on myeloid and tumor cells. Science 250:1132, 1990

179. Phillips ML, Nudelman E, Gaeta FCA, Perez M, Singhal AK, Hakomori S-I, Paulson JC: ELAM-I mediates cell adhesion by recognition of a carbohydrate ligand, Sialyl-Le". Science 250: 1130, 1990

180. Larsen E, Palabrica T, Sajer S , Gilbert GE, Wagner DD, Furie BC, Furie B: PADGEM-dependent adhesion of platelets to monocytes and neutrophils is mediated by a lineage-specific carbohydrate, LNF I11 (CD15). Cell 63:467, 1990

181. Larsen GR, Sako D, Ahem TJ, Shaffer M, Erban J, Sajer SA, Gibson RM, Wagner DD, Furie BC, Furie B: P-selectin and E- selectin. Distinct but overlapping leukocyte ligand specificities. J Biol Chem 267:11104, 1992

182. Zhou Q, Moore KL, Smith DF, Varki A, McEver RP, Cum- mings RD: The selectin GMP-140 binds to sialylated, fucosylated lactosaminoglycans on both myeloid and nonmyeloid cells. J Cell Biol 115:557, 1991

183. Polley MJ, Phillips ML, Wayner E, Nudelman E, Singhal AK, Hakomori S-I, Paulson JC: CD62 and endothelial cell-leukocyte adhesion molecule 1 (ELAM-1) recognize the same carbohydrate ligand, sialyl-Lewis X. Roc Natl Acad Sci USA 88:6224, 1991

184. Foxall C, Watson SR, Dowbenko D, Fennie C, Lasky LA, Kiso M, Hasegawa A, Asa D, Brandley BK: The three members of the selectin receptor family recognize a common carbohydrate epitope, the sialyl lewis" oligosaccharide. J Cell Biol 1172395, 1992

185. Moore KL, Varki A, McEver RF? GMP-140 binds to a glycoprotein receptor on human neutrophils: evidence for a lectin-like interaction. J Cell Biol 112:491, 1991

186. Bevilacqua MP, Nelson RM: Selectins. J Clin Invest 91:379, 1993

187. Bowen BR, Fennie C, Lasky LA: The MEL-14 antibody binds to the lectin domain of the murine peripheral lymph node homing receptor. J Cell Bioi 110:147, 1990

188. Geng JG, Bevilacqua MP, Moore KL, McIntyre TM, Pres- cott SM, Kim JM, Bliss GA, Zimmerman GA, McEver RP: Rapid neutrophil adhesion to activated endothelium mediated by GMP- 140. Nature 343:757, 1990

189. Graves BJ, Crowther RL, Chadran C, Rumberger JM, Li S, Huang K-S, Presky DM, Familetti PC, Wolitzky BA, Bums D: Insight into the E-selectinnigand interaction from the crystal structure and mutagenesis of the lec/EGF domains. Nature 367:532, 1994

190. Weis WI, Drickamer K, Hendrickson WA: Structure of a C-type mannose-binding protein complexed with an oligosaccharide. Nature 360: 127, 1992

191. Erbe DV, Wolitzky BA, Presta LG, Norton CR, Ramos RJ, Bums DK, Rumberger JM, Rao BNN, Foxall C, Brandley BK, Lasky LA: Identification of an E-selectin region critical for carbohydrate recognition and cell adhesion. J Cell Biol 119:215, 1992

192. Erbe DV, Watson SR, Presta LG, Wolitzky BA, Foxall C, Brandley BK, Lasky LA: P- and E-selectin use common sites for carbohydrate ligand recognition and cell adhesion. J Cell Biol 120:1227, 1993

193. Weis WI, Kahn R, Fourme R, Drickamer K, Hendrickson WA: Structure of the calcium-dependent lectin domain from a rat mannose-binding protein determined by MAD phasing. Science 254:1608, 1991

194. Kansas GS, Saunders KB, Ley K, Zakrzewicz A, Gibson RM, Furie BC, Furie B, Tedder TF: A role for the epidermal growth factor-like domain of P-selectin in ligand recognition and cell adhesion. J Cell Biol 124:609, 1994

195. Siegelman MH, Cheng IC, Weissman IL, Wakeland EK: The mouse lymph node homing receptor is identical with the lymphocyte cell surface marker Ly-22: Role of the EGF domain in endothelial binding. Cell 61:611, 1990

196. Pigott R, Needham LA, Edwards RM, Walker C, Power C: Structural and functional studies of the endothelial activation antigen endothelial leukocyte adhesion molecule-1 using a panel of monoclonal antibodies. J Immunol 147:130, 1991

197. Gibson RM, Kansas GS, Tedder TF, Furie B, Furie BC: Lectin and epidermal growth factor domains of P-selectin at physiologic density are the recognition unit for leukocyte binding. Blood 85:1.51, 1995

198. Murphy JF, McGregor JL: Two sites on P-selectin (the lectin and epidermal growth factor-like domains) are involved in the adhesion of monocytes to thrombin-activated endothelial cells. Biochem J 303:619, 1994

199. Jutila MA, Watts G, Walcheck B, Kansas GS: Characteriza- tion of a functionally important and evolutionarily well-conserved epitope mapped to the short consensus repeats of E-selectin and L- selectin. J E X ~ Med 175:1565, 1992

200. Li SH, Bums DK, Rumberger JM, Presky DH, Wilkinson VL, Anostario M Jr, Wolitzky BA, Norton CR, Familletti PC, Kim KJ, Goldstein AL, Cox DC, Huang K-S: Consensus repeat domains of E-selectin enhance ligand binding. J Biol Chem 269:4431, 1994

201. Watson SR, Imai Y, Fennie C, Geoffrey J, Singer M, Rosen SD, Lasky LA: The complement binding-like domains of the murine homing receptor facilitate lectin activity. J Cell Biol 115:235, 1991

202. Ageter-Shaw M, Cole JL, Klickstein LB, Wong WW, Fearon DT, Lalley PA, Weis JH: Expansion of the complement receptor gene family. J Immunol 138:3488, 1987

203. Picker LJ, Wamock RA, Bums AR, Doerschuk CM, Berg EL, Butcher EC: The neutrophil selectin LECAM- 1 presents carbohydrate ligands to the vascular selectins ELAM-I and GMP-140. Cell 66:921, 1991

204. Burns AR, Doerschuk CM: Quantitation of L-selectin and CD18 expression on rabbit neutrophils during CD18-independent and CD18-dependent emigration in the lung. J Immunol 153:3177, 1994

20.5. Erlandsen SL, Hasslen SR, Nelson RD: Detection and spatial distribution of the 0 2 integrin (Mac-I) and L-selectin (LECAM-I) adherence receptors on human neutrophils by high resolution field emission SEM. J Histochem Cytochem 41:327, 1993

206. Hasslen SR, von Andrian UH, Butcher EC, Nelson RD, Erlandsen SL: Spatial distribution of L-selectin (CD62L) on human lymphocytes and transfected murine L1-2 cells. Histochem J 27:547, 1995

207. von Andrian UH, Hasslen SR, Nelson RD, Erlandsen SL, Butcher EC: A central role for microvillous receptor presentation in leukocyte adhesion under flow. Cell 82:989, 1995

208. Moore KL, Patel KD, Breuhl RE, Fugang L, Johnson DL, Lichenstein HS, Cummings RD, Bainton DF, McEver RP: P-selectin





glycoprotein ligand-I mediates rolling of human neutrophils on P- selectin. J Cell Biol 128:661, 1995

209. Berlin C, Bargatze RF, Campbell JJ, von Andrian UH, Szabo MC, Hasslen SR, Nelson RD, Berg EL, Erlandsen SL, Butcher EC: a 4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80:413, 1995

210. Pavalko FM, Walker DM, Graham L, Goheen M, Doerschuk CM, Kansas GS: The cytoplasmic domain of L-selectin interacts with cytoskeletal proteins via a-actinin: Receptor positioning in microvilli does not require interaction with a-actinin. J Cell Biol 129:1155, 1995

21 1. Pavalko FM, Otey CA: Role of adhesion molecule cytoplasmic domains in mediating interactions with the cytoskeleton. Proc SOC Exp Biol Med 205282, 1994

212. Button E, Shapland C, Lawson D: Actin, its associated proteins and metastasis. Cell Motil Cytoskel 30:247, 1995

213. Yoshida M, Westlin WF, Wang N, Ingber DE, Rosenzweig A, Resnick N, Gimbrone MA: Leukocyte adhesion to vascular endothelium induces E-selectin linkage to the actin cytoskeleton. J Cell Biol 133:445, 1996

214. Kansas GS, Pavalko FM: The cytoplasmic domains of E- and P-selectin do not constitutively interact with a-actinin and are not essential for leukocyte adhesion. J Immunol 157:321, 1996

215. Schaller MD, Parsons J T Focal adhesion kmase and associated proteins. Curr Opin Cell Biol 6:705, 1994

216. Zachary I, Rozengurt E: Focal adhesion kinase (p125FAK): A point of convergence in the action of neuropeptides, integrins, and oncogenes. Cell 71:891, 1992

217. Kaplanski G, Farnarier C, Benoliel AM, Foa C, Kaplanski S, Bongrand P: A novel role for E- and P-selectins: Shape control of endothelial cell monolayers. J Cell Sci 107:2449, 1994

218. Varki A: Selectin ligands. Proc Natl Acad Sci USA 91:7390, 1994

219. Imai Y, Lasky LA, Rosen SD: Sulphation requirement for GlyCAM-I, an endothelial ligand for L-selectin. Nature 361555, 1993

220. Pouyani T, Seed B: PSGL-1 recognition of P-selectin is controlled by a tyrosine sulfation consensus at the PSGL-I amino terminus. Cell 83:333, 1995

221. Sako D, Comess KM, Barone KM, Camphausen RT, Cum- ming DA, Shaw GD: A sulfated peptide segment at the amino term- nus of PSGL-1 is critical for P-selectin binding. Cell 83:323, 1995

222. Wilkins PP, Moore KL, McEver RP, Cummings RD: Tyro- sine sulfation of P-selectin glycoprotein ligand-l is required for high affinity binding to P-selectin. J Biol Chem 270:22677, 1995

223. Norgard KE, Moore KL, Diaz S, Stults N, Ushiyama S, McEver RP, Cummings RD, Varki A: Characterization of a specific ligand for P-selectin on myeloid cells. A minor glycoprotein with sialylated 0-linked oligosaccharides. J Biol Chem 268: 12764, 1993

224. Tyrrell D, James P, Rao N, Foxali C, Abbas S, Dasgupta F, Nashed M, Hasegawa A, Kiso M, Asa D, Kidd J, Brandley BK: Structural requirements for the carbohydrate ligand of E-selectin. Roc Natl Acad Sci USA 88:10372, 1991

225. Kukowska-Latallo JF, Larsen RD, Nair RP, Lowe JB: A cloned human cDNA determines expression of a mouse stage-specific embryonic antigen and the Lewis blood group a(1,3/1,4) fucosyltransferase. Genes Dev 4: 1288, 1990

226. Lowe B, Stoolman LM, Rajan PN, Larsen RD, Berhend TL, Marks RM: ELAM- 1 -dependent cell adhesion to vascular endothelium determined by transfected human fucosyltransferase cDNA. Cell 63:475, 1990

227. Goelz SE, Hession C, Goff D, Griffiths B, Tizard R, New- man B, Chi-Rosso G, Lobb R: ELlT: A gene that directs the expression of an ELAM-I ligand. Cell 63:1349, 1990

228. Lowe JB, Kukowska-Latallo JF, Nair RP, Larsen RD, Marks RM, Macher BA, Kelly RJ, Emst LK: Molecular cloning of a human fucosyltransferase gene that determines expression of the Lewis x and VIM-2 epitopes but not ELAM-1-dependent cell adhesion. J Biol Chem 266:17467, 1991

229. Goelz S, Kumar R, Potvin B, Sundaram S, Brickelmaier M, Stanley P: Differential expression of an E-selectin ligand (SLex) by two Chinese hamster ovary cell lines transfected with the same alpha (1,3)-fucosyltransferase gene (ELFT). J Biol Chem 269: 1033. 1994

230. Kumar R, Potvin B, Muller WA, Stanley P: Cloning of a human alpha( 1,3)-fucosyltransferase gene that encodes ELFT but does not confer ELAM-I recognition on Chinese hamster ovary cell transfectants. J Biol Chem 266:21777, 1991

23 1 . Weston BW, Smith PL, Kelly RJ, Lowe JB: Molecular cloning of a fourth member of a human alpha (1,3)fucosyltransferase gene family. Multiple homologous sequences that determine expression of the Lewis x, sialyl Lewis x, and difucosyl sialyl Lewis X

epitopes. J Biol Chem 267:24575, 1992 232. Weston BW, Nair RP, Larsen RD, Lowe JB: Isolation of

a novel human alpha (1,3)fucosyltransferase gene and molecular comparison to the human Lewis blood group alpha (1,3/1,4)fucosyltransferase gene. Syntenic, homologous, nonallelic genes encoding enzymes with distinct acceptor substrate specificities. J Biol Chem 267:4152, 1992

233. Natsuka S , Gersten KM, Zenita K, Kannagi R, Lowe JB: Molecular cloning of a cDNA encoding a novel human leukocyte alpha-l,3-fucosyltransferase capable of synthesizing the sialyl Lewis X determinant. J Biol Chem 269:16789, 1994

234. Sasaki S, Kurata K, Funuyama K, Nagata M, Watanabe E, Ohta S, Hanai N, Nishi T: Expression cloning of a novel alpha 1.3- fucosyltransferase that is involved in biosynthesis of the sialyl Lewis X carbohydrate determinants in leukocytes. J Biol Chem 269: 14730, 1994

235. Knibbs RN, Craig RA, Natsuka S, Chang A, Cameron M, Lowe JB, Stoolman LM: The fucosyltransferase FucT-VI1 regulates E-selectin ligand synthesis in human T cells. J Cell Biol 133:911, 1996

236. Wagers AJ, Lowe JB, S. KG: An important role for the a1,3-fucosyltransferase FucT-VI1 in leukocyte adhesion to E-selectin. Blood 88:2125, 1996

237. Li F, Wilkins PP, Crawley S, Weinstein J, Cummings RD, McEver RP: Post-translational modifications of recombinant P-selectin glycoprotein ligand-l required for binding to P- and E-selectin. J Biol Chem 271:3255, 1996

238. Potvin B, Kumar R, Howard DR, Stanley P: Transfection of a human alpha-( 1,3)fucosyltransferase gene into Chinese hamster ovary cells. Complications arise from activation of endogenous alpha-(] ,3)fucosyltransferases. J Biol Chem 265:1615, 1990

239. Smith PL, Gersten KM, Petryniak B, Kelly RJ, Rogers C, Natsuka N, Alford I11 JA, Scheidegger EP, Natsuka S , Lowe JB: Expression of the 01 1,3 fucosyltransferase Fuc-TVII in lymphoid aggregate high endothelial venules correlates with expression of L- selectin ligands. J Biol Chem 271:8250, 1996

240. Maly P, Thall AD, Petryniak B, Rogers CE, Smith PL. Marks RM, Kelly RJ, Gersten KM, Cheng G, Saunders TL, Camper SA, Camphousen RT, Sullivan FX, Isogai Y, Hindsgaul 0, von Andrian UH, Lowe JB: The FucT-VI1 a ( 1,3) fucosyltransferase lo- cus controls lymphocyte homing and blood leukocyte emigration through an essential role in L-, E-, and P-selectin ligand biosynthesis. Cell 86:643, 1996

24 I . Fukushima K, Hirota M, Terasaki PI, Wakisaka A, Togashi H, Chia D, Suyama N, Fukushi Y, Nudelman E, Hakomori S: Char- acterization of sialosylated Lewisx as a new tumor-associated antigen. Cancer Res 44:5279, 1984





242. Ohmori K, Takada A, Ohwaki I, Takahashi N, Yoshitaka F, Michiyuki M, Kiso M, Hasegawa A, Kannagi M, Kannagi R: A distinct type of sialyl Lewis X antigen defined by a novel monoclonal antibody is selectively expressed on helper memory T cells. Blood 82:2797, 1993

243. Hanski C, Sheehan J, Kiehntopf M, Stolze B, Stein H, Riecken EO: Increased number of accessible sugar epitopes defined with monoclonal antibody AM-3 on colonic mucins is associated with malignant transformation of colonic mucosa. Cancer Res 51:5342, 1991

244. Muroi K, Suda T, Nojiri H, Ema H, Amemiya Y, Miura Y, Nakauchi H, Singhal A, Hakomori S: Reactivity profiles of leukemic myeloblasts with monoclonal antibodies directed to sialosyl-Le(x) and other lacto-series type 2 chain antigens: Absence of reactivity with normal hematopoietic progenitor cells. Blood 79:713, 1992

245. Fukushi Y, Kannagi R, Hakamori S, Shepard T, Kulander BG, Singer JW: Localization and distribution of difucoganglioside (VI3NeuAcV333Fuc2nLc6) in normal and tumor tissues defined by its monoclonal antibody FH6. Cancer Res 45:3711, 1985

246. Sawada M, Takada A, Ohwaki I, Takahashi N, Tateno H, Sakamoto J, Kannagi R: Specific expression of a complex sialyl Lewis X antigen on high endothelial venules of human lymph nodes: possible candidate for L-selectin ligand. Biochem Biophys Res Com- mun 193:337, 1993

247. Knapp W, Majdic 0, Stockinger H, Bettelheim P, Liszka K, Koller U, Peschel C: Monoclonal antibodies to human myelo- monocyte differentiation antigens in the diagnosis of acute myeloid leukemia. Med Oncol Tum Pharm 1:257, 1984

248. Knapp W, Majdic 0, Blanchard D, Chia W, Feizi T, Gyimesi E, Hirabayashi Y, Howard CJ, Kniep B, Thompson JS, Wiels J, Wijdenes J: Cdw65 Cluster Workshop Report. New York, NY, Ox- ford University Press, 1995

249. Kniep B, Peter-Katalinic J, Muthing J, Majdic 0, Pick1 WF, Knapp W: The CDw65 monoclonal antibodies VIM-8 and VIM-I 1 bind to the neutral glycolipid V3FucnLcSCer. J Biochem 1996 (in press)

250. Watson SR, Imai Y, Fennie C, Geoffrey JS, Rosen SD, Lasky LA: A homing receptor-IgG chimera as a probe for adhesive ligands of lymph node high endothelial venules. J Cell Biol 1102221, 1990

25 1. Levinovitz A, Muhlhoff J, Isenmann S, Vestweber D: Identi- fication of a glycoprotein ligand for E-selectin on mouse myeloid cells. J Cell Biol 121:449, 1993

252. Walchek B, Watts G, Jutila MA: Bovine y16 T cells bind E-selectin via a novel glycoprotein receptor: First characterization of a IymphocyteE-selectin interaction in an animal model. J Exp Med 178:853, 1993

253. Moore KL, Stults NL, Diaz S, Smith DF, Cummings RD, Varki A, McEver RP: Identification of a specific glycoprotein ligand for P-selectin (CD62) on myeloid cells. J Cell Biol 118:445, 1992

254. Streeter PR, Rouse BTN, Butcher EC: Immunologic and functional characterization of a vascular addressin involved in lymphocyte homing into peripheral lymph nodes. J Cell Biol 107:1853, 1988

255. Sako D, Chang X-J, Barone KM, Vachino G, White HM, Shaw G, Veldman GM, Bean KB, Ahem TJ, Furie B, Cumming DA, Larsen GA: Expression cloning of a functional glycoprotein ligand for P-selectin. Cell 75:1179, 1993

256. Yang J, Galipeau J, Kozak CA, Furie BC, Furie B: Mouse P-selectin glycoprotein ligand- 1 : Molecular cloning, chromosomal localization and expression of a functional P-selectin receptor. Blood 87:4176, 1996

257. Moore KL, Eaton SF, Lyons DE, Lichenstein HS, Cum- mings RD, McEver RP: The P-selectin glycoprotein ligand from

human neutrophils displays sialylated, fucosylated, 0-linked poly- N-acetyllactosamine. J Biol Chem 269:23318, 1994

258. Steininger CN, Eddy CA, Leimgruber RM, Mellors A, Wel- ply JK: The glycoprotease of Pasteurella haemolytica A I eliminates binding of myeloid cells to P-selectin but not to E-selectin. Biochem Biophys Res Commun 188:760, 1992

259. Mellors A, Lo RY: 0-sialoglycoprotease from Pasteurella haemolytica. Methods Enzymol 248:728, 1995

260. Sutherland DR, Abdullah KM, Cyopick P, Mellors A Cleav- age of the cell-surface 0-sialoglycoproteins CD34, CD43, CD44, and CD45 by a novel glycoprotease from Pasteurella haemolytica. J Immunol 148:1458, 1992

261. Abdullah KM, Udoh EA, Shewen PE, Mellors A: A neutral glycoprotease of Pasteurella haemolytica AI specifically cleaves 0- sialoglycoproteins. Infect Immun 60:56, 1992

262. Norman KE, Moore KL, McEver RP, Ley K: Leukocyte rolling in vivo is mediated by P-selectin glycoprotein ligand-1. Blood 86:4417, 1995

263. Vachino G, Chang XJ, Veldman GM, Kumar R, Sako D, Fouser LA, Bemdt MC, Cumming DA: P-selectin glycoprotein ligand- 1 is the major counter-receptor for P-selectin on stimulated T cells and is widely distributed in non-functional form on many lymphocytic cells. J Biol Chem 270:21966, 1995

264. De Luca M, Dunlop LC, Andrews RK, Flannery J Jr, Ettling R, Cumming DA, Veldman GM, Bemdt MC: A novel cobra venom metalloproteinase, mocarhagin, cleaves a IO-amino acid peptide from the mature N terminus of P-selectin glycoprotein ligand receptor, PSGL-I, and abolishes P-selectin binding. J Biol Chem 270:26734, 1995

265. Li F, Erickson HP, James JA, Moore KL, Cummings RD, McEver R P Visualization of P-selectin glycoprotein-1 as a highly extended molecule and mapping of protein epitopes for monoclonal antibodies. J Biol Chem 271:6342, 1996

266. Damle NK, Klussman K, Dietsch MT, Mohagheghpour N, Aruffo A: GMP-140 (P-selectidCD62) binds to chronically stimulated but not resting CD4+ T lymphocytes and regulates their production of proinflammatory cytokines. Eur J Immunol22: 1789, 1992

267. Aruffo A, Dietsch MT, Wan H, Hellstrom KE, Hellstrom I: Granule membrane protein 140 (GMP-140) binds to carcinomas and carcinoma-derived cell lines. Proc Natl Acad Sci USA 89:2292, 1992

268. Stone JP, Wagner DD: P-selectin mediates adhesion of platelets to neuroblastoma and small cell lung cancer. J Clin Invest 92:804, 1993

269. Berg EL. Robinson MK, Wamock RA, Butcher EC: The human peripheral lymph node vascular addressin is a ligand for LECAM-1, the peripheral lymph node homing receptor. J Cell Biol 114:343, 1991

270. Berg EL, McEvoy LM, Berlin C, Bargatze RF, Butcher EC: L-selectin-mediated lymphocyte rolling on MAdCAM-I. Nature 366:695, 1993

271. Hemmerich S, Butcher EC, Rosen SD: Sulfation-dependent recognition of high endothelial venules (HEV)-ligands by L-selectin and MECA-79, and adhesion-blocking monoclonal antibody. J Exp Med 180:2219, 1994

272. Hanninen A, Taylor C, Streeter PR, Stark LS, Sarte JM, Shizuru JA, Simell 0, Michie SA: Vascular addressins are induced on islet vessels during insulitis in nonobese diabetic mice and are involved in lymphoid cell binding to islet endothelium. J Clin Invest 92:2509, 1993

273. Michie SA, Streeter PR, Bolt PA, Butcher EC, Picker LJ: The human peripheral lymph node vascular addressin. An inducible endothelial antigen involved in lymphocyte homing. Am J Pathol 143:1688, 1993





274. Imai Y, Singer MS, Fennie C, Lasky LA, Rosen SD: Identi- fication of a carbohydrate-based endothelial ligand for a lymphocyte homing receptor. J Cell Biol 113:1213, 1991

275. Lasky LA, Singer MS, Dowbenko D, Imai Y, Henzel WJ, Grimley C, Fennie C, Gillett N, Watson SR, Rosen SD: An endothelial ligand for L-selectin is a novel mucin-like molecule. Cell 69:927, 1992

276. Brustein M, Kraal G, Mebius RE, Watson SR: Identification of a soluble form of a ligand for the lymphocyte homing receptor. J Exp Med 176:1415, 1992

277. Kikuta A, Rosen SD: Localization of ligands for L-selectin in mouse peripheral lymph node high endothelial cells by colloidal gold conjugates. Blood 84:3766, 1994

278. Dowbenko D, Kikuta A, Fennie C, Gillett N , LdSky LA: Glycosylation-dependent cell adhesion molecule 1 (GlyCAM 1) mucin is expressed by lactating mammary gland epithelial cells and is present in milk. J Clin Invest 92:952, 1993

279. Baumhueter S, Singer MS, Henzel W, Hemmerich S, Renz M, Rosen SD, Lasky LA: Binding of L-selectin to the vascular sialomucin CD34. Science 262:436, 1993

280. Simmons DL, Satterthwaite AB, Tenen DG, Seed B: Molec- ular cloning of a cDNA encoding CD34, a sialomucin of human hematopoietic stem cells. J Immunol 148:267, 1992

281. Baumhueter S, Dybdal N, Kyle C, Lasky LA: Global vascular expression of murine CD34, a sialomucin-like endothelial ligand for L-selectin. Blood 84:2554, 1994

282. Young PE, Baumheter S, Lasky LA: The sialomucin CD34 is expressed on hematopoietic cells and blood vessels during murine development. Blood 85:96, 1995

283. Fina L, Molgaard HV, Robertson D, Bradley NJ, Monaghan P, Delia D, Sutherland DR, Baker MA, Greaves M F Expression of the CD34 gene in vascular endothelial cells. Blood 75:2417, 1990

284. Suzuki A, Andrew DP, Gonzalo J-A, Fukumoto M, Spell- berg J, Hashiyama M, Suda T, Takimoto H, Genvin N, Webb I, Gutierrez-Ramos J-C, Molinex G, McNiece I, Ley K, Butcher E, May WS, Greaves MF, Amakawa R, Tada Y, Wakcham A, Mak TW: CD34 deficient mice have reduced eosinophil accumulation after allergen exposure and reveal a novel cross-reactive protein. Blood 87:3550, 1996

285. Pun KD, Finger EB, Gaudemack G, Springer TA: Sialomu- cin CD34 is the major L-selectin ligand in human tonsil high endothelial venules. J Cell Biol 131:261, 1995

286. Hemmerich S, Rosen SD: 6'-sulfated sialyl Lewis X is a major capping group of GlyCAM-1. Biochemistry 33:4830, 1994

287. Hemmerich S, Bertozzi CR, Leffler H, Rosen SD: Identifi- cation of the sulfated monosaccharides of GlyCAM-1, an endothelial-derived ligand for L-selectin. Biochemistry 33:4820, 1994

288. Hemmerich S, Leffler H, Rosen SD: Structure of the 0- glycans in GlyCAM- I , an endothelial-derived ligand for L-selectin. J Biol Chem 270:12035, 1995

289. Crommie D, Rosen SD: Biosynthesis of GlyCAM- I , a mucin-like ligand for L-selectin. J Biol Chem 270:22614, 1995

290. Mebius RE, Streeter PR, Breve J , Duijvestijn AM, Kraal G: The influence of afferent lymphatic vessel interruption on vascular addressin expression. J Cell Biol I15:85, 1991

291. Duijvestijn AM, Horst E, Pals ST, Rouse 3N, Steere AC, Picker LJ, Meijer CJ, Butcher EC: High endothelial differentiation in human lymphoid and inflammatory tissues defined by monoclonal antibody HECA-452. Am J Pathol 130:147, 1988

292. Berg EL, Robinson MK, Mansson 0, Butcher EC, Magnani JL: A carbohydrate domain common to both sialyl Le (a) and sialyl Le (X) is recognized by the endothelial cell leukocyte adhesion molecule ELAM-I. J Biol Chem 266: 14869, 199 1

293. Berg EL, Yoshino Y, Rott LS, Robinson MK, Warnock RA,

Kishimoto TK, Picker LJ, Butcher EC: The cutaneous lymphocyte antigen is a skin lymphocyte homing receptor for the vascular lectin endothelial cell-leukocyte adhesion molecule- 1. J Exp Med 174:1461, 1991

294. Berg EL, Magnani J, Wamock RA, Robinson MK, Butcher EC: Comparison of L-selectin and E-selectin ligand specificities: The L-selectin can bind the E-selectin ligands sialyl Le(x) and sialyl Le(a). Biochem Biophys Res Commun 184:1048, 1992

295. Picker L, Terstappen L, Rott L, Streeter P, Stein H, Butcher E: Differential expression of homing-associated adhesion molecules by T cell subsets in man. J Immunol 145:3247, 1990

296. Picker LJ, Martin RJ, Trumble A, Newman LS, Collins PA, Bergstresser PR, Leung DY: Differential expression of lymphocyte homing receptors by human memory/effector T cells in pulmonary versus cutaneous immune effector sites. Eur J Immunol 24: 1269, 1994

297. Gelb AB, Smoller BR, Warnke RA, Picker LJ: Lymphocytes infiltrating primary cutaneous neoplasms selectively express the cutaneous lymphocyte-associated antigen (CLA). Am J Pathol 142:1556, 1993

298. Rossiter H, van Reijsen F, Mudde GC, Kalthoff F, Bruijn- zeel-Koomen CAFM, Picker LJ, Kupper TS: Skin-disease related T cells bind to endothelial selectins: Expression of cutaneous lymphocyte antigen (CLA) predicts E-selectin but not P-selectin binding. Eur J Immunol 24:205, 1994

299. Kishimoto TK, Warnock RA, Jutila MA, Butcher EC, Lane C, Anderson DC, Smith CW: Antibodies against human neutrophil LECAM- 1 (LAM- 1 /Leu-S/DREG-56 antigen) and endothelial cell ELAM-1 inhibit a common CD18-independent adhesion pathway in vitro. Blood 78:805, 1991

300. Spertini 0, Luscinskas FW, Gimbrone MA Jr, Tedder TF: Monocyte attachment to activated human vascular endothelium in vitro is mediated by leukocyte adhesion molecule- 1 (L-selectin) under non-static conditions. J Exp Med 175:1789, 1992

301. Buttrum SM, Hatton R, Nash GB: Selectin-mediated rolling of neutrophils on immobilized platelets. Blood 82:1165, 1993

302. Lawrence MB, Bainton DF, Springer TA: Neutrophil tethering to and rolling on E-selectin are separable by requirement for L-selectin. Immunity 1: 137, 1994

303. Patel KD, Moore KL, Nollert MU, McEver RP: Neutrophils use both shared and distinct mechanisms to adhere to selectins under static and flow conditions. J Clin Invest 961887, 1995

304. Ma L, Raycroft L, Asa D, Anderson DC, Geng JG: A sialo- glycoprotein from human leukocytes functions as a ligand for P- selectin. J Biol Chem 269:27739, 1994

305. Asa D, Raycroft L, Ma L, Aeed PA, Kaytes PS, Elhammer AP, Geng J-G: The P-selectin glycoprotein ligand functions as a common human leukcoyte ligand for P- and E-selectins. J Biol Chem 270: I 1662, I995

306. Diacovo TG. Roth SJ, Morita CT, Rosat J-P, Brenner MB, Springer TA: Interactions of human (YIP and ylb T lymphocyte subsets in shear flow with E-selectin and P-selectin. J Exp Med 183:1193, 1996

307. Lenter M, Levinovitz A, Isenmann S, Vestweber D: Mono- specific and common glycoprotein ligands for E- and P-selectin on myeloid cells. J Cell Biol 125:471, 1994

308. Steegmaier M, Levinovitz A, lsenmann I, Borges E, Lenter M, Kocher HP, Kleuser B, Vestweber D: The E-selectin ligand ESL- I is a variant of a receptor for fibroblast growth factor. Nature 373:615, 1995

309. Burruss LW, Zuber ME, Lueddecke BA, Olwin BB: Identi- fication of a cysteine-rich receptor for fibroblast growth factors. Mol Cell Biol 12:5600, 1992

310. Stieber A, Mourelatos Z. Chen YJ, Le Douarin N, Gonatas




SELECTINS AND THEIR LIGANDS

NK. MG160, a membrane protein of the Golgi apparatus which is homologous to a fibroblast growth factor receptor and to a ligand for E-selectin, is found only in the Golgi apparatus and appears early in chicken embryo development. Exp Cell Res 219562, 1995

3 1 1. Hynes RO: Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 69: 1 1, 1992

312. Crockett-Torabi E, Sulenbarger B, Smith CW, Fantone JC: Activation of human neutrophils through L-selectin and Mac-1 molecules. J Immunol 154:2291, 1995

313. Laudanna C, Constantin G, Baron P, Scarpini E, Scarlato G, Cabrini G, Dechecchi C, Rossi F, Cassatella MA, Berton G: Sulfatides trigger increase of cytosolic free calcium and enhanced expression of tumor necrosis factor-alpha and interleukin-8 mRNA in human neutrophils. Evidence for a role of L-selectin as a signaling molecule. J Biol Chem 269:4021, 1994

314. Waddell TK, Fialkow L, Chan CK, Kishimoto TK, Downey GP: Potentiation of the oxidative burst of human neutrophils. A signaling role for L-selectin. J Biol Chem 269:18485, 1994

315. Po JL, Mazer B, Jensen GS: The L-selectin antibody FMC46 mediates rapid, transient increase in intracellular calcium in human peripheral blood mononuclear cells and Daudi lymphoma cells. Bio- chem Biophys Res Commun 217:1145, 1995

316. Waddell TK, Fialkow L, Chan CK, Kishimoto TK, Downey G P Signaling functions of L-selectin. Enhancement of tyrosine phosphorylation and activation of MAP kinase. J Biol Chem 270:15403, 1995

317. Lo SK, Lee S, Ramos RA, Lobb R, Rosa M, Chi-Rosso G, Wright SD: Endothelial-leukocyte adhesion molecule 1 stimulates the adhesive activity of leukocyte integrin CR3 (CD1 lbKD18, Mac- 1, amp2) on human neutrophils. J Exp Med 173:1493, 1991

318. Simon SI, Bums AR, Taylor AD, Gopalan PK, Lynam EB, Sklar LA, Smith CW: L-selectin (CD62L) cross-linking signals neutrophil adhesive functions via the Mac-I (CDllbKDI8) beta 2- integrin. J Immunol 155:1502, 1995

319. Nagata K, Tsuji T, Todoroki N, Katagiri Y, Tanoue K, Ya- mazaki H, Hanai N, Irimura T: Activated platelets induce superoxide anion release by monocytes and neutrophils through P-selectin (CD62). J Immunol 151:3267, 1993

320. Elstad MR, La Pine TR, Cowley FS, McEver RP, McIntyre TM, Prescott SM, Zimmerman GA: P-selectin regulates platelet- activating factor synthesis and phagocytosis by monocytes. J Immu- no1 155:2109, 1995

321. Weyrich AS, McIntyre TM, McEver RP, Prescott SM, Zim- merman GA: Monocyte tethering by P-selectin regulates monocyte chemotactic protein- 1 and tumor necrosis factor-a secretion. Signal integration and NF-KB translocation. J Clin Invest 952297, 1995

322. Weyrich AS, Elstad MR, McEver RP, McIntyre TM, Moore KL, Monissey JM, Prescott SM, Zimmerman GA: Activated platelets signal chemokine synthesis by human monocytes. J Clin Invest 97:1525, 1996

323. Lenardo MJ, Baltimore D: NF-KB: A pleiotropic mediator of inducible and tissue-specific gene control. Cell 58:227, 1989

324. Tozeren A, Ley K: How do selectins mediate leukocyte

325. Williams AF: Out of equilibrium. Nature 352:473, 1991 326. Walchek B, Kahn J, Fisher JM, Wang BB, Fisk RS, Payan

DG, Feehan C, Betageri R, Darlak K, Spatola AF, Kishimoto TK: Neutrophil rolling altered by inhibition of L-selectin shedding in vitro. Nature 380:720, 1996

327. Lawrence MB, Berg EL, Butcher EC, Springer TA: Rolling of lymphocytes and neutrophils on peripheral node addressin and subsequent arrest on ICAM-1 in shear flow. Eur J Immunol25:1025, 1995

328. Alon R, Hammer DA, Springer TA: Lifetime of the P-selectin-carbohydrate bond and its response to tensile force in hydrodynamic flow. Nature 374539, 1995

329. Hammer DA, Apte SM: Simulation of cell rolling and adhesion on surfaces in shear flow: General results and analysis of selectin-mediated neutrophil adhesion. Biophys J 63:35, 1992

330. Hammer DA, Tempelman LA, Apte S M Statistics of cell adhesion under hydrodynamic flow: Simulation and experiment. Blood Cells 19:261, 1993

33 1. Tempelman LA, Hammer DA: Receptor-mediated binding of IgE-sensitized rat basophilic leukemia cells to antigen-coated sub- strates under hydrodynamic flow. Biophys J 66:1231, 1994

332. DeGrendele HC, Estess P, Picker LJ, Siegelman MH: CD44 and its ligand hyaluronate mediate rolling under physiologic flow: A novel lymphocyte-endothelial cell primary adhesion pathway. J Exp Med 183:1119, 1996

333. Finger EB, Pun KD, Alon R, Lawrence MB, von Andrian UH, Springer TA: Adhesion through L-selectin requires a threshold hydrodynamic shear. Nature 379:266, 1996

334. Nguyen M, Strubel N, Bischoff J: A role for sialyl Lewis- WA glycoconjugates in capillary morphogenesis. Nature 365:267, 1993

335. Kraling BM, Razon MJ, Boon LM, Zurakowski D, Seachord C, Darveau RP, Mulliken JB, Corless CL, Bischoff J: E-selectin is present in proliferating endothelial in human hemangiomas. Am J Pathol 148:1181, 1996

336. Folkman J, Shing Y: Angiogenesis. J Biol Chem 267:10931, 1992

337. Folkman J: Angiogenesis in cancer, vascular, rheumatoid and other disease. Nature Med 1:27, 1995

338. Schweitzer KM, Drager AM, van der Valk P, Thijsen SFT, Zevenbergen A, Theijsmeijer AP, van der Schoot CE, Langenhuijsen MMAC: Constitutive expression of E-selectin and vascular cell adhesion molecule- 1 on endothelial cells of hematopoietic tissues. Am J Pathol 148:165, 1996

339. Zannettino AC, Bemdt MC, Butcher C, Butcher EC, Vadas MA, Simmons PJ: Primitive human hematopoietic progenitors adhere to P-selectin (CD62P). Blood 853466, 1995

340. Vose JM, Armitage JO: Clinical applications of hematopoietic growth factors. J Clin Oncol 13:1023, 1995

rolling in venules? Biophys J 63:700, 1992




1996 88: 3259-3287

GS Kansas Selectins and their ligands: current concepts and controversies

http://www.bloodjournal.org/content/88/9/3259.citation.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American


http://www.bloodjournal.org/content/88/9/3259.citation.full.html

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

http://www.bloodjournal.org/site/subscriptions/index.xhtml





Documents

BLOOD The Journal of - Semantic Scholar · Investigator of the American Hearr Association. Address reprint requests to Geoffrey S. Kansas, PhD, ... of VCAM-1 and ICAM-1, two other